Effects of aluminum chloride-catalyzed hydrothermal pretreatment on the structural characteristics of lignin and enzymatic hydrolysis

Bioresource Technology 206 (2016) 57–64

Contents lists available at ScienceDirect

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

Effects of aluminum chloride-catalyzed hydrothermal pretreatment on the structural characteristics of lignin and enzymatic hydrolysis Xiao-Jun Shen, Bing Wang, Pan-Li Huang, Jia-Long Wen ⇑, Run-Cang Sun Beijing Key Laboratory of Lignocellulosic Chemistry, Beijing Forestry University, Beijing 100083, China

h i g h l i g h t s

g r a p h i c a l a b s t r a c t


AlCl3-catalyzed pretreatment can

obtain digestible substrate for glucose and lignin.
Removal of hemicelluloses is correlated with the pretreatment severities.
The structural characteristic of the lignin was comprehensively elucidated.
The pretreatment greatly facilitates the enzymatic hydrolysis of the substrates.

a r t i c l e

i n f o

Article history: Received 13 November 2015 Received in revised form 6 January 2016 Accepted 7 January 2016 Available online 26 January 2016 Keywords: Eucalyptus camaldulensis AlCl3-catalyzed hydrothermal pretreatment Enzymatic saccharification Lignin Structural features

a b s t r a c t In this study, Eucalyptus camaldulensis was pretreated with 0.02 M aluminum chloride (AlCl3) at 140– 180 °C to obtain digestible substrates for glucose and lignin. The effects of AlCl3-catalyzed hydrothermal pretreatment on the degradation of carbohydrates, structural changes of lignin, crystallinity, morphologic changes, and cellulose conversion of the pretreated biomass have been investigated by HAPEC, HPLC, FTIR, XRD, CP/MAS NMR, SEM, and 2D-HSQC NMR. Results showed that the pretreatment significantly removed hemicelluloses and cleaved b-O-4 linkages of lignin at high temperatures. Under an optimum condition (at 170 °C for 1 h), almost all of hemicelluloses were removed and most of b-O-4 linkages in lignin were cleaved, and 77.8% cellulose conversion of the pretreated biomass was achieved, which was 7.3-fold higher than that of the original biomass. In short, this process was regarded as a promising approach to achieve an efficient conversion of lignocellulosic biomass to fermentable glucose and residual lignin. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction Facing the impending depletion of fossil resources and environmental problems, utilization of lignocellulosic biomass as an alternative for fossil carbon sources in fuel and chemicals is one of the greatest challenges in the current world (Alvira et al., 2010). Lignocellulosic biomass has a complex structure consisting of three main polymeric components, namely, cellulose, hemicelluloses, and lignin. Cellulose and hemicelluloses, which are both ⇑ Corresponding author. E-mail address: [email protected] (J.-L. Wen). http://dx.doi.org/10.1016/j.biortech.2016.01.031 0960-8524/Ó 2016 Elsevier Ltd. All rights reserved.

composed of monomer sugars, can be converted into fermentable sugars by pretreatment and enzymatic hydrolysis. Lignin is composed of randomly cross-linked phenylpropanoid units and fills in the spaces between the cellulose fibers and hemicelluloses to protect them from any kinds of degradation assisted by enzyme or catalysts (Sanderson, 2011). Moreover, lignin is an irregular and reticulated polymer, which greatly limits the further utilization of lignocellulosic biomass. Hence, an efficient pretreatment is crucial to break the lignin seal to increase the accessibility of the cellulase enzyme to the cellulose. The targets of a typical pretreatment are relocation of lignin and hemicelluloses, increase the porosity of the biomass so as to improve the enzymatic

58

X.-J. Shen et al. / Bioresource Technology 206 (2016) 57–64

digestibility and increase the conversion yield of mono-sugars (Wyman et al., 1992). Up to now, many pretreatments have been developed and investigated, including physical, chemical, biological methods, and a combination of these methods (Isroi et al., 2011; Kamireddy et al., 2013; Zhu et al., 2009; Lloyd and Wyman, 2005). Among these, hydrothermal pretreatment (also known as hydrothermolysis, uncatalyzed solvolysis, hot-compressed water treatment, or liquid hot water pretreatment) has been considered to be an economical and eco-friendly pretreatment that can improve sugar release performance of biomass and achieve some important bio-products, such as hemicelluloses, oligosaccharides and furfural (Garrote et al., 1999). During hydrothermal pretreatment, organic solvents and catalysts are not required and equipment corrosion can be minimized (Alvira et al., 2010). In fact, the ionization constant (Kw) of water rapidly increases with the elevated temperatures and is about one thousand times higher than that of water at room temperature (Kumar et al., 2011). Therefore, the high ionization constant promotes the hydrolysis of lignocellulosic biomass since acidic medium facilitates the cleavage of ether and ester bonds and hydrolysis of hemicelluloses (Savage, 1999). The hydrothermal pretreatment has been widely investigated for several lignocellulosic materials such as poplar (Yang et al., 2012); wheat straw (Satari Baboukani et al., 2012), sugarcane bagasse (Duarte et al., 2012). Unluckily, long reaction time, high temperature and high energy consumption of hydrothermal pretreatment restricts its large-scale application (Stephanopoulos, 2007). More importantly, as compared to pure water at high temperature alone, inorganic acid (sulfuric acid and phosphoric acid) forms more hydronium ions (H3O+) at lower temperatures. It has the advantage of recovering 91% of xylose after dilute sulfuric acid pretreatment (Lloyd and Wyman, 2005). Some researchers also have investigated the dilute sulfuric acid pretreatment of coastal Bermuda grass from 120 to 180 °C over a range of acid concentrations (0.3–1.2% w/w) and residence time (5–60 min) (Redding et al., 2011). The optimal pretreatment condition was found to be 1.2% dilute sulfuric acid for 60 min, yielding about 94% of the total sugars from the biomass. However, the dilute acid pretreatment required a special reactor to resist corrosion concerns. To overcome this shortcoming, the use of metal salt solutions instead of inorganic acid has been proposed (Liu et al., 2009). Recently, the enhancement effects of NaCl, KCl, CaCl2, MgCl2, FeCl3, FeCl2, Fe2(SO4)3, and FeSO4, on the production of furfural from xylans (hemicelluloses) and degradation of cellulose have been evidenced (Liu et al., 2009; Yu et al., 2011; Cai et al., 2014). As shown by a previous report (Ma et al., 2012), the conversion ratio of cellulose increased from 40.4% (without catalyst) to 86.8%, and the content of glucose in the final products increased from 25.9% (without catalyst) to 64.5% after the extremely low AlCl3 catalyzed pretreatment. Although some researchers investigated the effects of metal salts solution on the enzymatic hydrolysis of lignocellulosic biomass (Kamireddy et al., 2013), the effects of metal salts solution on the structural features of lignin, and structural features of residual lignin in the substrates as well as the corresponding enzymatic hydrolysis has been generally ignored. Lignin is principally comprised of three main lignin building blocks, p-hydroxyphenyl, guaiacyl, and syringyl units, linked by carbon–carbon (e.g., b-b, b-5, b-1 and 5–5) and ether bonds (e.g., a-O-4 and b-O-4) (Wen et al., 2013). In fact, the existence of lignin after pretreatment inhibits subsequent enzymatic hydrolysis, and the inhibitory mechanism was proposed as follows: Lignin limits the enzyme access to cellulose by forming a physical barrier; meanwhile, it also non-productively adsorbs enzyme, which probably reduces the enzymatic efficiency (Kumar and Wyman, 2013; Kumar et al., 2012). However, the structural characteristic of the residual lignin

in the substrates, which affects the efficiency of enzymatic hydrolysis of the pretreated biomass, still needs to be deeply investigated. In this study, low concentration of aqueous aluminum chloride (AlCl3) was used to pretreat the feedstock (Eucalyptus camaldulensis) for understanding the effects of AlCl3 pretreatment on the chemical structures of lignin and subsequent enzymatic hydrolysis of the pretreated substrates (see Fig. 1). The liquid fractions obtained after pretreatment were analyzed by a high performance liquid chromatography (HPLC) and a high performance anion exchange liquid chromatography (HPAEC). The chemical and morphological characterizations of original and the pretreated residues were analyzed by using different analytical techniques, such as Fourier Transform Infrared (FTIR), X-ray diffraction (XRD), crosspolarization magic angle spinning carbon-13 nuclear magnetic resonance (CP/MAS 13C NMR), and scanning electron microscopy (SEM). Moreover, structural features of lignin, such as S/G ratios and the percentage of major substructures (inter-coupling bonds, b-O-4, b-b, b-5, etc.), were qualitatively and quantitatively acquired according to two-dimensional hetero-nuclear single-quantum coherence nuclear magnetic resonance (2D-HSQC NMR). In short, all the results will provide useful information in the value-added applications of Eucalyptus species for producing of glucose and residual lignin for the production of phenol formaldehyde adhesive.

2. Methods 2.1. Materials The raw biomass, E. camaldulensis, was obtained from Guangxi province, China. It was cut into pieces, milled and screened in a mill (DFT-200A, Shanghai) to obtain a 20–40 mesh fraction. The composition of Eucalyptus wood was 43.5% glucan, 15.7% xylan, 0.6% galactan, 0.8% mannan, 26.3% Klason lignin and 3.94% acetyl groups in terms of dry weight, which was analyzed by the standard NREL method (Sluiter et al., 2008b). AlCl36H2O (>97%) and CH3COONa (>99%) were purchased from Xilong Chemical Co., Ltd (Guangdong, China). H2SO4 (95–98%) was supplied by Beijing Chemical Company (Beijing, China). Liquid-state cellulase (100 FPU/mL) were supplied by Novozymes, Beijing, China.

2.2. AlCl3-catalyzed hydrothermal pretreatment in a batch reactor The pretreatment experiments of the biomass were performed in a batch reactor (100 mL internal volume, Sen Long Instruments Company, Beijing, China). The reactor was made of Hastelloy C-276 to mitigate the effects of acid corrosion at high temperatures. A 7 g (oven dried) quantity of dry biomass was added to 70 ml of 0.02 M aqueous AlCl3 (equivalent to 1:10, solid:liquid ratio). The batch reactor was heated by electrical heating, the temperature was kept constant at the target temperature of 140, 150,160, 170 or 180 °C for 1 h, and the reactor was stirred at 1000 rpm in order to provide homogeneous mixing in the batch reactor. After the desired temperature achieved, the reaction time was initiated and the temperature in the reactor was maintained constant. When the reaction was finished, the reactor was immediately cooled by passing the cold water into the external jacket until ambient temperature. After releasing the gas and pressure, the reaction mixture was filtrated with a Buchner funnel. The filtrate was stored in refrigerator for further analysis, and the solid residue was washed thoroughly with distilled water until the liquid in the filter flask became colorless and further freeze-dried by a 216V-230 freeze dryer (Thermo Fisher Scientific, Shanghai, China). The pretreated substrates were labeled as S140, S150, S160, S170 and S180, respectively, corresponding

59

X.-J. Shen et al. / Bioresource Technology 206 (2016) 57–64

Fig. 1. Schematic illustration of the experimental procedure.

to the pretreatment at various temperatures. S0 represents for the original biomass. 2.3. Preparation of residual lignin from original and the pretreated substrates To maximize remove the carbohydrates from original and the pretreated biomass and remain lignin as residual lignin, original and the pretreated biomass were undergone ball-milling and excessive amounts cellulase hydrolysis. The ball-milling process was performed in a planetary ball mill (FritschGMBH, Idar-Oberstein, Germany) for 2 h. 0.5 g control and pretreated substrates was milled at a time. The milling bowl was composed of zirconium dioxide (80 mL) and contained 25 zirconium dioxide balls (0.5 cm diameter). The milling was conducted at room temperature under N2 atmosphere with a milling frequency of 450 rpm. The ball-milled substrates (0.5 g) were suspended in sodium acetate buffer (25 mL, pH 4.8) with loading excessive celluclast (50 FPU/g substrate). The reaction mixture was incubated at 50 °C in a rotary shaker (150 rpm) for 72 h. Residual lignin was washed extensively with acidic water (pH = 2.0) and sodium acetate buffer (pH = 4.8) and then freeze-dried for further analysis. The residual lignin from the pretreated substrates were labeled as L140, L150, L160, L170 and L180, respectively, corresponding to the pretreatment at various temperatures. L0 represents for the residual lignin from original biomass. 2.4. Enzymatic hydrolysis Enzymatic hydrolysis was carried out at 2% of substrate (w/v) in 10 mL of 50 mM sodium acetate buffer (pH 4.8) using shaking

incubators (ZWYR-2102C) (Shanghai, China) at 150 rpm for 72 h. Celluclast was provided from novozymes (Beijing, China) and employed at the activity of 15 FPU/g substrate for all the samples. The hydrolyzates was analyzed by using HPAEC system with an integral amperometric detector and CarboPac PA100 (4  250 mm, Dionex) analytical column according to the literature (Sun et al., 2014). Cellulose conversion (%) on the basis of glucose release was calculated according to Eq. (1) (Rocha et al., 2013):

Cellulose conversion ð%Þ ¼

mglucose  f h  100 minitial  yi

ð1Þ

where cellulose conversion (%) is the enzymatic conversion of cellulose (%), mglucose is the glucose mass in the enzymatic hydrolysate (g), minitial is the initial dry mass of original and the pretreated substrates (g), yi is the cellulose content in the pretreated bagasse sample (%) (see Table 1), fh is the conversion factor taking into account water addition upon hydrolysis (fh = 0.9 for glucose to glucan).

2.5. Characterization of residual lignin Heteronuclear single quantum coherence (HSQC) NMR spectra of the lignin residues were recorded on a Bruker AVIII 400 MHz spectrometer at 25 °C in DMSO-d6. Fifty milligrams (50 mg) of samples were dissolved in 0.5 mL of DMSO-d6, then, the dissolved samples were transferred to NMR sample tubes. The sequence was conducted as previously (Wen et al., 2013).

60

X.-J. Shen et al. / Bioresource Technology 206 (2016) 57–64

Table 1 Solid yield, chemical compositions of the original and pretreated substrates.a Temp. (°C) e

Control 140 150 160 170 180

Solid yield (%)

pH

Cellulose (%)

Hemicelluloses (%)

Klason lignin (%)

Cellulose recoveryb (%)

Hemicelluloses recoveryc (%)

Lignin recoveryd (%)

100 72.7 67.7 63.2 60.3 54.2

3.55 2.12 2.06 1.99 1.95 1.89

43.5 50.6 49.9 48.6 46.6 43.2

17.1 7.3 2.8 0.6 0.5 0.5

26.3 32.4 36.4 42.6 45.9 51.6

100.0 84.6 77.7 70.6 64.6 53.8

100.0 31.1 11.0 2.1 1.8 1.7

100.0 89.5 93.7 102.2 105.1 106.2

a

Reaction condition:7 g substrate in 70 ml 0.02 mol/L AlCl3 solution, 1 h. Cellulose recovery (%) = Cellulose recovered in solid residue after pretreatment (g)/Initial amount of Cellulose (g)  100 in the raw material. Hemicelluloses recovery (%) = Hemicelluloses recovered in solid residue after pretreatment (g)/Initial amount of Hemicelluloses (g)  100 in the raw material. d Lignin recovery (%) = Lignin recovered in solid residue after pretreatment (g)/Initial amount of lignin (g)  100 in the raw material. ðg=LÞ43 e The content of acetyl groups of the biomass was measured with following formula: Acetyl group % in biomass ¼ acetic acid in the hydrolysate  100% where 43 is the 60 molecular weight of acetyl groups, 43 is the molecular weight of acetic acid. b

c

2.6. Characterization of original and the pretreated substrates The chemical compositions (%, w/w) of all the samples were determined according to the NREL standard analytical method (NREL/TP-510–42618) (Sluiter et al., 2008b). In addition, the acetyl groups in the raw biomass was released by acid hydrolysis method as same as chemical compositions of the biomass (NREL/TP-510– 42618) and its content was measured according to the analytical method (NREL/TP-510–42623). The contents of furfural, hydroxymethylfurfural (HMF), acetic acid, formic acid, levulinic acid and lactic acid in the hydrolysates were determined according to the analytical method (NREL/TP-510–42623) (Sluiter et al., 2008a). FT-IR spectra were scanned using a Thermo Scientific Nicolet iN10 FT-IR Microscope (Thermo Nicolet Corporation, Madison, WI, USA) with liquid nitrogen-cooled MCT detector. The spectra were recorded in the region of 4000–800 cm1 at a resolution of 4 cm1 (64 scans per sample). X-ray diffraction (XRD) in reflection mode was recorded using an XRD-6000 (Shimadzu, Japan) with Cu Ka radiation (k = 1.5418 Å) generated at 40 kV and 40 mA. The diffraction patterns were measured from 2° to 45° at a scanning speed of 2°/min. The microstructural changes and surface characteristics of all the substrates were analyzed with a scanning electron microscope (SEM) (Hitachi S-3400N II, HITACHI Company, Japan) operating at 10 kV acceleration voltage. All samples were coated with gold prior to acquiring images. Solid-state crosspolarization/magic angle spinning (CP/MAS) 13C NMR spectra of the substrates were obtained at 100.6 MHz using a Bruker AV-III 400 M spectrometer (Germany). 3. Results and discussion 3.1. Solid yield and composition analysis, and characteristics of the liquid from the pretreated substrates E. camaldulensis was subjected to 0.02 M aqueous AlCl3 at 140–180 °C for 1.0 h, and the pretreated residues and liquids were separated. Solid yield, chemical compositions and pH values of the liquid are listed in Table 1. As shown in Table 1, the pH of the initial solution without pretreatment was 3.55, while it sharply reduced to 1.89 in the pretreated slurries. The reason for this decrease is that the high temperature decreases the viscosity of the solution and increases the ionic mobility as well as the ionization constant (Kw) of water (Savage, 1999). In other words, AlCl3 released more hydronium ions (H3O+) during the pretreatment, which led to the cleavage of acetyl linkages of hemicelluloses and formed acetic acid. Inevitably, this cause the decrease of the pH values of the pretreated liquids. The solid yield after the pretreatment declined from 72.7% to 54.2% as the temperatures elevated from 140 to 180 °C. The

decreased yield is mostly ascribed to the severe degradation of hemicelluloses and slight dissociation of lignin during the pretreatment. The compositional changes of original and the pretreated biomass could be revealed by the detailed compositional analysis. The major component of the untreated E. camaldulensis is cellulose (expressed as glucan), followed by hemicelluloses (expressed as xylan with small amounts of galactose) and lignin (expressed as Klason lignin). The composition of Eucalyptus wood was 43.5% Glucan (Cellulose), 15.67% Xylan, 0.6% Galactan, 0.8% Mannan, 26.3% Klason lignin and 3.9% acetyl groups in terms of dry weight, which was analyzed by the standard NREL method (Sluiter et al., 2008b). After the pretreatment, the contents of hemicelluloses in the pretreated samples rapidly decreased with elevated pretreatment temperature. The reason for this decrease is that the water autoionization and hydrolysis of AlCl3 can generate hydronium ions (H3O+), leading to the degradation of hemicelluloses by selective hydrolysis of glycosidic linkages, and then liberating O-acetyl groups and other acid moieties (Gütsch et al., 2012). The hydronium ions were further generated from the released acetic acid, which in turn enhanced the degradation of hemicelluloses at high temperatures, especially temperatures higher than 160 °C. By contrast, the content of cellulose firstly increased from 43.5% (raw biomass) to 50.6% at 140 °C, while decreased to 43.2% at 180 °C, suggesting that some amounts of cellulose were degraded under the higher temperatures. In addition, it was observed that the content of Klason lignin increased from 32.4% to 51.6% as the pretreatment temperature increased from 140 to 180 °C. The relatively high lignin content was mostly attributed to the significant degradation and concomitant loss of hemicelluloses under the high pretreatment temperatures. Furthermore, the lignin recovery in the pretreated substrates was over 100% under the high temperatures, suggesting that partial Klason lignin actually was pseudo-lignin. Explanation for the emergence of pseudo-lignin was possibly that condensation of sugar-derived products and lignin resulted in the accumulation of pseudo-lignin at high temperatures, which increased the relative amount of Klason lignin (Sannigrahi et al., 2011). During the AlCl3-catalyzed hydrothermal process, pentoses were degraded into furfural and further into formic acid, while the decomposition of hexoses resulted in the formation of HMF and further to formic acid as well as levulinic acid. Meanwhile, acetic acid was typically generated from the cleavage of the acetyl groups in the hemicelluloses components, which resulted in a decrease of pH value (Table 1). These compounds have been defined as potential inhibitors for further fermentations (Jonsson et al., 2013). The inhibitors formed during the pretreatment process are depicted in Table 2. Acetic acid was ubiquitous in the hydrolysate, and its content increased from 2.63 g/L at 140 °C to 5.23 g/L at 180 °C. This increase of acetic acid over 150 °C suggested that the hemicelluloses were mostly hydrolyzed during

61

X.-J. Shen et al. / Bioresource Technology 206 (2016) 57–64 Table 2 Inhibitors in the hydrolysate during the AlCl3-catalyzed hydrothermal pretreatment (unit, g/L). T (°C)

140 150 160 170 180 a b

Sugar analysisa (g/L)

Inhibitor analysisb (g/L)

Glu

Xyl

Man

Rha

Ara

Gal

FF

HMF

FA

AA

LA

LAA

0.39 0.63 0.61 0.45 0.48

1.96 0.94 0.07 0.01 0.00

0.26 0.07 0.07 0.04 0.02

0.05 0.03 0.00 0.00 0.00

0.04 0.00 0.00 0.00 0.00

0.02 0.00 0.00 0.00 0.00

1.99 3.05 2.85 2.47 1.31

0.14 0.53 0.80 1.53 1.09

0.00 0.00 1.90 2.27 3.67

2.63 3.54 3.67 4.27 5.23

0 0.38 0.93 2.47 5.23

2.06 7.14 9.07 11.42 12.08

Glu = Glucose, Xyl = Xylose, Man = Mannose, Rha = Rhamnose, Ara = Arabinose, Gal = Galactose. FF = Furfuran, HMF = Hydroxymethylfurfural, FA = Formic acid, AA = Acetic acid, LA = Levulinic acid, LAA = Lactic acid.

the pretreatment. It was found that the yield of HMF and furfural increased with the raised pretreatment temperature. For instance, the concentration of HMF increased from 0.14 to 1.53 g/L as the temperature increased from 140 to 170 °C, suggesting that more glucose were converted into HMF at the high temperatures, while its concentration decreased to 1.09 g/L with the temperature further increasing to 180 °C. Meanwhile, the concentration of formic acid and levulinic acid reached maximum, which is attributable to the fact that furfural can be degraded through hydrolytic fission of the aldehyde group into formic acid, and HMF can be converted into levulinic acid and formic acid under the acidic hydrothermal conditions (Nitsos et al., 2013). Furthermore, as a degradation product from carbohydrates, lactic acid (LAA), its content increased from 2.06–12.08 g/L with the elevated pretreatment temperature from 140–180 °C, which is in agreement with a previous study (Sitompul et al., 2014). 3.2. FT-IR spectra of original and the pretreated substrates FT-IR spectroscopy is frequently used to investigate the structural changes of the substrates during the pretreatment. Fig. S1 shows the FT-IR spectra of original and the pretreated substrates. The signals observed at 3348, 2903, 1107, 899 cm1 represent for O–H stretching of lignin and polysaccharides (cellulose and survived hemicelluloses), C–H stretching of amorphous and crystalline cellulose, the C–O and C–C stretching or C–OH of cellulose and hemicelluloses, b-glycosidic linkages between glucose units in cellulose and hemicelluloses (Sun et al., 2014), respectively. Additionally, an intense band at 1739 cm1 (C@O a stretching vibrations of the acetyl ester unit in hemicelluloses) was observed. However, this band became inconspicuous in Fig. S1, which was attributable to the removal and/or deacetylation of hemicelluloses during the process. Specially, the band almost disappeared at 180 °C, suggesting that a remarkable de-esterification of hemicelluloses occurred, as confirmed by the aforementioned composition analysis. Furthermore, the bands corresponding to aromatic skeletal vibrations in lignin (1595, 1502, and 1422 cm1), breathing vibration of syringyl and condensed guaiacyl (1326 cm1), C–H out-of-plane stretching in H units (829 cm1), and the C–H deformation combined with aromatic ring vibration (1454 cm1), could be distinctly observed in the spectra of original and the pretreated substrates, indicating that the basic aromatic structures of lignin were not significantly damaged during the processes under the conditions given. 3.3. X-ray diffraction and CP/MAS

13

C-NMR spectra

The crystallinity of the lignocellulosic biomass also plays an important role in enzymatic hydrolysis (Yang et al., 2012). The relationship between crystallinity and enzymatic hydrolysis has been discussed previously (Sun et al., 2014). In this study, the CrIs of original and the pretreated substrates were determined by XRD patterns (Fig. S2). As shown in Fig. S2, the region around 16.5°, 22.5°, and 35.0° is due to the (1 0 1), (0 0 2), and (0 4 0) characteristic

planes of cellulose I, while the peak of the amorphous cellulose appeared at 18.5° (Segal et al., 1959). The CrI of the original substrate was determined to be 32.1%, while CrI of the pretreated substrates increased slightly to 37.9% as the temperatures elevated to 170 °C. This increased trend indicated that the amorphous hemicelluloses were mostly removed after the pretreatment, as revealed by FT-IR spectra and compositional analysis of the substrates. However, the CrIs of the pretreated substrates declined as the pretreatment temperature reached to 180 °C, which was probably due to the partly degradation of crystalline cellulose at a higher temperature. CP/MAS solid-state 13C-NMR technique is a powerful tool to investigate the structural changes of biomass during different pretreatments. In this study, CP/MAS 13C-NMR spectra (Fig. S3) were used to observe the chemical and structural changes of the substrates under different temperatures. As shown in Fig. S3, most of the distinctly signals are distributed from 50.0 to 160.0 ppm, which are mostly attributed to cellulose, hemicelluloses and lignin. The peak at 171.0 ppm is mainly assigned to C@O groups in acetyl groups of hemicelluloses, while the methyl in acetyl groups is observed at 20.2 ppm. After the pretreatment, it was observed that the signal intensities of acetyl groups from hemicelluloses were sharply weaken under the high temperatures given, indicating that the hemicelluloses were susceptible to be degraded during the process, which correlates well to the aforementioned compositional analysis (the content of hemicelluloses rapidly decreased from 17.1% to 0.5% with the increase of pretreatment temperature) and the FT-IR analysis (the peak at 1734 cm1 was almost disappeared with the increased temperature). The region of C-4 of crystalline cellulose is distributed in the region of 86.0–92.0 ppm, while the region between 80.0 and 86.0 ppm is derived from C-4 of the amorphous cellulose, respectively. The intensity of signals for amorphous cellulose reduced with the increase of the pretreatment temperature. Concerning lignin, the signals primarily appeared in the downfield of aromatic regions. For example, the peak at 202.2 ppm is mainly assigned to carbonyl groups in the substrates. Additionally, the weak peak at 195.1 ppm is attributed to the aAC@O and c-CHO in cinnamaldehyde (Wen et al., 2013). With the increase of the pretreatment temperature, the signal at 202.2 ppm was gradually decreased and the signal at 195.1 ppm was slightly increased, suggesting that more cinnamaldehyde compounds were generated with the increase of temperature during the process. In addition, it was found that the signal for etherified S3,5 (151.9 ppm) reduced while that for non-etherified S3,5 (148.0 ppm) increased as the pretreatment temperature rose, implying that an extensive cleavage of b-O-4 linkages occurred in the process with the increment of the pretreatment temperature (Wen et al., 2014). 3.4. SEM analysis of original and the pretreated substrates The morphological changes of original and the pretreated substrates have been investigated by SEM imaging (Fig. S4). Initially,

62

X.-J. Shen et al. / Bioresource Technology 206 (2016) 57–64

E. camaldulensis without pretreatment appeared a smooth and regular surface, which indicated a laminated and highly ordered surface structure of cellulose (Figs. S4a1 and S4a2). The pretreated substrate at 140 °C still exhibited a similar morphological structure as the raw material on the whole (Figs. S4b1 and S4b2), while the surface of the pretreated substrate began to appear slight cracks and expose the internal structure of the lignocellulosic residue. The substrate pretreated at 150 °C showed some holes on the surface, indicating that the original structure begins to be loose (Fig. S4c). Furthermore, it was found that many holes and loose structure appeared on the surface of the substrate pretreated at 160 °C (Fig. S4d). The phenomenon implied that a considerable amount of hemicelluloses has been removed during the AlCl3-catalyzed hydrothermal pretreatment, which correlated well to the compositional analysis and the FT-IR spectra. When the pretreatment temperature was reached to 170 and 180 °C, an increasing number of collapsed areas, debris of collapsed structures in small sizes were observed, and the most significant change was numerous spherical droplets with kinds of size distribution appeared on the surface of the pretreated substrates (Figs. S4e and S4f), which might play a physical barrier in inhibiting the access of enzyme to cellulose and directly affect the subsequent enzymatic hydrolysis of the substrate. The appeared spherical droplet suggested that the lignin was melted, repolymerized, and finally formed spherical droplet for minimizing its surface contact area (Trajano et al., 2013). Another explanation for the appearance of spherical droplets is possibly due to the formation of lignin-like materials (pseudo-lignin) during the AlCl3-catalyzed hydrothermal process, which was derived from the dehydrated carbohydrates (Kumar et al., 2013; Sannigrahi et al., 2011). 3.5. 2D-HSQC NMR spectral analysis of residual lignin To further understand the structural features and changes of the residual lignin obtained from the enzyme hydrolysis process, 2D-HSQC NMR was applied for analysis of the lignins. Generally, the 2D-HSQC spectra can be divided into side-chain (dC/dH 50.0–9 0.0/2.50–6.00) and aromatic regions (dC/dH 100.0–150.0/5.50–8.5 0). The main substructures of the lignins are shown in Fig. S6. The assignments of cross-signals in the HSQC spectra of lignin are assigned based on the previous publications and listed in Table S1 (Wen et al., 2013). In the side-chain region of the spectra, the signals of different interunit linkages are depicted in Fig. S5A, such as b-aryl-ether (b-O-4, A), resinol (b-b, B), and phenylcoumaran (b-5, C). In this region, the cross-peaks of methoxy groups (OCH3, dC/dH 55.7/3.70) and b-O-4 substructures (A/A0 ) were the most predominant. The cross-signals of Ca–Ha in b-O-4 substructure are explicitly located at dC/dH 71.6/4.83, whereas the cross-signals at dC/dH 83.9/4.30 and 85.9/4.11 are attributed to its Cb–Hb correlations linked to G and S units in b-O-4 substructures, respectively. Meanwhile, the obvious signal at dC/dH 59.5/3.69 is attributed to the Cc–Hc in b-O-4 substructure. In addition, resinol (b-b, B) was also detected, and the Ca–Ha, Cb–Hb and Cc–Hc correlations were observed at dC/dH 84.9/4.64, 53.5/3.05, 71.0/4.16 and 3.79, respectively. Moreover, phenylcoumaran (b-5, C) substructures were found at Ca–Ha (dC/dH 86.8/5.48), Cb–Hb (dC/dH 52.4/3.45), and Cc–Hc (dC/dH 62.3/3.70). Other distinct signal, which locates at dC/dH 61.2/4.10 ppm in the side-chain region, was assigned to the Cc–Hc correlation of cinnamyl alcohol end groups (I). In the aromatic regions of the 2D-HSQC spectra (Fig. S5B), correlated signals from p-hydroxyphenyl (H), guaiacyl (G/G0 ), and syringyl (S/S0 ) lignin units were observed. Under the mild conditions, no p-hydroxyphenyl (H) was observed in the spectra of lignin fractions. As the temperature rose, signal for the C2,6–H2,6 of H-type lignin unit became more and more obvious, which implied that the

demethoxylation of G-S type lignin units occurred under high temperatures. The normal S-type lignin unit exhibited a prominent signal for the C2,6–H2,6 correlation at dC/dH 104.0/6.72, while the correlated signal for the Ca-oxidized S unit structure appeared at dC/dH 106.3/7.31 (C2,6–H2,6). By contrast, the G unit showed different correlations for C2–H2, C5–H5 and C6–H6 at dC/dH 111.0/6.99, 114.8/6.68 + 6.91 and 119.8/6.80, respectively. Two observed signals at C-5 are attributed to different substituents at the C-4 position (etherified and no-etherified C-4 in aromatic ring) (Wen et al., 2013). With an increment in pretreatment temperature, signal intensity of C6–H6 in guaiacyl unit (G) gradually was decreased and almost disappeared at high temperatures, which is attributable to the fact that the condensation reaction probably occurred at C-6 of guaiacyl unit in lignin. Moreover, guaiacyl unit (G) exhibited correlation for C5–H5 (dC/dH 115.1/6.95) when the pretreatment temperature was lower than 170 °C, however, with further increase of the pretreatment temperature, no signal for etherified guaiacyl unit was observed, indicating that the etherified guaiacyl unit can only survive at relatively low temperatures. Besides the signals for normal aromatic rings of the lignin, some condensed lignin structures were probably formed under severe conditions (S2,6, Fig. S5B). The condensed S-type lignin unit was presented around dC/dH 105.3/6.45. A previous study showed that condensed lignin structure in G-type lignin was C-2 of guaiacyl unit linked at the C-5 to other lignin side chains (Wen et al., 2014). As can be seen, the condensed S-type lignin unit structure was rapidly increased, suggesting that the condensed S-type lignin unit was formed at high temperatures. However, the detailed whole structures of the condensed lignin formed by S units remained unknown. Quantification of the residual lignin by 2D-HSQC NMR method can provide explicit structural evolution of lignin during this pretreatment under different temperatures. According to the computing method of the literatures, the different linkages could be expressed by a comparative mode (Sette et al., 2011; Wen et al., 2013). As shown in Fig. S5B, the relative content of b-O-4 linkage in the residual lignin is 55.7/100Ar, while it has decreased to 40.4/100Ar as the temperature increased to 140 °C, suggesting that b-O-4 linkage was cleaved to some extent during the AlCl3 pretreatment. When temperature further increased to 180 °C, the relative content of b-O-4 linkage was dramatically decreased to only 7.1/100Ar, implying again that the pretreatment led to the depolymerization of lignin at high temperature of 180 °C. All the data herein indicated that the cleavage of b-O-4 aryl ether is the main reaction during the AlCl3 pretreatment at the higher temperature. In addition, the contents of other carbon–carbon linkages (e.g., b-b and b-5) also diminished with the increased temperature. Moreover, changes of the S/G ratios were also indicated prominent structural alteration to evaluate the pretreatment. The S/G ratios increased as the pretreatment temperature rose and reached a maximum ratio (S/G = 11.2) in L160, suggesting that the G-type lignin units were more easily degraded under the temperatures given. However, with the pretreatment temperature further increased from 160 to 180 °C, the S/G ratios gradually decreased (S/G = 4.2) in L180, implying that both G-type and S-type lignin were degraded at relatively high temperatures. 3.6. Enzymatic saccharification The enzymatic digestibility is used to evaluate the pretreatment efficiency. Fig. 2 shows the cellulose conversion of the raw material without pretreatment and the pretreated substrates obtained under different pretreatment conditions. The AlCl3-catalyzed hydrothermal pretreatment could effectively improve the enzymatic digestibility of E. camaldulensis. As can be seen, the cellulose digestibility was just 10.6% for the raw material without pretreatment after 72 h enzymatic hydrolysis. Interestingly, the cellulose

X.-J. Shen et al. / Bioresource Technology 206 (2016) 57–64

63

Fig. 2. Glucose yields of enzymatic hydrolysis of original and the pretreated substrates under different pretreatment temperatures.

conversion was enhanced from 14.7% to 77.8% as the temperature increased from 140 to 180 °C. The significant effect of the AlCl3-catalyzed hydrothermal pretreatment on the cellulose conversion was highly related to the degradation of plenty of hemicelluloses (Table 1) and the cleavage of lignin–carbohydrate bonds during the pretreatment process, which could remove the barriers for the accessibility of enzyme to cellulose. As is wellknown, many factors influence the enzymatic digestibility of lignocellulosic biomass, such as lignin barrier, hemicelluloses content, and porosity (Alvira et al., 2010). The presence of lignin has a negative effect on the yields of enzymatic hydrolysis since nonproductive adsorption of lignin on enzyme leads to the formation of lignin–enzyme complex, which is considered to be ineffective for enzymatic hydrolysis process (Ko et al., 2015). However, in this study, the cellulose digestibility was still high (53.4–77.8%) in the presence of high content of lignin (42.6–51.6%) of the substrates. This suggested that the content of residual lignin in the pretreated substrates has slight effect on the cellulose digestibility in this study. The cellulose conversion of the substrates pretreated at 140 and 150 °C reached just to 14.7 and 27.4%, respectively, whereas the substrates pretreated at 160–180 °C achieved high digestibility of cellulose in the substrates (53.4–77.8%). The reason for this increase is related to the following reasons: (1) The decreased content of hemicelluloses was probably highly related to the enhanced cellulose digestibility (Table 1 and Fig. 2); (2) Aluminum ion (Al3+) eliminated the lignin inhibition through formation of lignin–metal complex and more active sites of cellulose were accessible for subsequent enzymatic hydrolysis (Kamireddy et al., 2013). However, it was also observed that the initial cellulose conversion (<12 h) of the substrate pretreated at 180 °C is high as same as that 170 °C, while the final cellulose conversion of the substrate is only 53.4%, which is lower than that (77.8%) of the substrate pretreated at 170 °C. The reason for this decrease is probably ascribed to the existence of more pseudo-lignin or condensed lignin on the surface the substrates under harsh condition, as supported by the SEM pictures and 2D-HSQC spectra (Figs. S4 and S5), which restricts the final enzymatic conversion of cellulose into glucose (Kumar et al., 2013). In short, AlCl3 pretreatment significantly removed most of hemicelluloses at 160–180 °C, resulting in the increase of surface area of the substrates and thus providing high accessibility for enzyme, as evidenced by the compositional analysis, the FT-IR, and CP/MAS 13C-NMR spectral analyses, while the hemicelluloses were mostly removed during the single hydrothermal pretreatment performed on the Eucalyptus wood at 200 °C (Gütsch et al., 2012).

Based on the data presented in Fig. 2 and S4, the SEM images demonstrated that the pretreatment significantly increase the exposure of the internal structure of the plant cell, leading to enhance of the surface area and pore volume of the solid residues, which improved enzyme access to cellulose (Zhu et al., 2009). As shown in Table 1, over 99% hemicelluloses were removed, therefore, this pretreatment was a promising approach for the effective degradation and removal of hemicelluloses to increase the enzymatic digestibility of the substrates. These results are in good agreement with the previous reports on the evaluation of the process to reduce biomass recalcitrance using different physicochemical pretreatments for enhanced enzymatic digestibility, such as liquid hot water (Ko et al., 2015), steam explosion and alkaline pretreatment (Sun et al., 2014). Based on the above data, it can be concluded that the formation of lignin-metal complex and the removal of recalcitrance barriers (removing hemicelluloses, cleaving lignin and lignin-carbohydrates complex) finally led to increase of the surface area of the substrates, which improved the access of cellulase to the ‘‘open’’ cellulose structure.

4. Conclusions The AlCl3-catalyzed hydrothermal pretreatment enhanced the enzymatic hydrolysis of E. camaldulensis by removing its chemical and physical barriers and changing the enzyme accessibility. After the pretreatment, cellulose bundles were broken-up and disassemble, which was favorable for enzymatic hydrolysis. Meanwhile, this pretreatment resulted in the cleavage of different linkages in lignin (e.g., b-O-4, b-b, and b-5). The maximum yield of glucose (77.8%) was achieved at 170 °C for 1 h, which was 7.3 times higher than that of the raw material and indicated that the pretreatment remarkably enhanced the enzymatic hydrolysis efficiency and facilitated the collection of the residual lignin.

Acknowledgements This work was supported by the grants from the ‘National Natural Science Foundation of China’ – ‘China’ (31430092, 31500486, and 31110103902), ‘International Science & Technology Cooperation Program of China’ – ‘China’ (2015DFG31860) and ‘Fundamental Research Funds for the Central Universities’ – ‘China’ (BLX2014-37).

64

X.-J. Shen et al. / Bioresource Technology 206 (2016) 57–64

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biortech.2016.01. 031.

References Alvira, P., Tomas-Pejo, E., Ballesteros, M., Negro, M.J., 2010. Pretreatment technologies for an efficient bioethanol production process based on enzymatic hydrolysis: a review. Bioresour. Technol. 101, 4851–4861. Cai, C.M., Nagane, N., Kumar, R., Wyman, C.E., 2014. Coupling of metal halides with a co-solvent to achieve co-production of furfural and hmf from lignocellulosic biomass. Abstr. Pap. Am. Chem. Soc. 247, 1. Duarte, C.L., Ribeiro, M.A., Oikawa, H., Mori, M.N., Napolitano, C.M., Galvão, C.A., 2012. Electron beam combined with hydrothermal treatment for enhancing the enzymatic convertibility of sugarcane bagasse. Radiat. Phys. Chem. 81, 1008– 1011. Garrote, G., Domínguez, H., Parajó, J.C., 1999. Mild autohydrolysis: an environmentally friendly technology for xylooligosaccharide production from wood. J. Chem. Technol. Biotechnol. 74, 1101–1109. Gütsch, J.S., Nousiainen, T., Sixta, H., 2012. Comparative evaluation of autohydrolysis and acid-catalyzed hydrolysis of Eucalyptus globulus wood. Bioresour. Technol. 109, 77–85. Isroi, Millati, R., Syamsiah, S., Niklasson, C., Cahyanto, M.N., Lundquist, K., Taherzadeh, M.J., 2011. Biological pretreatment of lignocelluloses with whiterot fungi and its applications: a review. Bioresources 6, 5224–5259. Jonsson, L.J., Alriksson, B., Nilvebrant, N.O., 2013. Bioconversion of lignocellulose: inhibitors and detoxification. Biotechnol. Biofuels 6. Kamireddy, S.R., Li, J., Tucker, M., Degenstein, J., Ji, Y., 2013. Effects and mechanism of metal chloride salts on pretreatment and enzymatic digestibility of corn stover. Ind. Eng. Chem. Res. 52, 1775–1782. Ko, J.K., Kim, Y., Ximenes, E., Ladisch, M.R., 2015. Effect of liquid hot water pretreatment severity on properties of hardwood lignin and enzymatic hydrolysis of cellulose. Biotechnol. Bioeng. 112, 252–262. Kumar, L., Arantes, V., Chandra, R., Saddler, J., 2012. The lignin present in steam pretreated softwood binds enzymes and limits cellulose accessibility. Bioresour. Technol. 103, 201–208. Kumar, R., Hu, F., Sannigrahi, P., Jung, S., Ragauskas, A.J., Wyman, C.E., 2013. Carbohydrate derived-pseudo-lignin can retard cellulose biological conversion. Biotechnol. Biofuels 110, 737–753. Kumar, R., Wyman, C.E., 2013. Physical and chemical features of pretreated biomass that influence macro-/micro-accessibility and biological processing. In: Aqueous Pretreatment of Plant Biomass for Biological and Chemical Conversion to Fuels and Chemicals. John Wiley & Sons, Ltd., pp. 281–310. Kumar, S., Kothari, U., Kong, L., Lee, Y.Y., Gupta, R.B., 2011. Hydrothermal pretreatment of switchgrass and corn stover for production of ethanol and carbon microspheres. Biomass Bioenergy 35, 956–968. Liu, L., Sun, J., Cai, C., Wang, S., Pei, H., Zhang, J., 2009. Corn stover pretreatment by inorganic salts and its effects on hemicellulose and cellulose degradation. Bioresour. Technol. 100, 5865–5871. Lloyd, T.A., Wyman, C.E., 2005. Combined sugar yields for dilute sulfuric acid pretreatment of corn stover followed by enzymatic hydrolysis of the remaining solids. Bioresour. Technol. 96, 1967–1977. Ma, Y., Ji, W., Zhu, X., Tian, L., Wan, X., 2012. Effect of extremely low AlCl3 on hydrolysis of cellulose in high temperature liquid water. Biomass Bioenergy 39, 106–111.

Nitsos, C.K., Matis, K.A., Triantafyllidis, K.S., 2013. Optimization of hydrothermal pretreatment of lignocellulosic biomass in the bioethanol production process. ChemSusChem 6, 110–122. Redding, A.P., Wang, Z., Keshwani, D.R., Cheng, J.J., 2011. High temperature dilute acid pretreatment of coastal Bermuda grass for enzymatic hydrolysis. Bioresour. Technol. 102, 1415–1424. Rocha, G.J., Silva, V.F., Martín, C., Gonçalves, A.R., Nascimento, V.M., Souto-Maior, A. M., 2013. Effect of xylan and lignin removal by hydrothermal pretreatment on enzymatic conversion of sugarcane bagasse cellulose for second generation ethanol production. Sugar Tech. 15, 390–398. Sanderson, K., 2011. Lignocellulose: a chewy problem. Nature 474, S12–S14. Sannigrahi, P., Kim, D.H., Jung, S., Ragauskas, A., 2011. Pseudo-lignin and pretreatment chemistry. Energy Environ. Sci. 4, 1306–1310. Satari Baboukani, B., Vossoughi, M., Alemzadeh, I., 2012. Optimisation of dilute-acid pretreatment conditions for enhancement sugar recovery and enzymatic hydrolysis of wheat straw. Biosyst. Eng. 111, 166–174. Savage, P.E., 1999. Organic chemical reactions in supercritical water. Chem. Rev. 99, 603–622. Segal, L., Creely, J.J., Martin, A.E., Conrad, C.M., 1959. An empirical method for estimating the degree of crystallinity of native cellulose using the X-ray diffractometer. Text. Res. J. 29, 786–794. Sette, M., Wechselberger, R., Crestini, C., 2011. Elucidation of lignin structure by quantitative 2D NMR. Chem. Eur. J. 17, 9529–9535. Sitompul, J.P., Simangunsong, R.F., Asrizal, A.A., Alisyahbana, H., Lee, H.W., Rasrendra, C.B., 2014. Catalytic conversion of empty fruit bunch of palm oil for producing lactic acid. Proc. Chem. 9, 88–93. Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Templeton, D., 2008a. Determination of sugars, byproducts, and degradation products in liquid fraction process samples. NREL/TP-510-42623, Laboratory Analytical Procedure (LAPs). National Renewable Energy Laboratory, Golden CO. Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Templeton, D., Crocker, D., 2008b. Determination of structural carbohydrates and lignin in biomass. NREL/ TP-510-42618, Laboratory Analytical Procedure (LAPs). National Renewable Energy Laboratory, Golden CO. Stephanopoulos, G., 2007. Challenges in engineering microbes for biofuels production. Science 315, 801–804. Sun, S.L., Wen, J.L., Ma, M.G., Sun, R.C., 2014. Enhanced enzymatic digestibility of bamboo by a combined system of multiple steam explosion and alkaline treatments. Appl. Energy 136, 519–526. Trajano, H.L., Engle, N.L., Foston, M., Ragauskas, A.J., Tschaplinski, T.J., Wyman, C.E., 2013. The fate of lignin during hydrothermal pretreatment. Biotechnol. Biofuels, 6. Wen, J.L., Sun, S.L., Xue, B.L., Sun, R.C., 2013. Recent advances in characterization of lignin polymer by solution-state nuclear magnetic resonance (NMR) methodology. Materials 6, 359–391. Wen, J.L., Sun, S.L., Yuan, T.Q., Xu, F., Sun, R.C., 2014. Understanding the chemical and structural transformations of lignin macromolecule during torrefaction. Appl. Energy 121, 1–9. Wyman, C.E., Spindler, D.D., Grohmann, K., 1992. Simultaneous saccharification and fermentation of several lignocellulosic feedstocks to fuel ethanol. Biomass Bioenergy 3, 301–307. Yang, H., Wang, K., Song, X., Xu, F., Sun, R.C., 2012. Enhanced enzymatic hydrolysis of triploid poplar following stepwise acidic pretreatment and alkaline fractionation. Process Biochem. 47, 619–625. Yu, Q., Zhuang, X., Yuan, Z., Qi, W., Wang, Q., Tan, X., 2011. The effect of metal salts on the decomposition of sweet sorghum bagasse in flow-through liquid hot water. Bioresour. Technol. 102, 3445–3450. Zhu, J.Y., Wang, G.S., Pan, X.J., Gleisner, R., 2009. Specific surface to evaluate the efficiencies of milling and pretreatment of wood for enzymatic saccharification. Chem. Eng. Sci. 64, 474–485.