- Email: [email protected]
Isolation and characterization of hemicelluloses extracted by hydrothermal pretreatment
Bioresource Technology 114 (2012) 677–683
Contents lists available at SciVerse ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Isolation and characterization of hemicelluloses extracted by hydrothermal pretreatment Ming-Guo Ma a,⇑, Ning Jia a, Jie-Fang Zhu b, Shu-Ming Li a, Feng Peng a, Run-Cang Sun a,c a
Institute of Biomass Chemistry and Technology, College of Materials Science and Technology, Beijing Forestry University, Beijing 100083, PR China Department of Chemistry, the Ångström Laboratory, Uppsala University, Uppsala 75121, Sweden c State Key Laboratory of Pulp and Paper Engineering, South China University of Technology, Guangzhou 510640, PR China b
a r t i c l e
i n f o
Article history: Received 8 January 2012 Received in revised form 14 March 2012 Accepted 16 March 2012 Available online 23 March 2012 Keywords: Hemicelluloses Pretreatment Hydrothermal Isolation Organic alkali (DMF)
a b s t r a c t The dewaxed sample from Triploid of Populus tomentosa Carr. was extracted by using organic alkaline solvent (Dimethylformamide, DMF) via hydrothermal pretreatment. Neutral sugar compositions and molecular weight analysis demonstrated that the hemicellulosic fractions with a higher Uro/Xyl ratio, namely the more branched hemicelluloses, had higher molecular weights. Interestingly, these results were different from the previous report, in which the ratio of Uro/Xyl in the water-soluble hemicellulosic fraction was more than that of the alkali-soluble hemicellulosic fraction. Spectroscopy (FTIR, 1H NMR, 13C NMR, and HSQC) analysis indicated that the hemicellulosic fractions were mainly composed of (1 ? 4)-linked a-D-glucan from starch and (1 ? 4)-linked b-D-xylan attached with minor amounts of branched sugars from hemicelluloses. In addition, thermal analysis implied that linear hemicelluloses showed more thermal stability than the branched ones during pyrolysis. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction Lignocelluloses has been receiving considerable attention as low cost and renewable feedstocks for producing biofuels, biobased chemicals, and high value-added biomaterials to meet global energy and chemical needs (Himmel et al., 2007). The major components of lignocelluloses are the polysaccharides (cellulose and hemicelluloses), and lignin. It is well known that lignin, as a network polymer, binds with hemicelluloses and cellulose to form a tight compact structure (Adler, 1977). It makes the components of lignocelluloses difficult to be separated and thus restrict its applications. Therefore, development of pretreatment methods for the extraction and isolation of lignocelluloses is of great importance for broadening its industrial applications. Until now, some successful pretreatment methods such as alkali extraction (Sun et al., 2000a), steam explosion (Biermann et al., 1984), hot water (Leppänen et al., 2011), enzymatic hydrolysis (Nhuan et al., 2011), bioconversion method (Himmel et al., 2007), and the hydrothermal method (Feria et al., 2012; Pin´kowska et al., 2011; Pronyk and Mazza, 2012) were employed for the extraction and isolation of lignocelluloses. Among of these pretreatment methods, the alkali extraction included the concentrated solutions of sodium or potassium hydroxide, and alkaline hydrogen peroxide solution. The main advantages of the alkali extraction are easy to act and cost-effective. Alkaline peroxide ⇑ Corresponding author. Tel.: +86 10 62336592; fax: +86 10 62336972. E-mail address: [email protected] (M.-G. Ma). 0960-8524/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biortech.2012.03.048
treatment could separate cellulose, hemicelluloses, and lignin, and achieve complete utilization of lignocelluloses without the negative impact on the environment (Sun et al., 2000b). In addition, the hemicellulosic fractions obtained from alkali by ultrasound-assisted extraction showed a relatively lower content of associated lignin, a higher molecular weight, and a slightly higher thermal stability than those obtained without ultrasonic irradiation (Sun and Tomkinson, 2002). The steam explosion is another potential method to isolate of hemicelluloses from cereal straw and wood samples (Glasser et al., 2000). Higher yields of hemicelluloses can be achieved at shorter extraction times and lower temperatures. However, compared with the conventional method, the hydrothermal method is a promising technology for the preparation of lignocelluloses due to its advantages of short reaction time, high percent conversion, and relatively low reaction temperature. In recent years, rapid progress has been made in the hydrothermal pretreatment for lignocelluloses (Feria et al., 2012; Goto et al., 2004; López et al., 2004; Makishima et al., 2006; Pin´kowska et al., 2011; Pronyk and Mazza, 2012; Sasaki et al., 2003). For example, hydrothermal pretreatment was usually operated by exposing lignocelluloses to a chemical at certain temperature for a period of time depending on the physiochemical structure of the biomass (López et al., 2004). In addition, hydrothermal pretreatment to produce a series of xylooligosaccharides had been reported as an effective fractionation method for the major components (cellulose, hemicelluloses, and lignin) of lignocelluloses (Goto et al., 2004; Sasaki et al., 2003). Makishima et al. (2006)
678
M.-G. Ma et al. / Bioresource Technology 114 (2012) 677–683
reported that the waste medium for Enokitake (Flammulina velutipes) mushroom cultivation mainly consisting of corncob was experimented by hydrothermal method to recover soluble sugars. Feria et al. (2012) investigated the integral valorization of Leucaena diversifolia by hydrothermal treatment at 160–180 °C for 0– 30 min followed by ethanol–soda–anthraquinone delignification. Pronyk and Mazza (2012) reported the fractionation of triticale, wheat, barley, oats, canola, and mustard straws for the production of carbohydrates and lignins by hydrothermal treatment at 165 °C for 20–30 min. Hemicelluloses are complex constituents in the plant cell wall. Unlike cellulose, hemicelluloses are not single homogeneous but polymers consist of various different sugar units, arranged in different proportions and with different substituents (Aspinall and Mahomed, 1954). Moreover, hemicelluloses are closely associated with cellulose by hydrogen bonds and are linked to lignin by covalent bonds (mainly a-benzyl ether linkages) (Freudenberg, 1965). Hemicelluloses are useful as gels, films, coatings, adhesives, gelling, stabilizing and viscosity-enhancing additives in food and pharmacy (Ebringerová, 2005). Krishna et al. (2000) reported the bioconversion of hemicelluloses into ethanol by fermentation. Therefore, understanding the chemical compositions and structural characteristics of hemicelluloses is useful to improve its industrial applications. Triploid of Populus tomentosa Carr., a kind of fast-growing poplar widely planted in China for preventing wind erosion and control desertification, has a considerable economical and ecological importance. The aim of this study was to investigate the influences of organic alkaline solvent (Dimethylformamide, DMF) and hydrothermal pretreatment on the hemicellulosic fractions from Triploid of Populus tomentosa Carr. The structural characteristics and physicochemical and thermal properties of the hemicellulosic fractions were investigated.
2. Methods 2.1. Materials Chips of Triploid of Populus tomentosa Carr., 4 years old, were obtained from Shandong province, China. They were ground to pass through a 20 mesh sieve. After being dried at 60 °C for 16 h, the poplar powder was dewaxed with toluene/ethanol (2:1, v/v) in a Soxhlet apparatus for 6 h. The dewaxed sample was further dried in a cabinet oven with air circulation at 60 °C for 16 h and stored. All standard chemicals used were of analytical grade and purchased from Sigma Chemical Company (Beijing, China).
2.2. Isolation of Hemicelluloses In order to study structural differences of the hemicelluloses present in Populus tomentosa Carr., hemicellulosic fractions were isolated by sequential extractions according to the scheme in Fig. 1. The dewaxed sample was extracted by 0%, 10%, 30%, 50%, 70%, and 100% DMF aqueous solution with a solid to liquid ratio of 1:20 (g mL 1), respectively. The mixture was transferred into a 50-mL Teflon-lined stainless steel autoclave and maintained at 160 °C for 24 h. After the treatment period, the insoluble residues were separated from the solution by filtration, washed with distilled water until the pH of filtrates was neutral, and then dried at 60 °C for 12 h. The filtrates were adjusted to pH 6.0 with 6.0 M HCl, and then concentrated under reduced pressure and precipitated in three volumes of ethanol. After filtration, the precipitated fractions were washed with 70% ethanol and freeze-dried. It should be noted that the six hemicellulosic fractions (H0, H10, H30, H50, H70, H100) were obtained by precipitation in three volumes of ethanol from the 0%, 10%, 30%, 50%, 70%, and 100% DMF-soluble hemicelluloses, respectively.
Populus tomentosa Carr. Dried 60 °C
16 h.
Dried sample Dewaxed with toluene/ethanol (2:1, v/v) for 6 h. Dewaxed sample Extracted by 0%, 10%, 30%, 50%, 70%, and 100% DMF solution with a solid to liquid ratio of 1:20 (g mL-1), and then transferred into a 50-mL Teflon-lined stainless steel autoclave.at 160 °C for 24 h. Residue(Cellulose)
Filtrate (Discarded)
Filtrate Neutralized to pH 6.0 with 6.0 M HCl, concentrated under reduced pressure, and then precipitated in 3 volumes of ethanol. Pellet Washed with 70% ethanol and freeze-dried. Hemicelluosic fractions (H0, H10, H30, H50, H70, and H100)
Fig. 1. Scheme for isolation of hemicelluloses from Populus tomentosa Carr.
679
M.-G. Ma et al. / Bioresource Technology 114 (2012) 677–683
2.3. Chemical characterization
3. Results and discussion
The neutral sugars and uronic acids in the hemicellulosic fractions were obtained by hydrolysis with 1.0 M H2SO4 at 105 °C for 2.5 h. After hydrolysis, the samples were filtered and diluted 50-fold, and analyzed by high-performance anion-exchange chromatography (HPAEC) using a Dionex ICS3000 gradient pump, amperometric detector, AS50 autosampler and a Carbopac™ PA-20 column (4 250 mm, Dionex) and a guard PA-20 column (3 30 mm, Dionex). Neutral sugars and uronic acids were separated in 5 mM NaOH isocratic eluent (carbonate free and purged with nitrogen) for 20 min, followed by a 0–75 mM NaAc gradient in 5 mM NaOH for 15 min. Then the columns were washed with 200 mM NaOH to remove carbonate for 10 min, and followed a 5 min elution with 5 mM NaOH to re-equilibrate the column before the next injection. The total analysis time was 50 min, and the flow rate was 0.4 mL/min. Calibration was performed with standard solutions of L-arabinose, D-glucose, D-xylose, D-glucose, D-mannose, D-galactose, glucuronic acid, and galacturonic acids. The molecular weights of the hemicellulosic fractions were estimated by gel permeation chromatography (GPC) on a PL aquagelOH 50 column (300 7.7 mm, polymer laboratories Ltd.), calibrated with PL pullulan polysaccharide standards (peak average molecular weights of 783, 12,200, 100,000, and 1600,000 g mol 1, Polymer Laboratories Ltd.). A flow rate of 0.5 mL min 1 was maintained. The eluents were 0.02 M NaCl in 0.005 M sodium phosphate buffer, at pH 7.5. Detection was achieved with a Knauer differential refractometer. The column oven was kept at 30 °C. Hemicelluloses were dissolved in 0.005 M sodium phosphate buffer with 0.02 M NaCl, pH 7.5, at a concentration of 0.1% before measurement.
3.1. Content of neutral sugars and uronic acids
2.4. Spectroscopic and thermal characterization Fourier transform infrared spectra (FTIR) of hemicellulosic samples were recorded from an FT-IR spectrophotometer (Tensor 27, Brucker, Germany) using the KBr disk method. Thirty-two scans were taken of each sample recorded from the range 4000– 400 cm 1 at a resolution of 2 cm 1 in the transmission mode. The solution-state 1H NMR and 13C NMR spectra were recorded on a Bruker AVIII 400 MHz spectrometer using 20 mg for 1H and 80 mg for 13C of hemicellulose sample in 1.0 mL D2O. Additionally, to increase solubility of alkali-extractable hemicelluloses H5 in D2O, a few drops of sodium deuteroxide (7.5 M NaOD) were added. The chemical shifts were calibrated relative to the signals from D2O, used as an internal standard, at 4.70 ppm for the 1H NMR spectrum. The acquisition time (AQ) was 3.9 s, and the relaxation time was 1.0 s. The 13C NMR spectrum was recorded at 25 °C after 30,000 scans. A 30° pulse flipping angle, a 9.2 ls pulse width, a 1.89 s delay time and a 1.36 s AQ between scans were used. Heteronuclear Single Quantum Coherence (HSQC) experiment was also obtained on a Bruker AVIII 400 MHz spectrometer after 128 scans with 20 mg sample dissolved in 1.0 mL D2O. The spectral widths for HSQC were 5000 and 20,000 Hz for the 1H and 13C dimensions, respectively. The number of collected complex points was 1024 for the 1H dimension with a cycle delay of 1.5 s. The number of scans was 128, and 256 time increments were always recorded in 13C dimension. The 1JC–H used was 146 Hz. Prior to Fourier transformation, the data matrices were zero filled up to 1024 points in the 13C dimension. Thermal behavior of the hemicellulosic samples was performed using thermogravimetric analysis (TGA) and differential thermal analysis (DTA) on a simultaneous thermal analyzer (DTG-60, Shimadzu, Japan). The samples weighed between 8 and 13 mg and were heated from room temperature to 600 °C at a heating rate of 10 °C min 1 in flowing nitrogen.
To characterize the solubilized hemicelluloses, the six hemicellulosic fractions were performed for the determination of their constituent sugars. The results about the composition of neutral sugars and content of uronic acids in the six hemicellulosic fractions are listed in Table 1. As shown, the hot water-soluble hemicellulosic fraction (H0) contained a major amount of xylose (52.9%) and glucose (40.7%). Galactose (3.2%), arabinose (1.4%), and rhamnose (1.4%) were present in smaller amounts together with traces of uronic acids (0.3%), mainly originated from glucuronic acids (GlcpA) or 4-O-methyl-glucuronic acids (4-O-Me-a-DGlcpA) as side chains. This high percentage of xylose and glucose was taken to demonstrate correspondingly more xylans and glucans. The presence of hemicellulosic and pectic polysaccharides in hot water-soluble fraction from agricultural residues has been widely demonstrated. For example, Lawther et al. (1995) reported that wheat straw yields of 4.6% in hot water-soluble polysaccharides, in which xylose and galactose were the major sugar constituents. Xu et al. (2007) indicated that the hot water treatment probably released some amounts of pectic polysaccharides exception for noticeable amounts of b-glucan. Glucose (63.4%), galactose (19.3%), arabinose (8.2%), and xylose (7.0%) are the major sugar compositions of the alkali-soluble hemicellulosic fraction (H100). The value of xylose in H100 decreased from 52.9% to 7.0%, compared with that of the hot water-soluble hemicellulosic fraction (H0). The lower value of xylose was different from the previous reports (Xu et al., 2007). Moreover, the amount of galactose (19.3%) and uronic acids (1.6%) in H100 was greater than that of H0 (3.2% and 0.3%). The alkali-soluble hemicellulosic fraction in the literatures was obtained in the existence of water. However, the hemicellulosic fraction (H100) was produced in pure DMF solution. It can be concluded that pure DMF solution can probably solubilize arabinan and galactan. The intrinsic and detailed formation mechanism needs to be further explored. The further four hemicellulosic fractions (H10–H70) were obtained in the mixed solvents of water and DMF using 10%, 30%, 50%, and 70% DMF solution, respectively. Clearly, glucose (53.4–63.0%) and xylose (33.2–39.5%) were the predominant sugar components. This result demonstrated that the hydrothermal pretreatment favored the extraction of xylose with high percentage. Xylose content (H10–H70) was different from that in H100, which was due to the absence of water. In the literature, more than 90% of the xylan and nearly 50% of the lignin were extracted in a flow-through reactor at 165 °C (Pronyk and Mazza, 2012). In our study, the hemicelluloses were obtained by hydrothermal pretreatment at 160 °C for 24 h. The Teflon-lined stainless steel autoclave was used in the hydrothermal pretreatment, which was different from those (a flow-through reactor) in the literatures (Feria et al., 2012; Pronyk and Mazza, 2012). Galactose (1.3–3.9%), arabinose
Table 1 Neutral sugars and uronic acids (relative percent hemicellulosic sample, w/w) and Uro/Xyl ratios of hemicellulosic fractions. Sugars (%)
Rhamnose Arabinose Galactose Glucose Xylose Uronic acids Uro/Xyl
Hemicellulosic fractions H0
H10
H30
H50
H70
H100
1.42 1.40 3.22 40.74 52.91 0.29 0.01
0.06 0.20 2.13 63.04 33.20 1.36 0.04
0.03 0.52 2.43 55.38 39.48 2.15 0.05
0.34 0.62 1.29 58.39 35.89 3.46 0.10
0.24 5.12 3.86 53.41 34.55 2.76 0.08
0.49 8.19 19.26 63.40 7.00 1.58 0.23
680
M.-G. Ma et al. / Bioresource Technology 114 (2012) 677–683
(0.2–5.1%), and uronic acids (1.4–2.8%) were present in substantial amounts, and rhamnose (0.03–0.24%) was observed as small sugar constituents. Evidently, glucose content was increased and xylose content was decreased, compared with that of the hot water-soluble hemicellulosic fraction (H0), indicating that xylans is the predominant hemicelluloses in the cell walls of poplar wood (Sun et al., 2001). The predominant components of xylose and the substantial amount of uronic acids, mainly originated from glucuronic acids (GlcpA) or 4-O-methyl-glucuronic acids (4-O-Me-a-D-GlcpA) as side chains, demonstrated that the hemicelluloses with the mixed solvents of water and DMF pretreatment probably consisted mainly of glucuronoxylans. Furthermore, the four hemicellulosic fractions were also rich in glucose. The existence of glucose probably arose from b-glucans and xyloglucans under the hydrothermal treatment through organic alkaline solvent. For example, Wilkie et al. (1977), who supported the existence of b-glucans, stated that the hemicelluloses from delignified bamboo contained xylans and glucose-rich fractions. In addition, xyloglucan was a hemicellulosic polysaccharide in all higher plants, where they represent a quantitatively major building material of the primary cell wall (Ebringerová et al., 2005). The relatively high quantity of glucose and xylose implied that the hemicellulosic fractions were mixed polysaccharides of glucan and xylan. The increased glucose was also potentially considered as the direct consequence of the favorable effect of the hydrothermal pretreatment through organic alkaline solvent. However, further evidence to approve of this supposition needs additional experiments including the effect of the types of materials and alkali on the sugar composition of hemicelluloses. In general, the Uro/Xyl ratios were indicative of the degree of linearity or branching of hemicelluloses. It should be noted that the ratio of Uro/Xyl (0.04–0.10) in the hemicellulosic fractions (H10–H70) is more than that the value (0.01) in the water-soluble hemicellulosic fraction (H0), but lower than the value (0.23) in the alkali-soluble hemicellulosic fraction (H100), revealing that the H10–H70 had more branched chains than H0 and less branched chains than H100. From the ratio of uronic acids to xylose (0.01– 0.23) in the six hemicellulosic fractions, it can be indicated that the H0 seemed to be more linear and had less uronic acids than those of the H10–H100. These results are different from the previous report (Xu et al., 2007), in which the ratio of Uro/Xyl in the watersoluble hemicellulosic fraction was more than those of the alkalisoluble hemicellulosic fractions. This phenomenon suggested that the hydrothermal pretreatment method at high temperature for a longer period may result in the difference between the water-soluble hemicellulosic fraction and the alkali-soluble hemicellulosic fraction. In addition, the value of Uro/Xyl in H10–H70 increased from 0.04 to 0.05 to 0.10 firstly, and then decreased from 0.10 to 0.08 with the increasing ratio of DMF/water. It can be concluded that the ratio of DMF/water play an important role in the sugar compositions and structure of hemicellulosic fractions and the appropriate ratio of DMF to water is important for the pretreatment of lignocelluloses. 3.2. Molecular weight distribution In this study, the molecular weights of the hemicellulosic fractions were determined by gel permeation chromatography (GPC). As reported, GPC is effective in estimating the molecular weight of unknown polymers of similar or identical chemical structures to those used to calibrate columns (Himmel et al., 1989). The values of the weight-average (Mw) and number-average (Mn) molecular weights as well as polydispersity (Mw/Mn) are given in Table 2. Obviously, the water-soluble hemicellulosic fraction showed a relatively lower degree of polymerization with Mw values of 3960 g mol 1 than the alkali-soluble hemicellulosic fraction (H100) with Mw of 5600 g mol 1. This implied that the pretreatment
Table 2 Weight-average (Mw) and number-average (Mn) molecular weights and polydispersity (Mw/Mn) of the hemicellulosic fractions. Hemicellulosic preparations
Mw Mn Mw/Mn
H0
H10
H30
H50
H70
H100
3960 1690 2.34
3330 1370 2.43
4690 1520 3.08
24,840 2310 10.75
17,060 1730 9.86
5600 3340 1.68
with hot water released only hemicellulosic polymers with low molecular weights and a slight degradation of macromolecules structure of hemicelluloses probably occurred. Additionally, the experimental results indicated that the extraction with an increasing DMF concentration from 10% to 50% led to grow of Mw from 3330 to 24,840 g mol-1, suggesting that increasing DMF concentration (from 10% to 50%) favored the dissolution of hemicelluloses with large molecular size. Conversely, as the DMF concentration was further increased from 50% to 70%, the Mw value decreased from 24,840 to 17,060 g mol 1. Clearly, the hemicellulosic fractions (H50 and H70), obtained from 50% and 70% DMF concentrations, showed a very much higher degree of polymerization with Mw values between 24,840 and 17,060 g mol 1 than the other hemicellulosic fractions (H0, H10, H30, and H100) with Mw from 3330 to 5600 g mol 1. These results implied that the hemicellulosic fractions (H50 and H70) using relatively high DMF concentration can result in the dissolution of large hemicellulosic molecules from the materials. Furthermore, on the basis of the Uro/Xyl ratio and molecular weights results, it may be concluded that the hemicellulosic fractions with a higher Uro/ Xyl ratio, namely the more branched hemicelluloses, had higher molecular weights, except that the hemicellulosic fraction (H100) with pure DMF pretreatment. These results were consistent with the previous reports. Hoffmann et al. (1991) indicated the highly branched xylan fractions to be of higher molecular weight than their less branched fractions. The hemicellulosic fraction (H100) with a higher Uro/Xyl ratio had lower molecular weights, which might due to the differences in the techniques of hydrothermal pretreatment and pure DMF solvent condition. These results also indicated that the addition of DMF favored the extraction of hemicellulosic fractions with high Mw molecular weights and the appropriate DMF concentration was important for the obtained hemicellulosic fraction with appropriate Mw molecular weights. The molecular weight distribution curves of the six hemicellulosic fractions are also shown in Fig. 2, and the results of the polydispersity were calculated by Mw/Mn. As can be seen from Fig. 2, the polydispersity of the water-soluble hemicellulosic fraction (H0) and the alkali-soluble hemicellulosic fraction (H100) were 2.34 and 1.68, respectively, which declared a narrow distribution of molecular size. A much higher polydispersity (2.43–10.75) was observed in four hemicellulosic fractions (H10–H70), compared with the H0 (2.34) and H100 (1.68), which showed a broad distribution of molecular weights, implying that the polymers with the mixed solvents of DMF and water generally had a wider range of molecular weights than those of the water-soluble fraction and alkali-soluble fraction. The polydispersity was consistent with the Mw results. Xu et al. (2007) also reported the similar results that the polydispersity of hot water-soluble hemicellulosic fraction was lower than that of alkali-soluble hemicellulosic fraction.
3.3. FTIR spectra Infrared spectroscopy (FTIR), which is quite extensively applied in plant cell wall analysis, has been proved to be useful for studying functional groups of carbohydrates (Kacˇuráková and Wilson,
M.-G. Ma et al. / Bioresource Technology 114 (2012) 677–683
681
Fig. 2. Molecular weight distribution curves of six hemicellulosic fractions.
2001). In this study, FTIR spectroscopy was used to evaluate the structural features among the six hemicellulosic fractions. FTIR spectra of water-soluble hemicellulosic fraction (H0) and alkali-soluble hemicellulosic fraction (H10–H100) are shown in Fig. S1 (see Fig. S1 in Supplementary data), respectively. The signal at 3408 cm 1 is assigned to the OH stretching vibrations of the hemicelluloses, and two bands at 2923 and 2855 cm 1 are assigned to C–H stretching. In addition, the intense band at 1621 cm 1 was due to the bending mode of the absorbed water, since the hemicelluloses usually have a strong affinity for water, and in the solid state these macromolecules may have disordered structures, which can easily be hydrated (Kauráková et al., 1998). A prominent absorbance at 1043 cm 1 was attributed to the C–O–C stretching of glycosidic linkages in xylans, indicating a dominant xylan of the fractionated hemicelluloses (Kacurakova et al., 1994). The small band at 892 cm 1, which arose from the C1 group frequency, was indicative of b-glycosidic linkage between the sugar units (Gupta et al., 1987). FTIR spectrum of alkali-soluble hemicellulosic fraction (H100) has slight difference, compared with the hot water-soluble fraction (H0). The presence of a small signal at 1736 cm 1 is as-
signed to the acetyl, uronic, and ferulic ester groups of the polysaccharides. The signal at 1736 cm 1 in the spectra of the hot watersoluble fraction (H0) disappeared, indicating that the acetyl group converted into acetic acid via hydrothermal pretreatment. Interestingly, the presence of this signal at 1736 cm 1 in the spectra of the alkali-soluble hemicellulosic fraction (H100) likely indicated that the pure organic alkali solution didnot include the groups, which can cleave this ester bond from hemicelluloses (see Fig. S1f in Supplementary data). The peak intensity at 1621 cm 1 decreased from Fig. S1a to Fig. S1f. The strong band at 1464 cm 1 was clearly observed. FTIR spectra of the four hemicellulosic fractions in the mixed solvents of DMF to water (H10–H70) are illustrated in Fig. S1b–e (see Fig. S1b–e in Supplementary data). As expected, the four hemicellulosic fractions showed very similar spectra, indicating a similar structure of the four hemicellulosic fractions. However, compared with H0 and H100, there are some difference in the peaks and peak intensities. The peak intensity of hemicellulosic fractions in the mixed solvents of DMF to water (H10–H70) was stronger than that of hot water-soluble hemicellulosic fraction (H0). The other
682
M.-G. Ma et al. / Bioresource Technology 114 (2012) 677–683
bands at 1464 and 1366 cm 1 were associated with C–H, OH or CH2 bending vibrations for the typical hemicelluloses. The signal at 1736 cm 1 disappears, verifying that the mixed solvents of DMF to water pretreatment completely cleaved this ester bond from hemicelluloses. As shown, the peak intensity of the H10 at 1043 cm 1 was lower than those of the other hemicellulosic fractions, indicating that the H10 released few amounts of xylans in accordance with the result of the previous sugar analysis. A small band at 1157 cm 1 was assigned to C–O and C–O–C stretching with some contribution of OH bending in arabinoxylans. One can clearly observe that the peak intensity increased with the increasing DMF concentration, suggesting that increasing alkali concentration probably increased the amounts of arabinoxylans. In addition, the shoulder peaks at 938 and 840 cm 1, probably due to the signal overlapping, indicated the existence of a-glycosidic linkage (Huang and Zhang, 2009). 3.4. NMR spectra Nuclear magnetic resonance (NMR) spectroscopy has significant advances in understanding complicated structure. To further elucidate the structural features of the polymers, the hemicellulosic fraction (H70) with 70% DMF solution pretreatment was investigated using one-dimensional (1D 1H and 13C) and the two-dimensional (2D HSQC) NMR spectroscopy. NMR spectrum is supposed to assay and identify the polymer backbone and the type of sidechain branching along the backbone. The 1H and 13C NMR and 2D HSQC spectra of H70 are given in Figs. S2–S4, respectively (see Figs. S2–S4 in Supplementary data). The signals for 1H and 13C NMR were assigned on the basis of the HSQC spectrum and the assignment data of 1H and 13C NMR spectra are given in Table 3. As can be seen from Fig. S2 (see Fig. S2 in Supplementary data), the anomeric 1H NMR signals in the spectrum of H70 were found in the spectral region of 4.3–5.6 ppm (Chiarini et al., 2004). The relevant signals occurred in two regions, namely, the anomeric region (d 5.6–4.9 for a-anomers and d 4.9–4.3 for b-anomers) and the ring proton region (d 4.5–3.0). To confirm the structural features, the H70 was also characterized by 13C NMR, as shown in Fig. S3 (see Fig. S3 in Supplementary data). The 13C NMR spectrum of H70 showed five main signals at d 101.9 (C-1), 72.9 (C-2), 74.4 (C-3), 75.7 (C-4), and 63.0 ppm (C-5), corresponding to the (1 ? 4)-linked b-D-Xylp units. Other main signals at d 101.9 (C-1), 74.2 (C-2), 74.0 (C-3), 79.5 (C-4), 71.2 (C-5) and 60.4 ppm (C-6) are attributed to the (1 ? 4)-linked a-D-Glcp units. From Fig. S4 (see Fig. S4 in Supplementary data), the predominant five signals gave HSQC 13C/1H cross-peaks at 101.9/4.30, 72.9/3.16, 74.4/3.38, 75.8/ 3.64, and 63.0/3.95 + 3.26 ppm (Table 3), which are assigned to C-1, C-2, C3, C-4, and C-5 of the (1 ? 4)-linked b-D-xylan, respectively (Sun et al., 2001). In addition, the cross-peaks at 101.9/5.07, 72.4/3.42, 74.0/3.75, 79.7/3.34, 71.2/3.70, and 60.4/3.69 ppm correspond to C-1, C-2, C-3, C-4, C-5, and C-6 of the (1 ? 4)-linked a-D-glucan, i.e., amylose (Huang and Zhang, 2009; Lisboa et al., 2005). A strong
Table 3 1 H and 13C chemical shift (ppm) assignments for hemicellulosic fraction H70. Sugar residues
X
a
Gb a b c d
Chemical shift (H/C) 1
2
3
4
5axc
5eqd
4.30 101.9 5.07 101.9
3.16 72.9 3.42 72.4
3.38 74.4 3.75 74.0
3.64 75.8 3.34 79.7
3.26 63.0 3.70 71.2
3.95 63.0
X, (1 ? 4)-linked b-D-xylan. G, (1 ? 4)-linked a-D-glucan. ax, axial. eq, equatorial.
6
signal at 4.70 ppm is attributed to the residual solvent (HDO), while the strong signals at 36.4/2.21 and 170.8/8.39 ppm are assigned to the DMF solution. Thus, from the results of NMR and the previous sugar analysis, it can be concluded that the hemicellulosic fraction (H70) is mainly composed of (1 ? 4)-linked a-Dglucan and (1 ? 4)-linked b-D-xylan attached with minor amounts substituted sugars and glucuronic acids. 3.5. Thermal analysis Thermogravimetric analysis (TGA) and differential thermal analysis (DTA) were used to investigate the thermal stability and the physicochemical properties of the hemicelluloses fractions. Fig. S5 (see Fig. S5 in Supplementary data) shows the typical TGA and DTA curves of the hot water-soluble hemicellulosic fraction (H0) and the hemicellulosic fraction (H70) with 70% DMF solution pretreatment. As shown, the weight loss occurred at the beginning is attributed to the evaporation of water, indicating that the water probably adsorbed on the surface of the material. From the TGA curves, it can be seen that the weight losses of the H0 and H70 occurred mainly in the temperature range of 208–350 °C and 200–330 °C, respectively. At 10% weight loss, the decomposing temperature of the two hemicellulosic fractions (H0 and H70) occurred at 188 and 173 °C, respectively. The total weight losses of 47.7% and 83.8% were observed with increasing DMF concentration from 0% to 70%, indicating the H0 had a higher thermal stability than that of the H70. The thermal stability of the water-soluble hemicellulosic fraction appeared to be higher than that of the hemicellulosic fraction with organic alkali–water mixed solvents pretreatment, which corresponded to the increasing trends of their molecular weights from 3960 to 17,060 g mol 1 in Table 2, implying that linear hemicelluloses showed the characterization of more thermal stability than the branched ones during pyrolysis. Furthermore, DTA curves for the hemicellulosic fractions (H0 and H70) exhibit a single small endothermic and two exothermic peaks. The exothermic peaks, which represent heat released from the hemicellulosic fractions, were observed at 291 and 427 °C for H0 and at 313 and 499 °C for H70, respectively. In general, the DTA curves of hemicellulosic fraction showed two degradation peaks: the first peak was related to initial hemicellulose loss at lower temperature, and the second one related the organic material oxidation (Choi et al., 2009). 4. Conclusions In summary, the influences of DMF with different concentrations and hydrothermal pretreatment on structural and properties of the hemicelluloses were researched. The hydrothermal pretreatment favored the extraction of xylose with high percentage. The H10–H70 had more branched chains than H0 and less branched chains than H100. The addition of DMF favored the extraction of hemicelluloses with high molecular weights Mw. The hemicellulosic fractions (H70) were mainly composed of (1 ? 4)-linked a-Dglucan from amylose and (1 ? 4)-linked b-D-xylan from minor amounts of branched sugars. The values of molecular weights and branching of the hemicelluloses played an important role in their thermal stability. Acknowledgements
3.69 60.4
Financial support from National Natural Science Foundation of China (31070511), the Program for New Century Excellent Talents in University, and Major State Basic Research Development Program of China (973 Program) (No. 2010CB732204) is gratefully acknowledged.
M.-G. Ma et al. / Bioresource Technology 114 (2012) 677–683
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. 2012.03.048.
References Adler, E., 1977. Lignin chemistry: past, present and future. Wood Sci. Technol. 3, 169–218. Aspinall, G.O., Mahomed, R.S., 1954. The constitution of a wheat-straw xylan. J. Chem. Soc. 6, 1731–1734. Biermann, C.J., Schultz, T.P., McGinnis, C.D., 1984. Rapid steam hydrolysis/extraction of mixed hardwoods as biomass. J. Wood Chem. Technol. 4, 111–128. Chiarini, L., Cescutti, P., Drigo, L., Impallomeni, G., Herasimenka, Y., Bevivino, A., Dalmastri, C., Tabacchioni, S., Manno, G., Zanetti, F., Rizzo, R., 2004. Exopolysaccharides produced by Burkholderia cenocepacia recA lineages IIIA and IIIB. J. Cyst. Fibros. 3, 165–172. Choi, M.J., Soottitantawat, A., Nuchuchua, O., Min, S.G., Ruktanonchai, U., 2009. Physical and light oxidative properties of eugenol encapsulated by molecular inclusion and emulsion–diffusion method. Food Res. Int. 42, 148–156. Ebringerová, A., 2005. Structural diversity and application potential of hemicelluloses. Macromol. Symp. 232, 1–12. Ebringerová, A., Hromádková, Z., Heinze, T., 2005. Hemicellulose. Adv. Polym. Sci. 186, 1–67. Feria, M.J., Alfaro, A., López, F., Pérez, A., García, J.C., Rivera, A., 2012. Integral valorization of Leucaena diversifolia by hydrothermal and pulp processing. Bioresour. Technol. 103, 381–388. Freudenberg, K., 1965. Lignin: its constitution and formation from phydroxycinnamyl alcohols. Science 148, 595–600. Glasser, W.G., Kaar, W.E., Jain, R.K., Sealey, J., 2000. Isolation options for noncellulosic heteropolysaccharides (HetPS). Cellulose 7, 299–317. Goto, M., Obuchi, R., Hirose, T., Sakaki, T., Shibata, M., 2004. Hydrothermal conversion of municipal organic waste into resources. Bioresour. Technol. 93, 279–284. Gupta, S., Madan, R.N., Bansal, M.C., 1987. Chemical composition of Pinus caribaea hemicellulose. Tappi J. 70, 113–114. Himmel, M.E., Ding, S.Y., Johnson, D.K., Adney, W.S., Nimlos, M.R., Brady, J.W., Foust, T.D., 2007. Biomass recalcitrance: engineering plants and enzymes for biofuels production. Science 315, 804–807. Himmel, M.E., Tatsumoto, K., Oh, K.K., Grohmann, K., Johnson, D.K., Chum, H.L., 1989. Molecular weight distribution of aspen lignins estimated by universal calibration. In: Glasser, W.G., Sarkanen, S. (Eds.), Lignin Properties and Materials. ACS, Washington, DC, pp. 82–100. Hoffmann, R.A., Roza, M., Maat, J., Kamerling, J.P., Vliegenthart, J.F.G., 1991. Structural characteristics of the cold-water-soluble arabinoxylans from the white flour of the soft wheat variety Kadet. Carbohydr. Polym. 15, 415–430. Huang, Z.P., Zhang, L.N., 2009. Chemical structures of water-soluble polysaccharides from rhizoma Panacis japonici. Carbohydr. Res. 344, 1136–1140. Kacˇuráková, M., Belton, P.S., Wilson, R.H., Hirsch, J., Ebringerova, A., 1998. Hydration properties of xylan-type structures: an FTIR study of xylooligosaccharides. J. Sci. Food Agric. 77, 38–44.
683
Kacˇuráková, M., Ebringerova, A., Hirsch, J., Horomadkova, Z., 1994. Infrared study of arabinoxylans. J. Sci. Food Agric. 66, 423–427. Kacˇuráková, M., Wilson, R.H., 2001. Developments in mid-infrared FT-IR spectroscopy of selected carbohydrates. Carbohydr. Polym. 44, 291–303. Krishna, S.H., Chowdary, G.V., 2000. Optimization of simultaneous saccharification and fermentation for the production of ethanol from lignocellulosic biomass. J. Agric. Food Chem. 48, 1971–1976. Lawther, J.M., Sun, R.C., Banks, W.B., 1995. Extraction, fractionation, and characterization of structural polysaccharides from wheat straw. J. Agric. Food Chem. 43, 667–675. Leppänen, K., Spetz, P., Pranovich, A., Hartonen, K., Kitunen, V., Ilvesniemi, H., 2011. Pressurized hot water extraction of Norway spruce hemicelluloses using a flowthrough system. Wood Sci. Technol. 45, 223–236. Lisboa, S.A., Evtuguin, D.V., Neto, C.P., Goodfellow, B.J., 2005. Isolation and structural characterization of polysaccharides dissolved in Eucalyptus globulus kraft black liquors. Carbohydr. Polym. 60, 77–85. López, M.J., Nichols, N.N., Dien, B.S., Moreno, J., Bothast, R.J., 2004. Isolation of microorganisms for biological detoxification of lignocellulosic hydrolysates. Appl. Microbiol. Biotechnol. 64, 125–131. Makishima, S., Nozaki, K., Mizuno, M., Netsu, E., Shinji, K., Shibayama, T., Kanda, T., Amano, Y., 2006. Recovery of soluble sugars from waste medium for Enokitake (Flammulina velutipes) mushroom cultivation with hydrothermal reaction and enzyme digestion. J. Appl. Glycosci. 53, 261–266. Nhuan, P.N., Justin, M., David, B.J., Caye, D., 2011. Fractionation of corn fiber treated by soaking in aqueous ammonia (SAA) for isolation of hemicellulose B and production of C5 sugars by enzyme hydrolysis. Appl. Biochem. Biotechnol. 164, 1390–1404. Pin´kowska, H., Wolak, P., Złon´ciska, A., 2011. Hydrothermal decomposition of xylan as a model substance for plant biomass waste-Hydrothermolysis in subcritical water. Biomass Bioenergy 35, 3902–3912. Pronyk, C., Mazza, G., 2012. Fractionation of triticale, wheat, barley, oats, canola, and mustard straws for the production of carbohydrates and lignins. Bioresour. Technol. 106, 117–124. Sasaki, M., Adschiri, T., Arai, K., 2003. Fractionation of sugarcane bagasse by hydrothermal treatment. Bioresour. Technol. 86, 301–304. Sun, R.C., Fang, J.M., Tomkinson, J., Geng, Z.C., Liu, J.C., 2001. Fractional isolation, physico-chemical characterization and homogeneous esterification of hemicelluloses from fast-growing poplar wood. Carbohydr. Polym. 44, 29–39. Sun, R.C., Tomkinson, J., 2002. Comparative study of lignins isolated by alkali and ultrasound-assisted alkali extractions from wheat straw. Ultrason. Sonochem. 9, 85–93. Sun, R.C., Tomkinson, J., Geng, Z.C., Wang, N.J., 2000a. Comparative studies of hemicelluloses solubilized during the treatment of maize stems with peroxymonosulfuric acid, peroxyformic acid, peracetic acid, and hydrogen peroxide. Part 1. Yield and chemical characterization. Holzforschung 54, 349– 356. Sun, R.C., Tomkinson, J., Ma, B.L., Liang, S.F., 2000b. Comparative study of hemicelluloses from rice straw by alkali and hydrogen peroxide treatments. Carbohydr. Polym. 42, 111–122. Wilkie, K.C.B., Woo, S.L., 1977. A heteroxylan and hemicellulosic materials from bamboo leaves, and a reconsideration of the general nature of commonly occurring xylans and other hemicelluloses. Carbohydr. Res. 49, 145–162. Xu, F., Sun, J.X., Geng, Z.C., Liu, C.F., Ren, J.L., Sun, R.C., Fowler, P., Baird, M.S., 2007. Comparative study of water-soluble and alkali-soluble hemicelluloses from perennial ryegrass leaves (Lolium peree). Carbohydr. Polym. 67, 56–65.
Contents lists available at SciVerse ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Isolation and characterization of hemicelluloses extracted by hydrothermal pretreatment Ming-Guo Ma a,⇑, Ning Jia a, Jie-Fang Zhu b, Shu-Ming Li a, Feng Peng a, Run-Cang Sun a,c a
Institute of Biomass Chemistry and Technology, College of Materials Science and Technology, Beijing Forestry University, Beijing 100083, PR China Department of Chemistry, the Ångström Laboratory, Uppsala University, Uppsala 75121, Sweden c State Key Laboratory of Pulp and Paper Engineering, South China University of Technology, Guangzhou 510640, PR China b
a r t i c l e
i n f o
Article history: Received 8 January 2012 Received in revised form 14 March 2012 Accepted 16 March 2012 Available online 23 March 2012 Keywords: Hemicelluloses Pretreatment Hydrothermal Isolation Organic alkali (DMF)
a b s t r a c t The dewaxed sample from Triploid of Populus tomentosa Carr. was extracted by using organic alkaline solvent (Dimethylformamide, DMF) via hydrothermal pretreatment. Neutral sugar compositions and molecular weight analysis demonstrated that the hemicellulosic fractions with a higher Uro/Xyl ratio, namely the more branched hemicelluloses, had higher molecular weights. Interestingly, these results were different from the previous report, in which the ratio of Uro/Xyl in the water-soluble hemicellulosic fraction was more than that of the alkali-soluble hemicellulosic fraction. Spectroscopy (FTIR, 1H NMR, 13C NMR, and HSQC) analysis indicated that the hemicellulosic fractions were mainly composed of (1 ? 4)-linked a-D-glucan from starch and (1 ? 4)-linked b-D-xylan attached with minor amounts of branched sugars from hemicelluloses. In addition, thermal analysis implied that linear hemicelluloses showed more thermal stability than the branched ones during pyrolysis. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction Lignocelluloses has been receiving considerable attention as low cost and renewable feedstocks for producing biofuels, biobased chemicals, and high value-added biomaterials to meet global energy and chemical needs (Himmel et al., 2007). The major components of lignocelluloses are the polysaccharides (cellulose and hemicelluloses), and lignin. It is well known that lignin, as a network polymer, binds with hemicelluloses and cellulose to form a tight compact structure (Adler, 1977). It makes the components of lignocelluloses difficult to be separated and thus restrict its applications. Therefore, development of pretreatment methods for the extraction and isolation of lignocelluloses is of great importance for broadening its industrial applications. Until now, some successful pretreatment methods such as alkali extraction (Sun et al., 2000a), steam explosion (Biermann et al., 1984), hot water (Leppänen et al., 2011), enzymatic hydrolysis (Nhuan et al., 2011), bioconversion method (Himmel et al., 2007), and the hydrothermal method (Feria et al., 2012; Pin´kowska et al., 2011; Pronyk and Mazza, 2012) were employed for the extraction and isolation of lignocelluloses. Among of these pretreatment methods, the alkali extraction included the concentrated solutions of sodium or potassium hydroxide, and alkaline hydrogen peroxide solution. The main advantages of the alkali extraction are easy to act and cost-effective. Alkaline peroxide ⇑ Corresponding author. Tel.: +86 10 62336592; fax: +86 10 62336972. E-mail address: [email protected] (M.-G. Ma). 0960-8524/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biortech.2012.03.048
treatment could separate cellulose, hemicelluloses, and lignin, and achieve complete utilization of lignocelluloses without the negative impact on the environment (Sun et al., 2000b). In addition, the hemicellulosic fractions obtained from alkali by ultrasound-assisted extraction showed a relatively lower content of associated lignin, a higher molecular weight, and a slightly higher thermal stability than those obtained without ultrasonic irradiation (Sun and Tomkinson, 2002). The steam explosion is another potential method to isolate of hemicelluloses from cereal straw and wood samples (Glasser et al., 2000). Higher yields of hemicelluloses can be achieved at shorter extraction times and lower temperatures. However, compared with the conventional method, the hydrothermal method is a promising technology for the preparation of lignocelluloses due to its advantages of short reaction time, high percent conversion, and relatively low reaction temperature. In recent years, rapid progress has been made in the hydrothermal pretreatment for lignocelluloses (Feria et al., 2012; Goto et al., 2004; López et al., 2004; Makishima et al., 2006; Pin´kowska et al., 2011; Pronyk and Mazza, 2012; Sasaki et al., 2003). For example, hydrothermal pretreatment was usually operated by exposing lignocelluloses to a chemical at certain temperature for a period of time depending on the physiochemical structure of the biomass (López et al., 2004). In addition, hydrothermal pretreatment to produce a series of xylooligosaccharides had been reported as an effective fractionation method for the major components (cellulose, hemicelluloses, and lignin) of lignocelluloses (Goto et al., 2004; Sasaki et al., 2003). Makishima et al. (2006)
678
M.-G. Ma et al. / Bioresource Technology 114 (2012) 677–683
reported that the waste medium for Enokitake (Flammulina velutipes) mushroom cultivation mainly consisting of corncob was experimented by hydrothermal method to recover soluble sugars. Feria et al. (2012) investigated the integral valorization of Leucaena diversifolia by hydrothermal treatment at 160–180 °C for 0– 30 min followed by ethanol–soda–anthraquinone delignification. Pronyk and Mazza (2012) reported the fractionation of triticale, wheat, barley, oats, canola, and mustard straws for the production of carbohydrates and lignins by hydrothermal treatment at 165 °C for 20–30 min. Hemicelluloses are complex constituents in the plant cell wall. Unlike cellulose, hemicelluloses are not single homogeneous but polymers consist of various different sugar units, arranged in different proportions and with different substituents (Aspinall and Mahomed, 1954). Moreover, hemicelluloses are closely associated with cellulose by hydrogen bonds and are linked to lignin by covalent bonds (mainly a-benzyl ether linkages) (Freudenberg, 1965). Hemicelluloses are useful as gels, films, coatings, adhesives, gelling, stabilizing and viscosity-enhancing additives in food and pharmacy (Ebringerová, 2005). Krishna et al. (2000) reported the bioconversion of hemicelluloses into ethanol by fermentation. Therefore, understanding the chemical compositions and structural characteristics of hemicelluloses is useful to improve its industrial applications. Triploid of Populus tomentosa Carr., a kind of fast-growing poplar widely planted in China for preventing wind erosion and control desertification, has a considerable economical and ecological importance. The aim of this study was to investigate the influences of organic alkaline solvent (Dimethylformamide, DMF) and hydrothermal pretreatment on the hemicellulosic fractions from Triploid of Populus tomentosa Carr. The structural characteristics and physicochemical and thermal properties of the hemicellulosic fractions were investigated.
2. Methods 2.1. Materials Chips of Triploid of Populus tomentosa Carr., 4 years old, were obtained from Shandong province, China. They were ground to pass through a 20 mesh sieve. After being dried at 60 °C for 16 h, the poplar powder was dewaxed with toluene/ethanol (2:1, v/v) in a Soxhlet apparatus for 6 h. The dewaxed sample was further dried in a cabinet oven with air circulation at 60 °C for 16 h and stored. All standard chemicals used were of analytical grade and purchased from Sigma Chemical Company (Beijing, China).
2.2. Isolation of Hemicelluloses In order to study structural differences of the hemicelluloses present in Populus tomentosa Carr., hemicellulosic fractions were isolated by sequential extractions according to the scheme in Fig. 1. The dewaxed sample was extracted by 0%, 10%, 30%, 50%, 70%, and 100% DMF aqueous solution with a solid to liquid ratio of 1:20 (g mL 1), respectively. The mixture was transferred into a 50-mL Teflon-lined stainless steel autoclave and maintained at 160 °C for 24 h. After the treatment period, the insoluble residues were separated from the solution by filtration, washed with distilled water until the pH of filtrates was neutral, and then dried at 60 °C for 12 h. The filtrates were adjusted to pH 6.0 with 6.0 M HCl, and then concentrated under reduced pressure and precipitated in three volumes of ethanol. After filtration, the precipitated fractions were washed with 70% ethanol and freeze-dried. It should be noted that the six hemicellulosic fractions (H0, H10, H30, H50, H70, H100) were obtained by precipitation in three volumes of ethanol from the 0%, 10%, 30%, 50%, 70%, and 100% DMF-soluble hemicelluloses, respectively.
Populus tomentosa Carr. Dried 60 °C
16 h.
Dried sample Dewaxed with toluene/ethanol (2:1, v/v) for 6 h. Dewaxed sample Extracted by 0%, 10%, 30%, 50%, 70%, and 100% DMF solution with a solid to liquid ratio of 1:20 (g mL-1), and then transferred into a 50-mL Teflon-lined stainless steel autoclave.at 160 °C for 24 h. Residue(Cellulose)
Filtrate (Discarded)
Filtrate Neutralized to pH 6.0 with 6.0 M HCl, concentrated under reduced pressure, and then precipitated in 3 volumes of ethanol. Pellet Washed with 70% ethanol and freeze-dried. Hemicelluosic fractions (H0, H10, H30, H50, H70, and H100)
Fig. 1. Scheme for isolation of hemicelluloses from Populus tomentosa Carr.
679
M.-G. Ma et al. / Bioresource Technology 114 (2012) 677–683
2.3. Chemical characterization
3. Results and discussion
The neutral sugars and uronic acids in the hemicellulosic fractions were obtained by hydrolysis with 1.0 M H2SO4 at 105 °C for 2.5 h. After hydrolysis, the samples were filtered and diluted 50-fold, and analyzed by high-performance anion-exchange chromatography (HPAEC) using a Dionex ICS3000 gradient pump, amperometric detector, AS50 autosampler and a Carbopac™ PA-20 column (4 250 mm, Dionex) and a guard PA-20 column (3 30 mm, Dionex). Neutral sugars and uronic acids were separated in 5 mM NaOH isocratic eluent (carbonate free and purged with nitrogen) for 20 min, followed by a 0–75 mM NaAc gradient in 5 mM NaOH for 15 min. Then the columns were washed with 200 mM NaOH to remove carbonate for 10 min, and followed a 5 min elution with 5 mM NaOH to re-equilibrate the column before the next injection. The total analysis time was 50 min, and the flow rate was 0.4 mL/min. Calibration was performed with standard solutions of L-arabinose, D-glucose, D-xylose, D-glucose, D-mannose, D-galactose, glucuronic acid, and galacturonic acids. The molecular weights of the hemicellulosic fractions were estimated by gel permeation chromatography (GPC) on a PL aquagelOH 50 column (300 7.7 mm, polymer laboratories Ltd.), calibrated with PL pullulan polysaccharide standards (peak average molecular weights of 783, 12,200, 100,000, and 1600,000 g mol 1, Polymer Laboratories Ltd.). A flow rate of 0.5 mL min 1 was maintained. The eluents were 0.02 M NaCl in 0.005 M sodium phosphate buffer, at pH 7.5. Detection was achieved with a Knauer differential refractometer. The column oven was kept at 30 °C. Hemicelluloses were dissolved in 0.005 M sodium phosphate buffer with 0.02 M NaCl, pH 7.5, at a concentration of 0.1% before measurement.
3.1. Content of neutral sugars and uronic acids
2.4. Spectroscopic and thermal characterization Fourier transform infrared spectra (FTIR) of hemicellulosic samples were recorded from an FT-IR spectrophotometer (Tensor 27, Brucker, Germany) using the KBr disk method. Thirty-two scans were taken of each sample recorded from the range 4000– 400 cm 1 at a resolution of 2 cm 1 in the transmission mode. The solution-state 1H NMR and 13C NMR spectra were recorded on a Bruker AVIII 400 MHz spectrometer using 20 mg for 1H and 80 mg for 13C of hemicellulose sample in 1.0 mL D2O. Additionally, to increase solubility of alkali-extractable hemicelluloses H5 in D2O, a few drops of sodium deuteroxide (7.5 M NaOD) were added. The chemical shifts were calibrated relative to the signals from D2O, used as an internal standard, at 4.70 ppm for the 1H NMR spectrum. The acquisition time (AQ) was 3.9 s, and the relaxation time was 1.0 s. The 13C NMR spectrum was recorded at 25 °C after 30,000 scans. A 30° pulse flipping angle, a 9.2 ls pulse width, a 1.89 s delay time and a 1.36 s AQ between scans were used. Heteronuclear Single Quantum Coherence (HSQC) experiment was also obtained on a Bruker AVIII 400 MHz spectrometer after 128 scans with 20 mg sample dissolved in 1.0 mL D2O. The spectral widths for HSQC were 5000 and 20,000 Hz for the 1H and 13C dimensions, respectively. The number of collected complex points was 1024 for the 1H dimension with a cycle delay of 1.5 s. The number of scans was 128, and 256 time increments were always recorded in 13C dimension. The 1JC–H used was 146 Hz. Prior to Fourier transformation, the data matrices were zero filled up to 1024 points in the 13C dimension. Thermal behavior of the hemicellulosic samples was performed using thermogravimetric analysis (TGA) and differential thermal analysis (DTA) on a simultaneous thermal analyzer (DTG-60, Shimadzu, Japan). The samples weighed between 8 and 13 mg and were heated from room temperature to 600 °C at a heating rate of 10 °C min 1 in flowing nitrogen.
To characterize the solubilized hemicelluloses, the six hemicellulosic fractions were performed for the determination of their constituent sugars. The results about the composition of neutral sugars and content of uronic acids in the six hemicellulosic fractions are listed in Table 1. As shown, the hot water-soluble hemicellulosic fraction (H0) contained a major amount of xylose (52.9%) and glucose (40.7%). Galactose (3.2%), arabinose (1.4%), and rhamnose (1.4%) were present in smaller amounts together with traces of uronic acids (0.3%), mainly originated from glucuronic acids (GlcpA) or 4-O-methyl-glucuronic acids (4-O-Me-a-DGlcpA) as side chains. This high percentage of xylose and glucose was taken to demonstrate correspondingly more xylans and glucans. The presence of hemicellulosic and pectic polysaccharides in hot water-soluble fraction from agricultural residues has been widely demonstrated. For example, Lawther et al. (1995) reported that wheat straw yields of 4.6% in hot water-soluble polysaccharides, in which xylose and galactose were the major sugar constituents. Xu et al. (2007) indicated that the hot water treatment probably released some amounts of pectic polysaccharides exception for noticeable amounts of b-glucan. Glucose (63.4%), galactose (19.3%), arabinose (8.2%), and xylose (7.0%) are the major sugar compositions of the alkali-soluble hemicellulosic fraction (H100). The value of xylose in H100 decreased from 52.9% to 7.0%, compared with that of the hot water-soluble hemicellulosic fraction (H0). The lower value of xylose was different from the previous reports (Xu et al., 2007). Moreover, the amount of galactose (19.3%) and uronic acids (1.6%) in H100 was greater than that of H0 (3.2% and 0.3%). The alkali-soluble hemicellulosic fraction in the literatures was obtained in the existence of water. However, the hemicellulosic fraction (H100) was produced in pure DMF solution. It can be concluded that pure DMF solution can probably solubilize arabinan and galactan. The intrinsic and detailed formation mechanism needs to be further explored. The further four hemicellulosic fractions (H10–H70) were obtained in the mixed solvents of water and DMF using 10%, 30%, 50%, and 70% DMF solution, respectively. Clearly, glucose (53.4–63.0%) and xylose (33.2–39.5%) were the predominant sugar components. This result demonstrated that the hydrothermal pretreatment favored the extraction of xylose with high percentage. Xylose content (H10–H70) was different from that in H100, which was due to the absence of water. In the literature, more than 90% of the xylan and nearly 50% of the lignin were extracted in a flow-through reactor at 165 °C (Pronyk and Mazza, 2012). In our study, the hemicelluloses were obtained by hydrothermal pretreatment at 160 °C for 24 h. The Teflon-lined stainless steel autoclave was used in the hydrothermal pretreatment, which was different from those (a flow-through reactor) in the literatures (Feria et al., 2012; Pronyk and Mazza, 2012). Galactose (1.3–3.9%), arabinose
Table 1 Neutral sugars and uronic acids (relative percent hemicellulosic sample, w/w) and Uro/Xyl ratios of hemicellulosic fractions. Sugars (%)
Rhamnose Arabinose Galactose Glucose Xylose Uronic acids Uro/Xyl
Hemicellulosic fractions H0
H10
H30
H50
H70
H100
1.42 1.40 3.22 40.74 52.91 0.29 0.01
0.06 0.20 2.13 63.04 33.20 1.36 0.04
0.03 0.52 2.43 55.38 39.48 2.15 0.05
0.34 0.62 1.29 58.39 35.89 3.46 0.10
0.24 5.12 3.86 53.41 34.55 2.76 0.08
0.49 8.19 19.26 63.40 7.00 1.58 0.23
680
M.-G. Ma et al. / Bioresource Technology 114 (2012) 677–683
(0.2–5.1%), and uronic acids (1.4–2.8%) were present in substantial amounts, and rhamnose (0.03–0.24%) was observed as small sugar constituents. Evidently, glucose content was increased and xylose content was decreased, compared with that of the hot water-soluble hemicellulosic fraction (H0), indicating that xylans is the predominant hemicelluloses in the cell walls of poplar wood (Sun et al., 2001). The predominant components of xylose and the substantial amount of uronic acids, mainly originated from glucuronic acids (GlcpA) or 4-O-methyl-glucuronic acids (4-O-Me-a-D-GlcpA) as side chains, demonstrated that the hemicelluloses with the mixed solvents of water and DMF pretreatment probably consisted mainly of glucuronoxylans. Furthermore, the four hemicellulosic fractions were also rich in glucose. The existence of glucose probably arose from b-glucans and xyloglucans under the hydrothermal treatment through organic alkaline solvent. For example, Wilkie et al. (1977), who supported the existence of b-glucans, stated that the hemicelluloses from delignified bamboo contained xylans and glucose-rich fractions. In addition, xyloglucan was a hemicellulosic polysaccharide in all higher plants, where they represent a quantitatively major building material of the primary cell wall (Ebringerová et al., 2005). The relatively high quantity of glucose and xylose implied that the hemicellulosic fractions were mixed polysaccharides of glucan and xylan. The increased glucose was also potentially considered as the direct consequence of the favorable effect of the hydrothermal pretreatment through organic alkaline solvent. However, further evidence to approve of this supposition needs additional experiments including the effect of the types of materials and alkali on the sugar composition of hemicelluloses. In general, the Uro/Xyl ratios were indicative of the degree of linearity or branching of hemicelluloses. It should be noted that the ratio of Uro/Xyl (0.04–0.10) in the hemicellulosic fractions (H10–H70) is more than that the value (0.01) in the water-soluble hemicellulosic fraction (H0), but lower than the value (0.23) in the alkali-soluble hemicellulosic fraction (H100), revealing that the H10–H70 had more branched chains than H0 and less branched chains than H100. From the ratio of uronic acids to xylose (0.01– 0.23) in the six hemicellulosic fractions, it can be indicated that the H0 seemed to be more linear and had less uronic acids than those of the H10–H100. These results are different from the previous report (Xu et al., 2007), in which the ratio of Uro/Xyl in the watersoluble hemicellulosic fraction was more than those of the alkalisoluble hemicellulosic fractions. This phenomenon suggested that the hydrothermal pretreatment method at high temperature for a longer period may result in the difference between the water-soluble hemicellulosic fraction and the alkali-soluble hemicellulosic fraction. In addition, the value of Uro/Xyl in H10–H70 increased from 0.04 to 0.05 to 0.10 firstly, and then decreased from 0.10 to 0.08 with the increasing ratio of DMF/water. It can be concluded that the ratio of DMF/water play an important role in the sugar compositions and structure of hemicellulosic fractions and the appropriate ratio of DMF to water is important for the pretreatment of lignocelluloses. 3.2. Molecular weight distribution In this study, the molecular weights of the hemicellulosic fractions were determined by gel permeation chromatography (GPC). As reported, GPC is effective in estimating the molecular weight of unknown polymers of similar or identical chemical structures to those used to calibrate columns (Himmel et al., 1989). The values of the weight-average (Mw) and number-average (Mn) molecular weights as well as polydispersity (Mw/Mn) are given in Table 2. Obviously, the water-soluble hemicellulosic fraction showed a relatively lower degree of polymerization with Mw values of 3960 g mol 1 than the alkali-soluble hemicellulosic fraction (H100) with Mw of 5600 g mol 1. This implied that the pretreatment
Table 2 Weight-average (Mw) and number-average (Mn) molecular weights and polydispersity (Mw/Mn) of the hemicellulosic fractions. Hemicellulosic preparations
Mw Mn Mw/Mn
H0
H10
H30
H50
H70
H100
3960 1690 2.34
3330 1370 2.43
4690 1520 3.08
24,840 2310 10.75
17,060 1730 9.86
5600 3340 1.68
with hot water released only hemicellulosic polymers with low molecular weights and a slight degradation of macromolecules structure of hemicelluloses probably occurred. Additionally, the experimental results indicated that the extraction with an increasing DMF concentration from 10% to 50% led to grow of Mw from 3330 to 24,840 g mol-1, suggesting that increasing DMF concentration (from 10% to 50%) favored the dissolution of hemicelluloses with large molecular size. Conversely, as the DMF concentration was further increased from 50% to 70%, the Mw value decreased from 24,840 to 17,060 g mol 1. Clearly, the hemicellulosic fractions (H50 and H70), obtained from 50% and 70% DMF concentrations, showed a very much higher degree of polymerization with Mw values between 24,840 and 17,060 g mol 1 than the other hemicellulosic fractions (H0, H10, H30, and H100) with Mw from 3330 to 5600 g mol 1. These results implied that the hemicellulosic fractions (H50 and H70) using relatively high DMF concentration can result in the dissolution of large hemicellulosic molecules from the materials. Furthermore, on the basis of the Uro/Xyl ratio and molecular weights results, it may be concluded that the hemicellulosic fractions with a higher Uro/ Xyl ratio, namely the more branched hemicelluloses, had higher molecular weights, except that the hemicellulosic fraction (H100) with pure DMF pretreatment. These results were consistent with the previous reports. Hoffmann et al. (1991) indicated the highly branched xylan fractions to be of higher molecular weight than their less branched fractions. The hemicellulosic fraction (H100) with a higher Uro/Xyl ratio had lower molecular weights, which might due to the differences in the techniques of hydrothermal pretreatment and pure DMF solvent condition. These results also indicated that the addition of DMF favored the extraction of hemicellulosic fractions with high Mw molecular weights and the appropriate DMF concentration was important for the obtained hemicellulosic fraction with appropriate Mw molecular weights. The molecular weight distribution curves of the six hemicellulosic fractions are also shown in Fig. 2, and the results of the polydispersity were calculated by Mw/Mn. As can be seen from Fig. 2, the polydispersity of the water-soluble hemicellulosic fraction (H0) and the alkali-soluble hemicellulosic fraction (H100) were 2.34 and 1.68, respectively, which declared a narrow distribution of molecular size. A much higher polydispersity (2.43–10.75) was observed in four hemicellulosic fractions (H10–H70), compared with the H0 (2.34) and H100 (1.68), which showed a broad distribution of molecular weights, implying that the polymers with the mixed solvents of DMF and water generally had a wider range of molecular weights than those of the water-soluble fraction and alkali-soluble fraction. The polydispersity was consistent with the Mw results. Xu et al. (2007) also reported the similar results that the polydispersity of hot water-soluble hemicellulosic fraction was lower than that of alkali-soluble hemicellulosic fraction.
3.3. FTIR spectra Infrared spectroscopy (FTIR), which is quite extensively applied in plant cell wall analysis, has been proved to be useful for studying functional groups of carbohydrates (Kacˇuráková and Wilson,
M.-G. Ma et al. / Bioresource Technology 114 (2012) 677–683
681
Fig. 2. Molecular weight distribution curves of six hemicellulosic fractions.
2001). In this study, FTIR spectroscopy was used to evaluate the structural features among the six hemicellulosic fractions. FTIR spectra of water-soluble hemicellulosic fraction (H0) and alkali-soluble hemicellulosic fraction (H10–H100) are shown in Fig. S1 (see Fig. S1 in Supplementary data), respectively. The signal at 3408 cm 1 is assigned to the OH stretching vibrations of the hemicelluloses, and two bands at 2923 and 2855 cm 1 are assigned to C–H stretching. In addition, the intense band at 1621 cm 1 was due to the bending mode of the absorbed water, since the hemicelluloses usually have a strong affinity for water, and in the solid state these macromolecules may have disordered structures, which can easily be hydrated (Kauráková et al., 1998). A prominent absorbance at 1043 cm 1 was attributed to the C–O–C stretching of glycosidic linkages in xylans, indicating a dominant xylan of the fractionated hemicelluloses (Kacurakova et al., 1994). The small band at 892 cm 1, which arose from the C1 group frequency, was indicative of b-glycosidic linkage between the sugar units (Gupta et al., 1987). FTIR spectrum of alkali-soluble hemicellulosic fraction (H100) has slight difference, compared with the hot water-soluble fraction (H0). The presence of a small signal at 1736 cm 1 is as-
signed to the acetyl, uronic, and ferulic ester groups of the polysaccharides. The signal at 1736 cm 1 in the spectra of the hot watersoluble fraction (H0) disappeared, indicating that the acetyl group converted into acetic acid via hydrothermal pretreatment. Interestingly, the presence of this signal at 1736 cm 1 in the spectra of the alkali-soluble hemicellulosic fraction (H100) likely indicated that the pure organic alkali solution didnot include the groups, which can cleave this ester bond from hemicelluloses (see Fig. S1f in Supplementary data). The peak intensity at 1621 cm 1 decreased from Fig. S1a to Fig. S1f. The strong band at 1464 cm 1 was clearly observed. FTIR spectra of the four hemicellulosic fractions in the mixed solvents of DMF to water (H10–H70) are illustrated in Fig. S1b–e (see Fig. S1b–e in Supplementary data). As expected, the four hemicellulosic fractions showed very similar spectra, indicating a similar structure of the four hemicellulosic fractions. However, compared with H0 and H100, there are some difference in the peaks and peak intensities. The peak intensity of hemicellulosic fractions in the mixed solvents of DMF to water (H10–H70) was stronger than that of hot water-soluble hemicellulosic fraction (H0). The other
682
M.-G. Ma et al. / Bioresource Technology 114 (2012) 677–683
bands at 1464 and 1366 cm 1 were associated with C–H, OH or CH2 bending vibrations for the typical hemicelluloses. The signal at 1736 cm 1 disappears, verifying that the mixed solvents of DMF to water pretreatment completely cleaved this ester bond from hemicelluloses. As shown, the peak intensity of the H10 at 1043 cm 1 was lower than those of the other hemicellulosic fractions, indicating that the H10 released few amounts of xylans in accordance with the result of the previous sugar analysis. A small band at 1157 cm 1 was assigned to C–O and C–O–C stretching with some contribution of OH bending in arabinoxylans. One can clearly observe that the peak intensity increased with the increasing DMF concentration, suggesting that increasing alkali concentration probably increased the amounts of arabinoxylans. In addition, the shoulder peaks at 938 and 840 cm 1, probably due to the signal overlapping, indicated the existence of a-glycosidic linkage (Huang and Zhang, 2009). 3.4. NMR spectra Nuclear magnetic resonance (NMR) spectroscopy has significant advances in understanding complicated structure. To further elucidate the structural features of the polymers, the hemicellulosic fraction (H70) with 70% DMF solution pretreatment was investigated using one-dimensional (1D 1H and 13C) and the two-dimensional (2D HSQC) NMR spectroscopy. NMR spectrum is supposed to assay and identify the polymer backbone and the type of sidechain branching along the backbone. The 1H and 13C NMR and 2D HSQC spectra of H70 are given in Figs. S2–S4, respectively (see Figs. S2–S4 in Supplementary data). The signals for 1H and 13C NMR were assigned on the basis of the HSQC spectrum and the assignment data of 1H and 13C NMR spectra are given in Table 3. As can be seen from Fig. S2 (see Fig. S2 in Supplementary data), the anomeric 1H NMR signals in the spectrum of H70 were found in the spectral region of 4.3–5.6 ppm (Chiarini et al., 2004). The relevant signals occurred in two regions, namely, the anomeric region (d 5.6–4.9 for a-anomers and d 4.9–4.3 for b-anomers) and the ring proton region (d 4.5–3.0). To confirm the structural features, the H70 was also characterized by 13C NMR, as shown in Fig. S3 (see Fig. S3 in Supplementary data). The 13C NMR spectrum of H70 showed five main signals at d 101.9 (C-1), 72.9 (C-2), 74.4 (C-3), 75.7 (C-4), and 63.0 ppm (C-5), corresponding to the (1 ? 4)-linked b-D-Xylp units. Other main signals at d 101.9 (C-1), 74.2 (C-2), 74.0 (C-3), 79.5 (C-4), 71.2 (C-5) and 60.4 ppm (C-6) are attributed to the (1 ? 4)-linked a-D-Glcp units. From Fig. S4 (see Fig. S4 in Supplementary data), the predominant five signals gave HSQC 13C/1H cross-peaks at 101.9/4.30, 72.9/3.16, 74.4/3.38, 75.8/ 3.64, and 63.0/3.95 + 3.26 ppm (Table 3), which are assigned to C-1, C-2, C3, C-4, and C-5 of the (1 ? 4)-linked b-D-xylan, respectively (Sun et al., 2001). In addition, the cross-peaks at 101.9/5.07, 72.4/3.42, 74.0/3.75, 79.7/3.34, 71.2/3.70, and 60.4/3.69 ppm correspond to C-1, C-2, C-3, C-4, C-5, and C-6 of the (1 ? 4)-linked a-D-glucan, i.e., amylose (Huang and Zhang, 2009; Lisboa et al., 2005). A strong
Table 3 1 H and 13C chemical shift (ppm) assignments for hemicellulosic fraction H70. Sugar residues
X
a
Gb a b c d
Chemical shift (H/C) 1
2
3
4
5axc
5eqd
4.30 101.9 5.07 101.9
3.16 72.9 3.42 72.4
3.38 74.4 3.75 74.0
3.64 75.8 3.34 79.7
3.26 63.0 3.70 71.2
3.95 63.0
X, (1 ? 4)-linked b-D-xylan. G, (1 ? 4)-linked a-D-glucan. ax, axial. eq, equatorial.
6
signal at 4.70 ppm is attributed to the residual solvent (HDO), while the strong signals at 36.4/2.21 and 170.8/8.39 ppm are assigned to the DMF solution. Thus, from the results of NMR and the previous sugar analysis, it can be concluded that the hemicellulosic fraction (H70) is mainly composed of (1 ? 4)-linked a-Dglucan and (1 ? 4)-linked b-D-xylan attached with minor amounts substituted sugars and glucuronic acids. 3.5. Thermal analysis Thermogravimetric analysis (TGA) and differential thermal analysis (DTA) were used to investigate the thermal stability and the physicochemical properties of the hemicelluloses fractions. Fig. S5 (see Fig. S5 in Supplementary data) shows the typical TGA and DTA curves of the hot water-soluble hemicellulosic fraction (H0) and the hemicellulosic fraction (H70) with 70% DMF solution pretreatment. As shown, the weight loss occurred at the beginning is attributed to the evaporation of water, indicating that the water probably adsorbed on the surface of the material. From the TGA curves, it can be seen that the weight losses of the H0 and H70 occurred mainly in the temperature range of 208–350 °C and 200–330 °C, respectively. At 10% weight loss, the decomposing temperature of the two hemicellulosic fractions (H0 and H70) occurred at 188 and 173 °C, respectively. The total weight losses of 47.7% and 83.8% were observed with increasing DMF concentration from 0% to 70%, indicating the H0 had a higher thermal stability than that of the H70. The thermal stability of the water-soluble hemicellulosic fraction appeared to be higher than that of the hemicellulosic fraction with organic alkali–water mixed solvents pretreatment, which corresponded to the increasing trends of their molecular weights from 3960 to 17,060 g mol 1 in Table 2, implying that linear hemicelluloses showed the characterization of more thermal stability than the branched ones during pyrolysis. Furthermore, DTA curves for the hemicellulosic fractions (H0 and H70) exhibit a single small endothermic and two exothermic peaks. The exothermic peaks, which represent heat released from the hemicellulosic fractions, were observed at 291 and 427 °C for H0 and at 313 and 499 °C for H70, respectively. In general, the DTA curves of hemicellulosic fraction showed two degradation peaks: the first peak was related to initial hemicellulose loss at lower temperature, and the second one related the organic material oxidation (Choi et al., 2009). 4. Conclusions In summary, the influences of DMF with different concentrations and hydrothermal pretreatment on structural and properties of the hemicelluloses were researched. The hydrothermal pretreatment favored the extraction of xylose with high percentage. The H10–H70 had more branched chains than H0 and less branched chains than H100. The addition of DMF favored the extraction of hemicelluloses with high molecular weights Mw. The hemicellulosic fractions (H70) were mainly composed of (1 ? 4)-linked a-Dglucan from amylose and (1 ? 4)-linked b-D-xylan from minor amounts of branched sugars. The values of molecular weights and branching of the hemicelluloses played an important role in their thermal stability. Acknowledgements
3.69 60.4
Financial support from National Natural Science Foundation of China (31070511), the Program for New Century Excellent Talents in University, and Major State Basic Research Development Program of China (973 Program) (No. 2010CB732204) is gratefully acknowledged.
M.-G. Ma et al. / Bioresource Technology 114 (2012) 677–683
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. 2012.03.048.
References Adler, E., 1977. Lignin chemistry: past, present and future. Wood Sci. Technol. 3, 169–218. Aspinall, G.O., Mahomed, R.S., 1954. The constitution of a wheat-straw xylan. J. Chem. Soc. 6, 1731–1734. Biermann, C.J., Schultz, T.P., McGinnis, C.D., 1984. Rapid steam hydrolysis/extraction of mixed hardwoods as biomass. J. Wood Chem. Technol. 4, 111–128. Chiarini, L., Cescutti, P., Drigo, L., Impallomeni, G., Herasimenka, Y., Bevivino, A., Dalmastri, C., Tabacchioni, S., Manno, G., Zanetti, F., Rizzo, R., 2004. Exopolysaccharides produced by Burkholderia cenocepacia recA lineages IIIA and IIIB. J. Cyst. Fibros. 3, 165–172. Choi, M.J., Soottitantawat, A., Nuchuchua, O., Min, S.G., Ruktanonchai, U., 2009. Physical and light oxidative properties of eugenol encapsulated by molecular inclusion and emulsion–diffusion method. Food Res. Int. 42, 148–156. Ebringerová, A., 2005. Structural diversity and application potential of hemicelluloses. Macromol. Symp. 232, 1–12. Ebringerová, A., Hromádková, Z., Heinze, T., 2005. Hemicellulose. Adv. Polym. Sci. 186, 1–67. Feria, M.J., Alfaro, A., López, F., Pérez, A., García, J.C., Rivera, A., 2012. Integral valorization of Leucaena diversifolia by hydrothermal and pulp processing. Bioresour. Technol. 103, 381–388. Freudenberg, K., 1965. Lignin: its constitution and formation from phydroxycinnamyl alcohols. Science 148, 595–600. Glasser, W.G., Kaar, W.E., Jain, R.K., Sealey, J., 2000. Isolation options for noncellulosic heteropolysaccharides (HetPS). Cellulose 7, 299–317. Goto, M., Obuchi, R., Hirose, T., Sakaki, T., Shibata, M., 2004. Hydrothermal conversion of municipal organic waste into resources. Bioresour. Technol. 93, 279–284. Gupta, S., Madan, R.N., Bansal, M.C., 1987. Chemical composition of Pinus caribaea hemicellulose. Tappi J. 70, 113–114. Himmel, M.E., Ding, S.Y., Johnson, D.K., Adney, W.S., Nimlos, M.R., Brady, J.W., Foust, T.D., 2007. Biomass recalcitrance: engineering plants and enzymes for biofuels production. Science 315, 804–807. Himmel, M.E., Tatsumoto, K., Oh, K.K., Grohmann, K., Johnson, D.K., Chum, H.L., 1989. Molecular weight distribution of aspen lignins estimated by universal calibration. In: Glasser, W.G., Sarkanen, S. (Eds.), Lignin Properties and Materials. ACS, Washington, DC, pp. 82–100. Hoffmann, R.A., Roza, M., Maat, J., Kamerling, J.P., Vliegenthart, J.F.G., 1991. Structural characteristics of the cold-water-soluble arabinoxylans from the white flour of the soft wheat variety Kadet. Carbohydr. Polym. 15, 415–430. Huang, Z.P., Zhang, L.N., 2009. Chemical structures of water-soluble polysaccharides from rhizoma Panacis japonici. Carbohydr. Res. 344, 1136–1140. Kacˇuráková, M., Belton, P.S., Wilson, R.H., Hirsch, J., Ebringerova, A., 1998. Hydration properties of xylan-type structures: an FTIR study of xylooligosaccharides. J. Sci. Food Agric. 77, 38–44.
683
Kacˇuráková, M., Ebringerova, A., Hirsch, J., Horomadkova, Z., 1994. Infrared study of arabinoxylans. J. Sci. Food Agric. 66, 423–427. Kacˇuráková, M., Wilson, R.H., 2001. Developments in mid-infrared FT-IR spectroscopy of selected carbohydrates. Carbohydr. Polym. 44, 291–303. Krishna, S.H., Chowdary, G.V., 2000. Optimization of simultaneous saccharification and fermentation for the production of ethanol from lignocellulosic biomass. J. Agric. Food Chem. 48, 1971–1976. Lawther, J.M., Sun, R.C., Banks, W.B., 1995. Extraction, fractionation, and characterization of structural polysaccharides from wheat straw. J. Agric. Food Chem. 43, 667–675. Leppänen, K., Spetz, P., Pranovich, A., Hartonen, K., Kitunen, V., Ilvesniemi, H., 2011. Pressurized hot water extraction of Norway spruce hemicelluloses using a flowthrough system. Wood Sci. Technol. 45, 223–236. Lisboa, S.A., Evtuguin, D.V., Neto, C.P., Goodfellow, B.J., 2005. Isolation and structural characterization of polysaccharides dissolved in Eucalyptus globulus kraft black liquors. Carbohydr. Polym. 60, 77–85. López, M.J., Nichols, N.N., Dien, B.S., Moreno, J., Bothast, R.J., 2004. Isolation of microorganisms for biological detoxification of lignocellulosic hydrolysates. Appl. Microbiol. Biotechnol. 64, 125–131. Makishima, S., Nozaki, K., Mizuno, M., Netsu, E., Shinji, K., Shibayama, T., Kanda, T., Amano, Y., 2006. Recovery of soluble sugars from waste medium for Enokitake (Flammulina velutipes) mushroom cultivation with hydrothermal reaction and enzyme digestion. J. Appl. Glycosci. 53, 261–266. Nhuan, P.N., Justin, M., David, B.J., Caye, D., 2011. Fractionation of corn fiber treated by soaking in aqueous ammonia (SAA) for isolation of hemicellulose B and production of C5 sugars by enzyme hydrolysis. Appl. Biochem. Biotechnol. 164, 1390–1404. Pin´kowska, H., Wolak, P., Złon´ciska, A., 2011. Hydrothermal decomposition of xylan as a model substance for plant biomass waste-Hydrothermolysis in subcritical water. Biomass Bioenergy 35, 3902–3912. Pronyk, C., Mazza, G., 2012. Fractionation of triticale, wheat, barley, oats, canola, and mustard straws for the production of carbohydrates and lignins. Bioresour. Technol. 106, 117–124. Sasaki, M., Adschiri, T., Arai, K., 2003. Fractionation of sugarcane bagasse by hydrothermal treatment. Bioresour. Technol. 86, 301–304. Sun, R.C., Fang, J.M., Tomkinson, J., Geng, Z.C., Liu, J.C., 2001. Fractional isolation, physico-chemical characterization and homogeneous esterification of hemicelluloses from fast-growing poplar wood. Carbohydr. Polym. 44, 29–39. Sun, R.C., Tomkinson, J., 2002. Comparative study of lignins isolated by alkali and ultrasound-assisted alkali extractions from wheat straw. Ultrason. Sonochem. 9, 85–93. Sun, R.C., Tomkinson, J., Geng, Z.C., Wang, N.J., 2000a. Comparative studies of hemicelluloses solubilized during the treatment of maize stems with peroxymonosulfuric acid, peroxyformic acid, peracetic acid, and hydrogen peroxide. Part 1. Yield and chemical characterization. Holzforschung 54, 349– 356. Sun, R.C., Tomkinson, J., Ma, B.L., Liang, S.F., 2000b. Comparative study of hemicelluloses from rice straw by alkali and hydrogen peroxide treatments. Carbohydr. Polym. 42, 111–122. Wilkie, K.C.B., Woo, S.L., 1977. A heteroxylan and hemicellulosic materials from bamboo leaves, and a reconsideration of the general nature of commonly occurring xylans and other hemicelluloses. Carbohydr. Res. 49, 145–162. Xu, F., Sun, J.X., Geng, Z.C., Liu, C.F., Ren, J.L., Sun, R.C., Fowler, P., Baird, M.S., 2007. Comparative study of water-soluble and alkali-soluble hemicelluloses from perennial ryegrass leaves (Lolium peree). Carbohydr. Polym. 67, 56–65.