Carbohydrate Polymers 126 (2015) 185–191
Contents lists available at ScienceDirect
Carbohydrate Polymers journal homepage: www.elsevier.com/locate/carbpol
Fractionation and characterization of saccharides and lignin components in wood prehydrolysis liquor from dissolving pulp production Zhaojiang Wang a,∗ , Xiaojun Wang a , Jungang Jiang a , Yingjuan Fu a , Menghua Qin a,b,c,∗ a
Key Laboratory of Paper Science & Technology, Qilu University of Technology, Jinan 250353, China Laboratory of Organic Chemistry, Taishan University, Taian 271021, China c Huatai Group Corp. Ltd., Dongying 257335, Shandong, China b
a r t i c l e
i n f o
Article history: Received 22 July 2014 Received in revised form 28 February 2015 Accepted 3 March 2015 Available online 14 March 2015 Keywords: Oligosaccharides Lignin Fractionation Membrane Prehydrolysis
a b s t r a c t Saccharides and lignin components in prehydrolysis liquor (PHL) from kraft-based dissolving pulp production was characterized after being fractionated using membrane filtration. The results showed that the membrane filtration provided a method for organics fractionation with considerable recovery rate, but exhibited some disadvantages. Besides the limited ability in purifying oligosaccharides (OS) due to the overlaps of molecular weight distribution with lignin components, the membrane filtration could not improve the homogeneity of OS as indicated by the analysis of chemical compositions and the degree of polymerization (DP), which may be ascribed to the linear conformation of OS. The characterization of lignin components indicated a great potential for polymer industry because of the remarkable content of phenolic hydroxyl groups (PhOH), especially for low molecular weight (LMW) fraction. It was concluded the organics in PHL provided streams of value-added chemicals. However, the practical significance thereof can be realized and maximized only when they are successfully and completely fractionated. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction The prehydrolysis-kraft process currently attracts widespread attention because it not only produces high grade dissolving pulp but also provides a potential way for high-value utilization of hemicellulose (Buranov & Mazza, 2010; Nabarlatz, Ebringerová, & Montané, 2007; Sun et al., 2005). This process fits perfectly into the forest biorefinery concept if these hemicellulosic oligosaccharides (OS) in prehydrolysis liquor (PHL) could be fully isolated and purified. However, the prehydrolysis step that serves the purpose of hemicelluloses extraction is ineluctably accompanied with the depolymerization of other component of biomass, namely lignin degradation. This makes the OS separation a challenging task because of the presence of lignin components (Bujanovic, Goundalkar, & Amidon, 2012; Tunc & van Heiningen, 2011).
∗ Corresponding authors at: Key Laboratory of Paper Science & Technology, Qilu University of Technology, University Avenue No.3801, Changqing, Jinan 250353, Shandong, China. Tel.: +86 531 89631168; fax: +86 531 89631117. E-mail addresses: [email protected], [email protected] (Z. Wang), [email protected] (M. Qin). http://dx.doi.org/10.1016/j.carbpol.2015.03.012 0144-8617/© 2015 Elsevier Ltd. All rights reserved.
It was reported the lignin reactions during prehydrolysis mainly involved the homolytic cleavage of aryl ether bonds, the generation of new phenolic units and condensation reactions occurred at extremely high temperature (Bardet, Robert, & Lundquist, 1985). These reactions release a diverse palette of degradation byproducts. Liu, Ni, Fatehi, and Saeed (2011) found that the depolymerized lignin in PHL had a lower average molecular weight but a significant increase of the phenolic hydroxyl groups (PhOH) compared to dioxane lignin. Similar result was reported by Leschinsky, Zuckerstätter, Weber, Patt, and Sixta (2008) that the extensive lignin degradation occurs during prehydrolysis through homolytic cleavage of the aryl-ether bonds, resulting in a strong increase of PhOH and a decrease in aliphatic hydroxyl groups. Another study of Leschinsky, Weber, Patt, and Sixta (2009) reported that the formation of the condensed lignin in PHL is characterized by a higher molecular weight and insoluble at low temperature. His study also indicated a broad distribution of molecular weight of lignin components in PHL. Besides these lignin oligomers, a variety of low molecular weight (LMW) aromatic compounds were present as a result of the cleavage of the ␣-O-4 linkages of lignin. Goundalkar, Bujanovic, and Amidon (2010) demonstrated that vanillin and syringaldehyde were the major aromatic compounds in maple PHL. Besides these degradation byproducts of lignin, the dissolving organics in PHL
186
Z. Wang et al. / Carbohydrate Polymers 126 (2015) 185–191
also include furfurals and organic acids that are derived from carbohydrate degradation (Cara et al., 2012; Liu, Hu, Jahan, & Ni, 2013; Saeed et al., 2012), and therefore make the compositions of PHL more complicated. Various methods have been developed for OS separation by removing non-saccharide contaminants, namely lignin components, e.g. polymer flocculation (Duarte, Ramarao, & Amidon, 2010), solvent extraction (Vázquez, Garrote, Alonso, Domínguez, & Parajó, 2005), adsorption by surface active materials (Liu, Fatehi, & Ni, 2012; Montané, Nabarlatz, Martorell, Torné-Fernández, & Fierro, 2006), ion exchange or chromatography techniques (Cara et al., 2012; Palm & Zacchi, 2004), membrane filtration (Al Manasrah, Kallioinen, Ilvesniemi, & Mänttäri, 2012; Buranov & Mazza, 2010), and combination of these techniques (Chen, Wang, Fu, Li, & Qin, 2014; Shen, Kaur, Baktash, He, & Ni, 2013). However, their performances remain to be improved and currently far from the requirement of industrial application in terms of process cost and recovery yield as reviewed by Qing, Li, Kumar, and Wyman (2013). Although the separation strategies of OS from PHL were extensively studied, the characterization of the dissolving organics is rather limited. It is believed that a better understanding of the molecular information about the dissolving organics in PHL is of vital importance for biorefinery and will undoubtedly facilitate the OS separation. The present work addresses the fractionation and characterization of organic compounds in PHL. This is an effort not only to provide fundamental information beneficial to OS purification, but also to wood biorefinery. 2. Materials and methods Fig. 1. Flow diagram for the fractionation of PHL by MF and UF.
2.1. Materials The fast-growing aspen species, Populus × euramericana ‘Neva’, was selected as raw materials in the present study because it is widely used in dissolving pulp mills through prehydrolysis kraft process in North China. Poplar wood chips were prepared from debarked wood logs harvested from the southwest region of Shandong province, China. Microporous membrane made of mixed cellulose esters with pore size of 0.45 m and Ultracel regenerated cellulose membrane with molecular weight cut off (MWCO) 1.0, 3.0, 10.0, 30.0 kD were provided by Merck Millipore, Billerica, MA. All chemicals were analytical-reagent grade from Sigma-Aldrich, Inc. The chemicals included 2 N Folin–Ciocalteu’s phenol reagent and 3,5-dinitrosalicylic acid (DNS), anhydrous sodium carbonate, and 4-hydroxybenzoic acid. Dextran with weight average molecular weight from 0.18 to 36.3 kD (American Polymer Standards Corporation, Mentor, OH) was used as standard samples for the size exclusion chromatography (SEC) analysis. Xylose and xylooligosaccharides (XOS) purchased from Megzyme Ireland with degree of polymerization (DP) from 2 to 6 were used as standard for high performance anion exchange chromatography with pulsed amperometric detector (HPAEC-PAD) to get DP profile of saccharides in PHL.
was withdrawn. The composition of aspen wood before and after treatment as well as the PHL composition were listed in Table 1. 2.3. Fractionation by membrane filtration Batch filtration processing was conducted to fractionate the organics in PHL according to the scheme in Fig. 1. Prior to ultrafiltration (UF), the PHL was subjected to microfiltration (MF) using 0.45 m membrane to eliminate insoluble particles. The UF was performed in dead-end mode using a stirred cell (Millipore) at ambient temperature and 0.3 MPa by compressed nitrogen. Flat sheet Millipore membranes with an effective membrane area of 40 cm2 were employed. A magnetic stirrer bar was employed at a stirring rate of 100 rpm. Reverse stirring was also applied every 10 min to ensure that the feed solution was well mixed. The nominal MWCO of membranes used was 30, 10, 3 and 1 kD, respectively. The PHL starting volume was approximately 150 mL. In each UF step, the volume concentration ratio was controlled to be 10, and the permeate was collected and filtered using the next lower MWCO membrane. 2.4. Analytical methods
2.2. Prehydrolysis and PHL preparation The prehydrolysis was carried out in a 23 L pulp digester, using 1.0 kg of oven-dried poplar chips. Deionized water was added in order to reach 6:1 of liquid to wood ratio. The digester temperature was ramped from room temperature to 170 ◦ C and held for 60 min. This is a representative reaction severity widely applied in industrial process that corresponds to P-factor of 550 (Duarte, Ramarao, Amidon, & Ferreira, 2011). At the end of prehydrolysis, the digester was cooled, depressurized and the reaction mixture
The reducing sugar in solution was determined using DNS method (Miller, 1959). While a strict definition of an OS is not established, saccharides with DP greater than 2 but soluble in PHL were considered as OS in present study. The OS content was measured by an indirect method based on quantitative acid hydrolysis of the liquid samples with 4% w/w of H2 SO4 at 121 ◦ C for 60 min according to technical report from NREL (Sluiter et al., 2006). The OS concentration was expressed as the increase of monosaccharides (MS) which were determined by HPAEC-PAD. This system
Z. Wang et al. / Carbohydrate Polymers 126 (2015) 185–191
187
Table 1 Chemical compositions of aspen wood before and after prehydrolysis as well as the compositions of the PHL. Untreated wood (1.0 kg)
Residual solid (0.81 kg)
PHL (6.0 L)a Oligosaccharides (18.89 g/L)
Arabinan 0.38% Galactan 0.71% Glucan 40.7% Xylan 16.4% Mannan 3.8% Lignin 23.5%
a b c d e f g h i
Arabinan ndb Galactan nd Glucan 49.2% Xylan 5.3% Mannan 1.7% Lignin 26.1%
AOSc 5.7%d GalOSe 7.6% GluOSf 8.9% XOSg 70.5% MOSh 7.2%
Monosaccharides (2.46 g/L) Arabinose 24.9% Galactose 15.9% Glucose 10.8% Xylose 40.2% Mannose 8.1%
Non-saccharides (15.34 g/L) Lignin 7.12 g/L Formic acid 2.23 g/L Acetic acid 4.10 g/L Furfural 1.64 g/L HMFi 0.25 g/L
Theoretical volume of PHL. Not detected. Arabinooligosaccharides. All wt%. Galactooligosaccharides. Glucooligosaccharides. Xylooligosaccharides. Mannooligosaccharides. 5-Hydroxymethyl furfural.
consisted of a Dionex (Sunnyvale, CA) HPLC system (ICS-5000) equipped with a GP40 gradient pump, an anion exchange column (CarboPac PA20 and a guard) and an ED40 electrochemical detector. Aliquots (25 L) were injected manually after passing through a 0.2 m nylon syringe filter. Gradient was set as 2 mM NaOH isocratic with a step to 200 mM NaOH at 10 min to regenerate the column at a flow rate of 0.5 mL/min and 25 ◦ C. MS were quantified with reference to standards using the same analytical procedure. The DP profile of OS was obtained by HPAEC-PAD system same to MS determination, but with different column (CarboPac PA1 and a guard) and elution conditions. Gradient for DP profile of OS was set as 0–300 mM NaOAc in 100 mM NaOH from 0 to 30 min at a flow rate of 0.5 mL/min and 25 ◦ C. The column was reconditioned using 100 mM NaOH after each analysis. The degradation byproducts of carbohydrate in HPL such as formic acid, acetic acid, furfural and HMF were analyzed by HPLC equipped with Waters C18 symmetry column (4.6 × 150 mm, 5 m) at 30 ◦ C with 0.1% H3 PO4 (v/v) as eluent at 0.5 mL/min. All samples were analyzed in duplicates. The molecule weight distribution of the OS and lignin in was analyzed by SEC using a HPLC system comprised of Shimadzu LC-20T, Shimadzu SB-803 HQ column (8 × 300 mm, 6 m), a UV detector and a refraction index (RI) detector with pure water as mobile phase at 0.5 mL/min and 35 ◦ C. The lignin content was measured according to a spectrometric method at a wavelength of 205 nm (TAPPI UM-250). The PhOH content of lignin was determined according to Folin–Ciocalteau assay (Blainski, Lopes, & De Mello, 2013). All samples were analyzed in duplicates. The LMW degradation byproducts of lignin in PHL were determined by gas chromatography mass spectrometry (GC/MS) after being extracted using methyl tert-butyl ether (MTBE) as solvent. The extraction procedures involved introducing 15 mL MTBE into 30 mL PHL, shaking generally for 5 min, and the collection of extract. Extraction was repeated for three times for a better separation. The extract containing LMW degradation byproducts of lignin was concentrated and subjected to GC/MS analysis using Agilent 7890 GC equipped with Agilent 5975 C detector. A capillary column HP-1 (6–7 m × 0.53 mm, 0.15 m) was used. The temperature program used during the GC/MS analysis was as follows: starting from 40 ◦ C, ramp to 280 ◦ C in 24 min and hold for 20 min. The mass spectra were recorded in the interval 35–250 amu. Organics were identified by comparing their spectra with those available in the NIST libraries.
3. Results and discussion 3.1. Distribution of saccharides and lignin components after fractionation The recovery rate after fractionation was determined for lignin and saccharides (Fig. 2). The recovery rate of saccharides was 85%, slightly less than that of lignin. This suggested that the saccharides may contributed more than lignin to the fouling of membranes. The relative yield of saccharides and lignin among individual fractions after MF and UF is given in Table 2. It was observed that more than half of organics could pass through the 1 kD membrane and present in fraction P4 . MF is effective in lignin separation as indicated by the lowest content of saccharides in fraction R0 . For UF treatment, the relative yield of saccharides showed an increase from 4.21% in R1 to 24.43% in R4 , while the variation of the lignin yield was not significant. This led to highest saccharides content of 84.03% in R4 . However, because more than half saccharides was found in P4 , the basically unchanged saccharides content in P4 and original PHL represents a drawback for UF in saccharides purification.
Fig. 2. Recovery rate of saccharides and lignin in fractions obtained from membrane filtration.
188
Z. Wang et al. / Carbohydrate Polymers 126 (2015) 185–191
Table 2 Relative yield and saccharides content in all fractions. Fractions
PHL R0 R1 R2 R3 R4 P4
Relative yield (%)
Saccharides content (%)
Saccharides
Lignin
3.38 4.12 3.36 12.7 24.4 51.9
17.3 6.37 2.26 9.37 9.98 54.7
69.0 29.5 58.2 76.2 74.4 84.0 67.1
3.2. Chemical characterizations of saccharides The content of OS and MS as well as the compositions of saccharides in all fractions were listed in Table 3. It was observed that the OS and MS accounted 87.2% and 12.8%, respectively, for the saccharides in PHL. By membrane fractionation, the OS became dominant in all retentates from R0 to R4 . However, the permeate P4 contained 27.9% of MS, 15.1% higher than that in the original PHL. As to the saccharides compositions, arabinose, galactose, glucose, xylose and mannose were derived from the hydrolysis of carbohydrate, mainly hemicellulose as indicated by the higher proportion of xylose in PHL. The results in Table 3 also suggested that every fraction was the mixture of five saccharides. This meant saccharides in PHL had a broad molecular weight distribution. Hence, it is impossible for membrane fractionation to obtain fraction composed of homogeneous saccharides. It was reported that almost each oligosaccharide chain has a reducing end on its terminal residue. Because the aldehyde or ketone group of this terminal residue is not fixed into a ring structure, it is free to undergo oxidation–reduction reactions with chemical reagents to form products that can be detected by colorimetric methods. The most widely used method for colorimetric quantification of reducing ends is the DNS assay, in which DNS reacts with reducing ends of saccharides to form red–brown 3-amino-5-nitrosalicylate, quantified by comparison of its absorbance at 560 nm. To get the accurate and reliable information of reducing ends for saccharides mixtures in PHL, DNS assay was performed on MF and UF permeates as they were. To evaluate the validity of DNS assay, all permeates were subjected to quantitative acid hydrolysis using 4% sulfuric acid which converts all saccharides into MS, and the resulting MS was determined by ion chromatography to get the quantity of sugar units. Fig. 3 gives the correlation between the quantity of reducing ends and the total saccharides as MS. Interestingly, the linear correlation in Fig. 3 indicated that the quantity of reducing ends was equal to the resulting MS, especially for saccharides in LMW permeates. But that’s obviously impossible in theory. This can be interpreted by the well documented variations of the equivalence between amino-nitrosalicylate produced and the number of reducing ends for different saccharides (Miller, 1959; Rivers, Gracheck, Woodford, & Emert, 1984). Further, the reactivity of DNS varies to OS with
Fig. 3. Correlation between the quantity of reducing ends and quantity of MS in MF/UF permeates.
different DP as demonstrated by the study of Jeffries, Yang, and Davis (1998) that DNS assay showed higher reactivity to XOS with higher DP. This may be the reason why the quantity of the reducing ends of saccharides with high molecular weight (HMW) was basically equal to the quantity of total saccharides as MS. Nevertheless, our study suggested that the DNS assay cannot be used for the determination of reducing ends of saccharides mixture with various DP in PHL. However, it seems that DNS assay provides a rapid method for determination of total saccharides in PHL as indicated by the linear correlation in Fig. 3.
Fig. 4. PhOH content in softwood kraft lignin (SKL), hardwood kraft lignin (HKL), organosolv lignin, insoluble lignin (Insol) from hot water extract of maple, and fractions of R0 ∼ P4 from membrane fractionation. The data of SKL, HKL, OL and Insol was adapted from the study of Kubo et al. (2005).
Table 3 Chemical characterization of saccharides obtained by membrane fractionation. Fraction
PHL R0 R1 R2 R3 R4 P4
Saccharides (%)
Saccharides compositions (%)
OS
MS
Arabinose
Galactose
Glucose
Xylose
Mannose
87.2 97.3 97.5 97.8 98.0 97.7 72.1
12.8 2.7 2.5 2.2 2.0 2.3 27.9
5.73 4.0 1.2 0.8 0.8 1.5 10.3
7.65 1.0 2.0 8.1 1.9 9.9 9.6
8.91 7.5 5.7 5.0 7.8 13.5 8.6
70.51 86.6 57.6 67.2 84.0 69.8 61.5
7.20 0.8 33.6 18.9 5.5 5.4 10.0
Z. Wang et al. / Carbohydrate Polymers 126 (2015) 185–191
189
Fig. 5. Extractable LMW organnic compounds in PHL detected by GC/MS, the value following * represents concentration (mg/L), the value in bracket is the abundance.
3.3. Chemical characterizations of lignin components During prehydrolysis, the chemical reaction occurred to lignin was mainly the homolytic cleavage of the aryl-ether bonds. The resulting dissolving lignin showed a strong increase in PhOH. Phenolic lignin has high valued applications in pharmacy and polymer synthesis. PhOH content in different lignin fractions was given in Fig. 4. It was found that PhOH content was dependent on the molecular weight, and further degradation resulted in higher PhOH content. Same to the insoluble lignin from hot water extract of hardwood maple (Bujanovic et al., 2012), the insoluble aspen lignin obtained from MF showed lower PhOH content than that registered for kraft and organosolv lignin (Kubo, Gilber, & Kadla, 2005). However, the increase of PhOH content of lignin in LMW fractions was quite remarkable. Based on these results, it was concluded that the lignin in PHL may have a great potential for polymer industry. The LMW organics in permeate P4 that derived from lignin degradation was determined by GC/MS after being extracted from PHL to give an insight into the nature and abundance of these compounds. Fig. 5 presents structures and concentrations determined by GC/MS. It was well known that substantial amount of p-hydroxybenzoic acid are attached to aspen lignin, primary by ester linkages (Bardet et al., 1985). As expected, it was found that phydroxybenzoic acid was dominant in PHL with the concentration of 25.65 mg/L, constituted 83.87% of all extractable organics based on their relative abundance. Syringaresinol, the second abundant organic compound in PHL, are mostly lignin derived, since it was absent in the native wood extract as reported by Goundalkar et al. (2010). Evidences of syringaresinol formation from birch lignin during acidolysis (Lundquist, 1973) and from aspen lignin during steam hydrolysis (Bardet et al., 1985) had been documented. It may be attributed to the cleavage of ␣-O-4 ether linkage in lignin, in a mild acidic environment of prehydrolysis. There were observed syringaldehyde and vanillin as well as their acids in significant amount, constituted 7.3% of the total extractable organics. Conneryl alcohol was identified in aspen PHL, however, the presence of conneryl alcohol has not been reported until now to our knowledge. Coniferyl alcohol and sinapyl alcohol in PHL were not surprising, as they are biosynthetic precursors of the three common monolignols to lignin or lignans. Also, there was observed p-hydroquinone in trace amount in the PHL, which may be an acid hydrolysis product
(a)
(b)
Fig. 6. (a) Comparison of SEC elution profiles of MF permeate P0 obtained by UV detector and RI detector. (b) RI detected SEC elution profile RI detected SEC elution profiles of UF permeates.
190
Z. Wang et al. / Carbohydrate Polymers 126 (2015) 185–191
Fig. 7. Dionex IC chromatogram of saccharides in UF permeates.
of p-hydroquinone glycoside, identified earlier in hot water extract of sugar maple (Goundalkar et al., 2010). The nature of the LMW organic compounds detected in the present study will be used to determine feasible means to sequester the extracts and provide streams of compounds that may be used as value-added chemicals. This is undoubtedly an effort to enhance and broaden the scope of the wood biorefinery of aspen. 3.4. Molecular weight distribution of saccharides and lignin components In order to isolate saccharides from lignin components as higher value added products, it is essential to assess the molecular weight distribution of these origanics. The permeates from MF and UF were analyzed by SEC for determination of molecular weight distribution using conventional calibration method. Fig. 6a gives the elution profile of organics in P0 obtained by RI detector and UV detector, respectively. The UV detected elution profile reflects the molecular weight distribution of lignin, while the RI detected elution profile reflects the molecular weight distribution of saccharides. The elution profiles showed the overlapping of saccharides and lignin molecular weight distributions, especially in the regions around 7.0 kD and 0.15 kD, and this can be helpful in interpreting the incapability of UF in saccharides purification from PHL. The UV detected profile in Fig. 6a suggested the majority of lignin molecules were smaller than saccharides. The peak around 6.0 kD and 3.0 kD was both RI and UV detectable, and may result from lignin-carbohydrate complex which had been reported by Aimi, Matsumoto, and Meshitsuka (2005). To evaluate the performance of UF on saccharides fractionation, all permeates from UF were analyzed by SEC with RI detector (Fig. 6b). There was observed a broad molecular weight distribution in all permeates with three peaks corresponding to at 32 kD, 0.55 kD and 0.15 kD. Astonishingly, HMW saccharides were found in every permeate of UF, even in P4 obtained from 1.0 kD MWCO membrane. It was concluded that UF did not improve the homogeneity of saccharides due to the disability of stopping high molecular saccharides passing through membranes. The disability may be ascribed to the linear conformation of saccharides. 3.5. DP profiles for saccharides HPAEC-PAD was applied to acquire the DP profiles of saccharides because of its novel ability in separating OS over a wide DP
range. A comparison of DP profiles of saccharides in P1 and P3 was therefore made and given in Fig. 7. By using XOS with DP from 2 to 6 as standard to quantify XOS, it was observed that XOS with DP from 2 to 6 accounted for 56.8% (wt%) of all saccharides in P1. Suppose the DP of saccharides in Fig. 7 increases gradually with the elution time, the chromatogram of P1 showed that saccharides with DP up to 12 were separated well with near baseline resolution. From the molecular weight distribution of saccharides in UF permeates given in Fig. 6b, it was deduced that a large amount of saccharides with molecular weight larger than 2.0 kD could not be detected by HPAEC-PAD. This may ascribed to the fact that the relative electrochemical responses of saccharides decreases with DP. From the DP profiles of P1 and P3 in Fig. 7, UF resulted in the distinct reduction of both lower DP saccharides and high DP saccharides. This is consistent with the results of SEC discussed above, and verified the disability of UF in improving the homogeneity of the saccharides.
4. Conclusion The present study evaluated the performance of membrane fractionation of organics in PHL based on the systematic chemical characterization of saccharides and lignin components. The results of chemical characterization indicated that saccharides in PHL consisted of 12.8% MS and 87.2% OS with XOS as the dominant component. SEC analysis showed a broad molecular weight distribution of OS with DP up to 200, while the molecular weight of lignin components was relatively low but overlaps with saccharides in the regions around 7.0 kD and 0.15 kD. The characterization of lignin components indicated a great potential for polymer industry because of the remarkable content of PhOH, especially for LMW fractions. It was concluded that the membrane filtration was nonselectivity to organics in PHL, and therefore cannot be applied in saccharides purification unless the molecules of lignin components or saccharides were modified prior to membrane fractionation.
Acknowledgements The authors would like to acknowledge financial support from the National Natural Science Foundation of China (31300492, 31370581), the Department of Science and Technology of Shandong Province (2013BSB01278), and the Yellow River Mouth Scholar Program (DYRC20120105).
Z. Wang et al. / Carbohydrate Polymers 126 (2015) 185–191
References Aimi, H., Matsumoto, Y., & Meshitsuka, G. (2005). Structure of small lignin fragments retained in water-soluble polysaccharides extracted from birch MWL isolation residue. Journal of Wood Science, 51(3), 303–308. Al Manasrah, M., Kallioinen, M., Ilvesniemi, H., & Mänttäri, M. (2012). Recovery of galactoglucomannan from wood hydrolysate using regenerated cellulose ultrafiltration membranes. Bioresource Technology, 114, 375–381. Bardet, M., Robert, D. R., & Lundquist, K. (1985). On the reactions and degradation of the lignin during steam hydrolysis of aspen wood. Svensk Papperstidning, 88(6), 61–67. Blainski, A., Lopes, G. C., & De Mello, J. C. P. (2013). Application and analysis of the folin ciocalteu method for the determination of the total phenolic content from Limonium brasiliense L. Molecules, 18(6), 6852–6865. Bujanovic, B. M., Goundalkar, M. J., & Amidon, T. E. (2012). Increasing the value of a biorefinery based on hot-water extraction: Lignin products. Tappi Journal, 11, 19–26. Buranov, A. U., & Mazza, G. (2010). Extraction and characterization of hemicelluloses from flax shives by different methods. Carbohydrate Polymers, 79(1), 17–25. Cara, C., Ruiz, E., Carvalheiro, F., Moura, P., Ballesteros, I., Castro, E., et al. (2012). Production, purification and characterisation of oligosaccharides from olive tree pruning autohydrolysis. Industrial Crops and Products, 40, 225–231. Chen, X., Wang, Z., Fu, Y., Li, Z., & Qin, M. (2014). Specific lignin precipitation for oligosaccharides recovery from hot water wood extract. Bioresource Technology, 152, 31–37. Duarte, G., Ramarao, B., & Amidon, T. (2010). Polymer induced flocculation and separation of particulates from extracts of lignocellulosic materials. Bioresource Technology, 101(22), 8526–8534. Duarte, G. V., Ramarao, B. V., Amidon, T. E., & Ferreira, P. T. (2011). Effect of hot water extraction on hardwood kraft pulp fibers (Acer saccharum, sugar maple). Industrial & Engineering Chemistry Research, 50(17), 9949–9959. Goundalkar, M. J., Bujanovic, B., & Amidon, T. E. (2010). Analysis of non-carbohydrate based low-molecular weight organic compounds dissolved during hot-water extraction of sugar maple. Cellulose Chemistry & Technology, 44(1), 27–33. Jeffries, T. W., Yang, V. W., & Davis, M. W. (1998). Comparative study of xylanase kinetics using dinitrosalicylic, arsenomolybdate, and ion chromatographic assays. Applied Biochemistry and Biotechnology, 70(1), 257–265. Kubo, S., Gilber, R. D., & Kadla, J. F. (2005). Natural fiber, biopolymers, and biocomposites. Boca Raton, FL: CRC Taylor & Francis Group. Leschinsky, M., Weber, H. K., Patt, R., & Sixta, H. (2009). Formation of insoluble components during autohydrolysis of Eucalyptus globulus. Lenzinger Berichte, 87, 16–25. Leschinsky, M., Zuckerstätter, G., Weber, H. K., Patt, R., & Sixta, H. (2008). Effect of autohydrolysis of Eucalyptus globulus wood on lignin structure. Part 1: Comparison of different lignin fractions formed during water prehydrolysis. Holzforschung, 62(6), 645–652. Liu, H., Hu, H., Jahan, M. S., & Ni, Y. (2013). Furfural formation from the pre-hydrolysis liquor of a hardwood kraft-based dissolving pulp production process. Bioresource Technology, 131, 315–320.
191
Liu, X., Fatehi, P., & Ni, Y. (2012). Removal of inhibitors from pre-hydrolysis liquor of kraft-based dissolving pulp production process using adsorption and flocculation processes. Bioresource Technology, 116, 492–496. Liu, Z., Ni, Y., Fatehi, P., & Saeed, A. (2011). Isolation and cationization of hemicelluloses from pre-hydrolysis liquor of kraft-based dissolving pulp production process. Biomass and Bioenergy, 35(5), 1789–1796. Lundquist, K. (1973). Acid degradation of lignin. Part VIII. Low molecular weight phenols from acidolysis of birch lignin. Acta Chemica Scandinavica, 27, 2597–2606. Miller, G. L. (1959). Use of dinitrosalicylic acid reagent for determination of reducing sugar. Analytical Chemistry, 31(3), 426–428. Montané, D., Nabarlatz, D., Martorell, A., Torné-Fernández, V., & Fierro, V. (2006). Removal of lignin and associated impurities from xylo-oligosaccharides by activated carbon adsorption. Industrial & Engineering Chemistry Research, 45(7), 2294–2302. Nabarlatz, D., Ebringerová, A., & Montané, D. (2007). Autohydrolysis of agricultural by-products for the production of xylo-oligosaccharides. Carbohydrate Polymers, 69(1), 20–28. Palm, M., & Zacchi, G. (2004). Separation of hemicellulosic oligomers from steamtreated spruce wood using gel filtration. Separation and Purification Technology, 36(3), 191–201. Qing, Q., Li, H., Kumar, R., & Wyman, C. E. (2013). Xylooligosaccharides production, quantification, and characterization in context of lignocellulosic biomass pretreatment. In C. E. Wyman (Ed.), Aqueous pretreatment of plant biomass for biological and chemical conversion to fuels and chemicals (pp. 391–415). Chichester, UK: John Wiley & Sons. Rivers, D. B., Gracheck, S. J., Woodford, L. C., & Emert, G. H. (1984). Limitations of the DNS assay for reducing sugars from saccharified lignocellulosics. Biotechnology and Bioengineering, 26(7), 800–802. Saeed, A., Jahan, M. S., Li, H., Liu, Z., Ni, Y., & van Heiningen, A. (2012). Mass balances of components dissolved in the pre-hydrolysis liquor of kraft-based dissolving pulp production process from Canadian hardwoods. Biomass and Bioenergy, 39, 14–19. Shen, J., Kaur, I., Baktash, M. M., He, Z., & Ni, Y. (2013). A combined process of activated carbon adsorption, ion exchange resin treatment and membrane concentration for recovery of dissolved organics in pre-hydrolysis liquor of the kraft-based dissolving pulp production process. Bioresource Technology, 127, 59–65. Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., & Templeton, D. (2006). Determination of sugars, byproducts, and degradation products in liquid fraction process samples. Golden, CO: National Renewable Energy Laboratory. Sun, X., Xu, F., Sun, R., Geng, Z., Fowler, P., & Baird, M. (2005). Characteristics of degraded hemicellulosic polymers obtained from steam exploded wheat straw. Carbohydrate Polymers, 60(1), 15–26. Tunc, M. S., & van Heiningen, A. R. (2011). Characterization and molecular weight distribution of carbohydrates isolated from the autohydrolysis extract of mixed southern hardwoods. Carbohydrate Polymers, 83(1), 8–13. Vázquez, M., Garrote, G., Alonso, J., Domínguez, H., & Parajó, J. (2005). Refining of autohydrolysis liquors for manufacturing xylooligosaccharides: Evaluation of operational strategies. Bioresource Technology, 96(8), 889–896.
Contents lists available at ScienceDirect
Carbohydrate Polymers journal homepage: www.elsevier.com/locate/carbpol
Fractionation and characterization of saccharides and lignin components in wood prehydrolysis liquor from dissolving pulp production Zhaojiang Wang a,∗ , Xiaojun Wang a , Jungang Jiang a , Yingjuan Fu a , Menghua Qin a,b,c,∗ a
Key Laboratory of Paper Science & Technology, Qilu University of Technology, Jinan 250353, China Laboratory of Organic Chemistry, Taishan University, Taian 271021, China c Huatai Group Corp. Ltd., Dongying 257335, Shandong, China b
a r t i c l e
i n f o
Article history: Received 22 July 2014 Received in revised form 28 February 2015 Accepted 3 March 2015 Available online 14 March 2015 Keywords: Oligosaccharides Lignin Fractionation Membrane Prehydrolysis
a b s t r a c t Saccharides and lignin components in prehydrolysis liquor (PHL) from kraft-based dissolving pulp production was characterized after being fractionated using membrane filtration. The results showed that the membrane filtration provided a method for organics fractionation with considerable recovery rate, but exhibited some disadvantages. Besides the limited ability in purifying oligosaccharides (OS) due to the overlaps of molecular weight distribution with lignin components, the membrane filtration could not improve the homogeneity of OS as indicated by the analysis of chemical compositions and the degree of polymerization (DP), which may be ascribed to the linear conformation of OS. The characterization of lignin components indicated a great potential for polymer industry because of the remarkable content of phenolic hydroxyl groups (PhOH), especially for low molecular weight (LMW) fraction. It was concluded the organics in PHL provided streams of value-added chemicals. However, the practical significance thereof can be realized and maximized only when they are successfully and completely fractionated. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction The prehydrolysis-kraft process currently attracts widespread attention because it not only produces high grade dissolving pulp but also provides a potential way for high-value utilization of hemicellulose (Buranov & Mazza, 2010; Nabarlatz, Ebringerová, & Montané, 2007; Sun et al., 2005). This process fits perfectly into the forest biorefinery concept if these hemicellulosic oligosaccharides (OS) in prehydrolysis liquor (PHL) could be fully isolated and purified. However, the prehydrolysis step that serves the purpose of hemicelluloses extraction is ineluctably accompanied with the depolymerization of other component of biomass, namely lignin degradation. This makes the OS separation a challenging task because of the presence of lignin components (Bujanovic, Goundalkar, & Amidon, 2012; Tunc & van Heiningen, 2011).
∗ Corresponding authors at: Key Laboratory of Paper Science & Technology, Qilu University of Technology, University Avenue No.3801, Changqing, Jinan 250353, Shandong, China. Tel.: +86 531 89631168; fax: +86 531 89631117. E-mail addresses: [email protected], [email protected] (Z. Wang), [email protected] (M. Qin). http://dx.doi.org/10.1016/j.carbpol.2015.03.012 0144-8617/© 2015 Elsevier Ltd. All rights reserved.
It was reported the lignin reactions during prehydrolysis mainly involved the homolytic cleavage of aryl ether bonds, the generation of new phenolic units and condensation reactions occurred at extremely high temperature (Bardet, Robert, & Lundquist, 1985). These reactions release a diverse palette of degradation byproducts. Liu, Ni, Fatehi, and Saeed (2011) found that the depolymerized lignin in PHL had a lower average molecular weight but a significant increase of the phenolic hydroxyl groups (PhOH) compared to dioxane lignin. Similar result was reported by Leschinsky, Zuckerstätter, Weber, Patt, and Sixta (2008) that the extensive lignin degradation occurs during prehydrolysis through homolytic cleavage of the aryl-ether bonds, resulting in a strong increase of PhOH and a decrease in aliphatic hydroxyl groups. Another study of Leschinsky, Weber, Patt, and Sixta (2009) reported that the formation of the condensed lignin in PHL is characterized by a higher molecular weight and insoluble at low temperature. His study also indicated a broad distribution of molecular weight of lignin components in PHL. Besides these lignin oligomers, a variety of low molecular weight (LMW) aromatic compounds were present as a result of the cleavage of the ␣-O-4 linkages of lignin. Goundalkar, Bujanovic, and Amidon (2010) demonstrated that vanillin and syringaldehyde were the major aromatic compounds in maple PHL. Besides these degradation byproducts of lignin, the dissolving organics in PHL
186
Z. Wang et al. / Carbohydrate Polymers 126 (2015) 185–191
also include furfurals and organic acids that are derived from carbohydrate degradation (Cara et al., 2012; Liu, Hu, Jahan, & Ni, 2013; Saeed et al., 2012), and therefore make the compositions of PHL more complicated. Various methods have been developed for OS separation by removing non-saccharide contaminants, namely lignin components, e.g. polymer flocculation (Duarte, Ramarao, & Amidon, 2010), solvent extraction (Vázquez, Garrote, Alonso, Domínguez, & Parajó, 2005), adsorption by surface active materials (Liu, Fatehi, & Ni, 2012; Montané, Nabarlatz, Martorell, Torné-Fernández, & Fierro, 2006), ion exchange or chromatography techniques (Cara et al., 2012; Palm & Zacchi, 2004), membrane filtration (Al Manasrah, Kallioinen, Ilvesniemi, & Mänttäri, 2012; Buranov & Mazza, 2010), and combination of these techniques (Chen, Wang, Fu, Li, & Qin, 2014; Shen, Kaur, Baktash, He, & Ni, 2013). However, their performances remain to be improved and currently far from the requirement of industrial application in terms of process cost and recovery yield as reviewed by Qing, Li, Kumar, and Wyman (2013). Although the separation strategies of OS from PHL were extensively studied, the characterization of the dissolving organics is rather limited. It is believed that a better understanding of the molecular information about the dissolving organics in PHL is of vital importance for biorefinery and will undoubtedly facilitate the OS separation. The present work addresses the fractionation and characterization of organic compounds in PHL. This is an effort not only to provide fundamental information beneficial to OS purification, but also to wood biorefinery. 2. Materials and methods Fig. 1. Flow diagram for the fractionation of PHL by MF and UF.
2.1. Materials The fast-growing aspen species, Populus × euramericana ‘Neva’, was selected as raw materials in the present study because it is widely used in dissolving pulp mills through prehydrolysis kraft process in North China. Poplar wood chips were prepared from debarked wood logs harvested from the southwest region of Shandong province, China. Microporous membrane made of mixed cellulose esters with pore size of 0.45 m and Ultracel regenerated cellulose membrane with molecular weight cut off (MWCO) 1.0, 3.0, 10.0, 30.0 kD were provided by Merck Millipore, Billerica, MA. All chemicals were analytical-reagent grade from Sigma-Aldrich, Inc. The chemicals included 2 N Folin–Ciocalteu’s phenol reagent and 3,5-dinitrosalicylic acid (DNS), anhydrous sodium carbonate, and 4-hydroxybenzoic acid. Dextran with weight average molecular weight from 0.18 to 36.3 kD (American Polymer Standards Corporation, Mentor, OH) was used as standard samples for the size exclusion chromatography (SEC) analysis. Xylose and xylooligosaccharides (XOS) purchased from Megzyme Ireland with degree of polymerization (DP) from 2 to 6 were used as standard for high performance anion exchange chromatography with pulsed amperometric detector (HPAEC-PAD) to get DP profile of saccharides in PHL.
was withdrawn. The composition of aspen wood before and after treatment as well as the PHL composition were listed in Table 1. 2.3. Fractionation by membrane filtration Batch filtration processing was conducted to fractionate the organics in PHL according to the scheme in Fig. 1. Prior to ultrafiltration (UF), the PHL was subjected to microfiltration (MF) using 0.45 m membrane to eliminate insoluble particles. The UF was performed in dead-end mode using a stirred cell (Millipore) at ambient temperature and 0.3 MPa by compressed nitrogen. Flat sheet Millipore membranes with an effective membrane area of 40 cm2 were employed. A magnetic stirrer bar was employed at a stirring rate of 100 rpm. Reverse stirring was also applied every 10 min to ensure that the feed solution was well mixed. The nominal MWCO of membranes used was 30, 10, 3 and 1 kD, respectively. The PHL starting volume was approximately 150 mL. In each UF step, the volume concentration ratio was controlled to be 10, and the permeate was collected and filtered using the next lower MWCO membrane. 2.4. Analytical methods
2.2. Prehydrolysis and PHL preparation The prehydrolysis was carried out in a 23 L pulp digester, using 1.0 kg of oven-dried poplar chips. Deionized water was added in order to reach 6:1 of liquid to wood ratio. The digester temperature was ramped from room temperature to 170 ◦ C and held for 60 min. This is a representative reaction severity widely applied in industrial process that corresponds to P-factor of 550 (Duarte, Ramarao, Amidon, & Ferreira, 2011). At the end of prehydrolysis, the digester was cooled, depressurized and the reaction mixture
The reducing sugar in solution was determined using DNS method (Miller, 1959). While a strict definition of an OS is not established, saccharides with DP greater than 2 but soluble in PHL were considered as OS in present study. The OS content was measured by an indirect method based on quantitative acid hydrolysis of the liquid samples with 4% w/w of H2 SO4 at 121 ◦ C for 60 min according to technical report from NREL (Sluiter et al., 2006). The OS concentration was expressed as the increase of monosaccharides (MS) which were determined by HPAEC-PAD. This system
Z. Wang et al. / Carbohydrate Polymers 126 (2015) 185–191
187
Table 1 Chemical compositions of aspen wood before and after prehydrolysis as well as the compositions of the PHL. Untreated wood (1.0 kg)
Residual solid (0.81 kg)
PHL (6.0 L)a Oligosaccharides (18.89 g/L)
Arabinan 0.38% Galactan 0.71% Glucan 40.7% Xylan 16.4% Mannan 3.8% Lignin 23.5%
a b c d e f g h i
Arabinan ndb Galactan nd Glucan 49.2% Xylan 5.3% Mannan 1.7% Lignin 26.1%
AOSc 5.7%d GalOSe 7.6% GluOSf 8.9% XOSg 70.5% MOSh 7.2%
Monosaccharides (2.46 g/L) Arabinose 24.9% Galactose 15.9% Glucose 10.8% Xylose 40.2% Mannose 8.1%
Non-saccharides (15.34 g/L) Lignin 7.12 g/L Formic acid 2.23 g/L Acetic acid 4.10 g/L Furfural 1.64 g/L HMFi 0.25 g/L
Theoretical volume of PHL. Not detected. Arabinooligosaccharides. All wt%. Galactooligosaccharides. Glucooligosaccharides. Xylooligosaccharides. Mannooligosaccharides. 5-Hydroxymethyl furfural.
consisted of a Dionex (Sunnyvale, CA) HPLC system (ICS-5000) equipped with a GP40 gradient pump, an anion exchange column (CarboPac PA20 and a guard) and an ED40 electrochemical detector. Aliquots (25 L) were injected manually after passing through a 0.2 m nylon syringe filter. Gradient was set as 2 mM NaOH isocratic with a step to 200 mM NaOH at 10 min to regenerate the column at a flow rate of 0.5 mL/min and 25 ◦ C. MS were quantified with reference to standards using the same analytical procedure. The DP profile of OS was obtained by HPAEC-PAD system same to MS determination, but with different column (CarboPac PA1 and a guard) and elution conditions. Gradient for DP profile of OS was set as 0–300 mM NaOAc in 100 mM NaOH from 0 to 30 min at a flow rate of 0.5 mL/min and 25 ◦ C. The column was reconditioned using 100 mM NaOH after each analysis. The degradation byproducts of carbohydrate in HPL such as formic acid, acetic acid, furfural and HMF were analyzed by HPLC equipped with Waters C18 symmetry column (4.6 × 150 mm, 5 m) at 30 ◦ C with 0.1% H3 PO4 (v/v) as eluent at 0.5 mL/min. All samples were analyzed in duplicates. The molecule weight distribution of the OS and lignin in was analyzed by SEC using a HPLC system comprised of Shimadzu LC-20T, Shimadzu SB-803 HQ column (8 × 300 mm, 6 m), a UV detector and a refraction index (RI) detector with pure water as mobile phase at 0.5 mL/min and 35 ◦ C. The lignin content was measured according to a spectrometric method at a wavelength of 205 nm (TAPPI UM-250). The PhOH content of lignin was determined according to Folin–Ciocalteau assay (Blainski, Lopes, & De Mello, 2013). All samples were analyzed in duplicates. The LMW degradation byproducts of lignin in PHL were determined by gas chromatography mass spectrometry (GC/MS) after being extracted using methyl tert-butyl ether (MTBE) as solvent. The extraction procedures involved introducing 15 mL MTBE into 30 mL PHL, shaking generally for 5 min, and the collection of extract. Extraction was repeated for three times for a better separation. The extract containing LMW degradation byproducts of lignin was concentrated and subjected to GC/MS analysis using Agilent 7890 GC equipped with Agilent 5975 C detector. A capillary column HP-1 (6–7 m × 0.53 mm, 0.15 m) was used. The temperature program used during the GC/MS analysis was as follows: starting from 40 ◦ C, ramp to 280 ◦ C in 24 min and hold for 20 min. The mass spectra were recorded in the interval 35–250 amu. Organics were identified by comparing their spectra with those available in the NIST libraries.
3. Results and discussion 3.1. Distribution of saccharides and lignin components after fractionation The recovery rate after fractionation was determined for lignin and saccharides (Fig. 2). The recovery rate of saccharides was 85%, slightly less than that of lignin. This suggested that the saccharides may contributed more than lignin to the fouling of membranes. The relative yield of saccharides and lignin among individual fractions after MF and UF is given in Table 2. It was observed that more than half of organics could pass through the 1 kD membrane and present in fraction P4 . MF is effective in lignin separation as indicated by the lowest content of saccharides in fraction R0 . For UF treatment, the relative yield of saccharides showed an increase from 4.21% in R1 to 24.43% in R4 , while the variation of the lignin yield was not significant. This led to highest saccharides content of 84.03% in R4 . However, because more than half saccharides was found in P4 , the basically unchanged saccharides content in P4 and original PHL represents a drawback for UF in saccharides purification.
Fig. 2. Recovery rate of saccharides and lignin in fractions obtained from membrane filtration.
188
Z. Wang et al. / Carbohydrate Polymers 126 (2015) 185–191
Table 2 Relative yield and saccharides content in all fractions. Fractions
PHL R0 R1 R2 R3 R4 P4
Relative yield (%)
Saccharides content (%)
Saccharides
Lignin
3.38 4.12 3.36 12.7 24.4 51.9
17.3 6.37 2.26 9.37 9.98 54.7
69.0 29.5 58.2 76.2 74.4 84.0 67.1
3.2. Chemical characterizations of saccharides The content of OS and MS as well as the compositions of saccharides in all fractions were listed in Table 3. It was observed that the OS and MS accounted 87.2% and 12.8%, respectively, for the saccharides in PHL. By membrane fractionation, the OS became dominant in all retentates from R0 to R4 . However, the permeate P4 contained 27.9% of MS, 15.1% higher than that in the original PHL. As to the saccharides compositions, arabinose, galactose, glucose, xylose and mannose were derived from the hydrolysis of carbohydrate, mainly hemicellulose as indicated by the higher proportion of xylose in PHL. The results in Table 3 also suggested that every fraction was the mixture of five saccharides. This meant saccharides in PHL had a broad molecular weight distribution. Hence, it is impossible for membrane fractionation to obtain fraction composed of homogeneous saccharides. It was reported that almost each oligosaccharide chain has a reducing end on its terminal residue. Because the aldehyde or ketone group of this terminal residue is not fixed into a ring structure, it is free to undergo oxidation–reduction reactions with chemical reagents to form products that can be detected by colorimetric methods. The most widely used method for colorimetric quantification of reducing ends is the DNS assay, in which DNS reacts with reducing ends of saccharides to form red–brown 3-amino-5-nitrosalicylate, quantified by comparison of its absorbance at 560 nm. To get the accurate and reliable information of reducing ends for saccharides mixtures in PHL, DNS assay was performed on MF and UF permeates as they were. To evaluate the validity of DNS assay, all permeates were subjected to quantitative acid hydrolysis using 4% sulfuric acid which converts all saccharides into MS, and the resulting MS was determined by ion chromatography to get the quantity of sugar units. Fig. 3 gives the correlation between the quantity of reducing ends and the total saccharides as MS. Interestingly, the linear correlation in Fig. 3 indicated that the quantity of reducing ends was equal to the resulting MS, especially for saccharides in LMW permeates. But that’s obviously impossible in theory. This can be interpreted by the well documented variations of the equivalence between amino-nitrosalicylate produced and the number of reducing ends for different saccharides (Miller, 1959; Rivers, Gracheck, Woodford, & Emert, 1984). Further, the reactivity of DNS varies to OS with
Fig. 3. Correlation between the quantity of reducing ends and quantity of MS in MF/UF permeates.
different DP as demonstrated by the study of Jeffries, Yang, and Davis (1998) that DNS assay showed higher reactivity to XOS with higher DP. This may be the reason why the quantity of the reducing ends of saccharides with high molecular weight (HMW) was basically equal to the quantity of total saccharides as MS. Nevertheless, our study suggested that the DNS assay cannot be used for the determination of reducing ends of saccharides mixture with various DP in PHL. However, it seems that DNS assay provides a rapid method for determination of total saccharides in PHL as indicated by the linear correlation in Fig. 3.
Fig. 4. PhOH content in softwood kraft lignin (SKL), hardwood kraft lignin (HKL), organosolv lignin, insoluble lignin (Insol) from hot water extract of maple, and fractions of R0 ∼ P4 from membrane fractionation. The data of SKL, HKL, OL and Insol was adapted from the study of Kubo et al. (2005).
Table 3 Chemical characterization of saccharides obtained by membrane fractionation. Fraction
PHL R0 R1 R2 R3 R4 P4
Saccharides (%)
Saccharides compositions (%)
OS
MS
Arabinose
Galactose
Glucose
Xylose
Mannose
87.2 97.3 97.5 97.8 98.0 97.7 72.1
12.8 2.7 2.5 2.2 2.0 2.3 27.9
5.73 4.0 1.2 0.8 0.8 1.5 10.3
7.65 1.0 2.0 8.1 1.9 9.9 9.6
8.91 7.5 5.7 5.0 7.8 13.5 8.6
70.51 86.6 57.6 67.2 84.0 69.8 61.5
7.20 0.8 33.6 18.9 5.5 5.4 10.0
Z. Wang et al. / Carbohydrate Polymers 126 (2015) 185–191
189
Fig. 5. Extractable LMW organnic compounds in PHL detected by GC/MS, the value following * represents concentration (mg/L), the value in bracket is the abundance.
3.3. Chemical characterizations of lignin components During prehydrolysis, the chemical reaction occurred to lignin was mainly the homolytic cleavage of the aryl-ether bonds. The resulting dissolving lignin showed a strong increase in PhOH. Phenolic lignin has high valued applications in pharmacy and polymer synthesis. PhOH content in different lignin fractions was given in Fig. 4. It was found that PhOH content was dependent on the molecular weight, and further degradation resulted in higher PhOH content. Same to the insoluble lignin from hot water extract of hardwood maple (Bujanovic et al., 2012), the insoluble aspen lignin obtained from MF showed lower PhOH content than that registered for kraft and organosolv lignin (Kubo, Gilber, & Kadla, 2005). However, the increase of PhOH content of lignin in LMW fractions was quite remarkable. Based on these results, it was concluded that the lignin in PHL may have a great potential for polymer industry. The LMW organics in permeate P4 that derived from lignin degradation was determined by GC/MS after being extracted from PHL to give an insight into the nature and abundance of these compounds. Fig. 5 presents structures and concentrations determined by GC/MS. It was well known that substantial amount of p-hydroxybenzoic acid are attached to aspen lignin, primary by ester linkages (Bardet et al., 1985). As expected, it was found that phydroxybenzoic acid was dominant in PHL with the concentration of 25.65 mg/L, constituted 83.87% of all extractable organics based on their relative abundance. Syringaresinol, the second abundant organic compound in PHL, are mostly lignin derived, since it was absent in the native wood extract as reported by Goundalkar et al. (2010). Evidences of syringaresinol formation from birch lignin during acidolysis (Lundquist, 1973) and from aspen lignin during steam hydrolysis (Bardet et al., 1985) had been documented. It may be attributed to the cleavage of ␣-O-4 ether linkage in lignin, in a mild acidic environment of prehydrolysis. There were observed syringaldehyde and vanillin as well as their acids in significant amount, constituted 7.3% of the total extractable organics. Conneryl alcohol was identified in aspen PHL, however, the presence of conneryl alcohol has not been reported until now to our knowledge. Coniferyl alcohol and sinapyl alcohol in PHL were not surprising, as they are biosynthetic precursors of the three common monolignols to lignin or lignans. Also, there was observed p-hydroquinone in trace amount in the PHL, which may be an acid hydrolysis product
(a)
(b)
Fig. 6. (a) Comparison of SEC elution profiles of MF permeate P0 obtained by UV detector and RI detector. (b) RI detected SEC elution profile RI detected SEC elution profiles of UF permeates.
190
Z. Wang et al. / Carbohydrate Polymers 126 (2015) 185–191
Fig. 7. Dionex IC chromatogram of saccharides in UF permeates.
of p-hydroquinone glycoside, identified earlier in hot water extract of sugar maple (Goundalkar et al., 2010). The nature of the LMW organic compounds detected in the present study will be used to determine feasible means to sequester the extracts and provide streams of compounds that may be used as value-added chemicals. This is undoubtedly an effort to enhance and broaden the scope of the wood biorefinery of aspen. 3.4. Molecular weight distribution of saccharides and lignin components In order to isolate saccharides from lignin components as higher value added products, it is essential to assess the molecular weight distribution of these origanics. The permeates from MF and UF were analyzed by SEC for determination of molecular weight distribution using conventional calibration method. Fig. 6a gives the elution profile of organics in P0 obtained by RI detector and UV detector, respectively. The UV detected elution profile reflects the molecular weight distribution of lignin, while the RI detected elution profile reflects the molecular weight distribution of saccharides. The elution profiles showed the overlapping of saccharides and lignin molecular weight distributions, especially in the regions around 7.0 kD and 0.15 kD, and this can be helpful in interpreting the incapability of UF in saccharides purification from PHL. The UV detected profile in Fig. 6a suggested the majority of lignin molecules were smaller than saccharides. The peak around 6.0 kD and 3.0 kD was both RI and UV detectable, and may result from lignin-carbohydrate complex which had been reported by Aimi, Matsumoto, and Meshitsuka (2005). To evaluate the performance of UF on saccharides fractionation, all permeates from UF were analyzed by SEC with RI detector (Fig. 6b). There was observed a broad molecular weight distribution in all permeates with three peaks corresponding to at 32 kD, 0.55 kD and 0.15 kD. Astonishingly, HMW saccharides were found in every permeate of UF, even in P4 obtained from 1.0 kD MWCO membrane. It was concluded that UF did not improve the homogeneity of saccharides due to the disability of stopping high molecular saccharides passing through membranes. The disability may be ascribed to the linear conformation of saccharides. 3.5. DP profiles for saccharides HPAEC-PAD was applied to acquire the DP profiles of saccharides because of its novel ability in separating OS over a wide DP
range. A comparison of DP profiles of saccharides in P1 and P3 was therefore made and given in Fig. 7. By using XOS with DP from 2 to 6 as standard to quantify XOS, it was observed that XOS with DP from 2 to 6 accounted for 56.8% (wt%) of all saccharides in P1. Suppose the DP of saccharides in Fig. 7 increases gradually with the elution time, the chromatogram of P1 showed that saccharides with DP up to 12 were separated well with near baseline resolution. From the molecular weight distribution of saccharides in UF permeates given in Fig. 6b, it was deduced that a large amount of saccharides with molecular weight larger than 2.0 kD could not be detected by HPAEC-PAD. This may ascribed to the fact that the relative electrochemical responses of saccharides decreases with DP. From the DP profiles of P1 and P3 in Fig. 7, UF resulted in the distinct reduction of both lower DP saccharides and high DP saccharides. This is consistent with the results of SEC discussed above, and verified the disability of UF in improving the homogeneity of the saccharides.
4. Conclusion The present study evaluated the performance of membrane fractionation of organics in PHL based on the systematic chemical characterization of saccharides and lignin components. The results of chemical characterization indicated that saccharides in PHL consisted of 12.8% MS and 87.2% OS with XOS as the dominant component. SEC analysis showed a broad molecular weight distribution of OS with DP up to 200, while the molecular weight of lignin components was relatively low but overlaps with saccharides in the regions around 7.0 kD and 0.15 kD. The characterization of lignin components indicated a great potential for polymer industry because of the remarkable content of PhOH, especially for LMW fractions. It was concluded that the membrane filtration was nonselectivity to organics in PHL, and therefore cannot be applied in saccharides purification unless the molecules of lignin components or saccharides were modified prior to membrane fractionation.
Acknowledgements The authors would like to acknowledge financial support from the National Natural Science Foundation of China (31300492, 31370581), the Department of Science and Technology of Shandong Province (2013BSB01278), and the Yellow River Mouth Scholar Program (DYRC20120105).
Z. Wang et al. / Carbohydrate Polymers 126 (2015) 185–191
References Aimi, H., Matsumoto, Y., & Meshitsuka, G. (2005). Structure of small lignin fragments retained in water-soluble polysaccharides extracted from birch MWL isolation residue. Journal of Wood Science, 51(3), 303–308. Al Manasrah, M., Kallioinen, M., Ilvesniemi, H., & Mänttäri, M. (2012). Recovery of galactoglucomannan from wood hydrolysate using regenerated cellulose ultrafiltration membranes. Bioresource Technology, 114, 375–381. Bardet, M., Robert, D. R., & Lundquist, K. (1985). On the reactions and degradation of the lignin during steam hydrolysis of aspen wood. Svensk Papperstidning, 88(6), 61–67. Blainski, A., Lopes, G. C., & De Mello, J. C. P. (2013). Application and analysis of the folin ciocalteu method for the determination of the total phenolic content from Limonium brasiliense L. Molecules, 18(6), 6852–6865. Bujanovic, B. M., Goundalkar, M. J., & Amidon, T. E. (2012). Increasing the value of a biorefinery based on hot-water extraction: Lignin products. Tappi Journal, 11, 19–26. Buranov, A. U., & Mazza, G. (2010). Extraction and characterization of hemicelluloses from flax shives by different methods. Carbohydrate Polymers, 79(1), 17–25. Cara, C., Ruiz, E., Carvalheiro, F., Moura, P., Ballesteros, I., Castro, E., et al. (2012). Production, purification and characterisation of oligosaccharides from olive tree pruning autohydrolysis. Industrial Crops and Products, 40, 225–231. Chen, X., Wang, Z., Fu, Y., Li, Z., & Qin, M. (2014). Specific lignin precipitation for oligosaccharides recovery from hot water wood extract. Bioresource Technology, 152, 31–37. Duarte, G., Ramarao, B., & Amidon, T. (2010). Polymer induced flocculation and separation of particulates from extracts of lignocellulosic materials. Bioresource Technology, 101(22), 8526–8534. Duarte, G. V., Ramarao, B. V., Amidon, T. E., & Ferreira, P. T. (2011). Effect of hot water extraction on hardwood kraft pulp fibers (Acer saccharum, sugar maple). Industrial & Engineering Chemistry Research, 50(17), 9949–9959. Goundalkar, M. J., Bujanovic, B., & Amidon, T. E. (2010). Analysis of non-carbohydrate based low-molecular weight organic compounds dissolved during hot-water extraction of sugar maple. Cellulose Chemistry & Technology, 44(1), 27–33. Jeffries, T. W., Yang, V. W., & Davis, M. W. (1998). Comparative study of xylanase kinetics using dinitrosalicylic, arsenomolybdate, and ion chromatographic assays. Applied Biochemistry and Biotechnology, 70(1), 257–265. Kubo, S., Gilber, R. D., & Kadla, J. F. (2005). Natural fiber, biopolymers, and biocomposites. Boca Raton, FL: CRC Taylor & Francis Group. Leschinsky, M., Weber, H. K., Patt, R., & Sixta, H. (2009). Formation of insoluble components during autohydrolysis of Eucalyptus globulus. Lenzinger Berichte, 87, 16–25. Leschinsky, M., Zuckerstätter, G., Weber, H. K., Patt, R., & Sixta, H. (2008). Effect of autohydrolysis of Eucalyptus globulus wood on lignin structure. Part 1: Comparison of different lignin fractions formed during water prehydrolysis. Holzforschung, 62(6), 645–652. Liu, H., Hu, H., Jahan, M. S., & Ni, Y. (2013). Furfural formation from the pre-hydrolysis liquor of a hardwood kraft-based dissolving pulp production process. Bioresource Technology, 131, 315–320.
191
Liu, X., Fatehi, P., & Ni, Y. (2012). Removal of inhibitors from pre-hydrolysis liquor of kraft-based dissolving pulp production process using adsorption and flocculation processes. Bioresource Technology, 116, 492–496. Liu, Z., Ni, Y., Fatehi, P., & Saeed, A. (2011). Isolation and cationization of hemicelluloses from pre-hydrolysis liquor of kraft-based dissolving pulp production process. Biomass and Bioenergy, 35(5), 1789–1796. Lundquist, K. (1973). Acid degradation of lignin. Part VIII. Low molecular weight phenols from acidolysis of birch lignin. Acta Chemica Scandinavica, 27, 2597–2606. Miller, G. L. (1959). Use of dinitrosalicylic acid reagent for determination of reducing sugar. Analytical Chemistry, 31(3), 426–428. Montané, D., Nabarlatz, D., Martorell, A., Torné-Fernández, V., & Fierro, V. (2006). Removal of lignin and associated impurities from xylo-oligosaccharides by activated carbon adsorption. Industrial & Engineering Chemistry Research, 45(7), 2294–2302. Nabarlatz, D., Ebringerová, A., & Montané, D. (2007). Autohydrolysis of agricultural by-products for the production of xylo-oligosaccharides. Carbohydrate Polymers, 69(1), 20–28. Palm, M., & Zacchi, G. (2004). Separation of hemicellulosic oligomers from steamtreated spruce wood using gel filtration. Separation and Purification Technology, 36(3), 191–201. Qing, Q., Li, H., Kumar, R., & Wyman, C. E. (2013). Xylooligosaccharides production, quantification, and characterization in context of lignocellulosic biomass pretreatment. In C. E. Wyman (Ed.), Aqueous pretreatment of plant biomass for biological and chemical conversion to fuels and chemicals (pp. 391–415). Chichester, UK: John Wiley & Sons. Rivers, D. B., Gracheck, S. J., Woodford, L. C., & Emert, G. H. (1984). Limitations of the DNS assay for reducing sugars from saccharified lignocellulosics. Biotechnology and Bioengineering, 26(7), 800–802. Saeed, A., Jahan, M. S., Li, H., Liu, Z., Ni, Y., & van Heiningen, A. (2012). Mass balances of components dissolved in the pre-hydrolysis liquor of kraft-based dissolving pulp production process from Canadian hardwoods. Biomass and Bioenergy, 39, 14–19. Shen, J., Kaur, I., Baktash, M. M., He, Z., & Ni, Y. (2013). A combined process of activated carbon adsorption, ion exchange resin treatment and membrane concentration for recovery of dissolved organics in pre-hydrolysis liquor of the kraft-based dissolving pulp production process. Bioresource Technology, 127, 59–65. Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., & Templeton, D. (2006). Determination of sugars, byproducts, and degradation products in liquid fraction process samples. Golden, CO: National Renewable Energy Laboratory. Sun, X., Xu, F., Sun, R., Geng, Z., Fowler, P., & Baird, M. (2005). Characteristics of degraded hemicellulosic polymers obtained from steam exploded wheat straw. Carbohydrate Polymers, 60(1), 15–26. Tunc, M. S., & van Heiningen, A. R. (2011). Characterization and molecular weight distribution of carbohydrates isolated from the autohydrolysis extract of mixed southern hardwoods. Carbohydrate Polymers, 83(1), 8–13. Vázquez, M., Garrote, G., Alonso, J., Domínguez, H., & Parajó, J. (2005). Refining of autohydrolysis liquors for manufacturing xylooligosaccharides: Evaluation of operational strategies. Bioresource Technology, 96(8), 889–896.
