Hemicellulose Composition in Different Cell Wall Fractions Obtained using a DMSO/LiCl Wood Solvent System and Enzyme Hydrolysis

In our previous studies on the relationships between lignin structure and hemicellulose composition in a wide range of hardwood species, we showed that the higher the syringyl ratio, the higher the xylan/mannan ratio. In this study, the relationship of the syringyl ratio and xylan/mannan ratio was further studied using fractions obtained by stepwise extraction of finely milled beech wood by use of aqueous dioxane and subsequent dimethyl sulfoxide (DMSO)/LiCl solvents with increasing LiCl concentration. Each soluble fraction obtained with a different LiCl concentration was compared with the corresponding insoluble residue, and it was shown that both the syringyl ratio and xylan/mannan ratio were always higher in the insoluble fraction. To analyze polysaccharides closely located with lignin, milled wood lignin (MWL), and cellulolytic enzyme lignin (CEL) obtained by enzyme treatment after DMSO/LiCl swelling were investigated. Although the xylan/mannan ratio of MWL was higher, that of CEL with more than 80% lignin yield was lower.     

Separation and Structural Characterization of Lignin from Hybrid Poplar Based on Complete Dissolution in DMSO/LiCl

In this paper, the physicochemical properties and structural features of six lignin preparations separated with a total dissolution of ball-milled wood in dimethyl sulfoxide and lithium chloride (DMSO/LiCl) followed by extraction with ethanol/water were investigated. These isolated lignin fractions were characterized using wet chemical analysis, FT-IR, and ¹H and¹³C-NMR techniques. Experimental results showed that separated lignin preparations were relatively free of associated polysaccharides. These lignin fractions were classified as guaiacyl-syringyl lignin type: mainly composed of guaiacyl units with noticeable amounts of syringyl units and fewer p-hydroxyphenyl units. The molar ratio of non-condensed guaiacyl units to syringyl units (G/S) decreased as the ratio of LiCl to poplar weight increased. The results also showed that these lignin preparations consisted mainly of β-O-4 ether bonds combined with small quantities of β-β′ and β-5 carbon-carbon linkages. Furthermore, considerable amounts of esterified p-hydroxybenzoic acids and minor amounts of esterified p-coumaric acid were also detected in these lignin fractions.     

Fractional Separation and Structural Characterization of Chlorophyll and Lignin from Perennial Ryegrass (L. perenne) and Cocksfoot Grass (D. glomerata)

Abstract

One chlorophyll rich fraction and two lignin preparations were separated from perennial grass and cocksfoot grass by sequential three‐stage treatments with 80% ethanol containing 0.2% NaOH, 2.5% H₂O₂?0.2% EDTA containing 1.5% NaOH, and 2.5% H₂O₂?0.2% TAED containing 1.0% NaOH at 75°C for 3 h, respectively, which released 4.6 and 3.6% chlorophyll rich fraction, 2.3 and 5.8%, and 0.9 and 1.0% lignin preparations, except for releasing 8.0 and 10.4%, 79.1 and 77.0%, and 12.9 and 12.5% of the original hemicelluloses, respectively. The lignin fractions obtained from the two different grasses had very similar molecular weights and structural composition. The NMR spectra of the lignin preparations revealed the presence of p‐hydroxyphenyl, guaiacyl, and syringyl structures, and the lignin in chlorophyll rich fraction contained more guaiacyl and syringyl units than p‐hydroxyphenyl unit, whereas the reverse trend was found in the two lignin preparations. The lignin preparations are distinguished with straw and wood lignins by relatively higher contents of p‐hydroxyphenyl unit and lower amounts of condensed units (β‐5 and 5‐5′) and resinol units (β‐β). This difference in distribution of structural units indicated some structural heterogeneity between grass and straw/or wood lignin.     

Rapid Hydrothermolysis of Poplar Wood: Comparison of Sapwood,Heartwood, Bark,and Isolated Lignin

Abstract

Rapid hydrothermolysis at 350°C of poplar sapwood, heartwood and bark, as well as the lignin extracted from the sapwood and heartwood, gives oil and water soluble fractions whose chemical analysis is correlated with the nature of the feed stock. Results from cellulose and Douglas fir are included, and the poplar bark and heartwood lignin are shown to give an ether soluble oil that consists mainly of guaiacyl units while the sapwood product is mainly syringyl. However, the chloroform soluble oil from both sapwood and heartwood was mainly syringyl with the additional presence of carbohydrate derived material in the heartwood oil. The acetone-only soluble residue from sapwood was shown by pyrolysis mass spectrometry to be mainly cellulose derived while that from heartwood was mainly from lignin. Thus there is a fundamental difference in the behaviour of the poplar sapwood and heartwood that may arise from different distributions of the guaiacyl and syringyl units, and of the binding to the cellulose. The origin of the rapid hydrothermolysis products with respect to the major wood components is discussed. An empirical approach to the calculation of the lignin content based on the rapid hydrothermolysis fractions is presented.     

Fractional Separation of Hemicelluloses and Lignin in High Yield and Purity from Mild Ball-Milled Periploca sepium

Abstract

A novel three-step procedure for separation of hemicelluloses and lignin with high yield and purity was proposed in this study, where wood is mildly milled and successively extracted to produce three hemicellulosic and lignin fractions representing the total hemicelluloses and lignin in wood. The sequential treatments of the mild ball-milled Periploca sepium with 80% aqueous dioxane containing 0.05 M HCl at 85°C for 4 h, DMSO at 85°C for 4 h, and 8% NaOH at 50°C for 3 h resulted in a total release of over 85% of the original hemicelluloses and 86% of the original lignin. In particular, approximately 36% of the original hemicelluloses and 50% of the original lignin were separated during the first mild acidolytic hydrolysis process after low intensity milling. The structure of the acidic dioxane-, DMSO-, and alkali-soluble hemicellulosic and lignin fractions were elucidated using wet chemical analysis, FT-IR, and solution-state ¹H, ¹³C, and ³¹P NMR techniques. Results showed that both the mild ball milling and the mild acidolysis under the conditions given did not affect the separated lignin macromolecular structure. On the other hand, the mild acidolytic hydrolysis condition did cause substantial hemicellulosic depolymerization exception for a significant cleavage the ether linkages between lignin and hemicelluloses. The acidic dioxane-soluble lignin fraction was structurally different from the DMSO- and alkali-soluble lignin preparations and may originate mainly from the primary wall, while the alkali-soluble lignin preparation was mainly released from the secondary wall of Periploca sepium. Furthermore, it was found that the acidic dioxane-soluble hemicelluloses mainly contained more branched and less acidic arabinoxylans, and the 8% NaOH-soluble hemicellulosic fraction H₃ was both less branched and less acidic in structure, whereas the DMSO-soluble hemicelluloses were more acidic but less branched and consisted mainly of 4-O-methylglucuronoarabinoxylan.     

Characteristics and Reactivity of Lignin in Acacia and Eucalyptus Woods

The chemical characteristics of lignin, including the Klason lignin content, acid-soluble lignin content, proportion of the aromatic ring types [syringyl ratio = syringyl/(syringyl + guaiacyl)], and proportion of the diastereomeric forms for the β-O-4 structure [erythro ratio = erythro/(erythro + threo)], and pulpability were investigated for 28 wood samples belonging to the genera Acacia and Eucalyptus. Although the lignin characteristics of these 28 woods varied widely, the chemical characteristics of the two genera could be categorized in two clearly distinguished groups on the basis of the syringyl ratio. Clear negative correlations were observed between both the syringyl and erythro ratios and the total lignin content (sum of the Klason and acid-soluble lignin contents) within each genus. In addition, the syringyl ratio correlated positively with the erythro ratio and acid-soluble lignin content, regardless of the genus. The existence of a clear, high correlation between the syringyl and erythro ratios supports the hypothesis that the aromatic ring type (syringyl ratio) is a decisive factor for controlling the diastereomeric forms of the β-O-4 structure (erythro ratio). Each of the wood samples was also subjected to Kraft pulping, and it was demonstrated that the woods with higher syringyl ratios were easier to delignify. This tendency is reasonably attributed to the high reactivity of the erythro-rich and syringyl-rich β-O-4 structures, and the low lignin content of these syringyl-rich woods.     

Analysis of Lignin Aromatic Structure in Wood Fractions Based on IR Spectroscopy

In our previous studies, a method of quantitative analysis of lignin structure by IR spectroscopy with direct use of wood meal was developed which can be apply to a wide range of wood species. Lignin structure deposition in different fractions in the same wood sample also has characteristic diversity and differences. In this study, quantitative analysis of lignin structure by IR spectroscopy was used to determine whether it can be applied to different fractions in the same wood sample. Finely milled beech wood was fractionated by stepwise extraction using aqueous dioxane and followed by subsequent dimethyl sulfoxide (DMSO)/LiCl solvents with increasing LiCl concentrations. The results showed that fractions of the peak area ratio of log(1595/1509) fit a linear relationship with different wood species very well. Therefore, log(1595/1509) could be used not only with different wood species but also with different fractions from the same wood. For 1509/1460, log(1275/1220), 1130/1030 and 835/(835+815), different fractions deviated from the original relationship obtained from different wood species.     

Long‐term lime pretreatment of poplar wood

Long‐term lime pretreatment has proven to increase digestibility of many herbaceous lignocellulose sources; but until this work, its effects had not been evaluated on wood, whose lignin content is higher, and therefore, more recalcitrant to enzymatic hydrolysis. In this study, the mild conditions of long‐term lime pretreatment (1‐atm pressure, temperatures ranging from 25 to 75°C, and reaction times between 1 and 12 weeks, with and without air) were systematically applied to poplar wood available in two batches with different lignin contents. These batches were designated as low‐lignin biomass (LLB) with lignin content of 21.4% and high‐lignin biomass (HLB) with lignin content of 29.1%. Full factorial designs resulted in 79 samples of pretreated poplar that were analyzed for lignin and carbohydrates pretreatment yields, and enzymatic digestibility (15 FPU/g glucan in raw biomass cellulose loading). After aerated lime pretreatment at 65°C for 4 weeks, and subsequent enzymatic hydrolysis, an overall yield of 0.76 g glucan + xylan recovered per gram glucan + xylan in raw biomass was obtained. This is equivalent to an increased poplar wood digestibility of 7.5‐fold compared with untreated biomass. Different batches of the feedstock resulted in different lignin and carbohydrates pretreatment yields; however, overall yields of carbohydrates (combining pretreatment and enzymatic hydrolysis) were similar. © 2010 American Institute of Chemical Engineers AIChE J, 2011     

Dissolution of Ethylenediamine Pretreated Pulp with High Lignin Content in LiCl/DMSO without Milling

Abstract

Various hardwood and softwood chemical pulps, including those with relatively high lignin content (up to ca. 10.5%), were completely dissolved without milling in lithium chloride/dimethyl sulfoxide (LiCl/DMSO) after a pretreatment with ethylenediamine (EDA). Because milling of the sample is not required, degradation of the cell wall components caused by milling does not take place. After the EDA pretreatment, the crystallinity of the pulps remained as high as the original pulps, although the crystal structure changed. This is the first time that transparent solutions of underivatized pulps with high lignin content were obtained in a simple organic solvent system. Interestingly, even in the case of coarse wood meal (40–80 mesh) about 70% could be dissolved after repeating the dissolving procedure two times. The formation of a pulp–EDA or wood–EDA complex seems to be critical for the dissolution in LiCl/DMSO. The nuclear magnetic resonance (NMR) spectrum of the EDA treated pulp solution had good resolution even though the degree of polymerization (DP) of the cellulose in the pulp is very high.     

Insoluble Lignin Models (4): Condensation Reactions of a Polymer-Bound Guaiacylpropanol Model

The reactions of syringyl alcohol with itself and with either a soluble phenol, 3-methoxy-4-hydroxyphenylpropane (guaiacylpropane, 13) or an insoluble polymer-bound phenol, guaiacylpropanol (14)¹, led to dimeric products. The low (11%), but significant yield of C5-Cα dimer 16 in the reaction of polymer-bound phenol with syringyl alcohol suggests that soluble lignin units are able to condense with insoluble lignin gels during the course of soda pulping. The yield of the C5-Cα dimer 15 in the soluble system was four times greater than that of the analogous heterogeneous dimer 16. Reactions of the polymer model 14 were not mass transfer limited.     

The Characteristics and Reactivity of Acid-Labile Aryl Ether Units in Wood Lignin

Wood lignin contains significant amounts of acid-labile aryl ether units, which play a significant role in lignin modification or delignification processes. We have evaluated the rate and reaction kinetics on the acid-catalyzed cleavages of aryl ether structures for wood lignin in situ based on the formation of phenolic hydroxyl groups. The content of acid-labile aryl ether units was quite uniform for a variety of softwood wood lignins (～4% per C₉ unit) and it varied appreciably among hardwood species, ranging from 4% for aspen to 9% for beech wood lignin. These variations, however, appear to be related to the content of syringyl units in wood lignin. The reactivity of these reactive aryl ether structures was noticeably higher for the spruce than for the aspen wood lignin. This difference in reactivity, based on the behavior of lignin model compound reactions, can be attributed to the influence of syringyl moieties in aspen wood lignin. It appears that most of the acid-labile aryl ether units in hardwood were associated with the syringy moiety being present as a benzyl unit, which is much less reactive than the corresponding guaiacyl moiety.     

Development of Mathematical Models for Describing Organosolv Pulping Kinetics of Fungally Pretreated Wood Samples

Abstract

Eucalyptus grandis wood samples decayed by white- and brown-rot fungi were cooked at 180°C for 5 to 100 min with methanol/water (78:22 v/v) containing 25 mM of CaCl₂ and 25 mM MgSO₄. Mathematical models for describing organosolv pulping kinetics of the fungally pretreated samples were developed to estimate delignification constants, and also rate constants for xylan and wood mass solubilization. These models had high R² values and were able to predict pulp yield, as well as lignin and xylan contents during the process at 99% confidence level. Delignification of undecayed control samples had a rate constant of 2.0 × 10^?2 min^?1 for the bulk phase. Most decayed samples presented delignification constants higher than those of the control. Some decayed samples exhibited an additional delignification phase at the beginning of the process, which was characterized by a low solubilization rate. These samples required insertion of an additional term in the model to represent this phase.     

Preparation of artificial wood films with controlled biodegradability

Artificial wood films containing cellulose, xylan, and lignin were easily prepared by the dissolution of wood components in 1‐ethyl‐3‐methylimidazolium acetate followed by reconstitution with distilled water. The composition and characteristics of wood films were highly controllable and predictable through the variation of the concentration of each component in the wood solution. The water vapor solubility of the wood films was increased when the xylan content was increased and the content of lignin was decreased. The biodegradability of the artificial wood films was investigated with cellulase from Trichoderma viride. The relative degradability of the wood film prepared with 5% cellulose and 5% lignin was 42%, whereas that of the wood film made with 5% cellulose and 5% xylan was 189%. The biodegradability of cellulose in the wood films correlated well with the content of xylan and lignin, and it was enhanced when the xylan content was increased and the content of lignin was decreased. © 2015 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2015 , 132, 42109.     

Fractional Isolation and Structural Characterization of Lignins from Oil Palm Trunk and Empty Fruit Bunch Fibers

Abstract

The lignins from dewaxed oil palm trunk and empty fruit bunch (EFB) fibers were fractionated into 95% ethanol soluble, cold and hot water soluble, and 1% NaOH soluble lignins, respectively. The chemical and structural composition of the lignin preparations was determined by using UV, GPC, FT-IR, ¹³C-NMR spectroscopy and nitrobenzene oxidation. The alkali soluble and 95% ethanol soluble lignin fractions were found to contain low amounts of chemically linked polysaccharides, 2.9-3.9% and 7.5-8.0%, respectively, while the water soluble lignin fractions showed significant amounts of bound polysaccharides (16.2-23.3%). All of the lignin fractions contained a high proportion of non-condensed syringyl units, together with small amounts of non-condensed guaiacyl and fewer p-hydroxyphenyl units. The lignin from oil palm EFB fiber contained a significant amount of esterified p-hydroxybenzoic acid and a minor quantity of esterified glucuronic acid. Trace of ferulic acids was both esterified and etherified to lignin side chains in the EFB fiber cell walls.     

Relationships between Hemicellulose Composition and Lignin Structure in Woods

The composition and absolute amount of neutral sugars were determined for 48 hardwood species (including 17 hardwoods of genera Acacia, 14 hardwoods of genera Eucalyptus, and 17 hardwoods of other genera) and 14 softwood species by alditol-acetate method, and their relationships to the syringyl ratio (syringyl/(syringyl+guaiacyl)) of lignin, which was determined by nitrobenzene oxidation, was investigated. In the hardwood species, an increase in the syringy ratio of lignin correlated with a tendency toward increased xylose/glucose, rhamnose/glucose, and arabinose/glucose ratios. However, the absolute amount of glucose in hardwood was maintained in a small range (0.4–0.5 g in 1 g sample), independent of changes in the syringyl ratio. In the softwood species, with increasing lignin content, the mannose/glucose ratio decreased, but the absolute amount of glucose remained almost constant. In both hardwood and softwood species, a strong correlation was suggested between lignin, indicated by higher syringy ratio, and hemicellulose, indicated by higher xylan/mannan ratio.     

Suitability of Hybrid Willow as a Source of Pulp

Hybrid willow (Salix spp.) is a potential source of pulp as demonstrated by its fibre morphology, chemical composition and pulping kinetics. Fibre length and cell wall thickness measurements on one and two years old clones ranged from 0·49 mm to 0·70 mm and 2·5 μm to 3·6 μm respectively. Fibre length variation by annual growth layer varied from 0·5 mm in the first growth layer to about 1·1 mm in the last growth layer and the maximum growth rate occurred in the first two to three years for fourteen years old clones. The clones were producing more fibre fraction as indicated by volumetric composition. The UV absorptfvities of milled wood lignin was found in the range of 12·17 -14·31 Lg^?1cm^?1 at 278 nm and the presence of syringyl and guaiacyl lignin was observed. ¹³C-NMR results of acetylated milled wood lignin showed clonal structural variations during lignification process and the Klason lignin content of one and two years old clones ranged from 20·8% to 26·1%. Bulk kraft delignification of mature wood resulted in 5–6% higher yield than juvenile wood from the same clone and the pulping activation energy ranged from 98 kJ.mol^?1 to 120 kJ.mor^?1.     

Interactions of the Beechvood Xylan Component with Other Cell Wall Polymers

Sodium chlorite holocellulose of beechwood was extracted in succession with aqueous ammonia solutions of increasing concentration (1–26% NH₄OH), and finally with 5% sodium hydroxide. The polymeric fractions obtained were composed mainly of 4-0-methylglucuronoxylan polymers which occur in two distinct molecular populations. The 1% NH₄OH-extract contained the most accessible polysaccharide fraction which represents a mixture of O-acetylated 4-0-methylglucuronoxylan, residual lignin, cellulose fragments and pectic polysaccharides of the rhamnogalacturonan type containing arabinan and galactan chains. A 4-0-methylglucuronoxylan-polygalacturonan complex with a minor proportion of neutral sugars and residual lignin was isolated from the 10% NH₄OH-extract. The results suggest that some of the residual lignin and pectic polysaccharides are bound by alkali-stable linkages to xylan and/or cellulose chains in the cell-wall complex.     

Characteristics of Dioxane Lignins Isolated at Different Ages of Nalita Wood (Trema orientalis)

Abstract

Nalita (Trema orientalis) is one of the fastest growing trees in the tropical countries. The structural characteristics of lignin isolated at different ages of Nalita wood (Trema orientalis) by acidolytic dioxane method were examined by UV, FTIR, ¹H‐NMR and ¹³C‐NMR spectroscopy, alkaline nitrobenzene oxidation, molecular weight determination, elemental and methoxyl analysis. The data were compared with aspen lignin. The structural analysis revealed that Nalita wood lignin is syringyl‐guaiacyl type. The methoxyl content in Nalita wood lignin was lower than aspen lignin. The C₉ formulas for 30‐months‐old Nalita was C₉H_9.31O_3.13(OCH₃)_1.27, whereas that of aspen was C₉H_8.94O_3.15(OCH₃)_1.47. The weight average molecular weight of Nalita wood lignin was decreased from 36,500 to 25,500 with increasing tree age from 12 to 30 months, whereas weight average molecular weight of aspen was 20,000. Both alcoholic and phenolic hydroxyl group in Nalita wood lignin is lower than aspen lignin.     

New aspects in cationization of lignocellulose materials. VII. Modification of spruce wood meal with quarternary ammonium groups

Trimethylammonium-2-hydroxypropyl (TMAHP) derivatives of spruce wood meal (SWM) and holocellulose of this specia were prepared by the reaction of wood meal with 3-chlor-2-hydroxy-propyltrimethylammoniumchloride (CHMAC) in alkaline medium. The TMAHP samples were fractionated and yields and exchange capacity (Q) of individual fractions were compared with beech and aspen fractions obtained under the same conditions. As it is evident from ¹³C-NMR spectroscopy and GPC analysis the water soluble fraction from TMAHP–SWM consists only of lignin–saccharide degradation products. The NaOH extracts of TMAHP–SWM and TMAHP–holocellulose as well as the water-soluble fraction from TMAHP–holocellulose are polymeric materials. From TMAHP–SWM only 3.1% of alkali-soluble material could be extracted, while from TMAHP–holocellulose 15.7% of water-soluble and 7.9% of alkali-soluble materials were obtained.     

Physicochemical characterization of lignin fractions sequentially isolated from bamboo (Dendrocalamus brandisii) with hot water and alkaline ethanol solution

Nine lignin fractions from bamboo (Dendrocalamus brandisii) were sequentially isolated with hot water at 80, 100, and 120°C for 3 h and 60% aqueous ethanol containing 0.25, 0.5, 1.0, 2.0, 3.0, and 5.0% NaOH at 80°C for 3 h. Molecular weight and purity analysis revealed that the lignin fractions isolated by hot water (L₁, L₂, and L₃) had lower weight-average molecular weights (between 1350 and 1490 g mol⁻¹) and contained much higher amounts of associated hemicelluloses (between 9.26 and 22.29%), while the lignin fractions isolated by alkaline aqueous ethanol (L₄, L₅, L₆, L₇, L₈, and L₉) had higher weight-average molecular weights (between 2830 and 3170 g mol⁻¹) and contained lower amounts of associated hemicelluloses (between 0.63 and 1.66%). Spectroscopy (UV, FTIR, ¹³C-NMR, and HSQC) analysis showed that the bamboo (Dendrocalamus brandisii) lignin was typical grass lignin, consisting of p-hydroxyphenyl (H), guaiacyl (G), and syringyl (S) units. The major interunit linkages presented in the alkaline aqueous ethanol extractable bamboo lignin were β-O-4′ aryl ether linkages (about 74.3%), followed by β-β′ resinol-type linkages and β-1′ spirodienone-type linkages (both for 7.8%), together with small amounts of β-5′ phenylcoumaran (6.8%) and p-hydroxycinnamyl alcohols end groups (3.1%). In addition, a small percentage (1.0%) of the lignin side-chain was found to be acetylated at the γ-carbon, predominantly over syringyl units. © 2012 Wiley Periodicals, Inc. J Appl Polym Sci, 2012