Effect of Simultaneous Steam Explosion and Alkaline Depolymerization on Corncob Lignin and Cellulose Structure

X. Ouyang,a,# L. Chen,a,# S. Zhang,b Q. Yuan,a W. Wang,a,c,* and R. J. Linhardtd,e,f aCollege of Life Science and Technology, Beijing University of Chemical Technology, Beijing, 100029, China bDepartment of Biological Systems Engineering, Washington State University, Pullman, WA 99164-6120, USA cState Key Laboratory of Pulp and Paper Engineering, South China University of Technology, Guangzhou, China, 510640 dDepartments of Chemistry and Chemical Biology, eChemical and Biological Engineering, fCenter for Biotechnology and Interdisciplinary Studies, Rensselaer Polytechnic Institute, 110 8th Street, Troy, NY 12180, USA


Introduction
Lignocellulose requires pretreatment before its biological conversion due to its inherent stability and recalcitrance.Over the past several decades, many lignocellulose pretreatment methods have been developed, including physical, chemical, and biological methods, as well as their combinations 1,2 .The objective of these pretreatment methods is to modify or fractionate lignin and/or hemicellulose, and disrupt the structure of cellulose to increase its enzymatic conversion to glucose 3,4 .From the standpoint of industrial applications, green processes and economic feasibility are important in critically evaluating pretreatment methods 5 .Consequently, it is essential to develop clean and low cost pretreatment methods for conversion of lignocellulose to its components at industrial scale for use in the biorefinery.
Alkaline reagents, including sodium hydroxide, potassium hydroxide, calcium hydroxide, and ammonium hydroxide, have been used extensively and improve the enzymatic digestion of the resultant cellulose, thus greatly improving sugar recovery 6,7,8 .The influence of alkaline pretreatment on enzymatic hydrolysis can be attributed to effective removal of lignin and to the swelling of cellulose structure 9,10,11 .When sugarcane bagasse was treated at 80 °C for 3 h with 0.25 M sodium hydroxide at a ratio of solid to liquid as 1 g to 20 mL, 78 % of lignin was removed 6 .Measurement of cellulose swelling indicated that the alteration of the crystalline hydrogen bond network improved the digestibility of cellulose 12 .The crystal index (CrI) of cellulose from Populus decreased by 5~7 %, 7~9 % and 5~9 % in sodium hydroxide, lime, and ammonia pretreatments, compared with untreated samples.These changes in CrI values were attributed to slight disruption of the cellulose crystalline structure 10 .After alkaline treatment, lignin can be collected from the alkaline solution and used as fuel or in the genera-tion of value-added products to decrease the cost of lignocellulose utilization 13,14 .Alkaline pretreatment provides a promising process for the lignocellulose industrial utilization.However, alkaline pretreatment usually requires the prolonged treatment of lignocellulose with a large volume of alkaline solution at high temperature.The resulting waste salts and the high-energy consumption constitute a major challenge for environmental protection and the economics of the pretreatment process.
Steam-explosion pretreatment of lignocelluloses is considered to be an efficient method for the removal of hemicellulose and for the disruption of cellulose crystallinity 2,5 .Steam-explosion has been applied for the pretreatment of various lignocelluloses, including wheat straw 15 , sunflower stalks 16 , Pinus patula 17 , rice straw 18 , switchgrass and sugarcane bagasse 19 , corn stover 20 , and Eucalyptus globulus 21 .We recently described a clean and effective pilot-scale process for the production of xylitol from hemicellulose using an acid-catalyzed steam explosion process 22,23 .Compared with the traditional boiling processes, this acid-catalyzed steam explosion process increased the temperature from 120 °C to 160~180 °C, and reduced the acid from 105-300 mg/100 g dry mass (1.5~2 %, liquid-to-solid ratios 7~15:1) to 9 mg/100 g dry mass (0.32 %, liquid-to-solid ratios 1.5:1), and time from 120 min to 5~10 min 22,23 .The reason for the reduction in the amount of acid used and the reduced process time was ascribed to the higher temperature and the force of steam explosion.These factors promote the degradation of hemicellulose to xylose improving our xylitol process.This process also substantially reduces waste salts and decreases energy consumption 24 .Subsequently, a high yield of high purity lignin was obtained from corncob by the combination of steam explosion followed by hot ethanol treatment, and this lignin represents a value-added product and improved the enzymatic hydrolysis of cellulose 25,26 .However, this process requires steam explosion to be followed by hot ethanol treatment for lignin isolation.On the basis of previous research, we report the development of a process, relying on the simultaneous steam explosion and alkaline depolymerization, carried out on corncob.Our newly developed single treatment process aims at removing lignin and hemicellulose from lignocellulose and affording cellulose with disrupted crystallinity.This new process also reduces the amount of resulting waste salts and reduces energy consumption.
Corn is one of the main crops in China, and there are about 20 million tons of corncobs generated annually from corn 27 .In addition to being a source of raw material for the production of xylitol, almost 50 % of corncobs are burned or simply rep-resent agricultural waste, representing a major environmental challenge.In the present paper, corncobs and sodium hydroxide were used to develop the simultaneous steam explosion and alkaline depolymerization process.The chemical composition of isolated lignin and the enzymatic hydrolysis of resulting cellulose were investigated to evaluate the effect of this pretreatment.The structural variation of the resulting lignin and cellulose were also analyzed to explore the chemical and physical mechanisms of pretreatment to provide guidance for the further development of pretreatment processes.

Lignocellulosic biomass
Untreated corncob (UC) was collected from Shandong Province and chipped to a size of 2 cm, and stored at room temperature.UC was dried in a cross-flow oven at 50 °C overnight before all experiments.The chemical composition of UC was 38.4 % glucose, 28.4 % xylose, 0.9 % arabinose, 17.9 % acid insoluble lignin (AIL), 5.3 % acid soluble lignin (ASL), and 2.5 % ash.

Simultaneous steam explosion and alkaline depolymerization
Prior to steam explosion, 100 g of UC was immersed in 3 %, 6 %, 9 % sodium hydroxide solution at the solid/liquid ratio of 1:6 for 2 hours.And then, the corncob was drained to obtain a solid with about 60 % absorbed water content.Simultaneous steam explosion and alkaline depolymerization (SSA) was carried out in the steam-explosion equipment according to Zhang 22 .The drained corncob was put in the steam-explosion equipment and heated to 180 °C, 190 °C, 200 °C (corresponding to 1.0 MPa, 1.3 MPa, 1.6 MPa, respectively).The steam pressure was held for 5 min, and the pressure was then reduced abruptly to standard atmospheric pressure.The severity factor of steam explosion can be calculated by an empirical equation 28 10 10 ( 100) Severity factor 14.75 where T is temperature of steam explosion (°C), and t is the retention time (min).The steam explosion experiments were carried out as follows.The t was 5 min and values of T, including 180 °C, 190 °C, 200 °C, were used corresponding to severity factors of 3.05, 3.45 and 3.64, respectively.The three NaOH concentrations were combined with three severity factors, representing a total of 9 different pretreatment conditions.

Chemical composition analysis of DC and L-SSA samples
The chemical compositions of all samples were analyzed by the National Renewable Energy Laboratory (NREL) method 29 .Dried sample (300 mg) was hydrolyzed in a thick-wall flask using 3 mL of sulfuric acid (72 %) for 30 min in a water bath with intermittent shaking every 5 min.After adding 84 mL water to dilute the 72 % H 2 SO 4 to 4 %, the sample was hydrolyzed at 121 °C for 60 min in an autoclave.The hydrolyzed liquid was separated by air pump filtration.The filter cake was dried at 105 °C for 5 h and weighed.A portion of the filter cake was placed in a muffle furnace (Zhonghuan, Tianjin, China) at 575 °C to determine the ash content.The hydrolyzate was analyzed by UV spectrophotometer (Shimadzu UV2450, Japan) to determine acid soluble lignin (ASL).Carbohydrate in the hydrolyzate was measured by high performance liquid chromatography (Hitachi, Tokyo, Japan) equipped with refractive index detector and a Sugar-pak 1 column (Waters, Milford, USA).All the samples were analyzed in triplicate.

Fourier transform infrared spectroscopy (FTIR)
The FTIR spectra of L-SSA samples were recorded in a Nicolet 6700 spectrophotometer in the scan range from 400 cm -1 to 4000 cm -1 with a resolution of 4 cm -1 to compare the molecular structures of samples.A 200-mg KBr disc containing 1 mg sample was used in FTIR measurements.The structural characteristics of the L-SSA samples were determined by a Bruker AV 600 spectrometer at 300 K.About 50 mg lignin was dissolved in 0.5 mL d 6 -DMSO.The spectral widths were 8389 Hz and 25655 Hz for 1 H and 13 C, respectively.The number of collected complex points was 2048 for 1 H dimension with a recycle delay of 1.5 s.The number of transients was 64 and 256 time increments were recorded in 13 C dimension.Data were analyzed by Bruker Topspin software.The DMSO peak (δ C /δ H = 39.5/2.49) was used as an internal standard of chemical shift.

Gel permeation chromatography (GPC)
The molecular weight of L-SSA was carried out in the Waters GPC system (Milford, Massachusetts, USA) equipped with a UV detector.The column was TSK G3000PWxl (TOSOH, Japan) and different molecular weight (210, 4300, 6800, 10000, 17000, 32000, 150000) polystyrene sulfonic acid salts (Sigma-Aldrich) were employed as standards.L-SSA samples were dissolved in 1 % sodium hydroxide solution and diluted to about 2 mg mL -1 with ultrapure water.GPC was performed at room temperature, elution relied on 0.1 M sodium nitrate solution at a flow rate of 1 mL min -1 , and 50 μL sample was injected into the GPC system.

Enzymatic hydrolysis of DC
DC samples were hydrolyzed in 50 mM citrate buffer (pH = 4.8) at 50 °C in an air incubator at 150 rpm.The DC substrate load was 5 %, corresponding to 5 g DC substrate mixed with 100 mL citrate buffer.Cellulase (Novozyme Cellic CTec2) activity was 206 FPU mL -1 and the enzyme loading was 18 FPU g -1 cellulose.The results of enzymatic hydrolysis were analyzed by HPLC as described in section Chemical composition analysis of DC and L-SSA samples.

Chemical composition analysis of samples after the simultaneous steam explosion and alkaline depolymerization
Simultaneous steam explosion and alkaline depolymerization was carried out under 9 different conditions.The chemical compositions of lignin from simultaneous steam explosion and alkaline depolymerization (L-SSA) are shown in Table 1.With increasing steam explosion severity and sodium hydroxide impregnation concentration, the xylose content significantly rose from 7 % to 14 %, while arabinose and glucose content only increased slightly.Xylose, arabinose, and a portion of the glucose are the main components of hemicellulose, which usually acts as an adhesive in lignocellulose through ether linkages, hydrogen bonding, and Van der Waals forces 31 .The high severity of steam explosion can break chemical bonds between hemicellulose and lignin or hemicellulose and cellulose, resulting in increasing hemicellulose-derived sugar content in L-SSA.Acid insoluble lignin (AIL) is the main component of L-SSA.However, AIL in all L-SSA samples decreases from 84 % to 68 % with the use of higher sodium hydroxide concentrations and higher severity factors.This result is mainly attributable to the base-catalyzed lignin depolymerization 32 .Hydroxyl groups show strong electronegativity and, thus, play a major role in the cleavage of the β-O-4 linkages in the phenolic units 33 .The chemical composition of L-SSA indicated that more carbohydrate existed in the L-SSA with the increased sodium hydroxide concentrations and in the case of increased steam explosion severity, which resulted for two reasons, more hemicellulose was depolymerized into monosaccharides, and more lignin was deploymerized into soluble lower-weight chemicals by harsh conditions 32 .In conclusion, lig-nin is the main part of chemical composition in L-SSA, and xylose content will increase with the improvement of sodium hydroxide concentration and steam explosion severity.
Table 2 shows the chemical compositions of DC after the SSA pretreatment.The enrichment of glucose in DC samples suggests that the corncob was well delignified under these different conditions.The xylose content shows an interesting variation under different experimental conditions.There is an obvious decrease in xylose content in DC at the highest severity factor of steam explosion, demonstrating that steam pressure has a much greater influence than sodium hydroxide concentration on the removal of hemicellulose from corncob.Hemicellulose is usually difficult to remove completely because of its chemical linkage with lignin or cellulose in the plant cell wall 34 .Steam explosion can provide high temperature and high pressure to enhance the removal of hemicellulose under basic conditions.In addition, the plant cell wall is destroyed when the high pressure is returned to standard pressure within a short time, resulting in the large-scale extraction of hemicellulose under the highest severity factor (3.64).The different effects of steam explosion and sodium hydroxide on lignin and hemicellulose removal can also be indicated by the percentage of removed lignin and hemicellulose (Table S1).The 5-minute steam explosion process facilitated by sodium hydroxide removes most (>80 %) of the lignin from corncob under the various experimental conditions.The steam explosion severity is the main factor in the SSA process for removing hemicellulose, low severity (3.05) removed 36.7 %~40.2 % of hemicellulose, and high severity (3.64) removed 83.9 %~89.7 %.The chemical composition of DC samples indicates that the SSA process is effective for the removal of lignin and hemicellulose from corncob in a short (5 min) reaction time.

Sample
Glucose (%) Xylose (%) Arabinose (%) ASL (%)  2 show that the different amounts of lignin and hemicellulose removed from corncob in one SSA treatment depends on operating conditions.The amount of sodium hydroxide used in the SSA process was greatly reduced in comparison with the boiling alkaline method 35 , as the base-treated corncobs were drained before steam explosion, while the boiling alkaline depolymerization method retained this sodium hydroxide.In addition, the SSA operating time was reduced to 5 min.Consequently, the process of simultaneous steam explosion and alkaline depolymerization shows potential for application in the industrial utilization of lignocellulose.

FTIR analysis of L-SSA
The characteristics of L-SSA are dependent on the delignification process and original plant species.FTIR is a convenient method to analyze structure of L-SSA samples.The relative amounts of the chemical groups in all L-SSA samples were calculated by comparing the intensity with the bands of lignin aromatic skeletal vibration at 1513 cm -1 36 and are shown in Table 3.Based on FTIR analysis, the relative amount of hydroxyl groups (3400 cm -1 ) in the samples increased with the increase of severity factors, caused by structural changes in lignin and an increase in carbohydrate content.The signals of β-linkages of cellulose (896 cm -1 ) and signals of Note: The intensity of aromatic skeleton vibration (1513 cm -1 ) was set as 1.00 and relative intensive of specific peaks were calculated.
cellulose (cellulose I and amorphous cellulose, 1425 cm -1 ) were stronger, indicating that the carbohydrate content increased in L-SSA.These results agree with the chemical composition analysis.The increasing carbonyl groups (1635 cm -1 , 1704 cm -1 ) of L-SSA indicate that the lignin is oxidized during the steam explosion.There was no positive correlation between lignin content and the relative intensity of p-hydroxy phenyl-guaiacyl-syringyl (HSG) units.The low coefficient of correlation between these values can be attributed to the different amounts of carbohydrate present in the L-SSA, which disturbs this coefficient.

Molecular weight average analysis of L-SSA
The values of weight average molecular weight (M w ) and number average molecular weight (M n ) were determined using GPC.Polystyrene sulfonic acid salts of different molecular weights were employed for preparing GPC calibration curves.The M w and M n of all L-SSA samples are shown in Table 4.A relationship between molecular weight and experimental conditions was observed that indicated the M w and M n of L-SSA samples decreased with an increase in steam-explosive severity factor.This suggests that the depolymerization of lignin takes place with the SSA process.A high sodium hydroxide impregnating concentration can also prompt lignin depolymerization.However, the molecular weight of the resulting L-SSA was still quite high, because lignin oligomers with high molecular weight were formed.Lignin de/re-polymerization happens simultaneously in the presence of sodium hydroxide.In addition, the high temperature and high pressure of steam explosion can enhance the lignin de/re-polymerization.

L-SSA of 2D HSQC NMR
The 2D NMR spectra of L-SSA with different sodium hydroxide concentrations and steam explosion severities are shown in Fig. 1.The assignments of lignin and carbohydrate signals are provided in Table S2 and Table S3 37,38,39 .In the side chain area of the L-SSA (δ C /δ H = 50-90/2.5-6.0),methoxyl groups are present in all samples at δ C /δ H 55.8/3.72.Obvious structural differences could be observed when treatment was performed with different steam explosion severities and different sodium hydroxide concentrations (Fig. 1 and Fig. S2).The linkages between lignin subunits were β-ether (mainly β-O-4 linkages, A and A' in Fig. 2) using a 3.05/3.45severity factor and 3 % sodium hydroxide concentration.Decrease in lignin inter-linkages and increase in carbohydrates signals can be observed in high severity factor and high sodium hydroxide concentration (Fig. 1 and Fig. S2).This interesting phenomenon of varied β-O-4 linkages was attributed to the depolymerization of β-O-4 bonds and the re-polymerization of lignin degradation products.Base-catalyzed depolymerization is often used to cleave the β-O-4 bonds and produce phenolic compounds, such as aldehydes, acids and alcohols 40 .However, the use of base catalysis on lignin depolymerization is not efficient due to a re-polymerization reaction.Unstable lignin intermediates can form the higher molecular weight products through a condensation reaction.Carbanions are likely formed during the depolymerization/re-polymerization on the basis of phenomenon that the decrease in β-O-4 content and the increase in C-C condensed structures 41 .
In the aromatic area (δ C /δ H = 100-135/5.5-8.5),signals of guaiacyl units (G), syringyl unit (S), and p-hydroxyphenyl unit (H) are observed in samples treated using 3.05/3.45severity factor with 3 % sodium hydroxide concentration, which indicates that the lignin from the corncob is typical GSH type lignin (Fig. 1).The intensity of G, S, and H subunits signals decreased with an increase in severity factor and sodium hydroxide concentration (Fig. 1 and Fig. S2).This was probably the result of decreasing the identical monomeric products in pyrolytic and recondensation reactions with increasing severity factors and sodium hydroxide concentrations 39 .
Carbohydrates signals of L-SSA could be observed both in the aliphatic region and the anomeric region of the spectra, which resulted primarily from the presence of hemicellulose depolymerization products, since there was no specific step to purify the lignin from the precipitate.With increasing steam explosion severity and increasing sodium hydroxide concentrations, more PhGlc linkages appeared in the HSQC spectrum, which indicated the enhancement of signal of cellulose binding to lignin (Fig. 1 and Fig. S2).This result agrees with the result of the chemical composition analysis of L-SSA (Table 1).S3).In the anomeric region, C
The SSA process efficiently removes most lignin and a portion of the carbohydrate, especially hemicellulose.The relative amount of carbohydrate in L-SSA samples shows a positive correlation with steam explosion severity and sodium hydroxide concentration..WAXS spectra of DC samples are shown in Fig. 3.
The WAXS peaks of DC samples shifted to lower angles, suggesting that the cellulose was distorted with low order structures 22 .All DC samples showed clear peaks at 16.4 o and 22.5 o , indicating that the cellulose of DC was cellulose I.However, the peaks of cellulose II could be observed when the sodium hydroxide concentration increased to 9 %.The cellulose I chains stack in a parallel arrangement style in plant cell, and cellulose II lies in an antiparallel arrangement style 44 .The reason for this structural conformation at high sodium hydroxide concentrations is the swelling of cellulose fiber.Additionally, in all DC samples the crystal index (CrI) increased with the increase of steam explosive severity factors (Fig. 4).At a given sodium hydroxide concentration, higher CrI was observed with increased severity factor (Fig. 4).This is because more amorphous hemicellulose and lignin are extracted by the increasing steam explosive severity factor.The WAXS results indicated that the DC pattern after the SSA pretreatment belonged to cellulose I with the sodium hydroxide concentration of 3 % and 6 %.However, DC 9-3.05 , DC 9-3.45 , DC 9-3.64 are composed of a mixture of cellulose I and cellulose II.Thus, the sodium hydroxide concentration and steam-explosive severity factor have combined effects on cellulose structures.

Effects of cellulose structural changes on enzymatic hydrolysis
Cellulose enzymatic hydrolysis is significantly influenced by its structure.Enzymatic hydrolysis experiments on all DC samples and UC sample were carried out to investigate the relationships between cellulose structure and enzymatic saccharification.DC 3-3.05 had the highest cellulose conversion (96 %), DC 6-3.05 , DC 6-3.45 , DC 6-3.64 also had high cellulose conversion (>90 %) (Fig. 5 and Fig. S3).Meanwhile, the cellulose digestibility of DC 9-3.64only reached 66 % after 72 h.This illustrates that there is no specific correlation between cellulose conversion and cellulose structures.DC 9-3.05 , DC 3-3.64 , DC 9-3.45 , DC 9-3.64 had low reducing sugar yield probably due to their high ratio of lignin (ASL+AIL):cellulose (glucose) (Table 2).Lignin has a negative charge at pH 4.8, while cellulase, including β-glucosidase I, EG III, EG II, and CBH II, are positively charged at pH 4.8 35 .Positively charged cellulase can be adsorbed by negatively charged lignin, therefore inhibiting the action of cellulase on DC 9-3.05 , DC 3-3.64 , DC 9-3.45 , DC 9-3.64 samples.This result also shows that saccharification is not only affected by cellulose structures, but also by other factors, such as lignin content, the accessibili-ty of cellulase to cellulose, and the degree of cellulose polymerization 45 .However, we observe a clear correlation between the rate of cellulose digestion and CrI (Fig. 4, Fig. 5) with a low CrI showing a high initial rate of enzymatic hydrolysis.A low CrI value indicates the presence of amorphous cellulose, which is easily accessed and hydrolyzed by cellulase.

Conclusion
The simultaneous steam explosion and sodium hydroxide pretreatment can result in the isolation of lignin or lignin and hemicellulose from corncob in reduced time and with lower sodium hydroxide consumption.This process significantly affects the structure of lignin and cellulose.High severity and high sodium hydroxide concentrations can lead to the breakdown of C-O linkages and re-polymerization of lignin substructures.Carbohydrates were bound with lignin by phenyl glycoside linkages.More carbohydrate was released from corncob, since the hemicellulose was dissolved under high severity and high sodium hydroxide concentrations.A well-delignified corncob shows a high cellulose content.Cellulose is also swelled by sodium hydroxide concentrations resulting in a change in cellulose crystal structure, and higher saccharification performance.

F i g . 1 -
2D HSQc NMR spectrum of l-SSA samples isolated from corncob by different steam explosion severities and sodium hydroxide-catalyzed concentrations.l-SSA 3-3.05 , L-SSA 3-3.45 and L-SSA 3-3.64 represent the spectrum of the different operating conditions.
Cellulose structure of DC WAXS has been widely employed to determine the cellulose structural characteristics.The diffraction patterns of DC showed its structure following the SSA process.The diffraction peaks of cellulose I, at 2θ of 15.1 o , 16.8 o , 21.0 o , 22.8 o and 34.6 o , correspond to the 11 ¯0, 110, 012, 200 and 004 crystallographic planes 42 .Cellulose II shows diffraction peaks at 12.2 o , 20.0 o , 22.2 o , corresponding to the 1 11 ¯0, 110 and 200 lattice planes

F i g . 3 -
WAXS patterns of the delignified samples DC 3-3.05 , DC 6-3.05 , DC 9-3.05 , DC 3-3.45 , DC 6-3.45 , DC 9-3.45 , DC 3-3.64 , DC 6-3.64 , DC 9-3.64 F i g .4 -Relationship of cellulose crI and steam explosion severity at various sodium hydroxide concentrations This work was supported by the National Key Research and Development Program of China (No. 2016yFD0400604), State Key laboratory of Pulp and Paper Engineering (201760), Natural Science Foundation of china (NSFc 21576153), Beijing Natural Science Foundation (5162019), National High-tech Research and Development Program (2015AA021001) and chinese 111 Project (B13005).A b b r e v i a t i o n SSA -Simultaneous steam-explosion and alkalinedepolymerization L-SSA -Lignin from simultaneous steam-explosion and alkaline-depolymerization WAXS -Wide angle X-ray scattering DC -Delignified corncob UC -Untreated corncob AIL -Acid insoluble lignin ASL -Acid soluble lignin

F i g . 5 -
Enzymatic hydrolysis of Dc samples at different steam explosion severities of 3.05 (black), 3.45 (red) and 3.64 (blue) S u p p l e m e n t a r y M a t e r i a l s Ta b l e S 1 -Percentage of removed lignin and hemicelluloses in corncob after simultaneous steam explosion and alkaline depolymerization Sample Lignin (%) a Hemicellulose (%) b DC

Table 1 and
Table Ta b l e 2 -The lignin and carbohydrates content of delignified corncob (DC) 13te: a lignin includes acid soluble lignin and acid insoluble lignin; b hemicellulose includes xylose and arabinose.Ta b l e S 2 -Assignments of13C-1 H cross-signals in 2D HSQc NMR spectra of lignin fractions Assignments of the associated carbohydrate13C-1 H cross-signals in 2D H NMRSQc spectra of lignin fractions