Kinetics of Hardwood Carbohydrate Degradation during Kraft Pulp

Article pubs.acs.org/IECR

Kinetics of Hardwood Carbohydrate Degradation during Kraft Pulp Cooking Ricardo B. Santos,*,† Hasan Jameel,† Hou-min Chang,† and Peter W. Hart‡ †

North Carolina State University, Raleigh, North Carolina 27695, United States MWV Corporation, Atlanta, Georgia 30309, United States

‡

ABSTRACT: Most of the studies on hardwood carbohydrate degradation focus upon the understanding of carbohydrate behavior of a single wood species. These studies tend to determine the activation energies associated with the three different cooking phases and for the different reactions that participate in carbohydrate degradation. In the current study, a variety of hardwood species were comprehensively characterized and the kinetics of carbohydrate degradation was studied. The kinetics of glucan, xylan, and total carbohydrate dissolution during the bulk phase of the kraft pulping process were investigated. A wide range of carbohydrate dissolution rates was obtained and correlated to chemical features and delignification rates for nine different hardwood species. It was determined that carbohydrate dissolution was dependent upon the rate of delignification. Species with high carbohydrate dissolution also presented high lignin removal rates. Our results indicate that the presence of lignin carbohydrate complexes positively influences pulping process selectivity during the bulk reaction phase.

1. INTRODUCTION The kraft process, established in 1879 by Carl Dahl in Germany, is currently the predominant chemical pathway for producing pulp. The kraft pulping process releases fiber from wood chips by degrading and solubilizing lignin. Lignin removal results from the reactions of sodium hydroxide and sodium sulfide (white liquor) with the lignin in wood. Lignin removal occurs in three distinct phases, the initial phase, bulk phase, and residual phase (as shown in Figure 1) with potentially as much

have a yield of about or even less than 50% on oven dry wood entering the process. In order to minimize carbohydrate yield losses, it is important to have a comprehensive understanding of both process reaction kinetics and of the starting raw material. Further, interactions between process conditions and intermediate reaction products from the utilized raw material can be critical and may require special attention. Three different reaction steps are known to result in the degradation of carbohydrates during kraft pulping. These reaction steps include initiation, peeling, and end-group stabilization reactions. The difference in carbohydrate reactivity depends upon the differences in their chemical and supramolecular composition.3 Acetyl groups from xylan and galactoglucommannans are readily hydrolyzed at the beginning of the cook. Carbohydrate yield loss occurs mainly due to extensive degradation of hemicelluloses which have both a low degree of polymerization and a significant percentage of amorphous regions. When compared to hemicelluloses, cellulose is relatively resistant to kraft pulping liquor (white liquor). Carbohydrate degradation of softwood has been the focus of a large number of studies which produced a variety of kinetic equations with different degrees of complexity.2,4−9 In general, kinetic models that start with wood chips and focus upon the dissolution of the wood components have produced models which better represent industrial practice rather than the ones using sawdust as starting material. These types of models frequently examine pulp viscosity and the degree of polymerization (DP) of various wood components with emphasis on cellulose, xylan, and glucomannan degradation. The use of xylan and glucomannan

Figure 1. Cooking phases.

as about 70% of the total delignification occurring during the bulk phase.1,2 The addition of sodium sulfide to the sodium hydroxide in white liquor (the distinction between kraft and soda pulping) results in a significant strength advantage in the resulting pulp as compared to other chemical pulping processes. Disadvantages of the kraft process are a loss in pulp yield caused by carbohydrate solubilization and degradation during the alkaline cooking process and the characteristic sulfur odor associated with the use of sodium sulfide in the process. Typically, bleachable grade brownstock kraft pulps © 2012 American Chemical Society

Received: Revised: Accepted: Published: 12192

April 24, 2012 August 16, 2012 August 26, 2012 August 26, 2012 dx.doi.org/10.1021/ie301071n | Ind. Eng. Chem. Res. 2012, 51, 12192−12198

Industrial & Engineering Chemistry Research

Article

was determined for the bulk phase and correlated to wood features such as lignin structure in an attempt to better understand losses during the kraft process.

in kinetics studies is due to the significant presence of those hemicelluloses in hardwoods and softwood, respectively. While the other hemicelluloses, e.g. for softwoods, arabinoglucuronoxylan and arabinogalactan, are ready degraded in the initial phase, xylan and glucomannan will persist in greater content to the end of the cook. Activation energy for cellulose and hemicellulose degradation at the different degradation stages (initiation, peeling, and end-group stabilization) are usually the center of typical kinetic investigations. Young and Liss4 have determined that the activation energy for the peeling reactions of spruce glucomannan has a value of 103 kJ/mol. They also concluded that higher concentration of hydroxide ions results in an increase in the end-group stabilization reactions which lead to higher glucomannan yields. The same results were also found by Wigell et al.5 when studying peeling, termination, and alkaline hydrolysis of pine wood meal. Wigell et al.5 found similar activation energies for peeling and end-group stabilization reactions with values of 111 and 110 kJ/mol, respectively. Their results were in agreement with a previous study performed by Kondo and Sarkanen.6 Kondo and Sarkanen6 found that the hydrogen sulfide ion concentration does not have a negative effect on the initial degradation of western hemlock hemicellulose. They also found that initiation and end-group stabilization had the same activation energy of 93 kJ/mol. By using a more sophisticated model, Andersson et al.7 predicted the cellulose, glucomannan, and xylan content of pulp over the three reaction phases. They assumed that each component reacted as multiple first order kinetic reactions in parallel with each other over the three reaction phases. They found activation energies of the combined carbohydrates (cellulose, glucomannan, and xylan) in the initial cooking phase to be of 50 and 144 kJ/mol for both the bulk and residual phases. The rapid first order initiation reactions result in the rapid decrease in residual lignin during the initial phase while the bulk phase reactions account for the majority of the delignification as a slightly slower reaction rate. When it comes to hardwood kinetics modeling, a much smaller number of studies are available. Additionally, these studies have been conducted in a more superficial manner than the softwood studies. The activation energy for total carbohydrates has been reported to be 220 kJ/mol for alkaline hydrolysis reactions.8 In the same study, the authors found a much lower number for the activation energy associated with end-group stabilization (55 kJ/mol) and peeling reactions (68 kJ/mol). In a study where the authors11 divided total carbohydrates into two components (cellulose and hemicellulose) it was found that the activation energy for dissolution of cellulose is higher than the activation energy for hemicellulose, independent of the cooking phase. Cellulose dissolution exhibited an activation energy value of 125 kJ/ mol in the initial phase while the hemicellulose activation energy was only 50 kJ/mol. In the residual phase, the cellulose dissolution activation energy was determined to be 160 kJ/mol and that of hemicellulose was only 118 kJ/mol. Most of the studies on hardwood carbohydrate degradation have focused upon the understanding of carbohydrate behavior of single wood species. Additionally, these studies tend to determine the activation energies associated with the three different cooking phases and for the different reactions that participate in carbohydrate degradation. In the current study, a variety of hardwood species were comprehensively characterized and the kinetics of carbohydrate degradation was studied. The rate of carbohydrate degradation

2. EXPERIMENTAL SECTION 2.1. Raw Material. Eucalyptus nitens (EN), Eucalyptus globulus (EG), Eucalyptus urograndis (URO), sweet gum (Liquidambar styracif lua, SG), red maple (Acer rubrum, MA), red oak (Quercus rubra, RO), red alder (Alnus rubra, RA), cottonwood (Populus trichocarpa, CW), acacia (Acacia mangium, ACA), and loblolly pine (Pinus taeda, P), which was used as reference, were obtained from different pulp and paper mills around the world. Once chipped in a laboratory chipper, the bark and knots were removed from each of these samples. The cleaned and air-dried samples were then ground in a Wiley mill until all of the ground chips were passed through a 40−60 mesh sieve. The wood meals (40−60 mesh) were Soxhlet extracted for 24 h with benzene−ethanol 2:1 (v/v) to remove extractives,12 dried, and used for pulping and lignin isolation. 2.2. Sawdust Carbohydrate Degradation Rate Constant Determination. Carbohydrate degradation experiments were performed at 150 °C using E. nitens, E. globulus, E. urograndis, sweet gum, maple, red oak, red alder, cottonwood, acacia, aspen, and loblolly pine. Stainless steel autoclaves (50 mL) were filled with 3 g of OD wood sawdust and 30 mL of white liquor and sealed. Excess white liquor (liquor:wood ratio of 10:1) was used in order to maintain nearly constant reagent concentration during the experiments. The active alkali charge was 40%, with 25% sulfidity. It has been estimated that about 15−16% of the total alkali was consumed during the kinetic experiments. Preliminary data were collected to determine that the final alkali content was reasonably constant during 20−60 min reaction time frame associated with these experiments. The sealed autoclaves were placed on the bottom of an M&K digester. The M&K digester is a sealed pressure vessel equipped with a recirculating pump which removes process fluid from the bottom of the vessel passes it through a process heater to control liquid temperature and returns the liquid into the top of the digester. See Figure 2 below. The digester vessel containing

Figure 2. Apparatus used for kraft kinetics study.

the charged, sealed autoclave bombs was filled with white liquor as the process fluid. The white liquor used for heating had the same ionic concentration as the one used inside the 50 mL autoclaves. After the desired reaction time, the heating fluid was drained from the bottom of the digester vessel, and the whole apparatus was rapidly cooled down by running cold water through it. The samples were removed from the reactors (autoclaves) and washed with deionized water until a neutral 12193

dx.doi.org/10.1021/ie301071n | Ind. Eng. Chem. Res. 2012, 51, 12192−12198

Industrial & Engineering Chemistry Research

Article

Figure 3. Degradation curves of glucan (a), xylan (b), and total carbohydrates (c) for acacia, cottonwood, E.globulus, E. nitens, and pine.

pH was obtained. Cooking yield and lignin content were determined according to the method of Dence and Lin.13 Estimated error for lignin content analysis was below 2%. 2.3. 13C NMR for Lignin Structure Quantification. NMR analysis was carried out for all the species using the isolated milled wood lignin (MWL). The analysis14 was performed using quantitative 13C NMR spectroscopy of the acetylated and nonacetylated lignin preparations in DMSO-d6 using a Shigemi NMR microtube. 13C NMR spectra were acquired on a Bruker AVANCE 300 MHz spectrometer equipped with a 5 mm QNP probe using an inverse-gated proton decoupling (IGD) sequence at 300 K using a 90° pulse width, 1.2 s acquisition time, and 1.7 s relaxation delay. Sample concentration was ca. 25%. Chromium(III) acetylacetonate (0.01 M) was added to the NMR tube prior to quantitative 13C NMR acquisition to provide complete relaxation of all nuclei. A total of 20 000 scans were collected. The data set was processed using an exponential multiplication (EM) window function, with zero-filling and line broadening of 10 Hz. MWL was isolated according to a modified protocol14 where all samples were extracted with 0.3% NaOH for 1 h to remove tannins. The ball milled wood was extracted with 96% aqueous dioxane, in accordance with the method of Bjorkman.15 The targeted milled wood lignin yield was 27%. Lignin carbohydrate complexes (LCCs) were characterized using MWL and following the protocol proposed by Balakshin et al.16 2D HSQC NMR spectra were acquired at sample concentration of ca. 10% on a Bruker AVANCE 300 MHz spectrometer equipped with BBI probe. The acquisition parameters used on the 300 MHz spectrometer were as follows: 160 transients (scans per block) were acquired using

1K data points in F2 (1H) dimension for an acquisition time of 151 ms and 256 data points in F1 (13C) for an acquisition time of 7.68 ms for total of 20 h. A coupling constant 1JC−H of 147 Hz was used. The 2D data set was processed with 1K × 1K data points using Qsine function in both dimensions. The LCC signals were integrated in the HSQC spectra, and the amount of LCC was calculated using the following equations16 Benzyl ether = 2DBE/2D90−78/5.7−3.0 ×13 C90−78 /13 C163−103 × 600

Phenyl glycoside = 2DPhGlc /2D103−96/5.5−3.8 × 13C103−96 /13 C163−103 × 600

where BE and PhGlc are the amounts of benzyl ether and phenyl glycoside linkages (per 100Ar); 2DBE and 2DPhGlc are the resonance (volume) of the signals of benzyl ether and phenyl glycoside LCC linkages in the 2D spectrum, 2D90−78/5.7−3.0 and 2D103−96/5.5−3.8 are the total resonance of the corresponding clusters in the 2D spectrum, 13C90−78 and 13 C103−96, and 13C163−103 are the resonance of the corresponding cluster in the 13C spectrum, and 600 is the amount of aromatic carbons in 100Ar. 2.4. Carbohydrate Analysis. The samples’ carbohydrate composition was determined by acid hydrolysis. A total of 0.1 g of sample was hydrolyzed with 1.5 mL of 72% H2SO4 at room temperature with occasional stirring for 2 h. The mixture was then diluted to 3% H2SO4 using deionized water, transferred to a vial, sealed, and heated to 120 °C for 1.5 h. The resulting 12194

dx.doi.org/10.1021/ie301071n | Ind. Eng. Chem. Res. 2012, 51, 12192−12198

Industrial & Engineering Chemistry Research

Article

suspension was filtered, and the filtrate was analyzed for monomeric sugars. Fructose was added as an internal standard. The monomeric sugar content was determined by injecting 2.5 mL samples into a high-performance anion-exchange chromatograph with pulsed amperometric detection (HPAEPAD) on a Dionex IC-3000 chromatography system. Sugars were separated using Carbo-Pac PA1 guard and analytical columns connected in series. Water was used as eluent at a flow rate of 1.0 mL/min. The column temperature was 18 °C. A postcolumn base 40 mM NaOH was added to improve detection by pulsed amperometry. The postcolumn flow rate was 1.0 mL/min. Estimated error for carbohydrates analysis was below 3%.

Figure 4. Glucan and xylan dissolution rate constant.

3. RESULTS AND DISCUSSION 3.1. Carbohydrates Degradation Rate Constant. The majority of the kraft delignification reactions occur during the bulk phase of cooking; therefore, the kinetics experiments were designed to investigate the effect of the bulk reaction cooking phase on carbohydrate degradation. As polymers such as carbohydrates and extractives consume part of the alkaline solution, an excess of alkali charge was necessary to isolate those reactions and maintain a reasonably constant alkali concentration during the bulk delignification reaction phase. To ensure that the experimental pulps were within the bulk delignification phase, the log of residual lignin was plotted and a straight line following the bulk phase was found for each of the species examined. This method was employed to verify that the study of 20−60 min reaction time samples were within the bulk delignification phase. Carbohydrate degradation by kraft liquor was measured at 150 °C for four different time periods (20, 30, 45, and 60 min). As a result, the perceived first order kinetic rate equation could be written as shown in eq 1 dC /dt = k′[C ]

Figure 5. Total carbohydrates dissolution rate constant.

their low degree of polymerization and lower degree of crystallinity (more amorphous regions). 3.2. Relationship between Carbohydrate Dissolution and Lignin Structure. Lignin structure obtained from 13C NMR was correlated with the carbohydrate dissolution rate values. As determined in previous work,17 a substantial species related variation in syringyl and guaiacyl content was found. Table 1 shows initial compositional and structural analysis for all species. It was determined that the rate of carbohydrate degradation for all three carbohydrate fractions (glucan, xylan, and total carbohydrates) was correlated to the S/G ratio of the lignin. The correlation between S/G and glucan dissolution, shown in Figure 6, is quite weak (R2 = 50%) but indicates a relationship between them. On the other hand, a much better correlation (R2 = 77%) between xylan dissolution rate and the S/G ratio was found (Figure 7). When the total carbohydrate dissolution rate, which includes the dissolution of all hemicelluloses and cellulose, was evaluated (Figure 8), even higher correlation was observed (R2 = 85%). The correlation of lignin structure (S/G) to carbohydrate dissolution rate appears to be an indirect relationship. The wood cell wall contains not only cellulose, hemicellulose, and lignin but also pectins, extractives, and trace metals. All of these components interact with each other and form a complex matrix. Lignin, in this matrix, is known to cross-link to different polysaccharides contributing to wood rigidity.18−22 This close interaction among wood cell wall components seems to be responsible for the higher degradation of carbohydrates for species containing high amounts of syringyl lignin units. Since higher S/G ratio provides greater lignin removal,17 it is reasonable to assume that carbohydrates are also being degraded at a greater rate. This is especially confirmed when looking at the superior correlation of S/G with xylan (Figure 7) and total carbohydrates (Figure 8) which include degradation of all hemicelluloses. Hemicelluloses are intermediates for

(1)

where C is the ratio of residual carbohydrate being examined and k′ is the pseudo first order rate constant. k′ was determined graphically from the experimental data by plotting C/Co × 100% as a function of time and fitting the data using standard kinetic techniques. Figure 3a−c shows examples of the carbohydrate degradation curves obtained from the kinetics experiments for ACA, CW, EG, EN, and P. There is a very good correlation between cooking time and percentage of glucan, xylan, and total carbohydrates (carbs) removal. The kinetics indicate first order reaction rates which is in agreement with the work of other authors.3,9 The degradation rate constant values were determined by best fitting these data to an exponential function. The best fit functions and the R2 values for these regressions are shown in Figure 3. A wide range of reaction rate constants was obtained for the various hardwoods examined (Figures 4 and 5). A high rate constant represents higher carbohydrate degradation and therefore higher yield loss at a specific reaction time. In general, the Eucalyptus species presented the highest degradation rates especially for xylan and total carbohydrates. The highest degradation rate constant values were obtained for E. globulus, followed by E. nitens, and E. urograndis. The lowest value was obtained for red alder. Glucan degradation was approximately constant among all the species studied and lower than xylan degradation. In general, hemicelluloses are more easily degraded under alkaline conditions than cellulose due to 12195

dx.doi.org/10.1021/ie301071n | Ind. Eng. Chem. Res. 2012, 51, 12192−12198

Industrial & Engineering Chemistry Research

Article

Table 1. Initial Species Sugar Composition, Lignin Content, and Lignin Structure sugar, %

lignin, %

species

Ara

Rha

Gal

Glc

Xyl

Man

total

Klason

soluble

total

S/G

E. nitens E. urograndis E. globulus cottonwood acacia red alder maple sweetgum red oak pine

0.3 0.2

0.4 0.3 0.3

0.3 0.7 0.5

0.4 0.4 0.4

41.8 48.5 46.1 44.8 46.9 40.8 43.5 37.8 41.6 41.9

15.9 10.7 14.0 13.6 12.6 16.6 13.9 16.8 18.1 6.1

1.2

0.4

0.6 0.8 1.2 0.6 0.7 0.7 0.6 0.5 0.6 2.3

60.1 60.2 62.2 61.8 61.2 59.8 61.4 57.7 63.4 63.0

22.3 24.5 20.9 19.6 25.1 22.7 24.6 25.3 23.8 27.0

3.2 2.1 3.0 1.9 1.7 1.7 1.3 1.9 3.9 0.3

25.5 26.6 23.9 21.5 26.8 24.4 25.9 27.2 27.7 27.3

2.59 1.76 2.73 1.41 1.18 1.37 1.27 1.63 2.12 0

1.8

0.8 2.2 1.0 1.0 2.4 2.0 3.1 10.9

where lignin and hemicellulose bonds establish an essential element of secondary cell wall formation.28 3.3. Carbohydrate Degradation versus Delignification Rate. The relationship between lignin and carbohydrate degradation is also confirmed in Figure 9 where delignification

Figure 6. Glucan degradation rate versus S/G ratio.

Figure 9. Carbohydrate dissolution rate versus lignin reaction rate.

reaction constants for the different wood species (for more detail, refer to Santos et al. 17 ) were plotted against carbohydrate degradation constants. A straight line was obtained which indicates higher carbohydrate degradation for species with high lignin removal for a fixed time. Nevertheless, lignin degradation is much greater that carbohydrate degradation. On average, lignin was degraded 13 times faster than the carbohydrates. 3.4. Bulk Phase Selectivity. To ensure that our data follows common and practical beliefs, Figure 10 analyzes selectivity for all the species. Selectivity was calculated using eq 2 and the time interval of 20−60 min.

Figure 7. Xylan degradation rate versus S/G ratio.

Figure 8. Total carbohydrate degradation rate versus S/G ratio.

cellulose association through lignin23−25 which gives interconnection to the secondary wall.26,27 Hemicelluloses in the matrix have been reported to be associated with condensed type lignin and, in a more advanced stage of cell wall growth, to noncondensed lignin moieties also. This association is believed to be an indication of lignin−carbohydrate complex formation

Figure 10. Species selectivity during the bulk reaction phase. 12196

dx.doi.org/10.1021/ie301071n | Ind. Eng. Chem. Res. 2012, 51, 12192−12198

Industrial & Engineering Chemistry Research

Article

Figure 11. Increase in process selectivity by increase in lignin carbohydrate complex (LCC).

showing that LCC content for softwood is lower than hardwoods. In the case of the bulk and even initial cooking stages higher amounts of LCCs present in wood may actually be beneficial to the pulping process even though LCCs have been shown to negatively impact the residual phase of cooking. It is believed that LCC linkages have the ability to stop peeling reactions of carbohydrates through stabilization of the reducing end groups. Peeling reactions occur when the polysaccharide end group is converted to a carboxylic acid. The reaction starts with isomerization of the end group to a ketose in which the glycosidic bond is in the β position to the carbonyl group. This structure is labile in alkali resulting in cleavage of the glycosidic bond. When the end group is attached to lignin as in an LCC, the aldehyde reducing end group is no longer available to initiate the peeling reaction. Figure 11 shows that this type of protection may occur during early stages of delignification. To the best of our knowledge this is the first time LCC content has been shown to have a beneficial impact on the early stages of the kraft pulping process. 3.5. General Considerations. Eucalyptus species are known to be one of the best species in terms of pulping performance (low κ number with high carbohydrate yield and a high degree of polymerization). In our previous study17 we have shown the dominance of Eucalyptus with respect to the rate of delignification. Eucalyptus species have the highest delignification rate which makes them very easy to delignify. In the present work, the fact that Eucalyptus species have a high rate of carbohydrate dissolution was, at first, unexpected. Upon closer examination, it was observed that process selectivity was fairly constant among all of the samples studied. It seems that delignification and carbohydrate dissolution rates are closely linked, and this association appears to be responsible for the higher carbohydrate dissolution rates being correlated with the higher delignification rates. Since lignin is known to be closely associated with carbohydrates, a closer investigation showed that the rate of carbohydrate dissolution follows the rate of wood delignification. The rate of delignification was 13 times higher than the rate for carbohydrate dissolution. Studies on the association of

Selectivity (%) = [100 − (Δtotal carbs lost/Δ total lignin removed)] × 100

(2)

Even though our data showed a higher degradation rate for carbohydrates for the species that are known to yield more pulp at the end of the cooking process (Eucalyptus), the analysis of selectivity showed hardwood species are rather constant in terms of carbohydrate degradation during the process. Again, it is possible to conclude that degradation of carbohydrates follows along with delignification, no matter the hardwood species in discussion. The difference in selectivity (which can also be seen as yield loss) results of an extension of cooking which mainly occurs during the residual cooking phase. As can be seen in Figure 10, the softwood (P) selectivity was lower than that of any and all of the hardwoods examined. The lower pine selectivity is due the fact that glucommanans (the major form of hemicelluloses in pine) have no protection against degradation reactions that occur during pulping. On the other hand xylan, which is the major hemicellulose in hardwood, is protected from degradation by uronic acids29 which prevent the peeling reaction from occurring by stabilizing the carbohydrate reducing end groups. Even though selectivity was determined to be fairly constant among the hardwood species studied, higher amounts of LCC linkages (ether + glycoside) appear to increase process selectivity in the bulk phase (Figure 11). LCC content is known to be responsible for low delignification and/or hard to remove lignin during the residual stage of cooking. 30 Information regarding the quantification of LCCs is very scarce; however, some quantification attempts can be found in the literature. While evaluating LCC linkages, Obst31 observed that pine and aspen had different LCC contents. While pine LCC content (ester + ether) was of 4.7 per 100 monomeric lignin units, aspen had only 0.9 per 100 monomeric lignin units. More recently Balakshin et al30 also found variation in LCC content among different wood species. While total LCC content (ether + phenyl glycoside + esters) in pine was determined to be 7.7, birch was determined to have 10.2 LCCs per 100 monomeric lignin units. These results confirm our data 12197

dx.doi.org/10.1021/ie301071n | Ind. Eng. Chem. Res. 2012, 51, 12192−12198

Industrial & Engineering Chemistry Research

Article

(8) Johansson, D. Carbohydrate degradation and dissolution during kraft cooking. Licentiate thesis, Karlstad University; 2008. (9) Lai, Y.-Z.; Sarkanen, K. V. Kinetics of alkaline hydrolysis of glycosidic bonds in cotton cellulose. Cellul. Chem. Technol. 1967, 1, 517. (10) Giudici, R.; Park, S. W. Kinetic model for kraft pulping of hardwood. Ind. Eng. Chem. Res. 1996, 35, 856. (11) Mirams, S.; Nguyen, K. L. Kinetics of kraft pulping of Eucalyptus globulus, fundamentals and applications in pulping, papermaking and chemical preparation; The AIChE Symposium Series; American Institute of Chemical Engineers: New York; 1996; Vol. 6, p 1. (12) TAPPI standard 1998−1999, T264 om-88. (13) Dence, C. W.; Lin, S. V. Methods of lignin chemistry; SpringerVerlag: Berlin, 1992. (14) Capanema, E. A.; Balakshin, M. Y.; Kadla, J. F. Chang, H-m. On isolation of milled wood lignin from eucalyptus wood. O Papel 2007, N5, 74. (15) Bjorkman, A. Studies on finely divided wood. Part I. Extraction of lignin with neutral solvents. Sven. Papperstidn. 1956, 59, 477. (16) Balakshin, M. Y.; Capanema, E. A.; Gracz, H.; Chang, H-m.; Jameel, H. Quantification of lignin-carbohydrate linkages with high resolution NMR spectroscopy. Planta 2011, 233, 1097. (17) Santos, R. B.; Capanema, E. A.; Balakshin, M. Y.; Chang, H-m.; Jameel, H. Effect of hardwoods characteristics on kraft pulping process: emphasis on lignin structure. BioResources 2011, 6, 3623. (18) Liu, F. P.; Rials, T. G. Relationship of wood surface energy to surface composition. Langmuir 1998, 14, 536. (19) Lawoko, M.; Henriksson, G.; Gellerstedt, G. Characterization of lignin-carbohydrate complexes (LCCs) of spruce wood (Picea abies L.) isolated with two methods. Holzforschung 2005, 60, 156. (20) Lawoko, M.; Henriksson, G.; Gellerstedt, G. Structural differences between the lignin-carbohydrate complexes present in wood and in chemical pulps. Biomacromolecules 2005, 6, 3467. (21) Henriksson, G.; Lawoko, M.; Martin, M. E. E.; Gellerstedt, G. Lignin-carbohydrate network in wood and pulps: a determinant for reactivity. Holzforschung 2007, 61, 668. (22) Whetten, R.; Sederoff, R. Lignin Biosynthesis. Plant Cell 1995, 7, 1001. (23) Fry, S. C. Cross linking of matrix polymer in the growing walls of angiosperms. Ann. Rev. Plant Physiol. 1986, 37, 165. (24) Newman, R. H. Nuclear magnetic resonance study of spatial relationships between chemical components in wood cell walls. Holzforschung 1992, 46, 205. (25) Iiyama, K.; Lam, TB-T.; Stone, B. A. Covalent cross-links in the cell wall. Plant Physiol. 1994, 104, 315. (26) Anterola, A. M.; Lewis, N. G. Trends in lignin modification: a comprehensive analysis of the effects of genetic manipulations/ mutations on lignification and vascular integrity. Phytochemistry 2002, 61, 221. (27) Ha, M.-A.; MacKinnon, I. M.; Sturkova, A.; Apperley, D. C.; McCann, M. C.; Turner, S. R.; Jarvis, M. C. Structure of cellulosedeficient secondary cell walls from the irx3 mutant of Arabidopsis thaliana. Phytochemistry 2002, 61, 7. (28) Ruel, K.; Chevalier, V. B.; Guillemin, F.; Sierra, J. B.; Joseleau, J.P. The wood cell wall ate the ultrastructural scale-Formation and topochemical organization. Maderas: Cienc. Tecnol. 2006, 8, 107. (29) Jiang, Z.-H.; Lierop, B. V.; Berry, R. Hexenuronic acid groups in pulping and bleaching chemistry. Tappi 2000, 83, 167. (30) Balakshin, M. Y.; Capanema, E. A.; Chang, H-m MWL fraction with a high concentration of lignin-carbohydrate linkages: isolation and 2D NMR spectroscopic analysis. Holzforschung 2007, 61, 1. (31) Obst, J. Frequency and alkali resistance of lignin-carbohydrate bonds in wood. Tappi. 1982, 65, 109. (32) Li, A.; Khraisheh, M. Bioenergy II: Bio-Ethanol from municipal solid waste (MSW): The role of biomass properties and structures during the ethanol conversion process. Int. J. Chem. React. Eng. 2010, 8, A85.

lignin and carbohydrates have demonstrated not only that covalent bonds exist between lignin and all major polysaccharides (arabinoglucuronoxylan, galactoglucomannan, glucomannan, pectins, and cellulose) but also that cross-linkages also exist among them.19,20,32 Phenyl glycosides, benzyl ethers, and benzyl esters have been suggested as the main types of lignin−carbohydrate bonds in wood.30 The high LCC content of six of the species studied here proved to be beneficial during the bulk cooking stage by stabilizing the carbohydrate reducing end group, thus avoiding peeling reactions which reduce carbohydrate yield.

4. CONCLUSIONS The rate of carbohydrate dissolution followed the rate of wood delignification with the rate of delignification being 13 times higher than the rates for carbohydrate dissolution. Higher delignification resulted in higher carbohydrate degradation. Nevertheless, bulk phase reactions showed similar selectivity for carbohydrates for each of the species studied. By understanding how the rate of delignification and the rate of carbohydrate degradation relate to one another, it may be possible for engineers to make more informed decisions about the best way to pulp and bleach specific wood specie to increase pulp yield. A well developed kinetic model of hardwood carbohydrate dissolution may also aid in this endeavor. A high LCC content appears to be beneficial during bulk cooking stage by stabilizing the carbohydrates reducing end group thus avoiding peeling reactions. It is well known that LCCs make pulping more difficult and negatively impact pulp yield during the residual phase of delignification. The finding that the presence of LLCs might actually be useful during the bulk delignification phase by preventing the initiation of the peeling reaction is novel. Further study should be performed to determine the extent of LLC end group stabilization and their overall impact upon peeling and yield loss during kraft pulping.

■

AUTHOR INFORMATION

Corresponding Author

*Tel.: (9l9) 515-7712. Fax: (919) 515-6302. E-mail: balleirini@ gmail.com. Notes

The authors declare no competing financial interest.

■

REFERENCES

(1) Gierer, J. Chemical aspects of kraft pulping. Wood Sci. Technol. 1980, 14, 241. (2) Gustafson, R.; Sleicher, C.; McKean, W. Theoretical model of the kraft pulping process. Ind. Eng. Chem. Process Des. Dev. 1983, 22, 87. (3) Sixta, H.; Rutkowska, E. W. Comprehensive kinetics study of delignification, carbohydrate degradation, cellulose chain scissions, and hexenuronic acid reactions during kraft pulping of Eucalytus globulus. Lenzinger Ber. 2006, 86, 32. (4) Young, R. A.; Liss, L. A kinetic study of the alkaline endwise degradation of gluco- and galactomannans. Cellul. Chem. Technol. 1978, 12, 399. (5) Wigell, A.; Brelid, H.; Theliander, H. Kinetic modeling of (galacto)glucomannan degradation during alkaline cooking of softwood. Nord. Pulp Pap. Res. J. 2007, 22, 495. (6) Kondo, R.; Sarkanen, S. V. Kinetics of lignin and hemicellulose dissolution during the initial stage of alkaline pulping. Holzforschung 1984, 38, 31. (7) Andersson, N.; Wilson, D. I.; Germgård, U. An improved kinetic model structure for softwood kraft cooking. Nord. Pulp. Pap. Res. J. 2003, 18, 200. 12198

dx.doi.org/10.1021/ie301071n | Ind. Eng. Chem. Res. 2012, 51, 12192−12198