SULPHATE AND BISULPHITE PULP YIELDS WITHINWOOD GROWTH ZONES OF. Picea mariana (Mill.) B.S.P. AND Pseudotsuga menziesii (Mirb.

Transcription

1 SULPHATE AND BISULPHITE PULP YIELDS WITHINWOOD GROWTH ZONES OF Picea mariana (Mill.) B.S.P. AND Pseudotsuga menziesii (Mirb.) Franco. by SHUI-TUNG CHIU B Sc. Chung-hsing University, Taiwan, A THESIS SUBMITTED IN PARTIAL FULFILMENT OF. THE REQUIREMENTS FOR THE DEGREE OF MASTER OF FORESTRY in the Department of Forestry We accept tbis thesis as conforming to the required standard. THE UNIVERSITY OF BRITISH COLUMBIA JUNE, 1968.

2 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and Study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by h.ils representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of f- Or-gS fr-\/ The University of British Columbia Vancouver 8, Canada Date Au-^^t S~, /f&l?

3 ABSTRACT Quantitative methods for sulphate and Na-base bisulphite micropulping and micro-analytical procedures were developed. Raw pulp and pulp carbohydrate yields were correlated with relative position within growth increments of black spruce and Douglas fir. Profiles varied at different pulp yield levels and with pulping process. No profiles were simply correlated with wood micro-specific gravities. Maximum raw pulp and pulp carbohydrate yields within growth increments shifted from earlywood to latewood as yield changed from high(80 ^ 57.) to low(45-57=,) levels.' Delignification rate differed within increments for both pulping processes. In the initial sulphate and Na-base bisulphite cooking stage, latewood lignin seemed to be more easily removed than that from earlywood. At high yield levels (80 *57<>), the pulp residual lignin contents based on oven-dry pulp followed similar patterns in that maxima were found in earlywood, abruptly decreasing in the transition zone then slightly increasing in the latewood portion. At low yield levels (45 57.), the residual lignin patterns varied slightly, or remained constant throughout the whole increment. Raw pulp yields, residual lignin contents and pulp carbohydrate yields (based on extractive-free water-free wood) were not significantly different for combined data of heartwood and sapwood, the two woods and two pulping processes, except for Na-base bisulphite pulp carbohydrate

4 yields which showed significantly higher values for sapwood. Sulphate raw pulp yields and residual lignin contents obtained by combining data from all cooking levels and wood zones were not significantly different between the two species examined, except for Douglas fir carbohydrate yield which was significantly higher than that of black spruce. For Na-base bisulphite pulping, Douglas fir raw pulp yields and pulp carbohydrate yields were highly significantly greater than those from black spruce, whereas pulp residual lignin was not significantly different.

5 - I l l - TABLE OF CONTENTS TITLE PAGE.... '. ABSTRACT..... ; i TABLE OF CONTENTS TABLES AND FIGURES ACKNOWLEDGMENTS iii vi x INTRODUCTION 1 REVIEW OF LITERATURE 3 A. Variation of Chemical Pulp Yields Among Coniferous Woods 3 B. Chemical Pulp Yield Differences Between Coniferous Wood Zones 4 C. Coniferous Earlywood-Latewood Chemical Pulp Yields 5 MATERIALS AND METHODS 8 A. Wood Sample Preparation 8 B. Latewood Percentage Determinations.9 C. Wood Micro-specific Gravity and Micro-lignin Determinations Wood micro-specific gravity methods 9 2. Wood micro-lignin methods 9 D. Determining Micro-pulp Yields Micro-pulping procedures 10 a. Wood sample preparation and pulping 10 b. Pulp yields, residual lignin and pulp carbohydrate yields 12

6 - iv Development of procedures 13 a. Micro-pulp yield 13 (1). Effect of wood section thickness on pulp yield 13 (2). Effect of screen mesh size on pulp yield (3). Effect of sample position on screen 14 (4). Determination of replication number 15 b. Residual micro-lignin 16 (1). Sample weight 18 (2). Digestion time and pulp solubility 19 (3). Elapsed time and relative lignin content RESULTS 21 A. Intra-incremental Micro-specific Gravities And Wood Micro-lignins 21 B. Raw pulp yields Within growth increments 22 a. Raw pulp yields. 22 b. Residual lignin contents 25 c. Pulp carbohydrate yields Between wood zones Between species 31 DISCUSSION 35 A. Variation Within Growth Increments Morphology and pulp yields Delignification and residual lignin contents

7 - V - 3. Effect of carbohydrate degradation on pulp yields Process and raw pulp yield profiles within increments at various yield levels 45 B. Variation Between Wood Zones 47 C. Variation Between Species 49 CONCLUSIONS 51 REFERENCES 53

8 - vi - TABLES AND FIGURES Table 1.. Coniferous earlywood (E) and latewood (L) chemical compositions from the literature based on oven-dry wood 64 Table 2. Coniferous earlywood (E)-latewood (L) raw pulp yields and residual lignins at various cooking times (T) from the literature and summary of present experimental values based on extractive-free water-free wood 65 Table 3. Description of stems and growth increments used in micropulping studies 66 Table 4. Absorptivity of lignins from woods and pulps of black spruce and Douglas fir 67 Table 5. Species, growth increment numbers, moisture content of wood sections and times at maximum temperature for sulphate and Na-base bisulphite cooks 68 Table 6. Sulphate and Na-base bisulphite cooking variables Table 7. Analysis of variance on the effect of various thicknesses of black spruce and Douglas fir wood sections on sulphate and Na-base bisulphite pulp yields 70 Table 8. Analysis of variance on the effect of screen mesh size on black spruce sulphate cooking 70 Table 9. Analysis of variance on effect of between screen positions on pulp yields 71 Table 10. Analysis of variance on the effect of within screen position on black spruce and Douglas fir sulphate and Na-base bisulphite pulp yields 72

9 vii - Table 11. Calculation of replication number from three positions within growth increment for two species and two pulping process 73 Table 12. Effect of elapsed time on residual lignin determinations for Douglas fir sulphate pulp and black spruce Na-base bisulphite pulp 74 Table 13. Distribution of wood lignin within increments of black spruce and Douglas fir based on extractive-free water-free weight 75 Table 14. Distribution of wood specific gravity within increments of black spruce and Douglas fir based on extractive-free oven-dry weight 76 Table 15. Sulphate raw pulp, residual lignin and carbohydrate yields within Increment No.31 (heartwood) of black spruce at three cooking times 77 Table 16. Sulphate raw pulp, residual lignin and carbohydrate yields within Increment No.47 (sapwood) of black spruce at one cooking time 78 Table 17. Sulphate raw pulp, residual lignin and carbohydrate yields within Increment No.30 (heartwood) of Douglas fir at three cooking times 79 Table 18. Sulphate raw pulp, residual lignin and carbohydrate yields within Increment No.57 (sapwood) of Douglas fir at one cooking time 80 Table 19. Na-base bisulphite raw pulp, residual lignin and carbohydrate yields within Increment No.31(heartwood) of black spruce at three cooking times 81

10 - VI11. - Table 20. Na-base bisulphite raw pulp, residual lignin and carbohydrate yields within Increment No.47 (sapwood) of black spruce at one cooking time 82 Table 21. Na-base bisulphite raw pulp, residual lignin and carbohydrate yields within Increment No.30 (heartwood) of Douglas fir at three cooking times 83 Table 22. Na-base bisulphite pulp, residual lignin and carbohydrate yields within Increment No.57 (sapwood) of Douglas fir at one cooking time 84 Table 23. Regressions of micro-specific gravity on position within growth increments 85 Table 24. Regressions of wood lignin on position within growth increments 86 Table 25. Regressions of sulphate pulp yield on position within growth increments 87 Table 26. Regressions of Na-base bisulphite pulp yield on position within growth increments 88 Table 27. Regressions of sulphate pulp carbohydrate yields on position within growth increments 89 Table 28. Regressions of Na-base bisulphite pulp carbohydrate yields on position within growth increments examined 90 Table 29. Analysis of variance on species, wood zone and specieswood zone interactions for raw pulp, residual lignin and carbohydrate yields in sulphate pulping (12 min max) 91

11 - ix - Table 30. Analysis of variance on species, wood zone and species-wood zone interactions for raw pulp, residual lignin and carbohydrate yields in Na-base bisulphite pulping (240 min max) 93 Table 31. Analysis of variance on species, growth zone and speciesgrowth zone interactions for raw pulp, residual lignin and carbohydrate yields in sulphate pulping by combining cooking times and wood zones 95 Table 32. Analysis of variance on species, growth zone and speciesgrowth zone interactions for raw pulp, residual lignin and carbohydrate yields in Na-base bisulphite pulping by combining cooking times and wood zones 97 Fig. 1. Relationship between various sulphate raw pulp factors and relative position within growth increments of black spruce and Douglas fir 99 Fig. 2. Relationship between various Na-base bisulphite raw pulp factors and relative position within growth increments of black spruce and Douglas fir 100

12 - X - ACKNOWLEDGMENTS The writer wishes to express his gratitude to Dr. J.W. Wilson, Professor, Faculty of Forestry, for his-guidance and supervision during planning, experimental phases and final reporting of this thesis, as well as his continuous encouragement during the past three years. The writer is also greatly indebted to Dr. R.W.. Wellwood, Professor and Dr. A. Kozak, Associate Professor, Faculty of Forestry; Dr. R. Branion, Assistant Professor, Department of Chemical Engineering; and Dr. R.W. Kennedy, Part-time Associate Professor, and Research Scientist, Forest Products Laboratory, Vancouver, B.C., for reviewing this document. Thanks are also extended to Mr. J;E. Tasman, Head, Analysis and Testing Division, Pulp and Paper Research Institute of Canada, Pointe Claire, P.Q., for conducting the micro-scale pulping trials. The writer also gratefully acknowledges the financial support from the Faculty of Forestry, University of British Columbia, the Canada Department of Forestry and the National Research Council of Canada. Finally, my wife 1 s continuing encouragement is also appreciated.

13 - 1 - INTRODUCTION I.tt has been shown that yields from chemical pulping of coniferous woods are correlated significantly with pulp and paper physical and chemical properties (2, 3, 33, 56, 82, 99). Other investigations have been carried out to determine coniferous wood pulp yields and properties by various pulping processes as regards differences between and within species, as well as between gross positions within individual trees (75). Pulp yields, under pulping conditions leaving 3 to 67, residual lignin, are shown to vary between and within species (12, 64, 75). For the same wood, pulp yields vary by wood zones (16, 71) and differ as regards tree growth environments (83). Similarly, pulp yields increase with age of tree for certain species (70). Some authors have found that pulp yield decreases with increasing wood specific gravity, while others report no or only slight correlation between these variables. As example, sulphite digestion of Norway spruce wood (Picea abies (Mill.) B.S.P.) with specific gravities between 0.35 and 0.54 indicated that pulp yield at the nominal 507, level decreased steadily with increasing specific gravity (47). In contrast, sulphate pulp yields with slash pine (Pinus elliotii Engelm.), loblolly pine (Pinus taeda L.) and longleaf pine (Pinus palustris Mill.) at the nominal 607, level showed that high specific gravity wood gives slightly higher pulp yields (16). In controlled experiments it is always found that pulp yield is highly correlated with residual lignin content for various coniferous woods ( 6, 15, 57).

14 - 2 - Pulp processing factors such as cooking liquor composition and liquor/wood ratio, rate of temperature rise and maximum temperature, pressure and time also greatly affect wood pulp yields (4,5,6,22,31,33, 35,48,53,84). This study was designed to keep wood and processing variables, such as cooking liquor composition, liquor to wood ratio, temperature rise and maximum temperature, as well as wood section size, constant, so as to achieve three purposes. These were to: a) Study raw pulp and carbohydrate yield patterns at different yield levels within black spruce (Picea mariana (Mill.) B.S.P.) and Douglas fir (Pseudotsuga menziesii (Mirb.) Franco.) heartwood growth increments. b) Contrast pulp yield variations between heartwood and sapwood growth increments taken from the same stem and pulped under the same conditions. c) Compare pulp yield variations between black spruce and Douglas fir examined at the intra-incremental level. Meeting these objectives required development of micro-pulping and micro-analytical techniques.

15 - 3 - REVIEW OF LITERATURE Pulping variations between and within coniferous species, wood zones and growth zones depend on morphological differences such as cell external dimensions, cell wall thickness and chemical composition, as well as cell wall molecular organization (21,37,39,45,55,67,76,95,96). These sources of variation seem to be related to environmental, physiological and genetic factors (45,100,101). A. Variation of Chemical Pulp Yields Among Coniferous Woods Various coniferous woods obviously differ in wood element size, element proportioning, chemical composition and morphological structures (59,96), all of which depend on heredity and growing conditions (100,101). The sulphite process is sensitive to properties of the starting material, wherein difficulties are introduced in pulping certain woods containing extractives which condense lignin, or that have penetration barriers (75). In contrast, sulphate pulping has no species restrictions, while acid bisulphite pulping above ph 4.0 is not restricted by wood extractives. Pulp yields at regular delignification levels largely relate to wood lignin and extractive contents, which differ between species. Cole, Zobel and Roberds (16) compared pulp yields and properties of three pine species by sampling an even-aged stand on a relatively uniform site. Their

16 - 4 - results showed that under the same kraft pulping conditions, loblolly pine gave higher yields than slash pine, and that slash pine gave higher yields than longleaf pine for nominal 50 and 60% yield levels. Their explanation was that the lower yields for slash and longleaf pine were caused by higher extractive contents than contained in loblolly pine wood. Variation in chemical composition within species is known to be significantly affected by environmental factors. Cellulose content, which is highly correlated with chemical pulp yield, is subject to variation as regards site, geographic region and growth rate (101). Trees with higher latewood portion give slightly higher pulp yields because latewood usually contains more cellulose (66). According to one author (75), lower cellulose content in fast growing trees gives lower pulp yields compared to trees with regular growth rate. In general, trees with high cellulose and hemicellulose contents give higher pulp yields in all commercial pulping processes. B. Chemical Pulp Yield Differences Between Coniferous Wood Zones Pulp yields are significantly related to wood chemical composition, which is known to vary in radial direction from the stem pith to periphery. Cellulose content was investigated by Wardrop (87) across wood zones of Monterey pine (Pinus radiata D. Don) and Douglas fir Cross and Bevan cellulose increased rapidly in the first 10 to 15 years from the pith outward and then increased slowly thereafter. Similar results for Douglas fir Cross and Bevan cellulose contents were also obtained by Kennedy

17 - 5 - and Jaworsky (46). The lowest cellulose yields were found in the first 15-year section from pith, following which cellulose content fluctuated in succeeding years. Zobel and McElwee (101) reported a significant increase in alpha-cellulose and water-resistant carbohydrates from corewood to mature wood of loblolly pine. The same species was examined as regards chemical composition by Stamm and Sanders (79), who showed that heartwood had considerably higher extractives and lower alpha-cellulose content than sapwood. At normal delignification levels, chemical pulps retain most of the cellulose portion so that pulp yields are expected to increase significantly from corewood to mature wood. In practice, yields from heartwood and sapwood are usually slightly different. This was confirmed for sulphite cooking of white spruce (Picea glauca (Moench.) Voss.) as 53 and 527,, respectively, for heartwood and sapwood (98). For species containing heartwood extractives which are sulphite pulping inhibitors, raw pulp yield differences between wood zones can be remarkably different with those from heartwood being higher. As for sulphate pulping, heartwood usually gives 2 to 37, lower yield than sapwood (75). C. Coniferous Earlywood-Latewood Chemical Pulp Yields Variations across growth increments relate to morphological and chemical differences. In coniferous woods, the conspicuous difference between earlywood and latewood is that the former is comprised of thinwalled tracheids with large cell lumens, whereas the latter has tracheids with thicker walls and smaller lumens in radial direction. Studies on

18 - 6 - heterogenety of the cell wall structure have shown that the P, and layers remain relatively constant in thickness from earlywood to latewood, whereas the layer which forms the bulk of the prosenchyma cell wall and has high cellulose content, significantly increases from earlywood to latewood (32). Similar patters have been found for specific gravity variations within increments of various coniferous woods (38,39,41). Comparisons of earlywood-latewood chemistry using data from the literature are shown in Table 1. In most cases, holocellulose is slightly higher and alpha-cellulose considerably higher in latewood while lignin content is significantly higher in the earlywood (1,32,36,58,60, 74,79). According to Table 1, data for different coniferous woods show higher extractives in earlywood (32,36,74). In optimum processing, pulp yields are directly correlated with holocellulose, and inversely related to wood lignin and extractive contents. During chemical pulping procedures, wood extractives are lost at various stages of cooking, but usually before advanced delignification (7). Quantitative analyses of earlywood and latewood pulp yields are rare and confusing. Data available from the literature for matched earlywood-latewood pulps made from coniferous woods are shown in Table 2. Quantitative differences in chemical pulp yields within growth zones were first reported in 1927 by Hagglund and Johnson (30). Three

19 - 7 - different cooking times were used with separated earlywood and latewood of Norway spruce pulped by Ca-base sulphite. Their results showed 51.3, 52.5 and 50.0% yields for earlywood corresponding to 51.0, 52.2 and 47.9% yields for latewood, respectively. Thereby, earlywood seems to have given higher yield at these cooking levels. Further, Ca-base sulphite processing of southern pine (Pinus ap.) sapwood by de Montigny and Maass (20) showed 51.5 and 44.37, yield for earlywood and latewood, respectively. Yean and Goring (98) subdivided western hemlock (Tsuga heterophylla (Raf.) Sarg.) growth zones into three parts and pulped these by Na-base bisulphite. Their results showed that pulp yield increased slightly from earlywood to the transition zone, then to the latewood with values at 43, 44 and 457=, respectively. As for sulphate pulping, yields within growth zones have been reported by several authors for different species. Investigations have shown that earlywood may or may not give slightly higher yield than latewood at commercial sulphate cooking levels. Watson and Dadswell (86) reported 51% yield for earlywood and 507. for latewood of Monterey pine. Ahlm and Leopold (1) obtained the opposite result with loblolly pine in that earlywood gave lower yield than latewood, with values corresponding to 46.0 and 48.1%,, respectively. This result was confirmed by Mcintosh (60), who obtained 45.6 and 47.17, yields, respectively, for earlywood and latewood from the same species.

20 - 8 - MATERIALS AND METHODS A. Wood Sample Preparation Mature wood samples were taken at breast height from two coniferous trees. The black spruce (Picea mariana (Mill.) B.S.P.) sample originated from St. Michels des Saints, P.Q., while the Douglas fir (Pseudotsuga menziesii (Mirb.) Franco) sample was obtained from the University of British Columbia campus. Tree characteristics and details on growth increments examined are given in Table 3. Green specimen blocks of each wood were taken from sapwood and heartwood. Wood blocks were about 4-in. long, %-in. wide in tangential direction and 1-in. in radial direction. The wood blocks were submerged in water at room temperature, and a vacuum was applied for ten hours or until the blocks sank. Saturation provided a satisfactory softening pretreatment for microtome sectioning. The growth increments chosen for examination were cut into serial tangential sections of 100 ^ 20 micron thickness, except where special thicknesses were required. Sections from one black spruce and one Douglas fir heartwood block were used for examining various sulphate pulp yield levels, while a second heartwood block of both species provided materials for Na-base bisulphite pulping, as well as specimens for intra-incremental specific

21 - 9 - gravity and wood lignin determinations. Single sapwood blocks were used for the same purposes. B. Latewood Percentage Determinations Latewood percentage was determined for the growth increments studied according to Mork's definition for spruce (65) which states that latewood cells are those in which two times the double tangential wall thickness is equal to or greater than the radial lumen diameter. C. Wood Micro-specific Gravity and Micro-lignin Determinations 1. Wood micro-specific gravity methods Wood specific gravity profiles were obtained for the growth increments studied. Ten positions were sampled within increments to represent earlywood, transition and latewood zones. Micro-specimens were sampled by punching out wet circular samples with a 3/16-in. diameter die. section sampled. Two specimens were punched from the centre of each Procedures for determining micro-specific gravity have been described in a previous report (13). The calculation is based on extractive-free, oven-dry weight and green volume. 2. Wood micro-lignin methods Lignin profiles across growth increments were determined by following the Johnson, Moore and Zank method (42). Briefly, ten ml. of 257o acetyl bromide-acetic acid is used to digest each extractive-free wood sample at 70 * l^c for 30 min. The mixture is cooled in cold

22 water and transferred to a 200 ml volumetric flask containing 9 ml of 2 M sodium hydroxide and 50 ml of acetic acid. One ml of 7.5 M hydroxylamine hydrochloride is then added and the mixture is diluted to 200 ml with acetic acid. The prepared solution is immediately measured for absorbance at 282 mu. Wood lignin absorptivity depends on species and growth zone as shown in Table 4.. Samples were taken from ten extractive-free wood sections for each increment. Specimens were punched from sections by the above mentioned 3/16-in. die and dried in vacuum over P^O^at room temperature. Wood sample weights of mg for earlywood and mg for latewood were obtained on a Cahn electrobalance having sensitivity of 0.01 mg and contained in a desiccator. Each position required one determination (95). D. Determining Micro-pulp Yields 1. Micro-pulping procedures a. Wood sample preparation and pulping Extractive-free wood specimens were prepared by treating the entire lot of specimens in sequence with diethyl ether, ethyl alcohol, and hot distilled water as described in a previous report (13). Sections from each heartwood increment were grouped as consecutive sets of three and randomly assigned into three groups for the three cooking levels to be examined. Each group contained 10 sections representing 10 positions across the increment. For sapwood, two groups of

23 positions were used for each increment for a single level of sulphate pulping and another similar set for a single Na-base bisulphite cook. Each wood section was divided into two equal pieces to provide two replications. The segmented extractive-free wood samples were vacuum dried over P^O^ at room temperature for three days to obtain constant weight. They were then rapidly moved to a desiccator, which was entered into the desiccated glove box containing the Cahn electrobalance. Weighings were done at the 100 mg range which has sensitivity of 0.01 mg. Each pair of wood specimens was randomly positioned on a 5 x 15-in., 150-mesh stainless steel screen which was rolled to less than 5/8-in. diameter. Each rolled set of wood samples was assigned to a given pulping procedure as shown in Table 5. For each pulping process three rolls contained materials for 10 positions, while two further rolls each represented 20 positions. Each roll was made up to equal wood weight by adding additional wood sections, which gave a constant liquor to wood ratio. The packages of wood samples were conditioned in a humidity chamber for two days under 80 F dry bulb temperature and 2 F wet bulb depression which provided 90% relative humidity. Moisture contents of 17.9 to 18.1%. were obtained as determined by taking weight differences before and after conditioning. The moisture content of each roll is shown in Table 5. The wood sample rolls were immediately sealed in polyethylene bags to avoid

24 moisture change before cooking. The micro-scale sulphate and Na-base bisulphite pulps were produced by the Pulp and Paper Research Institute of Canada, Pointe Claire, P.Q. A rocking electrically heated cooker holding six bombs with 5/8-in. internal diameter and 5-in. length was used. Two cooks were conducted under conditions shown in Table.6 with different times at maximum temperature to obtain yield levels desired. At prescribed times, the bombs were removed from the block and quenched in cool water. Liquors were drained away and the wrapped' parcels were placed in 200 ml beakers with distilled water. After this, the wash water was clean, indicating removal of all cooking liquor. Again the packages were sealed in polyethylene and returned to Vancouver for further processing. b. Pulp yields, residual lignin and pulp carbohydrate yields The pulped wood segments were air-dried, then packages were opened and individual samples were removed and stored in vials. samples were dried under vacuum over ^2^5 anc * we fgbed i- n Again the the oven-dry condition. Per cent pulp yields were calculated based on the initial extractive-free water-free weight. Residual lignins were determined for the cooked micro-samples by a modified Johnson, Moore and Zank technique (42). Carbohydrate yields were obtained by subtracting residual lignin content based on wood from average pulp yield for each position.

25 Development of procedures a. Micro-pulp yield As noted, the pulping procedures kept all cooking variables constant except time at maximum temperature so that pulp yield variations within and between growth zones could be compared. In order to evaluate wood variables which might affect micro-pulp yield and composition, a series of preliminary experiments were conducted to examine micro-pulping variables. These included study of effects of wood section thickness, screen mesh size and wood sample position in packages. In addition, early experiments provided information for calculating replication to ensure statistical validity, and allowed proper setting of time at maximum temperature to obtain desired yields. (1). Effect of wood section thickness on pulp yield For sulphate pulping of Norway spruce and Scots pine (Pinus sylvestris L.) to 45 to 557c yield level, i.e., 4 to 57, residual lignin, Colombo et al. (17) observed that wood chip thicknesses between 3 and 8mm did not cause differences in delignification degree and screened pulp yield at constant cooking time. Chip thicknesses above 8 mm, however, significantly affected screened pulp yields and residual lignin contents. The effect of wood section thickness on pulp yields was studied in a preliminary trial by using various thicknesses of sections randomly sampled from black spruce and Douglas fir earlywood and latewood and cooked in both sulphate and Na-base bisulphite liquors. The section thicknesses were 50-10, 110 * 10, 210 * 10 and microns. Specimens of

26 different thickness were randomly positioned on a 150-mesh screen, wrapped as packages and pulped to 50% target yield. The analysis of variance for sulphate and Na-base bisulphite pulps in Table 7 shows that pulp yields were not affected by section thicknesses from 50 to 410 microns. (2). Effect of screen mesh size on pulp yield In order to determine the effect of different screen mesh sizes on pulp yields, an experiment was designed to examine mesh sizes of 40, 100 and 150 when containing black spruce wood sections (100 ^ 20 microns) subjected to sulphate cooking. Results of analysis in Table 8 show that these mesh sizes had no significant effect on pulp yield. (3). Effect of sample position on screen In this experiment, 100 ^ 20 microns thickness wood sections were randomly positioned on a 150-mesh 5 x 15-in. screen. The between positions effect (depth in package) and within position (edge vs. central position) were studied with black spruce and Douglas fir wood sections cooked in both sulphate and Na-base bisulphite to the 55 ^- 5% yield level. Two paired adjacent sections from earlywood, transition wood and latewood of an increment were randomly sampled, divided into four subsections and randomly positioned on the screen. Residual increment sections were used to make up wood volume.

27 Analysis of variance was used to test effects of different screen position on pulp yields. The results in Table 9 show that between position effects within a package were non-significant, except for Douglas fir earlywood and transition zone. These exceptions might have been caused by wood variation between two adjacent sections rather than the between position effect, because at the 55 ^- 57» yield level Douglas fir pulp yields within increment increased markedly from earlywood to the initial latewood then decreased gradually to the end of the growth zone. As a further part of the trial each section was divided into four sub-sections and two of these were randomly positioned at screen edges and the other two were placed in central positions. The combined data for species and pulping processes were subjected to analysis of variance with results shown in Table 10. The effect of within position and its interaction with species and pulping process did not significantly affect pulp yield. (4). Determination of replication number An experimental series was carried out to determine suitable replication numbers by both pulping procedures for each position examined within a growth increment. As described in the previous experiments, two paired randomly selected adjacent sections from earlywood, transition zone and latewood of black spruce and Douglas fir were cut into eight subsections for each of the three positions within growth zones. The wood samples were cooked by both sulphate and Na-base bisulphite to 55 * 57. yield.

28 The replication number for various positions within growth zone was calculated by using the following equation (80): N = t x s /d Where: N» replication number t = Student's t-value at n-1 degrees of freedom and 957. probability level s = sample variance d = ( mean x 0.0p)/2, 57. allowable error (P) being used. Yield replication numbers calculated for species and pulping processes are shown in Table 11. The average calculated number is 0.95, or 1. By using two replications, the allowable error (P) was reduced to 3%. b. Residual micro-lignin The ultraviolet spectrophotometrie absorbance measurement for wood pulp residual lignin determinations was originally developed by Johnson et. a_l (42) for micro-wood samples. They observed difficulties when applying their method to pulps. The original method was re-examined and used for determining lignification patterns within coniferous wood growth zones by Wu and Wilson (96). An application of the method to wood pulp residual lignins was reported recently by Marton (61), who applied a correction factor to compensate for background absorbance at zero kappa number, which was claimed to give precision of ^ 2%. The Uohnson et. al. (42) procedure was followed in the present study for estimating pulp residual lignin, except that some modifications were

29 included. Briefly, a micro-pulp sample of known weight (about 30 mg) was placed in a 30 ml reaction tube with notched glass stopper. Ten ml of freshly distilled 25% acetyl bromide-acetic acid reagent was added. The reaction was done at 70 1 C for 30 min with gentle swirling at 10 min intervals. The digested pulp solution was immediately cooled to 13 C in a cold water bath for 10 ^- 2 min. The cooled solution was then transferred to a 200 ml volumetric flask into which had been placed 9 ml of 2 M sodium hydroxide and 50 ml of acetic acid. Five to 10 ml of additional acetic acid was used for completing the transfer. Then 1 ml of 7.5 M hydroxylamine hydrochloride was added to the volumetric flask and the contents were diluted to 200 ml with acetic acid. After storing the solution at 20 C for 71 ^ 1 hr, the absorbance was measured at maximum peak of 282 mu on a Spectronic 505 instrument. Lignin content based on moisture-free pulp weight was computed as follows : Lignin % = (Sample absorbance - Blank absorbance) X Liters X 100 Absorptivity X Sample weight (g ) The blank absorbance was measured for a solution prepared by the same procedures except omitting the pulp sample. Pulp residual lignin absorptivities were found to depend on species and pulping processes. The absorptivities were calculated on the basis of residual lignin content. Absorbtivity = Absorbancex Liters Water-free pulp weight (g ) x Fractional residual lignin

30 Where : fractional residual lignin = 0.15 x Micro-kappa number (9). The average residual lignin absorptivities for various pulps and associated standard deviations are shown in Table 4. (1). Sample weight A proper absorbance between 0.2 to 0.8 for pulp solutions requires 0. 6 to 6.0 mg lignin per 100 ml dilution. In these experiments, the pulp sample weight selected depended on cooking time and process. Thereby, 25 to 30 mg of moisture-free pulp was needed for high yield samples, 1. e. 157, residual lignin, and 40 to 50 mg for commercial paper pulp yield levels containing about 57o residual lignin. Marton (61) used 25 to 35 mg for the commercial yield level, but only diluted to a total of 100 ml instead of the 200 ml total dilution. The relationship of different dilution volumes and sample size was examined by the original authors (42), who showed that the 100 ml, 200 ml and 1000 ml acetic acid dilutions were not much different for Douglas fir wood samples. Marton (61) pointed out that pulp samples could be reduced to 5-10 mg by lowering the final dilution volume to ml, and reducing the amount of reagent accordingly. In these experiments, the digested lignin solution was diluted to 200 ml by reagent grade acetic acid. For preparation of samples before digestion, Johnson e_t a_l (42) ground wood samples in a micro Wiley mill to pass an 80-mesh screen. The wood meal was stored in a controlled humidity chamber until equilibrium was reached. A similar treatment was used by Wu (95). For

31 treatment of wood pulps, Mar ton (61) dried samples in a 110 C oven, ground these to pass through a 20-mesh screen, then stored in a desiccator over?2^5* * n t ' l e P r e s e n t experiments, air-dry samples, combining replications for each position, were cut by scissors into lxl mm squares. The samples were dried in vacuum over P^*-^ f r three days to obtain constant weight. Weighing was conducted by using a Cahn electrobalance of 50 mg range with mg sensitivity housed in a desiccator. (2). Digestion time and pulp solubility Wood samples were almost completely dissolved in 30 min cooking as found by Johnson et al. (42) and Wu (95). Further study by Wu (95) indicated that cooking time of in relative lignin content. 30 * 5 min only caused * 1.5% difference As regards wood pulp solubility, Marton (61) showed that solubility of pulp in the digesting reagent depends on the degree of delignification. Coniferous pulps having more than residual lignin were completely soluble in 30 min, whereas bleached pulps were difficult to completely dissolve, except for bleached sulphate pine pulp. A similar observation was made in the present study, in that wood pulp solubility depended on the pulping process. After 30 min of digestion, sulphate pulps could be completely dissolved at all levels of delignif ication, but both Na-base bisulphite pulps of 57o residual lignin, were incompletely solubilized. 50 ^ 57o yield, i.e. With these an insoluble residue remained even after digestion time was extended to 60 min. (3). Elapsed time and relative lignin content It had been stated that absorbance of wood lignin solutions is con-

32 stant for certain periods up to 5 hr after dilution for Douglas fir (42), and 24 hr for amabilis fir (Abies amabilis (Dougl.) Forbes) with solutions stored at 12 1 C (95). In early experiments, it was shown that solution absorbance changed considerably after dilution. In order to study the effect of storage time, high yield Douglas fir sulphate pulp and black spruce Na-base bisulphite pulp lignin solutions were prepared and absorbances were measured at different time intervals after dilution. The solutions were stored and maintained at 20 ^ 1 C. Results are shown in Table 12. An obvious difference in absorbance readings was observed between the two pulp solutions. The Douglas fir sulphate pulp solution significantly increased in absorbance during the first 20 hr of storage; whereas the black spruce bisulphite pulp solution apparently decreased absorbance during the first 20 hr of storage, which was accompanied by a shift in maximum peak from 270 mu to 282 mu within the first 10 hr. The increasing absorbance with increasing storage time of the Douglas fir sulphate pulp solution might be due to further hydrolysis of residual lignin macro-molecules. Absorption behaviour of the black spruce bisulphite pulp solution seems complicated, although equilibrium + in absorbance was approached after about 71-1 hr of storage. In this study, pulp solution absorbances were read after storage at 20 ^ 1 C for 71-1 hr following dilution, since these values gave more stable absorbance readings. The proper time for making readings is by no means settled, nor are reasons for time-dependent changes in absorbance understood.

33 RESULTS Wood lignin contents and specific gravities across black spruce Growth Increments No.31 (heartwood) and No.47 (sapwood) and Douglas fir Increments No.30 (heartwood) and No.57 (sapwood) are given in Tables 13 and 14, respectively. Raw pulp yields, residual lignin contents and carbohydrate yields across growth increments examined at different cooking levels by two pulping processes and for two wood species are presented in Tables 15 to 22. The relationship of raw pulp yields and pulp carbohydrate yields with relative position within growth increments for black spruce and Douglas fir are shown graphically as Fig. 1 and 2 for sulphate and Nabase bisulphite cooking, respectively. Data were fitted to the model : ln Y = b + b.x + b X 2 + b X 3 o 1 z 3 Where : Y = raw pulp yields or pulp carbohydrate yields X = relative position within growth increment from beginning of earlywood A. Intra-incremental Micro-specific Gravities And Wood Micro-lignins Micro-specific gravity profiles within increments showed minimum values in earlywood, then increased abruptly at the transition zone and approached a maximum in the early latewood (Figs. 1 and 2). From Table 14, specific gravity for black spruce had average values of (0.278 to 0.693) and (0.255 to 0.734) for Increments No.31 and No.47, respectively. Douglas fir had average specific gravities of

34 (0.214 to 0.801) for Increment No.30 and (0.237 to 0.700) for Increment No.57. Curvilinear regressions of the relationship between specific gravity and relative position within growth increments are shown in Table 23. The relationship of specific gravity and raw pulp yields within growth increments are shown graphically in Fig. 1 and 2. In Table 13 the average lignin contents are recorded as 23.56%, (20.81 to 26.57%) for black spruce increment No.31 and 23.26% (21.10 to 26.19%,) for Increment No.47. The average lignin content of Douglas fir Increment No.30 was 24.12% (21.70 to 27.47%) and 26.16% (22.64 to 30.45%) for Increment No.57. The curvilinear regressions given in Table 24 show that wood lignin contents are highly significantly correlated with relative position within growth increments. B. Raw Pulp Yields Average raw pulp yields differences for different cooking times, relative positions within growth increment, pulping process, wood zones and wood species are shown in Tables 15 to Within growth increment a. Raw pulp yields Results at various cooking levels for black spruce and Douglas fir

35 heartwood growth in Fig. 1 and 2. increments, as well as for pulping processes are shown Pulp yields decreased considerably as cooking time was increased. The curvilinear regressions between raw pulp yields and relative position within increments are shown in Tables 25 and 26 for sulphate and Na-base bisulphite pulping processes, respectively. In sulphate cooking of heartwood increments No.31 for black spruce and No.30 for Douglas fir, the three different yield patterns illustrate that the position within increments at which yield was a maximum changed slightly between different yield levels. As shown in Fig.l, at 3 min cooking time at maximum temperature the peak yield was at about the 40% relative position within increment, but was shifted to the 50 to 55%, relative position when maximum cooking time was increased to 12 min. Further cooking to 32 min moved the yield maximum to the 80% relative position. For sapwood increments of black spruce (No. 47) and Douglas fir (No. 57) maximum yields were at about 60%, relative position for the medium cooking time. Yields decreased toward earlywood and latewood for Douglas fir, but only slightly toward latewood and sharply toward earlywood for black spruce (Fig.l). Table 25 shows curvilinear regressions between sulphate raw pulp yields and relative position within increment. The correlations are Hereafter referred to as min max.

36 highly significant for different yield levels of heartwood increments and for sapwood increments, except that the black spruce high yield level (3 min max.) was not significantly correlated. For Na-base bisulphite pulping, the three cooking levels for raw pulp yields within heartwood increments showed similar behaviour for black spruce and Douglas fir (Fig.2). At high yield levels, i.e. 43 and 53 min max cooking time, raw pulp yields were slightly increased from earlywood to the 30 to 407. relative position, then abruptly decreased in the transition zone and almost remained constant within latewood. As for intermediate yield cooks, i.e., 240 min max cooking time, heartwood and sapwood increments showed similar behaviour in tha.t pulp yields were sharply increased from the earlywood initial low to a maximum at the transition zone of 40 to 607, relative position, then abruptly decreased toward the latewood zone. Pulp yields within latewood zones were more or less constant. Curvilinear regressions of Na-base bisulphite raw pulp yields and relative position within growth increments of black spruce and Douglas fir are given in Table 26. At high yield levels, i.e. 43 and 53 min max cooking time, raw pulp yields were either significantly or highly significantly correlated with position for both black spruce Increment No.31 and Douglas fir Increment No.57, where as the intermediate yield levels, i.e. 240 min max cooking time for heartwood and sapwood increments (No.47 and No.57 for black spruce and Douglas fir, respectively) were not significantly correlated.

37 In addition, the variation in pulp yield between earlywood and latewood was subjected to analysis of variance for each cooking level. These results are tabulated in Table 2 for comparison with various processes and yield levels from the literature. b. Residual lignin contents Residual lignins were calculated on water-free pulp and water-free wood basis as shown in Tables 15, 16 and Tables 17, 18 for black spruce and Douglas fir sulphate cooking, respectively. Similarly, Tables 19, 20 and Tables 21, 22 record values for black spruce and Douglas fir Nabase bisulphite cooking. The data from these tables show that the residual lignin distribution within increments are different between species and pulping processes. Black spruce sulphate pulp cooked for 3 and 12 min max gave maximum residual lignin based on over-dry pulp in earlywood at about the 207=. relative position. This greatly decreased to the transition zone at 607, relative position, then increased to the end of the growth increment. Douglas fir high-yield sulphate cooking (3 min max) showed the highest residual lignin content in earlywood, which sharply decreased to the transition zone, then remained constant through the latewood. At 12 min max cooking level, the residual lignin peak was in the transition zone and decreased toward the earlywood and latewood (Table 17, 18). The lowest yield level (32 min max) gave a very low, essentially equal residual lignin content throughout the whole increment (Table 17). Residual lignin contents within increments showed different

38 behaviour between species. Black spruce heartwood pulps from Increment No.31 at high-yield levels, i.e. 43 and 53 min max, gave residual lignins which decreased from earlywood to the transition zone, and then increased toward the latewood. A similar pattern was found when cooking to 240 min max, in that residual lignin was still at a minimum level in the transition zone (Table 19). The sapwood Increment No.47 cooked to 240 min max showed slightly higher residual lignin content in the transition zone (Table 20). When Douglas fir heartwood Increment No.30 was cooked to high-yield levels, i.e. 43 and 53 min max, the highest residual lignin contents (based on water-free pulp) were in the earlywood and then abruptly decreased toward the transition zone with minimum at the 60 to 807c relative position after which lignin increased through latewood (Table 21). At the intermediate yield level, i.e. 240 min max, the behaviour was identical for heartwood (No.30) and sapwood (No.57) increments in that residual lignin contents sharply decreased from earlywood to the transition zone then slightly increased in the latewood zone (Tables 21 and 22). In general, residual lignin contents based on oven-dry pulp were directly related to the different raw pulp yield levels and pulping process used. At the commercial paper pulp yield level (45 to 507o raw pulp yeild) residual lignin contents mostly remained constant through the whole increment or were only slightly higher at the transition zone. Earlywood and latewood residual lignin contents based on water-free wood were tested by analysis of variance for each cooking level examined. The earlywood and latewood residual lignin contents are tabulated in

39 Table 2 for comparison with various pulp yield levels, processes and species from the literature. c. Pulp carbohydrate yields Pulp carbohydrate yields were obtained by subtracting the residual lignin content from the average raw pulp yield, with both based on waterfree wood. increments. This was done for each position sampled within growth Results are tabulated in Tables 15 to 22 and plotted in Figs. 1 and 2 for sulphate and Na-base bisulphite pulps, respectively. The carbohydrate yield patterns within increments have similar shapes to those of average raw pulp yields, except at high-yield levels which show a slight shift of maxima toward the initial latewood (Figs. 1 and 2). Curvilinear regression of sulphate pulp carbohydrate yield and relative position within growth increments of black spruce and Douglas fir are given in Table 27. The correlations are highly significant for different yield levels of heartwood and for sapwood increments of both species examined. Table 28 shows curvilinear regressions between Na-base bisulphite pulp carbohydrate yields and relative position within growth increments. Black spruce and Douglas fir pulp carbohydrate yields are either highly significantly or significantly correlated with relative position for heartwood and sapwood increments except for the black spruce sapwood increment. 2. Between wood zones In order to test the raw pulp yields and pulp chemical composition between heartwood and sapwood increments, these wood sections were pulped

40 in the same set for each process. As shown in Table 5, sulphate pulps were cooked to 12 min max and Na-base bisulphite pulps were cooked to 240 min max. The results are plotted as part of Figs. 1 and 2. These graphs compare directly pulp and carbohydrate yields between heartwood and sapwood increments of the same stems. Black spruce and Douglas fir raw pulp yields, residual lignin contents and pulp carbohydrate yields based on extractive-free water-free wood were further analysed for comparisons between wood zones. These analyses are shown in Tables 29 and 30 for sulphate and Na-base bisulphite pulps. For sulphate pulping, the results of analysis of variance for 12 min max cooking of heartwood and sapwood increments of black spruce and Douglas fir are shown in Tables 29A, 29B, and 29C. These correspond to raw pulp yields, residual lignins and carbohydrate yields based on extractive-free water-free wood as related to intra-incremental position. The combined raw pulp yields for black spruce and Douglas fir data are 59.22% and 57.89% corresponding to heartwood and sapwood increments respectively. For separate species, black spruce heartwood Increment Hb.31 gave 57.44% and sapwood Increment No.47 gave 57.93%; Douglas fir heartwood increment No.30 was 60.99%. and sapwood Increment No.57 was 57.86%,. The analysis of variance in Table 29A shows that combination of heartwood and sapwood data for two species is not significant, but that the species X wood zone interaction is significant.