Increasing Softwood Pulp Yield by Minimizing Primary Peeling of Wood Carbohydrates Using Sodium Methyl Mercaptide Before and During Kraft Pulping

8 The objective of this work was to determine the effect of sodium methyl mercaptide (SMM) on the 9 minimization of peeling reactions of southern pine chips in the kraft pulping process. Two methods were 10 evaluated for SMM addition to the pulping process: 1) pre-treatment before pulping or 2) co-addition with white 11 liquor. The effect of SMM charge, pre-treatment temperature and time, and pH of pre-treatment liquor was 12 studied. 13 The experimental results showed about 1.5 to 2.5 % (on O.D. (oven dry) wood basis) increase in the 14 pulp yield after pre-treatment with or co-addition of 4.38% SMM (on O.D. wood basis). The use of 4.38% SMM 15 allowed a decrease of the white liquor effective alkali charge (EA, on O.D. wood basis) by 3%. 4.38% SMM 16 charge seemed to be optimum for the pre-treatment. Pre-treatment at lower pH resulted in a significant decrease 17 in yield and an increase in rejects. The increase in pulp yield was mostly due to the increased retention of 18 cellulose and xylan. The retention of galactoglucomannan was negligible. About 80% of the cellulose yield 19 increase is due to the suppression of primary peeling. The remainder (0.3-0.4% of the yield increase (on O.D. 20 wood basis) is due to reduced alkaline hydrolysis and subsequent secondary peeling.


24
Kraft pulping is the most widely used pulping technology due to flexibility in the choice of woody feed 25 stock species, highest fiber strength, and a highly efficient recovery system for the recovery of the pulping 26 chemicals and energy. However, kraft pulping leads to significant degradation of hemicelluloses as compared to 27 cellulose. Therefore, the pulp yield is usually less than 50% as opposed to a theoretical maximum of about 70 to 28 75% when both cellulose and hemicelluloses are not degraded. Since the cost of wood used for kraft pulping 29 process accounts for over 50% of the production cost of softwood kraft pulp, followed by about 25% for the 30 capital cost, it is clear that the most effective approach to decrease the production cost of softwood kraft pulp is 31 to increase the pulp yield (Kangas et al., 2014).

32
For softwoods, primary peeling leads to about 10% loss of cellulose (4% on O.D. wood basis) and 50% 33 loss of galactoglucomannan (8% on O.D. wood basis) (Paananen 2009). Primary peeling of cellulose and 34 galactoglucomannan occurs during the initial phase of kraft pulping at high alkalinity (pH ~ 14) and at 35 temperatures higher than about 80 0 C ( Montagna et al., 2013). Softwood xylan is more stable against peeling 36 because the easy cleavage of the arabinose side chain leads to an alkali stable metasaccharinic acid end group.

37
However, a small amount of xylan is still lost during impregnation due to limited solubility of xylan in caustic. respectively (Sjostrom E. 1993). However, these technologies are not generally adopted in industry because 53 sodium borohydride is very expensive, and the use of polysulfide results in the formation of corrosive sulfur-54 oxides.

55
The only technology that was very close to commercialization is the H2S pre-treatment. It was operated 56 on semi-commercial pilot scale with 13.5 tons per day capacity in a continuous two-vessel Kamyr digester at the 57 Harmac mill in Canada for 4 to 5 weeks and about 267 tons of pulp was produced (Cox 1974). However, despite 58 a 15 to 25% rate of return on investment (ROI), this technology was not commercially implemented presumably 59 due to safety concerns regarding the high pressure H2S system.

60
Building on the H2S pre-treatment concept, pre-treatment of wood chips with sodium methyl 61 mercaptide/methyl mercaptan (SMM/MM) was investigated. Based on personal discussions with renowned 62 wood chemist Professor Thomas Rosenau at BOKU, Vienna, Austria, and using the similarity between SMM 63 and NaHS, a plausible mechanism for the stabilization of reducing end groups by SMM/MM was proposed as 64 shown in Figure 1. However, since CH3Sis a stronger nucleophile than HS -, the SMM pre-treatment 65 temperature can be significantly lower than 140 0 C, the optimum temperature for the H2S pre-treatment. The process. In the first case, MM was used as an additive in polysulfide kraft pulping to increase the overall 76 sulfidity and accelerate the rate of delignification. When MM was added to the polysulfide-kraft process, the 77 pulp yield increased by 1% and the kappa number was reduced from 27.9 to 21.7. Since there is about 0.15% 78 (on O.D. wood basis) change in the pulp yield for each kappa unit (Sixta 2006), it shows that MM addition led

87
Pulping experiments were performed using southern pine chips obtained from International Paper and 88 WestRock US mills. The chips were screened using a chip classifying system and the fraction with thickness 89 and width ranging from 3 to 5 mm and 3/8 to 9/8 inch, respectively, was collected. Screened chips were then air-90 dried to about 90 to 95% consistency using a fan at room temperature. The air-dried wood chips were mixed and 91 stored in plastic bags.

92
All experiments were performed in quadruplet using four 235 mL cylindrical rocking digesters. About 93 30 grams of wood chips (on O.D. basis) was used in each digester. In order to avoid the handling of MM gas, an 94 aqueous solution of SMM (21% w/w) containing also a small amount of NaOH (0.4% w/w) with pH 12.9 was  At the end of pre-treatment, the digesters were cooled to room temperature (using a water bath) while 101 make-up white liquor was being prepared. The final pH of the pre-treatment liquor was between 9.5 to 10.5. The 102 total liquor-to-wood ratio after adding make-up white liquor was 4.5 L/kg. It should be noted that for some 103 experiments, the SMM solution was charged together with the make-up white liquor to simulate more practical 104 pulp mill conditions. After adding the make-up white liquor (Sulfidity: 30%, Causticization Efficiency: 80%), 105 the digesters were placed in the oil bath for an hour at 115 C to achieve impregnation of white liquor into chips.

106
At the end of impregnation, the digesters were removed from the oil bath and cooled to room temperature while 107 the oil bath was being heated to the cooking temperature of 170 C. The digesters were then immersed in the bath 108 and cooked to target H-factor of about 1960 hrs.

109
After cooking, the digesters were cooled in a water bath and their weights were recorded to check for 110 leakage. The digesters were stored in a refrigerator until the pulp was washed. After separating black liquor 111 from the macerated wood chips, the pulp was disintegrated and washed using a pulp disintegrator and a 75-mesh 112 nylon bag, respectively. The resulting pulp suspension was filtered using a buchner funnel to make a pulp pad 113 which was then air dried to determine the total pulp yield. The pulp was screened using a Somerville-type screen 114 with 0.15 mm width slots.

115
In order to obtain statistically meaningful data on pulp yield, the mean yield was calculated using four

136
It should be noted that when estimating the total composition of biomass samples using the two-step 137 hydrolysis process, the sum of individual components does generally not exactly add up to 100% of the initial mass. To account for this difference, the mass of carbohydrates measured using chromatographic analysis were 139 normalized using the normalization factor(Equation 1), which is defined as the ratio of the experimentally 140 measured mass of total carbohydrates (Equation 2) to the mass of total carbohydrates calculated by difference 141 between the pulp weight and total lignin weight (Equation 3). The measured masses of carbohydrates were then 142 divided by the normalization factor to estimate the normalized masses. Normalizing the carbohydrate 143 composition data using this method ensures that the sum of individual components adds up to 100% and implies 144 that all experimental and systematic errors associated with the biomass analysis are concentrated in the 145 carbohydrates composition data. This is reasonable because the analysis of total lignin content using the acid-146 hydrolysis procedure is more accurate and is less subject to errors. 147 148

Effect of various process variables on pulp yield 224
Effect of pre-treatment temperature 225 As shown in Figure 4, the pretreatment temperature was varied between 80 0 C to 130 0 C. For the experiments 226 with pretreatment at 80 0 C, the pretreatment time was increased to 90 minutes to compensate for the lower 227 pretreatment temperature. In addition, for the experiment involving co-addition of 4.38% SMM and 12% EA at 228 80 0 C, the usual white liquor impregnation (1 hour at 115 0 C) was omitted. Since the data points for all the 4.38% 229 SMM and 12% EA experiments lie on the same line of total yield vs kappa plot, it can be concluded that the pre-230 treatment temperature has no effect on the total pulp yield.

242
In industrial settings, pre-treatment of wood chips with SMM prior to pulping would require an additional 243 digester and/or additional piping, which may have a negative impact on the economic feasibility of the process. 244 Therefore, it was investigated whether SMM and white liquor could be charged together. Figure 5       that SMM stabilizes some cellulose before peeling but that the glucomannan stabilization by SMM is too slow 289 compared to primary peeling of glucomannan. The residual alkali was generally within the acceptable ranges of 290 4-10 g/L. Similarly, the amount of rejects was less than 0.1% except for two experiments, pre-treatment with 291 8.75% SMM liquor followed by pulping with 9 % EA (0.33% rejects) and pre-treatment with 4.38 % SMM (pH 292 10) liquor followed by pulping with 12% EA (0.23% rejects), most likely due to insufficient alkali during

388
The increase in cellulose yield is mostly due to less primary peeling of cellulose because of conversion of the 389 reducing end groups of cellulose into alkali-stable pyran and/or alditol structures. About 20% of the cellulose 390 yield increase is due to less secondary peeling because alkaline hydrolysis during kraft cooking is reduced by 391 the lower effective alkali charge when SMM is used.