PeroxideDelignification

Delignification of aspen wood using hydrogen peroxide and peroxymonosulfate

Edward L. Springer Chemical engineer, USDA Forest Service, Forest Products Laboratory, One Gifford Pinchot Dr., Madison, Wis. 53705-2398

ABSTRACT Treatment of finely divided aspen wood with peroxymonosulfate KEYWORDS at low pH, followed by alkaline extraction, resulted in nearly complete lignin Aspen removal. Treatment of the wood with hydrogen peroxide at optimum pH (pH Delignification 11), followed by alkaline extraction, removed at most 36% of the original Hydrogen lignin. Peroxymonosulfate can be easily produced by mixing hydrogen peroxide peroxide with concentrated sulfuric acid. Peroxymonosulfuric acid Sodium hydroxide Sulfuric acid

Wood and other lignocelluloses can be (3) studied the delignification of Delignification using readily delignified using organic southern pine kraft pulp with alkaline hydrogen peroxide peroxides such as peracetic and per- hydrogen peroxide. They also found formic acids (1). With the exception that approximately half the lignin One-gram samples of mature aspen of studies using alkaline hydrogen present in the pulp could be removed. wood, ground to pass through a No. peroxide, little research has been done In both cases, the most efficient 40 mesh screen. were treated at room on delignification with inorganic delignification occurred under reac­ temperature (22°C) with a 5.0% solu­ peroxides. The purpose of this work tion conditions where no stabilizers (to tion of hydrogen peroxide adjusted to was to determine whether acidic inhibit peroxide decomposition) were either pH 2.0 with sulfuric acid or to hydrogen peroxide could delignify present and the rate of peroxide pH 11.0 or 11.5 with sodium hydrox­ aspen wood and. if not, whether decomposition was at a maximum. ide. A liquor-to-wood ratio of 10:1 was peroxymonosulfuric acid made by For comparison, we initially studied used, and the treatment time was 3 reacting hydrogen peroxide and sul­ the delignification of aspen wood with days except for the pH 11.5 condition, furic acid could delignify the wood. alkaline hydrogen peroxide. where treatment time was 12 days. Aspen wood (Populus tremuloides Acidic hydrogen peroxide is a much Samples were also treated at the Michx.) was chosen for study because stronger oxidizing agent than alka­ natural pH (pH 5.5) of the 5.006 it is easily delignified. line hydrogen peroxide. Under acidic solution of hydrogen peroxide. The A few studies have been performed conditions, the lignin macromolecule samples were stirred initially but not on delignification of lignocellulosic is extensively degraded and dissolved agitated during treatment. After materials with alkaline hydrogen by hydrogen peroxide (4, 5). It seemed peroxide treatment, each sample was peroxide. Gould (2) found that approx­ likely, therefore, that acidic solutions extracted with 1.0% NaOH at 50°C for imately half the lignin present in of hydrogen peroxide should more 3 h. The results from these treatments agricultural residues, such as wheat readily delignify wood than alkaline are shown in Table I. straw, could be solubilized when the solutions. The most selective delignification residue was treated at 25°C with an occurred at pH 11.0, where 32% of the alkaline solution of hydrogen perox­ original lignin and 17% of the original ide. The delignification was most carbohydrate were removed. Divid­ effective at pH 11.5. McDonough et al. ing the total weight of carbohydrate

January 1990 TappiJournal 175 removed by the weight of lignin I. Delignification of 40-mesh aspen wood with 5% hydrogen peroxide at a liquor to wood ratio removed (C/L ratio) gives an index of of 10·1 the selectivity of lignin removal (Ta­ ble I). Increasing the pH to 11.5 and the treatment time to 12 days resulted in a small increase in lignin removal (to 36% of the original); however, the selectivity was slightly reduced (from C/L ratio 2.2 to 2.4). Lowering the pH to 2.0 increased the lignin removal only slightly from what it was at pH 5.5. Much more lignin was removed under alkaline conditions. Residual peroxide levels were not determined. To determine the effect of treat­ ment temperature on lignin removal, runs were made at 80°C. The results from these runs are also shown in Table I. The most effective delignifi­ cation again occurred at pH 11.0; however, the selectivity of lignin removal was somewhat reduced from II. Production of peroxymonosulfuric acid by mixing hydrogen peroxide and concentrated sulfuric that at room temperature. None of the acid at ice bath temperature conditions studied resulted in low lignin contents in the final residue. Since acidic hydrogen peroxide was ineffective in delignification of aspen wood, we decided to attempt to delig­ nify the wood with peroxymonosulfur­ ic acid produced by adding sulfuric acid to hydrogen peroxide.

Generation of peroxymonosulfate Adding concentrated sulfuric acid to Several experiments were per- Delignification solutions of hydrogen peroxide results formed to determine the conditions using peroxymonosulfate in the production of peroxymonosul­ needed to produce significant quanti­ furic acid (6). This acid is a much ties of peroxymonosulfuric acid. Cold Small quantities of the solutions stronger oxidizing agent than hydro­ hydrogen peroxide (2°C) was placed containing peroxymonosulfate were gen peroxide. It can also be produced in a vial in an ice bath: concentrated carefully diluted with distilled water by adding peroxydisulfuric acid salts sulfuric acid at room temperature and used to treat 1-g samples of the to concentrated sulfuric acid. Zakis was added, and the two were tho- No. 40 mesh aspen wood at room and Neiberte (7) found that they could roughly mixed. An aliquot of this temperature (22°C). The weight of the delignify spruce sawdust to low lignin mixture was titrated using the meth- peroxide-acid mixture used for a levels using peroxymonosulfuric acid od of Greenspan and MacKellar (8) to given total weight of solution (used to in 50% H2SO4 at 20°C. They produced determine the yield of peroxymono- treat 1.00 g of wood) is shown in the the peroxymonosulfuric acid by ad­ sulfuric acid produced. Results of third column of Table III. The con- ding the ammonium salt of peroxydi­ these procedures are shown in Table centration of the peroxymonosulfate

sulfuric acid to concentrated sulfuric 11. Using 32% H2O2 and a mole-to- anion in the initial solution is given in acid. A high concentration of sulfuric mole ratio of acid to peroxide of 1:1. Column 4, and the equivalent total acid seemed to be necessary for effec­ the yield of peroxymonosulfate was percentage of hydrogen peroxide in tive lignin removal. Based on these only 8%. This indicates that the yield the initial solution is given in Column findings, hydrogen peroxide and was probably negligible when a few 8. The treated residues were extracted sulfuric acid mixtures with a large drops of concentrated sulfuric acid as described previously with 1.0%

excess of acid present should delignify were added to 5.0% H2O2 to adjust the NaOH. finely divided aspen wood. Our pre­ pH to 2.0. Using more concentrated Initial runs were made using per- vious data indicate that simply adjust­ hydrogen peroxide and a higher mole- oxymonosulfate prepared by Proce­ ing a 5.0% solution of H2O2 to pH 2.0 to-mole ratio of acid to peroxide dure 1. This resulted in very dilute with concentrated sulfuric acid did greatly increased the yield of peroxy- solutions of peroxymonosulfate con­ not result in a solution that effectively monosulfate. taining large amounts of acid and delignified aspen wood. peroxide. In the first run (Table III), the initial concentration of peroxymo­

176 January 1990 Tappi Journal III. Delignification of 40-mesh aspen wood (1.00 g) using peroxymonosulfate at 22°C

nosulfate was only 0.45%, and the All the results mentioned were from the peroxide-acid mixtures. In equivalent total concentration of obtained using peroxymonosulfate Run 7, an equivalent quantity of hydrogen peroxide was 2.3%. Even produced using Procedure 1. Run 6 Oxone was substituted for the Proce­ under these conditions, a relatively shows the result of drastically increas­ dure 3 mixture in Run 6. All other large amount of lignin was removed. ing the peroxymonosulfate concentra­ conditions were identical except that The lignin in the residue (11.8%) was tion by using Procedure 3. Comparing some hydrogen peroxide was present much lower than that for the best of Runs 6 and 4, the same equivalent in Run 6 and the pH was much lower. the hydrogen peroxide runs (16.9%), concentration of hydrogen peroxide is The results were quite similar. For and the C/L ratio (1.5) was also much present: however, the initial concen­ Run 7, the yield of residue was a bit lower than that for the best of the tration of peroxymonosulfate was higher, and its lignin content was also peroxide runs (2.2). As illustrated by greatly increased in Run 6. The somewhat higher. These significant Runs 2 and 3, increasing the concen­ treatment time for Run 6 was reduced differences were probably the result tration of peroxymonosulfate and to one day. Despite this reduced of the much lower pH in Run 6. This extending the reaction time gave even treatment time, Run 6 shows a large is reflected in the large difference in better results. Increasing the liquor­ increase in lignin removal compared residue viscosities in Runs 6 and 7. to-wood ratio from 10:1 to 25:1 gave with Run 4. This is undoubtedly an The much higher hydrogen ion con­ a slight increase in selectivity from a effect of the much higher initial centration in Run 6 undoubtedly C/L ratio 1.5 to 1.3 (compare Runs 1 concentration of peroxymonosulfate. increased the rate of carbohydrate and 4); however, this increase in Another source of the peroxymono­ degradation and also increased the selectivity came at the expense of a sulfate anion is Oxone®, a commercial rate of lignin removal. If this lower much larger quantity of peroxymono­ product sold by DuPont (E. I. DuPont pH were taken into account. it ap­ sulfate per gram of wood (2.7 g of De Nemours and Company, Inc., pears that, with regard to delignifi­ mixture compared to 1.35 g). At this Wilmington, Del., 1988). Oxone is a cation, the source of the peroxymono­ higher liquor-to-wood ratio, increas­ triple salt containing potassium per­ sulfate anion is of no importance. ing the perosymonosulfate concentra­ oxymonosulfate (2 KHSO5· KHSO4· tion and treatment time also greatly K2SO4). Delignification ofaspen wood increased lignin removal (compare with Oxone was compared with delig­ Runs 4 and 5). nification using peroxymonosulfate

January 1990 Tappi Journal 177 Conclusions peroxide); for Run 2. the figure was lund, B., and Winter, L., 1989 Oxygen 1.9 g of lignin removed for each Delignification Conference Proceed­ and recommendations ings, TAPPI PRESS, Atlanta, p. 165. equivalent gram of hydrogen perox­ 4. Chang, H-M. and Allan, G. G., Lignin, Low pH solutions of the peroxymono­ ide consumed. Because only relatively (K. V. Sarkanen and C. H. Ludwig, sulfate anion are much more effective small amounts of oxidants were con­ Eds.), Wiley-Interscience, New York, in delignifying aspen wood than are sumed in delignifying the wood. it 1971, p. 469. alkaline solutions of hydrogen perox­ might be possible to use solutions of 5. Dence, C. W., Chemistry of Delignifica­ tion with Oxygen. Ozone, and Peroxides ide. This is probably because, under peroxymonosulfate generated by mix­ (J. S. Gratzl, J. Nakano, and R. P. Singh, these conditions, peroxymonosulfate ing hydrogen peroxide and sulfuric Eds.), Uni Publishers, Tokyo, 1980, p. is a much stronger oxidizing agent acid to improve the strength of high- 199. than is hydrogen peroxide. The very yield mechanical or chemimechanical 6. Hall, P,. E., Encyclopedia of Chemical Technology. Vol. 17, 3d edn., Wiley- low pH of the peroxymonosulfate pulps or even to produce chemical Interscience, New York, 1982, p. 14. solutions. produced by mixing hydro­ pulps. Such solutions of peroxymono­ 7. Zakis, G. F. and Neiberte, B. Ya., Khim. gen peroxide with sulfuric acid, sulfate might also be used to restore Drev. (Riga) 9: 109(1971). results in marked attack on the car­ or enhance the strength of unbleached 8. Greenspan, F. P. and MacKellar, D. G., bohydrate constituents of the wood, softwood kraft wastepaper. Peroxy­ Anal. Chem. 20(11): 1061(1948). resulting in low residue yields and low monosulfate might, in addition. be viscosities. A comparison of Runs 6 used to replace chlorine and chlorine The author wishes to thank Marilyn J. Effland and 7 in Table III suggests that dioxide in pulp bleaching. It also for performing much of the experimental work. The Forest Products Laboratory is maintained adjusting the pH of the solutions might be used to delignify agricultu­ in cooperation with the University of Wisconsin. upward with a base such as sodium ral residues, such as straw and corn This article was written and prepared by U.S. hydroxide might diminish the sever­ stover, to enhance their enzymatic Government employees on official time, and it is therefore in the public domain and not subject ity of the attack on the carbohydrates digestibility. to copyright. without greatly reducing the lignin­ The use of trade or firm names in this removing ability of the solutions. publication is for reader information and does Literature cited not imply endorsement by the U.S. Department For several of the peroxymonosul­ of Agriculture of any product or service. fate runs. the quantity of oxidants 1. Poppius, K., Laamanen, L., Sundquist, consumed was determined. For Run J., Wartiovaara, I., and Kauliomaki, S., Received for review April 24, 1989. Papier ja Puu 68(2): 87(1986). AcceptedMay15, 1989. 6, 1.0 g of lignin was removed from 2. Gould, J. M., Biotech. and Bioengr. 26: the wood for each gram of oxidant 46(1984). Presented at the TAPPI 1989 Wastepaper consumed (calculated as hydrogen 3. McDonough, T. J., Kirk, R. C., Back­ RecyclingSeminar.

178 January 1990 Tappi Journal