ORGANOSOLV PULPING ESA A review and distillation study related to peroxyacid MUURINEN pulping Department of Process Engineering

OULU 2000

ESA MUURINEN

ORGANOSOLV PULPING A review and distillation study related to peroxyacid pulping

Academic Dissertation to be presented with the assent of the Faculty of Technology, University of Oulu, for public discussion in Savonsali (Auditorium L 4), Linnanmaa, on June 30th, 2000, at 12 noon.

OULUN YLIOPISTO, OULU 2000 Copyright © 2000 Oulu University Library, 2000

Manuscript received 16 May 2000 Accepted 18 May 2000

Communicated by Professor Juhani Aittamaa Doctor Jorma Sundquist

ISBN 951-42-5661-1

ALSO AVAILABLE IN PRINTED FORMAT

ISBN 951-42-5660-3 ISSN 0355-3213 (URL: http://herkules.oulu.fi/issn03553213/)

OULU UNIVERSITY LIBRARY OULU 2000 To Paula and Dijon Muurinen, Esa, Organosolv pulping, a review and distillation study related to peroxyacid pulping Department of Process Engineering, University of Oulu, FIN-90014 University of Oulu, Finland 2000 Oulu, Finland (Manuscript received 16 May 2000)

Abstract

More than 900 papers related to organosolv pulping have been reviewed in this thesis. From the information included in those papers it can be concluded that organosolv pulping processes are still in a developing stage and are not yet ready to seriously threat the position of the kraft process as the main pulp manufacturing process in the world. Distillation seems to be the main alternative as the process for recovering the solvent in organosolv pulping. A good reason for this is that using simple distillation no potentially harmful components are introduced to the process. The effect of the feed composition on the operation of a separation sequence in an organosolv process using aqueous formic and acetic acids and corresponding peroxyacids was studied. When simple distillation was used as the separation method the effect was found to be significant. The unideal nature of the formic acid-acetic acid-water mixture, the separation of which was studied, makes the ternary composition space divide into four distillation regions. Which region the feed is located in, obviously determines the economy of the distillation sequence. Shortcut calculation methods cannot be recommended for use in designing a distillation sequence for the ternary mixture studied, but they give useful information for the comparison of such sequences. They were used to choose a limited number of alternatives for studies with rigorous calculation methods. Minimum work of separation can also be used to make a satisfying estimate for the relative easeness of separation of the formic acid-acetic acid-water mixture. Thermal integration using pinch technology was also tested and found to be very useful for decreasing the thermal energy consumptions of distillation processes. Thermodynamic efficiencies for separating the formic acid-acetic acid-water mixture by simple distillation were estimated. They were found to be lower than the average value for distillation presented in literature.

Keywords: solvent recovery, pinch analysis, formic acid, acetic acid

Acknowledgements

This research work was carried out in the Mass and Heat Transfer Process Laboratory of the Department of Process Engineering, University of Oulu, Finland, in 1994-1999. I would like to thank my supervisor, Professor Jorma Sohlo, for his encouragement and valuable comments during this research. I want to acknowledge Professor Juhani Aittamaa and Dr. Jorma Sundquist for their review of the thesis and for their useful recommendations. The financial support from Oulun yliopiston tukisäätiö, Jenny ja Antti Wihurin rahasto and Tekniikan edistämissäätiö is gratefully acknowledged.

Oulu, May 2000 Esa Muurinen

Contents

Abstract Acknowledgements Contents 1. Introduction...... 13 2. Pulping with organic chemicals...... 16 2.1. Alcohols...... 16 2.1.1. Methanol...... 16 2.1.1.1. Catalyzed methanol pulping...... 17 2.1.1.2. Methanol in sulphite pulping...... 20 2.1.1.3. The ASAM process...... 21 2.1.1.4. Methanol in soda pulping...... 29 2.1.1.5. The Organocell process...... 31 2.1.1.6. Other alkaline methanol processes...... 36 2.1.1.7. Methanol lignins...... 37 2.1.1.8. Summary...... 39 2.1.2. Ethanol...... 39 2.1.2.1. Catalyzed and autocatalyzed ethanol pulping...... 40 2.1.2.2. The Alcell process...... 49 2.1.2.3. Alkaline ethanol pulping...... 57 2.1.2.4. Ethanol in sulphite pulping...... 62 2.1.2.5. Simulation of ethanol pulping...... 63 2.1.2.6. The kinetics and mechanism of delignification...... 64 2.1.2.7. Ethanol lignins...... 65 2.1.2.8. Summary...... 66 2.1.3. Other alcohols...... 67 2.1.3.1. Propanol and isopropanol...... 67 2.1.3.2. Butanol...... 68 2.1.3.3. Glycols...... 70 2.1.3.4. Summary...... 73 2.2. Organic acids...... 73 2.2.1. Formic acid...... 73 2.2.1.1. Pulping and bleaching with formic acid...... 74 2.2.1.2. The Milox process...... 75 2.2.1.3. Milox bleaching...... 79 2.2.1.4. Solvent recovery in Milox pulping...... 80 2.2.1.5. The chemistry of Milox pulping...... 84 2.2.1.6. Summary...... 85 2.2.2. Acetic acid...... 85 2.2.2.1. Acetic acid pulping and bleaching...... 85 2.2.2.2. Peroxyacetic acid...... 92 2.2.2.3. Acetosolv pulping...... 96 2.2.2.4. Acetocell pulping...... 99 2.2.2.5. Formacell pulping...... 99 2.2.2.6. Summary...... 100 2.2.3. Other organic acids...... 100 2.2.4. Salts of organic acids...... 101 2.2.5. Summary...... 102 2.3. Phenol and cresols...... 102 2.3.1. Phenol...... 102 2.3.2. Cresols...... 106 2.3.3. Summary...... 108 2.4. Ethyl acetate...... 109 2.5. Amines and amine oxides...... 110 2.5.1. Ethylenediamine...... 110 2.5.2. Hexamethylene diamine...... 111 2.5.3. Methylamine...... 112 2.5.4. Ethanolamine...... 112 2.5.5. Amine oxides...... 113 2.6.6. Summary...... 113 2.6. Ketones...... 114 2.7. Dioxane...... 117 2.8. Other organic compounds...... 120 2.9. Summary...... 122 3. Solvent recovery in peroxyacid pulping...... 124 3.1. The search for feasible separation methods...... 125 3.1.1. Wells and Rose...... 125 3.1.2. Lee...... 126 3.1.3. Barnicki and Fair...... 127 3.1.4. Null...... 128 3.2. Feasible separation methods...... 128 3.2.1. Simple distillation...... 129 3.2.2. Azeotropic distillation...... 131 3.2.3. Extractive distillation...... 131 3.2.4. Liquid-liquid extraction...... 133 3.2.5. Membrane processes...... 134 3.2.6. Adsorption...... 134 3.2.7. Summary...... 135 3.3. The effect of the feed composition on the separation of the formic acid-acetic acid-water mixture...... 135 3.3.1. Minimum work of separation...... 135 3.3.2. Vapour-liquid equilibrium data for the formic acid-acetic acid-water mixture...... 138 3.3.3. Shortcut simulation of the first separation stage...... 142 3.3.3.1. Simple distillation at 20 kPa as the first separation stage...143 3.3.3.2. Simple distillation at 101.3 kPa as the first separation stage...... 145 3.3.3.3. Simple distillation at 300 kPa as the first separation stage...... 151 3.3.3.4. Extractive distillation as the first separation stage...... 156 3.3.3.5. The first separation stage for the peroxyformic acid pulping process...... 160 3.3.3.6. Summary of the simulation results for the first separation stage...... 161 3.3.4. Shortcut simulation of the second separation stage...... 164 3.3.4.1. Simple distillation at 20 kPa as the second separation stage...... 165 3.3.4.2. Simple distillation at 101.3 kPa as the second separation stage...... 169 3.3.4.3. Simple distillation at 300 kPa as the second separation stage...... 174 3.3.4.4. Extractive distillation at 101.3 kPa as the second separation stage...... 178 3.3.4.5. The second separation stages for the peroxyformic acid pulping process...... 178 3.3.4.6. Summary of the simulation results for the second separation stage...... 180 3.3.5. Shortcut simulation of the third separation stage...... 182 3.3.5.1. Simple distillation at 20 kPa as the third separation stage.. 182 3.3.5.2. Simple distillation at 101.3 kPa as the third separation stage...... 184 3.3.5.3. Simple distillation at 300 kPa as the third separation stage...... 187 3.3.5.4. Extractive distillation with n-methyl-2-pyrrolidone at 101.3 kPa as the third separation stage...... 191 3.3.5.5. The third separation stages for the peroxyformic acid pulping process...... 192 3.3.5.6. Summary of the simulation results for the third separation stage...... 194 3.3.6. Shortcut simulation of the fourth separation stage...... 196 3.3.6.1. Simple distillation at 20 kPa as the fourth separation stage...... 197 3.3.6.2. Simple distillation at 101.3 kPa as the fourth separation stage...... 197 3.3.6.3. Simple distillation at 300 kPa as the fourth separation stage...... 198 3.3.6.4. The fourth separation stages for the peroxyformic acid pulping process...... 199 3.3.6.5. Summary of the simulation results for the fourth separation stage...... 200 3.3.7. The recycle stream...... 200 3.3.8. Rigorous simulations...... 205 3.3.8.1. Feed containing 10% formic acid, 20% acetic acid and 70% water...... 206 3.3.8.2. Feed containing 10% formic acid, 10% acetic acid and 80% water...... 208 3.3.8.3. Feed containing 30% formic acid, 30% acetic acid and 40% water...... 210 3.3.8.4. Feed containing 30% formic acid, 60% acetic acid and 10% water...... 212 3.3.8.5. Feed containing 40% formic acid, 50% acetic acid and 10% water...... 215 3.3.8.6. Feed containing 12.1% formic acid, 0.2% acetic acid and 87.7% water...... 217 3.3.8.7. Summary of the rigorous calculations...... 220 3.3.9. Thermal integration of the separation processes...... 221 4. Discussion...... 251 4.1. Organosolv pulping...... 251 4.2. Solvent recovery in peroxyacid pulping...... 253 4.2.1. Feasible separation methods...... 253 4.2.2. Minimum work of separation...... 254 4.2.3. Phase equilibria...... 254 4.2.4. Shortcut simulations of the separation sequences...... 257 4.2.5. Rigorous simulations...... 259 4.2.6. Thermal integration using pinch technology...... 260 5. Conclusions...... 262 References...... 263 1. Introduction

Alkaline kraft pulping is the dominant chemical pulping process today, but it has some serious shortcomings. These are air and water pollution and the high investment costs. The air pollution is mainly caused by the release of organic sulphur compounds and dust particles. The main source of water pollution in the kraft process are the effluent water streams from pulp bleaching containing organic chlorine compounds. (Schweers 1972a, Schweers 1972b). The problems related to kraft pulping have been partly solved by using extended delignification, oxygen delignification, chlorine dioxide substitution and nonchlorine bleaching in order to minimize the use of chlorine in pulp bleaching. (Bowen & Hsu 1990) The odour problems of a kraft mill cannot be totally eliminated, but a partial elimination of the malodorous emissions is possible by using burning of the sulphur compounds or by absorbing them into alkaline or polythionate solutions. (Teder et al. 1990) The replacement of the kraft process could be possible if a new process having the following characteristics (Worster 1974, Palenius 1987, Franzreb 1989, Leopold 1993, Schroeter 1993) were introduced: - To avoid malodorous emissions the process should be totally sulphur free. - It should have pulping conditions that do not degrade cellulose and dissolved hemicelluloses. - It should be capable of solubilizing most of the lignin with very little loss of cellulose and only a modest loss of hemicelluloses. - It should not require higher temperatures, pressures or pulping times than used in kraft and sulphite pulping. - It should have an efficient and simple chemical recovery system. - It should not cause any environmental problems. - The optimal size of the process should be smaller than the optimal size of the kraft process. - Both hardwoods and softwoods should be possible raw materials. - Valuable by-products should be recovered. - The pulp quality should be at least equal to the quality of the kraft pulp. - The pulp should be bleached without chlorine chemicals. - The pulp yield should be high. - The specific energy consumption of the process should be low. - It should be possible to close the chemical cycle of the process. Organic chemicals were originally used to separate wood into its components in order to study lignin and carbohydrates (Klason 1893, Klason 1908, Klason & Fagerlind 1908, Hägglund & Urban 1928, Kleinert and Tayenthal 1931, Engel & Wedekind 1933, Aronovsky and Gortner 1936, Brickman et al. 1939). In recent years the search for pulping processes that could fulfil the requirements listed above has led to the development of several organosolv methods capable of producing pulp with properties near those of kraft pulp (Dahlmann and Schroeter 1990, Sundquist and Poppius-Levlin 1992, Lora and Pye 1992, Funaoka and Abe 1989, Gottlieb et al. 1992, Young 1989). In 1992 two organosolv processes were operating at full scale and two more were tested at pilot scale. The full scale processes were Organocell (methanol) and ASAM (methanol). The Acetosolv (acetic acid) and Milox (peroxyformic acid) processes were still at pilot scale (Teder & Olm 1992). In 1993 Schroeter (Schroeter 1993) reported that the operation of the Organocell plant was threatened by problems in chemical recovery. The problems associated with chemical recovery seemed at that time to be largely unanswered in all organosolv processes. According to Aziz and Goyal (Aziz & Goyal 1993) cleavage of alpha-ether linkages is the most important reaction in the lignin molecule breakdown during organosolv pulping. Cleavage of beta-ether linkages is believed to occur in the pulping liquor when the lignin has been removed from wood. According to McDonough (McDonough 1993) the alpha-ether linkage cleavage is the most important lignin breakdown in acidic organosolv pulping. Especially in alkaline processes the cleavage of beta-ether linkages also play a role. The easier delignification of hardwoods compared to softwood is primarily a result of differences in beta-ether reactivity, alpha-ether concentration, lignin content, and propensity to undergo condensation reactions. In general, the strength properties of organosolv pulps are inferior to those of corresponding kraft pulps. However, their strength properties can be improved by treatment with aqueous solutions of inorganic salts or organic solvents. (Young 1994) Studies of organosolv processes have shown that the washing operation has an important role for the whole process (Laxen 1987). A problem is the recovery of expensive solvent at the high temperatures required in some processes. Another problem is the solubility conditions of lignin. On the other hand soap and extractives cause much less problems than in conventional pulping. One of the most promising organosolv processes is pulping of birch with formic or acetic acid and corresponding peroxy acids (Laamanen et al. 1986, Poppius et al. 1986, Sundquist 1986, Laamanen et al. 1987, Poppius et al. 1987, Laamanen et al. 1988, Poppius et al. 1989, Poppius-Levlin et al. 1991, Koljonen et al. 1992, Sundquist and Poppius-Levlin 1992, Poppius-Levlin et al. 1993, Sundquist 1996). The method is called Milox pulping and uses no sulphur and chlorine chemicals. Recovering the solvent in Milox pulping economically has been shown to be difficult. The main reason for this is the nature of the solvent system. The spent pulping liquor contains water, formic acid, and acetic acid as the main volatile components. These three components form a ternary azeotrope. In addition, water and formic acid form a binary azeotrope. In order to find more economical process alternatives for the Milox method, a study was carried out on the effect of the feed liquor composition on the economy of a distillation plant separating the three-component mixture. 2. Pulping with organic chemicals

2.1. Alcohols

Alcohols have been used to split lignocelluloses into their components in order to study lignin (Pauly et al. 1934). Koll and co-workers (Koll et al. 1979) have treated birch with several organic solvents in the supercritical state. In the temperature range 523-553 K, at pressures around 10 MPa, alcohols showed the highest delignifying capability. Carbohydrates are also attacked. The highest ratio of lignin to carbohydrate degradation is achieved with ethanol. The optimum water content of the aqueous ethanol solvent is 25-50 %. With higher water content the wood is almost totally liquefied. More conventional alcohol pulping methods have also been developed.

2.1.1. Methanol

Methanol is most widely used as an additional chemical in pulping. It has been used in kraft, sulphite and soda pulping. Demonstration plants using the alkaline sulphite- anthraquinone-methanol process (ASAM) and the soda pulping method with methanol (Organocell) have been built. A full scale plant using the Organocell method has also been built. Catalyzed methanol pulping and various pretreatments with methanol have also been proposed. Methanol can also be used as a solvent when extracting the chemical components from a lignocellulosic material that has been dissociated by a steam explosion process (DeLong et al. 1990a, DeLong et al. 1990b). Muurinen and co-workers (Muurinen et al. 1988a) have made a preliminary comparison between 27 organosolv methods presented in the literature. When the comparison criteria were cooking conditions, pulp quality, solvent properties etc., alkali- methanol-anthraquinone pulping of spruce (Gasche 1985a, Gasche 1985b) was found out to be a very promising new pulping method. The relatively good strength properties of methanol pulps (ASAM and Organocell) have been explained by the distribution of hemicelluloses and DP (degree of polymerization) in their fibre walls. The largest portion of hemicelluloses has been found to be located in the outer layers of the primary wall while high DP values were determined in the outer secondary wall. (Bachner et al. 1993) If methanol pulps are bleached using xylanase enzyme, a thorough washing is required in order to prevent the negative effect of methanol on enzyme activity. (Pekarovicova et al. 1995) Jimenez and co-workers (Jimenez et al. 1997a) have used aqueous methanol to delignify wheat straw. Their results showed that pulps suitable for papermaking can be produced even without any catalysts. Bonn and co-workers (Bonn et al. 1988) have compared organosolv degradation of wheat straw with methanol and hydrothermal degradation. No qualitative or quantitative differences were found in the organic acid composition in the products of the two methods. It was therefore concluded that methanol does not influence the reaction mechanism of acid formation. Acetic acid, formic acid, glycolic acid and lactic acid were found in the degradation products of both methods. Methanol has also been used as an additive in kraft pulping liquors in order to increase the delignification rate. (Olm & Teder 1990, Norman et al. 1992) Compared to the sulphite process, methanol pulping will probably increase the costs of certain parts of the plant and also lead to an increase in primary energy consumption (Gasche 1990). The capital costs of a methanol pulping process will probably be in the same order of magnitude as for sulphite and kraft processes.

2.1.1.1. Catalyzed methanol pulping

The use of absolute methanol as a pulping liquor in the presence of sulphur dioxide has been proposed (Ettel & Hnetkovky 1965). Alkali or earth alkali formates were also added to the mixture. At 408 K, the pulp was only half delignified having a yield of 80-85 %. Paszner and Chang (Paszner & Chang 1982) have patented a method where high yield pulp is produced by cooking fragmented lignocellulose material at a temperature of 453- 513 K with aqueous methanol containing 80-98 v-% alcohol. 0.001-0.5 M of an alkaline earth metal salt catalyst is added to the mixture. Mineral acids, organic acids and acid reacting metal salts can also be used as the catalyst. The lignocellulosic material can be cooked using a liquor-to-wood ratio of 4:1 (kg:kg). The cooking liquor is a water-methanol mixture containing 50-80 w-% methanol. Magnesium, calcium or barium chloride or nitrate is used as a catalyst (0.005- 1 M). The cooking temperature is 453-483 K and the cooking time is less than two hours. Spruce pulp produced by this method has a high hemicellulose content but a low residual lignin content. The kappa number of the pulp is 63 and the brightness is about 50 %. The pulp can easily be bleached to a brightness of 80 %. (Chang & Paszner 1982) A mineral acid can also act as a catalyst (Paszner & Chang 1983a). Hydrochloric, sulphuric and phosphoric acids are useful catalysts in concentrations 0.0005-0.008 N. When spruce chips are delignified with aqueous methanol using calcium chloride as the catalyst, the pulp yield is 49 % and the kappa number is 15. The rate of delignification is highest when the temperature is above 573 K and the methanol content of the cooking liquor is more than 80 %. (Paszner & Chang 1983b) When the calcium chloride content of the cooking liquor is 0.05 %, dissolving pulp can be produced (Paszner & Chang 1983c). 98 % of the original a-cellulose can be preserved, even after bleaching. When magnesium sulphate is used as the catalyst, spruce chips can be delignified with a liquor containing 75-85 % methanol (Paszner & Behera 1984, Paszner & Behera 1985). After cooking for 45 minutes, the kappa number is 35 and the pulp yield after bleaching is about 60 %. The energy consumption of the process is low. Methanol is recovered using distillation and by-products can also be recovered. The process can be operated economically with a production between 50 to 80 tons of pulp per day. Neutral alkali earth metal catalyzed methanol pulping has been shown to be an economically feasible method for producing pulp from aspen wood. (Paszner & Cho 1987) Alkali earth metal salt, e.g. calcium chloride or magnesium sulphate, catalysed aqueous methanol (78 v-%) pulping of spruce wood indicates two distinct stages of delignification, both having first-order kinetics (Paszner & Behera 1989). Initial fast, bulk delignification removes 70 % of the lignin. Loss of residual lignin occurs at a slower rate in the residual delignification stage. Complete fibre liberation is observed at a pulp yield of 57 % and kappa number 72. When calcium or magnesium chlorides, sulphates or nitrates are used as catalysts (0.05 M), both hardwoods and softwoods can be delignified to a low kappa number using a methanol-to-water ratio of 80:20 (Paszner & Cho 1989). The viscosity of the pulp remains high. The pulp can be efficiently washed with pulping liquor and it can be bleached without chlorine. When lignin is precipitated from the black liquor (Abad & Paszner 1989), extractives will co-precipitate affecting the properties of the lignin. They also affect the fermentability of the sugars. Recovery of the extractives (resin and fatty acids etc.) will provide economic advantages for methanol and other organosolv pulping methods. If the pulping liquor is aqueous methanol (80 %), all the resin and fatty acids present in pine wood survive the high pulping temperature (478 K). 78 % of the resin acids and 72 % of the fatty acids are found in the black liquor. If lignin is completely precipitated from the black liquor, 98 % of the resin acids and 60 % of the fatty acids will also precipitate. Black cottonwood can be delignified with aqueous methanol (70 w-%). The liquor-to- wood ratio is 10:1, and the cooking temperature is 403-483 K (Tirtowidjojo et al. 1988). Sulphuric acid (£ 0.05 M) is used as a catalyst. In a flow-through reactor, pulp with a kappa number of 8 and a yield of 47 % can be produced in 180 minutes. The use of a flow-through reactor increases the rate of delignification because precipitation of dissolved lignin on pulp fibres upon cooling is eliminated. The pulp fibres are also less degraded. The recovered lignin has a high molecular weight and is likely to have good chemical feedstock characteristics. The lignin recovered from the black liquor after pulping in a flow-through reactor (Pla & Yan 1991) has been found to be very similar to the lignin from alkaline pulping liquors, when the pulping liquor is aqueous methanol (70 v-%) containing 0.01 M sulphuric acid as the catalyst. Aspen chips can be delignified with aqueous methanol (30-70 v-%) using sulphuric acid (0.01-0.05 M) or phosphoric acid (0.05 M) as the catalyst (Chum et al. 1988). When the pulping time is 150 minutes at 438 K, and the liquor-to-wood ratio is 4:1, the pulp produced can be digested by enzymes. The yield of fermentable sugars is 36-41 % on wood. Phosphoric acid has been found to be a better catalyst than sulphuric acid in the methanol pulping of poplar (Chum et al. 1990). It improves the recovery rate of sugars and lignin removal. In the organosolv treatment of aspen chips, more than 95 % of hexosan, 80 % of xylan and more than 90 % of lignin is dissolved in the pulping liquor, when sulphur dioxide is used as the catalyst (Chum et al. 1989). The treatment is carried out for 40- 200 minutes at 397-418 K with a liquor-to-wood ratio of 10:1. The cooking liquor contains 45- 70 v-% methanol and 0.2-1.7 w-% sulphur dioxide. Under these conditions, very little sugar degradation has been observed. The pulp produced has a low xylan content and is digestible with enzymes into glucose that can be fermented into ethanol. Cellulose containing raw materials can be pulped in aqueous methanol (60-80 %) in the presence of oxygen in amounts of 13.7-22.2 % of the o.d. wood at 408-433 K. The use of oxygen increases the selectivity of delignification and decreases the energy costs. Ethanol and n-propanol can be used to substitute methanol. (Deineko et al. 1988a) Pulp from wheat straw has been produced using a pulping liquor containing 0.9 w-% sodium hydroxide, 35.6 w-% methanol and 63.5 w-% water. The delignification can be carried out without any catalyst, but the addition of ammonium oxalate to the pulping liquor increases the pulp yield. Strength and optical properties of pulps produced by this method have been found to be better than the properties of conventional soda pulps. (Ray et al. 1993) The pulp produced at 438-443 K with aqueous methanol (50 v-%) and a small amount of phosphoric acid can be hydrolysed with enzymes to sugars (Chum et al. 1987). The pulp has low residual lignin and hemicellulose contents. The behaviour of extractives in pine wood during pulping with catalyzed 80 % aqueous methanol has been studied, and resin and fatty acids have been characterized before and after pulping trials (Quinde-Abad 1990). Large amounts of resin acids were found in cooking liquor after 5 minutes of cooking, and after 60 minutes, the black liquor contained 78.1 and 71.6 % of the resin and fatty acids respectively, while the pulp retained 11.7 and 8.2 % respectively. The beating resistance was doubled and the tear strength of the paper produced was increased, when water soluble hemicelluloses obtained by alkylation of aspen wood flour were added to fully bleached spruce organosolv pulps produced by using the method presented by Paszner and Chang (Paszner & Chang 1983a). (Antal et al. 1991) Huth and Cole (Huth & Cole 1994) have synthesized a water soluble lignin copolymer by modifying organosolv lignin at the phenolic hydroxyl position with poly(ethylene glycol)methyl ether (MPEG). The lignin was isolated from black cottonwood by pulping with aqueous methanol (70 %) using NaHSO4 as the catalyst. Ferraz and co-workers (Ferraz et al. 1996) have studied a pretreatment with lignolytic fungi followed by organosolv pulping of eucalyptus wood with aqueous methanol (78%) containing magnesium sulphate and calcium chloride. The delignification rate in pulping was improved by a one month pretreatment with Trametes versicolor. In a paper published in 1998, Paszner (Paszner 1998) calls the neutral alkali earth metal catalysed methanol pulping process the NAEM-ALPULP process. The pulping liquor consist of 80 % alcohol, 20 % water, and 0.012 % NAEM salt catalyst. Pulping is carried out in a continuous digester at high pressure (12.7-30.4 MPa). The pulp yield is up to 60 % and the strength properties are acceptable. The spent pulping liquor is evaporated. When the alcohol content of the liquor drops below 25 %, lignin precipitates and is recovered. The carbohydrates can be used to produce ethanol and xylitol. The total loss of methanol in the process is lower than the amount of methanol formed during pulping.

2.1.1.2. Methanol in sulphite pulping

It has been proved that inorganic additives influence and regulate the sulphite pulping process in presence of methanol (Schorning 1961). Varying the additives of Na2SO4 and NaHSO4 in methanol-water solutions of SO2 allows rapid pulping. A modified sulphite pulping process has been developed to produce high yield (53-70 %) fully defibered pulps from western hemlock (Bublitz & Hull 1981). Three bases (sodium, ammonium and magnesium) were compared, and only minor differences in pulping characteristics were found. The addition of methanol (30 w-%) to the sulphite pulping liquor offers the promise of rapid pulping of softwoods. Chip size has been shown to be an important factor influencing pulping rates in methanol-sulphite pulping (Bublitz & Hull 1982). The recommended size is a thickness of 4-6 mm. Mixed hardwood can be pulped retaining their advantages over softwoods of rapid delignification and higher yields at equivalent the kappa number. pH of the fresh pulping liquor is shown to be a major factor in determining delignification rates, acid liquors produce pulps most rapidly. No loss of methanol has been detected during the pulping. The addition of methanol to a conventional aqueous sulphite pulping process produces such benefits as rapid delignification and high yield of good quality pulp. These result in savings in wood costs, energy costs and investment costs. A minimum of 95 % recovery of methanol is needed to balance solvent make-up costs against savings in wood costs. About 98 % of methanol is potentially recoverable. Relative to aqueous sulphite pulps, methanol-sulphite pulps produce denser papers with higher tensile, burst and fold strengths, yet equivalent tearing strengths at equal pulp yields (Bublitz et al. 1983). Small quantities of DMSO (5 % on wood) can increase significantly the delignification rate of methanol-sulphite systems. Adding methanol to an acid sulphite pulping process reduces the pulping time from 5- 6 hours to one hour (Bublitz & Wilson 1983). Because the carbohydrates in wood are exposed to the pulping chemicals for a shorter time, they are less degraded, resulting in a high pulp yield of 60-65 %. Methanol improves the penetration of the pulping liquor to the chips. Small amounts of methanol are chemically combined with lignin, possibly preventing lignin condensation. Only negligible amounts of methanol are combined with wood carbohydrates. Therefore the loss of methanol due to chemical bonding does not prevent its recovery. Adding 30 % methanol to the sulphite pulping liquor increases the delignifying rate of western hemlock (Bublitz 1987, Bublitz 1998a, Bublitz 1998b). When the chemical charge is 20 % sulphur dioxide based on o.d. wood, and the cooking time is 60 minutes at 438-443 K, completely defibered pulp is produced. The pulp yield is 60-65 % and the brightness is normal. Pulp strength is better than that of conventional sulphite pulps in the same range of total yield. Only 8 kg of methanol is consumed per ton of pulp. Softwoods and hardwood can be delignified with a sulphite liquor containing methanol and 0.5-3 w-% sulphur dioxide (Chiang et al. 1987). Birch pulp with a kappa number of 20-25 and a yield of 44-46 % can be produced when the cooking temperature is 413-418 K, the cooking time is 100-200 minutes, the liquor contains 55 % methanol and 1.25- 1.5 % sulphur dioxide and the liquor-to-wood ratio is 6:1. With the same liquor-to-wood ratio, spruce pulp with a kappa number of 30-40 and a yield of 42-44 % can be produced when the cooking temperature is 433-438 K, the cooking time is 120-150 minutes and the liquor contains 60-65 % methanol and 3-3.5 % sulphur dioxide. When the pulp produced by the alkaline sulphite-anthraquinone method (ASAQ) is extracted with methanol (Oh et al. 1988), the strength properties become better than the properties of kraft and ASAM (alkaline sulphite-anthraquinone-methanol) pulps. The yield of pulp can be increased by adding 35-40 % methanol to the magnesium sulphite pulping liquor (Kordsachia et al. 1988). The strength of the pulp is also improved when the number of bonds between fibres increases. The delignification can proceed to a lower residual lignin content without impairing the quality of pulp. This decreases the chemical charge in bleaching. The addition of methanol to acidic sulphite (pH = 1.6) and bisulphite (pH = 3.5) pulping has been tested (Krull et al. 1989). In sulphite pulping, the methanol concentration 35 % gave the best yield and pulp properties. In bisulphite pulping, the optimal methanol concentration was 20-25 %. The modification of the magnesium sulphite process by adding methanol increases the yield and improves the pulp properties. The delignification can proceed to a lower lignin content and the cooking time is shorter. Pulp with good strength properties can be produced from pine by adding methanol to the magnesium sulphite process (Krull et al. 1991). The cooking temperature is 438 K and the digester pressure is 1.4 MPa. The pulping liquor contains 30 v-% methanol, 0.8 w-% magnesium oxide and 4 w-% sulphur dioxide. A short delignification (45-60 minutes) results in pulp having the kappa number 20. The pulp is easy to bleach.

2.1.1.3. The ASAM process

By adding methanol and anthraquinone to the alkaline sulphite liquor pulp with high yield, low residual lignin content, high brightness and good strength properties can be produced (Patt & Kordsachia 1986). The method is called the ASAM process. The pulp produced is easy to bleach and the yield after bleaching is 5 % higher than the yield of kraft pulp. In laboratory experiment chips are steamed for 20 minutes. After steaming, the pulping liquor is added and the temperature is raised to 388 K in 35 minutes. The pulping liquor is allowed to impregnate for 60 minutes. After impregnation, the temperature of the mixture is raised to 448 K, where the delignification continues 180 minutes. The liquor-to-wood ratio is 4:1. The liquor contains 35 % methanol and 25 % on wood inorganic chemicals. The inorganic chemicals used are sodium hydroxide, sodium carbonate and sodium sulphite. Most of the methanol (70-80 %) is recovered in the digester flashing after pulping. The same wood species can be used as raw-materials as in the kraft process. Compared to the kraft process, the ASAM process uses 10 % less inorganic chemicals, the cooking temperature is 5-10 K higher and the digester pressure (1.3-1.4 MPa) is higher. 70-80 % of the inorganic chemical charge is sodium sulphite (Patt et al. 1987a). The pulp yield is 14-18 % higher than in kraft pulping. The residual lignin content is low because no lignin condensation occurs. The pulp can be bleached without chlorine and it has strength properties superior to those of kraft pulp. The methanol content of the pulping liquor can be dropped from 35 % to 20 % without impairing the pulp quality. In the pressure relief of the digester, even 95 % of the methanol is recovered. By steam stripping the pulp, the recovery rate can be increased to over 97 %. The recovery of the inorganic chemicals is similar to the recovery in sulphite pulping. When 25 % on wood sodium sulphite and sodium hydroxide (80:20), 0.2 % on wood anthraquinone and 35 % methanol of liquor are used in ASAM pulping, all wood species can be delignified with the liquor-to-wood ratio 4:1 (Patt et al. 1987b). The cooking temperature is 448 K, and the digester pressure is 1.3-1.4 MPa. The kappa numbers of the softwood pulps are about 13, and the kappa numbers of the hardwood pulps are about 6. The optical and physical properties of the pulps are good. The ASAM method can be used to produce beech pulps with very low residual contents (Kordsachia & Patt 1987). The pulp can be bleached without chlorine chemicals and its strength properties are better than those of kraft and sulphite pulps. When sodium sulphite and sodium hydroxide are used as the inorganic pulping chemicals, the method is called ASAM I. If the chemicals are sodium sulphite and sodium carbonate, the method is called ASAM II. When sodium carbonate is used instead of sodium hydroxide, no recovery boiler is needed. Methanol improves the impregnation of the chemicals, acts as a buffer, prevents lignin from condensing and stabilizes carbohydrates (Patt & Kordsachia 1988). The beech pulp produced by the ASAM method has a kappa number between 10 and 20 and a yield between 49 % and 55 %. The inorganic chemical charge is 20-25 % on dry wood with the alkali ratio 80:20-70:30 (Na2SO3:NaOH/Na2CO3). The anthraquinone charge is 0.1 % on wood. The liquor-to-wood ratio is 3:1-4:1 and the liquor contains 15-20 % methanol. The pulp can be bleached to the brightness 98.2 ISO without chlorine. There is no difference in the qualities of pulps bleached without chlorine or with a conventional sequence. (Kordsachia & Patt 1988a) The environmental compatibility of the ASAM process is based on the following factors (Kordsachia & Patt 1988b): - Very small amounts volatile sulphur compounds are produced. - It is possible to avoid the emission of sulphur containing gases by closing the gas cycle. - Chlorine-free bleaching can be used. It has been discovered that the addition of methanol to sulphite pulping (the ASAM process) increases the rate of delignification and improves the selectivity (Paik et al. 1988). At kappa numbers 18-20, the yield of ASAM pulp is 0.5-1 % higher than the yield of kraft pulp. The brightness of unbleached pulp is 42-43 %. The best strength properties are achieved, when the pulping liquor contains 20 % methanol. Tensile and burst strengths are better, but the tear strength is lower than for kraft pulp. The ASAM process has been found to be well suited for pulping of pine, poplar, mossy oak and robinia (Winkler & Patt 1988). Pulp properties were found to be good, except for robinia. Commercial pulp production seemed to be most advantageous using poplar as the raw material. The ASAM poplar pulps have 10 % higher strength properties than the corresponding kraft pulps (Patt et al. 1989a). The ASAM pulp has also a higher yield and it is easier to bleach than the kraft pulp. In principle, it is possible to close the water circulation of the ASAM process because the pulp can be bleached without chlorine (Patt et al. 1989b). Methanol is almost completely recovered during the pressure relief of the digester and during the evaporation of the black liquor. Spruce pulp with the kappa number 27.6 can be bleached to the brightness 84.5 % ISO. After bleaching, the pulp yield is 49.1 %. Birch pulp with the kappa number 14.7 can be bleached to the brightness 90.3 % ISO and the yield after bleaching is 54.7 %. The suitability of FTIR spectroscopy has been investigated for feedback process control in ASAM pulping (Faix et al. 1989). The correlation between kappa numbers and IR responses for inorganic chemicals is good. Selectively to a low residual lignin content delignified poplar pulps can be produced by the ASAM process (Kordsachia et al. 1989). The pulp is easy to bleach and has good strength properties. Even pulps produced from unbarked wood can be bleached without chlorine. In October 1989, an ASAM pilot plant was started (Anon. 1990a, Anon. 1990b). The pilot plant can produce 1000 metric tons of pulp per year in a 10 m3 digester. The investment costs of the equipment were 16 million DM. The unit operations included in the pilot plant are cooking, washing, screening, bleaching, evaporation and distillation. The black liquor is evaporated to the solid content 50-60 %. Methanol is concentrated to 98 % in the distillation plant. The methanol loss in recovery is 0.7 %. The effluent waters from the ASAM process can easily meet the German environmental regulations (Anon. 1990c): - Chemical oxygen demand (COD) 70 kg/t ASAM 10 kg/t - Biological oxygen demand (BOD) 5 kg/t ASAM 3 kg/t - Adsorbable organic halides (AOX) 1 kg/t ASAM 0 Spruce pulps with kappa numbers of 9-35 have been produced in the pilot plant (Patt et al. 1990a). The cooking temperature was 453 K and the cooking time was 120-180 minutes. The inorganic chemical charge was 24-30 % and the anthraquinone charge was 0.1-0.15 % on wood. Methanol concentration in the pulping liquor was 9-14 %. The strength properties of the pulp can almost be preserved in chlorine free bleaching and they are slightly improved in conventional bleaching. The advantages and disadvantages of the ASAM process compared to the kraft process can be summarized as follows (Patt et al. 1990b): Advantages: - Higher yield in softwood pulping - Better pulp strength properties - Chlorine-free bleaching in a sequence with low chemical demand - No odour problems - Good heat economy - No or only a small caustizising system - No or minimized effluent treatment - Alkali can be recovered from the bleaching stages - Low fresh water consumption Disadvantages: - Higher temperature and pressure - Longer cooking time - Increased chemical demand for cooking - Higher shive content of the pulp - Methanol recovery is needed - Conversion plant for chemicals is needed. High yield birch pulps can be produced by the ASAM method (Kordsachia et al. 1990a). The pulp has excellent strength properties at kappa numbers of around 10. Because of the low residual lignin content and the high original brightness, the birch pulp is easy to bleach. A chlorine free sequence can be used to bleach the pulp to the brightness 90 % ISO. The pulp quality is not impaired. In birch pulping, the liquor-to- wood ratio 4:1 is used and the pulping temperature is 448 K. The pulping liquor contains 35 v-% methanol and 0.2 % on dry wood anthraquinone. The alkali charge is 25 % on wood calculated as sodium hydroxide. Sodium sulphite, sodium hydroxide and sodium carbonate can be used as alkali. Better results are achieved using hydroxide, but causticizing can be avoided by the use of carbonate. When pulping Spanish eucalyptus (Kordsachia et al. 1990b) the pulp produced has a lower kappa number and similar strength properties as the corresponding kraft pulp. The charge of bleaching chemicals is lower than in kraft pulping. The slow thermal degradation of ASAM waste liquor (Faix et al. 1990) has been studied between 303 and 1233 K using thermogravimetry-mass spectrometry (TG-MS). Dried liquors were heated on the thermobalance in argon atmosphere or in an argon/oxygen (70/30) mixture with a 20 K/min heating rate. The weight loss of the ASAM liquors is similar in both atmospheres up to 713 K. Although ten times more sulphur dioxide is liberated in the argon/oxygen atmosphere than in inert gas, the sulphur cannot be completely recovered as sulphur dioxide. Smelt bed pyrolysis with temperatures over 1223 K will be needed for an ASAM recovery cycle. When softwoods are delignified in the ASAM pilot plant, the liquor-to-wood ratio 5:1 is used and the digester pressure is 1.2-1.4 MPa (Schubert 1990). The typical composition of the pulping liquor calculated as sodium hydroxide on dry wood is: -Na2SO3 15-20 % -Na2CO3 0-7 % - NaOH 0-10 % - Anthraquinone 0.05-0.15 % -CH3OH 10-20 v-% in liquor. In spruce pulping (Schubert 1990), the lowest kappa number (17) and the highest brightness (32 % ISO) are achieved at the alkali ratio 17:3:4 (Na2SO3:NaOH:Na2CO3). The alkali ratio does not affect the strength properties of the pulp. Also the effect of the methanol amount used is small. In pine pulping, the anthraquinone charge can be decreased to 0.075 % without impairing the pulp quality. Softwoods can economically be pulped with the ASAM method to kappa numbers around 20, and hardwoods to kappa numbers around 10 (Schubert et al. 1990). With these grades of delignification, the pulp quality is equal to the quality of corresponding kraft pulps. The most significant differences between the ASAM and the kraft processes are (Kopfmann 1991a): - If sodium hydroxide is not used, no causticizing and no lime kiln are needed in the ASAM process. - Methanol recovery is needed in the ASAM process. - The use of methanol increases the digester pressure to 1.2-1.4 MPa. If an existing kraft mill is converted to an ASAM mill, the following modifications are needed (Kopfmann 1991a): - A new digester is needed because of the higher pressure. - The methanol recovery system has to be built. - Chemical conversion has to be used to recover alkali (Na2SO3 and Na2CO3). The investment costs of the ASAM process are similar to the costs of the kraft process (Kopfmann 1991a). The use of methanol in the ASAM process requires additional considerations concerning explosion-prevention and general safety at work (Kopfmann 1991b). At temperatures over 284 K, methanol forms above the liquid phase an inflammable vapour phase. To prevent explosions, the formation of an explosive vapour phase (5.5-31 v-% methanol) has to be prevented and all ignition sources have to be eliminated. Methanol is poisonous and respirators have to be used when handling it. Most of the methanol is recovered by condensing the vapours formed during digester flashing (Black 1991). The recovery rate can be increased by stripping the black liquor. Methanol is concentrated by distillation. Anthraquinone is consumed during pulping and cannot be recovered. After stripping, the black liquor is concentrated in an evaporation plant. The concentrate from evaporation is burned in a boiler under reducing conditions. Inorganic chemicals are recovered in the smelt of the boiler as sodium sulphide and sodium carbonate. The smelt is dissolved in water and the solution is causticized. The inorganic chemicals used in the ASAM process can be converted in a process that is similar to the chemical cycle in alkaline and neutral sulphite processes. When North American Douglas-fir is delignified with the ASAM method, lower kappa numbers (22 vs. 33) and higher yields (48 vs. 46) are achieved than in kraft pulping (Zimmermann et al. 1991). In conventional bleaching, the ASAM pulp needs half the amount of chlorine chemicals needed for kraft pulp. The yield loss of the ASAM pulp is also lower. Tear strength for the bleached pulps are about the same, but tensile strength for the ASAM pulp is about 20 % higher. ASAM pulp can also be bleached by the chlorine-free OZPN sequence (oxygen, ozone, peroxide with nitrilamine). When the maximum temperature of the ASAM pulping has been reached, about 60 % of the sodium sulphite has been consumed (Patt et al. 1991a, Patt et al. 1991b). 10-12 % has been converted to sulphate and 1-2 % to thiosulphate. At the end of the delignification, 75 % of the sodium sulphite has been consumed. The recovery rate of methanol is 99 %. About 1 % of methanol reacts irreversibly with lignin. About 0.02- 0.04 % of the total methanol is adsorbed or chemically bound in the pulp. When the maximum cooking temperature is reached, 50 % of the lignin has dissolved. About 30 % of the methoxyl groups in lignin are eliminated during ASAM cooking and 1% on dry wood of methanol is formed. Also xylanes form methanol. The molecular weight of the dissolved lignin is low (ca. 1800). The presence of methanol protects the carbohydrates. Methanol also improves anthraquinone solution and reduces the dissociation of cooking chemicals. The black liquor contains 15 % carbohydrates and their degradation products. ASAM pulps have been compared with methanol-sulphite (MS) pulps (Patt et al. 1991c). Both methods produce pine pulps that can be bleached without chlorine (alkali/oxygen, ozone, hydrogen peroxide). Unbleached MS pulp has a high original brightness and chlorine-free bleaching is easy. Good strength properties do require though a comparatively high residual lignin content. Also ASAM pulps are easy to bleach and their strength properties are superior to those of MS pulps. Softwood ASAM and kraft pulps can be bleached chlorine-free using oxygen, ozone, peracetic acid, Caroic acid and peroxide (Hammann et al. 1991). Strength losses occur mainly during peracetic acid bleaching and peroxide treatment. Due to its good strength properties and high initial brightness, ASAM pulp is better suited for chlorine-free bleaching. The use of methanol in pulping affects the design of the equipment (Hudeczek & Lindow 1992). In pulping, methanol increases the pressure and improves the solution of sulphur dioxide. Vapour formed in washing of the pulp contain methanol and the explosion risk is apparent. Methanol is a poisonous and inflammable liquid and has to be stored properly. Eucalyptus pulps produced by the ASAM method are easy to bleach when compared with kraft pulps (Kordsachia et al. 1992). Chlorine-free bleaching can also be used. The ASAM black liquor can be used to produce lignin-phenol gluing agents for particleboard production. (Pecina & Kuhne 1992b) Ozone bleaching of oxygen delignified pine ASAM paper grade pulp and beech ASAM dissolving pulps at high and medium consistency have been studied (Oltmann et al. 1992). Medium consistency bleaching resulted in similar delignification and pulp viscosity as ozone treatment at high consistency. Due to fibre deformations resulting from fluidization in the medium consistency bleaching, the pulps showed superior tear strength and lower tensile strength in comparison with those pulps bleached at high consistency. Experiences gained through three years’ operation of the ASAM pilot plant have been reported (Schubert et al. 1993). Spruce chips were delignified up to kappa numbers of 16 and pulps were bleached to brightnesses of over 88 % ISO using a OZP sequence. The strength properties of the ASAM pulps corresponded to those of the kraft pulps. The recovery rate of methanol was 99.3 %. The Borgards group (Borgards et al. 1993) has compared ASAM and kraft pulping of southern pine. When both processes were applied in order to obtain pulps at kappa levels 15, 20 and 25, ASAM pulping resulted in slightly higher yields. The ASAM pulps had much higher viscosities, better strength properties and higher brightnesses. When pulping Japanese red pine using the ASAM method, optimum results were obtained using 20 % methanol, sodium sulphite and sodium hydroxide at a ratio of 8:2, and a cooking time and temperature of 448 K and 150 minutes. Tall oil from the ASAM process was found to contain higher amounts of abietic acid and linoleic acid and lower levels of dehydroabietic acid, palmitic acid and oleic acid, when compared to the kraft process. (Paik et al. 1994a) Optimum pulping conditions for producing ASAM pulps from oak and birch have been determined by Paik (Paik et al. 1994b). Oak pulp with a kappa number of 19.7 and a screened yield of 43.1 % was obtained by using an active alkali (as Na2O) charge of 17 %, 20 % methanol addition, a cooking temperature of 438 K and cooking time of 150 minutes. When similar conditions, except for a 30 % methanol addition, were used for birch, pulp with a kappa number of 17.8 and a screened yield of 56.8 was obtained. The optical and strength properties of birch pulps were superior to those of oak pulps. When compared with corresponding kraft pulps, the ASAM pulps had higher yields and better strength properties. If an existing kraft batch digester is used to pulp beechwood by the ASAM method, the cooking temperature has to be reduced to 433 K in order to keep the operating pressure at 1.0-1.1 MPa. This results in the need to prolong the cooking time if low kappa number pulps are desirable. The beech ASAM pulps are easily bleached. Conventional bleaching sequences need only about half the amount of active chlorine compared with corresponding kraft pulps. (Obrocea et al. 1994) Oltmann et al. (Oltmann et al. 1994) and Kordsachia et al. (Kordsachia et al. 1994) have studied the transition points between pulping and bleaching. They produced softwood ASAM pulps with kappa numbers 33, 28, 22 and 19. These pulps were oxygen delignified to kappa numbers around 9 and then bleached in a ZEP sequence to brightnesses between 86 and 88 % ISO. Their results showed that it was favourable to stop pulping at a kappa number above 25, because the bleachability of ASAM pulps was good. The yield advantage obtained by stopping pulping at a relatively high lignin content was mainly kept during chlorine-free bleaching. Higher final brightness and better pulp viscosity were also achieved in this way. When pulping was stopped at a kappa number above 25, the pulps had higher tensile and burst strength after OZEP bleaching than the more intensively cooked pulps. When compared to the kraft process, the ASAM process requires two additional chemical recovery steps: methanol recovery and conversion of Na2S to Na2SO3. Methanol is flashed in the first stage of the evaporation plant. The condensate of the evaporator is stripped and then rectified. The methanol recovery rate is more than 99.5%. (Glasner et al. 1994) Kordsachia’s group (Kordsachia et al. 1995) has studied OZ(E)P bleaching of ASAM softwood pulp with a kappa number of 28. In order to avoid severe damage of the carbohydrates, the ozone charge should be below 1 %. When the first bleaching stage was an oxygen delignification to kappa number 9, only 0.9 % ozone and 1 % hydrogen peroxide were needed to achieve a brightness above 87 % ISO. Glasenapp et al. (Glasenapp et al. 1996) have studied totally chlorine-free bleaching of spruce ASAM pulps with and without an ozone stage in the bleaching sequence. The sequences were O(OP)AZP and OA(OP)P. Higher brightness levels were achieved with the sequence containing an ozone stage. Other pulp properties were similar. The chemical costs of the bleaching sequence with ozone were estimated to be about 50 % lower. According to Schild and co-workers (Schild et al. 1996) ASAM delignification of Eucalyptus globulus wood meal can be expressed by means of a kinetic model with three parallel pseudo-first order rate reactions. They calculated the activation energy of the fast initial delignification as 88 kJ/mol. For bulk delignification they obtained a value of 101 kJ/mol when paper grade pulp was produced. The Obrocea group (Obrocea et al. 1996) has studied bleaching of alkaline ethanol, alkaline methanol and ASAM beech pulps using the sequence CEHD, DED or D(EP)D. The ASAM pulp was found to have the highest bleachability. It could be bleached to full brightness with 40 % less chlorine than a corresponding kraft pulp. The strength of the ASAM pulp was only slightly lower than that of the kraft pulp. According to Patt (Patt et al. 1996), ASAM pulps can be peroxide bleached using as catalysts binuclear manganese complexes incorporating manganese in a high oxidation state. Considerable lignin degradation occurs at 423 K and the selectivity of the delignification is very good. Short totally chlorine-free sequences can be used to achieve high brightnesses without the use of ozone. Puthson’s group (Puthson et al. 1997) has compared three modifications of the ASAM process and the kraft process in order to delignify eucalyptus chips. These modifications were ASAM I with sodium hydroxide, ASAM II with sodium carbonate, and PH ASAM with a mild prehydrolysis. The strengths of unbleached and bleached ASAM pulps were found to be up to 20 % higher than those of the kraft pulps. The best results were obtained with the ASAM I process. According to Schubert et al. (Schubert et al. 1998), the ASAM process stabilises celluloses at acid sulphite pulp levels and hemicelluloses at kraft pulp levels. This explains the increased pulp yield. ASAM pulps can be produced at existing kraft mills. Due to the higher sulphidity level, only 33 % of the total production may be ASAM pulp if the recovery boiler is not changed. The development of the ASAM process has been reviewed by Patt and co-workers (Patt et al. 1998). According to them, commercial scale operation of the process is possible without a major risk. 2.1.1.4. Methanol in soda pulping

Methanol solutions of sodium hydroxide have been found to be more efficient delignifying agents than aqueous solutions with the same concentration. (Larocque & Maass 1941) Delignification by alkali-methanol cooking is more rapid and the pulp yield is 7-8 % higher than in kraft pulping (Nakano et al. 1976). One of the optimum conditions is a pulping liquor containing sodium hydroxide 40 g/dm3 and methanol 400 g/dm3, maximum temperature 433 K, cooking time 30-60 minutes. The alkali-methanol method results in more rapid dissolution of lignin and a higher retention of carbohydrates than kraft pulping. In alkali-methanol pulping of birch and beech (Nakano et al. 1977), the pulp yields are 4-5 % higher than in kraft pulping. With the same residual lignin content of pulp, only the tear strength is lower than for kraft pulp. The methanol loss in cooking is 13- 21 kg per ton of wood. Methanol can be formed from lignin and hemicelluloses 6-13 kg per ton of wood. Methanol is easily removed from the pulp by washing and can be recovered at the black liquor evaporation. In alkali-methanol pulping (Kobayashi et al. 1978) of hardwoods, where the liquor-to- wood ratio is 4:1, the pulping liquor contains 40 % (w/v) methanol and the active alkali content is 16 % calculated as sodium oxide, the yield of pulp is about 2 % higher at kappa numbers 20-25 than in kraft pulping. The delignification is slower than in kraft pulping but faster than in soda pulping. The strength properties of the pulp are near those of kraft pulps but the tear strength is lower. The price of the pulp is estimated to be slightly higher than the price of kraft pulp, which is mainly caused by the methanol recovery. The use of methanol requires closed equipment, increasing the investment costs. Based on the comparison the authors claimed that the alkali-methanol process will not be economically feasible in the near future. The rapid delignification in alkali-methanol pulping is due to the prevention of condensation through the methylation of active benzylalcohol groups in lignin molecule (Daima et al. 1978). When studying the behaviour of lignin and carbohydrates in alkali-methanol pulping, it has been found that the peeling-off reaction is suppressed (Daima et al. 1979). This may suggest that the isomerization rate of the glucose end group to a fructose end group is slower in alkali-methanol pulping than in soda pulping. When liquor-to-wood ratio of 15:1, a temperature of 433 K, a pulping time of 30-60 minutes, a methanol concentration of 40 % and sodium hydroxide concentration 40 g/dm3 are used in alkali-methanol pulping (Nakano et al. 1981), delignification is faster and the pulp yield is 4-5 % higher than in kraft pulping. The fast delignification is caused by the prevention of lignin condensation. The tear strength is lower but the other strength properties are equal to those of kraft pulp. The methanol loss is about 14 kg per ton of chips. 99 % of the methanol can be recovered by washing or vacuum evaporating the pulp. Pulps with good strength properties can be produced from hardwoods and softwood using a pulping liquor containing sodium hydroxide and 40 % methanol (Gasche 1985a, Gasche 1985b). The high residual lignin content of the pulp can be decreased by adding anthraquinone 0.5-2 kg per dry wood to the pulping liquor. When the liquor-to-wood ratio is 4:1, the active alkali charge is 22 % per dry wood, the pulping temperature is 433 K and the pulping time is 300 minutes, beech pulp with a quality near that of kraft pulp can be produced. Methanol is formed at 10 kg per ton of pulp from the methylether groups of lignin. The yields of hardwood pulps are 15 % higher than in sulphite pulping. Alen (Alen 1988a) has studied the formation of carboxylic acids during the soda pulping of birch wood in aqueous methanol, ethanol and isopropanol. In addition to formic and acetic acids, 20 monocarboxylic acids and 12 hydroxy dicarboxylic acids were identified. The total amount of formic and acetic acids were 55-60 % of total acids. The dicarboxylic acids represented only a minor fraction (4-6 %) of the total acids. The dominant hydroxy monocarboxylic acids were glycolic, lactic, 2-hydroxybutanoic, xyloisosaccharinic and glucoisosaccharinic acids. The formation of carboxylic acids during the soda pulping of pine wood in aqueous methanol, ethanol and isopropanol has also been studied (Alén 1988b). In addition to formic and acetic acids, 20 monocarboxylic acids and 12 hydroxy dicarboxylic acids were identified. The total amount of formic and acetic acids were 35-45 % of total acids. The dicarboxylic acids represented only a minor fraction (9-15 %) of the total acids. The dominant hydroxy monocarboxylic acids were glycolic, lactic, 3,4-dideoxypentonic and glucoisosaccharinic acids. The soda-anthraquinone-methanol method (Melo & Muller 1989) can be used to produce pulps suitable for papermaking. The adding of methanol produced higher yields, higher kappa numbers, lower viscosities, higher breaking lengths and lower tear factors. Methanol also improves the selectivity of delignification. The main process variable is however the amount of soda. Pine chips have been pulped in aqueous methanol, then treated in an additional pulping stage to which sodium hydroxide was added (Martinez & Sanjuan 1990). This process produced pulps with mechanical properties similar to those of industrial pine kraft pulps. It was found that the temperature in the first pulping stage affected pulp brightness and the extent of delignification. Methanol can be added to the soda pulping of sugar cane bagasse (Mogollon 1991). The best pulp is produced when the methanol concentration in the pulping liquor is 30 % and the sodium hydroxide content is 12 %. The pulping temperature is 433 K and the pulping time is 15 minutes. When the active alkali charge in alkali-methanol pulping (Abe 1993) is 15 % for hardwoods and 19 % for softwoods, the methanol charge is 50 v-% in the pulping liquor, the anthraquinone charge is 0.2 % on wood and the liquor-to-wood ratio is 5:1, delignification is faster and selectivity is better than in kraft pulping. The yield at the same kappa number is 5-10 % higher than the yield of a kraft pulp. Sugar cane bagasse can be pretreated with a soda-methanol liquor for the production of chemimechanical pulps (Ramos et al. 1993). The El-Sakhawy group (El-Sakhawy et al. 1995a) has studied pulping of sugar cane bagasse with methanol and ethanol. Their studies indicated that sodium hydroxide should be added to the pulping liquor in order to obtain easily bleachable pulps at a suitable pulping temperature. 2.1.1.5. The Organocell process

The Organocell process was originally called the MD process. Pulp produced by the MD process has strength properties near (90-95 %) those of kraft pulp (Edel 1984). The pulping has two stages. The first stage is cooking with aqueous methanol at 468 K. The main part of the sugars and 20 % of lignin is dissolved in this stage. The second cooking stage is carried out with a methanol-water-sodium hydroxide solution at 443 K. Methanol is recovered by evaporation and distillation. Lignin is precipitated in evaporation and it can be separated using a centrifuge. The lignin from the second stage can only be precipitated by decreasing the pH of the liquor. The energy content of dried lignin is 25 MJ/kg. The spent liquors from the two pulping stages can be fractionated by precipitation. The liquor from stage one is weakly acidic and contains as solvents only methanol and water. Liquor number two contains additionally sodium hydroxide, which makes the fractionation more complicated. The salt can be separated by ion exchange or electrolysis. Only 10-20 % of the carbohydrates can be precipitated. (Feehl 1984) Lignins from the MD process have been incubated with different types of fungi. White-rot fungi converted the lignin by splitting the aryl-ether linkages. A water soluble acid-precipitable polymerizate was formed. (Haars et al. 1985) The MD process starts with the impregnation of the liquor to the chips (Baumeister & Edel 1983a, Baumeister & Edel 1983b, Baumeister & Edel 1985). Both of the pulping stages are carried out in the same digester. The solvent of the first stage contains methanol 90 w-% in water. The pulping time is 40-120 minutes, the temperature is 453- 483 K, the pH is 3.8-4.9, the liquor-to-wood ratio is 7:1-10:1 and the pressure is 1.2-4 MPa. In the second stage, the water content of the solvent is increased and 5-30 % sodium hydroxide and 0.01-0.15 anthraquinone on dry wood is added. The pulping time is 10-80 minutes and the temperature is 423-463 K. The pulp is washed in two stages and methanol can be recovered from the spent washing liquor by stripping. Methanol is recovered from the spent liquor from the first pulping stage by flashing. Lignin is precipitated by neutralizing the liquor and it is separated by filtration. The remaining liquor is evaporated and the alkali is re-used. In 1987, the MD alias the Organocell process had reached the pilot plant stage (Anon. 1987a). The production capacity of the plant was five metric tons of bleached pulp per day. The pulping stages are continuously carried out in the same digester and the stages are separated by the density difference of the liquors. The alkali is recovered and the lignin is separated electrolytically. The precipitated lignin is separated by filtration and dried in a spray dryer. Lignin and alkali can be recovered electrolytically (Edel et al. 1986). The alkaline lignin solution is introduced into the anode chamber of a divided electrolytic cell and electrochemically acidified therein. Simultaneously the liquor is electrochemically concentrated in the cathode chamber. The cell is divided by an ion exchange membrane. The precipitation of the lignin starts at pH 9.5 and is complete at pH 4. The oxygen developed at the anode produces with the liquor and the precipitated lignin a foam which can be separated by a flotation installation. The power consumption of the electrolysis equipment is about 25 Ah per litre of liquor (Wabner et al. 1986). The investment costs are about 5000 DM per square meter of anode surface. The operating costs are about two times the price of sodium hydroxide. Feckl and Edel (Feckl & Edel 1987) reported that the production of the Organocell pilot plant was raised to six tons per day. The properties of the pulp were near those of kraft pulp. The temperature of the first pulping stage was 468 K and 15 % of the wood was dissolved. Half of the dissolved material was lignin. In the second pulping stage, the methanol content of the liquor was decreased to 30 % and the sodium hydroxide content was 5-10 %. The cooking temperature was 443 K and anthraquinone was used as the catalyst. Methanol was recovered by distilling the spent liquors. In the distillation of the spent liquor from the first pulping stage, lignin precipitated and it was separated by a centrifuge. After distilling the spent liquor from the second stage the remaining alkaline liquor was treated electrolytically to precipitate the lignin and to recover sodium hydroxide. The pilot plant was compatible with the environment and was operated around the clock in a highly populated residential area (Dahlmann & Schroeter 1989). Lindner and Wegener (Lindner & Wegener 1987) have analyzed lignins obtained from the Organocell process. At the end of the second pulping stage, the lignins had the highest oxygen contents which is an indication of oxidation. The first stage lignins contained 1.3-2.2 % nitrogen which is an indication of the presence of proteins. After the second stage, no nitrogen was found. Lignins separated from the spent liquors of the Organocell process are sulphur-free and of low molecular nature (Lindner & Wegener 1988a). They are chemically much less changed than kraft and sulphite lignins. All wood species can be delignified by the Organocell method (Lindner & Wegener 1988b). Pulp with kappa number 30 can be produced by pulping spruce using 50 % methanol in the first stage. The lignin dissolved in the first pulping stage contains a considerable amount of organic impurities. Second stage lignin showed characteristic changes in elemental composition and functional groups, except for the methoxyl groups, as reaction time progressed. The residual lignins are less changed than the dissolved lignins (Lindner & Wegener 1989). Lignin methylation by methanolic liquor is almost negligible. In the Organocell demonstration plant, pulp was bleached with the three-stage chlorine-free sequence EEP/D/EP (Edel 1989, Williams 1989). In addition to the bleaching, the demonstration plant contained wood storage, an impregnation unit, a digester, an evaporation train, chemical recovery (distillation and electrolysis), washers, a screen, a dryer and a boiler. The precipitated lignin was recovered using a filter press and a spray dryer. The low price and boiling point, as well as the easy separation of methanol from pulp, were good reasons for its use. A shortcoming of methanol is the high operating pressure of the equipment. Pulping was carried out in two stages because sodium hydroxide and anthraquinone in the second stage shortened the pulping time and made it possible to delignify softwoods. Spruce pulp with a kappa number of 30 can be produced. Aqueous methanol (50 %) was used in the prehydrolysis and impregnation (30 min) at 423 K. In the first pulping stage, the liquor contained 50 % methanol, the temperature was 463 K, the pulping time was 20 minutes, pH was 4-5, the liquor-to- wood ratio was 3.5:1 and the pressure was 2.8-3.2 MPa. In the second stage, the liquor contained 30 % methanol, the sodium hydroxide charge was 18-22 % on wood, the temperature was 443 K, the pulping time was 40 minutes and the liquor-to-wood ratio was 4:1. The pulp was washed at 388 K for three hours using 2.5 tons of washing liquor per ton of pulp. The evaporate from black liquor evaporation contained 70 % methanol and the concentrate was distilled to 90 %. The methanol loss was 0.8 %. During the first pulping stage (methanol-water), the lignin content of the secondary cell walls decrease slowly, but in the compound middle lamella only the reactivity of lignin increases. During the second stage (methanol-sodium hydroxide) the delignification proceeds fast in both layers but the residual lignin content in the compound middle lamella remains higher than in the secondary walls. (Fengel et al. 1989) In 1990 (Dahlmann & Schroeter 1990a, Dahlmann & Schroeter 1990b), the demonstration plant produced three metric tons per day of bleached (87 % ISO) pulp. The bleaching was carried out in three stages: oxygen with peroxide addition, chlorine dioxide and alkaline peroxide. The molecular weights of lignins isolated during delignification at the demonstration plant have been found to reveal substantial differences depending on the origin (first stage, second stage etc.). Lignins from the second cooking stage can be classified as low molecular lignins having molecular weight ranges similar to kraft lignins. (Lindner & Wegener 1990a) Dissolved and residual spruce lignin have been subjected to permanganate oxidation and thioacidolysis in order to study changes in lignin’s structure during Organocell pulping (Lindner & Wegener 1990b). Lignin degradation reactions known from soda pulping are dominant in the second pulping stage, including enol ether formation. The chemical effect of methanol is restricted to a minor methylation of the alpha-C-atom. Methanol also improves the liquor diffusion into the chips and prevents lignin condensation. The Organocell demonstration plant used the bleaching sequence OEP-D-P to produce pulp with a brightness of 88 % ISO (Pappens 1990). The amount of AOX compounds in the effluent waters from bleaching was less than 0.5 kg/ton of pulp. The two stage continuous digester had been designed with the help of Kamyr. The pulp was washed at the bottom of the digester. After bleaching, the pulp was mechanically dewatered to a solids content of 42 %. Spent liquors were evaporated resulting in a methanol recovery rate of 50 %. An electrolysis unit recovered the caustic soda from the second pulping stage, but in full scale a recovery boiler was planned to be installed. The modification of a sulphite mill to an Organocell mill, producing 430 metric tons of bleached pulp per day, was planned. The mill was estimated to start operation in 1992. (Zitzelsberger 1990) The electrolytic recovery of lignin has been found to modify the structure of lignin from the Organocell demonstration plant. These changes included polymerization reactions and reactions of the side-chain. (Lobbecke et al. 1990) When pulping spruce, most of the hemicelluloses and a part of the lignin is dissolved in the first pulping stage (Puls 1991). The dissolved hemicelluloses are well preserved and 25 % of them can be precipitated. Over 40 % of the precipitated hemicelluloses is mannane. In the first pulping stage, the alcohol-water mixture hydrolyses the a-aryl ether bonds of lignin, forming phenolic and benzyl alcohol groups (Schroeter 1991). Anthraquinone is added to the second stage, causing benzyl alcohol to oxidize to alpha-carbonyls, which are less susceptible to the secondary condensation reactions that occur in the presence of alkali. The effective delignification in the two-stage process is caused by lignin fragmentation during the first pulping stage and prevention of condensation reactions during the alkaline second pulping stage. It has been found that the first pulping stage of the Organocell process can be eliminated (Schroeter & Dahlmann 1991, Brodersen et al. 1992). This simplification results in a stronger pulp. The reduction of the methanol charge to 30 v-% results in a significant reduction of the operating pressure, decreasing the overall costs of the plant. The organic solvent only prevents the dissolved lignin from condensing. The delignification is caused by an acidic hydrolysis. In the one-stage Organocell process, the chips are impregnated with aqueous methanol (50 %). The cooking in the presence of alkali is carried out at 438 K. During the cooking, the alkali displaces the methanol in the chips. The cooking time is two hours and the alkali is recovered in a boiler. Using ozone bleaching a pulp brightness of 88 % ISO can be reached. (Leopold 1991) Lignin-powder can be obtained from the spent liquors by acidic precipitation. The lignins are chemically less altered than comparable technical lignins, and they have relatively low molecular weights, as well as low sugar contents. The lignins can be used to split emulsions, to produce vanillin and vanillic acid, and to protect agricultural seed from being eaten by birds. (Kopra-Schafer 1991) The electrolytic recovery of lignin in the Organocell process has been found to cause severe changes on the structure of the lignins. The molecular weights of the lignins increase due to polymerization reactions. Changes in aromatic and phenolic structures and side chains were also observed. The degree of modification were mostly dependent on the current densities applied. (Lobbecke et al. 1991). In 1991, a full scale Organocell digester was built (Anon. 1991a). It was the first full scale organosolv digester in the world. The investment costs of the plant producing 150000 metric tons of pulp per year were estimated to be 350 million DM. Wegener (Wegener et al. 1991) has subjected lignin obtained from the Organocell demonstration plant to hydroxypropylation in order to improve the reactivity of the lignin in isocyanate particle-board resins. The hydroxypropylation was effectively carried out with an almost total exchange of phenolic hydroxyl groups. Chemical conversion without the removal of, for example, autopolymeric products did not improve the reactivity of the lignin with isocyanate resins. A major drawback of the method was also the poor solubility of the lignin in isocyanate. Lignins obtained from the Organocell process can be converted and liquefied by hydropyrolysis in order to be used as raw materials in the chemical industry. The yields of oils produced depend strongly on the hydrogen pressure. The oils contain mainly phenol, m/p-cresol, o-cresol, 4-ethyl- and 4-propyl-phenol. (Meier & Faix 1991) Full scale production with the Organocell method was started in October 1992 (Young 1992a, Jende 1992, Anon. 1993). The annual production of the plant was 150000 metric tons of pulp. 65 % of the production was fluff pulp and 35 % paper grade pulp. Before cooking, the chips were steamed to remove air and impregnated with aqueous methanol. The one-stage cooking was carried out at 433-438 K. The liquor-to- wood ratio was 4.2:1. No methanol was added to the digester and its concentration in the pulping liquor was 25-30 %. The pulping liquor also contained sodium hydroxide 125 g/dm3 and anthraquinone 0.1 % on dry wood. The digester pressure was 2 MPa. The pulp was bleached with the sequence OE/PPP to the a brightness of 90-92 % ISO. The black liquor was flashed and the vapour formed was used to heat the evaporator train where the rest of the liquor was concentrated to a solid content of 55 %. Methanol was concentrated by distillation. No methanol loss occurred in cooking but the loss in the other parts of the process was 5-10 kg per ton of pulp. Alkali was recovered in a similar way as in kraft pulping. The process consumed 8 tons of steam and 780 kWh of power per ton of pulp. The power excess was 3-5 %. A linear scale-up of the process was not possible (Leopold 1992), because different pieces of equipment affect the pulp quality in different ways. Continuous pulping results in better pulp than batch pulping. The bleachability of the pulp depends on the pulp washing, which should be carried out at a high pH and at a high stock consistency. The lignins dissolved under alkaline conditions during Organocell pulping can be used to make lignin-phenol gluing agents. Up to 65 % of the pure phenol can be substituted with natural polyphenols obtained from the spent pulping liquors. The properties of the particle-boards using this gluing agent meet the demands of building materials. (Pecina & Kuhne 1992a) Meier’s group (Meier et al. 1992) has subjected Organocell lignins to catalytic hydropyrolysis. Using a palladium catalyst, they have produced up to 80 w-% oils as liquid products. The solid residue was only 1 % of the original material. Merkle et al. (Merkle et al. 1992) have acetylated Organocell lignins and studied the high-molecular-weight fractions by means of combined static and dynamic light scattering. Weight average molecular weights between two and ten millions were found. Low molecular weight lignin fractions were found to have a tendency to cluster together forming large aggregates. These aggregates are not stable, which leads to a high flexibility of the overall particle structure. The following goals for the air and water emissions of the Organocell full scale mill producing 430 metric tons of pulp per day have been set (Weidenmuller 1992): -SO2 0 t/a -NOx 490 t/a - CO 360 t/a - Dust 70 t/a - Chlorine 0.15 t/a - AOX 0.05 t/d In 1993 it was reported that there had been some problems with the operation of the full-scale digester. The higher delignification rate and the higher content of fines in wood in comparison with laboratory-scale pulping caused instability in the digester operation. Methanol can be recovered with a loss of only 10 kg per metric ton of pulp. Bayerische Zellstoff GmbH decided to start the full-scale production with Organocell fluff pulps. The price of fluff pulp is higher than the price of paper pulps and the strength of fibres in fluff pulp is of secondary importance. (Leopold 1993) The incorporation of Organocell lignin into polypropylene in amounts of 2-10 wt-% has been found to be a cheap and simple process to utilize it. Polypropylene films with sufficient tensile strengths can be obtained in the absence of commercial stabilizers. The addition of Organocell lignin makes the films partially biodegradable. (Kosikova et al. 1993a, Kosikova et al. 1993b, Kosikova et al. 1994) Organocell lignin can be used as a filler for polypropylene (PP) and poly(ethene-co- vinylacetate) (EVA) containing 13 wt-% vinylacetate. Both lignin-filled thermoplastics show matrix reinforcement with increasing lignin content. A 30 wt-% lignin addition to EVA doubles Young’s modulus maintaining high elongation at break. (Rosch & Mulhaupt 1994) Fagerlind and Agren (Fagerlind & Agren 1995) have invented a process for the evaporation of spent pulping liquors and recovery of volatile substances from the spent liquors of alkaline alcohol pulping processes. Evaporation and solvent recovery are carried out simultaneously in a multiple effect evaporation train where pure water, vapour condensate or steam may be injected to the warmer side of an evaporator effect in order to increase the condensing temperature of the vapour. Heat transfer in the evaporator effects is further improved by divided heat surfaces which makes an early separation of the alcohol, such as methanol, possible. A high organic solvent content in the spent liquor lowers the condensing temperature and impairs heat transfer. The Gliese group (Gliese et al. 1996) has studied enzymatic prebleaching of softwood Organocell pulps with xylanase. Their results showed that this pretreatment improved the bleachability of the pulps. Low enzyme dosages (2-6 IU/g) and short reaction times (30-60 min) are preferable in order to maintain pulp yield and to avoid losses in viscosity and strength properties. The treatment should also be carried out at low consistency in order to prevent a possible attack of the enzyme on the fibre surface, leading to deteriorating mechanical properties in pulp. The bleachability of softwood Organocell pulps can be increased by removing residual lignin from the pulp by hemicellulase enzyme treatment. The bleaching chemical consumption is also reduced. (Zimmermann & Tajana 1996) The quality of the pulp produced by the one-stage Organocell process at the Bayerische Zellstoff mill in Kelheim did not meet the market requirements. This lead to the bankruptcy of the company. According to Neumann (Neumann et al. 1997), the one- stage cooking of pine shows a serious lack of selectivity in delignification. In order to improve the process, they suggest that the caustic soda should be added at a later time in cooking. This would divide the cooking process into two stages. First a prehydrolysis in aqueous methanol at 433 K, then alkaline cooking at the same temperature would be conducted. After bleaching to 85 % ISO, the strength of the pine pulp is on the level of TCF softwood kraft pulps.

2.1.1.6. Other alkaline methanol processes

Sarkanen (Sarkanen 1982) has patented a method for selective defiberizing and delignifying of lignocellulose in the presence of alkali. The pulping liquor contains water, a water-miscible organic reagent and a sulphide or bisulphide compound selected from the group consisting of alkali metal sulphides and bisulphides, ammonium sulphide and ammonium bisulphide. The yield (70 %) is higher than in kraft pulping. Best results are achieved when the organic reagent is methanol and/or ethanol. The best sulphide or bisulphide compounds are sodium sulphide and/or ammonium sulphide. Both hardwoods and softwood can be delignified. When the water-organic reagent ratio is 50:50-80:20, the sulphide charge 0.2-1 M, the liquor-to-wood ratio 4:1-10:1, the original pH of the liquor 8.5-11, the cooking temperature 433-443 K and the cooking time 1-5 hours, a pulp is produced with higher burst and tensile strengths but lower tear strength than the kraft pulp. Ahmed and Kokta (Ahmed & Kokta 1992) have presented a pulping method where wood chips are impregnated with methanol, steam cooked at 453-468 K for 1-4 minutes, then subjected to explosive decompression and refined to produce pulp. The impregnation liquor contains 20-80 % methanol and 0-8 % inorganic alkali. Ku’s group (Ku et al. 1993) has compared alkaline methanol pulping and ethanol pulping of taiwania and eucalyptus. Their results showed that the alkaline process had a better delignification selectivity. The alkaline methanol eucalyptus pulps had higher yields and equal strength properties when compared to the corresponding kraft pulps.

2.1.1.7. Methanol lignins

Methanol lignin has been prepared by extracting spruce wood meal with absolute methanol and using hydrochloric acid as the catalyst (Brauns & Hibbert 1935). The methoxyl content of the lignin obtained was 21.6 %. The formula of the native lignin was estimated to be C47H52O16. When lignin was prepared on a larger scale, the presence of a second fraction, soluble in ether-dioxane, was detected (Compton et al. 1936). The original lignin and this second fraction can be obtained in a pure state by fractionation with dioxane and benzene. The benzene-dioxane-soluble fraction has a methoxyl to hydroxyl group ratio of 5:3. The composition of spruce methanol lignin prepared by the action of anhydrous methanol-hydrogen chloride on spruce meal has been found to vary with the temperature and time of extraction (Compton & Hibbert 1937). The reaction mixture contained two products, having a methoxyl content of 21.6 % and 24 %, respectively. Higher temperatures and longer time of heating favour the formation of the latter. Characteristics of lignins obtained from acidic methanol delignification of black cottonwood have been studied. At the beginning of the delignification, over 60 % of the dissolved lignin is water-soluble. As delignification progresses, the amount of water- insoluble lignin increases in the dissolved material. (Pla et al. 1986) Lignin can be liquefied by hydrogenolysis in the presence of a lower aliphatic alcohol and a catalytic composition of metal sulphides. When methanol is used in the presence of a catalytic mixture of the sulphides of divalent iron, copper and tin, the total yield of monophenols can be as high as 65 % and the total yield of cresols is about 45 %. (Urban & Engel 1988) Bennacer et al. (Bennacer et al. 1989) have studied lignin reactions with methanol, ethanol and DMSO discovering that the beta-ether bonds in lignin are the strongest and the glucoside bonds are the weakest. When methanol lignins solubilized in aqueous dioxane were treated with laccase enzyme, 13C-NMR analysis showed that the main difference between treated and untreated lignins is the increase of the aliphatic alcohol and ether carbon concentration during enzyme treatment. Laccase in its immobilized form seems to be suitable for carrying out conversion processes with lignin derivatives in organic media. (Milstein et al. 1990) Cook and Hess (Cook & Hess 1991) have used lignin obtained from catalytic methanol pulping to produce resins. The cooking liquor contained 70 w-% methanol and 10 w-% sulphuric acid. The resins were produced by treating the lignin with aqueous sodium hydroxide, aqueous formaldehyde and phenol. Up to 40 % of phenol or organosolv lignins in wood adhesives can be replaced with this resin without affecting the performance of the adhesive. Balogh and co-workers (Balogh et al. 1992) have isolated lignin from pine sawdust and studied it by using chemical analysis and infrared spectroscopy. The sawdust was pre- treated with a methanol-benzene mixture (1:1) and then pulped with methanol in the presence of hydrochloric acid (2 N). The isolated lignin presented incorporation of solvent molecules in the lignin. Similar results were obtained by pulping the pre-treated sawdust with ethanol and 1-propanol. The Robert group (Robert et al. 1993) has characterized organosolv lignins with the 13C NMR method. The lignins studied were produced among other things by phosphoric acid catalyzed methanol cooking of spruce and by a acetic acid catalyzed TMP method. Both methods produced lignin with high yield. When the mineral acid acts as catalyst lignin is degraded and condensed. The changes in lignin are negligible when using the organic acid as catalyst. Lignin has been isolated from pine using nine organic solvents with aqueous hydrochloric acid (2.0 N). The molecular weights of lignins obtained were determined. The results show that lignins isolated with alcohols (methanol, ethanol, 1-propanol, 1- butanol and 2-butanol) have similar molecular weights that are considerably higher than the molecular weights of the lignins obtained using other solvents (chloroform, acetone, dioxane and tetrahydrofuran). (Balogh 1993) When aspen chips are cooked in aqueous methanol (70 %) with or without a catalyst (0.01 M H2SO4), the most important reaction in delignification is the cleavage of arylethers and condensation of lignin (Lai & Mun 1993). Methanol prevents lignin condensation. Purina (Purina et al. 1994) has separated and analyzed the outer fibre cell wall layers P and S1 of Organocell and ASAM birch and pine pulp fibres. The residual lignin contents in layers P and S1 of Organocell and ASAM birch pulp fibres were determined to be 7.0 % and 3.3 %, which were 4.7 times the concentration in layers S2 and T. The residual lignin content of layers P and S1 in Organocell and ASAM pine pulp fibres were 6.3 % and 7.3 %, being two times the content in layers S2 and T. The hemicellulose contents were also analyzed and the distributions were found to be more even in pine pulps than in birch pulps.

2.1.1.8. Summary

Methanol improves penetration of the pulping liquor into the lignocellulosic material and therefore increases the delignification rate. It also prevents lignin from condensing during the process, which results in low pulp kappa numbers. Pulp quality is typically near the quality of a corresponding kraft pulp. Methanol is one of the components formed during pulping. This is a probable reason for the very low methanol losses reported. Methanol is a poisonous chemical and forms inflammable vapours at relatively low temperatures. Therefore a pulping process using it has to be carefully designed and operated. Methanol is a low boiling alcohol and therefore it can be relatively easily recovered by distillation. The most promising methanol pulping method is the ASAM process. The pulp quality seems to be acceptable and existing kraft mill equipment may be used. The presence of methanol however requires additional equipment and makes the recovery cycle more complicated and expensive than in a kraft process.

2.1.2. Ethanol

Ethanol has been used to separate the wood components in order to study them (Klason 1893a, Klason 1893b, Klason 1908, Klason & Fagerlind 1908, Holmberg & Runius 1925, Hagglund & Urban 1928, Kleinert & Tayenthal 1931, Rassow & Lude 1931, Ritter & Barbour 1935, Wacek & Morghen 1937, Wacek & David 1937, Brickman et al. 1939, Hunter et al. 1939, Sarkanen & Schuerch 1957, Sarkanen et al. 1981, Fullerton 1983, Vuorinen 1983, Sousa et al. 1986). Today it is one of the most promising organic pulping chemicals. In many ethanol pulping studies small, amounts of organic acids have been used as catalysts (Sarkanen 1980). When high pulping temperatures are applied, no catalyst addition is needed. These methods are probably catalysed by the organic acids released from wood. Lyublin et al. (Lyublin et al. 1988) state as a result of their studies that ethanol pulping could replace all the heavily polluting sulphite mills in the USSR. Lignin can be eliminated from exploded wood by washing with ethanol or other alcohols. (Delong & Delong 1991) Kleinert (Kleinert 1933) has shown that ethanol and water form azeotropic mixtures containing about 95 w-% at temperatures between 393 and 453 K. This behaviour will not cause problems in ethanol recovery because nearly pure ethanol is not needed in any method presented so far. In addition to chemical pulping, ethanol can also be used as an additive in mechanical pulping in order to reduce energy consumption. (Aravamuthan et al. 1993) Ethanol extraction can also be used as a bleaching stage in a sequence including chlorine dioxide treatment. (Brogdon et al. 1997) Ethanol pretreatment has been shown to improve the efficiency of enzymatic hydrolysis of wood. (Holtzapple & Humphrey 1984, Skachkov et al. 1991)

2.1.2.1. Catalyzed and autocatalyzed ethanol pulping

Campbell (Campbell 1929) examined the simultaneous effect of ethanol and hydrochloric acid on spruce wood. He found out that two main reactions were involved, namely, hydrolysis of carbohydrates, caused by the presence of alcohol, and the action of alcohol on lignin in the presence of hydrochloric acid. A method for pulping wood and other fibrous plant material using aqueous ethanol was patented 1932 (Kleinert & Tayenthal 1932). The water content of the pulping liquor could be varied between 20 and 75 %. Delignification was carried out under elevated pressure at temperatures above 423 K and the pulping liquor was changed several times during delignification. The pH of the liquor was adjusted by adding basic or acidic chemicals in concentrations less than 0.1 %. As a result of experiments where cotton was treated with water-ethanol mixtures of various concentrations, Kleinert (Kleinert 1940) has stated that ethanol protects cellulose during the delignification. The reaction conditions in aqueous ethanol-hydrochloric acid pulping of maple woodmeal have no effect on the yield of distillable oils. The mechanism of ethanol delignification (ethanolysis) have been shown to involve both depolymerization and polymerization reactions. Distillable oils are formed by cleavage of high molecular lignin aggregates, while simultaneous polymerization reactions yield a complex ethanol- insoluble polymer. (Hewson et al. 1941a) Delignification of maple wood by ethanolysis involves cleavage of lignin-lignin or lignin-carbohydrate linkages, or both, and is catalyzed by the presence of hydrogen or hydroxyl ions in the pulping liquor. The temperature increase accelerates the cleavage and dissolution of the lignin is facilitated by the presence of an appropriate solvent. (Hewson et al. 1941b) In 1971 Kleinert patented (Kleinert 1971) a process for pulping fibrous plant material using a mixture of water and ethanol as the pulping agent. Pulping was carried out in a counter-current manner. The fibrous material was immersed in the pulping liquor at elevated temperature and pressure. The spent cooking liquor was conducted into a multi- stage flash evaporator from which ethanol was recovered and recycled for make-up of the pulping liquor. The pulping liquor contained 20-75 w-% ethanol. The maximum pulping temperature was 453 K. After washing and screening, the pulp yield was 54 % and the lignin content of the pulp was 1.9 %. Aqueous ethanol in a medium concentration range can be used to delignify both hardwoods and softwoods (Kleinert 1974a). Delignification rates of hardwoods are considerably higher than those of softwoods. The major cause for the increased yields (50- 56 %) of ethanol pulps is carbohydrate retention. Strength properties of the ethanol pulps are in the same range as those of bisulphite pulps produced from the same wood. After distilling off ethanol from the black liquor, a major fraction of the solubilized lignin separates as a quasi-molten phase. Continuous ethanol pulping with recovery of both the solvent and the solubilized wood substances is technically and economically feasible. Ethanol delignification can be carried out at a pH near the neutral value (Kleinert 1974b). This reduces the thermal degradation of cellulose and hemicelluloses resulting in 4-4.5 % higher total yields than in kraft pulping. Ethanol is not consumed during pulping and the total loss of less than 1 % on wood occurs during recovery. Aqueous ethanol penetrates easily into the structure of freshlycut softwoods and hardwoods (Kleinert 1975a). This results in practically uniform delignification. The fibre separation on ethanol-water pulping is related to the removal of both lignin and hemicellulose fractions. In laboratory and pilot plant ethanol pulping of beech, the pulp yields before and after bleaching are 3.3 % and 4.1 % respectively higher than in kraft pulping (Kleinert 1976). The moisture content of the wood used was 30 % and the delignification was carried out at 458 K. The pulping time was 30 minutes and the liquor-to-wood ratio was 10:1. The desired pH value (4-7) of the aqueous ethanol solvent can be adjusted using ammonia or ammonium hydroxide (Kleinert 1978). Paszner and Chang (Paszner & Chang 1979) have presented a method for treating minced lignocellulose. The process consists of boiling the material in an acidified mixture of aqueous solvents. The mixture contains 30 to 70 % water and 70 to 30 % an organic solvent. Preferable solvents are ethanol and acetone. The pH of the pulping liquor is adjusted to 1.7-3.5 by adding a catalytic compound, for example, oxalic acid. The cooking time is preferably between 453 and 473 K. The minimum cooking time is three minutes. Baumeister and Edel (Baumeister & Edel 1980) found out that the results regarding pulp quality, published by others, could not be reproduced. By adding a catalyst to ethanol-water pulping they were able to improve the delignification and to reduce carbohydrate degradation. The strength properties of the pulps produced were comparable with sulphite pulps. A liquor-to-wood ratio of 6:1 was used. The minimum cooking time was 23 minutes at 478 K. The maximum digester pressure was 4.1 MPa. The pulping liquor contained 50 % ethanol and its pH was adjusted with acetic acid. Low pH values gave the lowest kappa numbers. A method has been proposed (Myerly et al. 1981), where chips are contacted in a pressure vessel with the solvent mixture. After delignification, the cellulosic cut is filtered and washed with fresh solvent. The filtrate and the spent washing liquor are combined and evaporated. The evaporation is carried out under reduced pressure, and the lignin separates as a solid and can be recovered. The pulping liquor contains 50 w-% ethanol and the pulping temperature is 473 K. The hardwood fibres produced have a tensile strength which is almost as good as kraft pulps. During aqueous ethanol pulping the pH of the pulping liquor decreases from the original 7 to 3.8. This can be explained by the cleavage of acetyl groups from the hemicelluloses. The dissolved ethanol water lignins showed no distinct differences from kraft lignins with the exception of the sulphur and ethoxyl contents. (Meier et al. 1981) Baumeister and Edel (Baumeister & Edel 1982) have patented a method for continuous pulping of plant fibre material, wherein the material was countercurrently treated with an organic solvent. The solvent was ethanol or isopropanol and the pulping liquor contained 40-60 % of it. The chips were impregnated before cooking, with the solvent at atmospheric pressure and at a temperature between 313 K and 353 K. Pulping was carried out at 453- 473 K. Ethanol was recovered from the spent pulping liquor using flash evaporation. When using the ethanol-water system, the pulping results are highly dependent on the maximum pulping temperature and the liquor-to-wood ratio (Young & Achmadi 1983). To achieve a low kappa number with reasonable carbohydrate retention, a liquor-to-wood ratio close to 20:1 is preferable. The secondary lignin condensation reaction can be avoided using a dilute pulping medium, a two stage cook or a continuous recycling of the pulping liquor. Aspen chips should be preconditioned in the ethanol-water liquor to improve the extent of delignification. The best approach is a gradual rise to the maximum pulping temperature from 373 K for one hour. The pulping continues for an hour at the maximum temperature 438 K. Pulps with kappa numbers around 30 and total yields near 50 % can be produced with this method. Single stage uncatalysed pulping of Eucalyptus globulus with aqueous ethanol (50 v- %) at the temperature of 438-453 K produces pulps with yields above 60 % and residual lignin contents of about 10 %. The pulps have low energy requirements for beating, but their strength properties are comparable only to bisulphite pulps. (Pereira et al. 1986) The ethanol pulping procedure has been studied with regard to its suitability for the production of dissolving pulp from beech wood (Peter & Hoeglinger 1986). Pulp of rather good quality was produced in a pilot plant. The viscose fibres processed from the pulps were comparable with those made of sulphite pulps. The dissolved solids are separated from the black liquor in a spray dryer and ethanol is concentrated in a distillation column. Ethanol is formed during the pulping and its recovery rate is 101- 102 %. A method for chemical recovery in ethanol pulping has been patented (Peter et al. 1984). The black liquor is flashed and evaporated before ethanol recovery by distillation. In an alternative method, the dissolved solids are removed from the black liquor by spray drying and ethanol is recovered by distillation. For the acidic ethanol process (Lonnberg et al. 1987a, Lonnberg et al. 1987b) the washing is proposed to start with HiHeat-type displacement of spent liquor with fresh cooking liquor (Laxen 1987). The pulp is then washed countercurrently with water. Birch wood has been treated in a flow apparatus with ethanol-, 1-butanol- and ethylenglycol-water (50/50 v/v) mixtures (Koell & Lenhardt 1987). Higher temperatures (498-523 K) can be used than in batch delignification. Hemicelluloses and lignin can be separated from the cellulose without severe decomposition of pentosans. The pulps were delignified by more than 90 % and can be bleached easily. Due to the high pulping temperature, the cellulose had low DPw values and may be used as dissolving pulps. Pulps with satisfactory strength properties were obtained by pulping rice straw with aqueous ethanol (1:1) and ethylenediamine (6 v-% on solvent) at 393 K.. The pulp yield was 51 % and the kappa number was about 23. (Guha et al. 1987) The effects of various inorganic additives on aqueous ethanol pulping have been studied. Well delignified pulps with good physical properties in fairly high yields were obtained. The functions of the additives were found to include reaction and depolymerization of the lignin and protection of polysaccharides by restricting their degradation by end-group peeling. The solvent was found to improve the selectivity of delignification by increasing the solubility of lignin and by reducing the hydrolysing power of the liquor. It also improves the transfer and adsorption of reagents in the wood. Among the additives tested, ammonium sulphide fulfilled its functions well. Good results were also obtained with sodium sulphite in ammonia and ammonium chloride. (Ivanow & Robert 1987) Laxen (Laxen et al. 1988) has studied the solvent recovery of acid ethanol pulping (Lonnberg et al. 1987a, Lonnberg et al. 1987b). They found that some formic and acetic acids are formed during pulping. When fresh solvent was initially used in brown stock washing as displacement liquor in order to avoid lignin precipitation, the acids were partly conveyed to the digester. The load of dissolved organic matter also increased in the cooking liquor. The solvent loss was calculated to be less than 15 kg per metric ton of pulp. The energy consumption was estimated to be at the same level as in kraft pulping. Park and Phillips (Park & Phillips 1988) have studied ammonium hydroxide catalyzed solvent pulping of poplar. Delignification parameters included in the study were the concentration of ammonium hydroxide, time and temperature of the reaction, and type of solvent. The addition of 0.82 M ammonium hydroxide to the pulping liquor increased delignification and decreased carbohydrate degradation. Further increase of the concentration had no additional effect. At low pulping temperatures the extent of delignification increased with reaction time. At high temperatures it was found that wood was relignified. No major differences between the three solvents studied (ethanol, butanol, phenol) were observed. A process where cellulosic material is delignified by rapidly heating it in a liquor containing water, ethanol and a buffer, has been patented (Roberts et al. 1988). Sodium bicarbonate is used as the buffer to maintain a neutral solvent extraction. Methylanthraquinone is also added to the liquor. The material is cooked in the liquor as it is heated from 423 K to the maximum temperature of 473-553 K. The mixture is then rapidly cooled to a temperature below 423 K. Optimum parameter values for larchwood pulping in aqueous ethanol solutions of sulphur dioxide are: pulping temperature 398-418 K, sulphur dioxide concentration 15- 24 % and pulping time 25-60 minutes. Under these conditions, a pulp with a yield of 57.6 % and a kappa number of 28.6 was produced. Ethanol can be recovered in two stages. The spent liquor is first treated in a cyclone. It is then fed into a settling tank to precipitate lignin followed by distillation. (Primakov & Barbash 1988, Primakov & Barbash 1989) Organosolv pulping does not require acidic or basic conditions to delignify wood (Faass et al. 1989). Relatively neutral pulping liquors at high temperatures for short residence times can be used without significantly damaging the carbohydrate pulp. Sodium bicarbonate can be used to buffer an ethanol-water-methylanthraquinone liquor. The maximum cooking temperatures (473-513 K) with low residence times (less than 20 minutes at maximum temperature) can be used to produce high quality pulps. Sinner et al. (Sinner et al. 1989) have patented a method where lignocellulosic materials are pulped with an aqueous mixture, containing ethanol or acetone 30-70 v-%, at 373-463 K and at elevated pressure. The pulping time is from two minutes to four hours. If mineral acids (0.001-1.0 N) are added to the mixture, the pulping is carried out at 443-493 K and the pulping time is 2-180 minutes. The black liquor is vacuum distilled to separate lignin. According to Sabatier et al. (Sabatier et al. 1989) the reaction pH is the most important parameter affecting the selectivity of bagasse ethanol pulping. Acid catalysts provoke carbohydrate hydrolysis. This leads to low pulp yields. Dissolving pulps can be produced by this method through adequate control of the pulping parameters. If more drastic conditions are applied to achieve a complete hydrolysis of the lignocellulosic material, it is possible to obtain a fermentable sugar solution. Aluminium chloride seems to be the most suitable catalyst. If paper grade pulps are produced, soda-ethanol pulping seems to be a better choice. Aravamuthan and co-workers (Aravamuthan et al. 1989) have compared acid catalyzed (H2SO4) ethanol delignification with uncatalyzed ethanol delignification and ethylene- diamine-ethanol-anthraquinone delignification. The sulphuric acid catalyzed ethanol-water (7:3) system was found to be more effective than the two others. Prehydrolysis of bagasse with 0.1 % sulphuric acid at 398 K for four hours has been found to remove 41 % of hemicellulose without remarkable degradation of lignin or cellulose. Aqueous ethanol (50 %) pulping of this prehydrolysed bagasse at 473 K for two hours results in the removal of more than 90 % of the lignin and most of the remaining hemicellulose. Degradation of cellulose during pulping is negligible. (Patel & Varshney 1989) Bamboo can easily be delignified with aqueous ethanol (50 %). The strength properties of the pulp can be improved by adding 1 % sodium xylene sulphonate. The tensile index of the pulp was comparable to that of a corresponding kraft pulp but tear and burst indexes were lower. (Singh et al. 1989) The Puech group (Puech et al. 1990) has studied the delignification of oak wood with aqueous ethanol (55 v-%) in a flow-through reactor. This method reduces to a minimum the secondary reactions generally affecting lignin in a closed reactor. Products ready for commercial use can be recovered from pine wood by extracting it with, for example, ethanol (Caperos Sierra et al. 1990). The products are mainly rosin and fatty acids. The extracted wood can be used for papermaking. Extracting the chips prior to pulping reduces the consumption of pulping chemicals. Nunez and Reinoso (Nunez & Reinoso 1990) have studied the conversion of the kappa number to residual lignin content for bagasse pulps produced by cooking in aqueous ethanol. At pulp yields below 60 %, they found the following linear correlations: insoluble klason lignin 0.123, kappa number 2.20; soluble klason lignin 0.016, kappa number 0.14; total klason lignin 0.139, kappa number 2.34. Pulp yields are increased and the use of organic solvent decreased when 20-65 w-% on wood of aluminium oxide is added to an aqueous ethanol pulping liquor (Komissarenkov et al. 1990). If the cooks are carried out in the presence of aluminium palmitate, or a 15- 25 % aqueous solution of aluminium propylate, the pulp yield is increased and the amount of residual lignin in pulp is decreased (Komissarenkov et al. 1991a). If the gamma-oxide of aluminium that has been modified by the ion of a metal (sodium, zinc, magnesium, cobalt, nickel, chromium or iron) is added to an aqueous ethanol pulping liquor, the amount of residual lignin in the pulp is reduced (Komissarenkov et al. 1991b). The effect of ethanol concentration and liquor-to-wood ratio has been investigated for autocatalyzed solvent pulping of a mixture of hardwoods (Goyal et al. 1991,Goyal et al. 1992). Delignification increased with decreasing ethanol concentration over the range studied (50-70 v-%). Optimum selectivity in terms of delignification and pulp viscosity was found at 60 % ethanol concentration. When pulping was carried out at 468 K in 60 v-% ethanol, the kappa number increased slightly when the liquor-to-wood ratio was increased to 4.5:1. Under these conditions, cellulose was almost completely resistant to dissolution. When softwood chips are delignified with oxygen in aqueous ethanol solutions, pulp is produced with a yield of 51 %, when the degree of delignification is 90 % (Deineko & Makarova 1991). The amount of oxygen needed is 14 % on dry wood. Up to 80 % of the solubilized lignin are polymers containing more than 5 % carbonyl and carboxyl groups. Pulp with optical and strength properties suitable for newsprint paper can be produced from sugar cane bagasse using a multistage method. In the first stage the bagasse is cooked in aqueous ethanol at 448-458 K. The resulting pulp is then subjected to a four stage alkali-oxygen extraction. The yield of the resulting pulp is high. (Sanjuan & Gomez 1991). According to Buryachenkov (Buryachenkov et al. 1991) acid-catalyzed aqueous ethanol pulping of aspen followed by washing, thickening and alkaline extraction produces pulps with yields of 50-55 % and kappa numbers of 10-20. Delignification in the presence of catalytic amounts of hydrochloric acid, aluminium chloride, phosphoric acid, hydroxylamine hydrochloride or oxalic acid at 428-448 K was found to follow first-stage kinetics. The most effective catalysts were phosphoric acid and hydroxylamine hydrochloride. These two catalysts gave pulps with 2-3 points higher yields at similar kappa numbers. Synthetic polymers can be grafted to wood polymers concurrent with the pulping. This method is called graft pulping (Young & Achmadi 1992). Free radicals are formed during pulping and they can be used to initiate copolymerization of synthetic monomers with lignocellulosic fibres. Vinyl monomers can be grafted to aspen wood pulp in ethanol based solvent pulping. Only low levels of polymer loading were obtained when single monomers were added to the pulping system. Better results were obtained by using a binary monomer system of acrylonitrile and styrene. Lignin was the main site of grafting. A high-yield (70 %) birch pulp has been obtained by using as the pulping liquor an aqueous ethanol solution containing sulphur dioxide and ammonium hydroxide. They have found that the fibre walls of this pulp have high hemicellulose contents, especially in the two outer layers. (Purina et al. 1992) The delignification rate of aqueous ethanol pulping can be increased by the addition of a sorbent to the pulping liquor. Komissarenkov et al. (Komissarenkov et al. 1992a) have proposed several additives to aspen pulping in an aqueous 20 % ethanol solution. The best results were obtained by using -aluminium oxide. The sorbents permitted a reduction in the ethanol concentration and reduced lignin condensation. Elnitskaya and co-workers (Elnitskaya et al. 1992) have proposed a pulping method where chips are delignified in aqueous ethanol in the presence of acetic acid. The residual lignin content of the pulp can be reduced and the cooking time shortened if the chips are ball-milled in the cooking liquor prior to cooking. The Komissarenkov group (Komissarenkov et al. 1992b) has invented a pulping method comprising cooking of wood chips in aqueous ethanol in the presence of an aluminium containing additive (10-40 w-% on wood). Delignification is improved when the additive consists of the product from the precipitation of kraft lignin with aluminium sulphate that has been pyrolysed. Hemp (Cannabis sativa) stem wood chips can be delignified in an ethanol:water (60/40 v/v) mixture (Tjeerdsma & Zomers 1993, Zomers et al. 1995). In autocatalyzed batch pulping, the best selectivity was obtained by cooking the chips 2.5 hours at 468 K. The pulp yield is 48.6 % and the kappa number is 24.8. When whole stem hemp is pulped with aqueous ethanol (60 v-%), bast fibers are overpulped, resulting in lower tear strength. Optimum pulp properties are only reached when core and bast fibers are separated. (Zomers et al. 1993, Zomers et al. 1995) Tjeerdsma (Tjeerdsma et al. 1994) has found second order polynomial regression models to be appropriate to describe the autocatalysed ethanol pulping process of hemp stem wood fibres as a function of time, temperature, ethanol concentration and liquor to wood ratio. The estimated models can be used to predict the kappa number, the yields and the DP of the pulp. Sansigolo and Curvelo (Sansigolo & Curvelo 1994, Sansigolo & Curvelo 1995) have studied aqueous ethanol pulping of eucalyptus in order to identify optimum pulping conditions. Their results showed that pulp yield increased with increasing temperature, and time at maximum temperature. The optimum cooking temperature was found to be 468 K. The pulp kappa number decreased when pulping at high temperatures and for longer periods. The relationship between yield and pulping time and between residual lignin and pulping time were found to be linear with regard to residual delignification, indicating approximately first-order kinetics. Liang et al. (Liang et al. 1994) have studied the bleaching of ethanol pulps produced using bagasse and wheat straw as raw materials. Both OH and HAP sequences were found to be suitable for the this task. Abramovitch and Iyanar (Abramovitch & Iyanar 1994) have studied the use of microwaves as the energy source in organosolv pulping of poplar and pine. Under optimum conditions, poplar pulps with yields of 45-60 % and residual lignin contents of 2-4 % were obtained using an aqueous pulping liquor containing 70 % ethanol or butanol. The delignification of pine was found to be more difficult and to result in pulps with higher residual lignin contents. Black liquors from oxygen-organosolv pulping of eucalyptus wood in aqueous ethanol (50 v-%) and aqueous acetic acid (80 v-%) have been characterized by Neto and co- workers (Neto et al. 1994). Lignin was precipitated and separated by filtration. The degradation products in solution were separated by simple extraction techniques. Five main groups were isolated: lignin (11-12 % on dry wood), sugars (10-17 % on dry wood), aliphatic acids (12.5 % on dry wood), aromatic acids (2 % on dry wood) and low molecular weight phenols (1 % on dry wood). Pulping of juvenile (9-year old) poplar with aqueous ethanol has been studied. According to these studies, the best pulping conditions are an ethanol concentration of 60 v-%, a pulping time of 210 minutes and a temperature of 468 K. Under these conditions pulps with low kappa numbers (20) and good yields (54 %) can be produced. (Sierra-Alvarez & Tjeerdsma 1995a, Sierra-Alvarez & Tjeerdsma 1995b) The Curvelo group (Curvelo et al. 1995) has studied the kinetics of pulping eucalyptus with aqueous ethanol (1:1 by volume). As in the kraft process, the kinetic behaviour shows three phases: initial, bulk and residual. The initial and residual phases are not selective for lignin removal. A major part of the delignification occurs in the bulk phase with high selectivity. Deineko (Deineko 1995) has studied the conversion of lignin during pulping of wood by oxygen in the presence of aqueous ethanol. The lignin transformations during the process have been found to be connected by the interaction of its reactivity centres both with the solvent components and oxygen. It has been shown that a biological pretreatment of aspen chips with fungi results in a kappa number reduction of 6-9 units after aqueous ethanol pulping. Kappa numbers from 10 to 13 can be reached when the chips are pretreated with Trametes villosus, Phanerochaete chrysosporium or Phanerochaete sanguinea followed by delignification with aqueous ethanol (45 v-%) containing 0.2 % hydrochloric acid on wood at 438 K. (Aleksandrova et al. 1995) Oxidation of the main components of wood can be carried out using molecular oxygen in aqueous ethanol. Methanol, 1-propanol and 2-propanol can also be used instead of ethanol. Oxygen pressure of 2 MPa, an oxidation temperature of 423 K and an ethanol concentration of 40 v-% were used to delignify spruce and aspen chips in a rocking digester. The rate of lignin destruction during the process was found to far exceed the rate of dissolution of carbohydrates. This suggests that no acidic or basic reagent are needed in this process in order to delignify wood. Both hardwood and softwood pulps can be produced in high yields. (Deinko et al. 1995) Gosselink (Gosselink et al. 1995) has analyzed the acid contents of spent liquor resulting from aqueous ethanol (60 v-%) pulping of fibre hemp. Acetic acid and formic acid were found to be the main acids responsible for lowering the pH of the spent liquor and autocatalyzing the the delignification process. Hemp core and hemp bast fibers contain respectively 4.3 w-% and 1.3 w-% acetyl groups and 0.2 w-% formyl groups. Trace amounts of glycolic acid, lactic acid, fumaric acid, levulinic acid, 5-hydroxy- methylfurfural and furfural were also found. The main source of these compounds is probably the polysaccharides of the raw material. Hakansson and co-workers (Hakansson et al. 1996) have used reed canary grass containing about 25 % leaves as the raw material in autocatalyzed ethanol pulping at 453 K. Kappa numbers below 35 were obtained by using ethanol concentrations of 50 to 60 % in the pulping liquor, and high liquor to raw material ratios (10 dm3/kg). The strength properties of grass pulps were lower than those of birch kraft pulps. About 85 % of the ethanol used in a batch cook of white birch (Betula papyrifera) at 408 K can be recovered by flushing from the digester. After condensing, this ethanol mixture may be used to prepare cooking liquor. The pulp yield and the mechanical properties are not affected by the use of recycled ethanol, but the kappa number is increased. (Li & Chen 1996) The Aleksandrova group (Aleksandrova et al. 1996) has studied the effect of fungal pretreatment on hydrochloric acid catalysed aqueous ethanol pulping. When aspen chips were treated with Trametes villosus for six days either a two point increase in pulp yield, an up to nine point reduction in kappa number, or a significant reduction in cooking time was achieved. When delignifying wheat straw at 348 K with a 60:40 (v/v) mixture of ethanol and water using sulphuric acid as a catalyst, optimum pulping occurs according to Sun (Sun et al. 1997) in 1.0 N sulphuric acid and with a two-hour cooking time. Billa’s group (Billa et al. 1997) has studied enzyme treatment of sulphuric acid (2 N) catalyzed aqueous ethanol (80 v-%) pulps. The treatment of wheat straw and sweet sorghum stem pulps with manganese peroxidase in oxalate or malonate buffers degraded 15-30 % of the beta-O-4 linked guaiacyl and syringyl lignin monomers. The catalysis in ethanol pulping can be performed by the acetic acid liberated from the hydrolysis of acetyl groups in wood. According to Araujo and Curvelo (Araujo & Curvelo 1997a) the change of the pH value of the digester content reduces the efficiency of the delignification reactions. When chloroacetate is used as a buffering agent in aqueous ethanol pulping of Eucalyptus urograndis wood, the efficiency of the delignification is improved. Aqueous ethanol (1:1/v:v) pulping of Mimosa hostilis Benth has been found to remove only 65 % of the original lignin at 463 K when the pulping time is 150 minutes. (Araujo & Curvelo 1997b) Girard and Heiningen (Girard & Heiningen 1997) have constructed mass balances for ethanol based birch cooks, obtaining an overall average closure of 98.0 ± 2.6 %. Lignin precipitation is a problem when determining the pulp yield. This problem can be eliminated by subjecting the pulp to a hot ethanol wash. The potential by-products of the process are lignin, methyl and ethyl acetates, methanol, acetic acid, and furfural. The effects of different cooking parameters on the pulping of eucalyptus in aqueous ethanol has been studied by Gilarranz et al. (Gilarranz et al. 1997, Botello et al. 1998). They have found that the amount of water used in the process has no significant effect on pulp properties. In addition, they found that no additive used (sulphuric acid, magnesium salts, and calcium salts) enhanced pulp properties enough to justify their use. Aqueous ethanol in the presence of hydrochloric acid can be used to delignify oil palm fibers of trunks, empty fruit bunches, and fronds. Pulps with kappa numbers around 20 are obtained when 20 % hydrochloric acid is added to a 60 % aqueous ethanol solution. (Suleman & Yusoff 1997)

2.1.2.2. The Alcell process

Diebold and co-workers (Diebold et al. 1978, Diebold et al. 1983, Diebold et al. 1988) have presented a method were wood chips are delignified using an aqueous solution of a lower aliphatic alcohol in a plurality of batch extraction vessels. The charge in each vessel is rapidly heated to the delignification temperature by recirculating a primary extraction liquor having a relatively high dissolved solids content. Thereafter, the charge is subjected to a series of once-through extractions or washes with successively cleaner liquors. The final extraction or wash is carried out with fresh solvent. The extraction liquor from one stage is used in another extraction stage in another vessel. When the extraction is completed, the liquor is drained from the vessel, the vessel is depressurized and the remaining solvent is steam stripped from the charge. The alcohol recovery system contains flash vaporization, condensation of the solvent vapours and vacuum stripping of the residual liquor with steam. Lignin and carbohydrates are then recovered from the alcohol-free extract. Aqueous ethanol (40-60 w-%) is the preferred solvent. The pulping temperature is 453-483 K and the pressure is 2-3.5 MPa. The extraction time is about 5 minutes in each stage. A typical commercial plant of this process, originally called the APR process, has been sized at 136 metric tons per day (Katzen et al. 1980a). The plant would use nine batch extractors, each operating at a three-hour cycle. The fresh pulping liquor containing ethanol 50 v-% is pumped at its atmospheric boiling point into the extractor to pre-soak the chips and purge the air. The liquor is then drained back to the fresh liquor accumulator, leaving saturated ethanol-water vapour behind. In primary extraction, the liquor (solid content 5-7 w-%) from the primary extraction accumulator (473 K and 2.86 MPa) is rapidly circulated through the extractor for less than five minutes. During this isothermal stage 70-80 % of lignin is removed. The flow from the extractor is diverted to the recovery feed accumulator, from which a continuous flow of liquor, containing about 10 % solids, is taken to the recovery cycle. In secondary extraction, the solvent is taken from the extractor undergoing tertiary extraction. This liquor flows once through the extractor to the primary extraction accumulator. In tertiary extraction, the feed to the extractor is switched so that liquor is discharged from the fresh liquor accumulator to another extractor as secondary extraction liquor. At the end of the tertiary extraction, the liquor is drained back to the fresh liquor accumulator, leaving saturated ethanol-water vapour behind. Controlled depressurization vents the extractor vapour to the blowdown condenser for recovery and recycling. The residual pulp in the extractor is steam stripped at atmospheric pressure to remove the remaining ethanol. Steam and stripped ethanol are fed to the blowdown condenser. Finally, the extractor is filled with water and the pulp flushed to a hold tank. The black liquor from the recovery feed accumulator is first fed to a flash drum (Katzen et al. 1980a). Vapours are condensed and fed to the fresh liquor accumulator. The black liquor slurry from the flash drum contains about 15 w-% dissolved and precipitated solids. This slurry is continuously stripped in the solvent recovery tower operating under vacuum. Steam and ethanol vapours from the top of the tower are condensed and fed to the fresh liquor accumulator. Tower bottoms is a slurry containing precipitated lignin suspended in an aqueous sugar solution. The lignin is separated using a clarifier and a centrifugal. The remaining aqueous sugar solution is concentrated to a solid content of about 60 w-% in a double-effect evaporator, to which heat is supplied by flash vapour from the black liquor. The costs of an APR plant, producing 150 tons of pulp per day, have been estimated (Katzen et al. 1980b). With no byproduct credits, the payback time of the investment was estimated to be 4.4 years and the return on fixed investment was estimated to be 13 %. If byproduct credits are taken into account, the corresponding values are 3.2 years and 22 %. A pilot plant for the APR process was started up in August 1983 (Lora & Aziz 1985). The main variables affecting the extent of delignification and cellulose degradation of unbleached APR pulps are: extraction time, extraction temperature, solvent composition and hydrogen ion concentration. Red oak pulp has been produced with a kappa number of 14 and a pulp yield of 43 %. Aspen and birch seemed to be easier to delignify. The brightness and the strength properties of the APR pulps are near those of corresponding kraft pulps. The commercial potential of fibres produced by the APR method has been tested in pilot paper machine trials. APR pulps can successfully replace the hardwood kraft component in printing and tissue grades. The potential of the process in producing dissolving pulp has been shown in pilot viscose production runs. At equal kappa numbers, the APR process produced in the pilot plant aspen pulps with screened yields 1.2-1.7 % higher than kraft pulps produced in the same plant (Lora et al. 1985). The ethanol pulps were easier to bleach than the kraft pulps. The strength properties of unbleached ethanol pulps were 62-82 % of kraft. After bleaching to 88 brightness, the difference between ethanol and kraft pulps was practically eliminated. In 1987, the APR process was renamed and called the Alcell process (Pye et al. 1987). The process was claimed to offer several advantages over the conventional methods: 1. It can economically operate at a smaller scale than the kraft process. This is mainly caused by the less complicated chemical recovery and by the need for less pollution control equipment. 2. The quality of pulp is competitive with conventional pulps. 3. The production costs are lower at any scale than the costs of kraft or sulphite pulps. 4. The process isolates lignin in a relatively pure form. 5. The hemicellulose saccharides are also recovered. 6. If the by-products can be marketed, the Alcell process would be more profitable than conventional pulping processes. 7. The process is environmentally friendly compared with the kraft and sulphite processes. Alcell pulps can be bleached to 90 % ISO. The only effluent from the process comes from the bleaching plant and can easily be treated (Evans 1987, Williamson 1987, Williamson 1988). Bleached Alcell pulps are stronger than both sulphite and TMP pulps and compare well with kraft pulps. Production costs in an Alcell mill are estimated to be 85 % of those in a kraft mill. The Alcell process gives pulp mills the ability for the capacity expansion at about 60 % of the cost of the kraft mill (Pye 1987). The process is most suitable for hardwoods and has been tested with red maple, aspen, birch, red oak, white oak and sweet gums. Pulping conditions can be changed to optimize papermaking conditions of a variety of pulps. Each oven dry ton of wood can yield about 53.5 % pulp, 18.4 % lignin and 21 % fuel. The Alcell process can, according to its developers, be used to (Anon. 1987b): 1. Expand an existing kraft mill. 2. Build a small mill capable of exploiting limited forest resources. 3. Build a less expensive greenfield pulp mill. 4. Produce pulp from non-wood feedstocks. In 1988, it was reported that Repap Enterprises was investing C$ 45,000,000 in an Alcell trial plant. Once the trial plant is operating satisfactorily it will be expanded to produce 230 tons of pulp per day and to include an oxygen-based bleaching sequence. (Fales 1988, Karl 1988) Lignin can be precipitated from the black liquor by diluting the liquor with water and an acid to form a solution with a pH value of 3 or less, an alcohol content of 30 v-% or less and a temperature of 348 K or less (Lora et al. 1988, Lora et al. 1990). The precipitated lignin is, after drying, in the form of a powder. The start-up of the Alcell demonstration facility producing 33 tons of pulp per day occurred in March 1989 (Mullinder 1989). The hardwood pulp produced at the start-up had a quality equivalent to kraft pulp. The unbleached pulp was clean, free of shives and had a kappa number of 22. In the Alcell demonstration plant in New Brunswick, Canada, hardwood chips were cooked in batch extractors with an aqueous ethanol liquor (Lora et al. 1989). The black liquor was flashed and then the lignin was recovered by precipitation (Lora et al. 1988) followed by settling, centrifugation (or filtration) and drying. The demonstration plant had only one extractor and four liquor accumulators (Pye 1989a). The pulp could be bleached without chlorine but chlorine dioxide had to be used. The pulps had a higher bending stiffness than kraft pulps, but they were more brittle than kraft pulps. The wood is cooked to quite high kappa numbers in order to preserve the strength. The Alcell process was found to be particularly suitable as a pulp source for nonintegrated pulp mills. Those mills often cannot site full-scale kraft pulping facilities due to environmental restrictions. (Pye 1989b) Lignin recovered from the Alcell process can be used as adhesives for waferboard. When the adhesive contains 20-30 % lignin, the product quality is identical to waferboards produced using 100 % phenolic resin as the adhesive. (Wu et al. 1989) Without a need for a recovery furnace, lime kiln, causticizers and brownstock washers, the Alcell process has a clear capital cost advantage over a comparably scaled kraft mill (Petty 1989, Pye 1990, Pye & Lora 1991). An Alcell mill becomes economically attractive with a pulp capacity of 300 metric tons per day or even less. When no recovery furnace and other high-maintenance equipment are used, an Alcell mill should have lower maintenance and operating labour costs than a kraft mill. Wood costs are also lower due to the higher pulp yield of the Alcell process. The recovery and sale of lignin and other by-products lead to a lower energy production in the process. This increases the demand for externally generated energy, increasing the utility costs of the mill. The cooking chemical (ethanol) costs for the Alcell mill are also higher than for a kraft mill. These two disadvantages should however be compensated for by the value of the recovered by-products. The spent pulping liquor, from which lignin has been precipitated, can be distilled to recover ethanol (Dillen 1990). Sodium hydroxide is fed to the top of the distillation column in order to prevent ester formation. Aqueous ethanol is taken as vapour from the top of the column and as liquid from a lower stage of the column. These streams are mixed and used as pulping liquor. A side stream is taken from the mid part of the column. It contains furfural and some of the more volatile compounds in wood. The bottom product contains carbohydrates and sodium salts of organic acids (acetic acid etc.). It is concentrated by evaporation to a solid content of 50 %. This syrup can be used as animal fodder or hydrolysed to be used in ethanol fermentation. Alcell pulps can be bleached from relatively high kappa numbers to high brightness and high cleanliness pulps with strength properties comparable to kraft hardwood pulps (Cronlund & Powers 1990). Organic chlorine compound levels in bleaching effluent streams and bleached pulp are low. Alcell pulps also show high potential for non- chlorine bleaching. The Alcell process produces hardwood pulp equal to commercial kraft pulp (table 1). The pulp properties have been determined for a furnish of 50 % maple, 35 % birch and 15 % poplar. 10 % kraft softwood pulp was also added to the Alcell pulp. (Harrison 1991, McCready 1991, Anon. 1991b).

Table 1. Physical properties of Alcell and commercial kraft pulp at a Canadian Standard Freeness of 400 (Harrison 1991, McCready 1991, Anon 1991b).

------Alcell Commercial kraft ------Tensile (km) 7.47 7.40 Tear (mN·m2/g) 7.20 6.75 Burst (kPa·m2/g) 5.08 5.18 Brightness (ISO) 88.7 89.6 ------

Compared with a kraft mill, the Alcell process is dependent on the sale of its by- products for competitive operation (Jamieson 1991, Karl 1991a, Karl 1991b). Of the by- products, furfural is the easiest to market. About 1.8 million kg per year would be distilled from the black liquor in a 300 tons per day plant. It can be used, for example, as a solvent in the production of lubricating oil. The Alcell lignin is a brown, hydrophobic powder. A 300 tons per day mill would precipitate about 100 tons per day of lignin at 3 % moisture content. It can be also used, for example, to produce glue for the waferboard industry. Other by-products that could have market value are acetic acid, and such essences and flavours as vanillin and ethyl acetate. The costs of Alcell pulping have been compared with the costs of conventional chemical pulping. The cooking liquor (ethanol) in Alcell is about 4-5 times more costly than caustic liquor, but Alcell produces pulp with an up to 2% higher yield. In bleaching costs, there no difference between the methods. The equipment used in Alcell pulping is simpler leading to reduced capital costs and lower labour costs. If lignin is marketed as a by-product, the energy costs of Alcell pulping will be higher than for conventional chemical pulping. (Henry 1991) The lignin obtained by the Alcell process has a low molecular weight, high purity, low glass transition temperature and relatively high decomposition temperature (Lora et al. 1991a). Mill scale trials have shown that this lignin can be used to partially replace phenol formaldehyde resins used in plywood and other applications. The process also produces a water soluble, very low molecular weight lignin fraction that can be used in the adhesives industry. Agricultural residues can be pulped using the Alcell method. Kenaf pulps have strength properties between those of commercial hardwood and softwood pulps. Sugar cane bagasse and sugar cane rind chips result in bleached pulps with similar physical properties, that are comparable to conventional mixed hardwood pulps. During cooking of rind chips 2.3 % per weight of dry chips of furfural is generated. The fact that no inorganic solids are recycled in the process, means that there is no accumulation of silica. (Winner et al. 1991a, Winner et al. 1991b, Lora & Pye 1992) One of the major advantages of the Alcell method is its recovery of by-products and the selling of these byproducts to reduce operating costs. Sulphur-free lignin, furfural and wood sugars can be recovered. The Alcell process needs only one-third of the 1000 ton per day production needed for a kraft mill to operate economically. The Alcell process has lower operating costs than a kraft mill because of the higher pulp yield. (Koncel 1991) Lignin obtained as a dry powder in the Alcell process has been tested as an alternative for petroleum-based resins in plywood. Mill trials have shown that 15 to 20 % of the resins needed can be substituted with this lignin without any changes in the strength properties of the panels. (Lora et al. 1991b) In designing a distributed control system (DCS) for the Alcell process there has been a need to take into account some unique components, such as: - an unproven process design, - unknown control requirements and - a potentially volatile environment. These have led to more emphasis on field and rack room wiring, use of a digital field communication bus, better definition of the information system requirement and partition of the plant utilities into their own controller modules. (Henry & Seely 1991, Henry & Seely 1992) In 1992, it was reported that an Alcell mill with a capacity of 300 tons of pulp per day was under design (Lora et al. 1992). The scale-up will be done after successful operation of the demonstration plant (table 2). Table 2. Characteristics of industrially produced Alcell pulps after 4000 revolutions in a PFI mill (Lora et al. 1992).

------Birch 50% maple 35% birch 15% poplar ------Unbleached kappa no. 27 30 Bleaching sequence EoDED EoDED Brightness, ISO 88 + 88 + Bulk density, cm3/g 1.32 1.36 Tear index, mN·m2/g 6.8 6.7 Burst index, kPa·m2/g 4.7 3.2 Breaking length, km 7.5 5.1 CSF, mL 420 416 ------

The Alcell process operates at high pressure (2.8 MPa) and temperature (468 K) using 60 % ethanol as the extraction medium (Detweiler 1992). These conditions are above the flash point of ethanol and result in potential hazards of fire, explosion and accidental release of toxic process materials. The fact that the process is operated indoors due to the climatic conditions in Northeastern Canada, greatly increases the hazard potential to personnel and equipment. These risks can be minimized by proper management of risks. The Process Safety Management technique can be useful to develop strategies to either prevent the event from occurring by excellence in process and equipment design, or operating and maintenance procedures, or to minimize the potential impact through the use of a competent consultant and experienced personnel. Alcell hardwood pulps with optimal strength properties have higher kappa numbers than kraft pulps, but they can be further delignified to low kappa numbers in an oxygen bleaching stage with excellent preservation of pulp strength properties (Cronlund & Powers 1991, Cronlund & Powers 1992). Ozone and other nonchlorine bleaching stages can also be used. The best strength preservation, economics and lowest levels of AOX in the effluent waters are however achieved with chlorine dioxide bleaching. In 1992 a simple chemical recovery was introduced for the Alcell method. Initially the black liquor was flashed, and then diluted to precipitate lignin. The lignin was recovered using a solid/liquid separation step and then dried. The remaining stream, containing water, ethanol, dissolved wood sugars, furfural and acetic acid as the main components, was fed to a distillation column. Ethanol was recovered as the top product from the column. Furfural was taken as a side stream and the bottom product was partly recycled to dilute the black liquor. (Lora & Pye 1992) Leszczynski (Leszczynski 1992) has reported that the Alcell process needs only one- third of the amount of water used in a kraft mill of similar size. The effluents of the process are easily purified. The residual lignins that remain with the fibres in Alcell pulps have higher molecular weight, less aromatic hydroxyl groups, more aliphatic hydroxyl groups and resemble native lignin to a greater extent than the lignins that are dissolved during pulping. (Lora et al. 1993a) In 1993, over 2000 cooks had been completed in the Alcell demonstration plant since start-up. Lora and others (Lora et al. 1993b) reported that the process seemed to be ideally suited for a closed cycle mill, because the cooking chemical (ethanol) was easily separated due to its volatility. Lignin obtained during Alcell pulping of North American hardwoods is a polyphenol characterized by its high purity, low molecular weight, low polydispersity, low glass transition temperature and relatively high decomposition temperature. The material can be used industrially as a partial replacement for phenolic resins in, for example, waferboards. Another potential field of use for this material is as a rubber tackifier and antioxidant. (Lora et al. 1993c) The physical properties of the lignin obtained as a co-product in Alcell pulping have been measured (Sellers et al. 1994, Sellers 1994). The lignin was used to partially substitute phenol in resole phenol-formaldehyde (PF) resins. The physical properties of the formulated resins were also measured and the resins were used as binders in strandboard production. The results indicate that Alcell lignin can be used to replace up to 35 % of the phenol in PF resins used in strandboard production. Maldas and Kokta (Maldas & Kokta 1994) have studied the use of Alcell lignin as a bonding agent for wood fibre-filled polystyrene composites. The mechanical properties of the composites were improved by the addition of lignin. The optimum concentration of bonding agents was found to be 2-3 %. Ni and Heiningen (Ni & Heiningen 1994, Ni & Heiningen 1996) have studied washing of Alcell pulps with aqueous ethanol. Displacement washing was found to offer much better results than a stirred batch operation. A pulp with an initial kappa number of 54 was delignified to a kappa number about 25 by extended displacement washing. Practical operation conditions for washing were found to be: dilution factor 3.0 cm3/g, ethanol concentration 50 %, temperature 336 K and residence time 6.17 minutes. Higher temperatures and ethanol concentrations improve the lignin removal. One of the disadvantages of the Alcell process is that it needs more purchased energy than a kraft mill. The Ronan group (Ronan et al. 1994) has performed a pinch analysis and found that the potential thermal energy savings in Alcell pulping are about 15 %. These savings can be reached primarily by modifying the heat exchanger network of the process. Because the analysis was made at an early stage of design these modifications could be made without an excessive amount of additional pipework. In 1994, it was reported that the defibration of chips occurs in the Alcell process in a high-pressure extractor at a temperature of 468 K and a pressure of 2.9 MPa. Lignin is recovered from the cooled extraction liquor by diluting it with water. Also acetic acid and furfural are obtained as by-products. The kappa number of the pulp is below 30 after washing. (Anon. 1994) The swelling of Alcell fibres in aqueous ethanol as a function of the ethanol concentration has been studied. The results showed that the swelling of the fibres decreased continuously as the ethanol concentration increased. (Ni & Heiningen 1995, Ni & Heiningen 1997) The solubility of the Alcell lignin has been found to strongly increase when the ethanol concentration of the solvent increases from 9.5 % to 47.5 %. When the ethanol concentration is increased from 47.5 %, the solubility increases much more slowly reaching a maximum at an ethanol concentration of about 70 %. This behaviour is explained by the theory that lignin exhibits the maximum solubility when the solubility parameter value of the solvent is close to that of the Alcell lignin and the hydrogen bonding capacity of the solutions with different ethanol contents is similar. (Ni & Hu 1995) In January 1995, it was reported that Repap Enterprises Inc. had bought an existing pulp mill in order to convert it to produce 142,000 metric tons of hardwood Alcell pulp per year. The mill was scheduled to begin production in 1997. (Anon. 1995) The commercial scale Alcell mill will be producing fully bleached market pulp and as byproducts also lignin, furfural and acetic acid. Acetic acid may be used to produce peroxyacetic acid for pulp bleaching and recovered at the mill. This may be a step towards a totally effluent-free mill. (Lora & Pye 1995) Alcell pulps show advantages in bleaching by totally chlorine free methods, such as ozone bleaching. They can be bleached with lower chemical charges than kraft pulps. (Singh et al. 1995, Zhang et al. 1996) Formaldehyde emissions can be significantly reduced when organosolv lignin fractions are used as a partial replacement for PF in the manufacture of waferboard and particleboard. A reduction of 38 % was obtained when 20 % PF resin solids were replaced with a lignin dispersion. (Senyo et al. 1996) Blends of PVC and Alcell lignin have been studied (Feldman et al. 1996). The tensile strength of the blend increased with increasing lignin content. The presence of lignin did not have a negative effect on the processing of the blends but led to breaking of intermolecular bonds in the PVC structure. Two lignin fractions can be recovered during the Alcell process when hardwoods are delignified. The first fraction is obtained by dilution and precipitation of the spent pulping liquor. The second fraction is obtained by concentrating the aqueous stream from distillation. This fraction has a lower molecular weight, a lower glass transition temperature and a higher carboxyl content than the first fraction. Potential uses of the second lignin fraction are in novolac resins, concrete superplasticizers and particleboard production. (Creamer et al. 1997) Wheat straw and reed can be used as raw materials for producing bleachable grade Alcell pulps that have properties competitive with commercially available hardwood kraft pulps. The total yield of saleable products, including pulp and by-products, is around 70 %. In conventional annual fibre pulping, the accumulation of silica is a problem. In Alcell pulping only about 15 % of the silica enters the pulping liquor. (Winner et al. 1997) An elemental chlorine-free (ECF) bleaching sequence for the Alcell process has been introduced (Ni & Heiningen 1998). The method consists of a wash by aqueous ethanol, followed by ozone bleaching of the pulp impregnated with an acidified ethanol solution, and then extraction with aqueous ethanol. Finally, the pulp is bleached in two chlorine dioxide stages. The physical properties of the pulp are at least similar to the corresponding ODED bleached pulp. When no caustic is used, the ECF sequence has an economic advantage over the ODED sequence. The load of the waste water treatment originating mostly from the oxygen stage can also be reduced. Ooi and Ni (Ooi & Ni 1998) have proposed a totally chlorine-free (TCF) bleaching sequence for the Alcell process. The brownstock is first displacement washed with aqueous ethanol (70%). Then it is impregnated with a 30 % ethanol solution acidified to pH 2 and delignified with 2 % ozone. In the next step, the pulp in again displacement washed with 70 % ethanol. The resulting pulp has a kappa number of 10 and can be further bleached using hydrogen peroxide and peracetic acid stages. In July 1998, it was reported that Repap Enterprises had terminated its December 1997 agreement to sell Alcell Technologies to Malaspina Capital Ltd. Repap could not afford to continue to support Alcell in anticipation of finding another buyer. (Anon 1998a)

2.1.2.3. Alkaline ethanol pulping

Marton and Granzow (Marton & Granzow 1982a) have patented a method for producing wood pulp using a mixture of ethanol and sodium hydroxide. The aqueous pulping liquor contains 10-25 w-% of sodium hydroxide and 10-60 v-% of ethanol. The pulping time is 0.5-4.0 hours and the pulping temperature is 373-473 K. Essentially, any wood can be pulped. Adding ethanol to a caustic soda cook greatly improves its selectivity with respect to lignin (Marton & Granzow 1982b). Likewise the presence of sodium hydroxide improves the delignifying ability of ethanol. The chips pulped with an aqueous mixture of ethanol and sodium hydroxide are easier to defiberize than those pulped with soda or by the kraft process. The organic solvent reduces the surface tension of the pulping liquor at high temperature promoting the penetration of the alkali into the chips and the diffusion of the breakdown products of lignin from the chips to the liquor. Simultaneously, the alcohol also degrades lignin and prevents it from condensing. Spruce pulps have been produced using the liquor-to-wood ratio 5:1 and the ethanol- water volume ratio 1:1. The cooking temperature can be varied from 423 K to 443 K. The use of ethanol allows considerably shorter pulping times to be applied. At the same yield, the residual lignin content of the ethanol-NaOH pulp is slightly higher than for kraft pulp. The strength of the pulp is also slightly lower but the brightness of the unbleached pulp is higher than for kraft pulp. Sarkanen (Sarkanen 1982) has presented an alkaline method for selective defiberizing and delignifying of lignocellulose. The pulping liquor contains water, a water-miscible organic reagent, and a sulphide or bisulphide compound. The organic reagent can be an aliphatic alcohol, for example, ethanol, or a ketone. Extremely high pulp yields compared to kraft pulping can be reached. Chiang and Sarkanen (Chiang & Sarkanen 1983) have found that 0.3 M ammonium sulphide in 30-50 % ethanol-water is a satisfactory pulping medium for delignifying industrial hemlock chips at the liquor-to-wood ratio 10:1. The optimal pulping temperature is 453 K. The extent of delignification increases with increased ammonium sulphide charge. A hemlock pulp with the kappa number 45.8 was found to be easily bleached by a conventional CEDED sequence to a brightness of 86 % G.E. in 55 % yield. The bleached pulp had a high viscosity value (1970 cP TAPPI) and both tensile and burst strengths were adequate, while the tear strength was low. Cottonwood chips were easier to delignify than hemlock chips. Lower ammonium sulphide charges were required and low kappa numbers around 15 were reached. The yield of cottonwood pulps were 12-13 % higher than those of corresponding kraft pulps. Unbleached cottonwood pulps beat fast to high tensile strength values. The tear strengths were comparable to kraft pulp, but the burst strengths were lower than those of cottonwood kraft pulps. Semichemical type pulps can be produced by cooking sugarcane bagasse in aqueous ethanol (40-60 w-%). These pulps have high yields, high kappa numbers and low strength properties. Adding sodium hydroxide to the pulping liquor increases the selectivity of delignification and results in pulps with good yields, low kappa numbers and acceptable strength properties. Adding anthraquinone as catalyst to the ethanol-water- sodium hydroxide system increases the pulp yields, decreases the kappa numbers and improves the strength properties. (Valladares et al. 1984) The addition of ethanol has a complex influence on the performance of various derivates of anthraquinone in alkaline pulping (Chacon & Lai 1985). Ethanol increases the solubility of the additive in the solvent system. With most additives it also stabilizes carbohydrates. A positive effect on the delignification efficiency can be observed with a small addition of ethanol, but high ethanol concentrations (>50%) hinder delignification. Simultaneous addition of ethanol and anthraquinone to sodium hydroxide cooking liquors has been found to offer advantages that are not possible with the use of each additive alone. The alkali consumption during pulping decreased by 10-25 % when ethanol was present in soda-anthraquinone pulping. The pulp yield increased by 1-4 % on wood. The optimum ethanol concentration in the pulping liquor is 25-30 %. The pulps produced by this method are somewhat weaker than kraft pulps. The feasibility study performed showed, however, that this method is not economically attractive. (Janson & Vuorisalo 1986) When comparing pulping of spruce with three methods: aqueous ethanol mixed with sodium hydroxide or a catalytic amount of calcium chloride or ethanol without any additional agents, it has been found that the kappa number is not a good indication of the pulp quality. Instead, the total lignin content should be used. The fiberising point of the delignification was at 9 % lignin content after anthraquinone catalysed ethanol-alkali cooking and at 16 % after calcium chloride catalysed ethanol cooking. Cooking with ethanol only at 468 K did not lead to fiberising at all. Simulation of the solvent recovery with flashing and distillation indicated that ethanol losses can be kept negligible if the energy consumption is allowed to be high. (Aittamaa et al. 1986, Lonnberg et al. 1987a) The alkaline ethanol pulps needed five bleaching stages (DEPDEPD), where the alkaline extraction steps were strengthened with peroxide, for bleaching to a brightness of around 85 % ISO (Lonnberg et al. 1987b). The acidic pulps needed only three bleaching stages (DEPD) to reach the same brightness. The alcohol pulps required short beating times and low beating load. The strength properties of the alkaline pulps were comparable with those of kraft pulps, but the acidic pulps were weaker. The acidic ethanol pulp had a particularly poor tear index. Tissot and Launay (Tissot & Launay 1987) have presented a two-stage soda pulping process where ethanol is added. The first stage (373 K) allows penetration of the cooking liquor to the chips, followed by the actual cooking stage (443 K). The pulping liquor contains 13 % ethanol and 10 % sodium hydroxide. The pulps produced are inferior to kraft pulps. The aqueous ethanol delignification of wheat straw in the presence of sodium hydroxide and anthraquinone can be divided into bulk delignification during the heating- up period and supplementary delignification. Yields of alkaline ethanol pulps are about 12 % higher than those of corresponding soda and soda-anthraquinone pulps. The physical properties of ethanol pulps are similar to those of conventional pulps. (Liang et al. 1988) When pulping bagasse in aqueous ethanol systems, acids have been found to be harmful in ethanol pulping, but alkali has shown good catalytic selectivity. The properties of the bagasse ethanol pulps were similar to those of corresponding soda pulps. (Weisheng et al. 1988) Eucalyptus has been subjected to two-stage organosolv pulping with ethanol (Sanjuan et al. 1989, Sanjuan & Chavez 1989). The first pulping stage used aqueous ethanol to which soda was added in the second stage. The pulps had kappa numbers between 15 and 44 and breaking lengths similar to those of kraft pulps. It was possible to extract tannin from the spent liquor of the first pulping stage with a yield of 6.2 % on dry wood. The use of a solution of ethanol, water and sodium hydroxide (15 %) as the pulping liquor for hornbeam wood results in rapid delignification (90 % in 2 hours at 443 K). The addition of anthraquinone improves both delignification and pulp yield. The ethanol- water solution transforms soda to a superbase and doubles polysaccharide adsorption of hydroxyl ions. The high ion concentration leads to rapid cleavage of the intermolecular lignin bonds and speeds up delignification. (Ivanow & Robert 1989) When pulping bagasse meal with organic solvents, the best results have been obtained with aqueous ethanol. Adding sodium hydroxide to the system improved the selectivity with respect to lignin. If the process was acid catalysed, the degradation of carbohydrates increased. The highest yields, lowest kappa numbers and best bleachability were achieved when anthraquinone was used as the catalyst. Pulp properties were better than those of corresponding soda pulps. (Bharati et al. 1989) Tubino and Filho (Tubino & Filho 1989, Tubino & Filho 1990) have presented a method where eucalyptus wood is pulped with an aqueous solution containing 20 % ethanol and 19-20 % sodium hydroxide. The pulping is carried out at 433 K for 65 minutes. Chen et al. (Chen et al. 1990) have introduced a two-stage ethanol pulping method. The first pre-treatment stage is vacuum impregnation of wood chips in aqueous ethanol with the presence of sodium hydroxide. In the second stage, the pre-treated chips are delignified with a sulphur dioxide-ethanol-water mixture. This process produces softwood pulps with a 6-10 % higher yield than kraft pulps. The fibre strengths approach a level which is about 90 % of kraft strength. The black liquor is flashed and evaporated. Ethanol is recovered by distillation. The concentrate from the evaporator plant is fed to a recovery boiler. A pulping method consisting of cooking wood chips in an aqueous mixture of sodium hydroxide, hydrazine and alcohol has been patented. Pulp yields are improved if the alcohol used is ethanol and its concentration in the mixture is 30-50 v-%. (Pazukhina & Shan 1990) A pulping method including cooking of chips in aqueous ethanol (25-50 v-%) in the presence of 10-20 g/dm3 sodium hydroxide at 438-443 K for 90-180 minutes has also been patented. Process selectivity and the mechanical strength of pulp are improved if, after the removal of the spent liquor, the pulp is treated with aqueous ethanol (40-60 %) at 323-353 K for 30-60 minutes. (Beigelman et al. 1991) The physical properties of wheat straw alkali-ethanol and soda pulps have been compared. Alkali-ethanol pulps have a rapid response to beating and satisfactory physical properties when the beating degree is above 40˚SR. (Liang & Xiao 1992) Ultra-high-yield (91.5 %) pulps with good mechanical and optical properties can be produced from aspen by impregnating the chips with aqueous ethanol and sodium hydroxide prior to steam explosion pulping (Ahmed et al. 1992). Girard and Chen (Girard & Chen 1992) have delignified white birch using a blend of caustic soda and denatured ethanol as the solvent. They used a liquor-to-wood ratio of 5:1, a total heating time of 165-170 minutes, a cooking temperature of 443-453 K, an alkaline charge of 15-20 w-% and an ethanol concentration of 20-50 v-%. The pulp quality is well comparable with hardwood kraft pulp, when the same active alkali charge is used. The organosolv pulp has a better brightness, and at a given breaking length, its tear index is higher than that of the kraft pulp. Sobral (Sobral et al. 1992) has studied a process called ASAE (alkaline sulphite anthraquinone ethanol) pulping. Compared with kraft pulping, the ASAE process is more selective and produces pulps with high yields (52 %) and low kappa numbers (13). In addition, ASAE pulps are easier to bleach than kraft pulps and have mechanical properties similar to those of kraft pulps. When studying alkaline ethanol pulping of Acacia confusa and Alnusa formosana Wang and Tseng (Wang & Tseng 1993) have found that the optimum pulping conditions are: alkali charge 16 %, temperature 433 K and time at maximum temperature 60 minutes. The yields of Acacia confusa pulps are 3-7 % higher than the yields of kraft and soda-anthraquinone pulps. The yields of Alnusa formosana pulps are somewhat lower. The beating time of ethanol-alkali pulps are 9-17 minutes shorter than those of kraft and soda-anthraquinone pulps at about 400 CSF, and it requires less beating energy. The strength properties of the ethanol-alkali pulps are better than those of soda- anthraquinone pulps but not so good as those of kraft pulps. When eucalyptus (Eucalyptus cilriodora) chips are cooked in an aqueous solution containing 8 % sodium hydroxide and 50 % ethanol at 453 K for four hours, they can be defiberated with a screened pulp yield of 65 % (Xiao et al. 1993). The pulping time can be reduced to three hours and the pulp quality can be improved by adding 0.05 % anthraquinone to the pulping liquor. The pulp yield is however reduced to 63.5 %. The delignification activation energy of eucalyptus is 42.0 kJ/mol. Adding inorganic reagents (e.g. NaOH) to one-stage aqueous ethanol pulping of sugar cane bagasse reduces the digester pressure and produces pulps with better mechanical properties (Sanjuan et al. 1993). The soda-ethanol pulps give a better response to bleaching than soda pulps. The most suitable bleaching sequences are OZP, OPD and O/PHP. When a brightness 80-85 % ISO is required, an unbleached bagasse pulp with a kappa number of 12-15 is needed. The pulping can also be carried out in two or more stages using aqueous ethanol in the first stage and aqueous soda-ethanol-anthraquinone mixture in the other stages. This would enable by-product recovery from the first stage. Martinez and Sanjuan (Martinez & Sanjuan 1993) have compared ethanol-water-soda pulping of Chihuahua pine with kraft pulping. They found that the organosolv pulps had strength properties similar to those of kraft pulps. A pulping method where chips are delignified using an aqueous solution of ethanol (30-40 v-%), ammonia (10-20 % on wood) and sulphur dioxide (5-30 % on wood) at 433- 438 K for 1-3 hours has been presented. Pulp yields are 67-70 % for hardwoods and 57- 60 % for softwoods. The delignification is carried out in a process called drip percolation pulping. Wood chips are drip wetted in the digester with the liquid being a pulping and washing liquor simultaneously. This liquid partly evaporates, condenses, drips onto the material being pulped, percolates through the layer and extracts dissolved components carrying them away. Aspen pulps with a yield of 70.6 %, a kappa number of 22.6 and a brightness of 62.5 % have been produced in a laboratory scale unit. The advantages of the method are claimed to be: pulping at low liquor ratios, simultaneous pulping and washing, formation of high concentrated spent liquors, closed cycle operation, a reduction of the energy consumption, prevention of pollution and compact equipment. (Krotov 1993) Alkaline ethanol pulping of wheat straw can be significantly improved by an acid catalyzed prehydrolysis. The breaking length was improved by 8 % and the tear index by 25 %. (Papatheophanus et al. 1995) Kulkarni’s group (Kulkarni et al. 1994) has delignified Agave sisalana under atmospheric conditions using aqueous ethanol (50 %) and alkali (2 %) at 353 K for four hours. The yield of unbleached pulp was 88.6 % and the kappa number was 37. The strength properties of the pulp were found to be comparable with corresponding kraft pulps. El-Sakhawy et al. (El-Sakhawy et al. 1995b) have studied pulping of cotton stalks with aqueous ethanol and alkaline ethanol processes. The alkaline ethanol method produced pulps with acceptable quality under mild conditions while the aqueous ethanol process required the use of a catalyst. When pulping wheat straw with aqueous ethanol, the addition of sodium hydroxide and anthraquinone improves the screened yield of pulp. (El-Sakhawy et al. 1996a) Kinetic studies of wheat straw pulping using the alkaline ethanol method have indicated that the delignification reaction rate increases directly with temperature. It also doubles if the alkali concentration is doubled. (El-Sakhawy et al. 1996b) Alkaline ethanol pulping of wheat straw, bagasse and cotton stalks produces pulps that are easier to bleach than corresponding soda and ethanol pulps. (El-Sakhawy et al. 1996c) Hultholm and co-workers (Hultholm et al. 1998) have presented a new pulping method they call the IDE-process. The method is a three stage process - Impregnation, Depolymerization (Delignification) and Extraction. The raw materials are impregnated with alkali in water and then cooked in an aqueous ethanol solution. The depolymerized lignin is extracted and washed out from the pulp. Pulp yields are 2-5 % higher compared to kraft pulps of the same raw materials. The addition of a 0.3 mole fraction of ethanol to the kraft pulping liquor when delignifying aspen or spruce has been found to improve the productivity of the process. This is due to the combined effect of faster delignification, increased pulp yield, and decreased activation energy of delignification. (Yoon et al. 1997, Yoon & Labosky 1998)

2.1.2.4. Ethanol in sulphite pulping

Ethanol can be used to extract xylose from birch sulphite waste liquor. When the extraction is carried out in three stages, the isolated xylose after one recrystallization is pure enough to be used in pharmaceutical and food products. (Saarnio & Kuusisto 1971) The yields of eugenol and isoeugenol can be improved during Mg-base alcohol- bisulphite pulping of softwoods by adding air to the digester and adjusting the alcohol concentration and the sulphite dosage. The pulps produced under favourable conditions for the production of these phenols have higher yields than kraft pulps at the same kappa number levels. Unbleached pulps have inferior strength properties to those of corresponding kraft pulps. The physical properties can be improved by alkali treatment. (Mun et al. 1988) Adding of alcohols to bisulphite cooking of aspen has been found to enhance delignification. Optimal sulphur dioxide dosage and alcohol (ethanol or isopropanol) concentration were 4-5 and 50 v-% respectively. When the yield of low molecular weight phenols was optimized, the sulphur dioxide dosage was 5-6 % and the alcohol (ethanol, propanol or isopropanol) concentration was 40-50 %. (Mun & Wi 1991a, Mun & Wi 1991b) Hemicellulose and cellulose can be converted to sugars in a three-stage process. Sulphur dioxide is used in the first stage to remove hemicellulose from hardwoods. In the next stage, a sulphur dioxide-ethanol-water mixture is used to remove lignin and the remaining hemicellulose. The third stage is an acid hydrolysis of cellulose to glucose. The use of the organosolv stage to remove lignin allows more efficient acid hydrolysis at less severe conditions. (Westmoreland & Jefcoat 1991) The Timermane group (Timermane et al. 1992) has studied the kinetics of birch delignification in aqueous ethanol with the addition of sulphur dioxide and ammonia. The results indicated that the process deviates from a first order reaction, but some correlation with the diffusion kinetics theory was found. The delignification rate was found to be limited by the diffusion rate of the delignification products. Agricultural residues have been cooked with 15 w-% ammonia and 10 w-% sulphur dioxide (on dry material) in aqueous ethanol (65 %). Their results indicated that this method produces pulps with higher yields than those obtained in kraft, soda or soda- anthraquinone pulping of agricultural residues. (Krotov & Frank 1993) The methanol using ASAM process has been modified by replacing methanol with ethanol. The new method is called ASAE (alkali-sulphite-anthraquinone-ethanol). Pine ASAE pulps were found to have better physical and optical properties than control kraft pulps with the exception of breaking length. (Kirci et al. 1994) Sanjuan (Sanjuan et al. 1995) has compared ethanol-magnesium sulphite pulping of whole kenaf with kraft and soda pulping. Using aqueous ethanol (50 %) and a total sulphur dioxide content of 4 %, they were able to produce pulps with higher yields than those produced with the conventional methods. The mechanical and optical properties of the kenaf pulps were found to be better than those of bagasse pulps.

2.1.2.5. Simulation of ethanol pulping

Muurinen et al. (Muurinen et al. 1988b) have proposed a solvent recovery process for alkaline ethanol pulping (Marton & Granzow 1982). Spent pulp washing liquor is evaporated and the residue is burned. Alkali is recovered by causticizing. Ethanol is concentrated by distillation. This process was simulated and the costs of the equipment and production were estimated. The costs of alkaline ethanol pulping was compared with the costs of an alkali-methanol-anthraquinone pulping (Gasche 1985a, Gasche 1985b) process having a similar recovery section. The difference between equipment costs is negligible but the production costs of ethanol pulping are lower. When peroxyformic acid pulping of birch (Poppius et al. 1987) and acetic acid pulping of aspen (Young & Davis 1986) were added to the comparison, alkaline ethanol pulping (Marton & Granzow 1982) still seemed to be the most economical of the processes studied. (Muurinen & Sohlo 1988b) Vanhanen and co-workers (Vanhanen et al. 1989) have preliminarily dimensioned a recovery process for ethanol-alkali-anthraquinone pulping using computer simulations. In the recovery process, pulp is washed after digester flashing, spent pulping and washing liquors are concentrated in a vacuum evaporator plant, and ethanol is concentrated by distillation. Methanol is recovered as a by-product by distilling the vapour formed during digester flashing. The energy consumption of the process was estimated to be near that of kraft pulping. The pulp manufacturing costs seem to be higher than the costs of a kraft process because of the higher chemical costs. New chemical industry simulators (e.g. HYSIMTM and ASPEN PLUSTM) appear to be capable of predicting vapour-liquid equilibria for ethanol-water and methanol-water systems (Barna et al. 1991). Evaporator train simulations performed with ASPEN PLUSTM are also able to handle model compounds representing lignin and sugars. 2.1.2.6. The kinetics and mechanism of delignification

The kinetics of beech wood delignification by ethanol-water mixtures have been studied (Kleinert 1967). Two different mechanisms both of first the order were found: relatively fast bulk delignification and slow residual delignification. The latter resulted in considerable cellulose losses. The pulping liquor contained 40 w-% ethanol at the start of the pulping. The bulk delignification in ethanol-water pulping is a composite phenomenon involving breakdown of high molecular lignin and solubilization of its breakdown products (Kleinert 1975b). The removal of wood moisture reduces the rate constant of delignification. Ethanol acts as a scavenger for the free radicals formed during pulping and reduces the extent of lignin condensation. It has been shown that organic solvent bulk delignification occurs in two steps which can be described by first-order kinetics. After testing several solvent systems, including the ethanol-water system, they have concluded that the mechanisms of bulk delignification do not change in various systems. (Hansen & April 1982) Heating aspen wood with 50-75 % ethanol-water mixtures for 30-90 minutes at 468 K and yields unbleached pulps with kappa numbers, viscosities and yields in commercial ranges. Solubilization primarily occurs through cleavage of beta ether linkages, yielding an isolated lignin with a phenolic hydroxyl content more than twice as high as the lignin as it exists in the wood. Some side chain rearrangement and ethoxylation of benzyl alcohol groups occurs. These reaction prevents the lignin molecule from recondensing during the pulping process. (Gallagher et al. 1989) The variation of the chemical component of bagasse during sodium hydroxide catalyzed ethanol pulping has been studied (Liang et al. 1989). The delignification process is divided in two phases: the bulk delignification phase and the supplementary delignification phase. The delignification rates are 70 % and 16 % respectively. The dissolved lignin to dissolved carbohydrate ratio is 1.52. The molecular weight of ethanol lignin is higher than that of bagasse kraft lignin. Delignification of birchwood with a 35:35:30 mixture of ethanol, water and acetic acid has been studied with respect to reaction time (15-380 min) and temperature (423- 463 K). The dissolved lignin was found to pass through stages of degradation and crosslinking. At higher temperatures, condensation became dominant over degradation. When the pulping time was increased, a part of the lignin was absorbed in the pulp fibres. (Aliev et al. 1990) Autocatalyzed delignification of birch in 60:40 ethanol:water in the temperature range 448-478 K has been studied. The delignification followed a bulk phase which was followed by a slower residual phase. When the reaction was assumed to be first order with respect to lignin concentration, an activation energy of 82.8 kJ/mol was calculated for the bulk delignification phase. The acidity of the system increased with increasing temperature and time. If this acidity effect is taken into account and the rate constant is assumed to be directly proportional to average hydrogen ion concentration, an activation energy of 67.8 kJ/mol was calculated. The low activation energy values point out that diffusion is a very important factor in the delignification process. (Goyal & Lora 1991) Deineko (Deineko 1992) has proposed a possible mechanism of wood delignification by oxygen in aqueous organic (e.g. ethanol) media. The initial rate determining stage of the process is an acid catalyzed eliminative reaction proceeding through the carbonium cation formation. The main products accumulated in the spent liquor from oxygen-ethanol pulping of spruce have been found to be partially oxidized lignin (17-20 %) and ether-soluble (1.8- 2.0 %) and ether-insoluble (21-24 %) substances. Lignin dissolution is suggested to be associated with lignin degradation by opening of the aromatic rings and rupture of the 1,2 carbon-carbon bonds of the propane side chains. Vanillin and vanillic acid were found in the spent liquor. Also glycolic, oxalic, pyropyruvic, glyceric, succinic, glutaric, muconic, formic and acetic acids were found. (Deineko et al. 1993) The mechanism of ethanol and alkaline methanol delignification of eucalyptus has been studied using scanning electron microscope observations of the resulting pulp. The findings suggested that delignification during ethanol pulping proceeds from the lumen wall toward the middle lamella, then reaching the cell corners. There was not enough evidence to suggest the same mechanism for the alkaline methanol pulping. (Wang et al. 1993) Balakshin and Deineko (Balakshin & Deineko 1995) have studied the delignification reactions during oxygen-ethanol pulping (pH=3, 413 K, 60 % ethanol) of spruce sawdust. According to their results, reactions of benzyl alcohol and benzyl aryl structures are the major reactions in delignification. While the aqueous ethanol delignification kinetics of wood occurs in three phases (initial, principal and residual), the delignification of sugar cane bagasse presents only the principal and residual phases. The Arrhenius activation energy for the principal stage has been calculated to be 101.9 kJ·mol-1. (Curvelo & Pereira 1995)

2.1.2.7. Ethanol lignins

Cooking of spruce wood meal in ethanol-hydrochloric acid yields, in addition to water- insoluble lignin, a mixture of distillable oils in amounts equal to about 8.2 % of the lignin present in wood. The mixture contains aldehydic, acidic, phenolic and neutral products. (Cramer et al. 1939) Schweers and Meier (Schweers & Meier 1979) have analytically characterized, through ethanol-water pulping isolated beech and spruce lignin. In comparison to alkali lignin the ethanol-water lignin has a significantly higher yield of monomeric phenols. The yield and composition of the phenol mixture does not make pyrolysis a technically promising process for destruction of the lignin. After studying the decomposition of beech and spruce by pyrolysis in a fluidized-sand bed equipment, Kaminsky (Kaminsky et al. 1980) found the method to be promising. Lignins isolated by ethanol-water pulping from beech, oak, birch, spruce and pine and alkali lignins from beech and spruce have been decomposed by alkaline oxidation under pressure, using oxygen as the oxidizing agent. Yields of the carbonylic products obtained from ethanol-water lignins were higher than those obtained from alkali lignins. Vanillin can be produced in amounts comparable with those obtained from sulphite waste liquors. (Meier & Schweers 1979) Lignins from beech, oak, birch, spruce and pine isolated by ethanol-water pulping have been degraded by catalytic hydrogenolysis. Because of the small yields of the monomeric phenolic products, no economic utilization of lignin using this method could be deduced. (Meier & Schweers 1981) Lignin has been isolated by pulping beech and pine wood with aqueous ethanol using 0.0-0.75 % on wood ammonium chloride as the catalyst (Lange et al. 1981). Increasing the catalyst charge was found to increase the methoxyl content of the lignins and to decrease the hydroxyl content. The catalyst also increased the lignin yield. Lignins isolated from beech wood and pine wood by the ethanol-water pulping method can be used as a wood glue either alone or together with phenolic resins (Ayla 1982). The use of a mixture of 90 % ethanol-lignin and 10 % of a phenolic resin leads to boiling proof bonds. Lange’s group (Lange et al. 1983) has separated ethanol-water lignins from birch and spruce into five fractions of increasing molecular weights. The deciduous wood character of the birch lignin fractions are more pronounced at higher molecular weights. The lignin recovered from Alcell pulping has a low molecular weight (about 1000), is highly hydrophobic and insoluble in neutral or acidic aqueous media, but soluble in moderate to strong alkaline solutions and certain organic solvents (Lora et al. 1989). It has many potential applications. It shows promise as a partial replacement of oil based resins in wood composites. Solubilized lignin recovered from the black liquor of with sodium bicarbonate buffered aqueous ethanol (50 %) pulping is relatively uncondensed with few carbohydrates associated with it (DeLange et al. 1991). Solubilized syringyl lignin removed first is more etherified than syringyl lignin removed later. Oxidation of lignin by hydrogen peroxide in ethanol has been found to result in 3- to 10-fold higher chemiluminescence than with oxidation of lignin in water. (Selentev 1991) Oliveira et al. (Oliveira et al. 1994) have extracted lignins from pinus caribaea hondurensis sawdust using ethanol, methanol, acetone, chloroform, tetrahydrofuran, 1,4- dioxane, 1-butanol, 2-butanol and 1-propanol) and made Langmuir monolayers of them. The behaviour of organosolv lignins was found to differ considerably from the behaviour of lignins from traditional pulping processes. The different behaviour of lignins extracted with different solvents was explained by the different values of the surface potentials and areas per molecule.

2.1.2.8. Summary

As with most of the organosolv pulping processes, ethanol pulping is more suitable for hardwoods than for softwoods. Aqueous ethanol penetrates easily into the structure of wood resulting in uniform delignification. Ethanol also reduces the surface tension of the pulping liquor which improves the diffusion of other pulping chemicals, for example, alkali. High pressures and temperatures used in ethanol pulping cause potential hazards of fire, explosion, and release of toxic chemicals. This has to be taken into account when designing the process. The most advanced ethanol pulping method is the Alcell process. It produces hardwood pulps of satisfactory quality. The solvent recovery is claimed to be easy. On the other hand, some researchers have found that black liquor from aqueous ethanol pulping contains, in addition to the components of the pulping liquor, lignin and sugars also aliphatic acids, aromatic acids and phenols. This highlights that the purification of the solvent may not be as easy as claimed. The addition of ethanol to conventional pulping liquors also seems to have positive effects on pulping productivity. The use of ethanol probably causes an increase in the production costs. If small amounts are added, ethanol is probably not recovered but lost during chemical recovery. Large amounts of added ethanol makes it necessary to invest in its recovery.

2.1.3. Other alcohols

Methanol and ethanol are the most popular alcohols used in pulping. However, several other alcohols have also been proposed to be used as pulping chemicals. Propanol, butanol and glycols appear in several papers which are therefore referred to here. Other alcohols proposed are benzyl alcohol (Heden & Holmberg 1936, Friedman & McCully 1938), tetrahydrofurfuryl alcohol (Johansson et al. 1987, Aaltonen et al. 1987, Aaltonen et al. 1989), glycerol (Hibbert & Phillips 1930, Demirbas 1998) and chloroethanol (Schutz 1941a, Nimz et al. 1986b, Pellinen & Nimz 1986).

2.1.3.1. Propanol and isopropanol

Semichemical and chemical pulps of Japanese cedar have been produced with sulphite solutions containing 40-50 % isopropanol. Adding the alcohol to sodium and magnesium sulphite solutions resulted in significant savings of energy for fiberizing semichemical pulps. Pulps produced by this method had better strength properties than corresponding conventional semichemical pulps. Chemical pulps could be produced with acidic magnesium sulphite and isopropanol. Pulp yield was 42-45 %. The strength properties of these pulps were lower than those of corresponding kraft pulps. (Sakai et al. 1983) Alcohol-magnesium bisulphite pulping of pine provides isoeugenol and eugenol as by-products (Sakai et al. 1987). The yield of softwood alcohol-bisulphite pulps is higher than that of kraft pulps at the same kappa number level. Of the alcohols tested, the best results were achieved with isopropanol and sec-butanol. Addition of 40 % isopropanol to the neutral sulphite pulping of pine accelerates the delignification, but has a negligible effect on neutral sulphite-anthraquinone pulping (Sakai & Uprichard 1987). Both the rate and selectivity of delignification are improved by isopropanol addition in magnesium bisulphite and magnesium acid-bisulphite pulping. Isopropanol-bisulphite pulps have better yields, lower kappa numbers and better papermaking properties than conventional bisulphite pulps. Isopropanol acid bisulphite pulps have similar strength properties to conventionally prepared pulps, but the rate and selectivity of delignification is improved. Deineko and Nikitina (Deineko & Nikitina 1989) have studied the use of oxygen as a delignifying agent during pulping of spruce sawdust in propanol and aqueous propanol. When oxygen was present in pulping with anhydrous alcohol, the amount of lignin and carbohydrates dissolved was more than twice the amount dissolved in an inert atmosphere. Pulps produced by cooking in 60 % propanol at 433 K for two hours are highly delignified, containing about 7 % lignin and have yields around 50 %. The advantages of oxygen-alcohol pulping over straight alcohol pulping include a lower cooking temperature and the possibility of using both hardwoods and softwoods (Deineko & Kikitina 1989). Mun (Mun 1989) has delignified a softwood (sugi) using aqueous isopropanol (50 v- %) and magnesium bisulphite. The lignin recovered from the black liquor exhibited a rather high molecular mass and only small modifications had occurred in its chemical structure. When sugi (Cryptomeria japonica) and buna (Fagus crenata) were pulped with 50 % aqueous isopropanol, high amounts of low-molecular-weight phenols were obtained. Higher or lower concentrations resulted in lower yields. (Kim et al. 1996)

2.1.3.2. Butanol

As with many other alcohols, butanol was originally used as a solvent to decompose wood in order to study lignin (Hagglund & Urban 1927). Aronovsky and Gortner (Aronovsky & Gortner 1936) have cooked aspen sawdust with aqueous solution of aliphatic monohydroxy and polyhydroxy alcohols. Normal butyl alcohol pulping resulted in a pulp comparable with commercial aspen soda pulp in strength characteristics. Softwoods were not pulped as easily as the aspen. The spent pulping liquor consisted of two layers. The top alcoholic layer contained the organic substances dissolved from wood, and the lower aqueous layer contained only small amounts of water soluble substances. The solubility of n-butyl alcohol in water is 8 - 9 parts per 100. Hardwoods and softwoods have been delignified with aqueous butanol. The hardwoods were pulped satisfactorily, but the softwoods were only partly pulped. Birch and basswood were most efficiently delignified. (McMillen et al. 1938) Bailey has patented (Bailey 1939) a process producing pure cellulose from wood by treating it with an aqueous mono-hydroxy alcohol having at least four carbon atoms. At least three of the carbon atoms have to be in a straight chain. The alcohol has to be substantially insoluble in water at temperatures below 373 K, but soluble therein at temperatures above 373 K. An inorganic alkali is also added to the pulping liquor in amounts from 2 % to 10 %. Pulping is carried out at 373-473 K. Aspen and pine have been digested in aqueous butanol that was buffered to prevent hydrolysis. It was found that all the lignin from aspen and 80 % of the lignin from pine was removed. The remaining pine lignin was easily dissolved in the same solvent in the presence of 2 % sodium hydroxide. Lignin studies indicated that lignin may exist partly in chemical combination with cellulose and partly free. (Bailey 1940a) The addition of 2 % alkali (NaOH) to a pulping liquor containing equal amounts of butanol and water considerably accelerates the delignification. Almost white pulps containing 0.33 % lignin were produced in yields of 43-50 %, by cooking aspen for 15 minutes and pine for 1 hour. Butanol was easily recovered. Only 10 % of the total liquor volume had to be distilled, because 95 % of the butanol was immiscible in the liquor and could be separated. Extraction with di-isopropyl ether can also be used to recover butanol from spent pulping liquor. (Bailey 1940b, Bailey 1942) If the butanol pulping liquor is made alkaline by adding sodium hydroxide, the spent liquor is separated into two layers upon cooling. The lignin is concentrated in the lower aqueous layer. The upper alcoholic layer can be directly used in the preparation of pulping liquor. (Bailey 1940c) Aqueous n-butanol (50 v-%) can be used to effectively delignify southern yellow pine meal. At 478 K and 2.2 MPa the original lignin content of wood (30.2 w-%) can be reduced to 16.1 % when the residence time is 12 hours. (Bowers & April 1977) McGee and April (McGee & April 1982) have compared aqueous ethanol and aqueous 1-butanol pulping of pine. They found that butanol pulping produced significantly higher delignification levels than ethanol pulping. Rice straw has been delignified with aqueous butanol in the presence of catalysts. Acid-catalyzed treatment with aqueous butanol (50 %) at 443 K for one hour resulted in almost complete (> 90 %) lignin removal. The lignocellulose components can be easily separated when pulping with butanol. The solvent phase contains lignin, the aqueous phase contains sugars and the cellulose remains in the solid phase. (Selvam et al. 1983) Ghose and co-workers (Ghose et al. 1983) have compared butanol and ethanol in pulping of rice straw. Both alcohols are effective delignifying agents for agricultural residues. Due to the higher swelling of the material and increased accessibility of solvents, pre-soaking improves the delignification process. Butanol has a larger molecular size and may be a better lignin solvent due to its polarity, but it is not as effective as ethanol in the decrystallization of cellulose. The influence of the solvent on the yields of dissolved lignins from Pinus Hondurensis Caribaea has been studied. The highest yield (81.5 %) based on Klason lignin was obtained with n-butanol. The other solvents were methanol, ethanol, n- propanol, 2-butanol, tetrahydrofuran, 1,4-dioxane, acetone and chloroform. (Curvelo et al. 1990b) The Terenteva group (Terenteva et al. 1990) has analyzed the structure of the fibrous material of aspenwood after delignification with aqueous butanol (50 %) in the presence of 0.5-1.0 v-% hydrochloric acid as the catalyst. To accelerate delignification, and to prevent lignin condensation, the chips were first extracted with aqueous butanol. The main factors affecting fibre structure were the amount of the catalyst used and the preliminary extraction process. A process for combined pulping and bleaching has been presented. Pulping is carried out in an alcohol solvent containing, for example, tert-butyl alcohol. Tert-butyl peroxide is used as the bleaching agent. Alcohol formed from the hydroperoxide during bleaching is recycled to the pulping. (Shaban & Suciu 1993)

2.1.3.3. Glycols

Glycols have been used to isolate lignin from wood (Fuchs 1929, Hibbert & Rowley 1930, Hibbert & Marion 1930, Rassow & Gabriel 1931a, Rassow & Gabriel 1931b, Rassow & Gabriel 1931c, Rassow & Gabriel 1931d, Gray et al. 1935, Freudenberg & Acker 1941) and to produce pulp (Anon. 1920, Rassow & Gabriel 1931d). Pulping of spruce with glycol using hydrochloric acid as the catalyst yields a lignin- free pulp containing 2 % pentosan. After removing the pentosan, the pulp yield is 41.8 %. (Rassow & Wagner 1931a, Rassow & Wagner 1931b) Annual plants can be delignified at high temperatures (403-433 K) and elevated pressure using glycol or glycerol as solvents. The delignification can be carried out without a catalyst but the addition of 0.2 % formic acid or acetic acid accelerates the process. (Erbring & Geinitz 1938) Erbring (Erbring 1939) has delignified annual plants and milled birch and pine woods with glycol using mineral or organic acids as catalysts. The pulps produced can easily be bleached with alkaline hydrogen peroxide and they can be used in paper production. Wheat straw, pine chips and milled spruce have been delignified at 423-433 K with glycol or glycerol in the presence of small amounts of hydrochloric acid or formic acid. Even aqueous glycol with water contents up to 50 % can be used. (Erbing & Peter 1941) Anhydrous ethylene glycol has been used to delignify spruce at 423 K. If aqueous glycol is used, the temperature has to be increased. The moisture content of the chips has to be 10-13 %. The delignification rate can be increased by adding organic acids, phenol or methanol, but these catalysts have a negative effect on pulp quality. Pulps produced by this method are soft. (Nakamura & Takauti 1941) Grondal and Zenczak (Grondal & Zenczak 1956) have patented a pulping process where triethylene glycol is used as the solvent. The pulping liquor contains anhydrous triethylene glycol and 0.03-0.5 % by weight of the wood chips of aluminium chloride. The pulping is carried out at 393-408 K and at atmospheric pressure for a period of 1-4 hours. If superatmospheric pressures are used, no catalyst is needed. Lignin is precipitated from the black liquor by adding water. The remaining liquor is concentrated by evaporating water therefrom and re-used as pulping liquor. A pulping process where ligno-cellulosic material is treated at atmospheric pressure and a temperature of 313-363 K with a liquor containing a glycol with two to three carbon atoms and an alkaline reagent has been patented by Meyer (Meyer et al. 1961). The alkaline reagent can be selected from the group consisting of alkali metal hydroxides and carbonates. The alkali charge is 5-50 % on wood calculated as sodium hydroxide. Alkali metal sulphides can also be used, the charge being 3-10 % on wood. Schwenzon (Schwenzon 1965) has made pulp using a liquor containing glycol and salicylic acid. The cooking temperature was 433-443 K and the method can easily pulp birch and poplar. He (Schwenzon 1966) has also used acetylsalicylic acid (3 %) in glycol. The poplar pulps he obtained were so bright that no bleaching was needed. The dissolved lignin precipitated by diluting the spent pulping liquor was very reactive. Eucalyptus (E. regnans) pulps have been prepared with a three per cent solution of salicylic acid in glycol at 443 K. The pulps have low kappa numbers and yields of about 50 %. The method is acid catalyzed and the papermaking properties of the pulps are similar to those of bisulphite pulps. (Nelson 1977) Ethylene-, propylene-, and butyleneglycols or other higher alcohols can be used to dissolve lignocellulosic materials. Prior to the alcohol treatment, the material has to be impregnated with acids (eg. sulphuric acid) and dried at a temperature below 373 K. The alcohol treatment is carried out at a temperature above 473 K for 2 to 60 minutes. The material is partly or totally dissolved during the treatment. The dissolved material can be used in production of plastics. Also cellulose can be produced with this method. (Unger et al. 1979) The Gast group (Gast et al. 1983) has studied the pulping efficiencies of monohydroxy and polyhydroxy alcohols. The most promising candidates were n-butanol and ethyleneglycol. The delignification efficiency of the ethylene glycol system was improved by catalyst [Al2(SO4)3 or AlCl 3] addition. The kinetics of birch pulping in ethylene glycol-water solutions (50:50) at temperatures 453-473 K has been studied. The delignification has two phases which can be described by first order reactions. The activation energies for the first and second delignification phases were determined to be 81.2 and 152 kJ/mol respectively. (Gast & Puls 1984) In a later paper, Gast and Puls (Gast & Puls 1985) reported that the activation energy of the second delignification phase is 135.5 kJ/mol. Sufficiently delignified pulps are obtained in pulping of hardwoods. Lignins recovered from the black liquor can be used as extenders in phenolic resin adhesives. A process for producing cellulose paste from wood or vegetable shavings has been patented (Bruniere & Galichon 1988). The lignocellulosic material is cooked in an aqueous solution containing 2-20 w-% of alkaline metal hydroxides, alkali metal salts, or alkaline earth metal salts at 423-473 K for 15-60 minutes. If the process is operated at atmospheric pressure, the aqueous solution may also contain 2-20 w-% of a solvent, for example, a glycol. A liquefaction process, the UDES-A process, for concentrated pastes of finely divided lignocellulosics has been presented (Vanasse et al. 1988). When ethylene glycol was used as the solvent in the process, only partial liquefaction was achieved. The probable cause of this behaviour is complex formation between ethylene glycol and cellulose. Thring and co-workers (Thring et al. 1989a) have isolated a prototype solvolysis lignin from poplar using ethylene glycol as a solvent. The characteristics of the lignin suggested that it is a native-type lignin. Ethylene glycol and kraft lignins have been fractionated by extraction with ethyl acetate and methanol. Glycol lignin fractions showed a lower carbon content than the kraft lignin fractions. Ethyl acetate was found to be a good solvent for selective extraction of low molecular weight material from glycol lignin. (Thring et al. 1989b) Burkart has patented (Burkart 1989) a method for removal of lignin and other non- carbohydrates as well as non-cellulosic carbohydrates from lignocellulosic materials. The material is first impregnated with a solution that is a reaction product of triethyleneglycol and an organic acid. After impregnation, the material is rapidly heated to a temperature between 392 and 403 K, where it is maintained for 2-5 minutes. Conventional techniques can be applied to separate the pulp and the spent liquor. The recovery of lignin from the black liquor after ethylene glycol pulping has been found to be more practically carried out by dilute acidification of the spent glycol liquor than by evaporation. At low temperatures, the lignin recovered is a gel-like material impossible to filtrate, but at higher temperatures (> 343 K) the lignin precipitates into fine particles and forces a very slow filtration. The optimum temperature for lignin precipitation is 323-333 K. One part of an acidic aqueous solution is added to three parts of black liquor. The concentration of hydrochloric acid in the precipitation liquor is 0.05 w-%. Gel permeation chromatography indicated that the average molecular weight of the glycol lignin is 4762 g/mol. Electron microscopy showed that the lignin contains essentially spherical particles of size range 0.5-2.5 m. (Thring et al. 1990a) After alkaline hydrolysis of glycol lignin, the lignin residue has higher carbon and lower oxygen contents than the original lignin. The calorific values of these residual lignins is increased during the depolymerization by alkaline hydrolysis. (Thring et al. 1990b) Thring (Thring et al. 1991) has fractionated glycol lignin by solvent extraction. Lignin and its fractions were characterized by determination of methoxyl and total hydroxyl groups. 13C NMR spectroscopy was also used. 26 % of the lignin was found to consist of a low molecular weight fraction soluble in ethyl acetate. This fraction is richer in guaiacyl units and hydroxyl groups than the other fractions. They (Thring et al. 1993a) have also characterized glycol lignins using 1H and 13C NMR spectroscopy. The lignins were fractionated by sequential solvent extraction. Fractions with the highest molecular weights were found to have structures linked by beta-O-4 type bonds. The lowest molecular weight fractions contained a high number of aliphatic side-chain structures. Up to 60 % of hardwood glycol lignin can be converted to liquid and gaseous products by thermal treatment in the presence of tetralin. At low treatment temperatures, syringols, guaiacols, aromatic aldehydes and ketones are formed. At higher temperatures, phenol, catechol and their methyl and ethyl derivates are the main monomers at high severities. (Thring et al. 1993b) Finely divided poplar wood or sawdust can be can be fractionated into cellulose, hemicellulose and lignin by a two-stage process. The first stage is autohydrolysis by aqueous-steam treatment. This pretreatment step removes over 90 % of the hemicelluloses and about 20 % of the lignin. The second stage is an organosolv treatment with aqueous ethylene glycol (6-10 w-%). This step yields 100 % of the cellulose fraction. (Thring et al. 1993c) When pulping aspen, birch and beech with ethylene glycol at a liquor ratio of 5:1, Rutkowski and co-workers (Rutkowski et al. 1993) obtained pulps at yields of 48-65 %, with kappa numbers of 10-63 and brightness up to 50 %. The best results were obtained using an 80 % aqueous solution of ethylene glycol at a maximum cooking temperature of 448-453 K and cooking time of 180 minutes. The results were further improved by adding 2 % sodium bisulphite to the cooking liquor and by evacuating all air from the digester. The Rutkowski (Rutkowski et al. 1994a, Rutkowski et al. 1994b) has proposed the adding of acetic acid as a modification for glycol pulping of aspen. The optimum pulping liquor composition was found to be 20 % water, 20-40 % glycol and 40-60 % acetic acid. When acetic acid is added to the pulping liquor lower pulping temperatures can be used. Glycol-acetic aspen and pine pulps are easily beatable but their strength properties are inferior to those of corresponding kraft pulps. Ethylene glycol has been proposed to be used as a pulping agent for bagasse. The equipment can be adapted from the beet sugar industry. Lignin is precipitated from the spent pulping liquor by adding water. Lignin is filtrated and the filtrate is distilled. The top product of the distillation contains organic acids and the bottom product contains the ethylene glycol. (Chaudhuri 1996) The losses of ethylene glycol during the pulping process are caused by a reaction of the solvent with wood chemicals, chemical changes in the substance itself and finally evaporation. (Surma-Slusarska 1998)

2.1.3.4. Summary

Compared to methanol and ethanol, very little attention has been paid to the use of other alcohols as pulping solvents. The probable reason for this is the fact that they all, except for ethylene glycol, are more expensive than methanol and ethanol. The relatively high boiling point (471 K) of ethylene glycol probably decreases its attractiveness.

2.2. Organic acids

2.2.1. Formic acid

Formic acid is a good solvent for the lignin and extractives present in wood. It also causes hydrolytic breakdown of wood polymers into smaller and more soluble molecules. (Sundquist 1996) Peroxyformic acid makes lignin more soluble by oxidizing lignin and making it more hydrophillic. As a highly selective chemical, peroxyformic acid does not react with cellulose and other wood polysaccharides. (Sundquist 1996) As with most of the organic solvents, formic acid has been used in addition to pulping also as solvent in order to study wood components (Pauly 1918a, Pauly 1918b, Pauly 1921, Freudenberg et al. 1936, Staudinger & Dreher 1936, Wright & Hibbert 1937, Pew 1957, Wang et al. 1990).

2.2.1.1. Pulping and bleaching with formic acid

Reznikov and Zilbergleit (Reznikov & Zilbergleit 1980) have reported that a pulp intermediate product is obtained by cooking wood chips with formic acid in the presence of hydrogen peroxide as the oxidant and sulphuric acid as the catalyst. Pulp can be produced from hardwoods and their barks by refluxing in aqueous formic acid (65-90 %) at atmospheric pressure. The resulting pulp has a high hemicellulose content and a very high holocellulose yield. The spent pulping liquor has a very low carbohydrate content and the molecular weight of the dissolved lignin is less than 1000. The wood pulps can be used in paper production or as animal feed. Bark pulp is useful, for example, as animal feed and as a replacement for tobacco. (Jordan 1982) Pulp can be produced with aqueous formic acid (80 %) containing a small amount of a catalyst. Pulping is carried out at atmospheric pressure and after 45 minutes cooking hardwoods result in pulps with kappa numbers of 60-65 and yields near 60 %. The delignification is a result of hydrolysis, where water catalyzed by formic acid and by the other catalyst reacts with lignin-hemicellulose bonds. Hemicellulose is further hydrolysed forming acetic acid and sugars. Formic acid is not consumed chemically and the only loss (10-20 kg per metric ton of pulp) is a result of acid left in pulp after washing. About 70 kg of formic acid per metric ton of pulp is formed during pulping of hardwoods. Lignin, whose molecular mass is less than 1000 g/mol, can be used to produce, for example, methanol, phenol and benzene. (Bucholtz & Jordan 1983) In pulping hardwood chips with formic acid, delignification can be intensified by the addition of an organic solvent. The solvent may be acetone, ethanol or methanol. The pulping temperature is 343-383 K and liquor-to-wood ratios from 2:1 to 5:1 can be used. The ratio of the organic solvent to formic acid is 0.15-0.5 and the pulping liquor contains less than 25 w-% water. The cellulose produced by this method is easily bleached. (Avela et al. 1988) Arnoldy and Petrus have patented (Arnoldy & Petrus 1989, Arnoldy 1989, Arnoldy & Petrus 1990, Arnoldy & Petrus 1992) a process for pulping lignocellulose-containing materials with a pulping medium containing at least 75 w-% of a solvent system. The solvent system contains 20-95 w-% formic acid, 5-80 w-% at least one C1-3 primary alcohol or ester of formic acid with primary alcohol, and optionally up to 70 w-% at least one component selected from acetic acid and C1-3 esters of acetic acid with primary alcohols. The pulping temperature is 413-473 K. When the solvent system contains 39 w-% formic acid, 50 w-% acetic acid and 11 w-% ethanol pine can be delignified at 453 K to a kappa number of 12. When Eucalyptus globulus is cooked for 90 minutes at 363 K with 80 v-% formic acid using the liquor-to-wood ratio 30:1 and catalyst concentration 0.44 %, pulp with a yield of 56 % and a kappa number of 22 is obtained. Eucalyptus grandis needs 99 % formic acid, a temperature of 368 K, 0.22 % catalyst and a liquor-to-wood ratio of 10:1 to produce pulp with a kappa number of 31 and a yield of 43 %. Hydrochloric acid has been used as the catalyst. (Baeza et al. 1991) Nimz and Schone (Nimz & Schone 1993) have applied for a patent for a pulping method where the delignification is carried out with acetic acid and formic acid. The pulping liquor contains 50-95 w-% acetic acid, less than 40 w-% formic acid and less than 50 w-% water. Fir and spruce pulping experiments have been carried out on a laboratory scale at a liquor-to-wood ratio of 10:1 and a temperature of 353 K for 60 minutes. The cooking liquor contained 50 v-% of formic acid or acetic acid and 50 v-% of a 25 v-% aqueous hydrogen peroxide solution. Sulphuric acid was used as a catalyst. Strength properties of the pulps produced were comparable to those of kraft and sulphite pulps at similar degrees of delignification. (Pen et al. 1993) Sun (Sun et al. 1994a, Sun et al. 1994b) has found that a peroxyformic acid pretreatment accelerates the rate of UV-peroxide bleaching of eucalypt kraft pulp. The best results were achieved when the bleaching was started with a nitric acid catalysed peroxyformic acid pretreatment followed by sodium hydroxide extraction and UV- peroxide bleaching. Fungal treatment of pine with Punctularia artropurpurascens or Ceriporiopsis subvermispora prior to pulping with a mixture of formic acid and acetone (7:3) has been found to slightly increase tensile and burst indexes of the resulting pulp, whilst tear indexes are reduced by up to 30 %. Pretreated samples require shorter cooking times and the resulting pulps have 29 % lower kappa numbers. (Ferraz et al. 1998a) Obrocea and Cimpoescu (Obrocea & Cimpoescu 1998) have studied delignification of spruce by peroxyformic acid. They obtained uniform delignification by cooking the chips for 2...4 hours at the maximum temperature. The best pulp yield was obtained when using the cooking temperature 363 K. A mixture of formic acid and peroxyformic acid has been successfully used to delignify eucalyptus wood and sugar cane bagasse. Resulting pulps have low kappa numbers. (Silva Perez et al. 1998)

2.2.1.2. The Milox process

The idea of using peroxyformic acid as a pulping solvent was introduced by Russian researchers (Reznikov & Zilbergleit 1980, Zilbergleit et al. 1981). The Milox process is a Finnish development of this idea. Sundquist (Sundquist 1985) presented in 1985 a two-stage atmospheric pulping method using formic acid and peroxyformic acid as solvents. In the first stage, formic acid and hydrogen peroxide form peroxyformic acid, which is a good delignifying agent. When the formic acid concentration in solvent was 60-80 %, and hydrogen peroxide was used, 20-60 % of the dried chips, pulps with kappa numbers of 2-20 and yields of 50-60 % were produced. The second stage of the method was alkaline peroxide bleaching. The amount of hydrogen peroxide needed could be reduced by pretreating the chips at reflux temperature. The brightness of bleached peroxyformic acid pulps was about 90 % ISO and birch pulps were reported to be nearly as good as corresponding kraft pulps. (Poppius et al. 1986) Pulps obtained from one-stage cooking with a high peroxide charge had high tensile strength but low tearing strength. Two-stage cooking, where the solvent in the first stage is aqueous formic acid and 5 % on wood of hydrogen peroxide is added to the second stage, resulted in better pulps. The strength properties of birch pulp could be even better than those of corresponding kraft pulp. When the pulp was bleached with the alkaline peroxide method to full brightness, the strength fell slightly. (Laamanen et al. 1986) The amount of hydrogen peroxide needed to produce fully bleached hardwood pulp with the two-stage formic and peroxyformic acid pulping and following alkaline peroxide bleaching can be as low as 2 % on wood. When softwood pulp is produced the hydrogen peroxide charge needed is 10 % on wood, which is too high for economic operation. (Sundquist 1986) The two-stage formic acid/peroxyformic acid process can be used to produce high viscosity (> 900 dm3/kg SCAN) fully bleached (90 % ISO) pulp with a reasonable yield (40-48 %) from both hardwoods and softwoods. The pulping stages were carried out at atmospheric pressure and temperatures below 373 K. The resulting pulps had kappa numbers between 5 and 35. (Sunquist et al. 1986) The hydrogen peroxide charge needed can be reduced by using a three-stage cooking method. The first and third cooking stages are carried out at 353 K with peroxyformic acid, and the second stage at the boiling point of the aqueous formic acid mixture. Another way to reduce hydrogen peroxide consumption is to carry out the only formic acid using stage at slightly elevated pressure (383 K). The hydrogen peroxide charge is also reduced when the liquor-to-wood ratio in the first and third stages of the three-stage process is reduced from 4:1 to 2:1. Hardwood and softwood pulps produced by this method can easily be bleached to full brightness with an alkaline hydrogen peroxide treatment. The minimum total hydrogen peroxide consumptions (cooking + bleaching) is 2.6 % on wood for birch and 7.9 % on wood for pine. (Poppius et al. 1987) The method has been patented in Finland (Laamanen et al. 1987), the USA (Laamanen et al. 1988) and Canada (Laamanen et al. 1991). Sundquist’s group (Sundquist et al. 1988) has examined problems related to organosolv processes using the peroxyformic acid process as an example. The most important problems are: raw-material limitations, pulping technology, washing and bleaching, pulp quality, solvent recovery, by-products and economic considerations. The environmental compatibility of many organosolv methods is very apparent, because the pulp has to be bleached using chlorine chemicals in many cases. The authors state that the peroxyformic acid process is completely free from the use of sulphur and chlorine chemicals and therefore friendly to the environment. Three-stage peroxyformic acid cooking can be used to produce birch pulps with yields of around 43 %, low kappa numbers (around 3) and high viscosities (1000-1400 dm3/kg). The method is now called the Milox process. The yield of non-chlorine bleaching to the brightness of 90 % ISO is 90-92 %. If the bleaching sequence contains only peroxide stages, the total hydrogen peroxide consumption in pulping and bleaching is 60-200 kg per metric ton of pulp. The hydrogen peroxide charge can be further decreased by replacing some of the peroxide bleaching stages with oxygen or ozone stages. (Poppius et al. 1989a) The spent pulping liquors from Milox pulping of both birch and pine can be used for grass preservation using the AIV system. This makes it possible to decrease the formic acid loss occurring during chemical recovery. (Sundquist et al. 1989) The operation of a Milox pilot plant was started in 1991. The equipment can be used to study the three-stage Milox pulping and the Milox bleaching of kraft pulps. The batch digester can be used to treat 250 kg of wood or 120 kg of kraft pulp. (Anon. 1992a) Besides the digester, there are two bleaching reactors treating 150 kg of pulp and a washing press in the pilot equipment. The digester has been surfaced with zirconium and the other parts of the equipment have been manufactured from stainless Duplex steel. The pulps produced at the pilot plant have slightly higher kappa numbers than those produced on a laboratory scale. The pilot pulps are more easily bleached to a higher brightness than the laboratory pulps. The pulp viscosity is decreased too much during bleaching. This is probably due to the high copper content of the water used. The strength properties are somewhat lower than those of the laboratory pulps because of the heavier mechanical treatment in the pilot plant. (Sundquist & Poppius-Levlin 1992) Birch pulp was produced in the first pilot plant experiments. The pulp was bleached with alkaline hydrogen peroxide to the brightness of 85 % and it was used to produce paper containing 60 % Milox pulp and 40 % softwood kraft pulp. With the exception of strength, the paper had properties similar to those of paper produced from kraft pulp. (Koljonen et al. 1992) In the pilot plant, dried birch chips are mixed with preheated formic acid and hydrogen peroxide. The mixture is heated to 333-353 K. When the oxidising potential of the solvent is nearly used, the temperature is raised to the boiling point (383 K) of the mixture. The total pulping time is 4-5 hours. The softened chips are removed from the digester, pressed and treated again with formic acid and hydrogen peroxide at a maximum temperature of 353 K. When the pulp has been pressed, washed with formic acid and screened in water, it is bleached with alkaline peroxide. The hydrogen peroxide charge in pulping is 2 % on wood and the kappa number of the pulp after digesting is 25-30. Pulps produced at laboratory scale have a lower kappa number. This is mainly caused by the use of recycled impure formic acid in the pilot plant. The washing is probably also not effective enough. (Poppius-Levlin et al. 1993) The hydrogen peroxide consumption in three-stage Milox pulping of softwoods can be decreased from 10 % to 4 % by properly adjusting the process conditions in the second pulping stage. The result is achieved by increasing the pulping temperature to 413 K and by cooking the chips for only one hour. The resulting pulps can be bleached using alkaline peroxide. (Seisto 1993) Seisto and Poppius-Levlin (Seisto 1994, Seisto & Poppius-Levlin 1995, Seisto et al. 1995, Seisto et al. 1996) have found that the two-stage Milox method is more suitable for pulping of agricultural plants than the three-stage method, which is used for pulping wood species. They have pulped tall fescue, reed canary grass, goat’s rue and common reed using a method comprising cooking with formic acid alone (373-393 K, 60-120 min), followed by reaction in an aqueous mixture of formic acid and hydrogen peroxide (353 K, 180 min). Pulps with low kappa numbers and high viscosities were produced using only 1.5 % hydrogen peroxide on grass. The yield of unbleached pulps was 40-50 % and the pulps were easily bleached with hydrogen peroxide only. The peroxyformic acid grass pulps can be used to replace 30-50 % of the short-fibre material in fine papers without any decrease in the strength properties of the paper. When the optical properties are more important than the strength properties, nearly 70 % grass pulp may be used. When the Milox method is used to delignify agricultural plants, the resulting pulp contains all the silicon present in the plant. This enables the use of a similar chemical recycling system as in a corresponding wood pulping process. The silica is dissolved during the alkaline peroxide bleaching. (Seisto & Sundquist 1994) Tuominen (Tuominen 1995) has calculated the chemical balance for a full scale Milox process. His results show that the acid concentration in the pulping liquor should be at least 83 w-% in order to keep the acid concentration in the first pulping stage high enough. Argyropoulos et al. (Argyropoulos et al. 1995) have used the three stage Milox method to delignify birch chips. Dissolved lignins were isolated after each stage and subjected to quantitative 31P NMR and oxidative degradation analyses. During the first peroxyformic acid pulping stage, approximately 45 % of the beta-O-4 aryl ether linkages were found to be cleaved. The magnitude of these reactions was intensified during the second formic acid pulping stage. The formic acid delignification was found to preferentially cleave the syringyl rich lignin structures in the secondary cell wall. Condensation reactions were found to be predominant in the second pulping stage. A preliminary design for a commercial size Milox plant with the capacity of 200000 metric tons of pulp per year has been made. When the process was compared with a kraft mill of similar size, the results pointed out that while the investment costs were the same, production costs were higher for the Milox process. (Pohjanvesi et al. 1995, Poppius-Levlin et al. 1995) Puuronen and co-workers (Puuronen et al. 1995, Puuronen et al. 1997) have modified the two stage Milox process by replacing the formic acid partly by acetic acid. Acetic acid was found to stabilize the peroxy acids in the pulping liquor, resulting in a longer preservation of the oxidation potential of the liquor. With this method pulps of good quality were produced from common reed and reed canary grass. According to Sundquist (Sundquist 1995), the main difference between the Milox method and conventional pulping methods is that no sulphur is used in the Milox process. At the present state of development, the costs of Milox pulping and the pulp quality do not differ enough from those of the kraft process in order to make the Milox method an attractive alternative for the pulping industry. Hemicelluloses obtained as by-products in the Milox process have been used to produce lactic acid. Batch fermentations with Lactobacillus pentosus produced 27 g/dm3 lactic acid and 11 g/dm3 acetic acid when the sugar content of the initial liquor was 41 g/dm3. (Perttunen et al. 1996) The investment costs of a Milox plant have been estimated to be the same as for a kraft mill of similar size. The Milox process would be about 20 % more expensive to run because of the higher chemical and energy costs. (Sundquist 1996) According to Seisto and Poppius-Levlin (Seisto & Poppius-Levlin 1997a), the effect of formic acid on birch carbohydrates and lignin depends on the time and temperature used in the treatments. At 353 K, the selectivity towards lignin decreases as a function of time. At 380 K, the delignification selectivity improves when longer treatment times are used. The selectivity of peroxyformic acid in a mixture of formic acid and hydrogen peroxide is high and is further improved at higher concentrations. The degree of deformation has been found to be greater in birch Milox fibres than in kraft fibres. This explains why the strength properties of Milox pulps are not as good as those of corresponding kraft pulps. The deformations in Milox fibres are estimated to be caused mechanically. Therefore the authors (Seisto & Poppius-Levlin 1997b) recommend that more attention should be given to mechanical treatments during Milox pulping in order to improve the strength properties of the pulps. According to Seisto (Seisto 1998), the addition of grass Milox pulp to paper furnish improves bulk and optical properties but reduces strength.

2.2.1.3. Milox bleaching

Chlorine chemicals have to be used to reach full brightness when bleaching kraft pulps. The reactivity of residual lignin in pulp towards alkaline peroxide bleaching can be improved by peroxyformic acid pretreatment. A kraft pulp with kappa number 33 reached a delignification degree of 20-80 % in peroxyformic acid treatment, depending on hydrogen peroxide charge, formic acid concentration and temperature. In the alkaline peroxide bleaching, following the peroxyformic acid treatment, a brightness 80-85 % was reached. When an OZPP bleaching sequence followed the peroxyformic acid pretreatment, pulp with a brightness 89 % ISO was produced. (Poppius et al. 1989b) The reactivity of residual lignins of both birch and pine kraft pulps in oxygen bleaching is significantly improved by pretreating the pulps with peroxyformic acid before bleaching. Peroxyformic acid creates oxidized structures and additional phenolic groups making alkaline peroxide or oxygen bleaching easier. The pretreated pulp can be bleached with alkaline peroxide to a brightness of 83-86 %. When the final bleaching is carried out sequentially with oxygen, ozone and alkaline peroxide, both pine and birch kraft pulps reach a brightness 90 % completely without chlorine chemicals. (Poppius- Levlin & Tuominen 1991, Poppius-Levlin et al. 1991a) The method is called Milox bleaching. When the peroxyformic acid treatment is followed by an alkaline peroxide stage, or an oxygen stage and an alkaline peroxide stage, the bleaching of birch kraft pulp consumes hydrogen peroxide at 40-45 kg per metric ton of pulp. The corresponding consumption in bleaching of pine kraft pulp is 75 kg. When an ozone stage is added to the bleaching sequence, the hydrogen peroxide consumption is decreased to 24 kg for birch pulp and to 39 kg for pine pulp. Both birch and pine pulps have similar strength and optical properties to the corresponding kraft pulps bleached using conventional methods. (Poppius-Levlin et al. 1991b) When bleaching is carried out in a concentrated formic acid solution, peroxyformic acid has a higher delignification rate than peroxyacetic acid, peroxypropionic acid, Caro’s acid or polyoxomolybdate. In aqueous solutions, peroxyformic acid reacts with water forming formic acid and hydrogen peroxide. In the case of peroxyacetic acid, the corresponding reaction is slower and the delignification rate is therefore as high as in concentrated acetic acid. The viscosity of bleached pulp was found to fall linearly with decreasing delignification pH due to degradation of the carbohydrates. (Jaaskelainen 1996)

2.2.1.4. Solvent recovery in Milox pulping

Muurinen (Muurinen 1986) has proposed a solvent recovery cycle for two-stage formic acid/peroxyformic acid pulping. After cooking, the pulp is washed in a five-stage washing plant. The spent washing liquor, together with the black liquor, is evaporated. Formic acid remaining in the evaporation residue is recovered by spray drying before the dissolved solids are burned. The recovered formic acid is concentrated by distillation. Muurinen and Sohlo (Muurinen & Sohlo 1988a) have proposed a solvent recovery cycle for three-stage formic acid/peroxyformic acid pulping of birch. After cooking, the pulp is first washed with fresh solvent in order to prevent lignin condensation, and then washed with water. The black liquor and the spent washing liquor are concentrated in separate vacuum evaporation plants. The evaporation residues are dried and burned. Formic acid recovered in the evaporation and drying is concentrated by pressure shift distillation. The investment and production costs of this process are quite high. Muurinen and Sohlo (Muurinen & Sohlo 1988b, Muurinen & Sohlo 1989) have proposed recovery cycles based on evaporation and distillation for alkali-methanol- anthraquinone (Gasche 1985a, Gasche 1985b), alkaline ethanol (Marton & Granzow 1982), acetic acid (Young & Davis 1986) and peroxyformic acid pulping processes. These processes were compared with kraft pulping. The results of the comparison pointed out that in 1988 organosolv processes were not yet serious competitors for kraft pulping. When ethanol-alkali-anthraquinone pulping (Laxén et al. 1988) and peroxyacetic acid pulping (Laamanen et al. 1987) were added to the comparison, alkaline ethanol pulping still seemed to be the most economic organosolv method. The investment costs of organosolv processes seemed to be near those of a kraft process, but the operating costs of organosolv pulping are higher due to high recovery costs. (Muurinen et al. 1989) Koivusaari (Koivusaari 1990) has in his master’s thesis studied the use of adsorption in the recovery section of the peroxyformic acid pulping process. The mordenite molecular sieve (AW300) tested can be used to pass the azeotropic point of the water- formic acid mixture. Lahtinen (Lahtinen 1991) has in his thesis studied the selectivity of entrainers suitable for extractive distillation of the formic acid-acetic acid-water mixture. For closer studies, he has chosen butyric acid, which can be used to separate the mixture in a three- stage distillation sequence. Nearly pure water is obtained as distillate from the first column to which butyric acid and the acid-water mixture are fed. The bottom product is fed to the second column, from which formic acid is obtained as distillate. The bottom product of the second column is fed to the third column, from which acetic acid is obtained as distillate. The bottom product of the third column is nearly pure butyric acid and is recycled to the first column. Muurinen and Sohlo (Muurinen & Sohlo 1991a, Muurinen & Sohlo 1991b, Muurinen & Sohlo 1992) have studied the effects of chip moisture content, liquor-to- wood ratio and formic acid concentration in pulping on the economy of a suggested Milox pulping and recovery process (figure 1). The optimum liquor-to-wood ratio is around 4:1, when the chip moisture content is 20 % and the formic acid concentration in pulping is 83.5 %. The operating costs of the process and the investment costs of the recovery cycle are nearly constant with formic acid concentrations 75-88 %. Higher concentrations increase the investment costs.

Chips

HCOOH Drying of chips

HCOOH Digester 1 Distillation 353 K or H2O2 101,3 kPa Adsorption

HCOOH Digester 2 Flashing 398 K to 172 kPa 101,3 kPa

HCOOH Digester 3 H2O2 353 K Evaporation 101,3 kPa

HCOOH Washing Spent liquor with Drying Boiler solvent

H2O Washing Evaporation with or Extraction water Distillation

Pulp HCOOH

Fig. 1. The flowsheet used in simulations.

The operating costs and investment costs of four alternative recovery cycles have been compared. Four separation methods were tested: pressure shift distillation, distillation of spent washing liquor instead of evaporation, extraction of spent washing liquor with isobutyl acetate after evaporation and adsorption. Distillation seemed to be the basic separation method although adsorption also seemed to be a promising alternative. The effect of various flow configurations in Milox pulping was also tested by calculating the amounts of spent pulping liquor. A counter-current flow configuration produces less spent pulping liquor than the co-current and crosscurrent configurations, and was therefore estimated to be the most economic configuration. (Muurinen & Sohlo 1991a) The Mattila group (Mattila et al. 1992) has patented a method for recovering formic acid from pulp. The pulp is first vacuum evaporated at 343-373 K and then washed with hot water. The washing can be replaced with steam stripping at 373-413 K. Lignin can be precipitated from the black liquor by adding water. The precipitated lignin is recovered by filtration. Most of the formic acid has to be evaporated from the black liquor before precipitation. (Vuori et al. 1990) After lignin removal formic acid and sugars can be recovered by multi-stage evaporation. The concentrated liquor from an evaporation stage is diluted by the condensate from the next stage. The condensate from the first stage is concentrated aqueous formic acid. The concentrate from the last stage is a nearly pure sugar-water solution. (Saari et al. 1992) The economy of the Milox process can be improved by selling a fraction of the black liquor to be used to produce animal feed. Changing the flow configuration of the three- stage Milox pulping to a system operating both co-currently and counter-currently may also improve the economy. The relatively high operating costs of the Milox process are mainly due to the high energy consumption especially in the recovery section. Using pinch technology (Linnhoff et al. 1982) to integrate the heat flows of the unit operations can cut the operating costs down by about 14 %. (Muurinen et al. 1992) The heat integration of the Milox process (fig. 1) using pinch technology reduces the need for external heating and cooling by about 40 % and 50 % respectively. (Muurinen & Sohlo 1993a, Muurinen 1994) When separation process selection methods found in literature were applied to the Milox process, it was found that distillation, azeotropic distillation, extractive distillation, extraction, membrane processes and adsorption are potential separation methods. A mordenite molecular sieve was tested on a laboratory scale and the results showed that the capacity and selectivity of the adsorbent were too low. (Muurinen et al. 1993a, Muurinen 1994) Pressure shift distillation, extractive distillation with butyric acid and three-stage distillation without a mass separating agent have been compared as possible methods for recovering formic acid and acetic from the spent liquors of the Milox process. A preliminary evaluation showed that pressure shift distillation is the most attractive separation method. The situation was unchanged even after a thermal integration carried out using pinch technology. (Muurinen & Sohlo 1993b) Acetic acid is formed during Milox pulping and has to be separated to avoid accumulation. The feed liquor to the recovery cycle contains less than 1 w-% of acetic acid. This makes the recovery and especially distillation complicated. If pulping could be carried out with a mixture of formic acid and acetic acid, significant savings could be made in separation costs. Preliminary laboratory experiments show that pulping with the mixture is possible. (Muurinen et al. 1993b, Muurinen & Sohlo 1997a) Muurinen and Sohlo (Muurinen & Sohlo 1994) have studied the thermal integration of a proposed Milox process configuration (fig. 2). Pressure shift distillation and extractive distillation with butyric acid were used as alternative separation methods. If a heat exchanger network is designed according to a pinch analysis, the primary energy consumption of a Milox process using pressure shift distillation as the separation method is reduced by 30 %. If extractive distillation is used, the corresponding reduction is 25 %. Lahtinen (Lahtinen et al. 1994) has studied the the vapour-liquid equilibrium of the ternary system water-formic acid-acetic acid. They also measured the equilibrium of n- methyl-2-pyrrolidone (NMP) with water and the acids. These results showed that NMP is a potential entrainer in extractive distillation of the ternary system.

HCOOH

PULPING + PULP DRYING SOLVENT WASHING WASHING HCOOH SPENT LIQUOR SPENT LIQUOR WATER CHIPS EVAPORAT- SEPARATION ACID ION

PRECIPITAT- EVAPORATI- ACETIC ACID ION ON

LIGNIN CARBOHYDRATES

Fig. 2. Milox pulping and recovery.

The use of pervaporation in separating the formic acid-acetic acid-water mixture has been studied using a polyvinyl alcohol membrane, which was recommended for organic acids with two or more carbons. The best results were achieved with solutions containing less than 15 % formic acid. The best separation factor values for the acids were 3-4. (Matinheikki & Sohlo 1994) Muurinen and Sohlo (Muurinen & Sohlo 1997b) have used shortcut simulation methods in order to study the effect of feed liquor composition on the economy of a distillation sequence separation a mixture of formic acid, acetic acid, and water. Very high or low water contents in the feed liquor seem to result in the lowest thermal energy consumptions. 2.2.1.5. The chemistry of Milox pulping

The molecular weight of residual lignin in peroxyformic acid pulp increases when delignification proceeds. At the same time, the amount of unsubstituted guaiacyl units, arylether bonds and free phenolic groups decreases. Peroxyformic acid residual lignin seems to be more oxidised than residual lignin in kraft pulp. (Hortling et al. 1987) Ede’s group (Ede et al. 1988) has found that lignin model compounds were completely consumed after treating with 85 % formic acid for one hour at reflux temperature. During the early stages of the reaction, primary, secondary and phenolic hydroxyl groups were partially converted to corresponding formates. 1H NMR spectra pointed out that during long reaction times, components with high molecular masses are formed. Phenolic diaryl methanols are formed during pulping as a result of acidic hydrolysis. Significant amounts of benzofurans are also formed. (Ede & Brunow 1989a) The primary reactions of lignin model compounds in 99 % formic acid are condensation reactions between the a carbon and available aromatic groups. These condensation reactions increase the polymerisation degree of lignin. Lignin depolymerisation is therefore probably not necessary in formic acid pulping. (Ede & Brunow 1989b) Residual lignins in Milox pulps have higher molecular weights than the dissolved lignins. A part of the residual lignins have lower molecular masses. This fraction is dissolved during enzymatic hydrolysis with cellulase and it is probably more accessible to the pulping chemicals than the other fraction, that is insoluble during enzymatic hydrolysis. (Hortling et al. 1989) When five moles of hydrogen peroxide per mole of lignin model compounds were used, about 20 % of the model compounds degraded to compounds with lower molecular masses. (Ede & Brunow 1989c) Lignin can be precipitated from the black liquors of the three-stage Milox pulping by adding water. beta-aryl ether bonds are hydrolysed during the second pulping stage and therefore most of the lignin is precipitated from the black liquor of this stage. Lignin with the lowest molecular mass is obtained from the first stage, and lignin with the highest molecular mass from the third stage. It is apparent that the lignin fraction with the highest molecular mass is dissolved later than the lighter fractions. The lignin dissolved from birch in the first pulping stage has a high xylose content. The highest glucose content for softwood lignins is observed in the second stage. (Hortling et al. 1990, Hortling et al. 1991) Zule et al. (Zule et al. 1995) have determined the C9 formula of Milox lignin to be C9H7.36O2.82(OCH3)1.20(OH)0.58. The average molecular weight of the lignin was found to be from 1000 to 3000. 2.2.1.6. Summary

Formic acid seems to be a good solvent for the lignin and extractives present in wood. It has been found that it also causes hydrolytic breakdown of wood polymers into smaller and more soluble compounds. It is claimed not to be consumed chemically during pulping, but its thermal instability makes it necessary to operate using short residence times in the unit operations. Peroxyformic acid has been found to make lignin more soluble by oxidizing it and making it more hydrophillic. It is a selective solvent and does not react with cellulose and other polysaccharides. The solvent recovery in formic acid pulping is relatively complicated due to the formation of a binary azeotrope of formic acid with water and a ternary azeotrope with water and acetic acid which is formed in small amounts during pulping. Sundquist and Poppius-Levlin (Sundquist & Poppius-Levlin 1998) have published a paper where they present a summary of the research related to the Milox process. They also present the following list of the most important areas for future development work: - improving the quality uniformity of the pulp - minimising the loss of formic acid - minimising the amount of water used in the process - assessing the value of by-products. This list shows that the industrial use of a formic acid-based pulping process seems to be unlikely in the near future, because several problems still have to be solved.

2.2.2. Acetic acid

Acetic acid, like most of the organic chemicals used to produce pulp, was originally used to isolate lignin from wood (Pauly 1918a, Pauly 1918b, Pauly 1921, Pauly 1943). Apostol and Kozlov (Apostol & Kozlov 1979) have studied the effect of impregnating birch chips with aqueous acetic acid. The swelling was found to be significantly lower in chips impregnated with various concentrations of acetic acid (9-90 %) than in chips impregnated with pure water. When the concentration of acetic acid was increased, the amount of liquid absorbed by the chips also increased. The acetic acid impregnated chips did not show any significant changes in density and compression strength. In addition to pulping and bleaching, acetic acid can be used to inhibit colour reversion in pulp. (DeLong 1990)

2.2.2.1. Acetic acid pulping and bleaching

According to Schutz and Knackstedt (Schutz & Knackstedt 1942) wood cannot be delignified with aqueous acetic acid at 379-380 K, but the addition of 1-2 % hydrochloric acid makes it possible. Also magnesium chloride can be used as a catalyst when pulping with aqueous acetic acid (90 %). Wood has been delignified with aqueous acetic acid containing sulphuric acid as a catalyst. Pulping temperature of 391 K, one hour pulping time, atmospheric pressure, wood moisture content of 15 % and a liquor-to-wood ratio of 8:1 were used. The optimum catalyst concentration was determined to be 3 %. The pulp properties were satisfactory and were slightly improved by adding 0.5-2.5 % acetic anhydride to the pulping liquor. (Wiltshire 1943) Koval and Slavik (Koval & Slavik 1963) have used a mixture of acetic acid, hydrochloric acid and acetone as a pulping liquor. The use of this solvent results in good- quality pulps. The main advantages of this process are: a low temperature of delignification, pulping is carried out in atmospheric pressure, a simple chemical recovery by distillation, a simple method of obtaining lignin by filtration, a relatively high pulp yield and no water pollution. The main drawback of the process is the difficulties in the complete recovery of acetic acid. It has been found that optimum yields of pulps with the best acetylation properties were obtained by pulping wood in acetic acid containing 10 % water and 0.2 % sulphuric acid for four hours at 423 K. Both hardwoods and softwoods can be used. The acetic acid celluloses have low pentosan and mannan contents. They give acetates with properties near those of acetates produced from cotton linters. (Herdle et al. 1964) Haas and Lang have patented (Haas & Lang 1971) a pulping method where lignocellulosic material is treated with an aqueous solution containing at leat 50 w-% acetic acid. Liquor-to-wood ratios from 1:1 to 12:1 can be used and the system is heated to a temperature between 423 K and 478 K. The pulping time is 30 minutes at the higher temperature and 16 hours at the lower temperature. The pulp is reported to have excellent quality. Acetic acid or formic acid (pH 2) pretreatment of kraft pulp before ozone bleaching has been found to result in bleached pulp with higher viscosity and lower kappa number than the corresponding water-pretreated pulp. The amount of ozone consumed in bleaching is also decreased. (Mbachu & Manley 1981) Laboratory scale pulping of aspen and spruce chips has shown that good delignification can be achieved by cooking with aqueous acetic acid (50-95 %) for 30-60 minutes at 448-493 K. The stronger conditions are needed for softwoods. Acetic acid can be recovered using liquid-liquid extraction with ethyl acetate. (Young et al. 1985, Young 1985) Nimz and Robert (Nimz & Robert 1985) have studied the lignins obtained from spruce chips by continuous extraction with 95 % acetic acid and 0.1 % hydrochloric acid using the 13C NMR method. The spectra indicated the presence of acetylated g-hydroxyl groups. The splitting of the beta-O-4 linkages is also confirmed. Good delignification (kappa numbers 10-40) has been achieved by cooking aspen chips in aqueous acetic acid (50-87.5 %). The pulp yields were 50-60 %. It may be necessary to wash the pulp with acetone to remove precipitated lignin after the cooking. The pulps had satisfactory to good water absorbencies. (Young & Davis 1986) Young and co-workers (Young et al. 1986) have studied delignification of spruce with acetic acid. Pulping at a high temperature (488-493 K) for a short time (1 h) provided the highest degree of delignification. When an acetic acid concentration of 87.5 % and a liquor-to-wood ratio of 8:1 was used, pulp with a kappa number of 16 and a yield of 46.2 % was produced. The tensile strength of the pulp was comparable to that of a spruce kraft pulp with the same lignin content. Acetic acid is assumed to both promote solvation of lignin fragments and to reduce swelling of the predominantly carbohydrate fibres. The use of acetic acid-water and acetic acid-water-phenol systems as pulping solvents has been studied. Beech chips were easily delignified by cooking with aqueous acetic acid at 433 K. The pulping rate was greatly increased by adding small amounts of phenols to the pulping liquor. The optimum strength properties for the pulp were achieved with high acetic acid concentrations (> 90 %). (Sano et al. 1986) Nimz and others (Nimz et al. 1986c) have patented a method where wood is pulped with aqueous acetic acid containing up to 10 % chloroethanol. The pulping time (1-8 h), the pulp yield (40-60 %), and the pulping temperature (373-388 K) depend on the chloroethanol and water (1-55 %) content of the pulping liquor. The solvent is recovered by vacuum distillation. Both softwoods and hardwoods can be pulped. Arylglycerol-beta-guaiacyl ethers have been used as model compounds to study the delignification reactions occurring during acetic acid pulping. Ether cleavage followed two reaction paths. The first one leads to the formation of Hibbert’s ketones, while the second one yields an enol acetate. (Davis et al. 1987) Davis (Davis 1987) has in his PhD thesis reported that acetic acid pulping trials gave the lowest kappa numbers (10-16), highest pulp yield, and the best strength properties, when cooking at a high temperature for less than 60 minutes, using a high acetic acid concentration (87.5 %). Softwoods needed more drastic pulping conditions than hardwoods. The best pulps had excellent tensile strength, marginal burst strength and low tear strength relative to reference kraft pulps. According to Zilbergleit and others (Zilbergleit et al. 1987a) lignins dissolved in acetic acid pulping of birch, aspen and pine have similar properties compared to those of kraft lignins. The acetic acid lignins were more active adsorbents than kraft lignins. Both softwood and hardwood acetic acid lignins have hydrophillic properties. The data obtained during hardwood and softwood aqueous acetic acid (75 %) pulping has indicated that the reactivity of lignin changes during delignification. (Zilbergleit et al. 1987b) Yasuda and Ito (Yasuda & Ito 1987) have studied the behaviour of softwood lignin in acetic acid pulping by using model compounds. The treatment of veratryl alcohol, creosol and 1-guaiacyl-3-propanol in 90 % acetic acid gave a condensation product with Ca-C6 linkage and a phenolic acetate in high yields. Papermaking properties of birch and poplar pulps produced by cooking the chips in 75 % aqueous acetic acid at 428 K for four hours have been studied. At pulp yields of 50.0 and 48.3 %, the strength properties of the two pulps were better than those of corresponding sulphite pulp. When the spent pulping liquor was reused for pulping with the addition of 33 % fresh acetic acid, no significant changes in pulp properties were observed over ten pulping cycles. (Simkhovich et al. 1987) About 50 % of phenolic phenylcoumaran was consumed during a three hour treatment of lignin model compounds with 90 % acetic acid at 453 K yielding phenylcoumarone, a stilbene derivate and a condensation product, but the non-phenolic model was very stable. Phenolic and non-phenolic 1,2-diaryl-1,3-propanediols were unstable. They degraded within an hour giving stilbene derivates as the main products. (Yasuda 1988) Muurinen and Sohlo (Muurinen & Sohlo 1988a) have estimated the investment and operating costs of acetic acid pulping and recovery, and compared them with the costs of peroxyformic acid pulping. The investment costs of the acetic acid process seemed to be lower than for the peroxyformic acid process, which in turn had lower operating costs. When methanol and ethanol pulping were added to the comparison (Muurinen & Sohlo 1988b), the acetic acid process seemed to have investment costs near those of alcohol processes. The operating costs were estimated to be much higher for the acetic acid process than for the other processes included in the comparison. Acetic acid (70 %) pulping of beechwood has been studied. The cooking temperature was 30-150 minutes and the temperature was 428-443 K. Liquor-to-wood ratios were varied between three and five. Pulps had yields of 42-60 % and kappa numbers of 20-60. The best results were obtained using a liquor-to-wood ratio of 4.5:1 and a pulping temperature of 438 K. (Obrocea & Gavrilescu 1988) Reznikov (Reznikov 1988) has pulped aspen and birch with aqueous acetic acid for 180 minutes at 418 K. The greatest degree of delignification was achieved at an acetic acid concentration of 60-90 %. Hardwoods can also be delignified at the gas phase using a temperature of 433 K and an acetic acid concentration of 75 %. It has been reported that up to 35 % of resin phenol in lignin-PF resins can be substituted with acetic acid lignin, making it possible to improve the strength properties of plastics and particle boards by up to 20 % and to reduce the amount of free phenol by 50 %. (Zilbergleit et al. 1989a) Cellulose, especially rayon, can be acetylated easily by acetic acid using pyridine containing sulfonyl chlorides. The maximum degree of substitution of the products was obtained with equimolar concentrations of p-toluenesulfonyl chloride and acetic acid. (Shimizu & Hayashi 1989) The Burov group (Burov et al. 1989a) has reported that the optimum concentration of acetic acid in pulping liquor is 25-35 %. The rest of the liquor should consist of equal parts of water and ethanol. Aspenwood treatment with this solvent mixture produced pulp with acceptable strength properties. The effects of acidity and the dielectric constant of the solvent system on the delignification of aspen, birch and pine with an acetic acid-ethanol-water mixture have been studied. Determination of the acidity and dielectric constant was found to be a good method for estimating the delignification capacity of a solvent. (Burov et al. 1989b) Deineko and Kostyukevich (Deineko & Kostyukevich 1989) have presented a method where softwood and hardwood chips are delignified in aqueous acetic acid at elevated temperatures in the presence of an oxidizing agent. Pulp properties are improved and less chemicals are needed if the oxidizer is oxygen (15.6-19.5 % on wood) and the process is conducted at 403-433 K using a liquor-to-wood ratio of 7:1-10:1. Birch and aspen have been delignified with 75 % acetic acid in the vapour phase. After delignification the pulps were treated at room temperature with a 2 % sodium hydroxide solution. Pulps with normal yields and low residual lignin contents could be produced using a liquor-to-wood ratio of 1:1-2:1. (Zilbergleit et al. 1989b) Treated with aqueous acetic acid (90 %) containing more than 5 % p-phenolsulfonic acid, hardwood chips have produced pulps with satisfactory delignification. The rate and selectivity was improved significantly when small amounts of phenol was added to the system. Pulps produced from birch chips had sheet properties comparable to those of kraft pulps. (Sano et al. 1989a) Hardwood pulps can be produced by treating the chips with 95 % acetic acid for 4-5 hours at atmospheric pressure. 1 % hydrochloric acid was added to act as a catalyst. The pulp can be bleached with sequences PEDED or C/DEDED. The physical properties of the bleached pulps are not as good as those of the corresponding kraft pulps. (Paik et al. 1989) Burov and co-workers have patented (Burov et al. 1989c) a pulping method consisting of cooking wood chips at elevated temperature in an aqueous organic solvent based on acetic acid. Pulp yields are increased and selective delignification is improved if the aqueous solution containing 30-95 v-% acetic acid also contains ethanol. The acetic acid-ethanol ratio should be 0.17:1-8:1. It has been reported that microwave irradiation of wood chips in solutions of organic solvents is a possible method for the production of pulps with low lignin contents. The best results obtained were produced with a solution of acetic acid (43 w-%), ethylene glycol (43 w-%) and water (14 w-%). When the irradiation was carried out for two minutes, pine chips produced pulp with a kappa number of 31 and a yield of 48 %. The corresponding values for aspen pulp were 15 and 54 %. (Davis & Young 1989, Davis & Young 1991b) Kin (Kin 1990) has reported that beech wood can be pulped at normal pressure with 80 % acetic acid containing inorganic acids as catalyst. The process yielded 45-49 % pulp, 19.7-20 % acetic lignin and 2.5-5.5 % furfural. The lignins contain 5.9-6.75 % phenolic groups. The pulp contains about 93 % alpha-cellulose and 4.8-5.2 % pentosans. A two-stage method for pulping of birch has been presented. In the first stage chips are treated with aqueous acetic acid (90.4-92.8 %) at atmospheric pressure for 30-90 minutes. The second stage is a 2-3 hour treatment with 90 % acetic acid containing 1.6- 2 % sulphuric acid as a catalyst. The pulp yields are 51-54 % and the strength properties are comparable with the corresponding commercial kraft pulps. (Sano et al. 1990) Burov (Burov et al. 1990) has studied the degradative capacity of acetic acid (0.5-20 v- %) in aqueous ethanol with respect to cellulose and lignin in aspenwood. The hydrogen ion activity in solutions containing 5-20 % acetic acid influenced changes in the cellulose DP during the delignification. The lignin molecular weight characteristics varied with acetic acid concentration and water-ethnol ratios (4:6-6:4). Casuarina wood can be pulped with 50 % aqueous acetic acid, but higher acetic acid concentrations (up to 90 %) accelerate the delignification, improve the delignification selectivity, increase the pulp yield, enhance carbohydrate protection, and facilitate bleaching. The best results were obtained by using a pulping temperature of 423 K and a pulping time of 2 hours. (El-Meadawy et al. 1990) Uraki’s group (Uraki et al. 1991) has tested the pretreatment stage of the two-stage acetic acid pulping method presented by Sano et al. (Sano et al. 1990) using six hardwood species. Birch was effectively delignified by refluxing in 90 % aqueous acetic acid for 60 minutes. The pulp yield and delignification of beech were improved by removing the pretreatment liquor containing water soluble extracts, which significantly inhibited the delignification. Permeation of the pretreatment liquor to chips under reduced pressure accelerated the pulping of alder and eucalyptus, but did not affect pulping of acacia. Poplar was delignified under the same conditions as beech. Acetic acid-based pulping processes using high concentrations of acetic acid (70 or 85 %) produce lignins with higher average molecular weights (Mw) than the processes using low acetic acid concentrations (33 or 43 %). The high acetic acid concentration is assumed to promote better solvation of large lignin fragments. (Young & Davis 1991, Davis & Young 1991a) Spruce chips have been delignified by oxygen in aqueous acetic acid (80 v-%) at 423 K for 150-240 minutes using a liquor-to-wood ratio of 10:1. The initial oxygen pressure was 1.5 MPa. Pulps with good mechanical properties were obtained after pulping for 180-210 minutes. Longer pulping times result in decreased pulp quality. (Deineko & Kostjukevich 1991) Gamma-radiation is able to penetrate into wood and to initiate destructive reactions. This kind of treatment may decrease the costs of hardwood pulping in acetic acid or ethanol. (Zilbergleit et al. 1991a) The addition of 5.3 w-% potassium or sodium halides per wood lignin has been found to accelerate delignification in acetic acid pulping. A basic study with model compounds suggested that the halides have a good effect on the retardation of condensation and on the facilitation of hydrolytic solvolysis. (Yasuda et al. 1991) Kaneko and co-workers (Kaneko et al. 1991) have studied aqueous acetic acid and acetic acid-water-phenol solutions in pulping of softwoods. Todomatsu (Abies sachalinensis) was easily delignified in 70-90 % aqueous acetic acid with or without phenols at 443-453 K. Karamatsu (Larix leptolepis) was difficult to delignify without phenols, which improved the delignification. The strength properties were comparable with those of corresponding kraft pulps. The conversion of pine and aspen lignin during pulping with acetic acid has been studied. The results indicated that lignin is subjected to intermolecular condensation in the liquid phase during pulping. The chemical composition of the acetic acid lignins indicated that delignification was accompanied by the degradation of alkylaryl ether linkages, alpha-beta and beta-alpha elimination reactions, condensation, methoxylation, partial acylation, and the accumulation of carbonyl and carboxyl groups. (Zilbergleit et al. 1991b) Furfural and hydroxymethyl furfural (HMF) are formed during acetic acid pulping. The formation of polymers from these highly reactive monomers is undesirable, because it can retard delignification. Zilbergleit and Glushko (Zilbergleit & Glushko 1991) have studied polymerization in 75 % acetic acid at 438 K. They found that furfural and HMF condensation in acetic acid was accompanied by polymerization and the oxidation of carbonyl and hydroxyl groups to carboxyl groups, followed by decarboxylation. The Akishima group (Akishima et al. 1991) has studied structural changes of amorphous cellulose during acid pulping. During acetic acid (93 %) pulping, partial acetylation of the amorphous cellulose occurred. Optimum conditions for acetic acid pulping of aspen and birch have been determined using breaking length, bursting strength, folding endurance and tearing strength as optimization parameters. For aspen, the optimum properties were achieved when pulping at 418 K for 2.25-2.28 hours in an aqueous solution containing 70 % acetic acid. If the cooking temperature is raised to 438 K, and the same acetic acid concentration is used, the optimum cooking time is 0.48 hours. The optimum cook parameters for birch pulping were found to be 423-424 K, 2.28-2.55 hours and 70 % acetic acid. Using these conditions, aspen pulps with yields of 52-61 % and birch pulps with yields of 48-53 % were obtained. (Zilbergleit et al. 1991c) Wood chips can be digested in a solution containing acetic acid, ethanol and water in three stages using an acetic acid concentration of 5-10 v-%. The first and second stages are carried out at 423-438 K for 45-90 minutes, and the third stage is carried out at 393- 413 K for 10-20 minutes. (Burov et al. 1991) Results achieved by Kostyukevich and Deineko (Kostyukevich & Deineko 1992) indicate that wood delignification by oxygen in 80 v-% aqueous acetic acid produces pulps with relatively high yields and good properties. The high carbohydrate content of the spent pulping liquor makes it, after lignin removal, an attractive raw material for microbiological processing. Vazquez (Vazquez et al. 1992) has fractionated Eucalyptus globulus wood at 383 K using aqueous acetic acid containing catalytic amounts of hydrochloric acid. The effects of time (5-6 h), acetic acid concentration (70-95 %) and catalyst concentration (0-0.2 %) on yield and composition of delignified material were examined. The yield and the lignin content of pulp were mainly influenced by catalyst concentration. The polysaccharide content depended on both the acetic acid and hydrochloric acid concentrations. The treatment time was predicted to be an insignificant variable in the range tested. Wood and other lignocellulosic material can be delignified with a mixture containing at least 90 w-% acetic acid, 2-8 w-% alkyl acetate (e.g. butyl acetate), 0-8 w-% water, and a catalyst. The catalyst may be potassium acetate, aluminium acetate, ammonium acetate, calcium acetate, magnesium acetate, sodium acetate, lithium acetate, acetone, a 1- 4 carbon alcohol, or a mixture of these. (Heckrodt & Thompson 1992) Dievskii and Burov (Dievskii & Burov 1993) have derived mathematical diffusion models taking into account the change in the gradient of lignin concentration. They used the models to analyze pulping of pine and aspen with a mixture of acetic acid and ethanol. Their results showed that it could be possible to increase the delignification rate by applying mechanical or chemimechanical pretreatments in order to improve internal mass transfer in the lignocellulosic materials. Oxygen-acetic acid cooks of spruce chips and sawdust have been carried out by Nikandrov (Nikandrov 1993) in a 1 dm3 autoclave at a maximum temperature of 403- 423 K, liquor-to-wood ratio of 50:1, and acetic acid concentration of 80 %. When the maximum cooking temperature was reached, 5.25 dm3 of oxygen was added. The selectivity of delignification was found to be independent of the particle size of the wood. Because there seem to be no problems with diffusion of the reagents into the chips or sawdust, oxygen-acetic acid pulping can probably be carried out in conventional digesters. Leonova and co-workers (Leonova et al. 1994) have delignified aspen chips in a cooking liquor containing equal volumes of glacial acetic acid and an 18 % aqueous formic acid solution. Sulphuric acid was used as a catalyst. Cooking was carried out at a liquor-to-wood ratio of 4:1 and a temperature of 353 K for 60 minutes. The pulp obtained had a yield of 73 %, a lignin content of 10 %, a breaking length of 5.1 km and a bursting strength of 137 kPa. Reducing the liquor-to-wood ratio to 2:1 did not reduce the delignification selectivity but the cooking time had to be increased to 90 minutes. Lindholm’s group (Lindholm et al. 1995) has extracted a pulping liquor from the light fraction of the pyrolysis of agricultural waste materials. The liquor contains about 90 % organic acids. 80 % of the acid content is acetic acid. The spent pulping liquor is returned to the pyrolysis and no further recovery stages are needed. Pulps with good quality can be produced from especially decorticated flax straw. When cooking for 1-3 hours at 373-383 K, kappa numbers between 10 and 20 are reached. Lignins from aqueous acetic acid pulping can, without any chemical modifications, be converted into fibers that are precursors of carbon fibres. After a thermal treatment, they can be meltspinned. The mechanical properties of these fibres are similar to those of fibres made from phenolated steam-exploded lignin. (Uraki et al. 1995)

2.2.2.2. Peroxyacetic acid

Poljak (Poljak 1948) has shown that pinewood can be quickly delignified at 333-353 K in one stage using an aqueous solution of peroxyacetic acid (10 %) as solvent. Lignin is almost completely dissolved and the carbohydrates are only slightly attacked. He (Poljak 1951) has also presented a method of predicting the cellulose content of woody material using peroxyacetic acid. After a 30 minute treatment of woodmeal with 10 % peroxyacetic acid at 348 K, the cellulose contains no lignin. Haas and others (Haas et al. 1955) have used the method presented by Poljak (Poljak 1948) to obtain ashfree holocellulose from beech, spruce and pine. A method for pulping of wood at 333-373 K with peroxyacetic acid having a concentration between 10 and 40 w-% has been presented. A liquor-to-wood ratio between 8:1 and 20:1 is used. The pulp yield is 44-56 % when hardwoods are delignified. (Wayman & Harris 1960) Leopold (Leopold 1961) has compared the chlorine-ethanolamine, chlorite and peroxyacetic acid methods for producing holocellulose from pine. His results showed that the peroxyacetic acid method, modified by a subsequent sodium borohydride step, gave a superior holocellulose. Holocelluloses produced by a modified chlorite oxidation process and by the peroxyacetic acid method have been compared. No differences were observed in the strength properties. (Thompson & Kaustinen 1964) Bailey and Dence (Bailey & Dence 1966) have found that maximum brightening of kraft and sulphite pulps occur in the pH range 7-9, when peroxyacetic acid is used as the bleaching agent. Consistency (3-6 %) and temperature (323-358 K) had little effect on pulp properties in this pH range. Increasing the peroxyacetic acid application in the 1-8 % range produced lower kappa numbers and viscosity values, and higher brightness pulps. Large increases in the strength properties of mechanical pulps as a result of a post- treatment with peroxyacetic acid have been observed. Strength properties were improved when peroxyacetic acid was used at low pH (3.5-4.0). When the pH was changed to 7.0- 8.0 peroxyacetic acid functioned as a bleaching agent. A pretreatment of wood chips before mechanical fiberizing resulted in a reduction in the power needed for fiberizing and higher pulp strength properties. (Stevens & Marton 1966) Yiannos and Secrist (Yiannos & Secrist 1968) have invented a two-stage process for obtaining a lignin derivative from lignocellulosic materials. In the first stage, the lignocellulosic material is impregnated with a strongly oxidizing compound at a temperature below 323 K and a pH between 4 and 5. In the second stage, lignin is dissolved in an alkaline solution. Lignin is recovered by precipitation. The oxidizing compound can be an organic peroxide, preferably peroxyacetic acid. Also peroxyformic acid can be used. The alkaline solution can be aqueous sodium hydroxide. A method for removing lignin from lignin containing materials has been presented. The material is first pretreated with alkali to swell the material and to limit carbohydrate losses. Lignin is then dissolved under oxidizing conditions. These conditions include the use of chlorine dioxide, peroxyformic acid, peroxyacetic acid or peroxypropionic acid, and the use of temperatures between 323 and 373 K. (Thompson & Kaustinen 1969) Farrand and Johnson (Farrand & Johnson 1972) have studied peroxyacetic acid oxidation of several phenolic compounds in aqueous acetic acid. The oxidation process was found to have competitive pathways involving hydroxylation at activated ring positions, demethoxylation, ortho and para quinone formation and aromatic ring cleavage. Loblolly pine has been delignified with 3 % peroxyacetic acid at 333 K and the dissolved lignin analysed. The results they have obtained show that the aromatic rings and unsaturated sidechains of lignin were degraded in the fraction having molecular weights 10000-16000. (Albrecht & Nicholls 1974) Oki et al. (Oki et al. 1974) have studied the mechanism of peroxyacetic acid degradation of guaiacylglycerol-beta-aryl ether units in softwood lignin. They treated guaiacylglycerol-beta-guaiacyl ether, guaiacylglycol-beta-guaiacyl ether and veratrylglycol-beta-guaiacyl ether with an aqueous acetic acid (50 %) solution containing 2.6 % peroxyacetic acid at 303 K for 48 hours. The main degradation products were guaiacol, vanillic acid, catechol, 2-methoxyphenoxyacetic acid, muconic acid. Acetic acid was also found to be a better solvent for peroxyacetic acid than ethanol or dioxane. Peroxyacetic acid is a highly selective delignifying agent. This is due to its ability to oxidize the electron-rich aromatic ring systems in lignin generating intermediate products which are often more easily oxidized than the initial ring system. Peroxyacetic acid in mildly alkaline media is a bleaching agent causing little material dissolution. (Johnson 1975) Nimz and Schwind (Nimz & Schwind 1981) have studied the oxidation of lignin and lignin model compounds with peroxyacetic acid. Both synthetic lignin and spruce lignin reacted smoothly at their alpha-carbonyl and olefinic groups. Dilignol structures showed a decreasing reactivity in the order phenylcoumaran, pinoresinol, beta-aryl ether structure. A method where cellulose material is treated with peroxyacetic acid and an alkaline reagent has been patented. In order to decrease the chemical consumption, the material was treated with alkaline reagent before the peroxyacetic acid treatment. (Kosaya et al. 1981) Birch and pine can be delignified using acetic acid and peroxyacetic acid in a two-stage process. In the first stage, chips were treated with acetic acid at atmospheric pressure in the presence of a catalyst, or at elevated pressure without any catalyst. Hydrochloric acid and sulphuric acid were used as catalysts. In the second stage, chips from the first stage were treated with a mixture of acetic acid and hydrogen peroxide. Citric acid was added to stabilize the peroxyformic acid formed through a reaction between acetic acid and hydrogen peroxide. The catalyzed method produced birch pulp with a kappa number of 2.4 and consumed hydrogen peroxide 2.4 % on wood. The corresponding pine pulp had a kappa number of 11.3 and 4.8 % on wood of hydrogen peroxide was consumed. When elevated pressure was used instead of a catalyst, only 1.2 % on wood of hydrogen peroxide was needed to produce birch pulp with a kappa number of 5.2. When producing pine pulp the hydrogen peroxide consumption was 3.5-4.8 and the kappa numbers were 13.1-7.7. The strength properties of the birch pulp produced at elevated pressure was near those of kraft pulps. (Kurvinen 1988) When formic acid and peroxyformic acid were replaced with acetic acid and peroxyacetic acid in the Milox process, a two-stage pulping produced better pulp than a three-stage process. The first stage could be carried out at atmospheric pressure using hydrochloric acid as a catalyst, or at high temperature and pressure without any catalyst. In the second stage, hydrogen peroxide is added to form peroxyacetic acid. Birch and pine pulps with low kappa numbers (2-11) and acceptable yields (43-46 %) were produced. The pulps were easily bleached to 90 % brightness with alkaline peroxide only. The unbleached birch pulps had high tensile indexes, which were significantly reduced during alkaline bleaching. (Poppius-Levlin et al. 1991c) Brolin and co-workers (Brolin et al. 1991) have studied the delignification of softwood kraft pulps with hydrogen peroxide, oxygen and ozone in acetic acid media. Oxygen delignification in acetic acid comprised both phenolic and non-phenolic structures but the selectivity of the method was limited. The addition of small amounts of hydrogen peroxide to the oxygen delignification in acetic acid improved the selectivity. The treatment of pulp with ozone in acetic acid was found to be more selective than a treatment with ozone in water. The selectivity was further improved by conventional alkaline oxygen treatment before ozonation. The peroxyacetic acid products used in early studies was an equilibrium product of hydrogen peroxide, acetic acid, peroxyacetic acid and water. The concentrations of acetic acid and hydrogen peroxide may be equal to or even higher than the concentration of peroxyacetic acid. The unreacted reagents increase the cost of using peroxyacetic acid and may damage the fibres. According to Hill and coworkers (Hill et al. 1992) these problems can be eliminated when distilled peroxyacetic acid is used. Peroxyacetic acid can also be used as a slime control agent instead of organic biocides in paper machines. (Rantakokko et al. 1994) The Chang group (Chang et al. 1994) has found that chlorine-containing bleaching agents can be replaced by a mixture of peroxymonosulphuric and peroxyacetic acids produced by mixing 50 % hydrogen peroxide, 93 % sulphuric acid and glacial acetic acid. A mixed peroxy acid bleaching stage can replace conventional stages in a bleaching sequence and improve the properties of the bleached pulp. The reactions of lignin model compounds with peroxyacetic acid and peroxymonosulphuric acid have been studied by Kawamoto (Kawamoto et al. 1994). According to Liebergott (Liebergott 1994) peroxyacids can replace ozone in bleaching sequences. The capital and operating costs of using peroxyacids are lower than those of using ozone. If the effluents from a conventional bleaching sequence are partially or totally recirculated, they have to be treated with significant amounts of sodium hydroxide and sulphuric acid. When a stage using distilled peroxyacetic acid is used as the first bleaching stage followed by a hydrogen peroxide stage, no sulphuric acid and significantly less sodium hydroxide is needed. (Basta et al. 1994) The Kawamoto group (Kawamoto et al. 1995) has studied the reactions of lignin and lignin model compounds with peroxyacetic acid and peroxymonosulphuric acid. Reactions of the model compounds with the more nucleophilic peroxyacetic acid produced more ring-opening products. The reaction with peroxymonosulphuric acid resulted in accumulation of hydroxylated products and p-quinone. A four-stage sequence for bleaching of aspen wood delignified with aqueous butanol has been presented. The first and third stages in the sequence are peroxyacetic acid stages, while the other stages use hydrogen peroxide. Pulp brightness from 82 to 86 % can be reached with low losses of pulp (5.5-6.3 %). The resulting bleached aspen pulp has good strength properties but is inferior to a corresponding kraft pulp. (Pazukhina et al. 1995) Lachenal and Delagoutte (Lachenal & Delagoutte 1996) have compared peroxymonosulphuric acid and peroxyacetic acid as bleaching chemicals. Examination of residual lignin in kraft pulp after treatment with the acids showed that only peroxymonosulphuric acid led to an increase in the number of free phenolic groups. Both acids generated new carbonyl groups. Peroxymonosulphuric acid was a more efficient delignification agent, but peroxyacetic acid had a more positive effect on brightness. The different chemical structure of the acids explain, according to the authors, these results. When oxygen-treated softwood kraft pulps are delignified with equilibrium peroxyacetic acid, the carbonyl group contents of the pulps increases to figures that are close to those of ozonated pulps. However, the viscosities of the pulps decreased only by 13-15 %. Similar results were obtained also when using peroxyformic acid as the delignifying agent. (Poppius-Levlin et al. 1996) When oxygen-treated pine kraft pulp is delignified with peroxyacetic acid, the main factors affecting the rate of the delignification reaction are peroxyacid charge, temperature, pH and the metal content of the pulp. (Jaaskelainen & Poppius-Levlin 1997) Ni and Entremont (Ni & Entremont 1997) have investigated the oxidation of lignin model compounds with peroxyacetic acid. They have concluded that the overall oxidation of the model compounds can be generally characterized by four reactions: formation of hydroquinone via oxidation of alpha-carbonyl compounds, oxidation of para- benzoquinone, demethylation and formation of other OCH3 containing compounds. The BOD7, COD and TOC loads and the toxity in effluents from a peroxyacetic acid bleaching stage are higher than those of effluents from an ozone stage when bleaching softwood kraft pulps. This is explained by the 2 % residual peroxyacetic acid content of the effluents. (Liukko & Poppius-Levlin 1997) Yuan and co-workers (Yuan et al. 1998a) have presented an improved peroxyacetic acid bleaching process. This improved stage has a final pH of around neutral when a conventional peroxyacetic acid stage has a final pH of around 5. This modification results in higher pulp brightnesses. A single peroxyacetic acid stage cannot produce a fully bleached pulp. The pulp brightness can be increased by applying sequentially a peroxyacetic acid and a hydrogen peroxide reinforced extraction stage. (Yuan et al. 1998) 20.6 % of wood components have been found to be transformed into volatile products during peroxyacetic acid delignification of fir. (Pazukhina & Ostrik 1998) When delignifying an acid-washed, oxygen-delignified, softwood kraft pulp with peroxyacetic acid, the Zhang group (Zhang et al. 1998) has found that Mn2+, Fe3+, Cu2+, V5+ and Co2+ catalyse wasteful decomposition of peroxyacetic acid when present at a concentration of 5 ppm on pulp. Water prehydrolysis of birch pulp has been found to improve delignification by acetic acid. The addition of 1...4 % hydrogen peroxide in the acetic acid liquor increases delignification but decreases the degree of polymerisation of the pulp. (Rutkowski et al. 1998) When studying bleaching with peroxyacetic acid Pazukhina and Teploukhova (Pazukhina & Teploukhova 1998) have found that the best results in terms of pulp whiteness are achieved by using hydrogen peroxide and chlorine dioxide treatments as the following bleaching stages after the peroxyacetic acid stage.

2.2.2.3. Acetosolv pulping

Nimz (Nimz & Casten 1985, Nimz et al. 1986a) has presented a pulping method where wood is continuously extracted for 2-5 hours at 383 K with 95 % acetic acid containing 0.1 % hydrochloric acid. At the end of the extraction, 3-5 % on wood of hydrogen peroxide is added and the mixture is stored for eight hours at 353 K. The hydrogen peroxide reacts with the acetic acid forming peroxyacetic acid. When beech is delignified, pulp with a kappa number of seven is obtained. Spruce pulps have kappa numbers below 30. The process is operated at atmospheric pressure and it is called Acetosolv pulping. Hemicelluloses and lignin are obtained from the Acetosolv process in a low condensed and relatively reactive form that can be converted into chemical raw materials. (Nimz & Casten 1986a) The Acetosolv method has been patented. According to the patent claims, wood and annual plants are cooked at 373-385 K with aqueous acetic acid containing 0.05-0.2 % hydrochloric acid in the beginning of the cook and 0.5-2 % hydrogen peroxide at the end. Pulp yields 40-60 % and kappa numbers 2-30 are obtained with pulping times from two to five hours. (Nimz & Casten 1986b) Acetosolv pulps can be bleached in two stages. Before the first stage the pulp (kappa number 59) is washed with hot 93 % acetic acid, which decreases the kappa number to 24. The first bleaching stage is a treatment with hydrogen peroxide followed by washing with hot acetic acid. After the first bleaching stage, the pulp has a kappa number of 3.8 and a brightness of 39.5 % ISO. The second stage is an ozone treatment, which decreases the kappa number to 0.27 and increases the brightness to 71.5 % ISO. (Nimz & Berg 1989, Nimz & Berg 1991) Acetic acid can be partly replaced in bleaching by an acetate. The same acetate (e.g. butyl acetate) can be used as an entrainer in azeotropic distillation which is used to recover acetic acid from the spent pulping and bleaching liquors. (Nimz et al. 1989a) Hardwood Acetosolv pulps have strength properties between sulphite and kraft pulps. Pulp bleaching can be carried out entirely with ozone (1.5-3 %) in acetic acid. The ozone can partly be replaced by nitrogen dioxide. (Nimz et al. 1989b) The Acetosolv process uses 93 % aqueous acetic acid containing 0.1-0.2 % hydrochloric acid as solvent. The pulping is carried out for 4-6 hours at 383 K. A rotating extractor is used as digester. The pulp is prebleached with ozone. The final bleaching is done with peroxyacetic acid. A pilot plant was started in early 1989. (Nimz et al. 1988, Nimz et al. 1989b) Ozone in acetic acid is a more selective bleaching agent than ozone in aqueous systems. This due to a higher solubility of ozone, reduced formation of OH radicals and higher solubility of lignin degradation products. (Nimz et al. 1989c) Curvelo et al. (Curvelo et al. 1990a, Curvelo et al. 1990c) have invented a method to analyze chloride ions in aqueous acetic acid. The method is based on the use of an electrode sensitive to chloride ions. The method is useful in analyzing Acetosolv pulping liquors, because too little hydrochloric acid (catalyst) leads to an incomplete rupture of the covalent bonds in lignin, and too much hydrochloric acid leads to increased recombination of the lignin fragments. Lignin degradation during Acetosolv pulping in the presence of various catalysts (HCl etc.) has been studied using dimeric lignin model compounds. Because lignin reacts as a donor of electrons during pulping, catalysts with a catalytic activity exceeding that of hydrochloric acid are recommended. (Savov 1990) Lignins recovered from Acetosolv pulping of sugar cane bagasse have been analyzed by high pressure size exclusion chromatography. The chromatograms were divided into three regions corresponding to high (0-20 %), middle (90-75 %) and low (4 %) molecular weight fractions. (Groote et al. 1991a) Groote and co-workers (Groote et al. 1991b) have analyzed lignins obtained by pulping sugar cane bagasse using the Acetosolv method. Results indicate a low recombination behaviour of the lignin during pulping and the high reactivity potential of the lignin. Using a sulfomethylation process the lignin yields a lignosulfonate presenting good quality as a dispersing agent. Basic variables affecting Acetosolv pulping have been studied. A soaking pretreatment of the chips in pulping liquor for 60 minutes increased the screened yield of pulp by 3 %, but did not affect the kappa number. The optimum pulping time for oakwood was 60 minutes cooking and 120 minutes extraction. The pulp produced had a yield of 41.4 % and a kappa number of 11. Bleachable grade pine pulp could not be produced. (Kim et al. 1991) Jakab and others (Jakab et al. 1991) have studied the thermal decomposition of kraft and acetosolv lignins using thermogravimetry/mass spectrometry (TG/MS) in the temperature range 303-1223 K in an inert atmosphere using 20 a K/min heating rate. During the slow thermal decomposition, cleavage of functional groups plays an important role and leads to the evolution of low molecular mass products and to the formation of char. Lignin obtained from sugarcane bagasse by Acetosolv pulping has been sulfomethylated. The lignin was reacted with sodium hydroxymethanesulfonate at pH 9 and 403 K to yield a product that had good potential as a dispersant. (Groote et al. 1992a) Groote and co-workers (Groote et al. 1992b) have pulped depithed sugar cane bagasse using the Acetosolv method. Based on their analysis, they suggested the following empirical formula for the lignin: C9H8.04O2.81(OCH3)0.92(OH)0.73phenolic. Bleaching of mongolian oak Acetosolv pulps with formic acid and conventional bleaching stages has been studied. When magnesium sulphate was added as a bleaching stabilizer, the pulp yield was increased. Sodium silicate improved the brightness. When peroxide bleaching was used, the pulp yield and strength properties were improved in comparison with conventional chlorine bleaching sequences. (Kim & Paik 1993) The Parajo group (Parajo et al. 1993) has optimized the operational conditions of the Acetosolv process when hydrochloric acid is used as a catalyst. The optimum conditions they determined were an acetic acid concentration of 88 %, a liquor to wood ratio of 8 and a catalyst concentration of 0.24 %. Under these conditions, solid residues containing 8 % lignin, 88.5 % polysaccharide and 84.7 % glucan compared to the original amounts were obtained at 45.4 % pulp yield. The economy of Acetosolv pulping of Eucalyptus globulus has been studied when hydrochloric acid is used as the catalyst. The profitability of the process depends on the prices of the byproducts. A positive profitability is obtained when the price of lignin is 0.48 DM/kg. (Parajo & Santos 1995) The Acetosolv process can be carried out in three stages. The cellulose containing raw material is initially with an aqueous acetic acid solution at an elevated temperature and under an elevated pressure. The resulting pulp is treated in a second stage with an aqueous acetic acid solution, with the addition of nitric acid and is then washed or extracted with water or with the pulping liquor. In a third stage, the pulp is treated with an ozone-containing gas. (Berg et al. 1995) Savov and co-workers (Savov et al. 1996) have compared the use of aluminium trichloride and hydrochloric acid as catalysts in Acetosolv pulping. Lignin decomposition was found to be greater when AlCl3 was used. The oxidation of sugar-cane bagasse Acetosolv lignin catalyzed by Co+2, Mn+2 and HBr produces an oxidized lignin up to 21 % more carbonyl groups. Kinetic studies have shown that lignin is oxidized by a two-step reaction. First, C-O linkages are cleaved and after two hours the formation of C=O linkages takes place. (Goncalves et al. 1997)

2.2.2.4. Acetocell pulping

During the operation of the Acetosolv pilot plant, severe material and corrosion problems occurred in the equipment. This was the reason why the original process was modified and called the Acetocell process. The main modification were (Gottlieb et al. 1992): - The acetic acid concentration in pulping liquor was decreased from 93 w-% to 80-90 w- %. - No catalyst was used. - The pulping temperature was raised from 383 K to 443-463 K. - The digester pressure was raised from atmospheric to 0.8-1.5 MPa. - The pulping time was reduced from 180-300 minutes to 120-180 minutes. - The carousel extractor was replaced by a conventional digester. Spruce chips were delignified by the Acetocell process to kappa numbers 15-20 at high yield. Treating the resulting pulp with ozone (1-2 % on pulp) and mild extraction, the kappa numbers could be reduced to 2-4. The completely chlorine-free bleaching was concluded with a hydrogen peroxide and/or peroxyacetic acid treatment. Final yields of bleached pulps were from 43 to 46 %. (Gottlieb et al. 1992, Nimz et al. 1992a, Nimz et al. 1992b) In 1993 Neumann and Balser (Neumann & Balser 1993) reported that the Acetocell process was being extended to technical scale. It has been demonstrated that annual plants may also be delignified using the process. The Acetocell pulps are also suitable for chemical production.

2.2.2.5. Formacell pulping

Nimz and Schone (Nimz & Schoene 1993, Nimz & Schoene 1994) have invented a process where lignocellulosic material is delignified under pressure with a mixture of acetic acid (50-95 w-%), formic acid (< 40 w-%) and water (< 50 w-%). The pulping temperature is between 403 K and 463 K. Liquor-to-wood ratios from 1:1 to 12:1 can be used. Continuous extraction is achieved by connecting 2-20 digesters in line. The extraction is carried out counter currently and the pulp is prewashed with fresh pulping liquor. Pulps with very low residual lignin contents are produced and they can be bleached to full brightness using ozone and peroxyacetic acid. Azeotropic distillation with butyl acetate is used to separate water from the acids. The method is called Formacell pulping. In addition to the reduction of kappa numbers, the pulp strength properties are improved, when 5-10 % of formic acid is added to aqueous acetic acid. Formacell pulps have lower tear strengths than kraft pulps, but this is in most cases compensated for by the higher values of burst and tensile strengths. Formacell pulps produced from annual plants have better strength properties than corresponding soda pulps. Spruce pulp is due to its low kappa number (3.6) easily bleached in two stages with ozone in acetic acid and with peroxyacetic acid in butyl acetate. In the case of Formacell pulps produced from hardwoods or annual plants, only two peroxyacetic acid stages are needed to bleach the pulps. (Nimz & Schoene 1993) Low kappa numbers of around 10 for beech and pine and around 2 for aspen can be obtained using the Formacell process. The losses of formic acid under the pulping conditions were determined to be less than 1.5 % at 433 K, and less than 2.8 % at 453 K. (Saake et al. 1995a) Low pulping temperatures and high acetic acid concentrations should be used in the Formacell process in order to preserve hemicelluloses for paper grade pulps. The use of higher temperatures and water concentrations in the pulping liquor results in dissolving pulps with hemicellulose contents below 3 %. (Saake et al. 1995b) Aspen Formacell pulp can be bleached to a brightness of 86 % in a purely organic solvent system. The bleaching steps used are ozone, peroxyacetic acid, and butyl acetate. (Saake et al. 1998)

2.2.2.6. Summary

Young (Young 1998) has reviewed acetic acid-based pulping processes. When comparing the economy of various processes he states the following: - The separation of acetic acid and water is the most expensive unit operation in the process. - For dilute acid systems, chemicals should be recovered by extraction with ethyl acetate. - For more concentrated acid systems distillation is the preferred recovery method. Acetic acid is assumed to both promote solvation of lignin fragments and to reduce swelling of pulp fibres. The addition of oxygen has been found to improve diffusion of the reagents into wood chips. Technical problems related to acetic acid pulping are corrosion of the equipment and a relatively expensive solvent recovery. The vapour-liquid equilibrium of acetic acid and water is unfavourable and leads to high numbers of equilibrium stages in distillation.

2.2.3. Other organic acids

Chloroacetic acid can be used to produce pulp at atmospheric pressure. Pinewood was quickly delignified and the pulp had a good viscosity. (Schutz 1940) The lignin recovered from the spent pulping liquor after delignification with aqueous chloroacetic acid is pure and has a high methoxyl content. The yield of sugars is twice as high as in sulphite pulping. (Schutz 1941b) Likon and Perdih (Likon & Perdih 1995) have tested the use of chloroacetic acid, iodoacetic acid, difluoroacetic acid, dichloroacetic acid, dibromoacetic acid, trifluoroacetic acid, trichloroacetic acid and tribromoacetic acid in the delignification of spruce wood meal. Those haloacetic acids stronger than acetic acid, especially dichloroacetic acid and trichloroacetic acid, are efficient delignifying agents at temperatures below 373 K even without the presence of water. Zule (Zule et al. 1995) has studied lignin isolated from the spent liquors of spruce pulping with trichloroacetic acid. The C9 formula of the lignin was calculated to be C9H6.05O2.56(OCH3)0.88(OHf)0.68(TCA)0.29. The average molecular weight of the lignin was determined to be vary in the range 2000-3000. Thioacetic acid has been used to totally dissolve lignin from spruce wood and to study lignin reactions using model compounds (Nimz 1969a, Nimz 1969b). Thioglycolic acid has been used to isolate lignin from lignocellulosic materials. The methoxyl content of softwood and hardwood lignins obtained by this mean is 14-17 % and 20-23 %, respectively. Also cellulose can be isolated, although it cannot be quantified. (Anon. 1935) Holmberg (Holmberg 1936) has isolated lignin from spruce using thioglycolic acid, thiolactic acid and several other thioacids. Based on the lignin analyses he has suggested the formula C9H8.90O2.85(OCH3)0.92 for thioacid lignins. Holmberg (Holmberg 1947) has isolated lignin from aspen with thioglycolic acid and thioacrylic acid. Using the results of the lignin analyses, he has drawn the conclusion that aspenwood contains several lignin fractions. The fractions with high methoxyl contents are easily dissolved with thioacids.

2.2.4. Salts of organic acids

Robert (Robert 1955) has presented a pulping method using sodium benzoate as the delignification agent. The pulping liquor was an aqueous mixture of sodium benzoate (40 %). When the pulping was carried out at a liquor to wood ratio 7 and 393 K for 10.5 hours, the yield of the resulting unbleached pulp was 43 %. Conventional bleaching stages were used and the strength properties of the bleached pulps were reported to be good. Woody material can be converted into chemicals by treating it under heat and pressure with a substantially neutral aqueous catalytic solution of an alkali metal salt of cymenesulfonic acid, toluenesulfonic acid, benzoic acid or salicylic acid. The products of this treatment are furfural, acetic acid, carbon dioxide, lignin and cellulose. (McKee 1961) 2.2.5. Summary

Formic acid, acetic acid and corresponding peroxy acids are the only organic acids that have been intensively studied as pulping solvents. Atmospheric pulping is possible when using organic acids as solvents. This makes the operation of a plant safer. Operating costs of pulping are reduced by the fact that catalysts are not necessarily needed. On the other hand, the relatively complicated solvent recovery increases the costs. Investment costs are probably relatively high due to the need for corrosion resistant materials.

2.3. Phenol and cresols

2.3.1. Phenol

Buhler (Buhler 1897) has in the late 19th century patented a method where lignocellulosic materials were treated with phenol at temperatures above 423 K. The cellulose was not attacked and it had a quality suitable for paper production. A pulping method where wood or other lignocellulosic material is delignified with phenol has also been patented. The solvent can be concentrated or diluted with water, alcohols, benzene, benzene derivates or carbohydrates. The mixture is heated in the presence of catalytic amounts (0.01 %) of hydrochloric acid. (Hartmuth 1920) Pulping can be carried out at 375 K with phenol diluted to 35-40 % with alcohol or water. The water addition decreases though the pulp quality. Good results have been achieved by cooking wood chips in concentrated phenol at 353-373 K in the presence of 0.1-1.0 sulphuric acid for 4-5 hours. The yield of pulp was about 45 %. Significant amounts of phenol were bound to the dissolved material from wood. (Legeler 1923) Kalb and Schoeller (Kalb & Schoeller 1923) have studied the possibility of determining the amount of cellulose in wood by phenol delignification. They found that delignification with phenol, in the presence of small amounts of mineral acids, is rapid and can be carried out at temperatures below 373 K. During the delignification, the polysaccharides are partly hydrolysed. This makes it impossible to use phenol delignification as the basis for a method determining the total carbohydrate content. Phenol lignins have been reported to be amorphous and dark coloured. They dissolve easily, for example, in glacial acetic acid, methanol, ethanol, amyl alcohol and acetone. They do not dissolve, for example, in water, chloroform, aqueous ammonia and dilute mineral acids. The phenol lignin contains 71.6 % carbon, 5.3 % hydrogen, 23.1 % oxygen and 12.1 % methoxyl. (Hillmer 1925) Wedekind and co-workers (Wedekind et al. 1931) have proposed the formula [(C8H7O2)(OCH3)(OH)(C6H4OH)]x for phenol lignin. Pulping of spruce meal with phenol, using hydrochloric acid as a catalyst, has given two chemically different lignin fractions. The ratio of ether-insoluble and ether-dioxane- soluble lignins was about 3 to 1. When compared with the formula of native lignin (Brauns & Hibbert 1933), the ether-insoluble phenol lignin contained three new free phenolic hydroxyl groups, while one phenol group had reacted with one hydroxyl group in the lignin, forming a phenyl-oxygen ether linkage. The analysis of the ether-dioxane- soluble phenol lignin indicated that a significantly larger quantity of phenol had condensed with the native lignin unit than in the case of ether-insoluble lignin. (Buckland et al. 1935) Fuchs (Fuchs 1936) has found that phenol lignin and methoxy glycol lignin are closely related substances that can be separated into several fractions. Phenol can be cleaved from phenol lignin using a method consisting of a treatment with hydriodic acid and acetic acid in the presence of phosphor. It has been shown that in condensations of phenol with lignin, the phenol reacts in the ortho position. Also the para position undergoes the same reaction, which has been proved by the isolation of p-hydroxybenzoic acid. (Kratzl et al. 1962) The Ploetz group (Ploetz et al. 1963) has patented a method where lignin is hydrolysed with hydrogen in the presence of aromatic compounds in two stages. The first stage is carried out at 703 K, and the partial pressure of hydrogen is 42.5 MPa. The second stage is carried out in the presence of lignin degradation products from the first stage containing about 38 % phenols. The temperature of the second stage is 753 K, and the partial pressure of hydrogen is 42.5 MPa. The product contains 21 w-% phenols, 36 w-% oils, 26 w-% gases and 18 w-% water. When spruce wood is treated with phenol, mainly lignin is reacting with phenol. Condensation reactions between phenol and the polymeric wood carbohydrates are not assumed to occur. Phenol has been found to condense with lignin. (Kratzl et al. 1964) Schweers and Pereira (Schweers & Pereira 1972) have studied the penetration of different wood species by phenol. Beech and spruce chips were well penetrated at atmospheric pressure, but pine chips needed elevated pressure for the penetration. Lignins dissolved during phenol pulping can be used to produce a phenolic blend by pyrolysis. This blend can be used as the pulping liquor. The phenol-lignin can also be reduced to polyoles, which can be transformed into polyurethanes. (Schweers & Rechy 1972) Phenol remaining in the pulp after delignification can be recovered by washing with methanol. When the methanol is recovered by distillation, only negligible phenol losses occur. The main source for phenol losses is the phenol recovery by distillation. (Schweers et al. 1972) Pine and beech woods can be delignified with phenol or phenol mixtures in the presence of catalytic amounts of hydrochloric acid. An atmospheric pulping produced pulps with yields of about 50 % and satisfactory mechanical properties. The pulp were bleachable with conventional bleaching sequences. (Schweers & Rechy 1973) Pine wood has been delignified with aqueous phenol (90 %) containing 0.05 % hydrochloric acid as catalyst. The cooking was carried out in three stages (383 K, 413 K, 443 K) using a liquor-to-wood ratio of 4:1. After cooking, the pulp was washed with methanol or isopropanol. The pulping liquor could be reused at least twice before it became necessary to separate the phenol lignin by distilling. (Schweers 1974) According to Vorher and Schweers (Vorher & Schweers 1975) phenol lignin can be converted to mixtures of phenol and lower methyl and methoxyl substituted phenols using pyrolysis. The yield of simple composition phenols is approximately 30 % which, together with the composition of the product, makes it possible to use it as a pulping chemical. Aqueous phenol (1:1) can reduce the lignin content of pine wood meal from 30 w-% to 3 w-%, when the material is treated with the solvent for two hours at 478 K in a stirred batch reactor. Aqueous n-butanol delignification was found to be less effective. The phenol recovery rate was shown to be 70-78 %. (April et al. 1979) According to Lipinsky (Lipinsky 1983), the temperature needed in phenol pulping is well below that which causes furfural formation. Johansson (Johansson et al. 1983) has presented a phenol pulping method where the lignocellulosic material is extracted leaving the cellulosic fraction untouched. The method is called the Battelle-Geneva process. Pulping is carried out at 373 K and atmospheric pressure. The pulp yield is around 40 %. According to Millner (Millner 1983) the Battelle-Geneva process has the following advantages: - several raw materials can be used - the pulp yield is high - the pulp quality can be adjusted - low temperatures and pressures can be used - hydrolysis of hemicelluloses and extraction of lignin is carried out in one process stage - the process produces phenol - the process is environmentally friendly - the investment costs are relatively low - the process can be economically used, even at low production rates. Vanharanta and Kauppila (Vanharanta & Kauppila 1984) have studied the feasibility of the Battelle-Geneva process. The pulping is carried out at 373 K and atmospheric pressure using aqueous phenol (40 %) as the solvent. When the spent pulping liquor is allowed to cool, it separates into two phases. The organic lignin containing phase separates from the aqueous hemicellulose containing phase. In bench-scale batch digestion, hydrochloric acid (1.85 % in water) was used as catalyst. After cooking for four hours, birch pulps with a yield of around 44 % and kappa numbers around 11 were obtained. The pulp was washed first with hot water to achieve high lignin and hemicellulose yields. To remove the residual phenol in pulp, it was washed in the second stage with dilute sodium hydroxide. The phenol present in the aqueous phase of the spent pulping liquor can be recovered by extraction with toluene. Phenol can be recovered from the organic phase by vacuum distillation. An analysis showed that good profitability could be achieved with this phenol pulping method. Both softwoods and hardwoods have been delignified with aqueous n-propyl alcohol (50 %) containing 45 % phenol and 67 % acetic acid. This method, called phenorganosolv pulping, easily separates woods into their main components. (Sano & Sasaya 1985) Tournier’s group (Tournier et al. 1986) has patented a pulping method where lignocellulosic material is treated in an acidic aqueous solution in the presence of phenol compounds. The solution is acidified by adding 0.5-4.0 w-% of hydrochloric acid, sulphuric acid, phosphoric acid or a strong organic acid. The phenol content of the pulping liquor is at least 40 w-%. The cooking is carried out at nearly atmospheric pressure and at the boiling point of the solution using a liquor-to-wood ratio 2:1-4:1. After pulping, the spent liquor is separated into two liquid phases. The phenol containing organic phase is distilled to recover phenol. The distillation residue is treated by pyrolysis to give phenol compounds in amounts so large that no external solvent make-up is needed. Acetone-soluble phenol pulping lignin from spruce has been examined by N.M.R. spectroscopy. The result have provided evidence that the beta-0-4-aryl ether linkage has erythro stereochemistry. (Hawkes et al. 1986) According to Laxén (Laxén 1987) lignin will precipitate both inside the fibres and on the surfaces of the fibres, if the pulp is washed directly with water after phenol pulping. This lignin will, however, dissolve in caustic or acetone. An alkaline washing stage can lower the kappa number of the pulp significantly, but the application of such a stage is expensive and makes it difficult to recover lignin as a by-product. The results of a laboratory pulping and washing simulation showed that after direct counter-current washing with water, 88 % of the lignin remained in pulp. If the spent pulping liquor was before washing displaced with fresh solvent, the remaining amount of lignin was 20 %. Ekman (Ekman 1988a, Ekman 1991a) has invented a method of separating lignin and phenol from the spent phenol pulping liquor. When the organic lignin and phenol containing phase of the spent liquor is extracted with hot water, two liquid phases are formed: a phenol containing aqueous phase and a lignin containing organic phase. When the aqueous phase is cooled, two liquid phases are formed: a phenol phase with a low water content and an aqueous phase with a low lignin content. The aqueous phase is recycled to the hot water extraction and the phenol phase is used in preparing the pulping liquor. Lignin is precipitated with water from the lignin containing organic phase. Ekman (Ekman 1988b, Ekman 1991b) has also proposed the method for recycling the solvent in phenol pulping. The phenol containing solution is first used as washing liquor for the pulp coming from digestion. Then the solution is used as pulping liquor and finally it is used in the lignocellulosic material soaking stage before pulping. In a Japanese patent (Shiraishi et al. 1988), a method is described where lignocellulosic material is immersed in aqueous chlorine (0.01-0.5 w-%). After washing with water, the material is cooked in aqueous phenol or cresol (1-50 w-%) at 373-423 K for 30-90 minutes. Pulps without cellulose degradation are produced in yields of 50-80 %. Araki et al. (Araki et al. 1989) have compared phenol pulping with other organosolv methods. Alkaline systems were found to be superior to acidic systems in delignification selectivity. The pulp qualities were similar in most cases, but pulps produced with acidic solvents had lower tear indexes. Spent pulping liquor from phenol pulping can be used as a lignin-containing raw material for producing phenols. The phenols obtained can be used as solvent in pulping. The lignin-containing material is mixed with a double ring aromatic hydrocarbon, and the mixture is decomposed at a temperature of 533-673 K, under pressure of 0.5-9.8 MPa, for 5-180 minutes. (Kakemoto et al. 1990) Vega and Bao (Vega & Bao 1993) have used phenol and dilute hydrochloric acid to delignify an annual plant (Ulex europeaus) at atmospheric pressure and 373 K. They examined the kinetics of the pre-hydrolysis and delignification processes. The hydrolysis of cellulose can be satisfactorily modelled with a simple kinetic model proposed by Saeman (Saeman 1945). The delignification process can be explained using a simple model of two parallel pseudo first-order reactions. The influence of reaction time, phenol concentration, and hydrochloric acid charge on pulp yield, total polysaccharide content, residual lignin in pulps, dissolved reducing sugars, and pulp yield versus residual lignin has been studied in the Battelle-Geneva process when pulping elephant grass. The experimental results were fitted to second- order linear models. The model specified maximum selectivity for a reaction time of 108 minutes, a phenol concentration of 53.2 %, and a hydrochloric acid charge of 0.0394 g/g elephant grass. (Vega et al. 1997a) The hydrochloric acid charge was found to be the most influential variable, followed by reaction time and phenol concentration. (Vega et al. 1997b) Jimenez and co-workers (Jimenez et al. 1997b) have developed a second-order polynomial model for phenol pulping of wheat straw using three independent process variables. According to the model, high yield and waste water pH can be ensured by setting the variables to low values. If high holocellulose and alpha-cellulose contents and low lignin contents are desired, then a long cooking time (120 min) at high temperature (473 K) and intermediate phenol concentration (65 %) has to be used.

2.3.2. Cresols

Birch and spruce have been delignified with aqueous cresols at 443-463 K for 2-3 hours. Birch pulp with strength properties near those of kraft pulp was obtained with a cresols- water ratio of 4:1. A solvent mixture composed from lignin degradation products gave similar results. Birch was more difficult to delignify and the pulps showed lower strength properties. However the addition of acetic acid improved the pulp quality. In general, the pulp yield was slightly higher than that of corresponding kraft pulps, but the residual lignin content was also higher. The most advantageous feature of this method is that the solvent can be produced by pyrolysis or hydrocracking of the lignin dissolved during pulping. After pulping, the solvent mixture was separated into two phases. The organic phase contained lignin and the aqueous phase contained sugars. (Sakakibara et al. 1983, Sakakibara et al. 1984) The addition of 10-50 w-% acetic acid to the aqueous cresol pulping liquor has been found to improve the selectivity of delignification and the pulp strength properties in spruce pulping. When birch was pulped, the addition of acetic acid had no effect. (Sakakibara & Edashige 1984) Takeyama and Sasaya (Takeyama & Sasaya 1988) have hydrocracked beech lignin separated in cresol pulping. Up to 20 % of the hydrocracked lignin could be separated by steam distillation. The volatile fraction contained phenol, o-, m- and p-cresols, o- and p- ethylphenols, 2,4-xylenol and p-propylphenol. According to the Sano group (Sano et al. 1988), beech and birch are easily delignified by treating with a cresol-water-acetic acid mixture for 2-3 hours at 443 K. The cresol to water ratio of the solvent mixture was 7:3 and the acetic acid concentration was 10 %. Kachi and Tsuchiya (Kachi & Tsuchiya 1988, Kachi & Tsuchiya 1989) have subjected wood chips to pulping with an aqueous solution of phenols from lignin degradation products. Both unbleached hardwood and softwood pulps were easily defibered even at high kappa numbers (30-60) and the yields were from 50 to 60 %. The pulps were washed with solvent and hot water. After oxygen delignification and a conventional three-stage bleaching, the pulps were bright and strong enough to be used for the production of printing papers. Wood can be separated into carbohydrates and lignin within 10-20 minutes with a concentrated sulphuric acid-cresol system. The separation includes the swelling of carbohydrates and the solvation of lignin. During the treatment, lignin is protected from the attack of sulphuric acid through solvating with cresol. The technique gives perfect separation of any lignocellulosic material because the tissue structures are destroyed by the swelling of carbohydrates. (Funaoka & Abe 1989) Miyazaki (Miyazaki et al. 1989) has presented a pulping method using a mixture of cresol and water (4:1) as solvent. Because the selectivity of the method was not good enough for hardwoods, and the pulping time needed for softwood was twice the time applied in conventional pulping, the effect of some additives was examined. Tertiary amines were found to effectively remove the acetic acid from hardwoods. Lewis acids combined with tertiary amines were found to be efficient catalysts in softwood pulping. A method where softwoods are delignified with aqueous cresols (70 %) containing acetic acid has been presented. The pulping was carried out at 453 K for 120 minutes. When the acetic acid addition was 40 % on wood, good delignification was achieved. The cresols were reused several times before recovery by distillation. (Sano 1989) Polyurethane has been prepared from lignin obtained by delignifying wood with aqueous cresols. Lignin in polyurethane retarded the thermal degradation of the polymer in the air. Polyurethane containing polyol and phosphorus were also produced and they were found to be nonflammable in air. (Hirose et al. 1989) Cresol delignification is affected by the solvent-water ratio and the pulping temperature. A high solvent ratio or a low temperature increase the pulp yield and improves its properties. The pulping time can be decreased by the addition of some acetic acid or by circulating the liquor. The yields of hardwood pulps are equal to those of kraft pulps, and the yields of softwood pulps are even higher than the yields of corresponding kraft pulps. The strength properties of cresol pulps are inferior to those of kraft pulps. (Takagi et al. 1989) Takagi and Kachi (Takagi & Kachi 1989) have studied the bleaching of cresol pulps washed with methanol. When conventional chlorine bleaching was used, cresol pulps were easier to bleach than kraft pulps. Beech cresol pulps and kraft pulps had equal yields, but pine cresol pulps had 15-20 % higher yields. Strength properties of bleached cresol pulps were somewhat inferior to those of kraft pulps. The cresol remaining in bleached pulps was less than 1 ppm. The sugar content of wood meals cooked with cresol and water layers of the spent pulping liquors has been studied. The sugar found in the meals was mainly glucose. When formic acid was added as a catalyst, the glucose content increased. The water layer contained mostly oligosaccharides, but the amount of monosaccharides increased with increasing cooking temperature and time. (Chang & Lee 1989) Nishiyama’s group (Nishiyama et al. 1989) has studied the recovery of cresols from the aqueous phase of an cresol pulping process. They have found that methyl isobutyl ketone, butyl acetate, i-amyl acetate and benzene can be used as a solvent in liquid-liquid extraction of cresols from water. Fir, spruce and larch chips have been delignified in aqueous cresols (70 %) containing various amounts of acetic acid for two hours at 453 K. Good results were obtained by adding 40 % on wood of acetic acid. Spruce and larch yielded pulps with lower kappa numbers than fir. Spruce and fir yielded pulps with satisfactory strength properties. (Sano et al. 1989b) Kachi et al. (Kachi et al. 1990) have presented a solvolysis pulping method using a solvent containing cresols and phenols produced by pyrolysis or hydrogenolysis of lignin and about 20 % water. Hardwood and softwood pulps produced by this method contain about 5 % lignin. The pulps can be almost completely defiberized even at relatively high lignin contents, which indicates the solvents’ effectiveness in swelling wood. The kinetics of delignification and carbohydrate dissolution in a cresol-water system have been studied. The results have indicated that, during delignification, the diffusion of the cooking liquor and the surface chemical reaction were the rate-determining steps in the early cooking period. The opposite was observed for carbohydrate dissolution, where the diffusion of the reaction products was the rate-determining step in the early cooking period. (Jeong & Park 1990) Nakamura and co-workers (Nakamura et al. 1991) have prepared polyurethanes from lignins obtained as a by-product in aqueous cresol pulping of Japanese beech. Polyurethanes prepared from only cresol lignin were hard and brittle. The adding of polyethylene glycol into the polyurethane system allows the preparation of different types of polyurethanes. Softwood pulps produced by organosolv pulping with aqueous cresols have been treated with sodium chlorite. The crystallinity, degree of polymerization and alpha- cellulose content were measured. An increase in the crystallinity was observed with increasing delignification. A further delignification though caused a slight destruction of the crystalline regions. The degree of polymerization and the alpha-cellulose content of the pulps decreased during delignification with sodium chlorite. (Uraki et al. 1993)

2.3.3. Summary

Phenol and cresols are chemicals which can be produced using wood as raw material. This makes on site production of the pulping chemicals possible and reduces the costs of pulp manufacturing. The costs of solvent recovery are probably relatively high because vacuum distillation has to be used. 2.4. Ethyl acetate

According to the Young group (Young et al. 1985), water can be partly replaced by ethyl acetate in acetic acid pulping. The addition of ethyl acetate has the following positive implications: 1. The amount of water required in the system is reduced. 2. The cooking time is reduced. 3. The pulping temperature for a given pressure is reduced. 4. Pulp strength is improved. 5. Pulp yield is increased. 6. The solubility of lignin in the solvent is enhanced. 7. The solvent recovery is simplified. Ethyl acetate is totally miscible with acetic acid and water. By proper adjustment of the solvent ratios, the system is, however, divided into two liquid phases. The ethyl acetate-acetic acid phase and the aqueous phase can be separated by decantation, which reduces the energy consumption of the solvent recovery. (Young et al. 1985, Anon. 1985, Smith 1987, Haggin 1995) Since ethyl acetate has good lignin solvent properties, it can be used to replace acetic acid as well as water. A wide variety of pulps can be produced. The pulp grade is determined by the ethyl acetate-acetic acid-water ratio in the solvent, the pulping temperature (433-473 K) and the pulping time (15-120 min). When aspen pulp produced by the ethyl acetate method is compared with a typical sulphite pulp, it can be noted that at equivalent kappa numbers the yield of the ester pulp is about 20 % higher. The strength properties are roughly equivalent, but the ester pulp has a 20 % higher breaking length. (Young & Baierl 1985, Young et al. 1987a) The spent pulping liquor is flashed to atmospheric pressure. When the liquor is cooled during flashing, it is divided into two liquid phases, if the acetic acid concentration is less than 22 w-%. The organic phase contains most of the dissolved lignin. The aqueous phase contains suspended lignin particles. The organic phase is reused as pulping liquor. The aqueous phase is further treated to separate lignin and traces of the organic solvents. The pulping liquor contains 5-95 w-% ethyl acetate, 5-80 w-% acetic acid and 10-60 % water. The liquor-to-wood ratio can be varied from 1:1 to 20:1. (Baierl et al. 1986) Generally, the strength properties of ester pulps are between kraft and sulphite pulps. The use of phase separation in recovery should result in lower investment costs than in conventional pulping. (Young et al. 1987c) According to Aziz and McDonough (Aziz & McDonough 1987), other hardwood species do not respond to ester pulping as well as aspen, and the process is unlikely to be capable of producing market-grade pulps from softwoods. The recovery section of the ester pulping process includes liquid-liquid phase separators, a filter to recover the lignin, distillation columns to recover volatile organic compounds, an extraction column for acetic acid separation, and a burning system to handle sugars etc. (Young et al. 1987b) Coppen (Coppen et al. 1988) has tested the ester pulping process using eucalyptus as raw material. He concludes that the claims for ester pulping cannot be extended to materials other than aspen. The use of the acetic acid-ethyl acetate-water solvent system for pulping wheat straw has been studied. Cooking conditions were: maximum pressure 745 kPa, temperature 429 K, pulping time 340 minutes, and pulp yield 50 %. The pulp produced by this method may be suitable as furnish for writing and printing papers. (Jiwei et al. 1988) According to Young (Young 1989), stronger reaction conditions are required in pulping softwoods. Pulping temperatures of 463-473 K and times of 60-120 minutes are required. The ethyl acetate-acetic acid-water ratio of 33:33:33 has been used. By proper selection of the pulping conditions, pine pulps of good quality have been obtained. Young (Young 1992b) has reported that the optimum ethyl acetate-acetic acid-water ratio is 15:70:15.

2.5. Amines and amine oxides

2.5.1. Ethylenediamine

Ethylene diamine can be used together with cadmium oxide to predict the behaviour of cellulose in the viscose process. The solvent has shown promise for rapid characterization of viscose filterability potential and end product colour. (Smith et al. 1963) Softwoods and hardwood can be delignified at rates faster than in the kraft process in a soda liquor containing 40 % ethylenediamine. Compared to kraft pulps, the softwood soda-ethylenediamine pulps showed lower bleached yields, higher tear strengths, lower bleach chemical consumptions and lower shrinkage during bleaching. For hardwood pulps, no tear strength improvements were observed. (Kubes et al. 1978) Aqueous ethylene diamine, alone or together with sodium hydroxide, can be used to delignify wood. Kappa numbers 17.8 and 24.5 can be reached for aspen and pine, respectively. The corresponding pulp yields are 56.0 and 47.2 %. The pulp strength properties are equivalent to those of corresponding kraft pulps. (Julien & Sun 1979) MacLeod’s group (MacLeod et al. 1981) has found that the addition of small amounts of ethylenediamine (EDA) to soda-anthraquinone (AQ) pulping liquor results in bleachable grade black spruce pulps with better tear strengths than soda-anthraquinone pulps. The soda-AQ-EDA pulps could be bleached to full brightness with less chlorine than the soda-AQ pulps. The higher the EDA charge (0.1-10 % on wood) was, the better the tear strength of the pulps. The addition of methanol and ethylenediamine or monoethanolamine to soda pulping liquor increases the rate and selectivity of delignification, with increasing concentration of the organic components. Pulps obtained at low amine charges have tear values superior to kraft pulps. At high amine levels, the alkali requirement is reduced, but the cellulose viscosity and pulp properties deteriorate. (Green & Sanyer 1982) The delignification of lignocellulosic material can be carried out using an aqueous solution of ethylenediamine and ethanol in combination with catalytic amounts of anthraquinone. The method is highly selective and can be used to delignify both hardwoods and softwoods. Softwoods can be delignified to kappa numbers as low as 20. (April et al. 1983) Amines and alcohols alone cannot delignify softwoods to very low lignin contents. Addition of ammonium sulphide to ethanol or ethylenediamine gives the desired extent of delignification for softwoods. Very high selectivity is observed in delignification. The strength properties of ethylenediamine-sulphide pulps are superior compared to ethanol- sulphide pulps. The best cooking temperature has been found to be 453 K. No significant effect on selectivity has been observed when the liquor-to-wood ratio is changed from 10:1 to 1.62:1. The optimum ammonium sulphide concentration in pulping liquor is 0.3 M. (April et al. 1983) Ethylenediamine can be removed from spent pulping liquor solids by sodium hydroxide distillation with only 0.77 % ethylenediamine remaining in the solids (April et al. 1983) Abbot and Bolker (Abbot & Bolker 1984) have studied the kinetics of soda- ethylenediamine pulping. Studies were also conducted with primary and tertiary amines. Their results suggest that primary and secondary amines play a similar role to that of hydrosulphide ion in kraft pulping. Zargarian et al. (Zargarian et al. 1988) have studied aqueous ethylenediamine-ethanol- anthraquinone (EDA-EtOH-AQ) pulping of pine. Yields at the same kappa numbers were higher than those of conventional pulping processes. In order to achieve good pulp quality and a low kappa number the EDA concentration in the pulping liquor had to be from 40 to 50 %. The liquor contained also 10 % ethanol and 1.5 % anthraquinone. With a 99 % recovery of the organic chemicals, the chemical costs of the process were estimated to be significantly higher than the costs of kraft pulping.

2.5.2. Hexamethylene diamine

High pulp yields, up to 68 %, have been achieved when pulping aspen with hexamethylene diamine. The addition of sodium hydroxide to the amine-water system enhances the rate of delignification but decreases the yields. Breaking lengths of the amine-sodium hydroxide pulps are generally as high as those of softwood kraft pulp, but the tear factors are lower. (Julien & Sun 1979) Hexamethylene diamine alias 1,6-hexanediamine alias 1,6-diaminohexane (HMDA) has been used to delignify aspen and pine at 473 K for 15 minutes. HMDA has been found to selectively remove lignin from the wood chips. The HMDA aspen pulps are even stronger than kraft aspen pulps, whereas the pine pulps are weaker. Both pulps are difficult to bleach to high brightness levels with conventional bleaching stages. (Julien 1979) Julien and Barna (Julien & Barna 1980) have studied the economy of an HMDA pulping plant, and compared it to a kraft mill. In the amine aspen process, the solvent is recovered by distillation. Spent pulping liquor is fed directly to a distillation column. The pulp is washed with water and the spent washing liquor is concentrated in an evaporator plant before feeding to the distillation column. Lignin is recovered from the distillation bottom product by spray drying. Potential advantages of the amine process include decreased investment costs and improved yield. The operating costs are however higher than in kraft pulping. Hou and An (Hou & An 1989) have pulped wheat straw for 2 hours at 423 K with HMDA (200 % on straw) using a liquor ratio of 1:8. Delignification was found to be a third-order reaction with an efficiency of 90.7 % at the defibration point with an activation energy of 66.51 kJ/mol.

2.5.3. Methylamine

Beer and Peter (Beer & Peter 1985) have digested spruce wood at supercritical conditions (423-453 K) with aqueous methylamine. A very soluble product was extracted. The digestion reaction was found to be of the first order. In a later study, Beer and Peter (Beer & Peter 1986) have used aqueous methylamine to separate lignin from pine chips by supercritical extraction. The extraction was carried out at 453 K and 11.0 MPa using a solvent containing 40 w-% methylamine and 60 w- % water. Lignin was recovered from the extract by isobaric cooling. The condensate was an aqueous solution containing 70-80 w-% lignin. Li and Kiran (Li & Kiran 1988) have compared several solvents for supercritical extraction of wood. The extractions in carbon dioxide, ethylene, nitrous oxide and n- butane were not substantial, and the interactions with wood components appeared to be nonreactive. The dissolutions in ammonia and methylamine were found to be results of reactive interactions. The results of extraction with binary solutions were strongly dependent on the compositions of the solvents. Methylamine was found to be more reactive with lignin, while ammonia and carbon dioxide-water mixtures were more reactive with carbohydrates. Spruce has been delignified by supercritical methylamine and by a mixture of methylamine and nitrous oxide at temperatures of 323-470 K and pressures of 0.4-27.5 MPa. The extent of lignin removal was found to increase with extraction time, temperature, pressure and the methylamine content of the extraction fluid. The extracted lignins precipitated as solvent-free solids. (Li & Kiran 1989a, Li & Kiran 1989b)

2.5.4. Ethanolamine

Wallis (Wallis 1976) has treated lignin model compounds with ethanolamine either at 473 K or in benzene solutions at 353 K. He found three types of condensation reactions of ethanolamine with lignin: - replacement of one hydroxyl group in cathecols at 473 K to give n-substituted aryl groups, - condensation with compounds containing carbonyl groups in benzene at 353 K to give oxazolidines or imines, and - condensation with compounds containing p-hydroxybenzyl alcohol moieties in benzene at 353 K. Wallis (Wallis 1978) has delignified eucalyptus and pine with monoethanolamine (MEA), diethanolamine (DEA) and triethanolamine (TEA). MEA showed the highest pulping rate and TEA the lowest. Eucalyptus was pulped rapidly with MEA at 443 K, with DEA at 473 K, and with TEA at 503 K. Pine was pulped rapidly only with DEA at 503 K. The pulp yields were 60-70 %. The yields of MEA pulps were 11-16 % higher than the yields of corresponding kraft pulps. The strength properties of the DEA pulps were comparable with those of kraft pulps. Pinus elliottii wood chips have been delignified with monoethanolamine at 473 K, producing pulps with similar strengths and significantly higher yields when compared to corresponding kraft pulps. When pulping to a kappa number of 34, 36 % (on wood) of the amine was consumed. Black liquor analyses showed that the liquor contains methanol, acetaldehyde-ammonia and several derivates of monoethanolamine. The results indicated that monoethanolamine enters into extensive chemical combination with wood components. (Wallis 1980) Obst and Sanyer (Obst & Sanyer 1980) have found that monoethanol amine, when added to an alkaline pulping liquor, increases the rate of beta-O-4 ether cleavage of blocked phenolic units in lignin. The extent of free phenolic beta-ether splitting is decreased in the presence of monoethanol amine. The reaction of the amine with the quinone methide intermediates may prevent condensation and thereby promote delignification.

2.5.5. Amine oxides

Cellulose and lignocellulose materials, for example, water extracted exploded wood from aspen, can be dissolved in n-methylmorpholine n-oxide or dimethylethanolamine n- oxide. The resulting solution can be used to spin fibers and to produce films. (Chanzy et al. 1986, Chanzy et al. 1987, Peguy 1989)

2.6.6. Summary

No studies related to the use of amines as pulping chemicals have been published since the 1980’s. This is probably due to the fact that amines have a tendency to enter into extensive chemical combinations with wood component. Therefore their recovery is difficult and expensive. 2.6. Ketones

DeHaas and Lang (DeHaas & Lang 1974) have studied pulping using various ketones and ammonia. They used acetone, methyl ethyl ketone, cyclohexanone and their mixtures. Compared with kraft pulping ketone-ammonia pulping of softwoods seemed to be more selective when the pulping liquor contained 30 % ketones and 18 % ammonia at a liquor-to-wood ratio of seven. The strength properties of the pulps obtained were generally similar to those of kraft pulps, with the exception of tear strength. The heat of evaporation of ketones is less than 25 % of the value for water and it has been suggested that the spent pulping liquor can be evaporated to a dry powder suitable for hydrogenation to phenols and by-product ketones. Significant yields of soluble material can be obtained by supercritical extraction of wood using, for example, acetone as solvent. The extracts contained phenols, cresols, guaiacols, eugenol, isoeugenol and coniferyl alcohol. (Calimli & Olcay 1978) Koll and Metzger (Koll & Metzger 1978) have used supercritical acetone (Tc = 508.5 K, pc = 4.7 MPa) to dissolve cellulose. About 98 % of the cellulose was dissolved in a ten hour extraction at 523-613 K. The extract contained mostly low molecular weight components. The use of underflow tar from acetone-ethanol-ammonia pulping as a low percentage additive to petroleum refinery feed has been proposed. (Griffith et al. 1980) Paszner and Chang (Paszner & Chang 1981) have patented a method using all types of lignocellulosic materials found in the nature as raw materials for producing natural lignin, paper grade pulps, pure xylose, pure mannose or dehydration products of the sugars. The pulping liquor contains 30-70 % of an organic solvent, which can be ethanol or acetone. The cooking temperature is about 473 K. The pH value of the solution is adjusted to 3.5 or less by a mild acid, e.g. formic acid, or an acid-salt mixture, for example HCl-KCl. The liquor-to-wood ratio of 4:1 is preferred. Cellulose-containing materials, for example wood or delignified pulps, can be rapidly saccharified to convert pentosans and hexosans to sugars. The process is conducted by cooking the material under pressure at 453-493 K with an acetone-water solvent mixture containing 0.05-0.25 w-% phosphoric, sulphuric or hydrochloric acids. The solvent mixture contains 60-70 v-% of acetone. A predominantly cellulosic material produces relatively pure glucose and whole wood yields mixed pentoses and hexoses. (Paszner & Chang 1983d) Cellulosic materials can be partially or totally hydrolysed using an aqueous acetone mixture containing a small amount of an acidic compound. The mixture contains from 70 to 100 v-% acetone and the process is performed at temperatures between 418 and 503 K. Acetone at high concentrations forms stable complexes with the sugars preventing their degradation. Lignin and sugars produced by this method are claimed to be commercially useful compounds. (Paszner & Chang 1984) Paszner and co-workers (Paszner et al. 1985) have presented a process called ACOS (acid catalysed organosolv saccharification). It uses aqueous acetone and a minor amount of a mineral acid to hydrolyse wood. All wood species can be used as raw materials and the process separates a natural lignin and fermentable sugars. When the ACOS process is conducted using aqueous acetone (60 %) and hydrochloric acid (0.025-0.1 N) at 473 K, Douglas-fir wood is completely dissolved. High quality, mostly low molecular weight lignin is obtained. (Cho & Paszner 1987) The Deineko group (Deineko et al. 1988b) has patented a pulping method consisting of cooking wood in an aqueous aprotic organic solvent at 403-433 K. The reagent use and environmental pollution can be reduced, while accelerating delignification and increasing pulp yield, if oxygen is present in the process in amounts of 12.6-23.2 w-% on wood. Liquor-to-wood ratios of 5:1-10:1 and solvent-to-water ratios of 1:1-5.7:1 can be used. The aprotic solvent can be acetone. When spruce sawdust is oxidated at 433 K for two hours in aqueous acetone (66 v-%) in the presence of molecular oxygen, nearly 90 % of the initial lignin in the sawdust is dissolved. (Deineko & Evtyugin 1988) Xylitol can be produced from vegetable material including hemicelluloses containing a xylane component. The raw material is pretreated with aqueous acetone or ethanol at 373-463 K for a period of from two minutes to four hours. About 20 % of the xylane containing hemicelluloses are dissolved during this first step. The residue is treated with a solution containing 50 % water and acetone or ethanol at 393-493 K for a period of from two minutes to six hours. The rest of the hemicelluloses are dissolved during this step. Xylose is recovered from the hemicelluloses containing solutions and reduced to xylitol. (Sinner et al. 1988) Oxidative delignification of spruce and aspen with oxygen in 60 % aqueous acetone has produced pulps in yields of 42-46 % and 56-62 % respectively. The residual lignin contents of the pulps were 5.8-6.7 % and 2.2-7.0 % respectively. (Deineko & Evtyugin 1989) Both hardwoods and softwoods can be delignified by oxygen in aqueous acetone without the use of any catalyst. The optimum process conditions are: liquor-to-wood ratio 50 dm3/kg, oxygen pressure 1.5 MPa, maximum temperature 423 K, temperature raising time 60 minutes and time at temperature 120 minutes. Spruce pulps containing 8.2 % lignin have been obtained at yields around 49 %. Acetone can be recovered by distillation. The remaining aqueous solution can be concentrated and dried. (Deineko et al. 1989, Zarubin et al. 1989) The temperature required for the anodic conversion of acetone soluble spruce lignin, in aqueous alkaline solutions, can be lowered by the addition of nitrobenzene or 1,3- dinitrobenzene. (Smith et al. 1989) Chang (Chang 1990) has presented a method for continuous countercurrent production of lignins and sugars from lignocellulosic materials. The raw material is continuously fed with a natural moisture content into a reactor. The cooking liquor containing 70-90 v- % acetone, water, and a small amount of sulphuric acid is fed into the reactor countercurrently with the raw material. The method is conducted at 423-483 K and a liquor-to-wood ratio of 5:1-15:1 is used. When aspen chips are treated with the method, 99.5 % of the original wood is dissolved. The pulp produced with oxygen in aqueous acetone is partially dehydrated. This causes strong hydrogen bondings between polysaccharide chains in the cell wall, leading to rigidity of the fibers. As a result of these pulp properties, the paper made from oxygen organosolv pulps have low strength properties. Washing the pulp after delignification with fresh solvent, followed by a treatment with 0.1 N sodium hydroxide favours the rehydration of the fibers resulting in paper strength properties near those of a kraft pulp. (Evtuguin et al. 1993a) Oxygen delignification in aqueous acetone has been proposed as a suitable method for transformation of hydrolysis lignin into microcrystalline cellulose and soluble active oxidized lignin. The main low molecular products are vanillin and vanillic acid. (Evtuguin et al. 1993b) Kostyukevich (Kostyukevich et al. 1993) has separated and analyzed lignins from oxygen pulping in a medium of water and acetone, ethanol or acetic acid. The molecular weights and polydispersities of the lignins were found to be similar to those of dioxane lignins and lower than those of alkali lignins. The type of solvent did not affect the molecular weight. Low-molecular weight products in spent liquors from oxygen-acetone pulping of Norway spruce and trembling aspen have been studied. Based on the composition of the identified reaction products they have proposed probable mechanisms for the oxidative and hydrolytic degradation of lignin. (Evtyugin & Deineko 1994) Evtyugin’s group (Evtyugin et al. 1994a) has compared oxygen-organosolv pulping of trembling aspen with corresponding kraft pulping. The organosolv cooks were carried out with 60 v-% acetone or ethanol, or 80 v-% acetic acid. Their results indicated that the higher yield of the oxygen-organosolv pulps is more due to reduced dissolution of cellulose in the pulping liquor than to preservation of the hemicelluloses. They (Evtyugin et al. 1994b) have also carried out comparative studies on the degree of degradation of the polysaccharide complex in the pulps. Results from these studies indicated that one of the main reasons for the low strength properties of oxygen- organosolv pulps is the depolymerization of the polysaccharide complex. It has been suggested that oxygen attacks double bonds of lignin during oxygen- organosolv delignification resulting in breaking and dissolution of the lignin. The pulping method is also called oxysolvolysis. (Vedernikov & Deineko 1994) Evtuguin et al. (Evtuguin et al. 1995) have used acetone, ethanol and acetic acid as solvents in oxygen delignification of aspen chips. The comparison of the results suggest that the acidity of the pulping liquor has a more important influence on the depolymerization of wood polysaccharides than the oxidation potential. Delignification of wood with oxygen in aqueous acetone partially removes bound water from pulp. This dehydration causes rigidity of the fibres and low strength properties of a paper produced using these pulps. When the pulp is washed with fresh pulping liquor after cooking and then treated with alkali, pulp fibres are rehydrated and the strength properties approach those of a kraft pulp. (Evtuguin et al. 1996) The delignification kinetics of Eucalyptus urograndis in acetone-water pulping has been found to be very similar to other hardwood pulping processes. Bulk and residual delignification phases have been identified. Most of the lignin removal occurs in the bulk phase. (Perez & Curvelo 1997) 2.7. Dioxane

Engel and Wedekind (Engel & Wedekind 1933) have patented a pulping method where dioxane is used as solvent in the presence of a catalyst. When pinewood meal is cooked in dioxane using a liquor-to-wood ratio of 8:1, for 5-10 hours at 353-363 K, in the presence on 1 % or less on dioxane of hydrochloric acid, pulp containing less than 3 % lignin is obtained in yields around 60 %. While soluble in water, dioxane can be easily washed from the pulp. Dioxane can be recovered by distilling the spent pulping liquor. Lignin can also be precipitated from the spent liquor by adding an ether. The ether is then separated from dioxane by distillation. Beech pulp produced by cooking in dioxane contained 87.8 % a-cellulose and 0.3 % lignin. The pulp yield was 41.3 % after pulping and 40.3 % after bleaching. (Wedekind 1936) The Pepper group (Pepper et al. 1959) has studied the isolation of a lignin fraction from spruce and aspen wood meals by low temperature (363-368 K) acidolysis, using a aqueous dioxane (90 %) containing hydrochloric acid (0.2 N). For both species, increased extraction times led to greater yields of lignin with lower methoxyl contents. When hydrolysing lignin with dioxane coniferyl alcohol and its 2,4-dinitrophenyl ether, coniferyl aldehyde and vanillin as corresponding 2,4-dinitrophenyl hydrazones have been identified. Also vanillinic acid was found in the hydrolysis product of spruce wood. From beech wood hydrolysis products, also syringic acid and from poplar hydrolysis products also p-hydroxy benzoic acid were found. (Sakakibara & Nakayama 1962a) In further studies, Sakakibara and Nakayama (Sakakibara & Nakayama 1962b) delignified sixteen species of wood with aqueous dioxane (50 %) at 453 K for 40 minutes. In the hydrolysis products from softwoods, they found coniferyl alcohol, p- coumar alcohol, coniferyl aldehyde and vanillic acid. The hydrolysis products of hardwoods were found to include sinapyl alcohol, coniferyl alcohol, coniferyl aldehyde, syringic acid and vanillic acid. Only poplar wood produced p-hydroxy benzoic acid during hydrolysis with aqueous dioxane. Acidolysis of spruce wood has been found to correspond to a pseudo-first order reaction mechanism. When the acidolysis with hydrochloric acid is carried out in anhydrous media, methanol shows higher yields for isolated lignin than dioxane. Adding up to 10 % water to dioxane increases the lignin yield making dioxane a better acidolysis medium than methanol. (Kosikova and Polcin 1973) Kayama and Takagi (Kayama & Takagi 1978) have studied alkaline dioxane pulping of oak wood. The aqueous pulping liquor contained 40 g/dm3 sodium hydroxide and 33- 50 v-% dioxane and the liquor-to-wood ratio was 4 dm3/kg. When the pulping was carried out at 453 K for 90 minutes, screened pulp yields varied from 45 to 48 %. The rate of delignification increased with increasing dioxane concentration in the pulping liquor but the pulp strength properties decreased. It has been reported that autohydrolysis of aspen and eucalyptus at temperatures 448- 493 K induced lignin solubility in aqueous dioxane (90 %). From aspen about 90 % of the lignin could be extracted after autohydrolysis with aqueous dioxane for 4-25 min at 468-488 K. Longer extraction times reduced lignin solubility. (Lora & Wayman 1978) Argyropoulos and Bolker (Argyropoulos & Bolker 1987) have isolated lignin from spruce sawdust with dioxane:water:hydrochloric acid (90:8:1.8 by volume). During batchwise extraction, irreversible recondensation of the lignin fragments took place. Pine wood and pine bark dioxane lignins have been compared. The Klason lignin content of bark was found to be 33 % lower than that in wood. The results also pointed to certain differences in the structure of bark and wood lignins. The nature of hemicelluloses bonded with lignin in pine and wood cell walls was predicted to be different. (Solar et al. 1988) The results obtained by analyzing two dimensional NMR spectra of dioxane lignin treated with lignin peroxidase have shown that the lignin may possess a three dimensional structure. (Gilardi et al. 1990) Montanari (Montanari et al. 1991) has determined the Klason lignin values and kappa numbers for wood residues from the dioxane-water pulping of Caribbean pine using hydrochloric acid as a catalyst. Yields of pulp and wood residues obtained with various pulping times, various concentrations of hydrochloric acid and related process variations were included in the factors studied. A proportional factor of 0.2 was determined. Degradation of lignin model compounds on acid treatment in aqueous dioxane has been studied in order to find the acid-catalyzed reactions of lignin during organosolv pulping. Hydrochloric and hydrobromic acids catalyse elimination of the hydrogen at the beta position of the lignin units and the release of formaldehyde from terminal methylol groups. (Karlsson & Lundquist 1991) Dovgan and Leonovich (Dovgan & Leonovich 1992) have studied the softening points of dioxane lignin from softwood, hardwood, and algae. The softening point curves of these lignin were found to be different. This is mainly caused by the differences in reactivity, molecular weights, and methoxyl group contents. A dehydropolymer has been produced by synthesizing from ferulic acid and dioxane lignin. The hydrodynamic behaviour was found to be similar for the lignin and the polymer. (Karmanov et al. 1992) Curvelo’s group (Curvelo et al. 1992a) has studied the delignification of Eucalyptus grandis by both batch and continuous pulping in aqueous dioxane. After a five hour extraction, the lignin removal was negligible in both processes. The lignin yield from the continuous process was lower. The pulp yield in continuous pulping was also lower due to a higher level of carbohydrate dissolution. Dioxane lignins from Pinus caribaea have been isolated using various catalytic concentrations of hydrochloric acid. They found an increase on the appearing of high molecular weight fractions, when the catalyst concentration was increased. (Curvelo et al. 1992b) Montanari et al. (Montanari et al. 1993) have extracted lignins from Pinus caribaea with aqueous dioxane (90 %) containing hydrochloric acid (1.0 N) using various extraction times (30-300 minutes). The lignin yields indicated that the delignification rate decreases with the extraction time. The analytical results pointed out that the delignification occurs mostly in the middle lamella. Eucalyptus chips have been extracted with aqueous dioxane (90 %) containing hydrochloric acid (2 N) at 362-363 K for 1-6 hours and the lignins obtained analyzed. The results indicated that all lignins have similar functional group distributions and that condensation reactions were not significant. (Curvelo et al. 1993) Solar and Kacik (Solar & Kacik 1993) have used dioxane to isolate lignins from Caribbean pine wood and bark in order to study the differences in these lignins. Their data indicated differences in condensed bis-guaiacyl structure contents of the lignins. They also found differences in the thermostability and the molecular weights of the compared lignins. The bark lignin is more crosslinked and contains more aromatic carbon linkages than wood lignin. The bark lignin has a lower thermostability. Lignins isolated from algae and sea-grasses by aqueous dioxane delignification have been studied. The isolated lignins were high molecular compounds with average molecular weights from 9000 to 10000 and a polydispersity degree from 1.9 to 2.3. (Dovgan’ & Medvedeva 1993a) The relatively low yield of products of nitrobenzene oxidation of algae and sea-grass lignins has indicated that this lignin has a higher degree of condensation than wood lignins. (Dovgan’ & Medvedeva 1993b) Dioxane lignins from algae and seagrasses are split by 27-41 % under the effect of nucleophilic agents (hydrazine in n-butanol or hydroxylamine in alkaline media). Phenols, oxyaromatic acids and higher fatty acids have been identified in the degradation products. (Dovgan’ & Medvedeva 1993c) Kanitskaya and co-workers (Kanitskaya et al. 1993) have bleached spruce dioxane lignin with hydrogen peroxide or sodium borohydride. Spectrophotometric, UV and NMR analyses showed that oxidative bleaching resulted in an increase in the content of non etherified fragments with an alpha-CO group. During sodium borohydride bleaching, the number of structures containing a carbonyl group decreased by 240 %. The content of non etherified aromatic fragments also decreased. The degree of lignin condensation was reduced and the degradation of the coniferyl aldehyde structure occurred during both types of bleaching. It has been found that base-catalysed dioxane pulping results in an increase in yield, a- cellulose, DP and a decrease in the hot alkali solubility. The favourable effects are induced by the formation of hydrogen bonds between the cellulose hydroxyls and the 1,4- oxygen atoms of dioxane. When borax is present during soda-dioxane pulping, the resulting cotton pulp is almost completely composed of pure alpha-cellulose. (Mostafa 1994) Povarova and Chupka (Povarova & Chupka 1994) have used chemiluminescence and micromanometry in order to study the effect of oxygen pressure and temperature on the oxidation of wood and dioxane lignin in alkaline (NaOH) and alkaline-ethanol media. An increase in temperature and pressure resulted in an increase in the oxidation rate. The oxygen consumption was found to be three-fold higher in the alkaline-ethanol media than in the alkaline media. Ultracentrifuging, viscometry and calorimetry have been used to study solvation of dioxane lignin in dimethyl sulfoxide, pyridine and dioxane. The results showed that the solvation effect became stronger when the solvent polarity increased. Solvation effects were negligible with pyridine and dioxane as solvents. (Bogolitsyn et al. 1994) Curvelo and Balogh (Curvelo & Balogh 1995) have compared continuous and batch pulping of eucalyptus with aqueous dioxane (90 %) containing hydrochloric acid (0.2 M) as the catalyst. The batch procedure was found to be the more selective one. The delignification rate constants were also higher in batch pulping. Dioxane lignins have been isolated from maple sawdust extracted with a solvent containing dioxane, water, and hydrochloric acid. The isolation was carried out using several methods. The results pointed to an increased lignin degradation occurring at high temperatures. (Solar & Kacik 1995)

2.8. Other organic compounds

Kuster and Daur (Kuster & Daur 1930) have used beta-naphthol in the presence of some hydrochloric acid to dissolve residual lignin from pulp. 2-naphthol has been used to prevent lignin self-condensation during the autohydrolysis of aspen wood. A highly delignified product could be obtained with 2 % 2-naphthol at autohydrolysis temperatures around 448 K for two hours. (Wayman & Lora 1978) Hossain (Hossain 1958, Hossain 1959) has found that dimethyl sulphoxide (DMSO), in the presence of small amounts of mineral acids, is a powerful delignifying agent for both high-lignin sulphite pulps and wood chips. The delignifying process is conducted at a temperature between 399 and 448 K in an aqueous solution containing 75 w-% DMSO in the presence of catalytic amounts of sulphuric acid or hydrochloric acid. A solution of 5 % sulphur dioxide (SO2), 2 % nitrogen dioxide (NO2) and 5 % chlorine (Cl2) dissolved in DMSO has shown good pulping properties. Cooking times were short and the temperature did not exceed 413 K. The pulp yield varied from 42 to 80 % depending on the pulping conditions. Both sawdust and chips of spruce could be pulped and the pulps were easy to bleach. (Clermont and Bender 1961) The efficiency of alkaline catalysis at wood and lignin oxidation increases considerably when the process is conducted by adding DMSO. Proper adjustment of the pH-values can considerably improve the selectivity of the process. (Chupka et al. 1993) Aqueous solution containing 35 % sodium m-xylenesulfonate acidified with small amounts of sulphuric acid (0.1-0.4 N) have been used to delignify aspenwood at temperature from 373 to 423 K. At 423 K, using a solution containing 0.4 N sulphuric acid, the wood substance went into solution in 300 seconds. At higher temperatures, substances precipitated from the solution onto the solid residue. (Springer et al. 1963) Acidified aqueous solutions of sulfolane (tetramethylene sulphone) have been proved to be highly effective delignifying agents for finely divided aspen wood. Sulfolane concentrations 300-900 g/dm3, sulphuric acid concentrations 0.05-0.40 N and isothermal reaction temperatures of 403-433 K were used. High sulfolane concentrations and the reaction temperature were most effective in lignin removal. High acid concentrations improved the delignification rate but, at 0.4 N, caused lignin to precipitate on the solid residues. (Springer & Zoch 1966) Starr and Casebier (Starr & Casebier 1970) have patented a pulping process where the sodium sulphide pulping liquor contains an organic solvent (30-70 %). The solvent is selected from the group consisting of alkyl polyols, cyclic sulfones, acyclic sulfones and mixtures thereof. Pulping is carried out at 433-463 K for 2.5-4 hours. A liquor-to-wood ratio of 5:1 is used. A pulping method where cellulosic material is extracted at 443-483 K under elevated pressure with an aqueous solution of dimethyl formamide in two stages has been patented. The extraction is conducted using liquor-to-wood ratios between 7:1 and 20:1. The pulp is washed after each extraction stage with fresh solvent. (Lenz et al. 1984) Sameshima and Takamura (Sameshima & Takamura 1987) have found that aqueous ammonium oxalate solutions (5 g/dm3) are good pulping agents for bast fibers containing pectin as the main bonding substance. Pulping was carried out at 358 K using a liquor-to-wood ratio of 25:1. The strength properties of pulps produced by this method were better than those of conventional soda pulps. It has been reported that high yield bamboo pulps can be produced using aqueous sodium xylenesulphonate as the pulping liquor. The solution containing 30-50 % sodium xylenesulphonate can be used several times before a simple recovery. Lignin is precipitated from the spent liquor by diluting it with water and the remaining liquid phase is returned to its original concentration by evaporation. The pulping temperature is 453-463 K and the resulting pulp can be used to produce, for example, dissolving pulp. (Lau 1941) Hardwoods and sugarcane bagasse have been delignified with approximately neutral 30- 40 % solutions of sodium xylenesulphonate. The same solutions were used for six or seven cooks before recovery. Poplar wood was delignified at 423 K for 11-12 hours, giving pulps with yields of about 52 %. Bagasse was cooked for three hours at 433 K. The yield of unbleached bagasse pulps was about 48 %. The installation costs of a mill using this method was estimated to be only one third of the costs for a kraft mill of similar size. This is mainly due to the simple chemical recovery with only precipitation of lignin and evaporation. (McKee 1946) McKee (McKee 1954) has compared his hydrotropic pulping process with the kraft process. The McKee hydrotropic process uses sodium xylenesulphonate as the pulping chemical and gives an about 5 % higher pulp yield than the kraft process. The alpha cellulose content of the McKee pulp is higher, and so the quality of the pulp can be judged to be better than the quality of kraft pulps. The chemical costs of the hydrotropic process are 75 % lower than for kraft pulping when compared in 1954. The cooking liquor of the hydrotropic process can be used five to six times before it has to be recovered. There are several advantages with the use of sodium xylenesulphonate: - Penetration of the chemicals into the chips is twice as rapid as in the kraft process. - It is neutral, odourless and non-volatile. - Air oxidation of the cellulose does not lower the pulp yield. - The chemical concentration does not decrease during the pulping cycle. Sodium xylenesulphonate is a selective delignifying agent and no sugars are present in the spent pulping liquor, which can be reused until it is saturated with lignin. The hydrotropic process can be operated without any effluents and malodorous gaseous emissions. (McKee 1954) The Hinrichs group (Hinrichs et al. 1957) has found that hydrotropic pulping, using a 30 % solution of sodium xylenesulphonate, is an effective method for dissolving lignin, but gives pulps with lower strength properties and brightness than alkaline pulping methods. Lignin is precipitated when the solution is diluted with water, leading to difficulties in pulp washing. McKee (McKee 1960) has listed the advantages and disadvantages of hydrotropic pulping using sodium xylenesulphonate. The advantages were: 1. The hydrotropic process requires only two-thirds of the equipment costs of a kraft mill. 2. It requires only two-thirds of the mill labour. 3. It has lower chemicals costs than conventional processes. 4. Its water consumption is low. 5. The pulp can be bleached with less chemicals than kraft pulps. 6. The pulp yield is higher than in conventional processes. 7. Valuable by-products can be recovered. 8. Both hardwoods, softwoods and mixtures of them can be used as raw-material. 9. The process is free from odours and aqueous effluents. A major disadvantage is the need for counter-currently operating screw presses for pulp washing. This type of equipment is more expensive than the filter presses commonly used in the pulp industry. (McKee 1960) Troitskii (Troitskii et al. 1989) has presented a pulping method consisting of treating softwood or hardwood chips in an aqueous alkaline solution at an elevated temperature and pressure. If the alkaline reagent is tetraethylammonium or tetrabutylammonium hydroxide, pulp yields are improved and residual lignin contents decreased. A reagent concentration of 0.75-1.25 M is needed for softwoods, and 0.5-1.0 M for hardwoods. Kuznetsov’s group (Kuznetsov et al. 1995) has proposed methods for processing silver-fir and larch barks. The bark is first activated by steam cracking. Silver-fir bark is thereafter subjected to benzene extraction in order to isolate silver-fir oil. The solid residue is subjected to catalytic pyrolysis in order to produce carbon sorbents. The activated larch bark is also extracted with benzene in order to isolate larch oil. The solid residue is activated by non-isobaric steam treatment, and the activated residue is subjected to water extraction in order to isolate tanning matter. The remaining solid residue is subjected to catalytic pyrolysis producing carbon sorbents.

2.9. Summary

Several organic chemicals have been proposed to be used as pulping chemicals, but only those processes using lower aliphatic alcohols or lower organic peroxyacids have survived. The most severe problems related to organosolv methods are the pulp quality, which typically is lower than the quality of corresponding kraft pulps, and solvent recovery. Organic solvents are more expensive than the inorganic chemicals used in kraft pulping. They have to be recirculated at high recovery rates. No single organosolv pulping method fulfils all the requirements for a process capable of replacing the kraft process. The ASAM process, which produces the best quality pulp, is a modified sulphate process, and therefore not sulphur free. Several organosolv solvent are selective pulping chemicals and do not degrade cellulose. Alcohols are solvents with high vapour pressures. This leads to high pressures in digesters which have to be redesigned. The simplicity of the recovery cycles depends on the number of by-products that are recovered. The recovery of by-products is necessary to balance the economy of most organosolv processes, but every product typically adds at least one unit operation to the recovery cycle. All the methods are capable of producing by-products. Some by-products, like alcohols and organic acids, are not especially valuable and some by-products, for example lignins, are produced in such large amounts that it is difficult to find markets for them. Environmental problems can be eliminated if the chemical cycle can be closed. This requires a TCF bleaching sequence which can be applied to most organosolv methods. The ASAM method is the only organosolv process producing both hardwood and softwood pulps with a quality comparable with corresponding kraft pulps. The optimal size of an organosolv pulping plant seem to be significantly smaller than the optimal size of a kraft pulping plant. This is mainly due to the fact that no recovery boiler is needed in organosolv processes. It is therefore to be presumed that an organosolv pulping plant will be built in an area with limited forest resources. Many organosolv methods are also capable of pulping annual plant which also point in that direction. The most probable recovery method is distillation, which is a unit operation with a high thermal energy consumption. It was therefore decided to study the possibilities for reducing distillation costs in one of the processes. The process chosen was the peroxyacid pulping method because it has been developed in Finland and therefore it is of particular interest. 3. Solvent recovery in peroxyacid pulping

As the strongest organic acid, formic acid is a good delignification solvent. The cooking process can be operated at atmospheric pressure. When the boiling temperature of formic acid is 374 K, the cooking temperature can be kept significantly lower than in other organosolv processes or in conventional pulping processes. The process is also suitable for pulping of nonwood materials. Silica does not cause problems in a process operated at acidic conditions (Rousu 1996). In order to replace an existing process, the new process should produce a better or cheaper product. When producing paper grade pulps, the peroxyformic acid (Milox) process at the present stage of development does not differ enough from the kraft process in terms of production costs or pulp quality. The major difference is that the peroxyformic acid process is totally sulphur-free. (Sundquist 1996) A major drawback in the peroxyformic acid process seems to be the relatively high energy consumption of the chemical recovery sequence and especially distillation. Therefore optional separation methods were evaluated and the reduction of distillation costs was studied. No exact data for the composition of the spent peroxyformic acid pulping liquor has been published. It is obvious that the liquor contains small amounts of several degradation products resulting from lignin and carbohydrates. The amount of acetic acid formed during birch pulping is around 4 % on wood (Muurinen et al. 1992). The amounts of other compounds are probably significantly lower. Therefore the liquor entering the separation section of the process was assumed to contain only water, formic acid and acetic acid. The effect of distillation feed composition on the economy of the process was studied. The composition of this liquor depends on the composition of the pulping liquor and on the operation conditions in the unit operations preceding distillation. This made it difficult to determine the desired separations for individual feed compositions. Therefore it was decided that the mole fraction of all products should be 98 %. 3.1. The search for feasible separation methods

When the dissolved solids have been removed from the spent pulping and washing liquors resulting from peroxyformic acid pulping of birch, the remaining liquor contains water, formic acid and acetic acid as the main components (Muurinen & Sohlo 1994). Available methods for the selection of feasible separation processes were applied to this ternary liquid mixture.

3.1.1. Wells and Rose

Wells and Rose (Wells & Rose 1986) have given rules (table 3) for the selection of a separation method. The rules are based on the following heuristics: - Avoid solid-handling in the process fluid if possible, because this is generally more expensive than the handling of liquids and gases. - Avoid adding a mass separating agent (MSA) if possible, because this has to be removed from the products. If a MSA has to be used, it should be removed in the next separation stage. The MSA should be a component already present, if possible. - Use processes which offer economy of scale.

Table 3. Rules for the selection of a separation process.

------1. Tabulate all components in their boiling point sequence. Indicate the mole fractions and the desired separations. 2. If the required separation can be achieved by simple splitting, take the appropriate action. 3. Consider the phase or phases of the feed mixture over a range of appropriate conditions. 4. Are there two or more phases present and can this be used as the basis for separation? 5. If solid and gas phases are present separate them by gas cleaning methods. 6. If the phases are solid and liquid separate them on the basis of solid particle size and content. 7. If the phases are liquid and gas separate them as follows: Choice number 1: gravity settlers Choice number 2: flash vaporization, partial condensation Choice number 3: simple distillation Choice number 4: other distillation methods, evaporation Choice number 5: change phase, chemical reaction etc. 8. Is only one phase present? ------Table 3. Cont.

------9. If the only phase is liquid separate it as follows: Choice number 1: settling if two or more liquid phases Choice number 2: fine particle separation, liquid-liquid extraction Choice number 3: sorptive mass transfer, change the phase to liquid/gas Choice number 4: liquid permeation, split and blend, chemical reaction, phase change. 10. If the only phase is gas separate it as follows: Choice number 1: gas permeation Choice number 2: adsorption, absorption Choice number 3: change of phase Choice number 4: other methods. 11. If the only phase is solid separate it as follows: Choice number 1: screening, sorting Choice number 2: chemical reaction, change of phase. 12. Review product specifications, recycle and feed streams. Consider the separation of small impurities with finishing methods. 13. Review the sequence of separations as follows: Choice number 1: remove early any component which may cause further undesired reaction or which is hazardous Choice number 2: remove early any component which requires the use of expensive materials of construction or high pressures or vacuum. Choice number 3: perform the most difficult separation as the last one Choice number 4: separate the main components early Choice number 5: perform the easiest separations first. 14. Review the selected separation sequence. Give particular attention to sequences involving technical extrapolation of equipment and equipment with high costs. ------

When this method is applied to the feed of the separation section of the peroxyformic acid pulping process, the use of ultrafiltration, pervaporation, flash evaporation, distillation, evaporation and crystallisation is suggested. This method favours separation processes where no mass separating agents are present.

3.1.2. Lee

In the selection method of Lee (Lee 1986), the pairs of components that have to be separated are identified using the following rule: - Separation is required when the maximum (or minimum) mass or mole ratio of two components in the available process streams is less (or greater) than that in the desired product. When the pairs to be separated are identified, feasible methods for each binary separation are listed. The appropriate separation methods can be found when the molecular and chemical characteristics of the components are known. The separation sequences obtained must be compared in order to select the best one. If the number of separation sequences is large it may be reduced using the following rules: - Separate components present in large amounts first. - Separate corrosive and hazardous components first. - Minimize the difference between the enthalpies of the product streams in order to reduce the work of energy recovery. - Perform the easiest separation first. Applying this method to the present separation problem suggests the following feasible separation methods: membrane processes, distillation, crystallisation, stripping, evaporation and flashing. Membrane processes and separation methods involving the use of an energy separating agent seem to be favoured by this selection method. Again no separation method using mass separating agents are suggested. This is due to the higher costs of those methods.

3.1.3. Barnicki and Fair

The separation process selection method presented by Barnicki and Fair (Barnicki & Fair 1990) is based on the application of the following rules: - Identify product streams and product specifications in order to ensure that no unnecessary separations are done. - Rank components in order of decreasing adjacent relative volatilities which give a strong indication of the ease of separation and the favorability of simple distillation. - Identify all azeotropic mixtures that may interfere with product specification, because azeotropes require special processing considerations. - Perform splits in the order specified by sequential heuristics. The following list of heuristics that may be used is based on the work of Nadgir and Liu (Nadgir & Liu 1983): - Separate corrosive and hazardous components first. - Separate reactive components first. - Separate azeotropic mixtures last. - Perform other difficult separations last, but before azeotropic separations. - Separate components in order of decreasing contents in the feed. - Favour 50/50 splits. - If the order of separations cannot be chosen using preceding heuristics, perform the separation with the smallest coefficient of difficulty of separation (CDS) first (Nath & Motard 1981):  X X  ln LK ⋅ HK  11− X − X   D   DB−  CDS = LK HK ⋅ ⋅+1   +   +  ln aLK, HK DB DB (1)

where CDS is the coefficient of difficulty of separation, XLK is the mole fraction of the light key component, XHK is the mole fraction of the heavy key component, aLK,HK is the relative volatility of the light key component with respect to the heavy key component, D is the overhead product flow rate and B is the bottom product flow rate.

When the split sequencing has been done, possible separation methods for each split are identified using component properties. Simple distillation should be used if possible. This method suggests the use of simple distillation, extractive distillation, azeotropic distillation, membrane processes, liquid-liquid extraction and adsorption as the methods to separate the formic acid-acetic acid-water mixture. The method seems to favour simple distillation, and separation methods involving the use of mass separating agents are not rejected. The reason for this is that the method takes into account also more complex feed solutions, for example, azeotropic mixtures.

3.1.4. Null

According to Null (Null 1987), there is no rote procedure by which separation processes are chosen, but certain guidelines can be followed in order to design the separation process sequence. The most important criterion in the design is economics. The first step in the selection of a separation process is to define the problem. The next step is to determine which separation processes are capable of accomplishing the separation. During this step component property differences, unit operation classification, scale of operation, design reliability, necessity for pilot plant tests, number of steps required, capital costs and energy requirements have to be taken into account. This method suggest the use of distillation, liquid-liquid extraction, crystallisation and adsorption as methods of solving the separation problem being considered.

3.2. Feasible separation methods

In order to determine the potential of the separation processes suggested by the selection methods, the properties of the components in the mixture and phase equilibria of the mixture have to be studied. Some useful properties of the components are listed in table 4. Table 4. Properties of formic acid, acetic acid and water (Wagner 1978, Wagner 1980, Matter-Muller & Stumm 1984, Weast 1978).

------Property HCOOH CH3COOH H2O ------melting point (K) 281.6 289.84 273.2 boiling point (K) 373.9 391.1 373.2 density (kg/m3) at 293 K 1220 1049 at 298 K 1213 998 liquid at 273 K 1000 solid at 273 K 907 viscosity (mPa·s) at 293 K 1.784 11.83 at 298 K 10.97 0.8949 at 313 K 8.18 steam at 293 K 96·10-6 heat of vaporization (J/g) at boiling point 900.2 694.8 2258.9 heat capacity (J/g·K) at 290 K 2.15 steam at 397 K 5.029 solid at 100 K 0.837 steam at 373 K 2.078 liquid at 298 K 4.179 steam at 273 K 2.06 dipole moment (D) 1.52 1.74 1.83 molecule diameter (Å) 5.5 5.3 3.0 ------

Evaporation and stripping are useful processes when dissolved solids are separated from spent pulping liquor, but liquid components cannot be selectively separated. Crystallization is mainly used, if a solid product is desired. The other separation methods are evaluated.

3.2.1. Simple distillation

Formic acid and water form maximum boiling azeotropes containing 77.5 % acid at 101.3 kPa and 83.2 % acid at 2.4 MPa (Wagner 1980). At the 101.3 kPa, the azeotropic mixture boils at 380.3 K, and at 2.4 MPa it boils at 407.8 K. This dependence upon pressure makes it possible to produce concentrated formic acid using pressure shift distillation (fig. 3). Formic acid, acetic acid and water form a saddle type ternary azeotrope boiling at 380.3 K and containing 39.3 % water, 48.2 % formic acid and 12.5 % acetic acid at 101.3 kPa (Hunsmann & Simmrock 1966). FORMIC ACID WATER

300 20 kPa kPa FEED

Fig. 3. Pressure shift distillation of binary mixture containing formic acid and water (Muurinen & Sohlo 1994).

The pressure shift distillation method can produce nearly pure water but cannot separate the acids. The feed liquor is pumped to a column operated at 300 kPa producing nearly pure water as distillate. The bottom product is fed to a vacuum (20 kPa) column producing nearly pure formic acid as distillate. The bottom product from the vacuum column is circulated to the pressurized column. When the acetic acid formed during delignification has to be separated, the distillate from the pressurized column can be extracted. No method capable of separating a feed mixture containing mainly water and less acetic acid than formic acid without any mass separating agent has been reported so far. If the feed to the separation process were to contain, for example, 60 wt-% acetic acid, 20 wt-% formic acid and 20 wt-% water, the mixture could be separated into three nearly pure products in a three stage distillation plant (fig. 4) using no mass separating agents (Jahn et al. 1979). The first and second columns are operated at atmospheric pressure. The distillate of the first column is nearly pure water and the bottom product is fed to the second column. The second bottom product is nearly pure acetic acid and the distillate is fed to the third column operated at 26 kPa. The third distillate is nearly pure formic acid and the bottom product is circulated to the first column. H2O CH3COOH HCOOH H2O HCOOH 100 100 26 kPa kPa kPa

CH3COOH

Fig. 4. A distillation sequence for the separation of an aqueous mixture containing formic acid and acetic acid (Muurinen & Sohlo 1994).

3.2.2. Azeotropic distillation

In azeotropic distillation an entrainer is added to a mixture to make distillation separation feasible. The entrainer forms an azeotrope with one of the components giving a respectable relative volatility between the azeotrope and the rest of the components. Hunsmann and Simmrock (Hunsmann & Simmrock 1966) have proposed the use of ethyl-n-butyl ether as the entrainer in order to separate water from formic acid and acetic acid. When water has been separated, the acids may be separated using 1-chloro butane (Hunsmann & Simmrock 1966) or chloroform (Othmer & Conti 1962) as an entrainer. Water may be separated from the acid mixture also by using n-propyl formate (Clarke & Othmer 1928), butyl formate, amyl formate, isoamyl formate (Ricard & Guinot 1928), ethylene dichloride (Nakamura et al. 1966), n-propyl acetate, n-butyl acetate, ethyl acetate (Econompoulos 1978), n-hexane (Horlenko 1973) and pentane (Anon. 1955) as entrainers. Acetic acid may be separated from the ternary mixture by an esterification-azeotropic distillation method using butanol as the esterification agent. Formic acid and water are removed from the system as a butyl formate-water azeotropic mixture. (Xu et al. 1990). When benzene is used as the entrainer, nearly pure acetic acid is produced by separating formic acid and water from the mixture (Null et al. 1967).

3.2.3. Extractive distillation

An extraneous material is added to the system also in extractive distillation. The function of this material, called solvent, is to enhance the relative volatility of the key components. The solvent is normally a high-boiling material, which in some cases may form loose chemical bonds with one of the key components. Solvents used to separate the formic acid-acetic acid-water mixture should not have very high boiling points. Formic acid decomposes slowly, even at low temperatures. When the temperature of formic acid (99.9 %) is raised from 313 K to 373 K, the rate constant of the decomposition reaction increases from 700 s-1 to 4.7·105 s-1 (Barham & Clark 1951). The presence of water in the system has been found to act as a negative catalyst on the decomposition reaction. Solvents proposed for separating the formic acid - acetic acid - water mixture are n- methyl-2-pyrrolidone (Cohen 1986), triethanol amine (Mahapatra et al. 1988), n,n- dimethylacetamide, isophorone, sulfolane, hexanoic acid and octanoic acid (Berg et al. 1990). Lahtinen (Lahtinen 1991) has proposed the use of butyric acid (fig. 5) as the solvent in the separation of this ternary mixture. The feed liquor and the solvent are introduced to a column operated at 150 kPa and producing nearly pure water as distillate. The bottom product is fed to the second column operated at 20 kPa and producing nearly pure formic acid as distillate. The second bottom product is fed to the third column operated at 71 kPa and producing nearly pure acetic acid as distillate. The third bottom product containing mainly butyric acid is circulated to the first column.

Makeup

H2O HCOOH CH3COOH 152 20.3 70.9 kPa kPa kPa

H2O CH3COOH HCOOH

C3H7COOH

Fig. 5. Extractive distillation of the formic acid-acetic acid-water mixture with butyric acid (Muurinen & Sohlo 1994).

N-formyl morpholine has been proposed for use as the solvent when separating water from formic acid (Bulow et al. 1977). Mixtures of di-n-butyl formamide or di-n-pentyl formamide with formamide, methyl formamide, dimethyl formamide or acetamide can also be used as solvents (Bott et al. 1984). Berg (Berg 1988) has proposed the use of ethylene carbonate or propylene carbonate either alone or mixed with heptanoic acid, benzoic acid, isophorone or 2-hydroxyacetophenone. Sulfones like thiophan sulphone, dimethyl sulphone, adiponitrile, phenyl sulphone and acetophenone are also feasible solvents (Berg & Yeh 1987). Acetyl salicylic acid (Berg 1990) can be used as the solvent in order to separate formic acid from acetic acid. Butyl benzoate or ethylene carbonate can be used as diluents. Carboxylic acids in the range of hexanoic acid to neodecanoic acid can be used as solvents with or without diluents such as methyl benzoate, cetophenone and nitrobenzene (Berg 1987).

3.2.4. Liquid-liquid extraction

Liquid-liquid extraction (LLE) is a separation method in which a liquid solution (the feed) is contacted with a second liquid (the solvent). The solvent is immiscible or nearly immiscible with the feed and extracts a desired component (the solute). Several solvents capable of extracting formic acid and acetic acid from dilute aqueous solutions have been presented: ethyl acetate (Garner et al. 1955), tri-n-octylamine (Jagirdar & Sharma 1980, Mishra & Tiwari 1988, Chen et al. 1989, Ziegenfuss & Maurer 1994, Kirsch et al. 1997, Kirsch & Maurer 1998), trioctyl-phosphine oxide (Golob et al. 1981, Ricker et al. 1979, Helsel 1977, Wardell & King 1978, Grinstead 1974, Rickelton & Boyle 1988), methylisobutyl-ketone (Pilhofer and Dichtl 1986, Kirsch & Maurer 1998), 1,1,1-trichloroethane (Marr et al. 1987), tri-n-decylamine (Wojtech et al. 1986), s-butyl-di-n-octylphosphine oxide (Weferling et al. 1988) and tri-n- butyl phosphate (Shah and Tiwari 1981). Tertiary and quaternary amines have also been suggested as solvents for acetic acid extraction (Ricker et al. 1980, Tamada et al. 1990, Tamada & King 1990a, Tamada & King 1990b, Yang et al. 1991, King 1992). Hohenschutz and co-workers (Hohenschutz et al. 1985) have proposed the use of di-n- butylformamide to extract formic acid from water and n-n-butyl-n-ethylhexylacetamide for the extraction of acetic acid. The Andrae group (Lorenz & Dichtl 1993) has presented a new extraction method for the recovery of acetic acid from aqueous streams containing 5-50 wt-% acetic acid. The main difference between this method and conventional methods is that a sidestream is taken from the distillation column recovering the solvent and recycled to the extraction column. According to the inventors, this makes it possible to handle higher acetic acid concentrations than in conventional processes. Ethyl acetate and methyltertiarybutyl ether can be used as solvents in this method. According to Van Brunt (Brunt 1992) di-2,4,4-trimethylpenthyl-n-octyl phosphine oxide, tri-2,4,4-dimethylpenthyl phosphine oxide and mixtures thereof diluted with at least one high molecular weight ketone can be used to extract acetic acid from aqueous solutions. Mixtures with acetic acid concentrations of up to 65 wt-% can be extracted. Gentry and Solazzo (Gentry & Solazzo 1995) have reported that an improved extraction based recovery method for formic acid and acetic acid from aqueous streams uses only one third of the energy of a conventional extraction system. This method uses a blend of phosphine oxides as the solvent. This blend is a high-boiling solvent while most commercial extraction systems use low-boiling solvents. The aqueous acid stream is extracted counter-currently with the solvent. The co-extracted water and any light impurities are removed from the solvent-acid mixture by stripping. In the next stage the acids are stripped overhead from the solvent which is recycled to the extractor. This type of design is claimed to be economic for aqueous streams containing 1-30 % acids.

3.2.5. Membrane processes

The molecule diameters of formic acid, acetic acid and water are between 3·10-10 and 5.5·10-10 m (table 4). According to Rautenbach and Albrecht (Rautenbach & Albrecht 1989) reverse osmosis could be used to separate molecules of these sizes. The pressure needed for reverse osmosis separation of aqueous organic solutions is commonly high and they suggest the use of pervaporation to separate such mixtures. Pervaporation is a membrane separation method where a membrane is employed to act as a selective barrier between the liquid mixture to be separated and the gas phase to which the separated components are transmitted. The separation can be brought about using low vacuum pressure in the gas phase, a sweeping flow of gas to remove the vapour permeate or a temperature difference. Acetic acid-water mixtures can be separated by pervaporation using a poly- (tetrafluoroethylene) membrane (Aptel et al. 1976) but no successful experiments to separate formic acid-water mixtures have been reported. Zheleznov and co-workers (Zheleznov et al. 1998) have studied the use of diffusion dialysis and neutralization dialysis in order to remove organic acids from dilute aqueous solutions. Diffusion dialysis uses water as a stripping agent and the driving force is the gradient of acid concentration. It actually transports the acid. Neutralization dialysis uses an aqueous base as dialysate. The organic acid anions of the feed are exchanged against OH-anions. The transport rate of acetic acid anions through the membrane in neutralization dialysis is significantly higher than the transport rate of acetic acid molecules in diffusion dialysis. The authors suggest neutralization dialysis to be combined with fermentation processes in order to improve fermenter performance.

3.2.6. Adsorption

The adsorption process is a selective concentration method for one or more components (adsorbates), that may be gases or liquids, at the surface of a microporous solid (adsorbent). Adsorption may be competitive with distillation when the relative volatility of the key components is below 1.25 (Ruthven 1984). Activated carbon and polymeric sorbents can be used as adsorbents in order to separate aqueous acetic acid mixtures (Baierl 1979, Kuo et al. 1987, Munson et al. 1987, Frierman et al. 1987). 3.2.7. Summary

Distillation seems to be the basic method for separating the ternary mixture formic acid- acetic acid-water. Membrane processes and adsorption can be excluded at this stage because no membranes or adsorbents suitable for this separation task have been presented. The use of mass separating agents involves a significant risk. Small amounts of the agent will always be present in the pulping liquor. Therefore the effect of a mass separating agent on the delignification process should be carefully evaluated before a selection is made. The availability of vapour-liquid equilibrium data guided the decision to focus the research on simple distillation and extractive distillation.

3.3. The effect of the feed composition on the separation of the formic acid-acetic acid-water mixture

In previous studies (Muurinen et al. 1993b) it has been pointed out that the small amount of acetic acid formed during peroxyformic acid pulping (Milox) has to be removed from the process in order to prevent the accumulation of acetic acid in the pulping liquor. The separation of this small amount of acetic acid makes the chemical recovery process more complicated and increases both operating and investment costs. This study was conducted in order to find out if it is possible to lower the energy requirements of the peroxyacid pulping process by increasing the amount of acetic acid in the pulping liquor. Solvent separation, including distillation, has been found to be the largest energy consumer in the total process. Therefore it was decided to study the effect of the feed liquor composition on the energy consumption of the separation process. The concentrations of formic acid, acetic acid and water were varied from 10 % to 80 % with 10 % intervals. Simple distillation and extractive distillation were considered to be the feasible separation methods. Initial screening of the solvent compositions was carried out by simulating these separation methods using shortcut methods. The minimum work of separation was calculated for all feed compositions in order to see if it can be used to predict which composition results in the lowest energy consumption.

3.3.1. Minimum work of separation

The minimum mechanical work required for separation of a homogeneous liquid mixture into pure products at constant temperature and constant pressure is (King 1986, p. 661)

WRTxxmin,TjFjFjF=- ln g Âj () (2) where Wmin,T is minimum work per mole of feed, R is gas constant, xjF is mole fraction of component j in feed and gjF is liquid activity coefficient of component j in feed mixture.

The activity coefficients of formic acid, acetic acid and water were calculated using the UNIFAC method (Fredenslund et al. 1977, Danner & Daubert 1983) and they are presented in table 5. The composition 26.0 % formic acid, 0.6 % acetic acid and 73.4 % water is assumed to be the composition of the feed stream to the separation section of the Milox recovery process (Muurinen & Sohlo 1993b)

Table 5. The activity coefficients (gi) of components (i) at various mixture compositions.

------x1 x2 x3 g1 g2 g3 ------0.8 0.1 0.1 1.0078 1.5818 1.0701 0.7 0.2 0.1 1.0302 1.3846 1.1058 0.7 0.1 0.2 1.0056 1.5998 1.0742 0.6 0.3 0.1 1.0630 1.2541 1.1529 0.6 0.2 0.2 1.0259 1.3966 1.1131 0.6 0.1 0.3 1.0029 1.6259 0.0748 0.5 0.4 0.1 1.1041 1.1650 1.2095 0.5 0.3 0.2 1.0567 1.2621 1.1637 0.5 0.2 0.3 1.0200 1.4135 1.1166 0.5 0.1 0.4 1.0004 1.6628 1.0722 0.4 0.5 0.1 1.1523 1.1032 1.2746 0.4 0.4 0.2 1.0956 1.1700 1.2240 0.4 0.3 0.3 1.0476 1.2728 1.1700 0.4 0.2 0.4 1.0130 1.4372 1.1160 0.4 0.1 0.5 0.9990 1.7148 1.0666 0.3 0.6 0.1 1.2066 1.0603 1.3479 0.3 0.5 0.2 1.1414 1.1057 1.2931 0.3 0.4 0.3 1.0832 1.1760 1.2334 0.3 0.3 0.4 1.0360 1.2874 1.1716 0.3 0.2 0.5 1.0053 1.4700 1.1115 0.3 0.1 0.6 0.9998 1.7872 1.0581 0.2 0.7 0.1 1.2669 1.0311 1.4291 0.2 0.6 0.2 1.1931 1.0608 1.3706 0.2 0.5 0.3 1.1254 1.1081 1.3058 0.2 0.4 0.4 1.0670 1.1839 1.2373 0.2 0.3 0.5 1.0223 1.3074 1.1683 0.2 0.2 0.6 0.9977 1.5155 1.1031 0.2 0.1 0.7 1.0044 1.8885 1.0474 ------Table 5. Cont.

------x1 x2 x3 g1 g2 g3 ------0.1 0.8 0.1 1.3328 1.0124 1.5185 0.1 0.7 0.2 1.2504 1.0298 1.4564 0.1 0.6 0.3 1.1733 1.0600 1.3868 0.1 0.5 0.4 1.1043 1.1107 1.3120 0.1 0.4 0.5 1.0471 1.1946 1.2352 0.1 0.3 0.6 1.0068 1.3349 1.1600 0.1 0.2 0.7 0.9912 1.5784 1.0911 0.1 0.1 0.8 1.0149 2.0311 1.0352 0.121 0.002 0.877 1.0950 2.8656 1.0046 ------

Using the activity coefficients presented in table 5, the minimum work of separation can be calculated at 298 K using equation 2. The gas constant value 8.314 J·K-1·mol-1 was used. The results are presented in table 6.

Table 6. Minimum work of separation (Wmin,T) at the temperature ( T ) 298 K for various mixtures containing formic acid (1), acetic acid (2) and water (3).

------x1 x2 x3 Wmin,298K kJámol-1 ------0.8 0.1 0.1 1.437 0.7 0.2 0.1 1.749 0.7 0.1 0.2 1.825 0.6 0.3 0.1 1.930 0.6 0.2 0.2 2.098 0.6 0.1 0.3 2.046 0.5 0.4 0.1 2.016 0.5 0.3 0.2 2.235 0.5 0.2 0.3 2.273 0.5 0.1 0.4 2.142 0.4 0.5 0.1 2.015 0.4 0.4 0.2 2.267 0.4 0.3 0.3 2.356 0.4 0.2 0.4 2.312 0.4 0.1 0.5 2.125 0.3 0.6 0.1 1.924 0.3 0.5 0.2 2.201 0.3 0.4 0.3 2.322 ______Table 6. Cont.

------x1 x2 x3 Wmin,298K kJámol-1 ------0.3 0.3 0.4 2.327 0.3 0.2 0.5 2.225 0.3 0.1 0.6 1.997 0.2 0.7 0.1 1.723 0.2 0.6 0.2 2.023 0.2 0.5 0.3 2.167 0.2 0.4 0.4 2.203 0.2 0.3 0.5 2.148 0.2 0.2 0.6 2.004 0.2 0.1 0.7 1.747 0.1 0.8 0.1 1.384 0.1 0.7 0.2 1.694 0.1 0.6 0.3 1.855 0.1 0.5 0.4 1.913 0.1 0.4 0.5 1.888 0.1 0.3 0.6 1.788 0.1 0.2 0.7 1.611 0.1 0.1 0.8 1.335 0.121 0.002 0.877 0.907 ______

When the value of the reference mixture is not taken into account, the average minimum work of separation is 1.981 kJ/mol. For a highly nonideal solution, like the formic acid-acetic acid-water mixture, the UNIFAC-based method of calculating the activity coefficients gives only a rough estimate of the real situation.

3.3.2. Vapour-liquid equilibrium data for the formic acid- acetic acid-water mixture

Lahtinen and co-workers (Lahtinen et al. 1994) have published vapour-liquid equilibrium data at the pressures 20 kPa, 101.3 kPa and 300 kPa for the binary pairs formic acid- water, formic acid-acetic acid and acetic acid-water. The total errors of the variables were estimated to be: temperature 0.1 K, pressure 0.1 %, liquid phase composition 0.5 % and vapour phase composition 1 %. The vapour-liquid equilibrium data was regressed using the PRO/II Regress program (Anon 1991c). The Wilson liquid activity coefficient model was used at atmospheric and higher pressures. It was combined with the Hayden-O’Connell model in order to take into account the vapour phase association. The binary interaction parameters for formic acid, acetic acid and water are presented in table 7.

Table 7. The energy terms of the Wilson model for binary pairs of water, formic acid and acetic acid at the pressures 101.3 kPa and 300 kPa.

------

Pressure (kPa) Components (l12 - l11)(l21 - l22) ------300 (1) water (2) formic acid -37.3714 432.742

300 (1) formic acid (2) acetic acid 29.9568 42.5987

300 (1) water (2) acetic acid 379.128 -17.3002

101.3 (1) water (2) formic acid -94.1944 538.953

101.3 (1) formic acid (2) acetic acid 289.184 -205.372

101.3 (1) water (2) acetic acid 1257.05 -524.041 ------

Table 8. Parameters of the eight parameter NRTL model for binary pairs of water, formic acid and acetic acid at the pressure 20 kPa.

------a a b b c c Components 12 21 12 21 12 21 a12 b12 ------(1) formic acid (2) acetic acid -0.739 1.177 443.4 -488.8 0.10 0.10 -1.0 -0.01

(1) water (2) acetic acid 0.887 0.103 -131.3 92.6 -0.06 -0.65 1.0 -0.01

(1) water (2) formic acid -0.641 0.348 54.7 -47.3 10.88 0.10 -1.0 0.01 ------At the pressure 20 kPa, the best regression results were achieved using the eight parameter form of the Non-random Two Liquid (NRTL) model together with the Hayden- O’Connell model. These regression results are presented in table 8. In order to study extractive distillation as the separation method for the formic acid- acetic acid-water system phase equilibrium data for binary pairs of the components with n-methyl-2-pyrrolidone (NMP) at 101.3 kPa were regressed using the Wilson/Hayden- O’Connell model (table 9). The data was produced by the Lahtinen group (Lahtinen et al. 1994).

Table 9. The energy terms of the Wilson model for binary pairs of NMP with water, formic acid and acetic acid at the pressure 101.3 kPa.

------

Components (l12 - l11)(l21 - l22) ------(1) acetic acid (2) NMP -441.344 -384.949

(1) formic acid (2) NMP -907.599 854.613

(1) water (2) NMP 184.431 -102.789 ------

The azeotropic compositions were predicted by performing equilibrium flash calculations in order to find similar compositions for the liquid and vapour phases present at equilibrium. The PRO/II 3.32 Flowsheet Simulation Program (Anon. 1991d) was used for the calculations and the results are shown in table 10.

Table 10. Azeotropic compositions of the binary mixture HCOOH-H2O and the ternary mixture HCOOH-CH3COOH-H2O.

------Pressure Azeotropic compositions kPa ------HCOOH-H2O HCOOH-CH3COOH-H2O %% ------20 47.0 - 53.0 30.8 - 19.2 - 50.0 101.3 59.2 - 40.8 55.1 - 20.5 - 24.4 300 75.2 - 24.8 75.8 - 3.5 - 20.7 ------1 0 J Binary azeotrope B Ternary azeotrope 0.8 0.2

0.6 0.4 J H2O B HCOOH 0.4 0.6

0.2 0.8

1 1 0.8 0.6 0.4 0.2 0

CH3COOH

Fig. 6. Distillation boundaries for the HCOOH-CH3COOH-H2O mixture at 20 kPa.

1 0 J Binary azeotrope 0.8 0.2 B Ternary azeotrope

0.6 0.4 H2O HCOOH 0.4 J 0.6

B 0.2 0.8

1 0 1 0.8 0.6 0.4 0.2 0

CH3COOH

Fig. 7. Distillation boundaries for the HCOOH-CH3COOH-H2O mixture at 101.3 kPa.

The azeotropic compositions were placed on ternary diagrams (fig. 6, 7 & 8). The distillation boundaries, which cannot be crossed by simple distillation, were drawn as straight lines combining the two azeotropes and the ternary azeotrope with each edge of the diagram. According to Fien and Liu (Fien & Liu 1994), the error in approximating true boundaries with straight lines is inconsequential, especially at the flowsheet- synthesis stage.

1 0 J Binary azeotrope B Ternary azeotrope 0.8 0.2

0.6 0.4 HCOOH H2O 0.4 0.6 J 0.2 B 0.8

1 0 1 0.8 0.6 0.4 0.2 0

CH3COOH

Fig. 8. Distillation boundaries for the HCOOH-CH3COOH-H2O mixture at 300 kPa.

3.3.3. Shortcut simulation of the first separation stage

Due to the availability of phase equilibrium data, simple distillation at the pressures 20 kPa, 101.3 kPa and 300 kPa and extractive distillation with n-methyl-2-pyrrolidone (NMP) at the pressure 101.3 kPa were studied as the main separation stages. They were simulated with various feed liquor compositions using shortcut calculation methods in the PRO/II Flowsheet Simulation Program version 3.32 (Anon. 1991d) where a generalised Fenske method is used to predict product distributions. The minimum reflux ratio is determined by the Underwood method. The number of theoretical trays, the actual reflux rates, and condenser and reboiler duties are determined using a Gilliland-type correlation. Typical heuristic rules for column sequencing are (Rose 1985): - Sequences removing components one by one in the column overheads should be preferred. - Sequences giving a more equimolar division between the products should be preferred. - If the feed mixture contains one component in excess, this component should be removed as early as possible. The feasible first separation stages for each feed composition were chosen using following rules: (1) The number of equilibrium stages was limited to 100 in order to keep the investment costs reasonable. (2) The most plentiful component should be the product of the first separation stage in order to keep the operating costs reasonable. (3) The molar flow of a product should be at least 10 % of the feed flow in order to confirm the operability of the column. If a separation method according to the simulation results did not fulfil these requirements, it was rejected. Specifications used in simulations were: (1) The molar fraction of one component in a product is 0.98. (2) The recovery rate of one component into a product is 98 %.

3.3.3.1. Simple distillation at 20 kPa as the first separation stage

The simulation results used to either accept or reject simple distillation at 20 kPa as the first stage for the separation of formic acid-acetic acid-water mixtures are presented in tables 11 and 12. When the feed liquor contains 10 % formic acid, simple distillation at 20 kPa is accepted as the first separation stage only for mixtures containing 70 % or 80 % water. The feed composition 10 % formic acid, 70 % acetic acid and 20 % water was rejected because the most plentiful component (acetic acid) could not be obtained as a product of the column. Other feed compositions were rejected because the number of equilibrium stages was higher than 100.

Table 11. Simulation results of simple distillation at 20 kPa as the first separation stage when the feed contains 10 % formic acid. Formic acid is FA, acetic acid is AA and water is W.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------10 % FA 80 % AA 1292 FA 9.97 - 10 % W

10 % FA 70 % AA 57 W 19.99 - 20 % W

10 % FA 60 % AA 1221 FA 9.98 - 30 % W ______Table 11. Cont.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------10 % FA 50 % AA 1122 FA 10.00 - 40 % W

10 % FA 40 % AA 987 FA 9.99 - 50 % W

10 % FA 30 % AA 151 W 40.00 - 60 % W

10 % FA 20 % AA 64 W 30.00 + 70 % W

10 % FA 10 % AA 41 W 20.00 + 80 % W ------

Table 12. Simulation results of simple distillation at 20 kPa as the first separation stage when the feed contains 20 % formic acid. Formic acid is FA, acetic acid is AA and water is W.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------20 % FA 70 % AA 1250 FA 20.00 - 10 % W

20 % FA 60 % AA 1228 FA 20.00 - 20 % W ______Table 12. Cont.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------20 % FA 50 % AA 1162 FA 20.00 - 30 % W

20 % FA 40 % AA 1058 FA 20.00 - 40 % W

20 % FA 30 % AA 1755 FA 20.00 - 50 % W

20 % FA 20 % AA 151 W 40.00 - 60 % W

20 % FA 10 % AA 66 W 30.00 + 70 % W ------

When the feed liquor contains 20 % formic acid, simple distillation can be accepted as the first separation stage only when the water content is 70 %. All the other feed compositions required a number of equilibrium stages above 100. When the feed liquor contains 30 %, 40 %, 50 %, 60 %, 70 % or 80 % formic acid simple distillation at 20 kPa cannot be accepted as the first separation stage. All the feed compositions required a number of equilibrium stages that was above 100.

3.3.3.2. Simple distillation at 101.3 kPa as the first separation stage

The simulation results used to either accept or reject simple distillation at 101.3 kPa as the first stage for the separation of formic acid-acetic acid-water mixtures are presented in tables 13 to 18. Table 13. Simulation results of simple distillation at 101.3 kPa as the first separation stage when the feed contains 10 % formic acid. Formic acid is FA, acetic acid is AA and water is W.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------10 % FA 80 % AA 129 AA 20.00 - 10 % W

10 % FA 70 % AA 39 W 20.00 - 20 % W

10 % FA 60 % AA 48 W 30.01 - 30 % W

10 % FA 50 % AA 59 W 40.00 - 40 % W

10 % FA 40 % AA 60 W 50.00 + 50 % W

10 % FA 30 % AA 50 W 40.02 + 60 % W

10 % FA 20 % AA 40 W 30.01 + 70 % W

10 % FA 10 % AA 27 W 20.01 + 80 % W ------

When the feed liquor contains 10 % formic acid, simple distillation at 101.3 kPa is accepted as the first separation stage for mixtures containing at least 50 % water. The feed composition 10 % formic acid, 80 % acetic acid and 10 % water was rejected because the number of equilibrium stages was above 100. The others were rejected because the most plentiful component could not be obtained as the product of the column. Table 14. Simulation results of simple distillation at 101.3 kPa as the first separation stage when the feed contains 20 % formic acid. Formic acid is FA, acetic acid is AA and water is W.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------20 % FA 70 % AA 74 AA 30.00 + 10 % W

20 % FA 60 % AA 66 W 19.99 - 20 % W

20 % FA 50 % AA 66 W 30.00 - 30 % W

20 % FA 40 % AA 64 W 40.00 + 40 % W

20 % FA 30 % AA 50 W 50.00 + 50 % W

20 % FA 20 % AA 43 W 40.01 + 60 % W

20 % FA 10 % AA 35 W 30.00 + 70 % W ------

When the feed liquor contains 20 % formic acid, simple distillation at 101.3 kPa can be accepted as the first separation stage when the water content is 10 % or at least 40 %. The others were rejected because the most plentiful component could not be obtained as the product of the column. Table 15. Simulation results of simple distillation at 101.3 kPa as the first separation stage when the feed contains 30 % formic acid. Formic acid is FA, acetic acid is AA and water is W.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------30 % FA 60 % AA 74 AA 40.00 + 10 % W

30 % FA 50 % AA 178 W 20.00 - 20 % W

30 % FA 40 % AA 123 W 30.00 - 30 % W

30 % FA 30 % AA 86 W 40.00 + 40 % W

30 % FA 20 % AA 69 W 50.00 + 50 % W

30 % FA 10 % AA 49 W 40.02 + 60 % W ------

When the feed liquor contains 30 % formic acid simple distillation at 101.3 kPa can be accepted as the first separation stage when the water content is 10 % or higher than 40 %. The other feed compositions were rejected because the number of equilibrium stages was above 100. Table 16. Simulation results of simple distillation at 101.3 kPa as the first separation stage when the feed contains 40 % formic acid. Formic acid is FA, acetic acid is AA and water is W.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------40 % FA 50 % AA 94 AA 50.00 + 10 % W

40 % FA 40 % AA 130 AA 40.00 - 20 % W

40 % FA 30 % AA 160 AA 30.00 - 30 % W

40 % FA 20 % AA 190 AA 20.00 - 40 % W

40 % FA 10 % AA 199 AA 9.99 - 50 % W ------

When the feed liquor contains 40 % formic acid, simple distillation at 101.3 kPa can be accepted as the first separation stage only when the water content is 10 %. The other feed compositions were rejected because the number of equilibrium stages was above 100. When the feed liquor contains 50 % formic acid, simple distillation at 101.3 kPa cannot be accepted as the first separation stage because the number of equilibrium stages is above 100. Table 17. Simulation results of simple distillation at 101.3 kPa as the first separation stage when the feed contains 60 % formic acid. Formic acid is FA, acetic acid is AA and water is W.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------60 % FA 30 % AA 45 FA 41.86 + 10 % W

60 % FA 20 % AA 58 FA 45.15 + 20 % W

60 % FA 10 % AA 90 FA 37.49 + 30 % W ------

When the feed liquor contains 60 % formic acid, simple distillation at 101.3 kPa can be accepted as the first separation stage for all liquor compositions.

Table 18. Simulation results of simple distillation at 101.3 kPa as the first separation stage when the feed contains 70 % or 80 % formic acid. Formic acid is FA, acetic acid is AA and water is W.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------70 % FA 20 % AA 38 FA 38.83 + 10 % W

70 % FA 10 % AA 50 FA 40.88 + 20 % W

80 % FA 10 % AA 36 FA 49.25 + 10 % W ------When the feed liquor contains 70 % or 80 % formic acid, simple distillation at 101.3 kPa can be accepted as the first separation stage for all liquor compositions.

3.3.3.3. Simple distillation at 300 kPa as the first separation stage

The simulation results used to either accept or reject simple distillation at 300 kPa as the first stage for the separation of formic acid-acetic acid-water mixtures are presented in tables 19 to 24.

Table 19. Simulation results of simple distillation at 300 kPa as the first separation stage when the feed contains 10 % formic acid. Formic acid is FA, acetic acid is AA and water is W.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------10 % FA 80 % AA 29 W 20.01 - 10 % W

10 % FA 70 % AA 44 W 20.00 - 20 % W

10 % FA 60 % AA 40 W 30.00 - 30 % W

10 % FA 50 % AA 37 W 40.00 - 40 % W

10 % FA 40 % AA 35 W 50.00 + 50 % W

10 % FA 30 % AA 32 W 40.00 + 60 % W ------Table 19. Cont.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------10 % FA 20 % AA 31 W 30.00 + 70 % W

10 % FA 10 % AA 29 W 20.01 + 80 % W ------

When the feed liquor contains 10 % formic acid, simple distillation at 300 kPa is accepted as the first separation stage for mixtures containing at least 50 % water. The other feed compositions were rejected because the most plentiful component (acetic acid) could not be obtained as a product of the column.

Table 20. Simulation results of simple distillation at 300 kPa as the first separation stage when the feed contains 20 % formic acid. Formic acid is FA, acetic acid is AA and water is W.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------20 % FA 70 % AA 58 AA 30.00 + 10 % W

20 % FA 60 % AA 63 W 20.00 - 20 % W

20 % FA 50 % AA 48 W 30.01 - 30 % W

20 % FA 40 % AA 44 W 40.00 + 40 % W ------Table 20. Cont.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------20 % FA 30 % AA 41 W 50.00 + 50 % W

20 % FA 20 % AA 38 W 40.00 + 60 % W

20 % FA 10 % AA 34 W 30.00 + 70 % W ------

When the feed liquor contains 20 % formic acid, simple distillation at 300 kPa can be accepted as the first separation stage when the water content is 10 % or higher than 40 %. The other feed compositions were rejected because the most plentiful component (acetic acid) could not be obtained as a product of the column.

Table 21. Simulation results of simple distillation at 300 kPa as the first separation stage when the feed contains 30 % formic acid. Formic acid is FA, acetic acid is AA and water is W.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------30 % FA 60 % AA 65 AA 40.00 + 10 % W

30 % FA 50 % AA 79 W 20.00 - 20 % W

30 % FA 40 % AA 68 W 30.00 - 30 % W ------Table 21. Cont.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------30 % FA 30 % AA 58 W 40.00 + 40 % W

30 % FA 20 % AA 48 W 50.00 + 50 % W

30 % FA 10 % AA 44 W 40.01 + 60 % W ------

When the feed liquor contains 30 % formic acid, simple distillation at 300 kPa can be accepted as the first separation stage when the water content is 10 % or higher than 40 %. The other feed compositions were rejected because the most plentiful component (acetic acid) could not be obtained as a product of the column.

Table 22. Simulation results of simple distillation at 300 kPa as the first separation stage when the feed contains 40 % formic acid. Formic acid is FA, acetic acid is AA and water is W.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------40 % FA 50 % AA 71 AA 50.00 + 10 % W

40 % FA 40 % AA 114 W 20.00 - 20 % W

40 % FA 30 % AA 86 W 30.00 - 30 % W ------Table 22. Cont.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------40 % FA 20 % AA 75 W 40.00 + 40 % W

40 % FA 10 % AA 64 W 50.00 + 50 % W ------

When the feed liquor contains 40 % formic acid, simple distillation at 101.3 kPa can be accepted as the first separation stage when the water content is 10 % or higher than 40 %. The other feed compositions were rejected because the most plentiful component (acetic acid) could not be obtained as a product of the column. When the feed liquor contains 50 % or 60 % formic acid, simple distillation at 300 kPa cannot be accepted as the first separation stage because the most plentiful component (formic acid) cannot be obtained as a product of the column.

Table 23. Simulation results of simple distillation at 300 kPa as the first separation stage when the feed contains 70 % or 80 % formic acid. Formic acid is FA, acetic acid is AA and water is W.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------70 % FA 20 % AA 105 AA 20.00 - 10 % W

70 % FA 10 % AA 128 AA 9.99 - 20 % W

80 % FA 10 % AA 57 FA 10.00 + 10 % W ------When the feed liquor contains 70 % formic acid, simple distillation at 300 kPa cannot be accepted as the first separation stage because the most plentiful component (formic acid) cannot be obtained as a product of the column. When the liquor contains 80 % formic acid, simple distillation at 300 kPa is an acceptable first separation stage.

3.3.3.4. Extractive distillation as the first separation stage

The simulation results used to either accept or reject extractive distillation with n-methyl- 2-pyrrolidone (NMP) at 101.3 kPa as the first stage for the separation of formic acid- acetic acid-water mixtures are presented in tables 24 to 27. The amount of NMP fed to the extractive distillation column is twice the molar flow of the feed. The ratio solvent/nonsolvent should be as low as possible to avoid excess costs (Fair 1987).

Table 24. Simulation results of extractive distillation with NMP at 101.3 kPa as the first separation stage when the feed contains 10 % formic acid. Formic acid is FA, acetic acid is AA and water is W.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------10 % FA 80 % AA 14 W 3.33 - 10 % W

10 % FA 70 % AA 13 W 6.67 - 20 % W

10 % FA 60 % AA 13 W 10.00 - 30 % W

10 % FA 50 % AA 12 W 13.33 - 40 % W

10 % FA 40 % AA 13 W 16.67 + 50 % W ______Table 24. Cont.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------10 % FA 30 % AA 13 W 19.98 + 60 % W

10 % FA 20 % AA 13 W 23.31 + 70 % W

10 % FA 10 % AA 13 W 26.64 + 80 % W ------

When the feed liquor contains 10 % formic acid, extractive distillation with NMP at 101.3 kPa is accepted as the first separation stage for mixtures containing at least 50 % water. When the water content of the feed liquor is below 50 % the most plentiful component (acetic acid) cannot be obtained as a product of the column.

Table 25. Simulation results of extractive distillation with NMP at 101.3 kPa as the first separation stage when the feed contains 20 % formic acid. Formic acid is FA, acetic acid is AA and water is W.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------20 % FA 70 % AA 14 W 3.33 - 10 % W

20 % FA 60 % AA 13 W 6.67 - 20 % W

20 % FA 50 % AA 13 W 10.00 - 30 % W ------Table 25. Cont.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------20 % FA 40 % AA 12 W 13.33 + 40 % W

20 % FA 30 % AA 13 W 16.67 + 50 % W

20 % FA 20 % AA 13 W 19.98 + 60 % W

20 % FA 10 % AA 13 W 23.31 + 70 % W ------

When the feed liquor contains 20 % formic acid, simple distillation at 300 kPa can be accepted as the first separation stage when the water content is at least 40 %. When the water content of the feed liquor is below 40 %, the most plentiful component (acetic acid) cannot be obtained as a product of the column.

Table 26. Simulation results of extractive distillation with NMP at 101.3 kPa as the first separation stage when the feed contains 30 % formic acid. Formic acid is FA, acetic acid is AA and water is W.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------30 % FA 60 % AA 14 W 3.33 - 10 % W

30 % FA 50 % AA 13 W 6.67 - 20 % W ______Table 26. Cont.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------30 % FA 40 % AA 13 W 10.00 - 30 % W

30 % FA 30 % AA 13 W 13.33 + 40 % W

30 % FA 20 % AA 13 W 16.67 + 50 % W

30 % FA 10 % AA 13 W 19.98 + 60 % W ------

When the feed liquor contains 30 % formic acid, extractive distillation with NMP at 101.3 kPa can be accepted as the first separation stage when the water content is at least 40 %. When the water content of the feed liquor is below 40 %, the most plentiful component (acetic acid) cannot be obtained as a product of the column.

Table 27. Simulation results of extractive distillation with NMP at 101.3 kPa as the first separation stage when the feed contains 40 % formic acid. Formic acid is FA, acetic acid is AA and water is W.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------40 % FA 50 % AA 14 W 3.33 - 10 % W

40 % FA 40 % AA 13 W 6.67 - 20 % W ______Table 27. Cont.

------Feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------40 % FA 30 % AA 13 W 10.00 - 30 % W

40 % FA 20 % AA 13 W 13.33 + 40 % W

40 % FA 10 % AA 13 W 16.67 + 50 % W ------

When the feed liquor contains 40 %, 50 %, 60 %, 70 % or 80 % formic acid, simple distillation at 101.3 kPa can be accepted as the first separation stage when the water content is at least 40 %. When the water content of the feed liquor is below 40 %, the most plentiful component (formic acid or acetic acid) cannot be obtained as a product of the column.

3.3.3.5. The first separation stage for the peroxyformic acid pulping process

The composition of the liquor fed to the separation sequence in the peroxyformic acid pulping process has been estimated to be 12.1 % formic acid, 0.2 % acetic acid and 87.7 % water. Simple distillation at the pressures 20, 101.3 and 300 kPa and extractive distillation with n-methyl-2-pyrrolidone (NMP) at 101.3 kPa were simulated as the first separation stage for this feed composition. Shortcut calculation methods were used. The results are shown in table 28. All four separation methods seem to be feasible first separation stages in the peroxyformic acid pulping process. Table 28. Simulation results for the first separation stage in the peroxyformic acid pulping process. SD is simple distillation and ED is extractive distillation with NMP.

------Separation number of trays Product Lowest product Accepted(+)/ method and at 1.2·Rmin flow/feed flow rejected(-) pressure (kPa) % ------SD 20 27 W 12.30 + SD 101.3 21 W 12.30 + SD 300 25 W 12.30 + ED 101.3 13 W 29.21 + ------

3.3.3.6. Summary of the simulations of the first separation stage

Feasible separation methods for the feed compositions studied are listed in tables 29 to 35. Simple distillation at 20 kPa seems to be a feasible first separation stage only when the water content of the feed is 70 % or higher. Water contents below 10 % and above 40 % seem to favour the use of simple distillation at atmospheric pressure, or at 300 kPa. Extractive distillation may be used at high water contents. When the feed contains 50 % formic acid no feasible separation methods were found.

Table 29. Feasible first separation stages when the feed contains 10 % formic acid. FA is formic acid, AA is acetic acid, W is water, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone.

------Feed composition Feasible separation methods FA AA W and pressures (kPa) %% % ------10 80 10 - 10 70 20 - 10 60 30 - 10 50 40 - 10 40 50 SD101.3,SD300,ED101.3 10 30 60 SD101.3,SD300,ED101.3 10 20 70 SD20,SD101.3,SD300,ED101.3 10 10 80 SD20,SD101.3,SD300,ED101.3 ------Table 30. Feasible first separation stages when the feed contains 20 % formic acid. FA is formic acid, AA is acetic acid, W is water, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone.

------Feed composition Feasible separation methods FA AA W and pressures (kPa) %% % ------20 70 10 SD101.3,SD 300 20 60 20 - 20 50 30 - 20 40 40 SD101.3,SD300,ED 101.3 20 30 50 SD101.3,SD300,ED101.3 20 20 60 SD101.3,SD300,ED101.3 20 10 70 SD20,SD101.3,SD300,ED101.3 ------

Table 31. Feasible first separation stages when the feed contains 30 % formic acid. FA is formic acid, AA is acetic acid, W is water, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone.

------Feed composition Feasible separation methods FA AA W and pressures (kPa) %% % ------30 60 10 SD 101.3, SD 300 30 50 20 - 30 40 30 - 30 30 40 SD 101.3, SD 300, ED 101.3 30 20 50 SD 101.3, SD 300, ED 101.3 30 10 60 SD 101.3, SD 300, ED 101.3 ------Table 32. Feasible first separation stages when the feed contains 40 % formic acid. FA is formic acid, AA is acetic acid, W is water, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone.

------Feed composition Feasible separation methods FA AA W and pressures (kPa) %% % ------40 50 10 SD 101.3, SD 300 40 40 20 - 40 30 30 - 40 20 40 SD 300, ED 101.3 40 10 50 SD 300, ED 101.3 ------

Table 33. Feasible first separation stages when the feed contains 50 % formic acid. FA is formic acid, AA is acetic acid, W is water, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone.

------Feed composition Feasible separation methods FA AA W and pressures (kPa) %% % ------50 40 10 - 50 30 20 - 50 20 30 - 50 10 40 ------

Table 34. Feasible first separation stages when the feed contains 60 % formic acid. FA is formic acid, AA is acetic acid, W is water, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone.

------Feed composition Feasible separation methods FA AA W and pressures (kPa) %% % ------60 30 10 SD 101.3

60 20 20 SD 101.3

60 10 30 SD 101.3 ------Table 35. Feasible first separation stages when the feed contains 70 % or 80 % formic acid. FA is formic acid, AA is acetic acid, W is water, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone.

------Feed composition Feasible separation methods FA AA W and pressures (kPa) %% % ------70 20 10 SD 101.3

70 10 20 SD 101.3

80 10 10 SD 101.3, SD 300 ------

The fact that no feasible separation methods were found for some feed compositions does not mean that these compositions are impossible to separate. The rejection criteria were chosen particularly in order to reject those separation methods and feed compositions that are likely to result in high separation energy consumptions and investment costs, when compared with other methods and feed compositions. All four separation methods seem to be feasible first stages for the estimated feed composition of the separation sequence of the peroxyformic acid pulping process.

3.3.4. Shortcut simulation of the second separation stage

Simple distillation at the pressures 20 kPa, 101.3 kPa and 300 kPa and extractive distillation with n-methyl-2-pyrrolidone (NMP) at the pressure 101.3 kPa were studied as potential second separation stages. They were simulated using shortcut calculation methods in the PRO/II Flowsheet Simulation Program version 3.32 (Anon. 1991d) where a generalised Fenske method is used to predict product distributions. The minimum reflux ratio is determined using the Underwood method. The number of theoretical trays, the actual reflux rates, and condenser and reboiler duties are determined using a Gilliland-type correlation. The feasible second separation stages for each feed composition were chosen using the following rules: (1) The number of equilibrium stages was limited to 100 in order to keep the investment costs reasonable. (2) The most plentiful component should be the product of the first separation stage in order to keep the operating costs reasonable. (3) The molar flow of a product should be at least 10 % of the feed flow in order to confirm the operability of the column. If a separation method according to the simulation results did not fulfil these requirements, it was rejected. Extractive distillation with NMP always removes water as the distillate. Therefore it was only considered as the separation method for those feed streams with water as the most plentiful component. Only simple distillation was considered as the second separation stage following the extractive distillation (Cohen 1986). Specifications used in the simulations were: (1) The molar fraction of one component in a product is 0.98. (2) The recovery rate of one component into a product is 98 %.

3.3.4.1. Simple distillation at 20 kPa as the second separation stage

The simulation results used to either accept or reject simple distillation at 20 kPa as the second separation stage following the accepted first stages for the separation of formic acid-acetic acid-water mixtures are presented in tables 36 and 37.

Table 36. Simulation results of simple distillation at 20 kPa as the second separation stage when simple distillation at 101.3 kPa is an accepted first separation stage. Formic acid is FA, acetic acid is AA and water is W. + is accepted and - is rejected.

------Original feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------10 % FA 40 % AA 1162 AA (+) 20.00 - 50 % W

10 % FA 30 % AA 1210 AA (+) 24.99 - 60 % W

10 % FA 20 % AA 1188 FA (-) 33.29 - 70 % W

10 % FA 10 % AA 1156 FA (+) 49.08 - 80 % W

20 % FA 70 % AA 369 FA (+) 35.23 - 10 % W ______Table 36. Cont.

------Original feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------20 % FA 40 % AA 959 AA (+) 33.30 - 40 % W

20 % FA 30 % AA 935 FA (-) 39.86 - 50 % W

20 % FA 20 % AA 911 FA (+) 49.64 - 60 % W

20 % FA 10 % AA 875 FA (+) 35.03 - 70 % W

30 % FA 60 % AA 51 FA (+) 26.40 + 10 % W

30 % FA 30 % AA 918 FA (+) 49.20 - 40 % W

30 % FA 20 % AA 906 FA (+) 40.70 - 50 % W

30 % FA 10 % AA 811 FA (+) 26.71 - 60 % W

40 % FA 50 % AA NO SOLUTION WAS FOUND 10 % W

60 % FA 30 % AA 939 FA (-) 31.87 - 10 % W ______Table 36. Cont.

------Original feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------60 % FA 20 % AA 877 FA (-) 27.86 - 20 % W

60 % FA 10 % AA 702 FA (-) 36.59 - 30 % W

70 % FA 20 % AA 929 FA (+) 48.28 - 10 % W

70 % FA 10 % AA 727 FA (-) 28.20 - 20 % W

80 % FA 10 % AA 810 FA (+) 39.43 - 10 % W ------

Simple distillation at 20 kPa can be accepted as the second separation stage after simple distillation at 101.3 kPa only when the original feed contains 30 % formic acid, 60 % acetic acid and 10 % water.

Table 37. Simulation results of simple distillation at 20 kPa as the second separation stage when simple distillation at 300 kPa is an accepted first separation stage. Formic acid is FA, acetic acid is AA and water is W. + is accepted and - is rejected.

------Original feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------10 % FA 40 % AA 1089 FA (-) 19.46 - 50 % W ______Table 37. Cont.

------Original feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------10 % FA 30 % AA 1071 FA (-) 24.38 - 60 % W

10 % FA 20 % AA 1038 FA (-) 32.53 - 70 % W

10 % FA 10 % AA 1021 FA (+) 48.03 - 80 % W

20 % FA 70 % AA 375 FA (+) 35.33 - 10 % W

20 % FA 40 % AA 963 FA (-) 32.73 - 40 % W

20 % FA 30 % AA 934 FA (-) 39.32 - 50 % W

20 % FA 20 % AA 911 FA (+) 49.15 - 60 % W

20 % FA 10 % AA 873 FA (+) 34.80 - 70 % W

30 % FA 60 % AA 51 FA (+) 26.48 + 10 % W

30 % FA 30 % AA 918 FA (+) 49.28 - 40 % W ______Table 37. Cont.

------Original feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------30 % FA 20 % AA 908 FA (+) 41.02 - 50 % W

30 % FA 10 % AA 358 FA (+) 01.50 - 60 % W

40 % FA 50 % AA 36 FA (+) 20.78 + 10 % W

40 % FA 20 % AA 852 FA (+) 33.88 - 40 % W

40 % FA 10 % AA 821 FA (-) 20.74 - 50 % W

80 % FA 10 % AA 770 FA (+) 22.08 - 10 % W ------

Simple distillation at 20 kPa can be accepted as the second separation stage after simple distillation at 300 kPa only when the original feed contains 30 % formic acid, 60 % acetic acid and 10 % water or 40 % formic acid, 50 % acetic acid and 10 % water.

3.3.4.2. Simple distillation at 101.3 kPa as the second separation stage

The simulation results used to either accept or reject simple distillation at 101.3 kPa as the second separation stage following the accepted first stages for the separation of formic acid-acetic acid-water mixtures are presented in tables 38, 39 and 40. Table 38. Simulation results of simple distillation at 101.3 kPa as the second separation stage when simple distillation at 20 kPa is an accepted first separation stage. Formic acid is FA, acetic acid is AA and water is W. + is accepted and - is rejected.

------Original feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------10 % FA 20 % AA 39 AA (+) 34.20 + 70 % W

10 % FA 10 % AA 124 AA (+) 48.95 - 80 % W

20 % FA 10 % AA 80 FA (+) 03.47 - 70 % W ------

Simple distillation at 101.3 kPa can be accepted as the second separation stage after simple distillation at 20 kPa only when the original feed contains 10 % formic acid, 20 % acetic acid and 70 % water.

Table 39. Simulation results of simple distillation at 101.3 kPa as the second separation stage when simple distillation at 300 kPa is an accepted first separation stage. Formic acid is FA, acetic acid is AA and water is W. + is accepted and - is rejected.

------Original feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------10 % FA 40 % AA 32 AA (+) 21.22 + 50 % W

10 % FA 30 % AA 36 AA (+) 26.68 + 60 % W ______Table 39. Cont.

------Original feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------10 % FA 20 % AA 42 AA (+) 35.47 + 70 % W

10 % FA 10 % AA 169 AA (-) 47.28 - 80 % W

20 % FA 70 % AA 144 FA (+) 35.30 - 10 % W

20 % FA 40 % AA 36 AA (+) 33.38 + 40 % W

20 % FA 30 % AA 41 AA (+) 40.18 + 50 % W

20 % FA 20 % AA 81 FA (+) 49.65 + 60 % W

20 % FA 10 % AA 49 FA (+) 34.80 + 70 % W

30 % FA 60 % AA 65 FA (+) 26.18 + 10 % W

30 % FA 30 % AA 53 FA (+) 49.95 + 40 % W

30 % FA 20 % AA 71 FA (+) 39.94 + 50 % W ______Table 39. Cont.

------Original feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------30 % FA 10 % AA 51 FA (+) 26.32 + 60 % W

40 % FA 50 % AA 22 FA (+) 22.56 + 10 % W

40 % FA 20 % AA 30 FA (+) 37.30 + 40 % W

40 % FA 10 % AA 26 FA (+) 20.62 + 50 % W

80 % FA 10 % AA 36 FA (+) 49.58 + 10 % W ------

Simple distillation at 101.3 kPa as the second separation stage after simple distillation at 300 kPa was rejected only when the original feed contains 10 % formic acid, 10 % acetic acid and 80 % water or 20 % formic acid, 70 % acetic acid and 10 % water. Simple distillation at 101.3 kPa can be accepted as the second separation stage after extractive distillation at 101.3 kPa only when the original feed contains 30 % formic acid and 20 or 30 % acetic acid, or when the feed contains 40 % formic acid, 20 % acetic acid and 40 % water. Table 40. Simulation results of simple distillation at 101.3 kPa as the second separation stage when extractive distillation with NMP at 101.3 kPa is an accepted first separation stage. Formic acid is FA, acetic acid is AA and water is W. + is accepted and - is rejected.

------Original feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------10 % FA 40 % AA 108 NMP (+) 21.14 - 50 % W

10 % FA 30 % AA 121 NMP (+) 18.17 - 60 % W

10 % FA 20 % AA 238 NMP (+) 14.86 - 70 % W

10 % FA 10 % AA 188 NMP (+) 06.67 - 80 % W

20 % FA 40 % AA 205 NMP (+) 24.01 - 40 % W

20 % FA 30 % AA 218 NMP (+) 21.34 - 50 % W

20 % FA 20 % AA 246 NMP (+) 18.32 - 60 % W

20 % FA 10 % AA 1221 NMP (+) 14.95 - 70 % W

30 % FA 30 % AA 56 NMP (+) 24.31 + 40 % W ______Table 40. Cont.

------Original feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------30 % FA 20 % AA 87 NMP (+) 21.51 + 50 % W

30 % FA 10 % AA 791 NMP (+) 18.41 - 60 % W

40 % FA 20 % AA 70 NMP (+) 24.52 + 40 % W

40 % FA 10 % AA 140 NMP (+) 21.62 - 50 % W ------

3.3.4.3. Simple distillation at 300 kPa as the second separation stage

The simulation results used to either accept or reject simple distillation at 300 kPa as the second separation stage following the accepted first stages for the separation of formic acid-acetic acid-water mixtures are presented in tables 41 and 42.

Table 41. Simulation results of simple distillation at 300 kPa as the second separation stage when simple distillation at 20 kPa is an accepted first separation stage. Formic acid is FA, acetic acid is AA and water is W. + is accepted and - is rejected.

------Original feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------10 % FA 20 % AA 46 AA (+) 34.20 + 70 % W ______Table 41. Cont.

------Original feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------10 % FA 10 % AA 51 AA (+) 48.95 + 80 % W

20 % FA 10 % AA 64 AA (-) 32.97 - 70 % W ------

Simple distillation at 300 kPa can be accepted as the second separation stage after simple distillation at 20 kPa when the original feed contains 10 % formic acid and 10 or 20 % acetic acid.

Table 42. Simulation results of simple distillation at 300 kPa as the second separation stage when simple distillation at 101.3 kPa is an accepted first separation stage. Formic acid is FA, acetic acid is AA and water is W. + is accepted and - is rejected.

------Original feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------10 % FA 40 % AA 20 AA (+) 19.52 + 50 % W

10 % FA 30 % AA 25 AA (+) 25.29 + 60 % W

10 % FA 20 % AA 32 AA (+) 40.72 + 70 % W

10 % FA 10 % AA 67 AA (-) 46.63 - 80 % W ______Table 42. Cont.

------Original feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------20 % FA 70 % AA 163 AA (-) 01.87 - 10 % W

20 % FA 40 % AA 44 AA (+) 33.78 + 40 % W

20 % FA 30 % AA 50 AA (+) 41.14 + 50 % W

20 % FA 20 % AA 54 AA (-) 48.86 - 60 % W

20 % FA 10 % AA 65 AA (-) 32.57 - 70 % W

30 % FA 60 % AA 165 AA (-) 01.18 - 10 % W

30 % FA 30 % AA 52 AA (+) 49.83 + 40 % W

30 % FA 20 % AA 56 AA (-) 39.92 - 50 % W

30 % FA 10 % AA 73 AA (-) 24.89 - 60 % W

40 % FA 50 % AA 144 FA (+) 37.92 - 10 % W ______Table 42. Cont.

------Original feed number of trays Product Lowest product Accepted(+)/ composition at 1.2·Rmin flow/feed flow rejected(-) % ------60 % FA 30 % AA 76 W (-) 17.13 - 10 % W

60 % FA 20 % AA 47 W (+) 36.06 + 20 % W

60 % FA 10 % AA 48 W (+) 47.42 + 30 % W

70 % FA 20 % AA 185 W (-) 16.32 - 10 % W

70 % FA 10 % AA 41 W (+) 48.43 + 20 % W

80 % FA 10 % AA 305 W (-) 20.26 - 10 % W ------

Simple distillation at 300 kPa can be accepted as the second separation stage after simple distillation at 101.3 kPa when the original feed contains 10 % formic acid and 20...40 % acetic acid, or when it contains 20 % formic acid and 30 or 40 % acetic acid, or when it contains 30 % formic acid, 30 % acetic acid and 40 % water, or when it contains 60 % formic acid and 10 or 20 % acetic acid, or when it contains 70 % formic acid, 10 % acetic acid and 20 % water. 3.3.4.4. Extractive distillation at 101.3 kPa as the second separation stage

The first separation stages did not result in a water rich product needing further treatment with any feed composition or separation method that was studied. Therefore extractive distillation with NMP at 101.3 kPa was not considered as a potential second separation stage.

3.3.4.5. The second separation stages for the peroxyformic acid pulping process

The composition of the liquor fed to separation sequence in the peroxyformic acid pulping process has been estimated to be 12.1 % formic acid, 0.2 % acetic acid and 87.7 % water. Simple distillation at the pressures 20, 101.3 and 300 kPa and extractive distillation with n-methyl-2-pyrrolidone (NMP) at 101.3 kPa were simulated as the second separation stage for this first stage feed composition. Shortcut calculation methods were used. Results are shown in tables 43 to 46.

Table 43. Simulation results for the second separation stage in the peroxyformic acid pulping process when simple distillation at 20 kPa is the first stage. SD is simple distillation and ED is extractive distillation with NMP. + is accepted and - is rejected.

------Method and 1.2 · minimum Product Lowest product Accepted(+)/ pressure number of trays flow/feed flows rejected(-) (kPa) % ------SD 101.3 24 FA (+) 07.24 - SD 300 22 W (-) 05.93 ------

The separation methods considered in this study are not feasible second stages following simple distillation at 20 kPa in the separation sequence of the peroxyformic acid process. Extractive distillation was not considered because water was not the most plentiful component in the feed. Table 44. Simulation results for the second separation stage in the peroxyformic acid pulping process when simple distillation at 101.3 kPa is the first stage. SD is simple distillation and ED is extractive distillation with NMP. + is accepted and - is rejected.

------Method and 1.2 · minimum Product Lowest product Accepted(+)/ pressure number of trays flow/feed flows rejected(-) (kPa) % ------SD 20 27 FA (+) 09.19 - SD 300 20 W (-) 07.64 ------

No separation method considered was a feasible second stage following simple distillation at 101.3 kPa when the selection criteria presented in chapter 3.3.3 were applied. Extractive distillation was not considered because water was not the most plentiful component in the feed.

Table 45. Simulation results for the second separation stage in the peroxyformic acid pulping process when simple distillation at 300 kPa is the first stage. SD is simple distillation and ED is extractive distillation with NMP. + is accepted and - is rejected.

------Method and 1.2 · minimum Product Lowest product Accepted(+)/ pressure number of trays flow/feed flows rejected(-) (kPa) % ------SD 20 26 FA (+) 08.05 - SD 101.3 44 W (-) 06.59 ------

The separation methods considered in this study are not feasible second stages following simple distillation at 300 kPa in the separation sequence of the peroxyformic acid process. Extractive distillation was not considered because water was not the most plentiful component in the feed. Table 46. Simulation results for the second separation stage in the peroxyformic acid pulping process when extractive distillation with NMP at 101.3 kPa is the first stage. SD is simple distillation and ED is extractive distillation with NMP. + is accepted and - is rejected.

------Method and 1.2 · minimum Product Lowest product Accepted(+)/ pressure number of trays flow/feed flows rejected(-) (kPa) % ------SD 101.3 319 FA (+) 07.92 ------

Even simple distillation at 101.3 kPa following extractive distillation seems to be an unfeasible second separation stage. No separation sequence for the peroxyformic acid process was, despite the results, rejected because in the actual process the specifications for the distillation columns are not necessarily as strict as in this study. With less strict specifications, the sequences might be accepted. Another reason for not rejecting any sequence was the fact that the reference process could not be totally rejected at this stage of the study.

3.3.4.6. Summary of the simulation results for the second separation stage

Feasible combinations of two separation stages for the feed compositions studied are listed in table 47.

Table 47. Feasible first two separation stages for various feed compositions. FA is formic acid, AA is acetic acid, W is water, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone.

------Feed composition Feasible separation methods and pressures (kPa) ------FA AA W 1. stage 2. stage %%% ------10 20 70 SD 20 SD 101.3 10 20 70 SD 20 SD 300 10 20 70 SD 101.3 SD 300 10 20 70 SD 300 SD 101.3 10 40 50 SD 101.3 SD 300 ______Table 47. Cont.

------Feed composition Feasible separation methods and pressures (kPa) ------FA AA W 1. stage 2. stage %%% ------10 40 50 SD 300 SD 101.3 10 30 60 SD 101.3 SD 300 10 30 60 SD 300 SD 101.3 10 10 80 SD 20 SD 300 20 40 40 SD 101.3 SD 300 20 40 40 SD 300 SD 101.3 20 30 50 SD 101.3 SD 300 20 30 50 SD 300 SD 101.3 20 20 60 SD 300 SD 101.3 20 10 70 SD 300 SD 101.3 30 60 10 SD 101.3 SD 20 30 60 10 SD 300 SD 20 30 60 10 SD 300 SD 101.3 30 30 40 SD 101.3 SD 300 30 30 40 SD 300 SD 101.3 30 30 40 ED 101.3 SD 101.3 30 20 50 SD 300 SD 101.3 30 20 50 ED 101.3 SD 101.3 30 10 60 SD 300 SD 101.3 40 50 10 SD 300 SD 20 40 50 10 SD 300 SD 101.3 40 20 40 SD 300 SD 101.3 40 20 40 ED 101.3 SD 101.3 40 10 50 SD 300 SD 101.3 60 20 20 SD 101.3 SD 300 60 10 30 SD 101.3 SD 300 70 10 20 SD 101.3 SD 300 80 10 10 SD 300 SD 101.3 ------

Simple distillation at 20 kPa seems to be a feasible second separation stage when the feed to the first stage contains 30 to 40 % formic acid and 10 % water. Distillation at atmospheric pressure or at 300 kPa are the most common separation methods feasible for use as the first two stages. All possible pairs of the four separation methods were accepted as potential separation sequences for the peroxyformic acid pulping process. 3.3.5. Shortcut simulation of the third separation stage

Simple distillation at the pressures 20 kPa, 101.3 kPa and 300 kPa and extractive distillation with n-methyl-2-pyrrolidone (NMP) at the pressure 101.3 kPa were studied as potential third separation stages. They were simulated using shortcut calculation methods in the PRO/II Flowsheet Simulation Program version 3.32 (Anon. 1991d) where a generalised Fenske method is used to predict product distributions. The minimum reflux ratio is determined using the Underwood method. The number of theoretical trays, the actual reflux rates, and condenser and reboiler duties are determined using a Gilliland-type correlation. The feasible third separation stages for each feed composition were chosen using the following rules: (1) The number of equilibrium stages was limited to 100 in order to keep the investment costs reasonable. (2) The most plentiful component should be the product of the separation stage. For the third separation stage it was no longer required that the flow of a product should be at least 10 % of the feed flow. The strict specifications used in the simulation of the first and second stages result in low flow rates of some components to the third stage. In the real separation process, the recovery rates may be lower, resulting in higher flow rates in the following stages. If a separation method according to the simulation results did not fulfil the requirements, it was rejected. Extractive distillation with NMP always removes water as the distillate. Therefore it was only considered as the separation method for those feed streams with water as the most plentiful component. The specifications used in simulations were: (1) The molar fraction of one component in a product is 0.98. (2) The recovery rate of one component into a product is 98 %.

3.3.5.1 Simple distillation at 20 kPa as the third separation stage

The simulation results used to either accept or reject simple distillation at 20 kPa as the third separation stage following the accepted first and second stages for the separation of formic acid-acetic acid-water mixtures are presented in table 48. Simple distillation at 20 kPa is not an acceptable third separation stage when simple distillation at 101.3 kPa is the first stage, because very high numbers of equilibrium stages are required for the separation. Simple distillation at 20 kPa can be accepted as the third separation stage after simple distillation at 300 kPa as the first stage and simple distillation at 101.3 kPa as the second stage only when the original feed contains 20 % formic acid, 20 % acetic acid and 60 % water or 40 % formic acid, 50 % acetic acid and 10 % water. Table 48. Simulation results of simple distillation at 20 kPa as the third separation stage when simple distillation at 300 kPa is an accepted first separation stage. Formic acid is FA, acetic acid is AA and water is W, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone. + is accepted and - is rejected.

------Original 2nd stage Number of Product Are further Accepted(+)/ feed and pressure trays at stages rejected(-) composition [kPa] 1.2·Rmin required? ------10 % FA 40 % AA SD 101.3 711 FA (+) NO - 50 % W

10 % FA 30 % AA SD 101.3 652 FA (+) NO - 60 % W

10 % FA 20 % AA SD 101.3 529 FA (+) NO - 70 % W

20 % FA 40 % AA SD 101.3 561 FA (+) NO - 40 % W

20 % FA 30 % AA SD 101.3 478 FA (+) NO - 50 % W

20 % FA 20 % AA SD 101.3 A THIRD SEPARATION STAGE IS NOT NEEDED 60 % W

20 % FA 10 % AA SD 101.3 793 FA (-) YES - 70 % W

30 % FA 60 % AA SD 101.3 NO SOLUTION WAS FOUND 10 % W

30 % FA 30 % AA SD 101.3 A THIRD SEPARATION STAGE IS NOT NEEDED 40 % W ______Table 48. Cont. ------Original 2nd stage Number of Product Are further Accepted(+)/ feed and pressure trays at stages rejected(-) composition [kPa] 1.2·Rmin required? ------30 % FA 20 % AA SD 101.3 A THIRD SEPARATION STAGE IS NOT NEEDED 50 % W

30 % FA 10 % AA SD 101.3 NO SOLUTION WAS FOUND 60 % W

40 % FA 50 % AA SD 101.3 44 FA (+) YES + 10 % W

40 % FA 20 % AA SD 101.3 945 FA (-) YES - 40 % W

40 % FA 10 % AA SD 101.3 836 FA (+) YES - 50 % W

80 % FA 10 % AA SD 101.3 813 FA (+) YES - 10 % W ------

Simple distillation at 20 kPa is not an acceptable third separation stage when it has been used also as the first stage, or when extractive distillation with n-methyl-2- pyrrolidone is used as the first stage, because the number of equilibrium stages needed for the separation is above 100.

3.3.5.2. Simple distillation at 101.3 kPa as the third separation stage

The simulation results used to either accept or reject simple distillation at 101.3 kPa as the third separation stage following the accepted first and second stages for the separation of formic acid-acetic acid-water mixtures are presented in tables 49...51. Table 49. Simulation results of simple distillation at 101.3 kPa as the third separation stage when simple distillation at 20 kPa is an accepted first separation stage. Formic acid is FA, acetic acid is AA and water is W, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone. + is accepted and - is rejected.

------Original 2nd stage Number of Product Are further Accepted(+)/ feed and pressure trays at stages rejected(-) composition [kPa] 1.2·Rmin required? ------10 % FA 20 % AA SD 300 13 FA (+) NO + 70 % W

10 % FA 10 % AA SD 300 11 FA (+) NO + 80 % W ------

Simple distillation at 101.3 kPa can be accepted as the third separation stage when simple distillation at 20 kPa is the first stage, simple distillation at 300 kPa is the second stage and the original feed contains 10 % formic acid and 10 or 20 % acetic acid. Simple distillation at 101.3 kPa can be accepted as the third separation stage after simple distillation at 300 kPa and 20 kPa as the first and second stages when the original feed contains 30 % formic acid, 60 % acetic acid and 10 % water or 40 % formic acid, 50 % acetic acid and 10 % water. Simple distillation at 101.3 kPa is an acceptable third separation stage for all those original feed compositions for which simple distillation at 101.3 kPa can be accepted as the first stage and a feasible second stage is found. When extractive distillation with n-methyl-2-pyrrolidone is used as the first separation stage the second stage is always simple distillation at 101.3 kPa. Therefore simple distillation at 101.3 kPa was not considered as the third separation stage. Table 50. Simulation results of simple distillation at 101.3 kPa as the third separation stage when simple distillation at 300 kPa is an accepted first separation stage. Formic acid is FA, acetic acid is AA and water is W, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone. + is accepted and - is rejected.

------Original 2nd stage Number of Product Are further Accepted(+)/ feed and pressure trays at stages rejected(-) composition [kPa] 1.2·Rmin required? ------30 % FA 60 % AA SD 20 14 W (+) NO + 10 % W

40 % FA 50 % AA SD 20 15 W (+) NO + 10 % W ------

Table 51. Simulation results of simple distillation at 101.3 kPa as the third separation stage when simple distillation at 101.3 kPa is an accepted first separation stage. Formic acid is FA, acetic acid is AA and water is W, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone. + is accepted and - is rejected.

------Original 2nd stage Number of Product Are further Accepted(+)/ feed and pressure trays at stages rejected(-) composition [kPa] 1.2·Rmin required? ------10 % FA 40 % AA SD 300 33 AA (+) YES + 50 % W

10 % FA 30 % AA SD 300 38 AA (+) YES + 60 % W

10 % FA 20 % AA SD 300 52 FA (+) NO + 70 % W

20 % FA 40 % AA SD 300 23 FA (+) NO + 40 % W ______Table 51. Cont.

------Original 2nd stage Number of Product Are further Accepted(+)/ feed and pressure trays at stages rejected(-) composition [kPa] 1.2·Rmin required? ------20 % FA 30 % AA SD 300 9 FA (+) NO + 50 % W

30 % FA 60 % AA SD 20 14 W (+) NO + 10 % W

30 % FA 30 % AA SD 300 14 FA (+) NO + 40 % W

60 % FA 20 % AA SD 300 48 AA (+) NO + 20 % W

60 % FA 10 % AA SD 300 31 FA (+) YES + 30 % W

70 % FA 10 % AA SD 300 48 FA (+) YES + 20 % W ------

3.3.5.3. Simple distillation at 300 kPa as the third separation stage

The simulation results used to either accept or reject simple distillation at 300 kPa as the third separation stage following the accepted first and second stages for the separation of formic acid-acetic acid-water mixtures are presented in tables 52...54. Simple distillation at 300 kPa can be accepted as the third separation stage when simple distillation at 20 kPa or 101.3 kPa is the first stage. When the first stage is extractive distillation with NMP, no solution was found. Table 52. Simulation results of simple distillation at 300 kPa as the third separation stage when simple distillation at 20 kPa is an accepted first separation stage. Formic acid is FA, acetic acid is AA and water is W, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone. + is accepted and - is rejected.

------Original 2nd stage Number of Product Are further Accepted(+)/ feed and pressure trays at stages rejected(-) composition [kPa] 1.2·Rmin required? ------10 % FA 20 % AA SD 101.3 21 FA (+) NO + 70 % W ------

Table 53. Simulation results of simple distillation at 300 kPa as the third separation stage when simple distillation at 101.3 kPa is an accepted first separation stage. Formic acid is FA, acetic acid is AA and water is W, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone. + is accepted and - is rejected.

------Original 2nd stage Number of Product Are further Accepted(+)/ feed and pressure trays at stages rejected(-) composition [kPa] 1.2·Rmin required? ------30 % FA 60 % AA SD 20 18 W (+) NO + 10 % W ------

Table 54. Simulation results of simple distillation at 300 kPa as the third separation stage when simple distillation at 300 kPa is an accepted first separation stage. Formic acid is FA, acetic acid is AA and water is W, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone. + is accepted and - is rejected.

------Original 2nd stage Number of Product Are further Accepted(+)/ feed and pressure trays at stages rejected(-) composition [kPa] 1.2·Rmin required? ------10 % FA 40 % AA SD 101.3 18 W (-) YES - 50 % W ______Table 54. Cont.

------Original 2nd stage Number of Product Are further Accepted(+)/ feed and pressure trays at stages rejected(-) composition [kPa] 1.2·Rmin required? ------10 % FA 30 % AA SD 101.3 22 W (-) YES - 60 % W

10 % FA 20 % AA SD 101.3 22 W (-) YES - 70 % W

20 % FA 40 % AA SD 101.3 8 W (-) YES - 40 % W

20 % FA 30 % AA SD 101.3 3 W (-) YES - 50 % W

20 % FA 20 % AA SD 101.3 A THIRD STAGE IS NOT REQUIRED 60 % W

20 % FA 10 % AA SD 101.3 46 W (-) YES - 70 % W

30 % FA 60 % AA SD 20 18 W (+) NO + 10 % W

30 % FA 60 % AA SD 101.3 20 W (+) NO + 10 % W

30 % FA 30 % AA SD 101.3 A THIRD STAGE IS NOT REQUIRED 40 % W

30 % FA 20 % AA SD 101.3 A THIRD STAGE IS NOT REQUIRED 50 % W ______Table 54. Cont.

------Original 2nd stage Number of Product Are further Accepted(+)/ feed and pressure trays at stages rejected(-) composition [kPa] 1.2·Rmin required? ------30 % FA 10 % AA SD 101.3 50 W (-) YES - 60 % W

40 % FA 50 % AA SD 20 18 W (+) NO + 10 % W

40 % FA 50 % AA SD 101.3 169 AA (-) YES - 10 % W

40 % FA 20 % AA SD 101.3 50 AA (+) NO + 40 % W

40 % FA 10 % AA SD 101.3 74 AA (-) NO - 50 % W

80 % FA 10 % AA SD 101.3 207 W (-) YES - 10 % W ------

When the original feed contains 30 % formic acid, 60 % acetic acid and 10 % water, simple distillation at 300 kPa can be accepted as the third separation stage when the first stage is simple distillation at 300 kPa and the second stage is simple distillation at 20 or 101.3 kPa. When the second stage is simple distillation at 20 kPa, simple distillation at 300 kPa can be accepted as the third stage for the original feed composition of 40 % formic acid, 50 % acetic acid and 10 % water. When the second stage is simple distillation at 101.3 kPa, simple distillation at 300 kPa can be accepted as the third stage for the original feed composition of 40 % formic acid, 20 % acetic acid and 40 % water. 3.3.5.4. Extractive distillation with n-methyl-2-pyrrolidone at 101.3 kPa as the third separation stage

The simulation results used to either accept or reject extractive distillation with n-methyl- 2-pyrrolidone (NMP) at 101.3 kPa as the third separation stage following the accepted first and second stages for the separation of formic acid-acetic acid-water mixtures are presented in tables 55 and 56. Extractive distillation was considered as the third separation stage only when water was the most plentiful component in the feed stream of the stage.

Table 55. Simulation results of extractive distillation with NMP at 101.3 kPa as the third separation stage when simple distillation at 101.3 kPa is an accepted first separation stage. Formic acid is FA, acetic acid is AA and water is W, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone. + is accepted and - is rejected.

------Original 2nd stage Number of Product Are further Accepted(+)/ feed and pressure trays at stages rejected(-) composition [kPa] 1.2·Rmin required? ------30 % FA 60 % AA SD 20 13 W (+) NO + 10 % W ------

Extractive distillation with NMP at 101.3 kPa can be accepted as the third separation stage following simple distillation at 101.3 kPa and 20 kPa as the first and second stages only when the original feed contains 30 % formic acid, 60 % acetic acid and 10 % water. Extractive distillation with NMP at 101.3 kPa can be accepted as the third separation stage when the original feed contains 30 % formic acid, 60 % acetic acid and 10 % water and simple distillation at 300 kPa is the first separation stage, followed by simple distillation at 20 kPa or 101.3 kPa as the second stage. It may also be accepted when simple distillation at 300 kPa and 20 kPa are the first and second stages, and the original feed contains 40 % formic acid, 50 % acetic acid and 10 % water. When the first separation stage was simple distillation at 20 kPa, or extractive distillation with NMP at 101.3 kPa, the extractive distillation method was not found to be a feasible third stage for any original feed compositions. Using extractive distillation as the third stage would have required the use of at least one additional separation stage. Table 56. Simulation results of extractive distillation with NMP at 101.3 kPa as the third separation stage when simple distillation at 300 kPa is an accepted first separation stage. Formic acid is FA, acetic acid is AA and water is W, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone. + is accepted and - is rejected.

------Original 2nd stage Number of Product Are further Accepted(+)/ feed and pressure trays at stages rejected(-) composition [kPa] 1.2·Rmin required? ------30 % FA 60 % AA SD 20 13 W (+) NO + 10 % W

30 % FA 60 % AA SD 101.3 13 W (+) NO + 10 % W

40 % FA 50 % AA SD 20 14 W (+) NO + 10 % W ------

3.3.5.5. The third separation stages for the peroxyformic acid pulping process

The composition of the liquor fed to the separation sequence in the peroxyformic acid pulping process has been estimated to be 12.1 % formic acid, 0.2 % acetic acid and 87.7 % water. Simple distillation at the pressures 20, 101.3 and 300 kPa and extractive distillation with n-methyl-2-pyrrolidone (NMP) at 101.3 kPa were simulated as the third separation stage for this first stage feed composition. Extractive distillation was considered as the third stage only when water was the most plentiful component in the feed to the stage. Shortcut calculation methods were used. Results are shown in tables 57 to 60. Only simple distillation at 20 kPa following simple distillation at 101.3 kPa was rejected because no solution was found. All those third stages that were considered can be accepted when simple distillation at 101.3 is the first separation stage. A third separation stage is not needed when the second stage is simple distillation at 101.3 kPa. When the second stage is simple distillation at 20 kPa, all the stages considered can be accepted. Simple distillation at 20 kPa or at 300 kPa seems to be a feasible third separation stage following extractive distillation with NMP at 101.3 kPa and simple distillation at 101.3 kPa as the first two stages. Table 57. Simulation results for the third separation stage in the peroxyformic acid pulping process when simple distillation at 20 kPa is the first stage. SD is simple distillation and ED is extractive distillation with NMP. + is accepted and - is rejected.

------Second stage Third stage Number of Product Are further Accepted(+)/ and pressure and pressure stages at stages rejected(-) (kPa) (kPa) 1.2·Rmin needed? ------SD 101.3 SD 20 NO SOLUTION WAS FOUND SD 101.3 SD 300 32 W (+) YES + SD 101.3 ED 101.3 11 W (+) YES + SD 300 SD 20 26 FA (+) YES + SD 300 SD 101.3 25 FA (+) YES + ------

Table 58. Simulation results for the third separation stage in the peroxyformic acid pulping process when simple distillation at 101.3 kPa is the first stage. SD is simple distillation and ED is extractive distillation with NMP. + is accepted and - is rejected.

------Second stage Third stage Number of Product Are further Accepted(+)/ and pressure and pressure stages at stages rejected(-) (kPa) (kPa) 1.2·Rmin needed? ------SD 20 SD 101.3 24 W (+) YES + SD 20 SD 300 26 W (+) YES + SD 20 ED 101.3 13 W (+) YES + SD 300 SD 20 29 FA (+) YES + SD 300 SD 101.3 28 FA (+) YES + ------

Table 59. Simulation results for the third separation stage in the peroxyformic acid pulping process when simple distillation at 300 kPa is the first stage. SD is simple distillation and ED is extractive distillation with NMP. + is accepted and - is rejected.

------Second stage Third stage Number of Product Are further Accepted(+)/ and pressure and pressure stages at stages rejected(-) (kPa) (kPa) 1.2·Rmin needed? ------SD 20 SD 101.3 26 W (+) YES + SD 20 SD 300 26 W (+) YES + SD 20 ED 101.3 12 W (+) YES + SD 101.3 A THIRD STAGE IS NOT NEEDED ------Table 60. Simulation results for the third separation stage in the peroxyformic acid pulping process when extractive distillation with NMP at 101.3 kPa is the first stage. SD is simple distillation and ED is extractive distillation with NMP. + is accepted and - is rejected.

------Second stage Third stage Number of Product Are further Accepted(+)/ and pressure and pressure stages at stages rejected(-) (kPa) (kPa) 1.2·Rmin needed? ------SD 101.3 SD 20 6 FA (+) YES + SD 101.3 SD 300 8 W (+) YES + ------

3.3.5.6. Summary of the simulation results for the third separation stage

Feasible combinations of three separation stages for the feed compositions studied are listed in table 61. Feasible three-stage separation processes for the original feed composition of the peroxyformic acid process are presented in table 62.

Table 61. Feasible three-stage separation processes for various feed compositions. FA is formic acid, AA is acetic acid, W is water, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone.

------Feed composition Feasible separation methods and pressures Are further (kPa) stages ------needed? FA AA W 1. stage 2. stage 3. stage %%% ------10 40 50 SD 101.3 SD 300 SD 101.3 YES 10 30 60 SD 101.3 SD 300 SD 101.3 YES 10 20 70 SD 20 SD 300 SD 101.3 NO 10 20 70 SD 20 SD 101.3 SD 300 NO 10 20 70 SD 101.3 SD 300 SD 101.3 NO 10 10 80 SD 20 SD 300 SD 101.3 NO 20 20 60 SD 300 SD 101.3 - NO 20 40 40 SD 101.3 SD 300 SD 101.3 NO 20 30 50 SD 101.3 SD 300 SD 101.3 NO 30 60 10 SD 101.3 SD 20 SD 101.3 NO 30 60 10 SD 101.3 SD 20 SD 300 NO 30 60 10 SD 101.3 SD 20 ED 101.3 NO ______Table 61. Cont.

------Feed composition Feasible separation methods and pressures Are further (kPa) stages ------needed? FA AA W 1. stage 2. stage 3. stage %%% ------30 60 10 SD 300 SD 20 SD 101.3 NO 30 60 10 SD 300 SD 20 SD 300 NO 30 60 10 SD 300 SD 20 ED 101.3 NO 30 60 10 SD 300 SD 101.3 SD 300 NO 30 60 10 SD 300 SD 101.3 ED 101.3 NO 30 30 40 SD 101.3 SD 300 SD 101.3 NO 30 30 40 SD 300 SD 101.3 - NO 30 20 50 SD 300 SD 101.3 - NO 40 50 10 SD 300 SD 20 SD 101.3 NO 40 50 10 SD 300 SD 20 SD 300 NO 40 50 10 SD 300 SD 20 ED 101.3 NO 40 50 10 SD 300 SD 101.3 SD 20 YES 40 20 40 SD 300 SD 101.3 SD 300 NO 60 20 20 SD 101.3 SD 300 SD 101.3 NO 60 10 30 SD 101.3 SD 300 SD 101.3 YES 70 10 20 SD 101.3 SD 300 SD 101.3 YES ------

Several feasible three-stage separation processes can be found when the feed stream to the process contains 10...40 % formic acid. At higher formic acid contents, only few possible separation processes are found. When the feed stream contains 30 % formic acid and 20 or 30 % acetic acid, or 20 % both formic and acetic acids and is distilled at 300 kPa followed by a distillation at atmospheric pressure, a third stage is not needed. When the original feed contains 10 % formic acid and 30 or 40 % acetic acid, a fourth separation stage has to be added to the SD101.3+SD300+SD101.3 sequence in order to get a formic acid product. When the original feed contains 40 % formic acid and 50 % acetic acid, a fourth separation stage has to be added to the SD300+SD101.3+SD20 sequence in order to get a water rich product. When the original feed contains 60 % formic acid and 10 % acetic acid, or 70 % formic acid and 10 % acetic acid, a fourth separation stage has to be added to the SD101.3+SD300+SD101.3 sequence in order to get an acetic acid rich product. Further separation stages for the peroxyformic acid process are needed because no acetic acid product is produced by the three stages. The acetic acid product may be extracted from a product stream containing mostly water. Even when the first two separation stages are simple distillations at 300 kPa and 101.3 kPa, an extraction stage is needed. When extractive distillation is the third separation stage, a fourth distillation stage is needed in order to recycle the mass transfer agent. Table 62. Feasible three-stage separation processes for the peroxyformic acid pulping process. FA is formic acid, AA is acetic acid, W is water, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone.

------Separation stages and pressures [kPa] Are further ------stages needed? 1. stage 2. stage 3. stage ------SD 20 SD 101.3 SD 300 YES SD 20 SD 101.3 ED 101.3 YES SD 20 SD 300 SD 20 YES SD 20 SD 300 SD 101.3 YES SD 101.3 SD 20 SD 101.3 YES SD 101.3 SD 20 SD 300 YES SD 101.3 SD 20 ED 101.3 YES SD 101.3 SD 300 SD 20 YES SD 101.3 SD 300 SD 101.3 YES SD 300 SD 20 SD 101.3 YES SD 300 SD 20 SD 300 YES SD 300 SD 20 ED 101.3 YES SD 300 SD 101.3 - NO ED 101.3 SD 101.3 SD 20 YES ED 101.3 SD 101.3 SD 300 YES ------

3.3.6. Shortcut simulation of the fourth separation stage

Simple distillation at the pressures 20 kPa, 101.3 kPa and 300 kPa was studied as a potential fourth separation stage. They were simulated using shortcut calculation methods in the PRO/II Flowsheet Simulation Program version 3.32 (Anon. 1991d) where a generalised Fenske method is used to predict product distributions. The minimum reflux ratio is determined using the Underwood method. The number of theoretical trays, the actual reflux rates, and condenser and reboiler duties are determined using a Gilliland-type correlation. The feasible fourth separation stages for each feed composition were chosen using the following rules: (1)The number of equilibrium stages was limited to 100 in order to keep the investment costs reasonable. (2)The most plentiful component should be the product of the separation stage. (3)No further separation stages should be needed. If a separation method according to the simulation results did not fulfil the requirements, it was rejected. Extractive distillation with NMP was not considered as the fourth separation stage because in order to recycle the mass transfer agent, a fifth stage would be needed. Specifications used in the simulations were: (1)The molar fraction of one component in a product is 0.98. (2)The recovery rate of one component into a product is 98 %.

3.3.6.1. Simple distillation at 20 kPa as the fourth separation stage

Using simple distillation as the fourth separation stage results in columns with several hundreds of equilibrium stages. Therefore it was rejected as the fourth stage.

3.3.6.2. Simple distillation at 101.3 kPa as the fourth separation stage

The simulation results used to either accept or reject simple distillation at 101.3 kPa as the fourth stage in a process separating formic acid-acetic acid-water mixtures are presented in table 63. Simple distillation at 101.3 kPa can be accepted as the fourth separation stage when the original feed contains 30 % formic acid and 60 % acetic acid and the three previous stages are simple distillation at 300 kPa, simple distillation at 20 kPa and extractive distillation with NMP at 101.3 kPa. It may also be accepted when the original feed contains 40 % formic acid and 50 % acetic acid, and the first three stages are simple distillations at 300 kPa, 101.3 kPa and 20 kPa.

Table 63. Simulation results of simple distillation at 101.3 kPa as the fourth separation stage. Formic acid is FA, acetic acid is AA and water is W, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone. + is accepted and - is rejected.

------Original 1st, 2nd and Number of Product Are further Accepted(+)/ feed 3rd stages trays at stages rejected(-) composition and pressures 1.2·Rmin required? [kPa] ------30 % FA SD101.3 60 % AA SD 20 1747 FA (+) NO - 10 % W ED 101.3

30 % FA SD 300 60 % AA SD 20 11 FA (+) NO + 10 % W ED 101.3 ------Table 63. Cont.

------Original 1st, 2nd and Number of Product Are further Accepted(+)/ feed 3rd stages trays at stages rejected(-) composition and pressures 1.2·Rmin required? [kPa] ------30 % FA SD 300 60 % AA SD 101.3 576 FA (+) NO - 10 % W ED 101.3

40 % FA SD 300 50 % AA SD 101.3 14 W (+) NO + 10 % W SD 20

40 % FA SD 300 50 % AA SD 20 1254 FA (+) YES - 10 % W ED 101.3 ------

3.3.6.3. Simple distillation at 300 kPa as the fourth separation stage

The simulation results used to either accept or reject simple distillation at 300 kPa as the fourth stage in a process separating formic acid-acetic acid-water mixtures are presented in table 64.

Table 64. Simulation results of simple distillation at 300 kPa as the fourth separation stage. Formic acid is FA, acetic acid is AA and water is W, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone. + is accepted and - is rejected.

------Original 1st, 2nd and Number of Product Are further Accepted(+)/ feed 3rd stages trays at stages rejected(-) composition and pressures 1.2·Rmin required? [kPa] ------10 % FA SD 101.3 40 % AA SD 300 57 FA (+) NO + 50 % W SD 101.3 ------Table 64. Cont.

------Original 1st, 2nd and Number of Product Are further Accepted(+)/ feed 3rd stages trays at stages rejected(-) composition and pressures 1.2·Rmin required? [kPa] ------10 % FA SD 101.3 30 % AA SD 300 60 FA (+) NO + 60 % W SD 101.3

40 % FA SD 300 50 % AA SD 101.3 18 W (+) NO + 10 % W SD 20

60 % FA SD 101.3 10 % AA SD 300 56 AA (-) NO - 30 % W SD 101.3

70 % FA SD 101.3 10 % AA SD 300 19 AA (+) NO + 20 % W SD 101.3 ------

Simple distillation at 300 kPa was rejected when the original feed contained 60 % formic acid and 10 % acetic acid and the three previous stages were simple distillations at 101.3 kPa, 300 kPa and 101.3 kPa. The reason for the rejection was that the fourth stage gave an acetic acid product, and acetic acid was not the most plentiful component in the feed of the stage.

3.3.6.4. The fourth separation stages for the peroxyformic acid pulping process

The composition of the liquor fed to the separation sequence in the peroxyformic acid pulping process has been estimated to be 12.1 % formic acid, 0.2 % acetic acid and 87.7 % water. Simple distillation at the pressures 20, 101.3 and 300 kPa were simulated as the fourth separation stage for this first stage feed composition. Extractive distillation was not considered as the fourth because its use would lead to the need for a fifth separation stage in order to recycle the mass transfer agent. Those separation cycles needing a fourth stage were rejected mainly because there was too high a number of equilibrium stages, and because an acetic acid product was not obtained. 3.3.6.5. Summary of the simulation results for the fourth separation stage

Feasible combinations of four separation stages for the original feed compositions studied are listed in table 65. No feasible four-stage separation processes for the peroxyformic acid process were found.

Table 65. Feasible four-stage separation processes for various feed compositions. FA is formic acid, AA is acetic acid, W is water, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone.

------Feed composition Feasible separation methods and pressures (kPa) ------FA AA W 1. stage 2. stage 3. stage 4. stage %%% ------10 40 50 SD 101.3 SD 300 SD 101.3 SD 300 10 30 60 SD 101.3 SD 300 SD 101.3 SD 300 30 60 10 SD 300 SD 20 ED 101.3 SD 101.3 40 50 10 SD 300 SD 101.3 SD 20 SD 101.3 40 50 10 SD 300 SD 101.3 SD 20 SD 300 70 10 20 SD 101.3 SD 300 SD 101.3 SD 300 ------

3.3.7. The recycle stream

In most separation sequences studied one of the products from the last separation stage is not pure enough and has to be recycled to the first separation stage. This recycle stream was added to the simulation models of the sequences accepted during the previous simulation stages. Feasible separation sequences were chosen using the following rules: (1) The number of equilibrium stages for one separation stage was limited to 100 in order to keep the investment costs reasonable. (2) The energy consumption of the sequence should be below the average consumption of the sequences. Specifications used in the simulations were: (1) The molar fraction of one component in a product is 0.98. (2) The recovery rate of one component into a product is 98 %. The simulation results are presented in table 66. Table 66. Simulation results for separation sequences with recycling. Formic acid is FA, acetic acid is AA and water is W, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone.

------Original 1st, 2nd 3rd Number of The energy consumption feed and 4th stages trays at of the separation composition and pressures 1.2·Rmin sequence [kPa] [MJ/kmol of feed] ------10 % FA SD 101.3 60 40 % AA SD 300 21 10.33·106 50 % W SD 101.3 33 SD 300 56

10 % FA SD 101.3 49 30 % AA SD 300 25 10.49·106 60 % W SD 101.3 40 SD 300 60

10 % FA SD 20 55 20 % AA SD 300 48 54.74 70 % W SD 101.3 34

10 % FA SD 20 67 20 % AA SD 101.3 41 67.41 70 % W SD 300 40

10 % FA SD 101.3 53 20 % AA SD 300 21 10.80·106 70 % W SD 101.3 60

10 % FA SD 20 41 10 % AA SD 300 54 44.19 80 % W SD 101.3 26

20 % FA SD 101.3 51 40 % AA SD 300 46 242.37 40 % W SD 101.3 42

20 % FA SD 101.3 46 30 % AA SD 300 51 236.84 50 % W SD 101.3 22

30 % FA SD 101.3 74 60 % AA SD 20 54 117.72 10 % W SD 101.3 14 ______Table 66. Cont.

------Original 1st, 2nd 3rd Number of The energy consumption feed and 4th stages trays at of the separation composition and pressures 1.2·Rmin sequence [kPa] [MJ/kmol of feed] ------30 % FA SD 101.3 71 60 % AA SD 20 54 158.13 10 % W SD 300 17

30 % FA SD 300 69 60 % AA SD 20 53 118.62 10 % W SD 101.3 14

30 % FA SD 300 66 60 % AA SD 20 54 155.97 10 % W SD 300 17

30 % FA SD 300 65 60 % AA SD 101.3 65 15.26·106 10 % W SD 300 18

30 % FA SD 300 69 60 % AA SD 20 54 158.11 10 % W ED 101.3 9 SD 101.3 11

30 % FA SD 101.3 280 30 % AA SD 300 49 286.28 40 % W SD 101.3 25

40 % FA SD 300 75 50 % AA SD 20 37 136.31 10 % W SD 101.3 15

40 % FA SD 300 71 50 % AA SD 20 37 174.60 10 % W SD 300 18

40 % FA SD 300 75 50 % AA SD 101.3 21 133.05 10 % W SD 20 46 SD 101.3 14 ______Table 66. Cont.

------Original 1st, 2nd 3rd Number of The energy consumption feed and 4th stages trays at of the separation composition and pressures 1.2·Rmin sequence [kPa] [MJ/kmol of feed] ------40 % FA SD 300 71 50 % AA SD 101.3 21 177.17 10 % W SD 20 46 SD 300 18

70 % FA SD 101.3 49 10 % AA SD 300 42 272.55 20 % W SD 101.3 50 SD 300 21

12.1 % FA SD 20 27 00.2 % AA SD 101.3 24 188.66 87.7 % W

12.1 % FA SD 101.3 21 00.2 % AA SD 20 27 0.92·106 87.7 % W

12.1 % FA SD 300 27 00.2 % AA SD 20 22 114.15 87.7 % W

12.1 % FA SD 20 21 00.2 % AA SD 101.3 15 5.41·106 87.7 % W ED 101.3 11 SD 101.3 57

12.1 % FA SD 101.3 17 00.2 % AA SD 20 33 1.95·106 87.7 % W ED 101.3 10 SD 101.3 143 ------

Three separation sequences with only two stages have been found during previous simulations. The energy consumptions of those sequences are presented in table 67. Table 67. The energy consumptions of separation sequences with two stages. Formic acid is FA, acetic acid is AA and water is W, SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone.

------Original 1st and 2nd The energy consumption feed stages and of the separation composition pressures sequence [kPa] [MJ/kmol of feed] ------20 % FA SD 300 20 % AA SD 101.3 177.87 60 % W

30 % FA SD 300 30 % AA SD 101.3 161.30 40 % W

30 % FA SD 300 20 % AA SD 101.3 246.70 50 % W ------

When energy consumptions higher than 1.0 TJ per kmol of feed are not taken into account, the average consumption calculated from the data presented in tables 66 and 67 is 162.99 MJ per kmol of feed. Those sequences having an energy consumption lower than the average consumption were chosen for simulations with rigorous calculation methods. If several sequences resulted in feasible energy consumptions for the same feed composition, only the sequence with the lowest consumption was chosen. Table 68 lists the sequences selected for further examination.

Table 68. Separation sequences selected for simulations with rigorous calculation methods. Formic acid is FA, acetic acid is AA and water is W. SD is simple distillation and ED is extractive distillation with n-methyl-2-pyrrolidone.

------Original Separation stages and pressures [kPa] feed ------compositions First Second Third Fourth ______10 % FA 20 % AA SD 20 SD 300 SD 101.3 - 70 % W ______Table 68. Cont.

------Original Separation stages and pressures [kPa] feed ------compositions First Second Third Fourth ______10 % FA 10 % AA SD 20 SD 300 SD 101.3 - 80 % W

30 % FA 30 % AA SD 300 SD 101.3 - - 40 % W

30 % FA 60 % AA SD 101.3 SD 20 SD 101.3 - 10 % W

40 % FA 50 % AA SD 300 SD 101.3 SD 20 SD 101.3 10 % W

12.1 % FA 00.2 % AA SD 300 SD 20 - - 87.7 % W ------

3.3.8. Rigorous simulations

Separation sequences for the feed streams shown in table 68 were designed and simulated using rigorous calculation methods. The PRO/II 3.32 Flowsheet Simulation Program (Anon. 1991d) was used. The sequences used in shortcut simulations could not be solved using rigorous methods. Therefore the sequences were redesigned using the following specifications: (1) The first distillation stage operates at the same pressure as in the sequence designed using shortcut methods. (2) The product purity is 98 %. (3) The highest possible recovery rate is used. (4) The feed flow rate is 100 kmol/h. The columns were chosen to be packed in order to minimize the risk of formic acid decomposition. Ceramic Super Intalox saddles with a nominal packing diameter of 50.8 mm and an HETP of 88.9 cm were used as the packing material. Heat exchangers were added to the separation sequences in order to raise the feed stream temperature of each column to its bubble point at the column pressure. Also pumps raising the feed stream pressures to the corresponding column pressure were added.

3.3.8.1. Feed containing 10 % formic acid, 20 % acetic acid and 70 % water

When the first distillation stage operates at 20 kPa, the feed stream containing 10 % formic acid, 20 % acetic acid and 70 % water can be separated using the sequence shown in figure 9.

H2O H2O HCOOH D2 D1 D3 D4

20 FH2 300 101 FH4 300 F kPa kPa kPa kPa

FH1 C1 C2 C3 C4 CH3COOH

B1 B2 B3 B4

CB1 CB2 CB3 CB4

Fig. 9. Separation sequence for a feed stream containing 10 % formic acid, 20 % acetic acid and 70 % water.

Column data, thermal energy consumptions and stream compositions are presented in tables 69, 70 and 71.

Table 69. Column data for the separation sequence shown in figure 9.

------Column 1 Column 2 Column 3 Column 4 ------Pressures (kPa) Top 20 300 101 300 Bottom 23 305 106 303 Temperatures (K) Top 329.2 407.0 374.0 419.8 Bottom 336.3 422.2 384.8 430.6 Equilibrium stages 28 41 30 15 Feed trays (from top) 12 12, 26 22 4 Reflux ratio 2.7 4.6 30.0 10.8 ______Table 69. Cont.

------Column 1 Column 2 Column 3 Column 4 ------Flow rates (kmol/h) Total feed 100.00 97.64 60.50 51.45 Distillate 34.29 37.13 9.06 31.95 Bottom 65.71 60.50 51.45 19.50 Specifications 1: Stream D1 D2 D3 B4 Component H2OH2O HCOOH CH3COOH Type fraction fraction fraction fraction Value 0.98 0.98 0.98 0.98 2: Stream D1 D2 D3 B4 Component H2OH2O HCOOH CH3COOH Type recovery recovery recovery recovery Value 0.48 0.90 0.39 0.57 Packed height (m) 23.11 34.67 24.89 11.56 Column inner diameter (m) 1.4 1.2 1.4 1.4 ------

Table 70. Thermal energy consumption of the separation sequence shown in figure 9.

------Unit Thermal energy consumption ------GJ/h MW ------FH1 0.346 0.096 FH2 0.520 0.144 FH4 0.283 0.079 CB1 5.377 1.494 CB2 8.065 2.240 CB3 5.657 1.571 CB4 9.150 2.542 29.40 8.166 ------Table 71. Stream compositions in the separation sequence shown in figure 9.

------Stream Flow rate HCOOH CH3COOH H2O kmol/h % % % ------F 100.0000 10.0000 20.0000 70.0000 B1 65.7091 14.7745 29.8373 55.3882 D1 34.2909 0.8510 1.1496 97.9995 B2 60.5046 37.6178 55.6995 6.6827 D2 37.1333 1.1308 0.8693 97.9998 B3 51.4468 26.9869 65.1541 7.8590 D3 9.0578 98.0000 1.9984 0.0016 B4 19.4960 1.9765 98.0000 0.0235 D4 31.9508 42.2479 45.1119 12.6401 ------

3.3.8.2. Feed containing 10 % formic acid, 10 % acetic acid and 80 % water

When the first distillation stage operates at 20 kPa, the feed stream containing 10 % formic acid, 10 % acetic acid and 80 % water can be separated using the sequence shown in figure 10.

H2 O H2O HCOOH D1 D2 D3 D4

FH2 FH4 F 20 300 101 300 kPa kPa kPa kPa

FH1 C1 C2 C3 C4 CH3COOH

B1 B2 B3 B4

CB1 CB2 CB3 CB4

Fig. 10. Separation sequence for a feed stream containing 10 % formic acid, 10 % acetic acid and 80 % water.

Column data, thermal energy consumptions and stream compositions are presented in tables 72, 73 and 74. Table 72. Column data for the separation sequence shown in figure 10.

------Column 1 Column 2 Column 3 Column 4 ------Pressures (kPa) Top 20 300 101 300 Bottom 25 305 105 302 Temperatures (K) Top 329.2 407.0 374.0 417.6 Bottom 337.7 419.4 382.9 430.6 Equilibrium stages 28 42 30 17 Feed trays (from top) 16 12, 29 22 5 Reflux ratio 1.1 2.3 20.0 18.7 Flow rates (kmol/h) Total feed 100.00 88.54 46.09 36.98 Distillate 39.18 42.44 9.12 27.73 Bottom 60.82 46.09 36.98 9.25 Specifications 1: Stream D1 D2 D3 B4 Component H2OH2O HCOOH CH3COOH Type fraction fraction fraction fraction Value 0.98 0.98 0.98 0.98 2: Stream D1 D2 D3 B4 Component H2OH2O HCOOH CH3COOH Type recovery recovery recovery recovery Value 0.48 0.87 0.39 0.54 Packed height (m) 23.11 35.56 24.89 13.34 Column inner diameter (m) 0.9 0.95 1.2 1.8 ------

Table 73. Thermal energy consumption of the separation sequence shown in figure 10.

------Unit Thermal energy consumption ------GJ/h MW ------FH1 0.319 0.089 FH2 0.420 0.117 FH4 0.180 0.050 CB1 3.536 0.982 CB2 5.393 1.498 CB3 3.852 1.070 CB4 14.012 3.892 27.712 7.698 ------Table 74. Stream compositions in the separation sequence shown in figure 10.

------Stream Flow rate HCOOH CH3COOH H2O kmol/h % % % ------F 100.0000 10.0000 10.0000 80.0000 B1 60.8164 15.6794 15.9179 68.4027 D1 39.1837 1.1851 0.8149 98.0000 B2 46.0943 49.6904 36.8259 13.4836 D2 42.4436 0.9846 1.0155 97.9999 B3 36.9793 37.7826 45.4110 16.8064 D3 9.1151 97.9999 1.9967 0.0034 B4 9.2526 1.9721 97.9994 0.0285 D4 27.7268 49.7327 27.8621 22.4053 ------

3.3.8.3. Feed containing 30 % formic acid, 30 % acetic acid and 40 % water

When the first distillation stage operates at 300 kPa, the feed stream containing 30 % formic acid, 30 % acetic acid and 40 % water can be separated using the sequence shown in figure 11.

H2O HCOOH D1 D2 D3

FH1 HCOOH F 300 20 300 kPa kPa FH3 kPa

C1 C2 C3 FH2 B1 B2 B3

CB1 CB2 CB3 CH3COOH

Fig. 11. Separation sequence for a feed stream containing 30 % formic acid, 30 % acetic acid and 40 % water. Column data, thermal energy consumptions and stream compositions are presented in tables 75, 76 and 77.

Table 75. Column data for the separation sequence shown in figure 11.

------Column 1 Column 2 Column 3 ------Pressures (kPa) Top 300 20 300 Bottom 301 22 306 Temperatures (K) Top 407.0 327.6 414.2 Bottom 418.9 335.1 431.6 Equilibrium stages 15 19 45 Feed trays (from top) 10, 11 8 23 Reflux ratio 11.8 5.9 48.7 Flow rates (kmol/h) Total feed 162.51 122.92 60.42 Distillate 39.59 60.42 31.28 Bottom 122.92 62.51 29.13 Specifications 1: Stream D1 D2 B3 Component H2O HCOOH CH3COOH CH3COOH Type fraction fraction fraction Value 0.98 0.98 0.98 2: Stream D1 D2 D3 Component H2O HCOOH HCOOH CH3COOH Type recovery recovery recovery Value 0.60 0.61 0.999 Packed height (m) 11.56 15.11 38.23 Column inner diameter (m) 1.8 2.4 2.8 ------Table 76. Thermal energy consumption of the separation sequence shown in figure 11.

------Unit Thermal energy consumption ------GJ/h MW ------FH1 1.294 0.359 FH2 0.557 0.155 FH3 0.725 0.201 CB1 19.523 5.423 CB2 7.929 2.203 CB3 34.258 9.516 64.286 17.857 ------

Table 77. Stream compositions in the separation sequence shown in figure 11.

------Stream Flow rate HCOOH CH3COOH H2O kmol/h % % % ------F 100.0000 30.0000 30.0000 40.0000 B1 122.9230 38.1681 40.7926 21.0393 D1 39.5852 1.2493 0.7510 97.9997 B2 62.5071 27.8506 32.7078 39.4416 D2 60.4158 48.8426 49.1574 2.0000 B3 29.1349 0.1022 99.8978 0.0000 D3 31.2812 94.2401 1.8970 3.8629 ------

3.3.8.4. Feed containing 30 % formic acid, 60 % acetic acid and 10 % water

When the first distillation stage operates at 101 kPa, the feed stream containing 30 % formic acid, 60 % acetic acid and 10 % water can be separated using the sequence shown in figure 12. HCOOH FH1 D1 D2 D3 D4 F FH4 H2O 101 20 FH3 101 101 kPa kPa kPa kPa

C1 C2 C3 C4

B1 B2 B3 B4

CB1 CH 3COOH CB2 CB3 CB4

Fig. 12. Separation sequence for a feed stream containing 30 % formic acid, 60 % acetic acid and 10 % water.

Column data, thermal energy consumptions and stream compositions are presented in tables 78, 79 and 80.

Table 78. Column data for the separation sequence shown in figure 12.

------Column 1 Column 2 Column 3 Column 4 ------Pressures (kPa) Top 101 20 101 101 Bottom 104 23 106 102 Temperatures (K) Top 381.7 327.7 374.0 373.5 Bottom 391.6 336.2 382.7 380.7 Equilibrium stages 16 20 22 14 Feed trays (from top) 5, 6, 10 8 21 13 Reflux ratio 3.7 7.5 36.0 5.5 ______Table 78. Cont.

------Column 1 Column 2 Column 3 Column 4 ------Flow rates (kmol/h) Total feed 551.54 491.10 307.07 184.03 Distillate 491.10 307.07 29.86 10.29 Bottom 60.44 184.03 277.21 173.74 Specifications 1: Stream B1 D3 D4 Component CH3COOH HCOOH H2O Type fraction reflux ratio fraction fraction Value 0.98 7.5 0.98 0.98 2: Stream B1 D2 D3 D4 Component CH3COOH HCOOH HCOOH H2O CH3COOH Type recovery recovery recovery recovery Value 0.22 0.75 0.20 0.12 Packed height (m) 12.45 16.00 17.78 10.67 Column inner diameter (m) 4.4 5.6 2.6 0.75 ------

Table 79. Thermal energy consumption of the separation sequence shown in figure 12.

------Unit Thermal energy consumption ------GJ/h MW ------FH1 1.090 0.303 FH3 2.069 0.575 FH4 0.873 0.243 CB1 59.268 16.463 CB2 54.750 15.208 CB3 23.456 6.516 CB4 2.660 0.739 144.166 40.047 ------Table 80. Stream compositions in the separation sequence shown in figure 12.

------Stream Flow rate HCOOH CH3COOH H2O kmol/h % % % ------F 100.0000 30.0000 60.0000 10.0000 B1 60.4424 1.9917 97.9994 0.0089 D1 491.0987 38.6851 42.7652 18.5498 B2 184.0312 23.7313 30.6080 45.6607 D2 307.0680 47.6471 50.0513 2.3017 B3 277.2068 42.2228 55.2276 2.5496 D3 29.8612 98.0011 1.9985 0.0004 B4 173.7418 25.1230 32.3160 42.5610 D4 10.2894 0.2324 1.7678 97.9999 ------

3.3.8.5. Feed containing 40 % formic acid, 50 % acetic acid and 10 % water

When the first distillation stage operates at 300 kPa, the feed stream containing 40 % formic acid, 50 % acetic acid and 10 % water can be separated using the sequence shown in figure 13.

HCOOH H2O FH2 D1 D2 D4 D3 FH4 300 101 20 101 kPa kPa kPa kPa FH1 F C1 C2 C3 C4

B4 B1 B2 B3 CH COOH CB1 3 CB2 CB3 CB4 FH3

Fig. 13. Separation sequence for a feed stream containing 40 % formic acid, 50 % acetic acid and 10 % water. Column data, thermal energy consumptions and stream compositions are presented in tables 81, 82 and 83.

Table 81. Column data for the separation sequence shown in figure 13.

------Column 1 Column 2 Column 3 Column 4 ------Pressures (kPa) Top 300 101 20 101 Bottom 303 108 25 105 Temperatures (K) Top 416.2 374.0 327.2 373.4 Bottom 430.7 381.9 337.9 381.2 Equilibrium stages 26 40 31 30 Feed trays (from top) 2, 4, 13 32 5 16 Reflux ratio 3.8 7.6 10.3 15.3 Flow rates (kmol/h) Total feed 284.40 234.19 194.59 114.36 Distillate 234.19 39.60 80.23 10.19 Bottom 50.21 194.59 114.36 104.17 Specifications 1: Stream B1 D2 D3 D4 Component CH3COOH HCOOH HCOOH H2O CH3COOH Type fraction fraction fraction fraction Value 0.98 0.98 0.98 0.98 2: Stream B1 D2 D3 D4 Component CH3COOH HCOOH HCOOH H2O CH3COOH Type recovery recovery recovery recovery Value 0.65 0.2485 0.67 0.20 Packed height (m) 21.34 33.78 25.78 24.89 Column inner diameter (m) 2.6 1.4 3.2 1.2 ------Table 82. Thermal energy consumption of the separation sequence shown in figure 13.

------Unit Thermal energy consumption ------GJ/h MW ------FH1 1.553 0.431 FH2 0.845 0.235 FH3 0.380 0.106 FH4 0.502 0.139 CB1 28.538 7.927 CB2 6.310 1.753 CB3 18.467 5.130 CB4 6.687 1.858 63.282 17.579 ------

Table 83. Stream compositions in the separation sequence shown in figure 13.

------Stream Flow rate HCOOH CH3COOH H2O kmol/h % % % ------F 100.0000 40.0000 50.0000 10.0000 B1 50.2100 1.9844 97.9998 0.0159 D1 234.1865 66.6788 11.3140 22.0072 B2 194.5911 60.3055 13.2114 26.4831 D2 39.5955 98.0001 1.9895 0.0104 B3 114.3630 45.4128 10.9286 43.6585 D3 80.2283 81.5347 16.4653 2.0000 B4 104.1730 49.6655 11.9915 38.3430 D4 10.1899 1.9365 0.0632 98.0003 ------

3.3.8.6. Feed containing 12.1 % formic acid, 0.2 % acetic acid and 87.7 % water

When the first distillation stage operates at 300 kPa, the estimated feed stream of the peroxyformic acid pulping process containing 12.1 % formic acid, 0.2 % acetic acid and 87.7 % water can be separated using the sequence shown in figure 14. H2O D1 D2

FH1 HCOOH F 300 20 kPa kPa

C1 C2 FH2 B1 B2

CB1 CB2

Fig. 14. Separation sequence for a feed stream containing 12.1 % formic acid, 0.2 % acetic acid and 87.7 % water.

Acetic acid cannot be separated from the mixture by the sequence shown in figure 14. It is concentrated in the distillate of the second column and cannot be separated from this stream using simple distillation. Alternative separation methods should be considered for the separation of acetic acid. Column data, thermal energy consumptions and stream compositions are presented in tables 84, 85 and 86.

Table 84. Column data for the separation sequence shown in figure 14.

------Column 1 Column 2 ------Pressures (kPa) Top 300 20 Bottom 304 24 Temperatures (K) Top 406.9 326.6 Bottom 415.9 336.3 Equilibrium stages 28 22 ______Table 84. Cont.

------Column 1 Column 2 ------Feed trays (from top) 13, 13 11 Reflux ratio 71.3 24.0 Flow rates (kmol/h) Total feed 141.13 52.53 Distillate 88.59 11.41 Bottom 52.53 41.12 Specifications 1: Stream D1 Component H2O Type fraction reflux ratio Value 0.98 24.0 2: Stream D1 D2 Component H2O HCOOH Type recovery recovery Value 0.87 0.29 Packed height (m) 23.11 17.78 Column inner diameter (m) 5.8 1.8 ------

Table 85. Thermal energy consumption of the separation sequence shown in figure 14.

------Unit Thermal energy consumption ------GJ/h MW ------FH1 0.931 0.259 FH2 0.338 0.094 CB1 248.483 69.023 CB2 5.449 1.514 255.201 70.890 ------Table 86. Stream compositions in the separation sequence shown in figure 14.

------Stream Flow rate HCOOH CH3COOH H2O kmol/h % % % ------F 100.0000 12.1000 0.2000 87.7000 B1 52.5336 73.5806 1.4851 24.9343 D1 88.5915 0.9986 0.0013 99.0001 B2 41.1223 66.7397 1.4138 31.8465 D2 11.4113 98.2325 1.7422 0.0252 ------

3.3.8.7. Summary of the rigorous calculations

The thermal energy consumptions of the separation sequences simulated using rigorous calculation methods are summarized in table 87.

Table 87. Thermal energy consumptions of the separation sequences presented in figures 9 to 14.

------Feed compositions (%) Sequence shown Thermal energy HCOOH CH3COOH H2O in figure consumption (MW) ------10 20 70 9 8.166 10 10 80 10 7.698 30 30 40 11 17.857 30 60 10 12 40.047 40 50 10 13 17.579 12.1 0.2 87.7 14 70.890 ------

The energy consumption of the separation sequence for a feed stream containing 30 % formic acid, 60 % acetic acid and 10 % water is considerably higher than for the other feed stream compositions. Therefore this sequence was rejected and not studied further. The energy consumption for separating the reference feed composition is even higher. 3.3.9. Thermal integration of the separation processes

The separation of the remaining feed compositions was further studied by performing a pinch analysis (Linnhoff et al. 1982) in order to estimate the total costs of the separation sequences. The analysis was carried out using the SYNSET program (Anon. 1992b). Streams included in the pinch analysis are shown in figures 15 to 19 and the corresponding stream data is presented in tables 88 to 92. All product streams (B1...B4, C1...C4) are cooled to 293 K which is also the supply temperature of the feed stream F.

U1 U2 U3 U4 D1 D2 D3

F 20 300 101 300 kPa kPa kPa kPa

C1 C2 C3 C4

B1 B3 B4 L1 L2 L3 L4

Fig. 15. Streams included in thermal integration when the feed stream contains 10 % formic acid, 20 % acetic acid and 70 % water.

Table 88. Stream data for the separation sequence in figure 15.

------Stream Type Heat capacity Supply Target flow rate temperature temperature MW/K K K ------U1 HOT 2.4829 329.8 329.2 U2 HOT 11.1778 407.2 407.0 U3 HOT 16.6083 374.1 374.0 U4 HOT 2.3101 420.9 419.8 D1 HOT 0.0007 329.2 293.0 D2 HOT 0.0008 407.0 293.0 D3 HOT 0.0003 374.0 293.0 B4 HOT 0.0009 430.6 293.0 F COLD 0.0024 293.0 332.5 L1 COLD 14.9360 336.2 336.3 ______Table 88. Cont.

------Stream Type Heat capacity Supply Target flow rate temperature temperature MW/K K K ------L2 COLD 2.2400 421.2 422.2 L3 COLD 2.6190 384.2 384.8 L4 COLD 12.7080 430.4 430.6 B1 COLD 0.0019 336.4 414.3 B3 COLD 0.0020 384.9 424.1 ------

Streams to be cooled in the process are called hot streams, and streams to be heated are called cold streams. The heat capacity flow rates are obtained by multiplying the heat capacity of a stream with its mass flow rate. For streams U and L, the heat of phase change is also taken into account.

U1 U2 U3 U4 D1 D2 D3

F 20 300 101 300 kPa kPa kPa kPa

C1 C2 C3 C4

B1 B3 B4 L1 L2 L3 L4

Fig. 16. Streams included in thermal integration when the feed stream contains 10 % formic acid, 10 % acetic acid and 80 % water. Table 89. Stream data for the separation sequence in figure 16.

------Stream Type Heat capacity Supply Target flow rate temperature temperature MW/K K K ------U1 HOT 1.3929 329.9 329.2 U2 HOT 7.4861 407.2 407.0 U3 HOT 11.3167 374.1 374.0 U4 HOT 6.4861 418.2 417.6 D1 HOT 0.0008 329.2 293.0 D2 HOT 0.0009 407.0 293.0 D3 HOT 0.0003 374.0 293.0 B4 HOT 0.0004 430.6 293.0 F COLD 0.0023 293.0 331.6 L1 COLD 1.9640 337.2 337.7 L2 COLD 2.4970 418.8 419.4 L3 COLD 3.5670 382.6 382.9 L4 COLD 19.4610 430.4 430.6 B1 COLD 0.0015 337.7 412.3 B3 COLD 0.0013 382.9 420.9 ------

U1 U2 U3 D1 D2 D3

F 300 20 300 kPa kPa kPa

C1 C2 C3

B2 B3 L1 L2 L3

Fig. 17. Streams included in thermal integration when the feed stream contains 30 % formic acid, 30 % acetic acid and 40 % water. Table 90. Stream data for the separation sequence in figure 17.

------Stream Type Heat capacity Supply Target flow rate temperature temperature MW/K K K ------U1 HOT 27.1194 407.2 407.0 U2 HOT 5.1111 328.1 327.6 U3 HOT 95.1194 414.3 414.2 D1 HOT 0.0008 407.0 293.0 D3 HOT 0.0011 414.2 293.0 B3 HOT 0.0014 431.6 293.0 D2 COLD 0.0019 327.6 421.9 F COLD 0.0032 293.0 417.1 L1 COLD 6.0260 418.0 418.9 L2 COLD 22.0250 335.0 335.1 L3 COLD 95.1610 431.5 431.6 B2 COLD 0.0018 335.1 417.2 ------

U1 U2 U3 U4 D2 D4 D3 300 101 20 101 kPa kPa kPa kPa

F C1 C2 C3 C4

B4 B1 B3 L1 L2 L3 L4

Fig. 18. Streams included in thermal integration when the feed stream contains 40 % formic acid, 50 % acetic acid and 10 % water. Table 91. Stream data for the separation sequence in figure 18.

------Stream Type Heat capacity Supply Target flow rate temperature temperature MW/K K K ------U1 HOT 79.1583 416.3 416.2 U2 HOT 10.0833 374.2 374.0 U3 HOT 9.0083 327.8 327.2 U4 HOT 6.2111 373.7 373.4 D2 HOT 0.0013 374.0 293.0 D4 HOT 0.0002 373.4 293.0 B1 HOT 0.0024 430.7 293.0 D3 COLD 0.0024 327.2 417.0 F COLD 0.0030 293.0 421.9 L1 COLD 39.6360 430.5 430.7 L2 COLD 8.7640 381.7 381.9 L3 COLD 51.2970 337.8 337.9 L4 COLD 18.5750 381.1 381.2 B3 COLD 0.0059 337.9 382.5 B4 COLD 0.0030 381.2 415.4 ------

U1 U2 D1 D2

300 20 F kPa kPa

C1 C2

B2

L1 L2

Fig. 19. Streams included in thermal integration when the feed stream contains 12.1 % formic acid, 0.2 % acetic acid and 87.7 % water. Table 92. Stream data for the separation sequence in figure 19.

------Stream Type Heat capacity Supply Target flow rate temperature temperature MW/K K K ------U1 HOT 690.5167 407.0 406.9 U2 HOT 8.1972 326.8 326.6 D1 HOT 0.0019 406.9 293.0 D2 HOT 0.0003 326.6 293.0 F COLD 0.0022 293.0 409.5 L1 COLD 690.2310 415.8 415.9 L2 COLD 3.0270 335.8 336.3 B2 COLD 0.0011 336.3 416.5 ------

The streams that are included in the analysis are placed in temperature intervals. The interval boundary temperatures are set at 0.5·∆Tmin below hot stream temperatures and ∆ 0.5· Tmin above cold stream temperatures. ∆Tmin is the minimum temperature difference. When it is 10 K, the streams can be placed in intervals as shown in figure 20 for the sequence separating a mixture containing 10 % formic acid, 20 % acetic acid and 70 % water (fig. 15 and table 88). When the stream population in every temperature interval is known, the intervals are analyzed by calculating their enthalpy balances using equation 3.

∆HTT=−()+ ()∑∑ CPCP− iii1 C Hi (3)

where ∆Hi is the enthalpy change at interval i, Ti is the upper boundary temperature of interval i, Ti+1 is the lower boundary temperature of interval i, CPC is the CP value of a cold stream and CPH is the CP value of a hot stream. An example of the temperature interval analysis is shown in table 93. 435.6K

I

435.4K L4 II

429.1K

III

427.2K

IV

426.2K L2 V B4 425.6K

VI

419.3K

VII

415.9K U4

VIII

418.8K

IX U2 402.2K

X D2 402.0K

XI

389.9K B3 XII

389.8K XIII

389.2K L3 XIV 369.1K U3

XV 369.0K D3

XVI

341.4K B1 XVII

341.3K XVIII

341.2K L1 XIX 337.5K

XX 324.8K U1

XXI D1 324.2K XXII

298.0K F XXIII

288.0K

Fig. 20. The placement on temperature intervals of the streams shown in table 88, when the minimum temperature difference is 10K. Table 93. Temperature interval analysis for the sequence separating the feed containing 10 % formic acid, 20 % acetic acid and 70 % water (fig. 15 and table 88).

------Interval Ti + Ti+1 SCPC - SCPH DHi Surplus number K MW/K MW or i deficit ------1 0.2 12.7100 2.5420 deficit 2 6.3 0.0000 0.0000 3 1.9 0.0020 0.0038 deficit 4 1.0 2.2423 2.2423 deficit 5 0.6 0.0020 0.0012 deficit 6 6.3 0.0011 0.0069 deficit 7 3.4 0.0030 0.0101 deficit 8 1.1 - 2.3071 - 2.5379 surplus 9 12.6 0.0030 0.0373 deficit 10 0.2 - 11.1748 - 2.2350 surplus 11 12.1 0.0022 0.0261 deficit 12 0.1 0.0001 1·10-5 deficit 13 0.6 2.6191 1.5715 deficit 14 20.1 0.0001 0.0029 deficit 15 0.1 - 16.6082 - 1.6608 surplus 16 27.6 - 0.0001 - 0.0039 surplus 17 0.1 - 0.0020 - 0.0002 surplus 18 0.1 14.9341 1.4934 deficit 19 3.7 - 0.0020 - 0.0074 surplus 20 12.7 0.0004 0.0054 deficit 21 0.6 - 2.4824 - 1.4895 surplus 22 26.2 - 0.0003 - 0.0079 surplus 23 10.0 - 0.0027 - 0.0273 surplus ------

Assuming that no heat is supplied to the hottest interval (1) from a hot utility, then the deficit 2.542 MW from interval 1 is cascaded into interval 2. There it joins the zero enthalpy change from interval 2, making 2.542 MW cascade into interval 3. Interval 3 has a 0.004 MW deficit. After accepting the 2.542 MW, it passes on a 2.545 deficit to interval 4. The heat cascade composed using this principle is shown in figure 21(a). The negative heat flows in figure 21(a) are thermodynamically infeasible. By adding 4.806 MW of heat a from hot utility as shown in figure 21(b) and cascading it through the system, all the heat flows between intervals are given a value of at least zero. As a result of this, the minimum utility requirements have been predicted. They are 4.806 MW for the hot utility and 4.833 MW for the cold utility. The position of the pinch has also been located. It is the interval boundary temperature 415.9 K where the heat flow between intervals is zero. This means that the pinch point temperature for hot streams is 420.9 K and for cold streams it is 410.9 K. FROM HOT UTILITY FROM HOT UTILITY 0 kW 435.6K 4.806MW

1 DH=2.542 MW 1 DH=2.542MW

435.4K -2.542MW 2.264MW

2 DH=0.000MW 2 DH=0.000MW 429.1K -2.542MW 2.264MW

3 DH=0.004MW 3 DH=0.004MW

427.2K -2.545MW 2.260MW

4 DH=2.242MW 4 DH=2.242MW

426.2K -4.788MW 0.018MW

5 DH=0.001MW 5 DH=0.001MW

425.6K -4.789MW 0.017MW

6 DH=0.007MW 6 DH=0.007MW

419.3K -4.796MW 0.010MW

7 DH=0.010MW 7 DH=0.010MW

415.9K -4.806MW 0.000MW

8 DH=-2.538MW 8 DH=-2.538MW 414.8K -2.268MW 2.538MW

9 DH=0.037MW 9 DH=0.037MW 402.2K -2.305MW 2.501MW

10 DH=-2.235MW 10 DH=-2.235MW

402.0K -0.070MW 4.736MW

11 DH=0.026MW 11 DH=0.026MW

389.9K -0.096MW 4.709MW

12 DH=1·10-5MW 12 DH=1·10-5MW 389.8K -0.096MW 4.709MW

13 DH=1.571MW 13 DH=1.571MW 389.2K -1.668MW 3.138MW

14 DH=0.003MW 14 DH=0.003MW

369.1K -1.671MW 3.135MW

15 DH=-1.661MW 15 DH=-1.661MW

369.0K -0.010MW 4.796MW

16 DH=-0.004MW 16 DH=-0.004MW

341.4K -0.006MW 4.800MW

17 DH=-2·10-4MW 17 DH=-2·10-4MW

341.3K -0.006MW 4.800MW

18 DH=1.493MW 18 DH=1.493MW 341.2K -1.499MW 3.307MW

19 DH=-0.007MW 19 DH=-0.007MW

337.5K -1.492MW 3.314MW

20 DH=0.005MW 20 DH=0.005MW 324.8K -1.497MW 3.309MW

21 DH=-1.489MW 21 DH=-1.489MW

324.2K -0.008MW 4.798MW

22 DH=-0.008MW 22 DH=-0.008MW

298.0K 0.000MW 4.806MW

23 DH=-0.027MW 23 DH=-0.027MW

288.0K 0.027MW 4.833MW

TO COLD UTILITY TO COLD UTILITY

INFEASIBLE PINCH , QHmin , QCmin (a) (b)

Fig. 21. The heat cascade for the temperature intervals presented in table 93. All the separation sequences were analyzed using the minimum approach temperature of 10 K. Data presented in table 94 were used when performing the analyses and constructing the heat exchanger networks with the SYNSET program. The results are presented in tables 95 and 96.

Table 94. Data used in pinch analysis.

------Property Value Reference ------Steam temperature 450 K Steam latent heat 1785 kJ/kg Perry & Chilton 1973 Steam specific cost 0.06 FIM/kg Water inlet temperature 278 K Water outlet temperature 333 K Water specific heat capacity 4187 kJ/m3K Perry & Chilton 1973 Water specific cost 5.6 FIM/m3 Isoaho 1998 Overall heat transfer coefficients: exchangers and coolers 264 W/m2K Anon. 1992b heaters 347 W/m2K Anon. 1992b Annual working hours 8500 h/a Annual rate of return 0.1 a-1 Anon. 1992b ------

Steam was supposed to be produced in a heavy fuel oil fired boiler with an overall efficiency of 90 %. When the heat content of heavy fuel oil is 40.6 MJ/kg (Kinnula 1986) and the price is 1.1 FIM/kg (Finnish Oil Association 1998), the price of energy is 0.03 FIM/MJ. When this is multiplied with 2.042 MJ/kg, which is the enthalpy change needed to produce the steam from water at 278 K, the steam specific cost of 0.06 FIM/kg was obtained. The products are cooled to 293 K. Therefore higher minimum approach temperatures can not be used if water is the cold utility to be used. The pinch analysis was carried out with the minimum approach temperature of 5 K in order to find out if lower minimum approach temperatures would have a positive effect on the energy consumptions of the separation sequences. The results are presented in tables 97 and 98. Table 95. Pinch analysis results when DTmin is 10 K (F is formic acid, A is acetic acid, and W is water).

------Original feed Separation steps Minimum utility Pinch temperatures composition requirements ------F A W 1st 2nd 3rd 4th Hot Cold Hot Cold % % % kPa kPa kPa kPa MW MW K K ------10 20 70 20 300 101 300 4.806 4.833 420.9 410.9 10 10 80 20 300 101 300 5.407 5.429 418.2 408.2 30 30 40 300 20 300 - 15.014 15.049 414.3 404.3 40 50 10 300 101 20 101 7.993 8.007 416.3 406.3 12.1 0.2 87.7 300 20 - - 69.054 69.097 407.0 397.0 ------

Table 96. Pinch analysis results when DTmin is 10 K (F is formic acid, A is acetic acid, and W is water).

------Original feed Separation steps Number of heat Annual costs composition exchangers ------F A W 1st 2nd 3rd 4th Minimum Actual Steam Water % % % kPa kPa kPa kPa FIM/a FIM/a ------10 20 70 20 300 101 300 16 18 5.2·106 6.5·106 10 10 80 20 300 101 300 16 18 5.8·106 6.0·106 30 30 40 300 20 300 - 13 16 18·106 14·106 40 50 10 300 101 20 101 16 19 13·106 15·106 12.1 0.2 87.7 300 20 - - 9 10 73·106 53·106 ------Table 97. Pinch analysis results when DTmin is 5 K (F is formic acid, A is acetic acid, and W is water).

------Original feed Separation steps Minimum utility Pinch temperatures composition requirements ------F A W 1st 2nd 3rd 4th Hot Cold Hot Cold % % % kPa kPa kPa kPa MW MW K K ------10 20 70 20 300 101 300 4.789 4.818 420.9 415.9 10 10 80 20 300 101 300 5.394 5.417 418.2 413.2 30 30 40 300 20 300 - 14.979 15.015 414.3 409.3 40 50 10 300 101 20 101 7.951 7.965 416.3 411.3 12.1 0.2 87.7 300 20 - - 69.039 69.080 407.0 402.0 ------

Table 98. Pinch analysis results when DTmin is 5 K (F is formic acid, A is acetic acid, and W is water).

------Original feed Separation steps Number of heat Annual costs composition exchangers ------F A W 1st 2nd 3rd 4th Minimum Actual Steam Water % % % kPa kPa kPa kPa FIM/a FIM/a ------10 20 70 20 300 101 300 16 17 5.2·106 6.3·106 10 10 80 20 300 101 300 16 17 5.8·106 5.9·106 30 30 40 300 20 300 - 13 16 18·106 14·106 40 50 10 300 101 20 101 16 19 13·106 14·106 12.1 0.2 87.7 300 20 - - 9 10 73·106 53·106 ------

In order to compare the processes, the total annual costs have to be calculated. The heat exchanger networks were designed for the separation processes using the SYNSET program. Results are shown in tables 99...108. Item costs of the heat exchangers were estimated using equation 4 (Rose 1985).

HEIC1000 1 f f f 0.. 73 0 30 A065. 1968 =+++()TpM()+ (4)

where HEIC1968 is the item cost of a heat exchanger in 1968 (US$), fT is the exchanger type factor, fp is the pressure factor, fM is the material factor and A is the heat transfer area (m2).

All the exchangers were supposed to be floating head exchangers which have the value 0.0 for the factor fT (Rose 1985). Factor fp also has the value 0.0 because the pressures of all streams are below 1.0 MPa (Rose 1985). Factor fM has the value 2.0 when stainless steel was chosen to be the construction material of the exchangers. According to Schillmoller (Schillmoller 1997) type 304L is the preferred material for storage tanks when formic acid is handled. This alloy is resistant to formic acid in any concentration at ambient temperatures. Type 316L can be used in all concentrations at ambient temperatures and up to the formic acid content of 10 % at boiling temperature. For intermediate strengths (30-70 %) of formic acid at elevated temperatures, alloys 20, 28, 904, or 825 can be considered. They can withstand mixtures of formic and acetic acids. Several high-nickel alloys, such as G-3, 625, and C-276, can be used for all concentrations at temperatures up to the boiling point. The nominal analyses of the above mentioned alloys are presented in table 99.

Table 99. Nominal analyses of some stainless steel and high-nickel alloys. (Schillmoller 1997) ______Component Nominal analysis, % ------304L 316L 20 28 904 825 G-3 625 C-276 ______C (max.) 0.03 0.03 0.03 0.02 0.02 0.025 0.015 0.10 0.015 Cr 18.0 17.5 20.0 27.0 20.5 21.0 22.0 21.5 16.0 Ni 10.0 13.5 37.5 31.0 25.0 42.0 41.0 61.0 57.0 Mo - 2.5 2.5 3.5 4.7 3.0 7.0 9.0 16.0 Cu - - 3.5 1.0 1.5 2.5 2.0 - - N ------Cb/Ta - - 0.30 - - - 0.30 3.65 - Others - - Cb - - Al,Ti Co,W Ti,Co Ti ______

Equation 4 gives the item costs in the year 1968. These were converted to year 1998 and FIM using equation 5.

PCI1998 IC1998 = CER ◊◊IC1968 PCI1968 (5)

where IC1968 is the item cost in 1968 (US$), IC1998 is the item cost in 1998 (FIM), CER is the currency exchange rate (FIM/US$) and PCIi is the Chemical Engineering plant cost index in year i. The currency exchange rate was 5.38 FIM/US$ in June 1998 (Bank of Finland 1998). The Chemical Engineering plant cost index had the value 113.6 in 1968 (Desai 1981) and the value 435.5 in January 1998 (Anon 1998b). The heat transfer areas of the exchangers were calculated by the SYNSET program. The heat exchanger data for the separation processes are presented in tables 100...104 when the minimum approach temperature is 5 K and in tables 105...109 when DTmin is 10 K.

Table 100. Heat exchanger data for the process separating a feed containing 10% formic acid, 20% acetic acid, and 70% water (figure 15) , when DTmin is 5 K.

------Streams Tin Tout AIC1998 connected K K m2 FIM ------water 278.0 324.2 D1 329.2 293.0 10.9 132861 water 278.0 333.0 D2 407.0 293.0 9.4 124815 water 278.0 333.0 D3 374.0 293.0 3.4 86293 water 278.0 333.0 B4 420.9 293.0 10.8 132337 steam 450.0 450.0 F 293.0 332.5 2.0 74296 steam 450.0 450.0 B1 336.4 414.3 6.2 105939 steam 450.0 450.0 B3 384.9 415.9 3.7 88616

U4 420.9 420.9 L3 384.2 384.8 163.6 555196

U4 420.9 420.9 L1 336.2 336.3 66.8 330099 water 278.0 324.8 U1 329.8 329.2 284.3 775617 ______Table 100. Cont.

------Streams Tin Tout AIC1998 connected K K m2 FIM ------water 278.0 333.0 U2 407.2 407.0 85.5 379681 water 278.0 333.0 U3 374.1 374.0 97.3 409006 water 278.0 333.0 U4 420.9 419.8 85.4 379427 steam 450.0 450.0 B3 415.9 424.1 1.6 70363 water 278.0 333.0 B4 430.6 420.9 0.3 53656 steam 450.0 450.0 L2 421.2 422.2 228.0 678006 steam 450.0 450.0 L4 430.4 430.6 375.4 920265 1434.7 5296474 ------

Table 101. Heat exchanger data for the process separating a feed containing 10% formic acid, 10% acetic acid, and 80% water (figure 16) , when DTmin is 5 K.

------Streams Tin Tout AIC1998 connected K K m2 FIM ------water 278.0 324.2 D1 329.2 293.0 12.5 141026 water 278.0 333.0 D2 407.0 293.0 10.8 132337 water 278.0 333.0 D3 374.0 293.0 3.4 86293 water 278.0 333.0 B4 418.2 293.0 5.1 98693 ______Table 101. Cont.

------Streams Tin Tout AIC1998 connected K K m2 FIM ------steam 450.0 450.0 F 293.0 331.6 1.9 73341 steam 450.0 450.0 B1 337.7 412.3 4.7 95926 steam 450.0 450.0 B3 382.9 413.2 2.2 76158

U4 418.2 418.2 L3 382.6 382.9 114.4 449383

U4 418.2 418.2 L1 337.2 337.7 46.1 269061 water 278.0 324.9 U1 329.9 329.2 186.0 599562 water 278.0 333.0 U2 407.2 407.0 57.3 303059 water 278.0 333.0 U3 374.1 374.0 66.3 328711 water 278.0 333.0 U4 418.2 417.6 133.9 492925 steam 450.0 450.0 B3 413.2 420.9 0.9 62502 water 278.0 333.0 B4 430.6 418.2 0.2 51689 steam 450.0 450.0 L2 418.8 419.4 139.6 505224 steam 450.0 450.0 L4 430.4 430.6 574.8 1199473 1359.9 4965364 ------Table 102. Heat exchanger data for the process separating a feed containing 30% formic acid, 30% acetic acid, and 40% water (figure 17) , when DTmin is 5 K.

------Streams Tin Tout AIC1998 connected K K m2 FIM ------water 278.0 323.1 U2 328.1 327.6 498.2 1097002 water 278.0 333.0 D1 407.0 293.0 9.9 127544 water 278.0 333.0 D3 414.2 293.0 12.4 140527 water 278.0 333.0 B3 414.3 293.0 16.2 158622 steam 450.0 450.0 D2 327.6 409.3 6.1 105300 steam 450.0 450.0 F 293.0 409.3 12.6 141524 steam 450.0 450.0 L2 335.0 335.1 55.2 296875 steam 450.0 450.0 B2 335.1 409.3 5.2 99373 water 278.0 333.0 U1 407.2 407.0 207.4 640228 water 278.0 333.0 U3 414.3 414.2 338.7 863661 steam 450.0 450.0 D2 409.3 421.9 2.0 74296 steam 450.0 450.0 F 409.3 417.1 2.0 74296 steam 450.0 450.0 B2 409.3 417.2 1.1 64917 water 278.0 333.0 B3 431.6 414.3 0.8 61225 ______Table 102. Cont.

------Streams Tin Tout AIC1998 connected K K m2 FIM ------steam 450.0 450.0 L1 418.0 418.9 495.1 1092743 steam 450.0 450.0 L3 431.5 431.6 1485.5 2184863 3148.5 7222997 ------

Table 103. Heat exchanger data for the process separating a feed containing 40% formic acid, 50% acetic acid, and 10% water (figure 18) , when DTmin is 5 K.

------Streams Tin Tout AIC1998 connected K K m2 FIM ------water 278.0 333.0 U2 374.2 374.0 118.0 457606 water 278.0 333.0 U4 373.7 373.4 110.0 439208 water 278.0 333.0 D2 374.0 293.0 14.9 152618 water 278.0 333.0 D4 373.4 293.0 2.4 77961 water 278.0 333.0 B1 416.3 293.0 27.8 206331 steam 450.0 450.0 D3 327.2 411.3 7.9 116305 steam 450.0 450.0 F 293.0 411.3 11.9 138010 steam 450.0 450.0 L2 381.7 381.9 74.0 349702 steam 450.0 450.0 L4 381.1 381.2 77.7 359515 ______Table 103. Cont.

------Streams Tin Tout AIC1998 connected K K m2 FIM ------steam 450.0 450.0 B3 337.9 382.5 8.6 120341 steam 450.0 450.0 B4 381.2 411.3 5.0 98009

U1 416.3 416.3 L3 337.8 337.9 247.8 713205 water 278.0 333.0 U1 416.3 416.2 276.5 762527 water 278.0 322.8 U3 327.8 327.2 1059.6 1762985 steam 450.0 450.0 D3 411.3 417.0 1.1 64917 steam 450.0 450.0 F 411.3 421.9 2.7 80570 steam 450.0 450.0 B4 411.3 415.4 1.0 63731 water 278.0 333.0 B1 430.7 416.3 1.1 64917 steam 450.0 450.0 L1 430.4 430.7 1176.9 1884311 3225.0 7912770 ------Table 104. Heat exchanger data for the process separating a feed containing 12.1% formic acid, 0.2% acetic acid, and 87.7% water (figure 19) , when DTmin is 5 K.

------Streams Tin Tout AIC1998 connected K K m2 FIM ------water 278.0 321.8 U2 326.8 326.6 324.1 840551 water 278.0 333.0 D1 406.9 293.0 22.3 184816 water 278.0 321.6 D2 326.6 293.0 4.6 95221 steam 450.0 450.0 F 293.0 402.0 7.4 113345 steam 450.0 450.0 L2 335.8 336.3 38.3 243647 steam 450.0 450.0 B2 336.3 402.0 2.7 80570 water 278.0 333.0 U1 407.0 406.9 2644.5 3158017 steam 450.0 450.0 F 402.0 409.5 1.1 64917 steam 450.0 450.0 B2 402.0 416.5 1.1 64917 steam 450.0 450.0 L1 415.8 415.9 5819.6 5242916 8865.6 10088918 ------Table 105. Heat exchanger data for the process separating a feed containing 10% formic acid, 20% acetic acid, and 70% water (figure 15) , when DTmin is 10 K.

------Streams Tin Tout AIC1998 connected K K m2 FIM ------water 278.0 319.2 D1 329.2 293.0 8.1 117470 water 278.0 333.0 D2 407.0 293.0 9.4 124815 water 278.0 333.0 D3 374.0 293.0 3.4 86293 water 278.0 333.0 B4 420.9 293.0 10.8 132337 steam 450.0 450.0 F 293.0 332.5 2.0 74296 steam 450.0 450.0 B1 336.4 410.9 5.7 102706 steam 450.0 450.0 B3 384.9 410.9 2.9 82253

U4 420.9 420.9 L3 384.2 384.8 163.6 555196

U4 420.9 420.9 L1 336.2 336.3 66.8 330099 water 278.0 319.8 U1 329.8 329.2 223.8 670404 water 278.0 333.0 U2 407.2 407.0 85.5 379681 water 278.0 333.0 U3 374.1 374.0 97.3 409006 water 278.0 333.0 U4 420.9 419.8 85.4 379427 steam 450.0 450.0 B1 410.9 414.3 0.5 56998 ______Table 105. Cont.

------Streams Tin Tout AIC1998 connected K K m2 FIM ------steam 450.0 450.0 B3 410.9 424.1 2.4 77961 water 278.0 333.0 B4 430.6 420.9 0.3 53656 steam 450.0 450.0 L2 421.2 422.2 228.0 678006 steam 450.0 450.0 L4 430.4 430.6 375.4 920265 1371.3 5230871 ------

Table 106. Heat exchanger data for the process separating a feed containing 10% formic acid, 10% acetic acid, and 80% water (figure 16) , when DTmin is 10 K.

------Streams Tin Tout AIC1998 connected K K m2 FIM ------water 278.0 319.2 D1 329.2 293.0 9.2 123710 water 278.0 333.0 D2 407.0 293.0 10.8 132337 water 278.0 333.0 D3 374.0 293.0 3.4 86293 water 278.0 333.0 B4 418.2 293.0 5.1 98693 steam 450.0 450.0 F 293.0 331.6 1.9 73341 steam 450.0 450.0 B1 337.7 408.2 4.3 93074 steam 450.0 450.0 B3 382.9 408.2 1.7 71376 ______Table 106. Cont.

------Streams Tin Tout AIC1998 connected K K m2 FIM ------U4 418.2 418.2 L3 382.6 382.9 114.4 449383

U4 418.2 418.2 L1 337.2 337.7 46.1 269061 water 278.0 319.9 U1 329.9 329.2 146.5 519879 water 278.0 333.0 U2 407.2 407.0 57.3 303059 water 278.0 333.0 U3 374.1 374.0 66.3 328711 water 278.0 333.0 U4 418.2 417.6 133.9 492925 steam 450.0 450.0 B1 408.2 412.3 0.4 55401 steam 450.0 450.0 B3 408.2 420.9 1.3 67182 water 278.0 333.0 B4 430.6 418.2 0.2 51689 steam 450.0 450.0 L2 418.8 419.4 139.6 505224 steam 450.0 450.0 L4 430.4 430.6 574.8 1199473 1317.1 4920813 ------Table 107. Heat exchanger data for the process separating a feed containing 30% formic acid, 30% acetic acid, and 40% water (figure 17) , when DTmin is 10 K.

------Streams Tin Tout AIC1998 connected K K m2 FIM ------water 278.0 318.1 U2 328.1 327.6 391.6 944630 water 278.0 333.0 D1 407.0 293.0 9.9 127544 water 278.0 333.0 D3 414.2 293.0 12.4 140527 water 278.0 333.0 B3 414.3 293.0 16.2 158622 steam 450.0 450.0 D2 327.6 404.3 5.4 100719 steam 450.0 450.0 F 293.0 404.3 11.5 135969 steam 450.0 450.0 L2 335.0 335.1 55.2 296875 steam 450.0 450.0 B2 335.1 404.3 4.7 95926 water 278.0 333.0 U1 407.2 407.0 207.4 640228 water 278.0 333.0 U3 414.3 414.2 338.7 863661 steam 450.0 450.0 D2 404.3 421.9 2.7 80570 steam 450.0 450.0 F 404.3 417.1 3.1 83896 steam 450.0 450.0 B2 404.3 417.2 1.7 71376 water 278.0 333.0 B3 431.6 414.3 0.8 61225 ______Table 107. Cont.

------Streams Tin Tout AIC1998 connected K K m2 FIM ------steam 450.0 450.0 L1 418.0 418.9 495.1 1092743 steam 450.0 450.0 L3 431.5 431.6 1485.5 2184863 3041.9 7079375 ------

Table 108. Heat exchanger data for the process separating a feed containing 40% formic acid, 50% acetic acid, and 10% water (figure 18) , when DTmin is 10 K.

------Streams Tin Tout AIC1998 connected K K m2 FIM ------water 278.0 333.0 U2 374.2 374.0 118.0 457606 water 278.0 333.0 U4 373.7 373.4 110.0 439208 water 278.0 333.0 D2 374.0 293.0 14.9 152618 water 278.0 333.0 D4 373.4 293.0 2.4 77961 water 278.0 333.0 B1 416.3 293.0 27.8 206331 steam 450.0 450.0 D3 327.2 406.3 7.1 111536 steam 450.0 450.0 F 293.0 406.3 10.9 132861 steam 450.0 450.0 L2 381.7 381.9 74.0 349702 steam 450.0 450.0 L4 381.1 381.2 77.7 359515 ______Table 108. Cont.

------Streams Tin Tout AIC1998 connected K K m2 FIM ------steam 450.0 450.0 B3 337.9 382.5 8.6 120341 steam 450.0 450.0 B4 381.2 406.3 3.9 90129

U1 416.3 416.3 L3 337.8 337.9 247.8 713205 water 278.0 333.0 U1 416.3 416.2 276.5 762527 water 278.0 317.8 U3 327.8 327.2 832.6 1513818 steam 450.0 450.0 D3 406.3 417.0 1.9 73341 steam 450.0 450.0 F 406.3 421.9 3.8 89376 steam 450.0 450.0 B4 406.3 415.4 2.0 74296 water 278.0 333.0 B1 430.7 416.3 1.1 64917 steam 450.0 450.0 L1 430.4 430.7 1176.9 1884311 2998.0 7673600 ------Table 109. Heat exchanger data for the process separating a feed containing 12.1% formic acid, 0.2% acetic acid, and 87.7% water (figure 19) , when DTmin is 10 K.

------Streams Tin Tout AIC1998 connected K K m2 FIM ------water 278.0 316.8 U2 326.8 326.6 254.5 724891 water 278.0 333.0 D1 406.9 293.0 22.3 184816 water 278.0 316.6 D2 326.6 293.0 3.4 86293 steam 450.0 450.0 F 293.0 397.0 6.8 109699 steam 450.0 450.0 L2 335.8 336.3 38.3 243647 steam 450.0 450.0 B2 336.3 397.0 2.4 77961 water 278.0 333.0 U1 407.0 406.9 2644.5 3158017 steam 450.0 450.0 F 397.0 409.5 1.7 71376 steam 450.0 450.0 B2 397.0 416.5 1.5 69328 steam 450.0 450.0 L1 415.8 415.9 5819.5 5242858 8794.8 9968962 ------

Column costs are composed of the shell cost and the cost of the internals. Shell costs can be estimated using equation 6 (Rose 1985).

SIC1968 = fpM f CB, shell (6)

where SIC1968 is the shell item costs in 1968 (US$) and CB,shell is the base cost (US$).

The base cost of a column shell can be estimated using equation 7 (Rose 1985). CF B, shell = 1000◊ exp() 1 . 33- 0 . 541 (7)

where

FLDL=-()()0. 778 0 . 000082 ln 3 . 281+- 0 . 9199 1 . 433 (8)

where L is the column height (m) and D is the column diameter (m).

If the columns are made of stainless steel the material factor fM has a value of 3.67. At operating pressures below 1.0 MPa, the value of the pressure factor fp is one. (Rose 1985) The cost of packing can be estimated using equation 9 (Rose 1985).

p 2 PC= D LCB, packing 4 (9)

where PC is the costs of packing in 1968 (US$) and CB is the packing price in 1968 (US$/m3).

The price for stoneware saddles with a nominal size of 50 mm was in 1968 180 US$/m3 (Rose 1985). The total installed cost of the distillation plants were estimated using equation 10 (Rose 1985).

TIC item cos ts f 40 , = Â()◊◊I (10) where TIC is the total installed cost in 1998 (FIM) and fI is a correction factor for inflation and currency.

The correction factor fI can be calculated using equation 11.

PCI1998 fI = CER ◊ PCI1968 (11)

where CER is the currency exchange rate (FIM/US$) and PCIi is the Chemical Engineering plant cost index in year i.

The same values for the currency exchange rate and the plant cost indexes were used as when calculating the heat exchanger costs. These resulted in a value of 20.6 for the correction factor. The total annual costs of the distillation plants were estimated using equation 12 (Anon. 1992b).

TAC= I◊ TIC+ OC (12)

where TAC is the total annual costs, TIC is the total installed cost of equipment, OC is the annual operating cost and I is the rate of return.

The annual operating cost was presumed to consist of the annual costs of steam and water. When the value 0.1 a-1 was used as the rate of return, the results shown in tables 110 and 111 were obtained.

Table 110. Total annual costs of the distillation sequences when DTmin is 5 K (F is formic acid, A is acetic acid, and W is water).

------Original feed Separation steps Total annual costs composition 106 FIM/a ------F A W 1st 2nd 3rd 4th % % % kPa kPa kPa kPa ------10 20 70 20 300 101 300 41.3 10 10 80 20 300 101 300 43.1 30 30 40 300 20 300 - 69.9 40 50 10 300 101 20 101 74.8 12.1 0.2 87.7 300 20 - - 153.5 ------

The lowest annual costs are achieved when the feed to the distillation sequence contains 10% formic acid, 20% acetic acid and 70% water. When the feed liquor contains 10 % formic acid, 10% acetic acid and 80% water, the annual costs are only slightly higher. The effect of the minimum approach temperature is negligible. Table 111. Total annual costs of the distillation sequences when DTmin is 10 K (F is formic acid, A is acetic acid, and W is water).

------Original feed Separation steps Total annual costs composition 106 FIM/a ------F A W 1st 2nd 3rd 4th % % % kPa kPa kPa kPa ------10 20 70 20 300 101 300 41.4 10 10 80 20 300 101 300 43.1 30 30 40 300 20 300 - 90.5 40 50 10 300 101 20 101 75.3 12.1 0.2 87.7 300 20 - - 153.7 ------4. Discussion

4.1. Organosolv pulping

Ethanol seems to be the first organic chemical whose potential as a pulping solvent was discovered. During the 100 years plus of organosolv pulping research it has been the most popular solvent, if the number of references is used as a measure. The distribution of papers referred in this work and dealing with the use of organic solvents in pulping and pulping chemistry is shown in tables 112 and 113. 55 % of the papers are related to the use of alcohols, and 80 % of the papers have been published during the 1980’s and 1990’s. The popularity of alcohols is probably based on their relatively low prices when compared to other organic chemicals. The large number of publications during the 1980’s and 1990’s is a result of the growing awareness of pollution problems related to conventional pulp production. During the same time, the kraft process has also been developed keeping it still highly competitive.

Table 112. The number of papers dealing with the use of organic solvents in pulping and pulping chemistry.

------Solvent Number of papers ------Methanol 186 Ethanol 259 Other alcohols 72 Formic acid 105 Acetic acid and ethyl acetate 156 Other organic acids 11 Phenol and cresols 58 Other organic chemicals 98 945 ------Table 113. The number of organosolv papers at various decades.

------Years Number of papers ------≤ 1900 3 1901 - 1910 2 1911 - 1920 6 1921 - 1930 14 1931 - 1940 43 1941 - 1950 18 1951 - 1960 15 1961 - 1970 27 1971 - 1980 59 1981 - 1990 375 1991 - 1998 383 945 ------

The quality of organosolv pulps is not competitive with the quality of corresponding kraft pulps. Particularly in alcohol pulping, the pulp quality can be improved by adding alkali to the system. By doing this, the capital cost advantage of the organosolv processes is lost, because the alkali has to be recovered using an expensive recovery boiler. Many authors state that the organosolv process they are presenting is environmentally friendlier than the kraft process. This is not necessarily true, because: - the kraft process has been improved and the environmental problems have been partially solved. - unlike the kraft process, the organosolv processes are not energy self-sufficient, which probably leads to an increased energy production by fossil fuels. - the use of organosolv methods in pulp bleaching may result in higher BOD7, COD and TOC loads and higher poisonous qualities in effluents when compared to, for example, ozone bleaching. (Liukko & Poppius-Levlin 1997) These problems can probably be solved by further research and better process design. In order to make organosolv pulping processes competitive with conventional methods the problems have to be solved. Solvent recovery has been discussed in very few papers. Solutions presented are conventional, including unit operations like evaporation and distillation. In many cases, recovery cycle designs seem to be based on incomplete data. Feed streams are presumed to be simple binary mixtures, which leads to simple designs. In reality, the streams entering the recovery cycles are multicomponent mixtures. Most components are present in very low concentrations. If closed liquid circulation is used they have to be removed in order to prevent their accumulation in the system. This will result in the use of complex multi-stage separation processes, which have high investment and operating costs. Distillation is the most popular unit operation proposed for use as the solvent purification stage. It is a major energy consumer in the process and has therefore to be designed carefully. In addition, it has to be thermally integrated to other parts of the process. When organosolv processes are evaluated using the list of requirements for a new pulping process presented in chapter 1, it seems to be unlikely that any process will be applied on an industrial scale in the near future. In order to be able to replace conventional pulping methods in existing plants, or in new investments, the organosolv methods should offer significant benefits. In order to reach that status, the processes have to be further developed. This may be done by optimising the existing process configurations, by designing new ones, or by adding new stages, e.g. fungal pretreatment (Ferraz et al. 1998b), to existing configurations.

4.2. Solvent recovery in peroxyacid pulping

4.2.1. Feasible separation methods

When the separation process selection methods found in the literature are applied to the ternary mixture formic acid-acetic acid-water, they suggest the use of the following unit operations as possible methods: - adsorption - azeotropic distillation - crystallisation - evaporation - extractive distillation - flash evaporation - liquid-liquid extraction - pervaporation - simple distillation - stripping - ultrafiltration Simple distillation, azeotropic distillation, extractive distillation and liquid-liquid extraction are separation processes that have been used in published studies for separating the ternary mixture. Research in this field has been almost completely focused on the purification of the product in acetic acid production or waste water treatment. The acetic acid product contains only small amounts of formic acid and water as impurities. Waste waters contain small amounts of both acids. The feed of the separation section in a pulping process using the Milox method contains water as the main component and significantly more formic acid than acetic acid. Therefore the separation methods presented in the literature cannot be directly applied. Simple distillation was chosen for further studies in order to keep the separation process simple. Extractive distillation was also chosen as a representative for processes using mass separating agents. The selection of a mass separating agent to be used in an organosolv process is critical. Small amounts of the agent follow the recovered solvent to the digester and the possible effect of that has to be studied before a mass separating agent can be selected.

4.2.2. Minimum work of separation

When minimum work of separation calculated for the feed compositions studied are ranked in increasing order, the results shown in table 114 are obtained for those compositions that reached the final stage of this research.

Table 114. Minimum work of separation rankings compared with the total annual costs (x1 is the molar fraction of formic acid in feed, 2 and 3 refer to acetic acid and water).

------x1 x2 x3 Wmin,298K Ranking out of 37 Total annual costs kJámol-1 106 FIM/a ------0.1 0.2 0.7 1.611 5. 41.4 0.1 0.1 0.8 1.335 2. 43.1 0.3 0.3 0.4 2.327 36. 90.5 0.4 0.5 0.1 2.015 19. 75.3 0.121 0.002 0.877 0.907 1. 153.7 ______

Minimum work of separation seems to give a reasonable first estimate for the ease of separating the ternary mixture of formic acid, acetic acid and water, when significant amounts of each three components are present

4.2.3. Phase equilibria

According to Walas (Walas 1985) the Wilson model is generally a superior representation for activity coefficient behaviour of both polar and nonpolar mixtures. It is capable of representing multicomponent behaviour with only binary parameters. This is probably the reason for the fact that the Wilson model gave best fits for the vapour- liquid equilibrium data at the pressures 101.3 and 300 kPa. The large number of parameters (8) in the NRTL model resulted in the best data fit at the system pressure of 20 kPa. According to Walas (Walas 1985), the NRTL gives more frequently the best fit than the Wilson model for aqueous organics. When the distillation feed mixtures that reached the final stage of this research are placed on ternary diagrams, the presentations shown on figures 22, 23 and 24 are obtained. Mixtures containing 10 % formic acid, 20 % acetic acid and 70 % water (10- 20-30) and 10 % formic acid, 10 % acetic acid and 80 % water (10-10-80) reached the lowest total costs when the distillation sequence started with a column operating at 20 kPa. Both mixtures are located in figure 22 in the distillation region which is limited by the boundaries drawn between the ternary azeotrope and the upper and left edges of the diagram. This suggests that it is favourable to start the distillation sequence with a vacuum column when the formic acid concentration in feed is below 30.8 % and the feed composition is located in the left triangle.

0 1 B Binary azeotrope 0.2 0.8 F J Ternary azeotrope H H 10-20-70 0.4 0.6 HCOOH F 10-10-80 H2O J B 30-30-40 0.4 0.6 40-50-10 0.2 0.8 Milox

1 0 1 0.8 0.6 0.4 0.2 0 CH3COOH

Fig. 22. Feed compositions placed on the ternary diagram at 20 kPa. 0 1 B Binary azeotrope 0.8 F 0.2 J Ternary azeotrope H H 10-20-70 0.4 0.6 F 10-10-80 H O HCOOH 2 30-30-40 0.4 B 0.6 40-50-10 J 0.2 0.8 Milox

1 0 1 0.8 0.6 0.4 0.2 0

CH3COOH

Fig. 23. Feed compositions placed on the ternary diagram at 101.3 kPa.

0 1 B Binary azeotrope 0.8 F 0.2 J Ternary azeotrope H H 10-20-70 0.6 0.4 HCOOH F 10-10-80 H2O 30-30-40 0.4 0.6 40-50-10 B 0.2 J 0.8 Milox

1 0 1 0.8 0.6 0.4 0.2 0 CH3COOH

Fig. 24. Feed compositions placed on the ternary diagram at 300 kPa.

No sequences that were selected for this study started with an atmospheric column, but three of the five feed mixtures are located in the distillation region limited by boundary lines between the ternary azeotrope and the upper and the left edge of the diagram (fig. 23). Mixtures containing 30 % formic acid, 30 % acetic acid and 40 % water (30-30-40), 40 % formic acid, 50 % acetic acid and 10 % water (40-50-10) and 12.1 % formic acid, 0.2 % acetic acid and 87.7 % water (Milox) started with a pressurized column. The mixture 30-30-40 is located on figure 24 in the distillation region limited by boundaries drawn between the ternary azeotrope and the upper and the left edge of the diagram. This suggest that it is favourable to start a distillation sequence when the formic acid concentration in the feed is below 75.8 %. The mixture 40-50-10 is located on one of the boundary lines. It is therefore obvious that the real boundary lines are not straight lines. The Milox feed mixture is located in the distillation region formed when lines are drawn between the azeotropes and between the ternary azeotrope and the upper edge of the diagram. The total costs of separating this mixture are significantly higher than the costs of separating the four other mixtures. Obviously the acetic acid concentration in the liquor fed to a distillation sequence should therefore be higher. This would place the Milox mixture on the same distillation region with the other mixtures. As a conclusion, it can be noted that vacuum distillation is a favourable first unit operation in a sequence separating the formic acid-acetic acid-water mixture when the formic acid concentration is below 30.8 %. Pressurized distillation is likely to be the best choice when the formic acid concentration is between 30.8 and 75.8 %. When the concentration is above 75.8 % other separation methods should be considered.

4.2.4. Shortcut simulations of the separation sequences

The total number of feed streams with different compositions was 37. When the first separation stage was simulated with these feed streams: - simple distillation at 20 kPa was accepted in four cases, - simple distillation in an atmospheric column was accepted in 21 cases, - simple distillation at 300 kPa was accepted in 18 cases and - extractive distillation using NMP was accepted in 14 cases. Vacuum distillation at 20 kPa seems to be an acceptable first separation stage only for feed streams with high water contents. When the formic acid content of the feed is 50 %, no feasible separation method is found. At 20 kPa, the number of equilibrium stages required is very high. At the higher pressures, the most plentiful component (formic acid) cannot be obtained as a product. This is an indication for the distillation boundaries on the ternary diagrams being curves instead of straight lines. When studying the second separation stage, 99 cases were simulated. The number of cases accepted for further studies was 40. When vacuum distillation was tried as the second separation stage after atmospheric distillation, no solution was found for the original feed composition 40 % formic acid, 50 % acetic acid and 10 % water. In this case, the distillate of the first column contains 80 % formic acid, 1 % acetic acid and 19 % water. If this composition is placed on the ternary diagram at 101.3 kPa published by Hunsmann and Simmrock (Hunsmann & Simmrock 1966), it is located on one of the distillation borders, which makes the separation impossible. Of the 86 different three stage separation sequences simulated, 46 could be accepted. Several possible combinations were found for feed streams containing 40 % formic acid or less. At higher formic acid concentrations only few possible three stage separation sequences were found. Again some cases could not be solved. This is probably caused by the location of the feed composition on an actual distillation border line. In some cases, a three stage separation sequence, which did not produce three pure (98 %) components, was accepted according to the constraints used in this research. A fourth stage was simulated for those cases. Only six four stage processes out of the 17 simulated could be accepted. Most of the rejections were based on far too high a number of equilibrium stages. Some of the sequences were also rejected because all three components were not obtained as products. A fifth stage was not considered. In the last stage of the shortcut simulations, recycle streams were added to the separation sequences. Those sequences having an energy consumption lower than the average were chosen for simulations with rigorous calculation methods. When the thermodynamic efficiency for the separation is determined as

W η = min thermodynamic Wactual (13)

where hthermodynamic is the thermodynamic efficiency, Wmin is the minimum work of separation and Wactual is the actual energy consumption of the separation sequence the efficiencies based on shortcut simulation results are presented in table 115.

Table 115. Thermodynamic efficiencies based on shortcut simulation results for separation of HCOOH-CH3COOH-H2O mixtures.

______Feed composition Minimum work Actual energy Thermo------work of consumption dynamic HCOOH CH3COOH H2O separation kJ/mol of feed efficiency % % % kJ/mol of feed ______10 20 70 1.611 54.74 0.029 10 10 80 1.335 44.19 0.030 30 30 40 2.327 161.30 0.014 30 60 10 1.924 117.72 0.016 40 50 10 2.015 133.05 0.015 12.1 0.2 87.7 0.907 114.15 0.008 ______

The average thermodynamic efficiency is 1.9 % for the separation sequences chosen for rigorous simulations. If the energy consumption of separating the reference mixture is not taken into account, the average efficiency is 2.1 %, which is a fairly typical value for distillation because, according to Null (Null 1980), the average thermodynamic efficiency of distillation is around 2.5 %.

4.2.5. Rigorous simulations

The sequences used in shortcut simulations could not be solved as such using rigorous calculation methods. This is probably caused by the highly unideal nature of the formic acid-acetic acid-water mixture. Therefore the sequences had to be redesigned. The specifications used were: 1. The pressure determined in shortcut calculations was used for each distillation stage. 2. The product purity was 98 %. 3. The highest possible recovery rate was used. According to the rigorous calculation results, four distillation stages were required in most cases to separate the mixtures. This is caused by the specifications used. Accepting lower product purities or recovery rates would decrease the number of distillation stages required. It also could make the compositions of the feed streams to the following stages more favourable for separation. The recovery rates in columns simulated by rigorous methods were significantly lower than the corresponding recovery rates in columns simulated by shortcut methods. This led to significantly higher recycle stream flow rates and higher energy consumptions. The energy consumption of separating the reference mixture is significantly higher than the energy consumption of the other distillation sequences. The composition of the estimated distillation feed in the peroxyformic acid process would be located in the ternary diagrams in a more favourable distillation region if the acetic acid content were increased.

Table 116. Thermodynamic efficiencies based on rigorous simulation results for separation of HCOOH-CH3COOH-H2O mixtures.

______Feed composition Minimum work Actual energy Thermo------work of consumption dynamic HCOOH CH3COOH H2O separation kJ/mol of feed efficiency % % % kJ/mol of feed ______10 20 70 1.611 293.98 0.005 10 10 80 1.335 277.13 0.005 30 30 40 2.327 642.85 0.004 30 60 10 1.924 1441.69 0.001 40 50 10 2.015 632.84 0.003 12.1 0.2 87.7 0.907 2552.04 0.0004 ______When the energy consumptions obtained by rigorous calculations are used to estimate the thermodynamic efficiencies of the distillation sequences according to equation 13, the results presented in table 116 are obtained. The effect of thermal integration is not taken into account in these results. The efficiencies are significantly lower than those based on shortcut simulations. This is caused by the unideal nature of the ternary mixture studied. The average thermodynamic efficiency is 0.3 %. If the energy consumption value for separating the reference mixture is not taken into account, the average efficiency is 0.4 %, which is a fairly low value.

4.2.6. Thermal integration using pinch technology

When the pinch analysis was performed, the separation sequence designed for the feed composition 10 % formic acid, 20 % acetic acid and 70 % water showed the lowest total annual costs. This liquor composition may be near the composition of spent washing liquor if an aqueous mixture of the acids is used as pulping liquor. The effect of the minimum temperature approach value was tested by using the values 5 K and 10 K. The effect was found to be negligible. The cost estimates are rough and cannot be used as the basis for any investment decisions. They are calculated in a similar way for all process alternatives, and can therefore be used to compare the alternatives. A probable reason for the lowest costs occuring with the feed composition 10 % formic acid, 20 % acetic acid and 70 % water is that water seems to be the component most easily separated from the mixture in the first separation stage. When a large amount of one component is separated, the volume fed to the following separation stage decreases significantly causing both operating and investment costs to also decrease. The composition 10 % formic acid, 20 % acetic acid and 70 % water is located in the most favourable distillation region in the ternary diagrams for the pressures 20 kPa, 101.3 kPa and 300 kPa (figures 22...24). The composition with the second best results (10 % formic acid, 10 % acetic acid and 80 % water) is also located in the same region. Therefore it is probable that it is favourable to have around 10 % formic acid, 10-20 % acetic acid and 70-80 % water in the feed of a distillation sequence separating the ternary mixture. These distillation feed liquor compositions can be obtained by using a mixture of formic acid and acetic acid as the pulping liquor and by reducing the amount of water fed to the process. The water amount can be reduced by pulping dried chips and by using less water consuming methods for pulp washing. A feed composition where the acetic acid content is higher than the formic acid content is also favourable because more acetic acid is formed during pulping and formic acid is less stable. It also seems to be favourable to start the distillation sequence with a vacuum column separating most of the water, followed by pressurised distillation, atmospheric distillation and pressurised distillation. Thermal integration decreases the energy consumptions of the distillation sequences by up to 40 % when compared to a situation with no integration. In some cases, the decrease is significantly lower because the temperature levels of the streams in those cases are not suitable for integration. Changing the overall design of the distillation sequences would probably make it possible to make greater savings using thermal integration. When the energy consumptions obtained after thermal integration are used to estimate the thermodynamic efficiencies of the distillation sequences according to equation 13, the results presented in table 117 are obtained.

Table 117. Thermodynamic efficiencies based on thermal integration results for separation of HCOOH-CH3COOH-H2O mixtures.

______Feed composition Minimum work Actual energy Thermo------work of consumption dynamic HCOOH CH3COOH H2O separation kJ/mol of feed efficiencies % % % kJ/mol of feed ______10 20 70 1.611 172.40 0.009 10 10 80 1.335 194.18 0.007 30 30 40 2.327 539.24 0.004 40 50 10 2.015 286.24 0.007 12.1 0.2 87.7 0.907 2485.40 0.0004 ______

Thermal integration has a significant effect on the thermodynamic efficiencies of most of the distillation sequences. The average efficiency is 0.5 %. If the energy consumption of separating the reference mixture is not taken into account, the average efficiency is 0.7 %, which still is significantly lower than the estimate 2.5 % given by Null (Null 1980). Optimising both the sequence and column designs would probably improve the efficiencies. However, the thermodynamic efficiency for the separation of this highly unideal ternary mixture is considerably below the average values for distillation. 5. Conclusions

The research related to organosolv pulping has been so far mainly focused on optimising the delignifying process. Very few papers dealing with the solvent recovery have been published. This, together with the fact that the organosolv pulping processes have very few advantages compared with the kraft process, indicates that they are not ready to compete with the kraft process at this stage of development. The future of organosolv processes is highly dependent on environmental legislation, which in the future may cause problems for the economic operation of present pulping processes. Simple distillation seems to be the preferable solution for recovering the solvents in the peroxyacid pulping process. If only formic acid and peroxyformic acid are used as solvents, the small amount of acetic acid formed during pulping causes the distillation feed composition to be located in a region not favourable for distillation. In order to make the separation more economic, the water content of the distillation feed should be decreased to around 70 %, which means that less water should be used in the preceding unit operations. The feed should also contain more acetic acid than formic acid. This can be achieved by using a mixture of formic acid and acetic acid as the delignifying solvent. The minimum work of separation can be used to estimate the ease of separating various ternary mixtures containing formic acid, acetic acid and water. Shortcut simulations can be used to to compare various distillation sequences. Because of the highly unideal nature of the mixture the results of shortcut simulations should not be used for designing a separation sequence for a particular feed composition. Pinch technology is a useful tool in process synthesis. Energy savings of up to 40 % can be achieved in a distillation plant where the formic acid-acetic acid-water mixture is separated. When the total pulp manufacturing process is integrated, the savings may be even higher. Thermodynamic efficiencies of simple distillation sequences separating the formic acid- acetic acid-water mixture seem to be less than 1 %, which is lower than the average value 2.5 % given for distillation in the literature. The significant effect of the distillation feed composition on the total costs of the separation sequence stresses the importance of simultaneous optimisation of all unit operations in an organosolv pulping process. References

Aaltonen O, Ylinen P, Johansson A & Oksanen H (1987) Valtion teknillinen tutkimuskeskus, assignee. Menetelma lignoselluloosapitoisen raaka-aineen delignifioimiseksi. FI patent 73013. Aaltonen O, Johansson A & Ylinen P (1989) Organosolv pulping: softwood delignification kinetics and selectivity in THFA HCl-pulping. In: Kennedy JF, Phillips GO & Williams PA (eds) Wood Processing and Utilization. Ellis Horwood, Chichester, p 62-68. Abad AQ & Paszner L (1989) Fate and recovery of resin- and fatty acids from cooking liquors on organosolv pulping of Pinus elliottii. 1989 Solvent Pulping Conference, Seattle, WA, p 1-6. Abbot J & Bolker HI (1984) Alkaline delignification with amines and alcohols. Svensk Papperstidning (12): R105-R109. Abe Z (1993) Alkali-alcohol-anthraquinone pulping. Seventh International Symposium on Wood and Pulping Chemistry Proc., Beijing, PR China, 3: 189-194. Abramovitch RA & Iyanar K (1994) Organosolv pulping using a microwave oven. Holzforschung 48(4): 349-354. Ahmed A & Kokta BV (1992) Nonsulfur process for preparing pulps for papermaking. CA patent 1.336.546. (ABIPST 66:18255(P)). Ahmed A, Kokta BB & Carrasco F (1992) Steam explosion pulping of aspen pretreated with alcohol mixed with a non-sulfur chemical. 1992 Solvent Pulping Symposium Notes, Boston, MA, p 67-72. Aittamaa J, Laxen T, Lonnberg B & Sjoholm R (1986) Studies on the alcohol pulping processes. XXII Eucepa Conf., Florence, Italy, 1: 7:1-7:20. Akishima Y, Isogai A, Onabe F & Usuda M (1991) Structural changes of amorphous cellulose by acid pulping treatments. Nordic Pulp and Paper Res J 6(4): 166-169. Albrecht JS & Nicholls GA (1974) Fractionation and examination of soluble products from peroxyacetic acid delignification of loblolly pine. Paperi ja Puu (11): 927-943. Aleksandrova G, Medvedeva S, Kanitskaya L, Babkin V & Zakazov A (1995) An ecologically sound pulping process involving microbiological delignification. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 2: 273-278. Aleksandrova G, Buryachenkov M, Medvedeva S, Zakazov A & Babkin V (1996) Organosolv delignification of wood with microbiological pretreatment. Iz VUZ Lesn Zh (1/2): 30-34. (ABPST 69:349(AS)). Alen R (1988a) Formation of aliphatic carboxylic acids from hardwood polysaccharides during alkaline delignification in aqueous alcohols. Cellul Chem Tech 22(4): 443-448. Alen R (1988b) Formation of aliphatic carboxylic acids from softwood polysaccharides during alkaline delignification in aqueous alcohols. Cellul Chem Tech 22(5): 507-512. Aliev RG, Burov AV & Rusakov AE (1990) Molecular weight characteristics of lignin isolated during organic solvent treatment of birchwood [betula]. Khim Drev (Riga) (3): 45-49. (ABIPST 62:5436). Anon (1920) Koln-Rottweil Akt.-Ges., assignee. Verfahren zur herstellung von zellstoff. German patent 329566. Anon (1935) Thioglycollic acid as a new lignin reagent. The Paper Industry 16, 787-788. Anon (1955) Celanese Corporation, assignee. Purifying acetic acid by azeotropic distillation. GB patent 727 078. Anon (1985) Organic acid pulping of wood reduces chemical costs and energy requirements. Paper Trade J, (11): 50-51. Anon (1987a) Die demonstrationsanlage für das Organocell-verfahren vor dem anlauf. Dtsch Papierwirtsch, (1): T62-T63. Anon (1987b) Radical new technology in chemical pulping. Paper 108(6): 19-20. Anon (1990a) Asam-pilot-zellstoffanlage in Baienfurt. Wochenbl Papierfabr 118(11/12): 524-529. Anon (1990b) Pilot ASAM plant - harbinger of a german pulp industry. Paper Tech 31(8): 36-37. Anon (1990c) ASAM-pilotanlage erfolgreich in betrieb. Allg Papier-Rundsch, (24/25): 712-716. Anon (1991a) Bayerisches richtfest fur den ersten Organocell-kocher der welt. Dtsch Papierwirtsch, (3): 50-56. Anon (1991b) Alcell can offer a green solution. Pulp Paper Int 33(4): 53-57. Anon (1991c) PRO/II regress user guide. Simulation Sciences Inc. Anon (1991d) PRO/II pc user guide. Simulation Sciences Inc. Anon (1992a) Ymparistoa saastava sellun valmistusprosessi koetehdasvaiheessa. Kemia- Kemi 19(2): p. 150. Anon (1992b) SYNSET user manual. ChemEng Software and Services Ltd. Anon (1993) Weltpremiere in Kehlheim. Wochenbl Papierfabr 121(1): 27-29. Anon (1994) Alcell process - technology of the future - today. Dtsch Papierwirtsch (4): 34- 35, 38-39. (ABIPST 66:8836(AS)). Anon (1995) Repap buys Atholville for Alcell mill. Pulp Pap Week 17(1): 7. (ABIPST 65:16836). Anon (1998a) Alcell sale terminated. PIMAs North American Papermaker (6): 18. (Paperbase 00512640). Anon (1998b) Economic indicators. Chem Eng 105(4): 180. Antal M, Ebringerova A & Micko MM (1991) Kationisierte hemicellulosen aus espenholzmehl und ihr einsatz in der papierherstellung. Das Papier 45(5): 232-235. Apostol AV & Kozlov VP (1979) Effect of impregnating birchwood chips with various concentrations of acetic acid on some physical and mechanical properties. Izv VUZ Lesnoi Zh, (1): 64-66. (ABIPC 50:9447). April GC (1983) Cemicals from wood by organic solvent delignification. The National Science Foundation, BER Report No. 311-124. April GC, Kamal MM, Reddy JA, Bowers GH & Hansen SM (1979) Delignification with aqueous-organic solvents: southern yellow pine. Tappi 62(5): 83-85. Aptel P, Challard N, Cuny J & Neel J (1976) Application of the pervaporation process to separate azeotropic mixtures. J Membr Sci 1: 271-287. Araki H, Iwata H, Aoyagi T & Kachi S (1989) Solvolysis pulping process: a comparative study on reactions and products. In: Kennedy JF, Phillips GO & Williams PA (eds) Wood Processing and Utilization. Ellis Horwood, Chichester, p 69-74. Araujo ES & Curvelo AAS (1997a) Buffered organosolv delignification, 1., effect of buffer concentration. 9th International Symposium on Wood and Pulping Chemistry Oral Presentations, Montreal, Canada, p J4 1-4. Araujo GT & Curvelo AAS (1997b) Organosolv pulping of (mimosa hostilis benth.), 1., kinetics of ethanol-water delignification. 9th International Symposium on Wood and Pulping Chemistry Poster Presentations, Montreal, Canada, p 4:1-4. Aravamuthan R, Chen W, Zargarian K & April G (1989) Chemicals from wood: prehydrolysis/organosolv methods. Biomass 20(3/4): 263-276. Aravamuthan RG, Valade JL, Law KN & Lanouette R (1993) Organosolv TMP, CTMP and CMP from aspen. 1993 Pulping Conference Proc., Atlanta, GA, 1: 193 - 204. Argyropoulos DS & Bolker HI (1987) Condensation of lignin in dioxane-water-HCl. J Wood Chem Tech 7(1): 1-23. Argyropoulos DS, Hortling B, Poppius-Levlin K, Sun Y & Mazur M (1995) Milox pulping: lignin charaterization by 31P NMR spectroscopy and oxidative degradation. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 1: 495-502. Arnoldy P (1989) Shell Internationale Research Maatschappij B.V., assignee. Process for pulping lignocellulose-containing material. GB patent 2.209.772A. Arnoldy P & Petrus L (1989) Shell Internationale Research Maatschappij B.V., assignee. Verfahren zum aufschließen lignozellulosehaltigen materials. DE patent 38.30.993.A1. Arnoldy P & Petrus L (1990) Shell Oil Company, assignee. Process for pulping lignocellulose-containing material. US patent 4.904.342. Arnoldy P & Petrus L (1992) Shell Canada Ltd., assignee. Process for pulping lignocellulose-containing material. CA patent 1.300.324. (ABIPST 66:11422(P)). Aronovsky SI & Gortner RA (1936) The cooking process, IX, pulping wood with alcohols and other organic reagents. Ind Eng Chem, p 1270-1276. Avela ES, Sandell E & Avela A (1988) Suomen Sokeri Oy, assignee. Menetelma sellumassan ja sokereiden tuottamiseksi lehtipuuhakkeesta. FI patent 76845. Ayla C (1982) Uber eigenschaften und abbarbaukeit von mit alkohol-wasser-gemischen isolierten ligninen, 7. mitteilung: verwendung von ethanol-wasser-ligninen zur herstellung von holzleimen. Holzforschung 36(2): 93-98. Aziz S & Goyal GC (1993) Kinetics of delignification from mechanistic and process control point of view in solvent pulping processes. 1993 Pulping Conference Proc., Atlanta, GA, 3: 917-920. Aziz S & McDonough TJ (1987) Ester pulping - a brief evaluation. Tappi Journal 70(3): 137-138. Bachner K, Fischer K & Baucker E (1993) Zusammenhang zwischen aufbau der zellwand und festigkeitseigenschaften bei faserstoffen von konventionellen und neuen aufschlußverfahren. Das Papier 47(10A): V30-V40. Baeza J, Urizar S, Erismann N, Freer J, Schmidt E & Duran N (1991) Organosolv pulping, V: formic acid delignification of Eucalyptus globulus and Eucalyptus grandis. Biores Tech 37(1): 1-6. Baierl KW (1979) Adsorber-regenerator. US patent 4.155.849. Baierl KW, Young RA & Young TR (1986) Biodyne Chemicals Inc., assignee. Menetelma lignoselluloosan liuottamiseksi. FI patent application 863069. Bailey AJ (1939) Reagents of the University of Minnesota, assignee. Alcoholic treatment of ligneous cellulosic material. US patent 2.166.540. Bailey AJ (1940a) The chemistry of lignin, I: the existence of a chemical bond between lignin and cellulose. Paper Trade J 110(1): 29-34. Bailey AJ (1940b) The chemistry of lignin, II: the butanolysis of wood. Paper Trade J 110(2): 29-32. Bailey AJ (1940c) The alcoholysis of wood. Pacific Pulp Paper Ind 14: 24-27. Bailey AJ (1942) Recovery of butanol in butanol pulping. Ind Eng Chem 34(4): 483-485. Bailey CW & Dence CW (1966) Peroxyacetic acid bleaching of chemical pulps. Tappi 49(1): 9-15. Balakshin M & Deineko I (1995) The role of individual lignin fragments during wood delignification with oxygen in water-ethanol mixture. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 1: 323-330. Balogh DT, Curvelo AAS & Groote RAMC (1992) Solvent effects on organosolv lignin from Pinus caribaea hondurensis. Holzforschung 46(4): 343-348. Balogh DT (1993) Molecular weights of organosolv lignins. In: Kennedy JF (ed) Cellulosics: Chemical, Biochemical and Material Aspects. Ellis Horwood, Chichester, p 279-284. Bank of Finland (1998) Http://www.bof.fi/env/fin/new/fixlist.stm. Barham HN & Clark LW (1951) The decomposition of formic acid at low temperatures. J Am Chem Soc 73: 4638-4640. Barna BA, Pylkkanen VA, Kivimaki JK & Yang Y (1991) Applications of process design software in organosolv pulping. 1991 Tappi Pulping Conf. Proc., Orlando, FL, 2: 637- 644. Barnicki SD & Fair JR (1990) Separation system synthesis: a knowledge based approach, 1: liquid mixture separations. Ind Eng Chem Res 29(3): 421-432. Basta J, Holtinger L, Hallstrom AS & Lundgren P (1994) Peroxo compounds in tcf bleaching. 1994 Pulping Conference Proc., San Diego, CA, p 953-956. Baumeister M & Edel E (1980) Athanol-wasser-aufschluss. Das Papier 34(10A): V9-V18. Baumeister M & Edel E (1982) MD-Papierfabriken Heinrich Nicolaus GMBH, assignee. Verfahren und vorrichtung zum kontinuierlichen aufschliessen von pflanzenfaser- material. EP patent 0.012.960.B1. Baumeister M & Edel E (1983a) MD-Organocell Gesellschaft für Zellstoff- und Umwelttechnik mbH, assignee. Menetelma ja laite kasvikuitumateriaalin liuottamiseksi yhtajaksoisesti. FI patent 71360. Baumeister M & Edel E (1983b) MD Verwaltungsgesellschaft Nicolaus GmbH & Co. KG, assignee. Verfahren und reactor zum kontinuierlichen aufschliessen von pflanzenfasermaterial. DE patent 32.12.767.A1. Baumeister M & Edel E (1985) MD-Verwaltungsgesellschaft Nicolaus G.m.b.H & Co. KG, assignee. Method and reactor for continuous extraction of vegetable-fiber material. CA patent 1.196.155. Beer R & Peter S (1985) High pressure extraction of lignin from wood. In: Penninger JML, Radosz M, McHugh MA & Krukonis VJ (eds) Supercritical Fluid Technology. Elsevier, Amsterdam, p 385-396. Beer R & Peter S (1986) Lignin-extraktion aus holz mit alkylaminen unter druck. Chem- Ing-Tech 58(1): 72-73. Beigelman AV, Burov AV, Maltsev SM & Soldatenkova TS (1991) LTITsBP, assignee. Method of producing pulp. USSR patent 1.677.121. (ABIPST 63:6467(U)). Bennacer EM, Kiruyshina MF & Zarubin MY (1989) Effect of organic solvents on cleavage of beta-alkyl-0-arylic and glucosidic bonds kinetics. 1989 International Symposium on Wood and Pulping Chemistry Poster Sessions, Raleigh, NC, p 139-146. Berg L (1987) Celanese Chemical Co., assignee. Separation of formic acid from acetic acid by extractive distillation. US patent 4.692.219. Berg L (1988) Dehydration of formic acid by extractive distillation. US patent 4.786.370. Berg L (1990) Hoechst Celanese Chemical Co., assignee. Separation of formic acid from acetic acid by extractive distillation with acetyl salicylic acid. US patent 4.909.907. Berg L & Yeh AI (1987) Dehydration of formic acid by extractive distillation. US patent 4.642.166. Berg L, Szabados RJ, Wendt KM & Yeh AI (1990) The dehydration of the lower fatty acids by extractive distillation. Chem Eng Comm 89: 113-131. Berg A, Janssen W, Balle S, Kunz RG & Klein W (1995) Acetocell GmbH & Co. KG, assignee. Delignification of cellulosic raw materials using acetic acid, nitric acid and ozone. US patent 5.385.641. Bharati R, Bharati S, Singh SP, Karunanidhi MSP, Baskaran S & Paramsivam M (1989) Prospects of organosolv pulping of bagasse. Silver Jubilee International Seminar & Workshop: Appropriate Technologies for Pulp & Paper Manufacture in Developing Countries, 2: paper 34. (ABIPST 67:13930(AC)). Billa E, Kurek B, Koutsoula E, Monties B & Koukios E (1997) Effect of manganese peroxidase on the cell wall components of wheat straw and sorghum organosolv pulps. 9th International Symposium on Wood and Pulping Chemistry Oral Presentations, Montreal, Canada, p I2-1-4. Black NP (1991) ASAM alkaline sulfite pulping process shows potential for large-scale application. Tappi J 74(4): 87-93. Bogolitsyn KG, Ryabeva NV, Volkova NN & Wadso I (1994) Conformational transformations of dioxane lignin in organic solvents. Izv VUZ Lesn Zh (5/6): 109- 117. (ABIPST 67:475(AS)). Bonn G, Oefner PJ & Bobleter O (1988) Analytical determination of organic acids formed during hydrothermal and organosolv degradation of lignocellulosic biomass. Fresenius Zeitsch Anal Chemie 331(1): 46-50. Borgards A, Patt R, Kordsachia O, Odermatt J & Hunter WD (1993) A comparison of ASAM and kraft pulping and ECF/TCF bleaching of southern pine. TAPPI 1993 Pulping Conference, Atlanta, GA, p 629-636. Botello JI, Gilarranz MA, Rodriguez F & Oliet M (1998) Wood fractionation by cooking in alcohol-water media. Fifth European Workshop on Lignocellulosics Chemistry for Ecologically Friendly Pulping and Bleaching Technologies, Aveiro, Portugal, p 217- 220. (Paperbase 00522858). Bott K, Kaibel G, Hartmann H, Irnich R & Bulow H (1984) BASF AG, assignee. Verfahren zur destillativen gewinnung von ameisensaure. DE patent 33.19.651.A1. Bowen IJ & Hsu JCL (1990) Overview of emerging technologies in pulping and bleaching. Tappi J 73(9): 205-217. Bowers GH & April GC (1977) Aqueous n-butanol delignification of southern yellow pine. Tappi 60(8): 102-104. Brauns F & Hibbert H (1933) Studies on lignin, X: the identity and structure of spruce lignins prepared by different methods. J Am Chem Soc 55: 4720-4727. Brauns F & Hibbert H (1935) Studies on lignin and related compounds, XII: methanol lignin. Can J Res 13B: 28-34. Brickman L, Pyle JJ, McCarthy JL & Hibbert H (1939) Studies on lignin and related compunds, XXXIX: the ethanolysis of spruce and maple woods. J Am Chem So 61: 868-869. Brodersen KH, Dahlmann G & Leopold H (1992) Organocell Gesellschaft fur Zellstoff- und Umwelttechnik mbH, assignee. Verfahren zum delignifizieren von pflanzenfasermaterial. DE patent 41.03.572.A1. Brogdon BN, Dimmel DR & McDonough TJ (1997) Bleaching of softwood kraft pulps using an ethanol-based partial sequence. 9th International Symposium on Wood and Pulping Chemistry Oral Presentations, Montreal, Canada, p A2-1-4. Brolin A, Gierer J & Zhang Y (1991) Delignification of softwood kraft pulps by oxygen- containing species in acetic acid media. 6th International Symposium on Wood and Pulping Chemistry Proc., Melbourne, Australia, 2: 205-209. Bruniere PMF & Galichon JPM (1988) Societe Tag Pulp Industries S.A., assignee. Process for the pulping of lignocellulose materials with alkali or alkaline earth metal hydroxide or salt and a solvent. US patent 4.790.905. Brunt V (1992) The University of South Carolina, assignee. Process for recovering acetic acid from aqueous acetic acid solutions. US patent 5.175.357. Bublitz WJ (1987) Full chemical pulping with mixtures of methanol and sulfite liquors. Solvent Pulping - Promises & Problems, Appleton, WI, p 42-54. Bublitz WJ (1998a) Full chemical pulping with mixtures of methanol and acid sulfite liquors. 1998 Tappi Pulping Conf., Montreal, Canada, 2: 691-702. (Paperbase 00523482). Bublitz WJ (1998b) Full chemical pulping with mixtures of methanol and acid sulfite liquors. 1998 Breaking the Pulp Yiels Barrier Symp., Atlanta, GA, p 229-239. (Paperbase 00505158). Bublitz WJ & Hull JL (1981) A rapid process for production of high yield, full chemical sulfite pulp. 1981 Tappi Pulping Conf., Denver, CO, p 305-315. Bublitz WJ & Hull JL (1982) Recent developments in solvent-sulfite pulping. 1982 International Sulfite Pulping Conf., Toronto, Canada, p 263-268. Bublitz WJ & Wilson KP (1983) The role of methanol in a methanol acid sulfite pulping process. 1983 Tappi Pulping Conference Proc., p 423-427. Bublitz WJ, Gidlof P, Hull JL & Wilson K (1983) Recent developments in solvent sulfite pulping. International Symposium on Wood and Pulping Chemistry, Tokyo, Japan, 3: 38-43. Bucholtz M & Jordan RK (1983) Formic acid woodpulping could yield valuable chemical products. Pulp & Paper 57(9): 102-104. Buckland IK, Brauns F & Hibbert H (1935) Studies on lignin and related compunds, XVI: phenol lignin from spruce wood, from freudenberg spruce lignin and willstatter spruce lignin. Can J Res 13B(2): 61-77. Buhler F (1897) Verfahren zur darstellung von zellstoff mittels theereölen, bzw. den phenolen der theereolen. German patent 94467. Bulow H, Hohenschutz H, Schmidt JE & Sachsze W (1977) BASF AG, assignee. Verfahren zur reindarstellung von ameisensaure. DT patent 25.45.730.A1. Burkart LF (1989) Le Tourneau College, assignee. Rapid dissolution of lignin and other non-carbohydrates from ligno-cellulosic materials impregnated with a reaction product of triethyleneglycol and an organic acid. US patent 4.826.566. Burov AV, Beigelman AV, Lukanina TL, Maltsev SM & Vinogradova LI (1989a) Organosolv delignification of wood. Bum Prom (2): 15-16. (ABIPST 60:6317). Burov AV, Lukanina TL & Beigelman AV (1989b) Effect of acidity and the dielectric constant of the acetic acid-ethanol-water system on wood delignification. Khim Drev (Riga) (6): 27-30. (ABIPST 62:4297). Burov AV, Beigelman AV & Lukanina TL (1989c) Method of producing pulp. USSR patent 1.509.467. (ABIPST 62:3757(U)). Burov AV, Lukanina TL & Aliev RG (1990) Degradation of cellulose and changes in the molecular weight characteristics of lignin during delignification of wood in aqueous- ethanol solutions of acetic acid. Khim Drev (Riga) (6): 53-57. (ABIPST 62:13775). Burov AV, Lukanina TL, Aliev RG & Beigelman AV (1991) Method of delignification of wood chips. Russian patent 1.622.469. (ABIPST 64:8355(T)). Buryachenkov MM, Zakazov AN & Babkin VA (1991) Kinetics of water-ethanol delignification of trembling aspen [populus tremula] in the presence of acid catalysts. Irkutsk, Russia, Institut khimii drevesiny, Issledovaniya v oblasti khimii drevesiny 9. (ABIPST 65:6569(AC)). Calimli A & Olcay A (1978) Supercritical-gas extraction of spruce wood. Holzforschung 32(1): 7-10. Campbell WG (1929) CXXXVI, the degradation of wood by simultaneous action of ethyl alcohol and hydrochloric acid. Biochem J 23: 1225-1232. Caperos Sierra A, Romero Salvador A & Garcia-Ochoa Soria F (1990) Extraction of wood with solvents. Tappi J 73(11): 221-224. Chacon GM & Lai YZ (1985) Effect of ethanol on soda-anthraquinone pulping. 1985 International Symposium on Wood and Pulping Chemistry Technical Papers, Vancouver, Canada, p 291-294. Chang RP (1990) Pierre A. Tonachel, assignee. Method for continuous countercurrent organosolv saccharification of comminuted lignocellulosic materials. US patent 4.941.944. Chang JP & Lee JY (1989) Separation of wood main components by solvolysis, 2: characteristics of cooked wood meal and soluble sugar in water layer. J Tappik 21(2): 18-24. (ABIPST 60:6320). Chang PC & Paszner L (1982) Bau- und Forschungsgesellschaft Thermoform A.C., assignee. Pulping of lignocellulose with aqueous methanol/calyst mixture. CA patent 1131415. Chang HM, Jameel H, Wang S & Geng ZP (1994) Mixed peroxy acids as bleaching agents. Proc. of the International Symposium on Fiber Science and Technology, Yokohama, Japan, p 228-229. Chanzy H, Paillet M & Peguy A (1986) Spinning of exploded wood from amine oxide solutions. Polym Comm 27: 171-172. Chanzy H, Paillet M, Peguy A & Vuong R (1987) Dissolution and spinning of exploded wood in amine oxide systems. In: Kennedy JF (ed) Wood and cellulosics: industrial utilization, biotechnology, structure and properties. Ellis Horwood, Chichester, p 573- 579. Chaudhuri P (1996) Solvent pulping of bagasse - a process and system concenpt. Tappi Pulping Conference Proc., p 583-594. Chen F, Tanaka H, Naka Y & O’shima E (1989) Extraction of lower carboxylic acids from aqueous solutions by tri-n-octylamine. J Chem Eng Japan 22(1): 6-11. Chen ZT, Chen YI, Diebel JF, Chiang VL, Barna BA & Moffitt GE (1990) Pulp characteristics and mill economics for a conceptual SO2-ethanol-water mill. 1990 Tappi Pulping Conf., Toronto, Canada, 2: 663-672. Chiang VL, Moffitt GE, Puumala RJ & Sako T (1987) Linear regression modelling and response surface analysis of sulfur dioxide organosolv pulping. Solvent Pulping - Promises & Problems, Appleton, WI, p 68-92. Chiang VL & Sarkanen KV (1983) Ammonium sulfide organosolv pulping. Wood Sci Tech 17: 217-226. Cho H & Paszner L (1987) Processing of lignins to converging chemical by-products, part I, macrostructure and shape of ACOS organosolv hydrolysis lignin. In: Granger C (ed) 6th Canadian Bioenergy Research and Desing Seminar, p 364-367. Chum HL, Johnson DK, Ratcliff M, Black S, Schroeder HA & Wallace K (1985) Comparison between lignins produced by steam explosion and organosolv pretreatments. 1985 International Symposium on Wood and Pulping Chemistry Technical Papers, Vancouver, Canada, p 223-226. Chum HL, Black S, Johnson DK & Sarkanen KV (1987) Organosolv pretreatment for biomass-to-ethanol and lignins. Solvent Pulping - Promises & Problems, Appleton, WI, p 35-41. Chum HL, Johnson DK, Black S, Baker J, Grohmann K, Sarkanen KV, Wallace K & Schroeder HA (1988) Organosolv pretreatment for enzymatic hydrolysis of poplars, I: enzyme hydrolysis of cellulosic residues. Biotech Bioeng 31(7): 643-649. Chum HL, Johnson DK & Black SK (1989) Organosolv pretreatment. 1989 International Symposium on Wood and Pulping Chemistry Proc., Raleigh, NC, p 691-696. Chum HL, Johnson DK & Black SK (1990) Organosolv pretreatment of poplars, 2: catalyst effect and the combined severity parameter. Ind Eng Chem Res 29(2): 156-162. Chupka EI, Selentiev DG, Semenov SG & Khodyreva NV (1993) Oxidation of wood and its components in water-organic media. Seventh International Symposium on Wood and Pulping Chemistry Proc., Beijing, PR China, 3: 373-382. Clarke HT & Othmer DF (1928) Method of the dehydration of formic acid. US patent 1.826.302. Clermont LP & Bender F (1961) Effect of solutions of nitrogen dioxide, sulfur dioxide, and chlorine in dimethylformamide and dimethylsulfoxide on wood. Pulp Paper Mag Can 62(1): T28-T34. Cohen LR (1986) Badger B.V., assignee. Method for separating carboxylic acids from mixtures with non-acids. US patent 4.576.683. Compton J, Greig M & Hibbert H (1936) Studies on lignin and related compounds, XXII: fractionation of methanol lignin. Can J Res 14B(4): 115-119. Compton J & Hibbert H (1937) Studies on lignin and related compounds, XXVII: methylation and structure of methanol lignin (spruce). Can J Res 15B: 38-45. Cook PM & Hess SL (1991) Eastman Kodak Company, assignee. Organosolv lignin- modified phenolic resins and method for their preparation. US patent 5.010.156. Coppen JJW, Gibbs JA & Palmer ER (1988) A small-scale pulping process for developing countries? Paper Tech Ind 29(6): 298-299. Coulson JM, Richardson JF & Sinnott RK (1983) Chemical engineering vol 6. Pergamon Press, Oxford. Cramer AB, Hunter MJ & Hibbert H (1939) Studies on lignin and related compounds, XXXV: the ethanolysis of spruce wood. J Am Chem Soc 61: 509-516. Creamer AW, Blackner BA & Lora JH (1997) Properties and potential applications of a low-molecular-weight lignin fraction from organosolv pulping. 9th International Symposium on Wood and Pulping Chemistry Poster Presentations, Montreal, Canada, p 21:1-4. Cronlund M & Powers J (1990) Bleaching responses of alcell organosolv pulps. AIChE 1990 Annual Meeting, Chicago, IL. Cronlund M & Powers J (1991) Bleaching responses of alcell organosolv pulps using conventional and nonchlorine bleaching sequences. 1991 Tappi Pulping Conference Proc., Orlando, FL, 2: 547-552. Cronlund M & Powers J (1992) Bleaching of alcell organosolv pulps using conventional and nonchlorine bleaching sequences. Tappi J 75(6): 189-194. Curvelo AAS & Balogh DT (1995) Organosolv lignin extraction of Eucalyptus grandis, part III: successive and batch processes. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 2: 219-223. Curvelo AAS & Pereira R (1995) Kinetics of ethanol-water delignification of sugar cane bagasse. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 2: 473-478. Curvelo AAS, Alaburda J, Botaro VR, Lechat JR & Groote RAMC (1990a) Acetosolv pulping of sugar cane bagasse, analysis of the chloride content in the Acetosolv process of wood and related materials. Tappi J 73(11): 217-220. Curvelo AAS, Groote RAMC & Balogh DT (1990b) Solvent effect on organosolv lignin from Pinus caribaea hondurensis (part II). 1st European Workshop on Lignocellulosics and Pulp, Bergedorf, FR Germany, p 59-65. Curvelo AAS, Alaburda J, Botaro VR, Lechat JR & Groote RAMC (1990c) Analysis of the chloride content in the acetosolv process. Cell Chem Tech 24(3): 371-379. Curvelo AAS, Balogh DT & Groote RAMC (1992a) Organosolv pulping of Eucalyptus grandis. 25th Congresso Anual Celulose Papel ABTCP, Sao Paulo, Brazil, p 213-221. (ABIPST 65:5047(M)). Curvelo AAS, Groote RAMC & Montanari E (1992b) Dioxan lignins from Pinus caribaea var. hondurensis, part 1: effect of catalyst concentration. Paperi ja Puu 74(4): 324-327. Curvelo AAS, Balogh DT & Groote RAMC (1993) Organosolv lignin extraction from Eucalyptus grandis, part II: characterization of lignins obtained by batch process. Seventh International Symposium on Wood and Pulping Chemistry Proc., Beijing, PR China, 3: 177-182. Curvelo AAS, Araujo ES & Sansigolo CA (1995) Kinetics of organosolv pulping of Eucalyptus globulus. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 2: 213-218. Dahlmann G & Schroeter MC (1989) The Organocell process - pulping with the environment in mind. 1989 Solvent Pulping Conference, Seattle, WA, p 7-15. Dahlmann G & Schroeter MC (1990a) Pulping of spruce and pine with alcohol and alkali by the Organocell process. 1990 Tappi Pulping Conference, Toronto, Canada, 2: 657- 661. Dahlmann G & Schroeter MC (1990b) The Organocell process - pulping with the environment in mind. Tappi J 73(4): 237-240. Daima H, Hosoya S & Nakano J (1978) Studies on alkali-methanol cooking (part III) behavior of lignin during cooking. Japan Tappi 32(4): 49-52. Daima H, Hosoya S, Nakano J & Ishizu A (1979) Studies on alkali-methanol cooking (IV) - model experiments for the behaviors of lignin and carbohydrate during cooking. Japan Tappi J 33(6): 54-58. Danner RP & Daubert TE (1983) Manual for predicting chemical process design data. AIChE, New York. Davis JL (1987) Acetic acid pulping of wood. Ph.D. Thesis, Univ. of Wisconsin. (ABIPC 59:2752(DU)). Davis JL & Young RA (1989) Microwave-assisted solvent pulping. 1989 International Symposium on Wood and Pulping Chemistry Proc., Raleigh, NC, p 651-653. Davis JL & Young RA (1991a) Production and characterization of lignins from acetic acid based pulping processes, part 1: lignin preparation and molecular weight distribution. Holzforschung 45(Suppl.): 61-70. Davis JL & Young RA (1991b) Microwave-assisted solvent pulping. Holzforschung 45(Suppl.): 71-77. Davis JL, Young RA & Deodhar SS (1986) Organic acid pulping of wood III, acetic acid pulping of spruce. Mokuzai Gakkaishi 32(11): 905-914. Davis JL, Nakatsubo F, Murakami K & Umezawa T (1987) Organic acid pulping of wood IV: reactions of arylglycerol-b-guaiacyl ethers. Mokuzai Gakkaishi 33(6): 478- 486. Deineko IP (1992) Chemistry of oxygen-organosolv pulping. In: Kennedy JF, Phillips GO & Williams PA (eds) Lignocellulosics - Science, Technology, Development and Use. Ellis Horwood Limited, Chichester, p 239-246. Deineko I (1995) The chemical conversion of lignin during oxygen-organosolv delignification. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 2: 225-229. Deineko IP & Evtyugin DV (1988) Wood oxidation by molecular oxygen in organic solvents. Khim. Drev. (Riga) (6): 51-55. (ABIPC 59:11199). Deineko IP & Evtyugin DV (1989) Oxygen pulping of chips in aqueous acetone solutions. Khim Drev (Riga) (3): 106-107. (ABIPST 61:1278). Deineko IP & Kikitina OV (1989) Oxygen pulping of chips in water-alcohol solutions. Lesn Zh (1): 128-130. (ABIPST 61:7190). Deineko IP & Kostjukevich NG (1991) Products of pulping by oxygen in aqueous acetic acid solution. 6th International Symposium on Wood and Pulping Chemistry Proc., Melbourne, Australia, 2: 129-130. Deineko IP & Kostyukevich NG (1989) Method of producing pulp. USSR patent 1.490.199. (ABIPST 62:3758(U)). Deineko IP & Makarova OV (1991) Products of wood destruction by oxygen-alcoholic pulping. 6th International Symposium on Wood and Pulping Chemistry Proc., Melbourne, Australia, 2: 125-127. Deineko IP & Nikitina OV (1989) Oxidation of wood by oxygen in low molecular weight alcohols. Khim Drev (Riga) (4): 60-63. (ABIPST 61:8252). Deineko IP, Nikitina OV & Zarubin MY (1988a) RSFSR, assignee. Method of producing pulp. USSR patent 1.440.995. (ABIPST 60:3891(U)). Deineko IP, Evtyugin DV & Zarubin MY (1988b) Method of producing pulp. USSR patent 1.397.581. (ABIPC 59:5829(U)). Deineko IP, Evtuguin DV & Zarubin MY (1989) Oxygen-aceton wood pulping, products characterization and mechanism. 1989 International Symposium on Wood and Pulping Chemistry Proc., Raleigh, NC, p 719-726. Deineko IP, Makarova OV & Pranovich AV (1993) Characteristics of products of oxygen- alcohol pulping. Khim Drev (1-3): 45-50. (ABIPST 65:6572(AS)). Deinko IP, Makarova OV & Zarubin MY (1992) Delignification of wood by oxygen in low- molecular-weight alcoholic media. Tappi J 75(9): 136-140. DeLange M, Trummer JA, Roberts RS, Muzzy JD & Gelbaum LT (1991) 13CNMR characterization of tulip poplar lignin solubilized by buffered solvent pulping. Holzforschung 45(5): 361-365. DeLong EA (1990) Process to dissociate and extract the lignin and the xylan from the primary wall and the middle lamella or lignocellulosic material which retains the structural integrity of the fibre core. US patent 4.908.099. Delong EA & Delong EP (1991) Manufacture of bleached cellulose. Japan Kokai patent 249.290/91. (ABIPST 63:7424). DeLong EA, DeLong EP, Ritchie GS & Rendall WA (1990a) Tigney Technology Inc., assignee. Method for extracting the chemical components from dissociated lignocellulosic material. US patent 4.908.098. DeLong EA, DeLong EP, Ritchie GS & Rendall WA (1990b) Tigney Technology Inc., assignee. Method for extracting the chemical components from dissociated lignocellulosic material. CA patent 1.275.286. Demirbas A (1998) Aqueous glycerol delignification of wood chips and groundwood. Biores Tech 63(2): 179-185. (ABPST 69: 353(AS)). Desai MB (1981) Preliminary cost estimating of process plants. Chem Eng 88(15): 65-70. Detweiler RR (1992) Safety issues in design and operation of solvent pulping mills. 1992 Solvent Pulping Symposium Notes, Boston, MA, p 87-96. Diebold VB, Cowan WF & Walsh JK (1978) C P Associates Limited, assignee. Solvent pulping process. US patent 4.100.016. Diebold VB, Walsh JK & Cowan WF (1983) C P Associates Limited, assignee. Liuotinuutto-sellutusmenetelmä. FI patent 65290. Diebold VB, Cowan WF & Walsh JK (1988) Repap Technologies Inc., assignee. Losungsmittelextraktionsverfahren zur herstellung von cellulosepulpe. DE patent 26.37.449.C2. Dievskii VA & Burov AV (1993) Diffusion model of organosolv pulping kinetics. Izv VUZ Lesn Zh (5/6): 136-140. (ABIPST 66:726(AS)). Dillen S (1990) Nya massaprocesser: “kottkvarn” och sprit loser miljoproblem. Svensk Papperstidn 93(6): 38-44. Dovgan IV & Leonovich AA (1992) Softening features of soluble lignin preparations. Lesn Zh (6): 90-95. (ABIPST 65:465). Dovgan IV & Medvedeva EI (1993a) Studies of algae and sea-grasses lignins, 2: dioxane lignins from algae and sea-grasses. Khim Drev (1-3): 86-91. Dovgan IV & Medvedeva EI (1993b) Studies of algae and sea-grasses lignins, 3: nitrobenzene oxidation of dioxane lignins of the algae and sea-grasses. Khim Drev (1- 3): 92-95. Dovgan IV & Medvedeva EI (1993c) Studies of algae and sea-grasses lignins, 4: degradation of algae lignins by nucleophilic agents. Khim Drev (1-3): 96-100. Econompoulos AP (1978) Acetic acid dehydration: process and control. Chem Engineer (3): 199-202. Ede R, Brunow G, Poppius K, Sundquist J & Hortling B (1988) Formic acid/peroxyformic acid pulping, part 1: reactions of b-aryl ether model compunds with formic acid. Nordic Pulp Paper Res J 3(3): 119-123. Ede R & Brunow G (1989a) Formic acid/peroxyformic acid pulping, II: synthesis of 3-aryl- 2,3-dihydro-7-methoxy-2-benzofuranmethanols-model compunds for lignin acidolysis products. Holzforschung 43(2): 127-129. Ede R & Brunow G (1989b) Formic acid/peroxyformic acid pulping, III: condensation reactions of b-aryl ether model compunds in formic acid. Holzforschung 43(5): 317- 322. Ede R & Brunow G (1989c) Reactions of b-aryl ether model compunds under formic acid and peroxyformic acid pulping conditions. 1989 International Symposium on Wood and Pulping Chemistry Proc., Raleigh, NC, p 139-143. Edel E (1984) Das MD-organosolv-zellstoffverfahren. Dtsch Papierwirtsch (1): 39-45. Edel E (1989) Das organocell-verfahren - bericht über den betrieb einer demonstrations- anlage. Papier 43(10A): V116-V123. Edel E, Feckl J, Grambow C, Huber A & Wabner D (1986) MD-Organocell Gesellschaft fur Zellstoff- und Umwelttechnik mbH, assignee. Process for obtaining lignin from alkaline solutions thereof. US patent 4.584.076. Ekman K (1988a) Neste Oy, assignee. Menetelma ligniinin ja fenolin erottamiseksi fenoliliuoksista. FI patent 77061. Ekman K (1988b) Neste Oy, assignee. Menetelma fenolin kierrattamiseksi lignoselluloosa- materiaalien fenolikeittoprosessissa. FI patent 77062. Ekman K (1991a) Neste Oy, assignee. Procedure for separating lignin and phenol from phenol solutions. CA patent 1.284.491. Ekman K (1991b) Neste Oy, assignee. Procedure for recirculating phenol in a lignocellulosic materials digestion process. CA patent 1.282.714. El-Meadawy SA, Shukry N & Nassar MA (1990) Pulping with organic acids, I: acetic acid pulping of Casuarina sp. Papir Celuloza 45(10): V63-V67. Elnitskaya ZP, Burov AV, Aliev RG, Luferova LB & Turushev SL (1992) LTITsBP, assignee. Method of producing pulp. Russian patent 1.721.152. (ABIPST 63:14253(U)). El-Sakhawy M, Fahmy Y, Ibrahim AA & Lonnberg B (1995a) Organosolv pulping, (1) alcool pulping of bagasse. Cellul Chem Technol 29: 615-629. El-Sakhawy M, Lonnberg B, Ibrahim AA & Fahmy Y (1995b) Organosolv pulping, (2), ethanol pulping of cotton stalks. Cellul Chem Technol 29: 315-329. El-Sakhawy M, Lonnberg B, Fahmy Y & Ibrahim AA (1996a) Organosolv pulping, (3), ethanol pulping of wheat straw. Cellul Chem Technol 30: 161-174. El-Sakhawy M, Lonnberg B, Ibrahim AA & Fahmy Y (1996b) Organosolv pulping, (4), kinetics of alkaline ethanol pulping of wheat straw. Cellul Chem Technol 30: 281-296. El-Sakhawy M, Lonnberg B, Fahmy Y & Ibrahim AA (1996c) Organosolv pulping, (5), bleachability and paper properties. Cellul Chem Technol 30: 483-495. Engel O & Wedekind E (1933) Verfahren zum aufschluss von pflanzenfaserstoffen. German patent 581.806. Erbing H & Peter H (1941) Till kannedom om lignin. Kolloid-Zeitschrift 96: 47-71. (Svensk Papperstidning 44(22): 509) Erbring H (1939) Ueber den aufschluß von faserrohstoffen durch mehrwertige alkohole. Papier-Fabrikant 37(20): 168-170. Erbring H & Geinitz H (1938) Uber die aufschließung pflanzlicher faserstoffe mit mehrwertigen alkoholen. Kolloid-Zeitschrift 84(2): 215-222. Ettel W & Hnetkovky V (1965) Ceskoslovenska academie ved, assignee. Verfahren zur gewinnung von nativem lignin und halbzellstoff aus nadel- oder laubholz. DE patent 1.176.464. Evans JCW (1987) Repap plans alcohol pulping system at its newcastle kraft and lwc mill. Pulp Paper 61(11): 75-77. Evtuguin DV, Silvy J, Robert A & Zarubin MY (1993a) The pecularities of papermaking properties of the oxygen-organosolv pulp. Seventh International Symposium on Wood and Pulping Chemistry Proc., Beijing, P.R. China, 3: 167-172. Evtuguin DV, Deineko IP & Sheng LG (1993b) Oxygen-organosolv delignification: application for the hydrolysis lignin utilization. Seventh International Symposium on Wood and Pulping Chemistry Proc., Beijing, P.R. China, 3: 577-582. Evtuguin DV, Shatalov AA & Kostukevich NG (1995) Degradation of polysaccharides under conditions of oxygen delignification in organic solvent media. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 2: 325-330. Evtuguin DV, Robert A, Zarubin MY & Deineko IP (1996) The effect of oxygen- organosolv pulping conditions on pulp properties. Cell Chem Tech 30(1/2): 105-115. Evtyugin DV & Deineko IP (1994) Low-molecular reaction products from oxygen-acetone wood delignification. Wood Chem (3):46-52. (ABIPST 66:7048(AS)). Evtyugin DV, Shatalov AA & Zarubin MY (1994a) Degradation of polysaccharides in the presence of oxygen in different types of aqueous organic media, (1) change in the chemical composition of the carbohydrate complex during oxygen-organosolv delignification. Izv VUZ Lesn Zh (3):64-70. (ABIPST 66:3609(AS)). Evtyugin DV, Shatalov AA & Zarubin MY (1994b) Degradation of polysaccharides in the presence of oxygen in different types of aqueous organic media, (1) depolymerization of the polysaccharide complex during oxygen-organosolv delignification. Izv VUZ Lesn Zh (3):70-76. (ABIPST 66:3608(AS)). Faass GS, Roberts RS & Muzzy JD (1989) Buffered solvent pulping. Holzforschung 43(4): 245-250. Fagerlind O & Agren E (1995) A. Ahlstrom Corporation, assignee. Process for the concentration of spent liquors. US patent 5.382.321. Fair JR (1987) Distillation. In: Rousseau RW (ed) Handbook of separation process technology. Wiley-Interscience, New York, p 229-339. Faix O, Schubert HL & Patt R (1989) Continuous process control of pulping by ftir spectroscopy. 1989 International Symposium on Wood and Pulping Chemistry Proc., Raleigh, NC, p 1-8. Faix O, Jakab E, Fuchs K, Patt R, Till F & Szekely T (1990) Thermal behaviour of ASAM waste liquors. Paperi Puu 72(3): 239-247. Fales G (1988) Repap readies alcell trial plant for further expansion. Paper Age 104(10): 40. (ABIPC 59:7193). Farrand JC & Johnson DC (1971) Peroxyacetic acid oxidation of 4-methylphenols and their methyl ethers. J Org Chem 36(23): 3606-3612. Feckl J & Edel E (1987) Organosolv pulping with addition of alkali - procedure and products. Fourth International Symposium on Wood and Pulping Chemistry Proc., Paris, France, 1: 369-372. Feehl J (1984) Studies of soluble by-products resulting from organosolv-pulping. In: Ferrero GL (ed) Anaerobic digestion and carbohydrate hydrolysis of waste. Elsevier Applied Science Publishers Ltd, Essex, p 440-443. Feldman D, Banu D, El-Raghi S & Lora J (1996) Rigid poly(vinyl chloride)-organosolv lignin blends for applications in building. J Appl Polym Sci 61(12): 2119-2128. (ABIPST 67:5651(AS)). Fengel D, Wegener G & Greune A (1989) Studies on the delignification of spruce wood by organosolv pulping using sem-exda and tem. Wood Sci Tech 23(2): 123-130. Ferraz A, Mendonca R, Cotrim AR & Silva FT (1996) The use of white-rot decay as a pretreatment for organosolv delignification of eucalyptus grandis wood. In: Srebotnik E & Messner K (eds) Biotechnology in pulp and paper industry. Facultas- Universitatsverlag, Vienna, p 221-224. Ferraz A, Rodriguez J, Freer J & Baeza J (1998a) Fungal-organosolv pulping of pinus radiata softwood. Proc. 7th International Conference on Biotechnology in the Pulp and Paper Industry, Vancouver, Canada, B: 29-32. (Paperbase 00527145). Ferraz A, Christov L & Akhtar M (1998b) Fungal pretreatment for organosolv pulping and dissolving pulp production. In: Young RA & Akhtar M (eds) Environmentally Friendly Technologies for the Pulpo and Paper Industry. John Wiley & Sons, New Youk, NY, p 421-447. Fien GJAF & Liu YA (1994) Heuristic synthesis and shortcut design of separation processes using residue curve maps: a review. In Eng Chem Res 33: 2505-2522. Finnish Oil Association (1998) Http://www.oil.fi/keskusliitto1.html. Franzreb JP (1989) Was erwartet die klassische zellstoffindustrie von den neuen herstellungsverfahren? Papier 43(10A): V94-V97. Fredenslund A, Gmehling J & Rasmussen P (1977) Vapor-liquid equilibria using UNIFAC, a group contribution method. Elsevier, Amsterdam. Freudenberg K, Janson A, Knopf E & Haag A (1936) Zur kenntnis des lignins (15. mitteil.). Ber Dtsch Chem Gesellsch 69: 1415-1425. Freudenberg K & Acker L (1941) Uber die einwirkung von glycolchlorhydrin auf fichtenlignin. Ber Dtsch Chem Gesellsch 74(8): 1400-1406. Friedman L & McCully CR (1938) Isolation of lignin with benzyl alcohol. Paper Trade J 107(26): 28-29. Frierman M, Kuo Y, Joshi D, Garcia AA & King CJ (1987) Use of adsorbents for recovery of acetic acid from aqueous solutions, part III - solvent regeneration. Separ Purif Meth 16(1): 91-102. Fuchs W (1929) Uber den aufschluss des lignins mit methylglykolischer salzsaure. Ber Dtsch Chem Gesellsch 62B: 2125-2133. Fuchs W (1936) Investigations concerning phenol lignin and methoxy glycol lignin from spruce wood. J Am Chem Soc 58: 673-680. Fullerton TJ (1983) The hydrolysis of hard beech sawdust with aqueous ethanol. FRI Bulletin 38. Funaoka M & Abe I (1989) Rapid separation of wood into carbohydrate and lignin with concentrated acid-phenol system. Tappi J 72(8): 145-149. Gallagher DK, Hergert HL, Cronlund M & Landucci LL (1989) Mechanism of delignification in an autocatalyzed solvolysis of aspen. 1989 International Symposium on Wood and Pulping Chemistry Proc., Raleigh, NC, p 709-718. Garner FH, Ellis SRM & Pearce CJ (1955) Extraction of acetic acid from water, 4: ternary- vapour-liquid equilibrium data. Chem Eng Sci 4: 273-278. Gasche UP (1985a) Schwefelfreie katalysierte verfahren zur erzeugung von zellstoff. Papier 39(10A): V1-V8. Gasche UP (1985b) Schwefelfreies verfahren zur erzeugung von zellstoff aus holz. Swiss Biotech 3(1a): 47-52. Gasche U (1990) Hat das sulfitverfahren heute noch eine existenzberechtigung?, vergleich des sulfitprozesses mit den neuen aufschlussverfahren. Papier 44(3): 85-91. Gast D, Ayla C & Puls J (1983) Component separation of lignocelluloses by organosolv- treatment. In: Strub A (ed) Energy from biomass, 2nd E.C. Conf. Applied Science Publishers Ltd, Barking, p 879-882. Gast D & Puls J (1984) Ethylene glycol-water pulping, kinetics of delignification. In: Ferrero GL (ed) Anaerobic digestion and carbohydrate hydrolysis of waste. Elsevier Applied Science Publishers Ltd, Essex, p 450-453. Gast D & Puls J (1985) Improvement of the ethylene glycol water system for the component separation of lignocelluloses. In: Palz W (ed) Energy from biomass, 3rd E.C. Conf. Elsevier Applied Science Publishers Ltd, Essex, p 949-951. Gentry JC & Solazzo, AJ (1995) Recovery of carboxylic acids from aqueous streams. Environmental Prog 14(1): 61-64. Ghose TK, Pannir Selvam PV & Ghosh P (1983) Catalytic solvent delignification of agricultural residues: organic catalysts. Biotech Bioeng 25(11): 2577-2590. Gilardi G, Harvey PJ, Cass AEG & Palmer JM (1990) NMR studies of lignin degradation. In: Kirk TK & Chang HM (eds) Biotechnology in pulp and paper manufacture, applications and fundamental investigations. Butterworth-Heinemann, Boston, MA, p 367-373. Gilarranz M, Oliet M, Rodriguez F, Poveda P, Villar J & Barbadillo P (1997) Preliminary study on the manufacture and bleaching of organosolv pulps from Eucalyptus globulus. Invest Tech Pap 34: 600-619. (ABIPST 68:11153(AS)). Girard RD & Chen R (1992) Optimization of an organosolv process for white birch. 1992 Solvent Pulping Symposium Notes, Boston, MA, p 73-78. Girard RD & Heiningen ARP (1997) Yield determination and mass balance closure of batch ethanol-based solvent cookings. Pulp & Paper Canada 98(8): 65-68. Glasenapp A, Kordsachia O, Odermatt J & Patt R (1996) TCF-bleiche von ASAM-fichten- zellstoffen mit und ohne ozon. Papier 50(6): 300-309. Glasner A, Reinharter E & Schubert, HL (1994) Chemical energy and water balance of ASAM pulping. TAPPI Pulping Conference, San Diego, CA, p 155-165. Gliese T, Kleemann S & Fischer K (1996) Enzymatic bleaching of organosolv-pulp. In: Srebotnik E & Messner K (eds) Biotechnology in pulp and paper industry. Facultas- Universitatsverlag, Vienna, p 63-67. Golob J, Grilc V & Zadnik B (1981) Extraction of acetic acid from dilute aqueous solutions with trioctylphosphine oxide. Ind Eng Chem Proc Des Dev 20: 433-435. Goncalves AR, Urbano MP, Cotrim AR & Silva FT (1997) Oxidation of lignin-containing liquor from acetosolv pulping of sugar-cane bagasse. 9th International Symposium on Wood and Pulping Chemistry Oral Presentations, Montreal, Canada, p I4 1-3. Gosselink RJA, Dam, JEG & Zomers FHA (1995) Combined hplc analysis of organic acids and furans during organosolv pulping of fiber hemp. J Wood Chem Tech 15(1): 1-25. Gottlieb K, Preuss AW, Meckel J & Berg A (1992) Acetocell pulping of spruce and chlorine-free bleaching. 1992 Solvent Pulping Symposium Notes, Boston, MA, p 35- 39. Goyal GC, Lora JH & Pye EK (1991) Autocatalyzed organosolv pulping of north american hardwoods: effect of pulping conditions on the pulp properties and characteristics of the soluble and residual lignin. 1991 Tappi Pulping Conf. Proc., Orlando, FL, 2: 627- 635. Goyal GC, Lora JH & Pye EK (1992) Autocatalyzed organosolv pulping of north american hardwoods: effect of pulping conditions on the pulp properties and characteristics of the soluble and residual lignin. Tappi J 75(2): 110-116. Goyal GC & Lora JH (1991) Kinetics of delignification and lignin characteristics in autocatalyzed organosolv pulping of hardwoods. 6th International Symposium on Wood and Pulping Chemistry Proc., Melbourne, Australia, 1: 205-212. Gray KR, King EG, Brauns F & Hibbert H (1935) Studies on lignin and related compounds, XIII, the structure and properties of glycol lignin. Can J Res 13B: 35-47. Green J & Sanyer N (1982) Alkaline pulping in aqueous alcohols and amines. Tappi 65(5): 133-137. Griffith WL, Compere AL, Rao TK & Griest WH (1980) Potential use of organosolv underflow tar lignin as a petrochemical refinery feed. Tappi Papermakers Conference, Atlanta, GA, p 181-185. Grinstead RR (1974) The Dow Chemical Company, assignee. Extraction of carboxylic acids from dilute aqueous solutions. US patent 3.816.524. Grondal BL & Zenczac P (1956) Research Corporation, assignee. Process of pulping cellulosic materials with triethylene glycol. US patent 2.772.968. Groote RAMC, Curvelo AAS, Frangiosa PC, Pinto MFT & Zambon MD (1991a) Molecular weight distribution of sugar cane bagasse lignin in the acetosolv process. 6th International Symposium on Wood and Pulping Chemistry Proc., Melbourne, Australia, 2: 385-391. Groote RAMC, Curvelo AAS, Alaburda J & Botaro VR (1991b) The reactivity of sugar- cane-bagasse lignin from the acetosolv pulping process. In: Faix O & Meier D (eds) Mitteilungen der Bundesforschungsanstalt fur Forst- und Holzwirtschaft, 168: 228-233. Groote RAMC, Curvelo AAS, Alaburda J & Botaro VR (1992a) Acetosolv pulping of sugarcane bagasse: use of the pulp and the lignin. Papel 53(1): 43-46. (ABIPST 63:2811). Groote RAMC, Curvelo AAS, Frangiosa PC & Zambon MD (1992b) Some properties of acetosolv sugar cane bagasse lignin. Cellul Chem Tech 26(1): 53-61. Guha SRD, Sirmokadam NN, Goudar TR & Dhopeshwar R (1987) Pulping of rice straw by pollution free organosolv process. IPPTA 24(1): 24-26. Haars A, Majcherczyk A, Trojanowski J & Huttermann A (1985) Bioconversion of organosolv lignins by different types of fungi. In: Palz W (ed) Energy from biomass, 3rd e.c. conference. Elsevier Applied Science Publishers Ltd, Essex, p. 973-977. Haas GG & Lang CJ (1971) Weyerhaeuser Company, assignee. Non-catalytic process for the production of cellulose from lignocellulosic materials using acetic acid. US patent 3.553.076. Haas GG & Lang CJ (1974) Delignification with ketones and ammonia. Tappi 57(5): 127- 130. Haas H, Schoch W & Strole U (1955) Herstellung von skelettsubstanzen mit peressigsaure. Papier 9(19/20): 469-475. Haggin J (1986) Ester pulping process avoids use of sulfur compounds. Chem Eng News 64(4): 25-26. Hagglund E & Urban H (1927) Zur kenntnis der ligninacetale I. Cellulosechemie 8(7): 69- 71. Hagglund E & Urban H (1928) Zur kenntnis der ligninacetale II. Cellulosechemie 9(6): 49- 53. Hakansson H, Sjoberg LA & Ahlgren P (1996) Autocatalyzed ethanol pulping of reed canary grass. Nord Pulp Pap Res J 11(2): 100-104. (ABIPST 67:3109(AS)). Hammann M, Kordsachia O & Patt R (1991) Bleaching of kraft pulp and asam pulp without chlorine containing chemicals. International Pulp Bleaching Conference Proc., Stockholm, Sweden, 3: 185-200. Hansen SM & April GC (1982) Prediction of solvent effects in aqueous-organic solvent delignification. Ind Eng Chem Prod Res Dev 21(4): 621-626. Harrison A (1991) Repap produces high-quality pulp at newcastle with alcell process. Pulp & Paper 65(2): 116-119. Hartmuth R (1920) Verfahren zur behandlung von holz oder zellulosehaltigen stoffen zwecks gewinnung von zellulose und kunstlichem harz, lack, asphalt u. dgl. German patent 326705. Hawkes GE, Smith CZ, Utley JHP & Chum HL (1986) Key structural features of acetone- soluble phenol-pulping lignins by 1H and 13C N.M.R. spectroscopy. Holzforschung 40(Suppl.): 115-123. Heckrodt W & Thompson N (1992) Biodyne Chemical Inc., assignee. Pulping. CA patent 2.055.040. (ABIPST 66:14811(P)). Heden S & Holmberg B (1936) Bisulfitkok med aromatiska alkoholer. Svensk Kemisk Tidskrift 48: 207-211. Helsel RW (1977) Waste recovery: removing carboxylic acids from aqueous wastes. Chem Eng Prog 73(5): 55-59. Henry WR (1991) The future of organosolv pulping. 1991 Tappi Pulping Conference Proc., Orlando, FL, 2: 653 - 656. Henry W & Seely B (1991) DCS challenges at the alcell pre-commercial facility. 77th CPPA Annual Meeting Prep., Montreal, Canada, p A395-A398. Henry W & Seely B (1992) DCS challenges at the alcell pre-commercial facility. Pulp & Paper Canada 93(12): 143-145. Herdle LL, Pancoast LH & MacClaren RH (1964) Acetylation celluloses from pulping of wood in acetic acid. Tappi 47(10): 617-620. Hewson WB, McCarthy JL & Hibbert H (1941a) Studies on lignin and related compounds, LVII, mechanism of the ethanolysis reaction. J Am Chem Soc 63: 3041-3045. Hewson WB, McCarthy JL & Hibbert H (1941b) Studies on lignin and related compounds, LVIII, the mechanism of the ethanolysis of maple wood at high temperatures. J Am Chem Soc 63: 3045-3048. Hibbert H & Marion L (1930) Studies on lignin and related compounds, II., glycol-lignin and glycol-ether-lignin. Can J Res 2(6): 364-375. Hibbert H & Phillips JB (1930) Studies on lignin and related compounds, III, glycerol- chlorohydrin-lignin. Can J Res 3: 65-69. Hibbert H & Rowley HJ (1930) Studies on lignin and related compounds, I., a new method for the isolation of spruce wood lignin. Can J Res 2(6): 357-363. Hill RT, Walsh PB & Hollie JA (1992) Part 1: Peracetic acid - an effective alternative for chlorine compound free delignification of kraft pulp. 1992 Pulping Conference Proc., Boston, MA, p 1219-1230. Hillmer A (1925) Die loslichkeit von lignin in phenolen. Cellulosechemie 6(11): 169- 187. Hinrichs DD, Knapp SB, Milliken JH & Wethern JD (1957) Sugarcane bagasse as a fibrous papermaking material, VI, modified alkaline and hydrotropic pulping of bagasse. Tappi 40(8): 626-633. Hirose S, Yano S, Hatakeyama H & Hatakeyama T (1989) Heat-resistant polyurethanes from solvolysis lignin. In: Glasser W & Sarkanen K (eds) Lignin properties & materials. ACS, Washington, DC, p 82-99. (ABIPST 60:8967(B)). Hohenschutz H, Schmidt J & Kiefer H (1985) BASF AG, assignee. Verfahren zur gewinnung von carbonsauren aus ihren waßrigen losungen. DE patent 25.45.658.C2. Holmberg B (1936) Lignin-untersuchungen, XI mitteil., fichtenholz und mercapto-sauren. Berichte der Deutschen Chemischen Gesellschaft 69: 115-119. Holmberg B (1947) Espenholzer und mercaptosauren. Arkiv for Kemi, Mineralogi och Geologi 24A(29): 1-11. Holmberg B & Runius S (1925) Alkoholys af tra. Svensk Kemisk Tidskrift 37(8): 189- 197. Holtzapple MT & Humphrey AE (1984) The effect of organosolv pretreatment on the enzymatic hydrolysis of poplar. Biotech Bioeng 26(7): 670-676. Horlenko T (1973) Celanese Corporation, assignee. Distillation process for recovering substantially anhydrous formic acid from aqueous solutions. US patent 3.718.545. Hortling B, Ranua M & Sundquist J (1987) Influence of the pulping process on the residual lignin in kraft and peroxyacid pulps. Fourth International Symposium on Wood and Pulping Chemistry Proc., Paris, France, 2: 203-210. Hortling B, Poppius K & Sundquist J (1989) Comparison of the lignins in kraft and peroxyacid fibers with lignin in corresponding spent liquors. 1989 International Symposium on Wood and Pulping Chemistry Proc., Raleigh, NC, p 279-289. Hortling B, Poppius-Levlin K, Kataja K, Backman I & Sundquist J (1990) Spent liquor lignins from peroxyformic acid pulping compared with products from milled wood and kraft lignin obtained under similar conditions (abstract). 1st European Workshop on Lignocellulosics and Pulp, Hamburg, Germany, p 5. Hortling B, Poppius K & Sundquist J (1991) Formic acid/peroxyformic acid pulping, IV, lignins isolated from spent liquors of three-stage peroxyformic acid pulping. Holzforschung 45(2): 109-120. Hossain S (1958) Action of dimethyl sulphoxide on wood and sulphite pulps. Pulp and Paper Magazine of Canada 59(8): 127-130. Hossain S (1959) Abitibi Power & Paper Company, assignee. Process for pulping lignocellulosic material. CA patent 573.329. Hou SZ & An YQ (1989) Study of the defibration point and kinetics of pulping wheat straw with HMDA (hexamethylene diamine). J Northwest Inst Light Ind 7(1): 69-79. (ABIPST 62:8467(U)). Hudeczek W & Lindow J (1992) Verwendung von methanol in der zellstoff-erzeugung - einflussfaktoren auf die planung. Papier 46(7): 335-340. Hultholm T, Robertsen L, Lonnberg B, Henricson K & Laxen T (1998) The IDE-process, a new concept suitable for softwood, hardwood and nonwood. Fifth European Workshop on Lignocellulosics Chemistry for Ecologically Friendly Pulping and Bleaching Technologies, Aveiro, Portugal, p 577-580. (Paperbase 00522911). Hunsmann W & Simmrock KH (1966) Trennung von wasser, ameisensaure und essigsaure durch azeotrop destillation. Chemie-Ingenieur-Technik 38(10): 1053-1059. Hunter MJ, Cramer AB & Hibbert H (1939) Studies on lignin and related compounds, XXXVI, ethanolysis of maple wood. J Am Chem Soc 61: 516-520. Huth SP & Cole BJW (1994) Synthesis and nmr characterization of a lignin-poly(ethylene glycol)methyl ether copolymer. Holzforschung 48(1): 23-28. Isoaho M (1998) Verbal communication. City of Oulu, Oulu. Ivanow T & Robert A (1987) Delignification via solvolysis: role of solvent and of different additives. 4th International Symposium on Wood and Pulping Chemistry Proc., Paris, France, 2: 189-192. (ABIPC 59:8216(M)). Ivanow T & Robert A (1989) Delignification of hornbeam [carpinus] wood by alkaline solvolysis; ethanol-water solutions. Papeterie (132): 22-27. (ABIPST 60:7418). Jaaskelainen AS (1996) Peroksiyhdisteiden kaytto havupuusulfaattimassan valkaisussa. PSC Comm 88. Jaaskelainen AS & Poppius-Levlin K (1997) Factors affecting the kinetics of pine kraft pulp delignification with peroxyacetic acid. 9th International Symposium on Wood and Pulping Chemistry Oral Presentations, Montreal, Canada, D5 1-5. Jagirdar GC & Sharma MM (1980) Recovery and separation of mixtures of organic acids from dilute aqueous solutions. J Sep Proc Tech 1(2): 40-43. Jahn H, Hackethal U & Fleck U (1979) VEB Chemische Werke Buna, assignee. Verfahren zur schleppmittelfreien destillativen auftrennung waßriger essigsaure-ameisensaure- gemische. DD patent 133.559. Jakab E, Faix O, Till F & Székely T (1991) Thermogravimetry/mass spectrometry of various lignosulfonates as well as of kraft and acetosolv lignin. Holzforschung 45(5): 355-360. Jamieson S (1991) Alcell pulping: world class research rigth here in canada. Pulp & Paper Canada 92(3): 16-18. Janson J & Vuorisalo R (1986) Alcohol-hydroxide-anthraquinone pulping. Paperi ja Puu 68(9): 610-615. Jende UL (1992) The industrial-scale implementations of the Organocell process in Kelheim. 1992 Solvent Pulping Symposium Notes, Boston, MA, p 79-85. Jeong HC & Park CY (1990) Kinetic studies on solvolysis cooking reaction. J TAPPIK 22(2): 31-44. (ABIPST 62:4318). Jimenez L, Maestre F, Torre MJ & Perez I (1997a) Organosolv pulping of wheat straw by use of methanol-water mixtures. Tappi J 80(12): 148-154. Jimenez L, Torre M, Maestre F & Perez I (1997b) Organosolv pulping of wheat straw by use of phenol. Bioresour Technol 60(3): 199-205. (ABIPST 68:3905(AS)). Jiwei C, Dazhen Z, Yulin W, Junsuo W, Ruisheng Z & Znongren S (1988) A preliminary study of organosolv pulping wheat straw with acid-acetate process. 1988 International Non-Wood Fiber Pulping and Papermaking Conference Proc., Beijing, PR China, 1: 180-186. Johansson A, Sachetto JP & Roman A (1983) A new process for the fractionation of biomass. In: Strub A (ed) Energy from biomass, 2nd e.c. conference. Applied Science Publishers Ltd, Barking, p 868-872. Johansson A, Ebeling K, Aaltonen O, Aaltonen P & Ylinen P (1987) Wood pulping with aqueous tetrahydrofurfurylalcohol. Paperi ja Puu 69(6): 500-504. Johnson DC (1975) Lignin reactions in delignification with peroxyacetic acid. 1st International Symposium on Delignification with Oxygen, Ozone and Peroxides, Raleigh, NC, p 217-228. Jordan RK (1982) Formic acid pulping. PCT patent 82/01902. Julien LM (1979) Characteristics of pulps obtained from HMDA pulping. 1979 Tappi Pulping Conference Proc., p 23-29. Julien LM & Barna BA (1980) Amine pulping with 1,6-diaminohexane. In: New process alternatives in the forest products industries. AIChE Symposium Series 76: 209-216. Julien LM & Sun BCH (1979) Pulping with amine and soda-amine systems. Tappi 62(8): 63-65. Kachi S & Tsuchiya H (1988) Characteristics of solvolysis pulping. CPPA Spring Conference Preprints Session 5 Paper 3, Jasper, Alta. (ABIPST 62:4320(M). Kachi S & Tsuchiya H (1989) Research and development of solvolysis pulping, (1), characteristics of the process and the pulp. Japan Tappi J 43(9): 905-910. (ABIPST 60:6328). Kachi S, Iwata H, Araki H & Aoyagi T (1990) Research and development of solvolysis pulpung, (6), delignification of wood chips with phenolic cooking solvent. Mokuzai Gakkaishi 36(2): 108-115. (ABIPST 61:9772) Kakemoto G, Sagara H, Suzuki N & Kachi S (1990) JGC Corporation and Japan Pulp & Paper Research Institute, assignees. Method of manufacturing phenols from lignin. US patent 4.900.873. Kalb L & Schoeller V (1923) Zur frage der cellulose-bestimmung mit phenol. Cellulosechemie 4(4): 37-40. Kaminsky K, Schweers W & Schwesinger H (1980) Properties and decomposition of lignins isolated by means of an alcoholic-water-mixture, part 3: decomposition by pyrolysis in a fluid-bed. Holzforschung 34(3): 73-76. Kaneko N, Sano Y, Sasaya T & Otani J (1991) Phenorganosolv pulping, (3), pulping of softwoods with acetic acid-water-phenols system. Japan Tappi J 45(12): 1392-1399. (ABIPST 63:1865). Kanitskaya LV, Zakazov AN, Rossinskii OA, Rpkhin AV & Babkin VA (1993) Structural changes in spruce [picea] dioxane lignin under the action of hydrogen peroxide and sodium borohydride. Izv Vyssh Uchebn Zaved Lesn Zh (2/3): 147-154. (ABIPST 65:6393(AS)). Karl W (1988) Repap’s Alcell process: is it all it’s cracked up to be? Pulp & Paper J 41(7): 17, 21. Karl W (1991a) By George, they’ve done it. Pulp & Paper J 44(1): 106-109. Karl W (1991b) Repap’s Alcell process. Paper 215(1): 46-47. Karlsson O & Lundquist K (1991) Chemical modification of lignins during organosolv pulping in acid media. In: Faix O & Meier D (eds) Mitteilungen der Bundesforschungsanstalt fur Forst- und Holzwirtschaft, 168: 66-68. Karmanov AP, Belyaev VY & Monakov YB (1992) Hydrodynamic properties of a dehydropolymer-macromolecular model of lignin. Khim Drev (Riga) (2-3): 25-31. (ABIPST 64:238). Katzen R, Frederickson R & Brush BF (1980a) The alcohol pulping & recovery process. Chem Eng Prog (2): 62-67. Katzen R, Frederickson RE & Brush BF (1980b) Alcohol pulping appears feasible for small incremental capacity. Pulp & Paper (8): 144-149. Kawamoto H, Chang H & Jameel H (1994) Reaction of lignin model compounds with peracetic acid or peroxymonosulfuric acid. The International Symposium on Fiber Science and Technology Proc., Yokohama, Japan, p 230. Kawamoto H, Chang HM & Jameel H (1995) Reactions of peroxyacids with lignin and lignin model compounds. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 1: 383-390. Kayama T & Takagi H (1978) Alkali-dioxane pulping (part I), cooking conditions and pulp properties. Japan Tappi J 32(1): 31-38. Kim DH & Paik KH (1993) Peroxide bleaching of [mongolian] oakquercus [ mongolica] acetosolv pulp. J TAPPIK 25(1): 53-59. (ABIPST 64:7596). Kim DH, Paik KH & Kang CH (1991) Acetosolv pulp at atmospheric pressure for the reduction of pollution and energy. J TAPPIK 23(2): 21-32. Kim YH, Shimokawa A, Sakai K & Takahashi I (1996) Effect of organic solvents on the yield of low-molecular-weight phenols during organosolv pulping. Proc. 1996 (63rd) Pulp and Paper Research Conference, p 164-169. (ABIPST 67:6460(AC)). Kin Z (1990) The acetolysis of beech wood. Tappi J 73(11): 237-238. King CJ (1986) Separation processes. Tata McGraw-Hill Publishing Company Limited, New Delhi. King CJ (1992) Amine-based systems for carboxylic acid recovery. Chemtech 22(5): 285 - 291. Kinnula T (1986) Sulfaattisellutehtaan mitoitus ja kannattavuus. Univ Oulu, Dept Process Eng, Report 103. Kirçi H, Bostançi S & Yalinkiliç MK (1994) A new modified pulping process alternative to sulphate method “alkali-sulfite-anthraquinone-ethanol (ASAE)”. Wood Sci Tech 28(2): 89-99. Kirsch T & Maurer G (1997) Distribution of binary mixtures of citric, acetic and oxalic acid between water and organic solutions of tri-n-octylamine, part I, organic solvent toluene. Fluid Phase Equilibria 131: 213-231. Kirsch T & Maurer G (1998) Distribution of binary mixtures of citric, acetic and oxalic acid between water and organic solutions of tri-n-octylamine, part II, organic solvent methylisobutylketone. Fluid Phase Equilibria 142: 215-230. Kirsch T, Ziegenfuss H & Maurer G (1997) Distribution of citric, acetic and oxalic acids between water and organic solutions of tri-n-octylamine. Fluid Phase Equilibria 129: 235-266. Klason P (1893a) Bidrag till kannedomen om sammansattningen af granens ved samt de kemiska processerna vid framstallning af cellulosa darur. Teknisk Tidskrift, Afdelningen for Kemi och Metallurgi 23(2): 17-22. Klason P (1893b) Framstallning af rent lignin ur granved och denna sednares kemiska sammansattning. Teknisk Tidskrift, Afdelningen for Kemi och Metallurgi 23: 55-56. Klason P (1908) Bidrag till kannedom om granvedens kemiska sammansattning, I. Arkiv for Kemi, Mineralogi och Geologi 3(5): 1-20. Klason P & Fagerlind O (1908) Bidrag till kannedom om granvedens kemiska sammansattning, II. Arkiv for Kemi, Mineralogi och Geologi 3(6): 1-10. Kleinert T (1933) Zur kenntnis der dampf-flussigkeitsgleichgewichte von aethylalkohol- wassergemischen bei temperaturen von 120 bis 180 C. Beihefte zu den zeitschriften des vereins deutscher chemiker nr 2. Verlag Chemie G.M.B.H., Berlin. Kleinert T (1940) Uber das verhalten von cellulose bei der druckerhitzung in alkohol- wassermischungen. Cellulosechemie 18(5): 114. Kleinert TN (1967) Thermischer holzaufschluss in alkohol-wassermischungen. Holzforschung und Holzverwertung 19(4): 60-65. Kleinert TN (1971) Organosolv pulping and recovery process. US patent 3.585.104. Kleinert TN (1974a) Organosolv pulping with aqueous alcohol. Tappi 57(8): 99-102. Kleinert TN (1974b) Der alkohol-organosolv-aufschluss des holzes. Das Oesterreichische Papier 11(11): 14-19. Kleinert TN (1975a) Ethanol-water delignification of sizable pieces of wood, disintegration into stringlike fiber bundles. Holzforschung 29(3): 107-109. Kleinert TN (1975b) Ethanol-water delignification of wood - rate constants and activation energy. Tappi 58(8): 170-171. Kleinert TN (1976) Der alkohol-wasser-aufschluss des holzes, technologie und wirtschaftliche bedeutung. Papier 30(10A): V18-V23. Kleinert TN (1978) Kontinuierliches aufschluss- und ruckgewinnungsverfahren fur pflanzliche faserrohstoffe zur herstellung von zellstoff im organischen losungsmittel. DE patent 26.44.155.C2. Kleinert T & Tayenthal K (1931) Uber neuere versuche zur trennung von cellulose und inkrusten verschiedener holzer. Angewandte Chemie 44(39): 788-791. Kleinert T & Tayenthal K (1932) Satt for uppslutning av fibrosa vaxtmaterial med alkohol- vatten-losningar. Sweden patent 75178. Kobayashi M, Sasaki N, Gobara K & Nagasawa T (1978) Economic evaluation of alkali- methanol cooking. Japan Tappi 32(9): 525-532. Koivusaari P (1990) Adsorption kaytto peroksimuurahaishappokeiton jateliemen regeneroinnissa. MSc. thesis, Univ Oulu, Dept Process Eng. Koljonen J, Pohjanvesi S, Sundquist J & Poppius-Levlin K (1992) Pilot scale development of the Milox pulping and bleaching process. The European Pulp and Paper Week. Bologna, Italy, 1: 108-122. Koll P, Bronstrup B & Metzger J (1979) Thermischer abbau von birkenholz mit superkritischen gasen (organischen losungsmittel) in einer hochdruck-, hochtemperatur- stromungsapparatur: die verflussigung von holz und weitere hinweise auf eine alternative zellstoffgewinnungstechnologie. Holzforschung 33(4): 112-116. Koll P & Lenhardt H (1987) Organosolv-aufschluss von birkenholz in einem durchstromten reaktor. Holzforschung 41(2): 89-96. Koll P & Metzger J (1978) Thermischer abbau von cellulose und chitin in uberkritischem aceton. Angew Chemie 90(10): 802-803. Komissarenkov AA, Aliev RG, Burov AV, Sharkov VV & Badilovskaya TF (1990) LTITsBP, assignee. Method of producing pulp. USSR patent 1.608.275. (ABIPST 62:13064(U)). Komissarenkov AA, Aliev RG, Burov AV & Sharkov VV (1991a) LTITsBP, assignee. Method for producing pulp. USSR patent 1.689.481. (ABIPST 63:9047(U)). Komissarenkov AA, Aliev RG, Burov AV & Filatova GA (1991b) LTITsBP, assignee. Method for producing pulp. USSR patent 1.689.480. (ABIPST 63:9048(U)). Komissarenkov AA, Aliev RG, Burov AV, Sharkov VV & Luferova LB (1992a) Effect of inorganic sorbents on the delignification of wood by the organosolv process. Les Zh (2): 104-108. (ABIPST 64:4770). Komissarenkov AA, Aliev RG, Burov AV, Sharkov VV & Luferova LB (1992b) LTITsBP, assignee. Method of producing a fibrous cellulosic intermediate product. Russia patent 1.721.153. (ABIPST 63:14255(U)). Koncel J (1991) Alcell pulping process moves to first commercial installation. Am Papermaker 54(1): 22-26. Kopfmann K (1991a) Conversion of an existing kraft pulp mill to an ASAM mill. 1991 Tappi Pulping Conference Proc., Orlando, FL, 2: 925-931. Kopfmann K (1991b) Explosions- und arbeitsschutz beim einsatz von methanol fur das ASAM-verfahren. Papier 45/7): 396-402. Kopra-Schafer M (1991) Utilization possibilities of Organocell lignins. In: Faix O & Meier D (eds) Mitteilungen der bundesforschungsanstalt fur forst- und holzwirtschaft, 168: 22 - 28. Kordsachia O & Patt R (1987) Untersuchungen zum aufschluss von buche nach dem ASAM verfahren und zur bleiche der zellstoffe in untershiedlichen sequenzen. Papier 41(7): 340- 351. Kordsachia O & Patt R (1988a) Full bleaching of ASAM pulps without chlorine compounds. Holzforschung 42(3): 203-209. Kordsachia O & Patt R (1988b) Holzaufschluss nach dem ASAM-verfahren. Holz- Zentralblatt 114(81): 1220-1222. Kordsachia O, Patt R & Rachor G (1988) Untersuchungen zum methanol-sulfitaufschluss. Papier 42(6): 261-269. Kordsachia O, Patt R & Mix N (1989) Aufschluss von pappelholz nach dem ASAM- verfahren. Papier 43(7): 293-301. Kordsachia O, Reipschlager B & Patt R.(1990a) ASAM pulping of birch wood and chlorine free pulp bleaching. Paperi ja Puu 72(1): 44-50. Kordsachia O, Wandinger B & Patt R (1990b) Study on asam pulping and chlorine-free bleaching of spanish eucalyptus. Invest Tec Papel (104): 257-277. (ABIPST 61:12159). Kordsachia O, Wandinger B & Patt R (1992) Some investigations on asam pulping and chlorine free bleaching of eucalyptus from Spain. Holz Roh Werkstoff 50(3): 85-91. Kordsachia O, Oltmann E, Odermatt J & Patt R (1994) TCF bleaching of ASAM softwood pulp. TAPPI Pulping Conference Proc., San Diego, CA, p 277-287. Kordsachia O, Oltmann E & Patt R (1995) Abgrenzung der prozessschritte zur herstellung chlorfrei gebleichter ASAM-zellstoffe, teil 2, untersuchungen zur grenzpunkt- bestimmung innerhalb der bleichsequenz OZ(E)P. Papier 49(6): 308-315. Kosaya GS, Mironova TY, Prokopeva MA & Kosheleva VD (1981) All-Union Scientific- Industrial Enterprises of the Cellulose-Paper Industry, assignee. Pulp. USSR patent 829.747. (CA 95:82684z). Kosikova B & Polcin J (1973) Isolation of lignin from spruce by acidolysis in dioxane. Wood Sci Tech 7: 308-316. Kosikova B, Demianova V & Kacurakova M (1993a) Sulfur-free lignins as composites of polypropylene films. J Appl Polym Sci 47: 1065-1073. Kosikova B, Kacurakova M & Demianova V (1993b) Photooxidation of the composite lignin/polypropylene films. Chem Papers 47(2): 132-136. Kosikova B, Demianova V & Mikulasova M (1994) Effect of lignin on degradation of polyolefins with Phanerochaete chrysosporium. Drev Vysk 39(3): 15-23. Kostyukevich NG (1991) Acetylation of wood components during oxygen-acetic acid pulping. St. Petersburg, Russia, Institut khimii drevesiny, Issledovaniya v oblasti khimii drevesiny 43. (ABIPST 65:6582(AC)). Kostyukevich NG & Deineko IP (1992) Delignification by oxygen in acetic acid as new method of pulping. In: Kennedy JF, Phillips GO & Williams PA (eds) Lignocellulosics: Science, Technology, Development and Use. Ellis Horwood Limited, Chichester, p. 331-337. Kostyukevich NG, Evtyugin DV & Nikandrov AB (1993) Molecular-weight characteristics of oxygen-organosolv lignins. Izv Vyssh Uchebn Zaved Lesn Zh (2/3): 131-137. (ABIPST 65:6396(AC)). Koval J & Slavik I (1963) Acetic acid in wood delignification. Papir Celulosa 18: 215- 216. (CA 60:12221). Kratzl K, Buchtela K, Gratzl J, Zauner J & Ettingshausen O (1962) Lignin and plastics, the reaction of lignin with phenol and isocyanates. Tappi 45(2): 113-119. Kratzl K, Zauner J & Claus P (1964) Reaktionen mit markierten ligninen und modellen, I. mitteilung: kondensation mit phenol. Holzforschung 18(1/2): 47-52. Krotov VS (1993) Drip percolation pulping with aqueous and aqueous organic ammonia and sulphur dioxide solutions: environment-orientated technology. Khim Drev (1-3): 39- 44. Krotov VS & Frank AY (1993) Ammonium-sulfite-alcohol pulping: basis of prospective technology for the processing of nonwoody raw material. Tsellyul Bum Karton (5): 10- 11. (ABIPST 65:797). Krull M, Rachor G, Patt R & Kordsachia O (1989) Einfluss von methanol auf den sauren sulfit- und bisulfitaufschluss von buchenholz. Papier 43(2): 41-49. Krull M, Kordsachia O & Patt R (1991) Der methanol-sulfitaufschluss von kiefer. Papier 45(11): 673-680. Ku YC, Wang IC & Kuo ML (1993) Alcohol pulping of plantation taiwania and eucalyptus, (1), pulping and pulp properties. Bull Taiwan For Res Inst 8(3): 169-176. (ABIPST 66:2178(AS)). Kubes GJ, Fleming BI, MacLeod JM & Bolker HI (1978) Sulphide-free alkaline delignification, IV, extension of soda-ethylenediamine pulping to various species. 1978 Tappi Pulping Conference, New Orleans, LA, p 27-31. Kulkarni HG, Sirmokadam NN, Shenvi AK, Hegde GM, Padwalkar SN & Guha SRD (1994) Organosolv pulping of Agave sisalana. IPPTA 6(1): 15-18. Kuo Y, Munson CL, Rixey WG, Garcia AA, Frierman M & King CJ (1987) Use of adsorbents for recovery of acetic acid from aqueous solutions, part I - factors governing capacity. Sep Purif Methods 16(1): 31-64. Kurvinen T (1988) Puun delignifiointi etikka- ja peroksoetikkahapolla. MSc. thesis, Univ Helsinki, Dept Polymer Chemistry. Kuster W & Daur R (1930) Ueber die einwirkung von aromatischen diazokorpern auf lignin und cellulose. Cellulosechemie 11(1): 4-6. Kuznetsov BN, Levdanskii VA, Polezhaeva NI & Shilkina TA (1995) The wasteless processing of wood bark by non-isobaric steam cracking and fractional extraction methods. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 1: 669-675. Laamanen L, Poppius K, Sundquist J, Wartiovaara I & Kauliomaki S (1986) Bleached pulp by peroxyacid/alkaline peroxide delignification, part II, papermaking properties of peroxy pulps. Paperi ja Puu 68(9): 622-628. Laamanen LA, Sundquist JJ, Wartiovaara IYP, Kauliomaki SVM & Poppius KJ (1987) Oy Keskuslaboratorio - Centrallaboratorium Ab, assignee. Menetelma valkaistun selluloosa- materiaalin valmistamiseksi ligniinipitoisesta raaka-aineesta. FI patent 74750. Laamanen LA, Sundquist JJ, Wartiovaara IYP Kauliomaki SV & Poppius KJ (1988) Oy Keskuslaboratorio - Centrallaboratorium Ab, assignee. Process for bleaching organic peroxyacid cooked material with an alkaline solution of hydrogen peroxide. US patent 4.793.898. Laamanen LA, Sundquist JJ, Wartiovaara IYP, Poppius KJ & Kauliomaki SV (1991) Oy Keskuslaboratorio - Cetrallaboratorium Ab, assignee. Process for preparing pulp out of lignocellulosic raw material. CA patent 1.284.557. Lachenal D & Delagoutte T (1996) Effect of peracids on lignin - application to bleaching. Fourth European Workshop on Lignocellulosics and Pulp, 203-210. Lahtinen I (1991) Laimean muurahaishappoliuoksen vakevointi ja puhdistus. MSc. thesis, Univ Oulu, Dept Process Eng. Lahtinen I, Jurva E, Matinheikki S & Sohlo J (1994) Fischer labodest model 602- laitteistolla maaritettyja hoyry-nestetasapainotietoja. Univ Oulu, Dept Process Eng, Report 142. Lai YZ & Mun SP (1993) The role of organic solvents in acidic delignification of aspen. Seventh International Symposium on Wood and Pulping Chemistry Proc., Beijing, PR China, 1: 266-271. Lange W, Schweers W & Beinhoff O (1981) Uber die eigenschaften und abbarbaukeit von mit alkohol-wasser-gemischen isolierten ligninen, 5. mitteilung: orientierende unter- suchungen uber ausbeuten und eigenschaften von unter verschiedenen aufschluss- bedingungen mit und ohne katalysatorzusatz isolierten ligninen. Holzforschung 35(3): 119-124. Lange W, Faix O & Beinhoff O (1983) Uber die eigenschaften und abbarbaukeit von mit alkohol-wasser-gemischen isolierten ligninen, VIII, the inhomogeneity of the lignins from birch and spruce wood. Holzforschung 37(2): 63-67. Larocque GL & Maass O (1941) The mechanism of the alkaline delignification of wood. Can J Res 19B(1): 1-16. Lau H (1941) Bamboo pulp by the use of a hydrotropic solvent. Paper Ind Paper World 23: 247-255. Laxen T (1987) The impact of organosolv processes on washing. Pulp Washing Symposium Prep., Mariehamn, Finland, 2: 503-512. Laxen T, Aittamaa J & Lonnberg B (1988) Chemical pulping of softwood chips by alcohols, part 3, solvent recovery and energy consumption. Paperi ja Puu 70(10): 891- 894. Lee HL (1986) Selection of separation processes. In: Calo JM & Henley EJ (eds) AIChE modular instructions series B: stagewise and mass transfer operations volume 6: separation processes. AIChE, New York, p 17-23. Legeler E (1923) Der holzaufschluss mittels phenol. Cellulosechemie 4(6): 61-62. Lenz J, Peter W & Krassig, H. (1984) Chemiefaser Lenzig Aktiengesellschaft, assignee. Verfahren zur herstellung von zellstoffen mit niedrigem ligningehalt. AT patent 373.303. Leonova MO, Levina LF, Suvorova SI, Shapiro IL & Pen RZ (1994) Low-temperature oxidative delignification of wood, (2), pulping of trembling aspen chips with forced impregnation and low liquor ratio. Izv VUZ Lesn Zh (3): 76-80. (ABIPST 66:3616(AS)). Leopold B (1961) Chemical composition and physical properties of wood fibers, I, preparation of holocellulose fibers from loblolly pinewood. Tappi 44(3): 230-232. Leopold H (1991) Derzeitiger entwicklungsstand des Organocell-verfahrens. Deutsche Papierwirtschaft (3): T48-T49. Leopold H (1992) Fiber properties of Organocell pulps. 1992 Solvent Pulping Symposium Notes, Boston, MA, p 41-43. Leopold H (1993a) Organocell - ein neues konzept zur faserherstellung. XXV EUCEPA Conference Proc., Vienna, Austria, 1: 87-94. Leopold H (1993b) Technologische eigenschaften des Organocell-zellstoffs neueste erkentnisse nach der inbetriebnahme der anlage in Kelheim. Papier 47(10A): V1-V5. Leszczynski C (1992) Alcell - an ecologically friendly pulping technology. Przeglad Papierniczy 48(5): 166-167. (ABIPST 64:3366). Li Q & Chen R (1996) Alkaline-ethanol pulping of white birch [betula papyrifera] with recovery and reuse of ethanol. 82e Congres annuel Section technique Association canadienne des pates et papiers: Faconner aujour dhui lenvironnement de demain, Montreal, Canada, B: 159-163. (ABIPST 67:844(AC)). Li L & Kiran E (1988) Interaction of supercritical fluids with lignocellulosic materials. Ind Eng Chem Res 27(7): 1301-1312. Li L & Kiran E (1989a) Supercritical fluid extraction of lignin from wood. In: Glasser W & Sarkanen K (eds) Lignin properties & materials. ACS, Washington, p 42-57. (ABIPST 60:8976(B)). Li L & Kiran E (1989b) Delignification of red spruce by supercritical methylamine and methylamine-nitrous oxide mixtures. Tappi J 72(4): 183-190. Liang W & Xiao X (1992) Studies on the properties of wheat straw alkali ethanol pulp. In: Kennedy JF, Phillips GO & Williams, PA (eds) Lignocellulosics - Science, Technology, Development and Use. Ellis Horwood Limited, Chichester, p. 447-456. Liang WZ, Zhou FC, Xiao XR & Wang ZH (1988) Organosolv pulping characteristics of wheat straw. 1988 International Non-Wood Fiber Pulping and Papermaking Conference Proc., Beijing, PR China, p 271-280. Liang WZ, Qia HX & Wang ZH (1989) Studies on ethanol pulping of bagasse. In: Kennedy JF, Phillips GO & Williams PA (eds) Wood processing and utilization. Ellis Horwood Limited, Chichester, p 93-98. Liang W, Xiao X, Chen M & Han Y (1994) The bleaching of ethanol pulp. Second International Nonwood Fibre Pulping and Papermaking Conference Proc., Shanghai, PR China, p 700-707. Liebergott N (1994) Peracid delignification and bleaching. 1994 Pulping Conference Proc., San Diego, CA, p 357-370. Likon M & Perdih A (1995) Haloacetic pulping. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 2: 243-248. Lindholm J, Yilmaz Y, Johansson A, Gullichsen J & Sipila K (1995) A new process for combined energy and pulp production from agricultural plants. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 2: 455-460. Lindner A & Wegener G (1987) Isolierung un charakterisierung von organosolv-ligninen. Holz Roh Werkstoff 45(9): 390. Lindner A & Wegener G (1988a) Isolierung und chemische eigenschaften von ligninen aus aufschlussen nach dem Organocell-verfahren und ihre bedeutung fur eine potentielle nichtenergetische verwertung. Papier 42(10A): V1-V8. Lindner A & Wegener G (1988b) Characterization of lignins from organosolv pulping according to the Organocell process, part 1, elemental analysis, nonlignin portions and functional groups. J Wood Chem Tech 8(3): 323-340. Lindner A & Wegener G (1989) Characterization of lignins from organosolv pulping according to the Organocell process, part 2, residual lignins. J Wood Chem Tech 9(4): 443-465. Lindner A & Wegener G (1990a) Characterization of lignins from organosolv pulping according to the Organocell process, part 3, molecular weight determination and investigation of fractions isolated by GPC. J Wood Chem Tech 10(3): 331-350. Lindner A & Wegener G (1990b) Characterization of lignins from organosolv pulping according to the Organocell process, part 4, permanganate oxidation and thioacidolysis. J Wood Chem Tech 10(3): 351-363. Linnhoff B, Townsend DW, Boland D, Hewitt GF, Thomas BEA, Guy AR & Marsland RH (1982) A user guide on process integration for the efficient use of energy. The Institution of Chemical Engineers, Rugby. Lipinsky ES (1983) The cellulose challenge: where do we go from here? Tappi J 66(10): 47-49. Liukko S & Poppius-Levlin K (1997) Characteristics of different peroxyacid effluents. 9th International Symposium on Wood and Pulping Chemistry Poster Presentations, Montreal, Canada, p 65:1-4. Lobbecke JG, Lindner A, Wegener G & Wabner DW (1990) Anodic modification of lignins during electrolysis in the Organocell process. Holzforschung 44(6): 401-406. Lobbecke JG, Lindner A, Wegener G & Wabner DW (1991) Effects of electrolysis on the chemistry of Organocell lignins. In: Faix O & Meier D (eds) Mitteilungen der bundesforschungsanstalt fur forst- und holzwirtschaft (168): 50-52. Lora JH & Aziz S (1985) Organosolv pulping: a versatile approach to wood refining. Tappi J 68(8): 94-97. Lora JH & Pye EK (1992) The Alcell process. 1992 Solvent Pulping Symposium Notes. Boston, MA, p 27-34. Lora JH & Pye EK (1995) Solvent pulping opens door to viable chlorine-free, effluent-free options. Pulp & Paper 69(2): 109-110. Lora JH & Wayman M (1978) Delignification of hardwoods by autohydrolysis and extraction. Tappi 61(6): 47-50. Lora JH, Cronlund M, Powers J, Orlowski G & Wu L (1985) A comparative study of organosolv and kraft pulps. International Symposium on Wood and Pulping Chemistry, Vancouver, Canada, p 181-182. Lora JH, Katzen R, Cronlund M & Wu CF (1988) Repap Technologies Inc., assignee. Recovery of lignin. US patent 4.764.596. Lora JH, Wu CF, Pye EK & Balatinecz JJ (1989) Characteristics and potential applications of lignin produced by an organosolv process. In: Lignin: properties and materials. American Chemical Society Symposium Series, 397: 312-323. Lora JH, Katzen R, Cronlund M & Wu CF (1990) Repap Technologies Inc., assignee. Process for lignin recovery. CA patent 1.267.648. Lora JH, Creamer AW, Wu LCF & Goyal GC (1991a) Chemicals generated during alcohol pulping: characteristics and applications. 6th International Symposium on Wood and Pulping Chemistry Proc., Melbourne, Australia, 2: 431-438. Lora JH, Wu LCF & Klein WR (1991b) Mill trials on the use of ALCELLTM lignin as an adhesive for structural panels (abstract). 25th International Particleboard/Composite Materials Symposium, Pullman, WA, p 352. Lora JH, Winner SR & Pye EK (1992) Industrial scale alcohol pulping. In: Lisius JD (ed) 1989 and 1990 Forest Products Symposium Proc., Tappi Press, Atlanta, p 35-39. Lora JH, Goyal GC & Raskin M (1993a) Characterization of residual lignins after Alcell pulping. Seventh International Symposium on Wood and Pulping Chemistry Proc., Beijing, PR China, 1: 327 - 336. Lora JH, Powers JL, Pye EK & Branch BA (1993b). Recent developments in Alcell technology - towards a closed cycle mill. 1993 Pulping Conference Proc., Atlanta, GA, 3: 921-924. Lora JH, Creamer AW, Wu LCF & Goyal GC (1993c) Industrial scale production of organosolv lignins: characteristics and applications. In: Kennedy JF (ed) Cellulosics: chemical, biochemical and material aspects. Ellis Horwood Limited, Chichester, p 251- 256. Lonnberg B, Laxen T & Sjoholm R (1987a) Chemical pulping of softwood chips by alcohols, 1., cooking. Paperi ja Puu 69(9): 757-762. Lonnberg B, Laxen T & Backlund A (1987b) Chemical pulping of softwood chips by alcohols, 2., bleaching and beating. Paperi ja Puu 69(10): 826-830. Lorenz M & Dichtl G (1993) QVF Glastechnik GmbH, assignee. Verfahren zur ruckgewinnung von hochsiedenden komponenten aus abwasser. DE patent 42.22.783.C1. Lyublin VS, Baranova LE & Germer EI (1988) Effectiveness of ethanol pulping. Bum Prom (3): 16-17. (ABIPC 59:1341). MacLeod JM, Kubes GJ, Fleming BI & Bolker HI (1981) Soda-anthraquinone- ethylenediamine pulping. Tappi 64(6): 77-80. Mahapatra A, Gaikar VG & Sharma MM (1988) New strategies in extractive distillation: use of aqueous solutions of hydrotropes and organic bases as solvent for organic acids. Sep Sci Tech 23(4&5): 429-436. Maldas D & Kokta BV (1994) Use of lignin as a bonding agent for wood fiber-filled polystyrene composites. Drevarsky Vyskum 39(3): 25-41. Marr R, Siebenhofer M, Resch M & Bunzenberger G (1987) Extraktionstechnik Gesellschaft fur Anlagebau m.b.H., assignee. Verfahren zur gewinnung von furfural und monocarbonsauren aus wassrigen losungen. EP patent 0.111.699.B1. Martinez R & Sanjuan R (1993) Study of an organosolv process (ethanol-water-soda). Investigacion y Tecnica del Papel (117): 520-531. (ABIPST 65:799). Martinez R & Sanjuan R (1990) Advantages of the organosolv process. Amatl 4(4): 9-18. (ABIPST 63:9629). Marton R & Granzow SG (1982a) Use of ethanol in alkaline pulping. PCT patent WO 82/01568. Marton R & Granzow S (1982b) Ethanol-alkali pulping. Tappi 65(6): 103-106. Matinheikki S & Sohlo J (1994) HCOOH-CH3COOH-H2O-seoksen pervaporaatioerotus PVA-kalvon avulla. Univ Oulu, Dept Process Eng, Report 143. Matter-Muller C & Stumm W (1984) Water properties. In: Grayson M (ed) Kirk-Othmer encyclopedia of chemical technology, John Wiley & Sons, New York, 1: 276-294. Mattila T, Saari K & Vuori A (1992) Kemira Oy, assignee. Menetelma alemman alifaattisen hapon ottamiseksi talteen. FI patent 85510. Mbachu RAD & Manley R (1981) The effect of acetic and formic acid pretreatment on pulp bleaching with ozone. Tappi 64(1): 67-70. McCready M (1991) Alcell pulping: it works! PaperAge 107(3): 14-15,19,33. McDonough TJ (1993) The chemistry of organosolv delignification. Tappi Journal 76(8): 186-193. McGee JK & April GC (1982) Chemicals from renewable resources: hemicellulose behavior during organosolv delignification of southern yellow pine. Chem Eng Comm 19: 49- 56. McKee RH (1946) Use of hydrotropic solutions in industry. Ind Eng Chem 38(4): 382- 384. McKee RH (1954) Comparison of wood pulping processes; outline of principle and procedure of the hydrotropic pulping process. Pulp Paper Mag Canada 55(2): 64-66. McKee RH (1960) A new american paper pulp process, part I - pulping by the full hydrotropic process. Paper Ind 42(4): 255-257,266. McKee RH (1961) McKee Development Corp., assignee. Process for producing chemicals from wood. CA patent 631.051. McMillen JM, Gortner RA, Schmitz H & Bailey AJ (1938) Butanol cooking of hardwoods and softwoods. Ind Eng Chem 30(12): 1407-1409. Meier D, Faix O & Lange W (1981) Uber eigenschaften und abbaubarkeit von mit alkohol- wasser-gemischen isolierten ligninen, 6. mitteilung: chemische und spektroskopische charakterisierung. Holzforschung 35(5): 247-252. Meier D, Ante R & Faix O (1992) Catalytic hydropyrolysis of lignin: influence of reaction conditions on the formation and composition of liquid products. Biores Tech 40: 171-177. Meier D & Faix O (1991) Hydroliquefaction of lignins. In: Faix O & Meier D (eds) Mitteilungen der Bundesforschungsanstalt fur Forst- und Holzwirtschaft, 168: 198-203. Meier D & Schweers W (1979) Uber eigenschaften und abbaubarkeit von mit alkohol- wasser-gemischen isolierten ligninen, 2. mitteilung: alkalische druckoxidation mit luftsauerstoff zur gewinnung von vanillin und syringaaldehyd. Holzforschung 33(6): 177-180. Meier D & Schweers W (1981) Uber eigenschaften und abbaubarkeit von mit alkohol- wasser-gemischen isolierten ligninen, 4. mitteilung: katalytische hydrogenolyse zur erzeugung monomerer phenole. Holzforschung 35(2): 81-85. Melo SR & Muller GP (1989) Effects of adding MeOH to soda-AQ pulping of eucalyptus. Celulosa Papel 5(1): 10-4. (ABIPST 60:10113). Merkle G, Auerbach S, Burchard W, Lindner A & Wegener G (1992) Light scattering of acetylated lignin. J Appl Polym Sci 45(3): 407-415. Meyer FJ, Humiston CG & Moyle CL (1961) The Dow Chemical Company, assignee. Wood pulping process. CA patent 618.297. Millner D (1983) Ny fraktioneringsmetod: cellulosafibrer, lignin och kemikalier ur biomassa. Cellulosa (8): 20-21. Milstein O, Huttermann A, Ludemann HD, Majcherczyk A & Nicklas B (1990) Enzymatic modification of lignin in organic solvents. In: Kirk TK & Chang HM (eds) Biotechnology in pulp and paper manufacture, applications and fundamental investigations. Butterworth-Heinemann, Boston, p 375-387. Mishra RK & Tiwari KK (1988) Liquid-liquid equilibrium in carboxylic acids (formic acid and acetic acid)-water-trioctylamine (with halogenated hydrocarbons) systems. Indian J Tech 26: 447-452. Miyazaki K, Iwanaga Y & Tsuchiya H (1989) Effect of amine on cresol pulping. In: Kennedy JF, Phillips GO & Williams PA (eds) Wood processing and utilization. Ellis Horwood Limited, Chichester, p 75-80. Mogollon G (1991) Effect of organic solvent on the alkaline delignification of sugar cane bagasse. 6th International Symposium on Wood and Pulping Chemistry Proc., Melbourne, Australia, 2: 251. Montanari S, Curvelo AA & Groote RAMC (1991) Organosolv delignification of pinus caribaea hondurensis: relationship between kappa number and klason lignin. Proc. 24 Congresso Anual de Celulose e Papel (ABTCP), p 137-151. (ABIPST 66:3619(AC)). Montanari S, Curvelo AAS & Groote RAMC (1993) Dioxan lignins from pinus caribaea var. hondurensis, part 2, effect of the extraction time. Paperi ja Puu 75(7): 512-514. Mostafa NYS (1994) Base-catalyzed dioxane and dioxane-borax pulping and the fine structure, chemical reactivity and viscose filterability of cotton cellulose. Cell Chem Tech 28: 171-175. Mullinder J (1989) Repap happy with Alcell results, in engineering phase for expansion. Pulp Paper J 42(7): 31. Mun SP (1989) Alcohol-bisulfite pulping: mechanism of delignification. In: Kennedy JF, Phillips GO & Williams PA (eds) Wood processing and utilization. Ellis Horwood Limited, Chichester, p 99-104. Mun SP & Wi H (1991a) Alcohol-bisulfite cooking of hyun-aspen [populus alba x glandulosa L.] wood, (1), determination of delignification conditions. J TAPPIK 23(1): 55-61. (ABIPST 62:8473). Mun SP & Wi H (1991b) Alcohol-bisulfite cooking of hyun-aspen [populus alba x glandulosa L.] wood, (2), conditions for the production of low-molecular-weight phenols. J TAPPIK 23(2): 15-20. (ABIPST 62:8472). Mun SP, Sakai K & Imamura H (1988) Delignification with organic solvents and sulphites, (5), conditions for producing low-molecular-weight phenols and preparation of chemical pulp, (6), pulp quality and improvement of pulp strength. Mokuzai Gakkaishi 34(3): 234-245. (ABIPC 59:6356). Munson CL, Garcia AA, Kuo Y, Frierman M & King CJ (1987) Use of adsorbents for recovery of acetic acid from aqueous solutions, part II - factors governing selectivity. Separ Purif Methods 16(1): 65-89. Muurinen E (1986) Peroksihappomenetelmalla toimivan sellunvalmistusprosessin keittoliuoksen talteenotto ja regenerointi. MSc. thesis, Univ Oulu, Dept Process Eng. Muurinen E (1994) Peroksimuurahaishappoprosessin regenerointi ja prosessisynteesi. Lic. Sc. thesis, Univ Oulu, Dept Process Eng. Muurinen E & Sohlo J (1988a) Liuottimen regenerointi organosolv prosesseista, 3. happokeitot. Univ Oulu, Dept Process Eng, Report 113. Muurinen E & Sohlo J (1988b) Liuottimen regenerointi organosolv prosesseista, 4. alkoholi- ja happokeittojen vertailu. Univ Oulu, Dept Process Eng, Report 114. Muurinen E & Sohlo J (1989) Recovery of chemicals from organosolv woodpulping processes. In: Kennedy JF, Phillips GO & Williams PA (eds) Wood processing and utilization. Ellis Horwood Limited, Chichester, p 81-86. Muurinen E & Sohlo J (1991a) Recovery of chemicals from peroxyformic acid pulping. 6th International Symposium on Wood and Pulping Chemistry Proc., Melbourne, Australia, 2: 169-173. Muurinen E & Sohlo J (1991b) Recovery of chemicals from peroxyformic acid pulping. 4th World Congress of Chemical Engineering Prep., Karlsruhe, Germany, 1: 3.1-10. Muurinen E & Sohlo J (1992) Recovery of chemicals from peroxyformic acid pulping. In: Kennedy JF, Phillips GO & Williams, PA (eds) Lignocellulosics - Science, Technology, Development and Use. Ellis Horwood Limited, Chichester, p 385-392. Muurinen E & Sohlo J (1993a) Heat integration of a new formic acid pulping and recovery process. In: Pilavachi PA (ed) Energy efficiency in process technology. Elsevier Science Publishers Ltd, Barking, p 996-1004. Muurinen E & Sohlo J (1993b) Peroksihappokeiton erotusmenetelmat 2. Univ Oulu, Dept Process Eng, Report 137. Muurinen E & Sohlo J (1994) Simulation of recovery methods for pulping with peroxyformic and peroxyacetic acids. Comp Chem Eng 18(Suppl.): 609-613. Muurinen E & Sohlo J (1997a) Separation methods for the peroxyacid pulping process. 9th International Symposium on Wood and Pulping Chemistry Poster Presentations, Montreal, Canada, p 75:1-4. Muurinen E & Sohlo J (1997b) Solvent recovery in peroxyacid pulping. In: Darton R (ed) Distillation and absorption 97 vol 2. IChemE, Rugby, p 543-552. Muurinen E, Kivela E & Sohlo J (1988a) Liuottimen regenerointi organosolv prosesseista, 1, kirjallisuuskatsaus ja tutkimuskohteiden valinta. Univ Oulu, Dept Process Eng, Report 111. Muurinen E, Tuomisto S & Sohlo J (1988b) Liuottimen regenerointi organosolv prosesseista, 2, alkoholikeitot. Univ Oulu, Dept Process Eng, Report 112. Muurinen E, Sohlo J, Vanhanen T & Kivela E (1989) Recovery of chemicals from organosolv woodpulping processes. 1989 International Symposium on Wood and Pulping Chemistry Poster Sessions, Raleigh, NC, p 223-225. Muurinen E, Jurva E & Sohlo J (1992) Peroksihappoprosessin regenerointi ja prosessisynteesi. Univ Oulu, Dept Process Eng, Report 122. Muurinen E, Silvennoinen J & Sohlo J (1993a) Peroksihappokeiton erotusmenetelmat. Univ Oulu, Dept Process Eng, Report 131. Muurinen E, Jurva E, Lahtinen I & Sohlo J (1993b) Peroxyacid pulping and recovery. Seventh International Symposium on Wood and Pulping Chemistry Proc., Beijing, PR China, 3: 195-200. Myerly RC, Nicholson MD, Katzen R & Taylor JM (1981) The forest refinery. Chemtech 11(3): 186-192. Nadgir VM & Liu YA (1983) Studies in chemical process design and synthesis: part V, a simple heuristic method for systematic synthesis of initial sequences for multicomponent separations. AIChE J 29(6): 926-934. Nakamura H & Takauti E (1941) Zellstoffherstellung mittels aethylenglykol. Cell Ind 17(3): 19-26. Nakamura T, Yamaguchi T & Murayama H (1966) Tekkosha Co., assignee. Process for the recovery of lower carboxylic acids from the product of the liquid phase oxidation of paraffinic hydrocarbons. GB patent 1.049.901. Nakamura K, Morck R, Reimann A, Kringstad KP & Hatakeyama H (1991) Mechanical properties of solvolysis lignin-derived polyurethanes. Polym Adv Tech 2(1): 41-47. Nakano J, Takatsuka C & Daima H (1976) Studies on alkali-methanol cooking, (part 1), behavior of dissolution of lignin and carbohydrate. Japan Tappi 30(12): 33-38. Nakano J, Takatsuka C, Daima H, Kobayashi M & Sasaki N (1977) Studies on alkali- methanol cooking, (2), strenght of pulp sheet and recovery of methanol. Japan Tappi 31(11): 762-766. (ABIPC 48:9255) Nakano J, Daima H, Hosoya S & Ishizu A (1981) Studies on alkali-methanol cooking. International Symposium on Wood and Pulping Chemistry, Stockholm, Sweden, 2: 72- 77. Nath R & Motard RL (1981) Evolutionary synthesis of separation processes. AIChE J 27(4): 578-587. Nelson PJ (1977) Pulping of wood with glycol solutions of salicylic acid derivates. Appita 31(1): 29-32. Neto CP, Evtuguin D & Robert A (1994) Chemicals generated during oxygen-organosolv pulping of wood. J Wood Chem Tech 14(3): 383-402. Neumann N & Balser K (1993) Acetocell - ein innovatives verfahren zur absolut schwefel- und chlorfreien zellstoffproduktion. Papier 47(10A): V16-V24. Neumann U, Kordsachia O & Patt R (1997) Aufschluss von kiefernholz im natronlauge/AQ/methanol-system. Papier 51 (11): 573-579. Ni Y & Entremont M (1997) Peroxyacetic acid oxidation of lignin model compounds. 9th International Symposium on Wood and Pulping Chemistry Oral Presentations, Montreal, Canada, p D-6 1-5. Ni Y & Heiningen ARP (1994) Lirnin removal of alcell pulp by washing with ethanol- water solution. 1994 Tappi Pulping Conference Proc., p 5-12. Ni Y & Heiningen ARP (1995) The swelling of pulp fibers derived from the Alcell process. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 2: 255-260. Ni Y & Heiningen ARP (1996) Lignin removal from Alcell pulp by washing with ethanol and water. Tappi J 79(3): 239-243. Ni Y & Heiningen ARP (1997) The swelling of pulp fibers derived from the ethanol-based organosolv process. Tappi J 80(1): 211-213. Ni Y & Heiningen ARP (1998) An ECF sequence for the Alcell process including an ethanol-assisted ozone stage. Tappi J 81(4): 141-144. Ni Y & Hu Q (1995) Alcell lignin solubility in ethanol-water mixtures. J Appl Polym Sci 57: 1441-1446. Nikandrov AB (1993) Effect of wood dimensions on the dissolution of lignin during oxygen-acetic acid pulping. Izv Vyssh Uchebn Zaved Lesn Zh (2/3): 87-90. (ABIPST 65:6586(AS)). Nimz H (1969a) Uber ein neues abbauverfahren des lignins. Chem Ber 102: 799-810. Nimz H (1969b) Spaltung vom aetherbindungen durch benachbarte mercaptogruppen, modellversuche mit thioessigsaure. Chem Ber 102: 3803-3817. Nimz HH & Berg A (1989) Kunz Holding GmbH & Co. KG, assignee. Verfahren zur delignifizierung von zellstoffen. EP patent 0.325.890.A1. Nimz HH & Berg A (1991) Acetocell GmbH & Co. KG, assignee. Lignin removal method using ozone and acetic acid. US patent 5.074.960. Nimz HH & Casten R (1985) Organosolv pulping with acetic acid. International Symposium on Wood and Pulping Chemistry Technical Papers, Vancouver, Canada, p 265-266. Nimz HH & Casten R (1986a) Chemical processing of lignocellulosics. Holz Roh Werkstoff 44: 207-212. Nimz HH & Casten R (1986b) Holzaufschluss mit essigsaure. DE patent 34.45.132.A1. Nimz HH & Robert D (1985) 13C NMR spectra of acetic acid lignins. International Symposium on Wood and Pulping Chemistry Technical Papers, Vancouver, Canada, p 267. Nimz HH & Schoene M (1993) Non-waste pulping and bleaching with acetic acid. Seventh International Symposium on Wood and Pulping Chemistry Proc., Beijing, PR China, 1: 258-265. Nimz HH & Schwind H (1981) Oxidation of lignin and lignin model compounds with peracetic acid. International Symposium on Wood and Pulping Chemistry, Stockholm, Sweden, 2: 105-111. Nimz HHH & Schone M (1993) Gebruder Kummerer Projekt Agentur GmbH, assignee. Puun kuidutus etikkahapolla siten, etta lisataan muurahaishappoja. FI patent appl 933729. Nimz HH & Schone M (1994) Gebruder Kummerer Projekt Agentur GmbH, assignee. Holzaufschluss mit essigsaure unter zusatz von ameisensaure. DE patent 42.28.171.A1. Nimz HH, Granzow C & Berg A (1986a) Acetosolv pulping. Holz Roh Werkstoff 44: 362. Nimz HH, Roth M, Tschirner U & Casten R (1986b) Organosolv pulping with chloroethanol. Holzforschung 40(Suppl.): 125-132. Nimz HH, Casten R & Roth M (1986c) Holzaufschluss mit essigsaure und chlorethanol. DE patent 34.23.024.A1. Nimz HH, Berg A, Granzow C & Casten R (1988) Schadstofffreie gewinnung von gebleichtem zellstoff, lignin und hemicellulosen nach dem Acetosolv-verfahren. Mitteilungen der Bundesforschungsanstalt fur Forst- und Holzwirtschaft (160): 343-355. Nimz HH, Berg A, Granzow C & Casten R (1989a) Acetosolv pulping and bleaching. Non- waste Technology, VTT Symposium 102, Espoo, Finland, 1: 399-408. Nimz HH, Berg A, Granzow C, Casten R & Muladi S (1989b) Zellstoffgewinnung und bleiche nach dem Acetosolv verfahren. Papier 43(10A): V102-V108. Nimz HH, Berg A & Casten R (1989c) Kunz Holding GmbH & Co. KG, assignee. Verfahren zur gewinnung von gebleichtem zellstoff. EP patent 0.325.891.A1. Nimz H, Berg A, Janssen W, Balle S, Kunz RG & Klein W (1992a) Acetocell GmbH & Co. KG, assignee. Verfahren zur delignifizierung von cellulosehaltigen rohstoffen. DE patent 41.07.357.C1. Nimz H, Berg A, Janssen W, Balle S, Kunz RG & Klein W (1992b) Acetocell GmbH & Co. KG, assignee. Verfahren zur delignifizierung von cellulosehaltigen rohstoffen. EP patent 0.503.304.A1. Nishiyama M, Hosokawa J & Yoshihara K (1989) Basic study on removal of cresols from waste water. Shikoku Kogyo Gijutsu Shikensho Hokoku 21(1): 1-8. (ABIPST 61:4976(U)). Norman E, Olm L & Teder A (1992) Methanol-reinforced kraft pulping. 1992 Solvent Pulping Symposium, Boston, MA, p 51-60. Null HR (1980) Energy economy in separation processes. Chem Eng Prog 76(8): 42-49. Null HR (1987) Selection of a separation process. In: Rousseau RW (ed) Handbook of separation process technology. John Wiley & Sons, New York, p 982-995. Null HR, Bowe LE & Binning RC (1967) Monsanto Company, assignee. Production of acetic acid and recovery by plural stage distillation. US patent 3.335.179. Nunez C & Reinoso M (1990) Aqueous alcohol pulping of bagasse - conversion from kappa number to residual lignin. 26th ATIPCA Congr Tec Celulosa Papel, Buenos Aires, Argentina, p 401-412. (ABIPST 64:6163(M)). Obrocea P & Cimpoescu G (1998) Contribution to sprucewood delignification with peroxyformic acid. Cellul Chem Technol 32(5-6): 517-525. (Paperbase 00530945). Obrocea P & Gavrilescu D (1988) Delignification of beechwood acetic acid solutions. Celuloza Hirtie 37(4): 157-159. (ABIPST 60:8200) Obrocea P, Gavrilescu D & Stancana R (1994) Organosolv pulping of wood in alkaline medium, I., ASAM pulping of beechwood at low temperature. Cell Chem Tech 28: 339- 349. Obrocea P, Gavrilescu D & Stancana R (1996) Bleachability of some beech [Fagus] organosolv pulps. Celul Hirtie 45(1): 41-45. (ABIPST 67:12655(AS)). Obst JR & Sanyer N (1980) Effect of quinones and amines on the cleavage rate of beta -O-4 ethers in lignin during alkaline pulping. Tappi 63(7): 111-114. Oh SW, Paik KH & Kang CH (1988) Comparison of asaq (alkaline sulfite/aq) pulp extracted with organic solvents and ASAM (alkaline sulfite/aq/meoh). J Tappik 20(3): 42-48. (ABIPST 60:5298) Oki T, Okubo K & Ishikawa H (1974) The oxidative degradation of guaiacylglycerol- - guaiacyl ether by peracetic acid. Mokuzai Gakkaishi 20(11): 549-557. Oliveira ON, Constantino CJL, Balogh DT & Curvelo AAS (1994) Langmuir monolayers of lignins from pinus caribaea hondurensis sawdust. Cell Chem Tech 28: 541-549. Olm L & Teder A (1990) Nya kokprocesser for blekt massa: mindre klor med metanol. Sv Papperstidn 93(10): 22-24. Oltmann E, Gause E, Kordsachia O & Patt R (1992) Ozone bleaching technology: a comparison between high and medium consistency, part I. Papier 46(7): 341-350. Oltmann E, Kordsachia O & Patt R (1994) Abgrenzung der prozessschritte zur herstellung chlorfrei gebleichter ASAM-zellstoffe, teil 1, untersuchungen zur grenzpunkt- bestimmung zwischen dem aufschluss und der chlorfreien bleichsequenz OZEP. Papier 48(2): 55-68. Ooi T & Ni Y (1998) Development of an ozone-based TCF sequence for bleaching a hardwood ALCELL pulp. Tappi J 81(5): 255-259. Othmer DF & Conti JJ (1962) Process of azeotropic distillation of formic acid from acetic acid. US patent 3.024.170. Paik KH, Oh SH & Koo JO (1988) Alkaline sulfite pulping with the additions of anthraquinone and methanol. J Tappik 20(2): 24-32. (ABIPST 60:502) Paik KH, Nahm WS, An BJ & Moon CK (1989) Separation of wood components by acetic acid under atmospheric conditions and their utilization. J Tappik 21(3): 35-47. Paik KH, Kim DH & Moon CK (1994a) Alkaline sulfite pulping with the additions of athraquinone and methanol, (1), pine ASAM pulping and the composition of tall oil recovered from ASAM black liquor. J TAPPIK 26(1): 37-44. (ABIPST 65:18822). Paik KH, Kim DH & Nho MB (1994b) Alkaline sulfite pulping with the additions of athraquinone and methanol, (2), oak and birch ASAM pulping. J TAPPIK 26(4): 51-57. (ABIPST 65:18821). Palenius I (1987) Was sollte von neuen aufschlussverfahren gefordert werden? Papier 41(10A): V1-V6. Selvam PV, Ghose TK & Ghosh P (1983) Catalytic solvent delignification of agricultural residues: inorganic catalysts. Process Biochem 18: 13-15. Papatheophanus MG, Koullas DP & Koukios EG (1995) Alkaline pulping of prehydrolysed wheat straw in aqueous organic solvent systems at low temperatures. Cell Chem Tech 29(1): 29-40. (ABIPST 66:15422(AS)). Pappens R (1990) Organocell: high quality, low pollution. Pulp Paper Int 32(3): 74-77. Parajo JC & Santos V (1995) Preliminary evaluation of acetic acid-based processes for wood utilization. Holz Roh Werkst 53: 347-353. Parajo JC, Alonso JL, Vazquez D & Santos V (1993) Optimization of catalysed Acetosolv fractionation of pine wood. Holzforschung 47(3): 188-196. Park JK & Phillips JA (1988) Ammonia catalyzed organosolv delignification of poplar. Chem Eng Comm 65: 187-205. Paszner L (1998) Catalyzed alcohol organosolv pulping. In: Young RA & Akhtar M (eds) Environmentally Friendly Technologies for the Pulp and Paper Industry. John Wiley & Sons, New York, NY, p 69-100. Paszner L & Behera NC (1984) Beating behaviour and sheet strenght development of coniferous organosolv fibers. XXI Eucepa International Conference, Torremolinos, Spain, 1: 83-98. Paszner L & Behera NC (1985) Beating behaviour and sheet strength development of coniferous organosolv fibers. Holzforschung 39(1): 51-61. Paszner L & Behera NC (1989) Topochemistry of softwood delignification by alkali earth metal catalysed organosolv pulping. Holzforschung 43(3): 159-168. Paszner L & Chang PC (1979) Bau- und Forschunggesellschaft Thermoform AG, assignee. Verfahren zum entlignifizieren und verzuckern von pflanzlichen lignocellulose- materialien unter mitverwendung organischer losungsmittel. PCT patent WO 79/00119. Paszner L & Chang PC (1981) Organosolv delignification and saccharification process for lignocellulosic plant materials. CA patent 1.100.266. Paszner L & Chang PC (1982) Bau- und Forschunggesellschaft Thermoform AG, assignee. Pulping of lignocellulose with aqueous alcohol and alkaline earth metal salt catalyst. PCT patent WO 82/00483. Paszner L & Chang PC (1983a) Bau- und Forschungsgesellschaft Thermoform AG, assignee. Lignocelluloseaufschlussverfahren mit einem gemisch von wasser und niedrig molekularem aliphatischem alkohol und metallsalz. DE patent 29.20.731.C2. Paszner L & Chang PC (1983b) Catalysed organosolv pulping of softwoods. International Symposium on Wood and Pulping Chemistry, Tsukuba Science City, Japan, Suppl.: 21- 28. Paszner L & Chang PC (1983c) Productions of high-yield dissolving pulp by the catalysed organosolv process. 1983 International Dissolving and Speciality Pulps Conference, Boston, MA, p 81-88. Paszner L & Chang PC (1983d) Thermoform Bau- und Forschungsgesellschaft, assignee. Organosolv delignification and saccharification process for lignocellulosic plant materials. US patent 4.409.032. Paszner L & Chang PC (1984) High efficiency organosolv saccharification process. US patent 4.470.851. Paszner L & Cho PHJ (1987) Innovations for hardwood utilization: catalyzed organosolv pulp and saccharification. Proc. of the Workshop on Aspen [Populus], Pulp, Paper and Chemicals, 1: 63-78. (ABIPST 67:2122(AC)). Paszner L & Cho HJ (1989) Organosolv pulping: acidic catalysis options and their effect on fiber quality and delignification. Tappi J 72(2): 135-142. Paszner L, Quinde AA & Meshgini M (1985) ACOS - accelerated hydrolysis of wood by acid catalysed organosolv means. International Symposium on Wood and Pulping Chemistry, Vancouver, Canada, p 235-240. Patel DP & Varshney AK (1989) The effect of presoaking and prehydrolysis in the organosolv delignification of bagasse. Indian J Tech 27(6): 285-288. Patt R & Kordsachia O (1986) Herstellung von zellstoffen unter verwendung von alkalischen sulfitlosungen mit zusatz von anthrachinon und methanol. Papier 40(10A): V1-V8. Patt R & Kordsachia O (1988) Anforderungen an neue zellstoffherstellungsverfahren. Wochenbl Papierfabr 116(17): 709-713. Patt R, Kordsachia O & Reuter G (1987a) Wirtschaftliche und technische aspekte des ASAM verfahrens. Deutsche Papierwirtschaft (3): T96-T102. Patt R, Kordsachia O & Knoblauch J (1987b) The ASAM process - alkaline sulfite, anthraquinone, methanol pulping. Fourth International Symposium on Wood and Pulping Chemistry, Paris, France, 1: 355-360. Patt R, Kordsachia O & Mix N (1989a) Uber die eignung von pappelholz zur zellstoffherstellung nach dem ASAM-verfahren. Holz-Zentralblatt 115(4): 1,22,26. Patt R, Kordsachia O & Kopfmann K (1989b) Wirtschaftliche, technologische und oekologische aspekte der zellstoffherstellung nach dem ASAM-verfahren. Papier 43(10A): V108-V115. Patt R, Kordsachia O, Schubert HL & Hammann M (1990a) Actual status of the ASAM- process. 24th Eucepa Conference: Control, Maintenance, Environment, Stockholm, Sweden, p 442-446. Patt R, Schubert HL, Kordsachia O, Oltmann E & Ridder W (1990b) The ASAM process - competitor to kraft pulping? Sunds Defibrator Symposium, Pori, Finland, p 1-17. Patt R, Knoblauch J, Faix O, Kordsachia O & Puls J (1991a) Lignin and carbohydrate reactions in alkaline sulfite, anthraquinone, methanol (ASAM) pulping. Papier 45(7): 389-396. Patt R, Knoblauch A, Faix O, Kordsachia O, Puls J & Robert D (1991b) Lignin and carbohydrate reactions in alkaline sulfite, anthraquinone, methanol (ASAM) pulping. 6th International Symposium on Wood and Pulping Chemistry Proc., Melbourne, Australia, 1: 609-617. Patt R, Schubert HL, Kordsachia O, Oltmann E & Krull M (1991c) Chlorfreie bleiche von sulfit- und ASAM-zellstoffen im labor- und pilot-masstab. Papier 45(10A): V8-V16. Patt R, Jaschinski T & Mielisch HJ (1996) Stabilizierte und katalysierte peroxidbleiche. Wochenbl Papierfabr 124(17):750, 752-755. (ABIPST 67:12657(AS)). Patt R, Kordsachia O & Schubert HL (1998) The ASAM process. In: Young RA & Akhtar M (eds) Environmentally Friendly Technologies for the Pulp and Paper Industry. John Wiley & Sons, New York, NY, p 101-131. Pauly H (1918a) Aktiengesellschaft fur Zellstoff- und Papierfabrikation in Aschaffenburg, assignee. Verfahren zur gewinnung der das sogenannte lignin bildenten stoffe aus holzarten. German patent 309551. Pauly H (1918b) Aktiengesellschaft fur Zellstoff- und Papierfabrikation, assignee. Satt att utvinna lignin ur cellulosahaltigt material. Swedish patent 45010. Pauly H (1921) Aktiengesellschaft fur Zellstoff- und Papierfabrikation in Aschaffenburg, assignee. Verfahren zur gewinnung der das sogenannte lignin bildenten stoffe aus zellulosehaltigen materialien. AT patent 83396. Pauly H (1943) Gewinnung huminfreier essigsaure-lignole. Berichte der Deutschen Chemischen Gesellschaft 76(9): 864-871. Pauly H, Foulon A, Hansen O, Haberstroh O, Bailom H & Sextl J (1934) Scheidung von lignin-komponenten. Berichte der Deutschen Chemischen Gesellschaft 67: 1177-1199. Pazukhina G & Ostrik S (1998) LIgnin destruction by peracetic acid during the fir cooking in moderately mixed liquids. Proc. Fifth European Workshop on Lignocellulosics Chemistry for Ecologically Friendly Pulping and Bleaching Technologies, Aveiro, Portugal, p 429-432. (Paperbase 00522248). Pazukhina GA & Shan DS (1990) LTA, assignee. Method of producing pulp. USSR patent 1.542.983. (ABIPST 62:5054(U)). Pazukhina GA & Teploukhova MV (1998) Bleaching of organosolvent pulp by oxygen- containing reagents. Lesn Zh (6): 75-78. (Paperbase 00542458). Pazukhina GA, Teploukhova MV & Khakimova FK (1995) Chlorine-free bleaching of organosolv pulp. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 2: 265-268. Pecina H & Kuhne G (1992a) Die eignung von Organocell-zellstoffablaugen als phenolsubstituent in phenolischen spanplattenleimen. Holzforsch Holzverwert 44(3): 39-43. Pecina H & Kuhne G (1992b) Lignin-phenol-bindemittel auf der basis von ASAM- dicklauge. Holz-Zentralblatt 118(28): 464-466. Peguy A (1989) New materials from cellulose or lignocellose solutions in amine oxide systems. In: Inagaki H & Phillips GO (eds) Cellulosics utilization. Elsevier Science Publishers Ltd, Barking, p 19-24. Pekarovicova A, Jonekova A & Pekarovic J (1995) Use of xylanase and ligninase in bleaching of beech [fagus] organosolv pulp. Przegl Papier 51(9): 451-454, 457. (ABIPST 67: 4413(AS)). Pellinen J & Nimz HH (1986) Extraction of lignin from birch wood using chloroethanol and acetic acid. Paperi ja Puu 68(8): 548-552. Pen RZ, Suvorova SI & Leonova MO (1993) Low-temperature oxidative delignification of wood and properties of the pulps. Izv Vyssh Uchebn Zaved Lesn Zh (2/3): 57-60. (ABIPST 65:6588(AS)). Pepper JM, Baylis PET & Adler E (1959) The isolation and properties of lignins obtained by the acidolysis of spruce and aspen woods in dioxane-water medium. Can J Chem 37: 1241-1248. Pereira H, Oliveira MF & Miranda I (1986) Kinetics of ethanol-water pulping and pulp properties of eucalyptus globulus lab. Appita 39(6): 455-458. Perez DS & Curvelo AAS (1997) Kinetics of acetone-water organosolv delignification of eucalyptus urograndis. 9th International Symposium on Wood and Pulping Chemistry Poster Presentations, Montreal, Canada, p 88:1-4. Perry RH & Chilton CH (1973) Chemical engineers handbook. McGraw-Hill Book Company, New York. Perttunen J, Sohlo J & Maentausta O (1996) The fermentation of birch hemicellulose liquor to lactic acid. In: Srebotnik E & Messner K (eds) Biotechnology in pulp and paper industry. Facultas-Universitatsverlag, Vienna, p 295-298. Peter W & Hoglinger O (1986) Herstellung von kunstfaserzellstoff nach dem organosolv- aufschlussverfahren. Lenzinger Ber 61: 12-16. Peter W, Krassig H & Lenz J (1984) Chemiefaser Lenzig Aktiengesellschaft, assignee. Verfahren zur aufarbeitung von gebrauchter aufschlussflussigkeit sowie anlagen zur durchfuhrung dieses verfahren. AT patent 373.932. Petty G (1989) Canadians pioneer a small-scale pulp mill. Paper Tech 30(2): 10-13. Pew JC (1957) Properties of powdered wood and isolation of lignin by cellulytic enzymes. Tappi 40(7): 553-558. Pilhofer T & Dichtl G (1986) Liquid-liquid extraction as an alternative to distillation for the recovery of organic material. ISEC86 Prep., Frankfurt am Main, FR Germany, 3: 723-729. Pla F & Yan JF (1991) Viscosity and functionality of alkali and organosolv lignins. Holzforschung 45(2): 121-125. Pla F, Dolk M, Yan JF & McCarthy JL (1986) Lignin, 23., macromolecular characteristics of alkali lignin and organosolv lignin from black cottonwood. Macromolecules 19(5): 1471-1477. Ploetz T, Richtzenhain H, Deiters W & Giesen J (1963) Feldmuhle Aktiengesellschaft, assignee. Verfahren zur herstellung von niedrigsiedenden spaltprodukten aus lignin durch kontinuierliche druckhydrierung in gegenwart organischer flussigkeiten. DBP patent 1.142.853. Pohjanvesi S, Saari K, Poppius-Levlin K & Sundquist J (1995) Technical and economical feasibility study of the milox process. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 2: 231-236. Poljak A (1948) Holzaufschluss mit peressigsaure. Angewandte Chemie 60A(2): 45-46. Poljak A (1951) Holzaufschluss mit peressigsaure II. Holzforschung 5(2): 31-33. Poppius K, Laamanen L, Sundquist J, Wartiovaara I & Kauliomaki S (1986) Bleached pulp by peroxyacid/alkaline peroxide delignification. Paperi ja Puu 68(2): 87-92. Poppius K, Laamanen L, Sundquist J, Wartiovaara I, Toikkanen L & Keranen E (1987) Multi-stage peroxyformic acid pulping. Fourth International Symposium on Wood and Pulping Chemistry, Paris, France, 2: 211-214. Poppius K, Sundquist J & Wartiovaara I (1989a) Chemical pulping of birch and pine chips by the three stage peroxyformic acid method. In: Kennedy JF, Phillips GO & Williams PA (eds) Wood processing and utilization. Ellis Horwood Limited, Chichester, p 87-92. Poppius K, Hortling B & Sundquist J (1989b) Chlorine-free bleaching of chemical pulps - the potential of organic peroxyacids. 1989 International Symposium on Wood and Pulping Chemistry, Raleigh, NC, p 145-150. Poppius-Levlin K & Tuominen I (1991) Koivusulfaattimassan valkaisu milox- menetelmalla. PSC Comm 24. Poppius-Levlin K, Toikkanen L, Tuominen I & Sunquist J (1991a) Increased reactivity of kraft pulps towards non-chlorine bleaching chemicals. 6th International Symposium on Wood and Pulping Chemistry Proc., Melbourne, Australia, 1: 99-106. Poppius-Levlin K, Toikkanen L, Tuominen I & Sundquist J (1991b) Chlorine-free bleaching of kraft pulps. International Pulp Bleaching Conference, Production of Bleached Chemical Pulp in the Future, Stockholm, Sweden, 3: 103-120. Poppius-Levlin K, Mustonen R, Huovila T & Sundquist J (1991c) Milox pulping with acetic acid/peroxyacetic acid. Paperi ja Puu 73(2): 154-158. Poppius-Levlin K, Hortling B & Sundquist J (1993) Milox pulping and bleaching - possibilities to avoid chlorine chemicals. Seventh International Symposium on Wood and Pulping Chemistry Proc., Beijing, PR China, 1: 214-222. Poppius-Levlin K, Tuominen I, Kauliomaki S, Pohjanvesi S, Saari K, Sundquist J & Harma T (1995) Milox-prosessin koelaitos vuosina 1990-1994; Laitteisto, massan valmistus ja ominaisuudet. PSC Comm 81. Poppius-Levlin K, Jaaskelainen AS & Sundquist J (1996) Delignification of oxygen-treated softwood kraft pulp by different peroxy compounds. Fourth European Workshop on Lignocellulosics and Pulp, p 216-222. Povarova AY & Chupka EI (1994) Effectiveness of acid-alkaline and oxidation-reduction catalysis during oxidation of wood in aqueous organic media, (3), effect of oxygen pressure and temperature on the rate of wood and lignin oxidation in alkaline media. Izv VUZ Lesn Zh (3): 81-85. (ABIPST 66: 3396(AS)). Primakov SF & Barbash VA (1988) Optimization of parameters for pulping with aqueous alcohol solutions of sulfur dioxide. Khim Drev (Riga) (2): 44-47. (ABIPC 59:4013). Primakov SF & Barbash VA (1989) Delignification of wood with aqueous alcohol solutions of sulfur dioxide. Bum Prom (1): 6-7. (ABIPST 60:6073). Puech JL, Sarni F, Labidi A, Mouttet B & Robert A (1990) Delignification of oak wood with an ethanol-water solution in a flow-through reactor. Holzforschung 44(5): 367- 371. Puls J (1991) Characterization of hemicelluloses from organosolv pulping. 6th International Symposium on Wood and Pulping Chemistry Proc., Melbourne, Australia, 1: 21-25. Purina LT, Treimanis AP & Timermane GB (1992) Structural and chemical composition characteristics of pulp fibers isolated from hardwood by organic-solvent delignification. Khim Drev (Riga) (1): 3-12. (ABIPST 63:12217). Purina L, Treimanis A & Timermane G (1994) Distribution of residual lignin and hemicelluloses in the outer and cetral layers of organosolv pulp fiber walls. Nordic Pulp Paper Res J (2): 101-119. Puthson P, Kordsachia O, Odermatt J, Zimmermann M & Patt R (1997) ASAM pulping of eucalyptus camaldulensis and TCF bleaching of the resulting pulp. Holzforschung 51(3): 257-262. Puuronen M, Muurinen E & Sohlo J (1995) Organosolv pulping of reeds. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 2: 467-472. Puuronen M, Muurinen E & Sohlo J (1997) Peltobiomassat erikoissellujen raaka-aineina. Univ Oulu, Dept Process Eng, Report 168. Pye K (1987) Alcell offers savings with quality pulps, by-products. PIMA 69(11): 21-23. Pye EK (1989a) The Alcell process. 1989 Solvent Pulping Conference Proc., Seattle, WA, 16-32. Pye EK (1989b) Alcohol pulping - environmentally sound opportunity for nonintegrated paper mills. 6th International Pulp Symposium Prep., 2: paper 6. (ABIPST 67:13950(AC)). Pye EK (1990) The Alcell process - a proven alternative to kraft pulping. 1990 Pulping Conference, Toronto, Canada, 991-996. Pye EK & Lora JH (1991) The Alcell process - a proven alternative to kraft pulping. Tappi J 74(3): 113-118. Pye EK, Klein WR, Lora JH & Cronlund M (1987) The AlcellTM process. Solvent Pulping - Promises & Problems Conference, Appleton, WI, 55-67. Quinde-Abad A (1990) Behavior of the major resin and fatty acids of slash pine (pinus elliottii) during organosolv pulping. PhD. thesis, Univ British Columbia. (ABIPST 64:743(DU)). Ramos J, Davalos F & Gonzales P (1993) Sugar cane bagasse pretreatment with soda- methanol for the chemimechanical pulp production. Seventh International Symposium on Wood and Pulping Chemistry Proc., Beijing, PR China, 3:274-279. Rantakokko J, Maunuksela J & Malone J (1994) Einsatz von peroxyessigsaure anstelle von bioziden - erfolgreicher versuch zur schleimbekampfung in einer papierfabrik. Papier 48(11): 681-686. Rassow B & Gabriel H (1931a) Ueber das fichtenholz-lignin. Cellulosechemie 12(8): 227- 235. Rassow B & Gabriel H (1931b) Ueber das fichtenholz-lignin. Cellulosechemie 12(9/10): 249-254. Rassow B & Gabriel H (1931c) Ueber das fichtenholz-lignin. Cellulosechemie 12(11): 290- 295. Rassow B & Gabriel H (1931d) Ueber das fichtenholz-lignin. Cellulosechemie 12(12): 318- 320. Rassow B & Lude R (1931) Ueber bambuslignin. Zeitschrift Angew Chemie 44(40): 827- 833. Rassow B & Wagner K (1931a) Ueber die einwirkung von glykol-salzsaure auf kiefernholz. Wochenbl Papierfabr 62(23A): 35-42. Rassow B & Wagner K (1931b) Ueber die einwirkung von glykol-salzsaure auf kiefernholz. Wochenbl Papierfabr 62(31): 741-743. Rautenbach R & Albrecht R (1989) Membrane processes. Ellis Horwood, Chichester. Ray AK, Rao NJ, Mohanty B & Garceau JJ (1993) Methanol based organosolv pulping of wheat straw of indian origin. 1993 Pulping Conference Proc., Atlanta, GA, 1: 281-286. Reznikov VM (1988) Acetic acid delignification of hardwoods. Bum Prom (9): 12-13. (ABIPC 59:9137). Reznikov VM & Zilbergleit MA (1980) Belorussian Technological Institute, assignee. Paper pulp intermediate product. USSR patent 761.647. (CA 93:222146m). Ricard E & Guinot HM (1928) Manufacture of anhydrous formic acid. US patent 1.896.100. Rickelton WA & Boyle RJ (1988) Solvent extraction with organophosphines - commercial & potential applications. Separ Sci Tech 23(12&13): 1227-1250. Ricker NL, Michaels JN & King CJ (1979) Solvent properties of organic bases for extraction of acetic acid from water. J Separ Process Tech 1(1): 36-41. Ricker NL, Pittman EF & King CJ (1980) Solvent extraction with amines for recovery of acetic acid from dilute aqueous industrial streams. J Separ Process Tech 1(2): 23-30. Ritter GJ & Barbour JH (1935) Effect of pretreatments of wood on the lignin determination, distribution of methoxyls in wood. Ind Eng Chem Anal Ed 7(4): 238- 240. Robert TA (1955) Hydrotropic pulping of a tropical wood. Tappi 38(9): 149A-150A. Robert D, Neirinck V, Pan X, Evtuguine D, Neto CP & Chum HL (1993) Production and characterization of organosolv lignins: 13C NMR as a process control. Seventh International Symposium on Wood and Pulping Chemistry Proc., Beijing, PR China, 2: 889-893. Roberts RS, Muzzy JD & Faass GS (1988) Georgia Tech Research Corp., assignee. Process for extracting lignin from lignocellulosic material using an aqueous organic solvent and an acid neutralizing agent. US patent 4.746.401. Ronan P, Rossiter A, Agar R & Cheung L (1994) Timely pinch analysis improves energy efficiency of new pulping process. 1994 Tappi Engineering Conference Proc., p 937- 951. Rosch J & Mulhaupt R (1994) Mechanical properties of organosolv-lignin-filled thermoplastics. Polym Bull 32: 361-365. Rose LM (1985) Distillation design in practice. Elsevier Science Publishers, Amsterdam. Rousu E (1996) Muurahaishappo keittokemikaalina, erikoisselluloosaa non-wood-raaka- aineista. Paperi ja Puu 78(10): 594- 596. Ruthven DM (1984) Principles of adsorption and adsorption processes. John Wiley & Sons, New York. Rutkowski J, Mroz W, Surma-Slusarska B & Perlinska-Sipa K (1993) Glycolic delignification of hardwood. Progress 93 Conference Proc., 1: 190-205. (ABIPST 67:13953(AC)). Rutkowski J, Mroz W & Perlinska-Sipa K (1994a) Glycol-acetic wood delignification. Cell Chem Tech 28: 621-628. Rutkowski J, Mroz W & Perlinska-Sipa K (1994b) Glycol-acetic acid delignification of wood. Przegl Papier. 50(9): 437-439. (ABIPST 66:3624(AS)). Rutkowski J, Wendelt P & Perlinska-Sipa K (1998) Contribution to the technology of birch wood pulping with acetic acid solutions. Cellul Chem Technol 32(1-2): 89-96. (Paperbase 00530350). Saake B, Lehnen R, Lummitsch S & Nimz HH (1995a) Production of dissolving and paper grade pulps by the formacell process. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 2: 237-242. Saake B, Lummitsch S, Mormanee R, Lehnen R & Nimz HH (1995b) Erzeugung von zellstoffen nach dem Formacell-verfahren. Papier 49(10A): V1-V7. Saake B, Lehnen R, Schmekal E, Neubauer A & Nimz HH (1998) Bleaching of formacell pulp from aspen wood with ozone and peracetic acid in organic solvents. Holzforschung 52(6): 643-650. (Paperbase 00526322). Saari K, Vuori A & Mattila T (1992) Kemira Oy, assignee. Menetelma alifaattisten happojen ja sokerien talteenottamiseksi. FI patent 85511. Saarnio J & Kuusisto R (1971) D-xylose from sulphite waste liquor. Paperi ja Puu 53(6): 367-370. Sabatier J, Irulegui A & Tomas J (1989) Ethanolic pulping of sugar cane bagasse. 1989 International Symposium on Wood and Pulping Chemistry Proc., Raleigh, NC, p 781- 783. Saeman JF (1945) Kinetics of wood saccharification. Ind Eng Chem 37: 43-54. Sakai K, Kuroda K & Kishimoto S (1972) Delignification in peracid bleaching, VI, reaction of the beta-aryl ether type model compounds with peracetic acid. Tappi 55(12): 1702-1706. Sakai K, Mun SP, Fukuda J & Imamura H (1983) Delignification with sulfites and an organic solvent - semichemical and chemical pulping of sugi (japanese cedar). 2nd International Symposium on Wood and Pulping Chemistry, Tsukuba Science City, Japan, 4: 194-197. Sakai K, Mun SP & Imamura H (1987) Organosolv pulping which produces valuable by- products. Fourth International Symposium on Wood and Pulping Chemistry, Paris, France, 2: 193-195. Sakai K & Uprichard JM (1987) Isopropanol-sulphite pulping studies on radiata pine. Appita 40(3): 193-200. Sakakibara A & Edashige Y (1984) Solvolysis pulping (part 2), solvolysis pulping of spruce wood (picea glehnii) with cresol-water system, and the pulp strength. Japan Tappi 38(3): 73-80. Sakakibara A & Nakayama N (1962a) Hydrolysis of lignin with dioxane and water, II, identification of hydrolysis products. J Japan Wood Res Soc 8: 153-166. Sakakibara A & Nakayama N (1962b) Hydrolysis of lignin with dioxane and water, III, Hydrolysis products of various woods and lignin preparations. J Japan Wood Res Soc 8: 157-162. Sakakibara A, Edashige Y, Sano Y & Takeyama H (1983) Solvolysis pulping with lignin degradation products. 1983 International Symposium on Wood and Pulping Chemistry, Tsukuba Science City, Japan, 3: 33-49. Sakakibara A, Edashige Y, Sano Y & Hatakeyama H (1984) Solvolysis pulping with cresols-water system. Holzforschung 38(3): 159-165. Salentev DG (1991) Chemiluminescence during oxidation of wood and its componets in water-organic systems. Institut khimii drevesiny, Issledovaniya v oblasti khimii drevesiny 14. (ABIPST 65:6594(AC)). Sameshima K & Takamura N (1987) The ammonium oxalate pulping of bast fiber. Fourth International Symposium on Wood and Pulping Chemistry, Paris, France, p 199-201. Sanjuan R & Chavez A (1989) Organosolv pulping with eucalyptus globulus. Amatl 3(3/4): 3-9. (ABIPST 61:12171). Sanjuan R & Gomez F (1991) Multistage pulping of sugarcane bagasse. Amatl 5(1/2): 3- 11. (ABIPST 63:13607). Sanjuan R, Chavez H & Vargas J (1989) Organosolv pulping of eucalyptus globulus. ABTCP Congr. Anual Celulose Papel 22nd, Sao Paulo, Brazil, p 165-176. (ABIPST 62:11080(M)). Sanjuan R, Prado J, Romero N & Irulegui A (1993) Low pollution alternative processes to obtain bleached pulp from sugar cane bagasse. Seventh International Symposium on Wood and Pulping Chemistry Proc., Beijing, PR China, 3: 280-283. Sanjuan R, Vargas J, Anzaldo J, Munguia L & Delgado E (1995) Comparative study of three processes to obtain pulp from kenaf (hibiscus cannabinus). The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 2: 449-454. Sano Y (1989) Solvolysis pulping of softwood. In: Kennedy JF, Phillips GO & Williams PA (eds) Wood processing and utilization. Ellis Horwood, Chichester, p 111-116. Sano Y & Sasaya T (1985) Studies on phenorganosolv pulping I, delignification of woods by modified organosolv pulping. Mokuzai Gakkaishi 31(10): 836-842. Sano Y, Umemoto M, Takahashi A & Sasaya T (1986) Phenorganosolv pulping II, pulping of beechwood with acetic acid-water-phenols system. Mokuzai Gakkaishi 32(12): 1003-1010. Sano Y, Sasaya T & Sakakibara A (1988) Solvolysis pulping of hardwoods. Japan Tappi J 42(5): 487-496. (ABIPC 59:9138). Sano Y, Maeda H & Sakashita Y (1989a) Pulping of wood at atmospheric pressure I., pulping of hardwoods with aqueous acetic acid containing a small amount of organic sulfonic acid. Mokuzai Gakkaishi 35(11): 991-995. Sano Y, Endo M & Sakashita Y (1989b) Solvolysis pulping of softwoods with aqueous cresols containing a small amount of acetic acid. Mokuzai Gakkaishi 35(9): 807-812. Sano Y, Nakamura M & Shimamoto S (1990) Pulping of wood at atmospheric pressure II, pulping of birch wood with aqueous acetic acid containing a small amount of sulfuric acid. Mokuzai Gakkaishi 36(3): 207-211. Sansigolo CA & Curvelo AAS (1994) Delignification in aqueous ethanol of eucalyptus globulus: characteristics of lignin and pulp. Papel 55(8): 22-29. (ABIPST 65:18825). Sansigolo CA & Curvelo AAS (1995) Kinetics of pulping in ethanol and water and properties of eucalyptus globulus pulp. Papel 56(8): 38-42, 44-45. (ABIPST 66:17375(AS)). Sarkanen KV (1980) Acid-catalyzed delignification of lignocellulosics in organic solvents. In: Sarkanen KV & Tillman DA (eds) Progress in biomass conversion, New York, 2: 127-144. Sarkanen KV (1982) University of Washington, assignee. Method and system for selective alkaline defiberization and delignification. US patent 4.329.200. Sarkanen K & Schuerch C (1957) Lignin structure, XI, quantitative study of the alcoholysis of lignin. J Am Chem Soc 79: 4203-4209. Sarkanen S, Teller DC, Hall J & McCarthy JL (1981) Lignin, 18, associative effects among organosolv lignin components. Macromolecules 14: 426-434. Savov K (1990) Donor-acceptor reactions during degradation of lignin in an acid medium. Tselul Khartiya 20(4): 12. (ABIPST 63:8357). Savov KA, Tzanova SE & Draganova RJ (1996) Comparison of the catalytic activities of aluminium trichloride and hydrochloric acid in the acetic acid-lignin system. Holzforschung 50(2): 188-192. (ABIPST 67:2125(AS)). Schild G, Muller W & Sixta H (1996) Prehydrolysis kraft and ASAM paper grade pulping of eucalypt wood, a kinetic study. Papier 50(1): 10-22. Schillmoller CM (1997) Control organic-acid corrosion with these metals and alloys. Chem Eng Prog 93(2): 66-71. Schorning P (1961) Untersuchungen zur steuerung des sulfitaufschlusses von holzern in methanol-wasser bei gegenwart von Na2 SO4 bzw NaHSO4 bzw Na2SO3. Faserforsch Textiltech 12(3): 105-108. Schroeter MC (1991) Possible lignin reactions in the Organocell pulping process. Tappi J 74(10): 197-200. Schroeter MC (1993) Solvent pulping can be an alternative to the kraft process. China Paper 1993 Proc., p 63-68. Schroeter MC & Dahlmann G (1991) Organocell simplifies the solvent pulping process. 1991 Pulping Conference Proc., Orlando, FL, 2: 646-651. Schubert HL (1990) Asam report 1. Kraftanlagen Aktiengesellschaft, Heidelberg. Schubert HL, Kordsachia O & Patt R (1990) Pilotanlagenversuche zum ASAM-verfahren. Wochenbl Papierfabr 118(22): 977-981. Schubert HL, Fuchs K, Patt R, Kordsachia O & Bobik M (1993) Der ASAM-prozess - eine industriereife zellstofftechnologie. Papier 47(10A): V6-V15. Schubert H, Kordsachia O, Shackford L & Shin N (1998) High yield and easy bleachability - the ASAM process. 1998 Breaking the Pulp Yield Barrier Symp., Atlanta, GA, p 241- 246. (ABPST 69:373(AC)). Schutz F (1940) Ueber den aufschluss von faserrohstoffen mit wasserhaltiger monochloressigsaure. Cellulosechemie 18(4): 76-83. Schutz F (1941a) Ueber den aufschluss von faserrohstoffen mit chlorhydrinen. Cellulosechemie 19(2): 33-38. Schutz F (1941b) Untersuchung der beim holzaufschluss mit wasserhaltiger chloressigsaure entstehenden lignine. Cellulosechemie 19(4): 87-89. Schutz F & Knackstedt W (1942) Holzaufschluss mit salzsaure oder chloriden als katalysator in essigsaurer losung. Cellulosechemie 20(1): 15-22. Schweers W (1972a) Belastung von luft und gewassern durch holzchemische prozesse. Forstarchiv 43: 43-49. Schweers W (1972b) Umweltbelastung durch holzchemische prozesse. Ber Landwirtsch 50: 312-324. Schweers WHM (1974) Phenol pulping - a potential sulfur-free papermaking process. Chemtech 4(8): 490-493. Schweers W & Meier D (1979) Ueber eigenschaften und abbaubarkeit von mit alkohol- wasser-gemischen isolierten ligninen, 1. mitteilung: halbmikropyrolyse im stationaren system. Holzforschung 33(1): 15-18. Schweers W & Pereira H (1972) Ueber den holzaufschluss mit phenolen, 1. mitt., trankung verschiedener holzer mit phenol. Holzforschung 26(2): 51-54. Schweers W & Rechy M (1972) Moglichkeiten eines schwefelfreien holzaufschlusses. Papier 26(10A): 585-590. Schweers W & Rechy M (1973) Ueber den holzaufschluss mit phenolen, 3. mitt., ueber den aufschluss von kiefern- und buchenholz. Papier 27(12): 636-639. Schweers W, Behler H & Beinhoff O (1972) Ueber den holzaufschluss mit phenolen, 2. mitt., vorlaufige untersuchungen uber die phenolbilanz. Holzforschung 26(3): 103-105. Schwenzon K (1965) Hydrotroper aufschluss von pflanzen. Zellstoff Papier 14(7): 199- 201. Schwenzon K (1966) Hydrotroper aufschluss von pflanzen. Zellstoff Papier 15(7): 202- 204. Seisto A (1993) Havupuumassan valmistus milox-menetelmalla. PSC Comm 55. Seisto A (1994) Peltokasvien keitto ja valkaisu milox-menetelmalla. PSC Comm 60. Seisto A (1998) Effect of peroxyacid delignification on pulp and papermaking properties. PSC Comm 107. Seisto A & Poppius-Levlin K (1995) Milox pulping of agricultural plants. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 1: 407-414. Seisto A & Poppius-Levlin K (1997a) Formic acid/peroxyformic acid pulping of birch - delignification selectivity and zero-span strength. Nord Pulp Pap Res J 12(3): 155-161. Seisto A & Poppius-Levlin K (1997b) Fibre characteristics and paper properties of formic acid/peroxyformic acid birch pulps. Nord Pulp Pap Res J 12(4): 237-243. Seisto A & Sundquist J (1994) Silikaatit milox-ruohomassan keitossa ja valkaisussa. PSC Comm 68. Seisto A, Poppius-Levlin K & Jousimaa T (1995) Grass pulp for papermaking by the peroxyformic acid pulping method. 1995 TAPPI Pulping Conference Proc., p 487-494. (ABIPST 66:7039(AC)). Seisto A, Poppius-Levlin K & Jousimaa T (1996) Grass pulp for papermaking by the milox method. PSC Comm 87. Sellers T (1994) Modification of phenolic resin with organosolv lignins and evaluation of strandboards made by the resin as binder. PhD. thesis, Univ Tokyo. (ABIPST 65:8714(BDU)). Sellers T, Lora JH & Okuma M (1994) Organosolv-lignin-modified phenolics as strandboard binder, I., organosolv lignin and modified phenolic resin. Mokuzai Gakkaishi 40(10): 1073-1078. Senyo WC, Creamer AW, Wu CF & Lora JH (1996) Use of organosolv lignin to reduce press-vent formaldehyde emissions in the manufacture of wood composites. For Prod J 46(6): 73-77. (ABIPST 67:4298(AS)). Shaban A & Suciu GD (1993) ABB Lummus Crest Inc., assignee. Process of pulping and bleaching fibrous plant material with tert-butyl alcohol and tert-butyl peroxide. US patent 5.252.183. (ABIPST 66:9747(P)). Shah DJ & Tiwari KK (1981) Recovery of acetic acid from dilute aqueous streams using liquid-liquid extraction with tri-n-butyl phosphate as solvent. J Separ Process Tech 2(4): 1-6. Shimizu Y & Hayashi J (1989) Esterification of cellulose with organic acids. In: Kennedy JF, Phillips GO & Williams PA (eds) Cellulose: structural and functional aspects. Ellis Horwood, Chichester, p 187-193. Shiraishi N, Fu S, Yokota T & Tsujimoto N (1988) Oji Paper Company Ltd., assignee. Solvolytic pulping of lignocellulose. Japan Kokai patent 21.993/88. (ABIPC 59:1199). Sierra-Alvarez R & Tjeerdsma B (1995a) Organosolv pulping of juvenile poplar wood. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 2: 207-211. Sierra-Alvarez R & Tjeerdsma B (1995b) Organosolv pulping of poplar [populus] wood from short-rotation intensive-culture plantations. Wood Fiber Sci 27(4): 395-401. (ABIPST 66: 17379(AS)). Silva Perez D, Terrones MGH, Grelier S, Nourmamode A, Castellan A, Ruggiero R & Machado AEH (1998) Peroxyformic acid pulping of eucalyptus grandis wood chips and sugar cane bagasse in one stage and characterization of the isolated lignins. J Wood Chem Technol 18(3): 333-365. (Paperbase 00522189). Simkhovich BS, Zilbergleit MA & Reznikov VM (1987) Papermaking properties of acetic acid pulp from hardwoods. Bum Prom (7): 25-26. (ABIPST 60:478(T)). Singh SV, Bhandari SS & Singh SP (1989) Organosolv pulping of Dendrocalamus strictus and Eucalyptus tereticornis with aqueous ethanol. Silver Jubilee International Seminar & Workshop: Appropriate Technologies for Pulp & Paper Manufacture in Developing Countries (IPPTA), paper no. 35. (ABIPST 67:13956(AC)). Singh UP, Berry JC, Lora JH & Pye EK (1995) TCF bleaching of pulps from the Alcell process. International Non-Chlorine Bleaching Conference Proc., 12, paper no. 2. (ABIPST 66:19398(AC)). Sinner M, Dietrichs HH, Puls J, Schweers W & Brachthauset KH (1988) Bau- und Forschungsgesellschaft Thermoform AG, assignee. Process for production of xylitol from lignocellulosic raw materials. US patent 4.742.814. Sinner M, Dietrichs HH, Puls J, Schweers W & Brachthauset KH (1989) Bau- und Forschungsgesellschaft Thermoform AG, assignee. Verfahren zum aufschluss von lignocellulosischen pflanzlichen rohstoffen zur gewinnung von zuckern, lignin und cellulose sowie dessen anwendung. DE patent 27.37.118.C2. Skachkov VM, Tanicheva RV, Mikhailov GS, Kozhevnikov PS & Sokov AN (1991) Enzymolysis of oxygen-solvolysis pulp. Gid Les Prom (7): 4-5. (ABIPST 63:8527). Smith DE (1987) Ester pulping of aspen [populus]. Workshop on Aspen Populus[ ], Pulp, Paper and Chemicals, 2: 172-176. (ABIPST 67:2127(AC)). Smith DK, Bampton RF & Alexander WJ (1963) Use of new solvents for evaluating chemical cellulose for the viscose process. Ind Eng Chem Fund Pro Des Prod Res 2(1): 57-62. Smith C, Utley JHP, Petrescu M & Viertler H (1989) Biomass electrochemistry: anodic oxidation of an organosolv lignin in the presence of nitroaromatics. J Appl Electrochem 19: 535-539. Sobral MF, Gomide JL & Colodette JL (1992) ASAE (alkaline sulfite anthraquinone ethanol) process for producing eucalyptus pulp. 25th Congresso Anual Celulose Papel ABTCP, Sao Paulo, Brazil, p 179-193. (ABIPST 65:5064(M)). Solar R & Kacik F (1993) Comparative study of caribean pine (pinus caribaea l.) wood and bark dioxane lignin. Holz Roh Werkst 51(5): 347-352. Solar R & Kacik F (1995) Alteration of maple [acer] wood lignin under conditions of treatment in dioxane-water-HCl agent. Drev Vysk 40(1): 3-16. (ABIPST 66:13615(AS)). Solar R, Melcer I & Kacik F (1988) Comparative study of pine wood and pine bark (pinus silvestris l.) dioxan lignins. Cell Chem Tech 22: 39-52. Sousa M, Schuchardt U & Rodrigues JAR (1986) Separation and identification of sugar cane bagasse constituents by the organosolv process. Ciencia Cultura 38(1): 181-188. (ABIPC 59:2016). Springer EL & Zoch LL (1966) Delignification of aspen wood with acidified aqueous solutions of sulfolane. Svensk Pappersatidn 69(16): 513-516. Springer EL, Harris JF & Neill WK (1963) Rate studies of hydrotropic delignification of aspenwood. Tappi 46(9): 551-555. Starr LD & Casebier RL (1970) Rayonier Incorporated, assignee. Process for pulping wood chips with sodium sulfide and organic solvent. US patent 3.513.068. Staudinger H & Dreher E (1936) Ueber hochpolymere verbindungen, 142. mitteil., ueber das lignin. Ber Deutschen Chem Ges 69: 1729-1737. Stevens WP & Marton R (1966) Effect of peracetic acid on the properties of high yield pulps. Tappi 49(10): 452-456. Suleman Y & Yusoff M (1997) thanol-water pulping of oil palm (Elaeis guineensis) fibers using hydrochloric acid as a catalyst and test-paper characterization. J Jpn Wood Res Soc 43(12): 1016-1021. (ABIPST 68:11919(AS)). Sun YP, Abbot J & Yates B (1994a) UV-peroxide bleaching of peroxyformic acid pretreated eucalypt kraft pulp - tcf bleaching. 48th Appita Annual General Conference Proc., p 155-162. Sun YP, Abbot J & Chen CL (1994b) UV-peroxide bleaching of peroxyformic acid pretreated eucalypt kraft pulp: tcf bleaching. Appita 47(6): 459-462. Sun R, Lawther J & Banks W (1997) Physico-chemical characterization of organosolv lignins from wheat straw. Cellul chem Technol 31(3/4): 199-212. (ABIPST 68:6582(AS)). Sundquist J (1985) The multiform nature of residual lignin in chemical pulps. 1985 International Symposium on Wood and Pulping Chemistry Technical Papers, Vancouver, BC, p 23-25. Sundquist J (1986) Bleached pulp without sulphur and chlorine chemicals by a peroxyacid/alkaline peroxide method - an overview. Paperi ja Puu 68(9): 616-620. Sundquist J (1995) Muurahaishappoa sellunvalmistukseen. Tekniikan Nakoalat (5): 48-49. Sundquist J (1996) Chemical pulping based on formic acid, summary of Milox research. Paperi ja Puu 78(3): 92-95. Sundquist J & Poppius-Levlin K (1992) Milox pulping and bleaching - the first pilot scale trials. 1992 Solvent Pulping Symposium Notes, Boston, MA, p 45-49. Sundquist J & Poppius-Levlin K (1998) Milox pulping and bleaching. In: Young RA 6 Akhtar M (eds) Environmentally Friendly Technologies for the Pulp nad Paper Industry. John Wiley & Sons, New York, NY, p 157-190. Sundquist J, Laamanen L, Wartiovaara I, Poppius K & Kauliomaki S (1986) Manufacture of bleached pulp without sulphur and chlorine - utopia or not? EUCEPA Symposium Proc., Helsinki, Finland, p 29-33. Sundquist J, Laamanen L & Poppius K (1988) Problems of non-conventional pulping in the light of peroxyformic acid cooking experiments. Paperi ja Puu 70(2): 143-148. Sundquist J, Poppius K, Hortling B, Setala J, Heikonen M & Moisio T (1989) Spent liquor from peroxyformic acid pulping as preserving agent for animal fodder. 1989 International Symposium on Wood and Pulping Chemistry Poster Sessions, Raleigh, NC, p 233-234. Surma-Slusarska B (1998) Balance of ethylene glycol in organosolv pulping of hardwood. Przegl Pap 54(12): 712-714. (Paperbase 00537502). Takagi H & Kachi S (1989) Research and development of solvolysis pulping, (4), bleaching of solvolysis pulp. Japan Tappi J 43(12): 1290-1297. (ABIPST 61:339). Takagi H, Wakai M & Araki H (1989) Research and development of solvolysis pulping, (3), cooking conditions and pulp properties. Japan Tappi J 43(11): 1171-1178. (ABIPST 60:9147). Takeyama H & Sasaya T (1988) Hydrocracking of solvolysis lignin (3). Enshurin Ken Hok (Hokkaido) 45(2): 637-651. (ABIPC 59:8077). Tamada JA & King CJ (1990a) Extraction of carboxylic acids with amine extractants, 2. chemical interactions and interpretation of data. Ind Eng Chem Res 29(7): 1327-1333. Tamada JA & King CJ (1990b) Extraction of carboxylic acids with amine extractants, 3. effect of temperature, water coextraction, and process considerations. Ind Eng Chem Res 29(7): 1333-1338. Tamada JA, Kertes AS & King CJ (1990) Extraction of carboxylic acids with amine extractants, 1. equilibria and law of mass action modelling. Ind Eng Chem Res 29(7): 1319-1326. Teder A & Olm L (1992) Alternativa kokprocesser: favoriten modifierat sulfatkok - utmanaren alkalisk sulfitprocess. Svensk Papperstidn 95(7): 26-32. Teder A, Tormund D & Ulmgren P (1990) Luktfri sulfatmassafabrik en mojlighet?; jakten pa H2S och CH3SH ger resultat. Svensk Papperstidn 93(7): 155-160. Terenteva EP, Zorina RI, Varvarskaya SV & Dievskii VA (1990) Changes in the structure of aspenwood [populus] during pulping by the organic solvent method. Khim Drev (Riga) (3): 41-44. (ABIPST 62:5721). Thompson NS & Kaustinen OA (1964) Some chemical and physical properties of pulps prepared by mild oxidative action. Tappi 47(3): 157-162. Thompson NS & Kaustinen OA (1969) The Institute of Paper Chemistry, assignee. Fibrous material and its treatment. CA patent 806.573. Thring RW, Chornet E, Overend RP & Heitz M (1989a) Production and hydrolytic depolymerization of ethylene glycol lignin. In: Glasser W & Sarkanen K (eds) Lignin properties & materials. ACS, Washington, p 228-244. (ABIPST 60:8994(B)). Thring RW, Chornet E, Vidal PF, Bouchard J & Overend RP (1989b) Solvent fractionation of ethylene glycol lignin. 1989 International Symposium on Wood and Pulping Chemistry Poster Sessions, Raleigh, NC, p 343-345. Thring RW, Chornet E & Overend RP (1990a) Recovery of a solvolytic lignin: effects of spent liquor/acid volume ratio, acid concentration and temperature. Biomass 23(4): 289- 305. Thring RW, Chornet E, Bouchard J, Vidal PF & Overend RP (1990b) Characterization of lignin residues derived from the alkaline hydrolysis of glycol lignin. Can J Chem 68(1): 82-89. Thring RW, Chornet E, Bouchard J, Vidal PF & Overend RP (1991) Evidence for the heterogeneity of glycol lignin. Ind Eng Chem Res 30(1): 232-240. Thring RW, Chornet E & Overend RP (1993a) Fractionation of woodmeal by prehydrolysis and thermal organosolv; alkaline depolymerization, chemical functionality, and molecular weight distribution of recovered lignins and their fractions. Can J Chem 71(6): 779-789. Thring RW, Chornet E & Overend RP (1993b) Thermolysis of glycol lignin in the presence of tetralin. Can J Chem Eng 71(2): 107-115. Thring RW, Chornet E & Overend RP (1993c) Fractionation of woodmeal by prehydrolysis and thermal organosolv; process strategy, recovery of constituents, and solvent fractionation of lignins so produced. Can J Chem Eng 71(2): 116-123. Timermane GB, Purina LT, Ioelovich MY & Treimanis AP (1992) Kinetics of hardwood delignification in an organic solvent. Khim Drev (Riga) (4-5): 36-40. (ABIPST 64:4790). Tirtowidjojo S, Sarkanen KV, Pla F & McCarthy JL (1988) Kinetics of organosolv delignification in batch- and flow-through reactors. Holzforschung 42(3): 177-183. Tissot M & Launay A (1987) Study of a two-phase soda cooking process of eucalyptus in presence of alcohol. 4th International Symposium on Wood and Pulping Chemistry, Paris, France, 1: 373-383. (ABIPC 59:8240(M)). Tjeerdsma BF & Zomers FHA (1993) Organosolv pulping of hemp. Seventh International Symposium on Wood and Pulping Chemistry Proc., Beijing, PR China, 1: 514-522. Tjeerdsma BF, Zomers FHA, Wilkinson EC & Sierra-Alvarez R (1994) Modelling organosolv pulping of hemp. Holzforschung 48(5): 415-422. Tournier H, Sachetto JP, Michel JP, Roman A, Johansson AA & Armanet JM (1986) Neste Oy, assignee. Menetelma puiden ja muiden lignoselluloosatuotteiden delignifioimiseksi. FI patent 70938. Troitskii VV, Kiryushina MF, Ermakova MI & Zarubin MY (1989) Method of producing pulp. USSR patent 1.461.800. (ABIPST 61:3412(U)). Tubino M & Filho O (1989) Hybrid process for pulping wood: pulping of eucalyptus wood with an aqueous solution of ethanol and sodium hydroxide having a low alcohol content. ABTCP Congr Anual Celulose Papel 22nd, Sao Paulo, Brazil, p 177-201. (ABIPST 62:11091(M)). Tubino M & Filho O (1990) Hybrid pulping process - pulping of eucalyptus with an aqueous solution of ethanol and sodium hydroxide with a low alcohol content. Papel 51(7): 33-47. (ABIPST 61:7209). Tuominen I (1995) Milox-prosessin kemikaalitaseen tarkastelu tuotantomittakaavassa. PSC Comm 80. Unger A, Kletzin J, Reichelt L & Poller S (1979) Institut fur Forstwissenschaften Eberswald, assignee. Verfahren zur auflosung von lignozellulosehaltigen materialien. DD patent 136.845. Uraki Y, Sano Y & Sasaya T (1991) Cooking of hardwoods with organosolv pulping in aqueous acetic acid containing sulfuric acid at atmospheric pressure. Japan Tappi J 45(9): 1018-1024. Uraki Y, Ogawa S, Sano Y, Sasaya T & Takai M (1993) Influence of delignification on physicochemical properties of solvolysis pulp. Sen-I-Gakkaishi 49(7): 367-372. Uraki Y, Kubo S, Nigo N, Sano Y & Sasaya T (1995) Preparation of carbon fibers from organosolv lignin obtained by aqueous acetic acid pulping. Holzforschung 49(4): 343- 350. (ABIPST 66:8564(AS)). Urban P & Engel DJ (1988) UOP Inc., assignee. Process for liquefaction of lignin. US patent 4.731.491. Valladares J, Rolz C, Bermudez ME, Batres FR & Custodio MA (1984) Pulping of sugar cane bagasse with a mixture of ethanol-water solution in the presence of sodium hydroxide and anthraquinone. In: Seaquist AJ & Cobb EC (eds) Nonwood plant fiber pulping, progress report. Tappi Press, Atlanta, 15: 23-28. Vanasse C, Chornet E & Overend RP (1988) Liquefaction of lignocellulosics in model solvents: creosote oil and ethylene glycol. Can J Chem Eng 66(1): 112-120. Vanhanen T, Muurinen E & Sohlo J (1989) Liuottimen regenerointi organosolv prosesseista, etanoli-alkali-AQ-prosessi ja sivutuotteiden talteenotto. Univ Oulu, Dept Process Eng, Report 115. Vanharanta H & Kauppila J (1984) Engineering aspects and feasibility of a wood pulping process using Batelle-Geneva aqueous phenol process by Rintekno Oy. Proc. of SPCI World Pulp & Paper Week, p 92-96. Vazquez D, Lage MA, Parajo JC & Vazquez G (1992) Fractionation of eucalyptus wood in acetic acid media. Biores Tech 40: 131-136. Vedernikov D & Deineko I (1994) Lignin model compounds conversions under oxygen- organosolv delignification conditions. Third European Workshop on Lignocellulosics and Pulp, p 226-227. (ABIPST 66:10000(AC)). Vega A & Bao M (1993) Organosolv fractionation of ulex europaeus with dilute hydrochloric acid and phenol, two simple kinetic models for pre-hydrolysis and delignification. Wood Sci Tech 27: 61-68. Vega A, Bao M & Lamas J (1997a) Application of factorial design to the modeling of organosolv delignification of miscanthus sinensis (elephant grass) with phenol and dilute acid solutions. Bioresour Technol 61(1): 1-7. (ABIPST 68:8498(AS)). Vega A, Bao M & Rodriguez J (1997b) Fractionating lignocellulose of miscanthus sinensis with aqueous phenol in acidic medium. Holz Roh Werkst 55(3): 189-194. (ABIPST 68:1125(AS)). Vorher W & Schweers WHM (1975) Utilization of phenol lignin. Proc. of the Eighth Cellulose Conf., Syracuse, NY, 1: 277-284. Vuori A, Mattila T & Saari K (1990) Kemira Oy, assignee. Menetelma ligniinin ottamiseksi talteen. FI patent appl. 902676. Vuorinen T (1983) Alkali-catalyzed oxidation of d-glucose with sodium 2- anthraquinonesulfonate in ethanol-water solutions. Carbohydr Res 116(1): 61-69. Wabner TW, Grambow C & Ober J (1986) Ein umweltfreundliches verfahren zur elektrochemischen lignin-ausfallung. Chem-Ing-Tech 58(4): 328-330. Wacek A & David E (1937) Ueber die einwirkung von lauge und natriumaethylat auf substituirte chalkone und dehydro-di-isoeugenol-methylaether, modellversuche zur ligninspaltung, II mitteil. Ber Deutschen Chem Gesells 70B(2): 190-195. Wacek A & Morghen I (1937) Spaltung von propenyl-brenzcatechin-aethern mit natriumalkoholat, modellvesuche zur ligninspaltung, I mitteil. Ber Deutschen Chem Gesells 70B(2): 183-189. Wagner FS (1978) Acetic acid. In: Grayson M (ed) Kirk-Othmer encyclopedia of chemical technology, John Wiley & Sons, New York, 11: 124-147. Wagner FS (1980) Formic acid and derivates. In: Grayson M (ed) Kirk-Othmer encyclopedia of chemical technology, John Wiley & Sons, New York, 24: 251-258. Walas SM (1985) Phase equilibria in chemical engineering. Butterworth, Boston. Wallis AFA (1976) Reaction of lignin model compounds with ethanolamine. Cell Chem Tech 10: 345-355. Wallis AFA (1978) Wood pulping with mono-, di- and triethanolamine. Appita 31(6): 443- 448. Wallis AFA (1980) Wood pulping with monoethanolamine in pressure vessels. Appita 33(5): 351-355. Wang H & Tseng HY (1993) Studies on the ethanol-alkali pulping of some Taiwan hardwoods. Seventh International Symposium on Wood and Pulping Chemistry Proc., Beijing, PR China, 1: 272-278. Wang WX, Chen ZH, Liu SC, Qiu ST & Wu YY (1990) Wheat-straw pulping and defibration point. Jiangsu Zaozhi (2/3): 10-14. (ABIPST 62:11093(U)). Wang IC, Kuo ML & Ku YC (1993) Alcohol pulping of plantation taiwania and eucalyptus, (2), scanning electron microscope observations. Bull Taiwan For Res Inst 8(3): 177-185. (ABIPST 66:2183(AS)). Wardell JM & King CJ (1978) Solvent equilibria for extraction of carboxylic acids from water. J Chem Eng Data 23(2): 144-148. Wayman M & Harris GR (1960) Columbia Cellulose Company Ltd., assignee. Pulping of lignocellulosic material. US patent 2.939.813. Wayman M & Lora JH (1978) Aspen autohydrolysis, the effects of 2-naphtol and other aromatic compounds. Tappi 61(6): 55-57. Weast RC (1978) CRC handbook of chemistry and physics. CRC Press Inc, West Palm Beach. Wedekind E (1936) Eigenschaften und verwendungsarten des buchenzellstoffes. Cellulosechemie 17(5/6): 45-49. Wedekind E, Engel O, Storch K & Tauber L (1931) Zur erkenntnis des lignins, 11. mitteilung, phenol-lignine und ihre acetylderivate. Cellulosechemie 12(6): 163-173. Weferling N, Pfuller P, Kanzler W, Schedler J & Thalhammer H (1988) Hoechst AG, assignee. Kontinuierliches verfahren zur extraktion von carbonsauren, aldehyden, ketonen, alkoholen und phenolen aus verdunnten wassrigen losungen. DE patent 36.26.968.A1. Wegener G, Rothe A & Lindner A (1991) Chemical modification of Organocell lignin by means of hydroxypropylation. In: Faix O & Meier D (eds) Mitteilungen der Bundesforschungsanstalt fur Forst- un Holzwirtscahft 168: 145-150. Weidenmuller R (1992) Schwefelfreier zellstoff - umweltgerecht und wirtschaftlich, das Organocell-verfahren. Papier 46(10): 611-614. Weisheng L, Hanqiang Z, Fengmei C & Jie G (1988) Organosolve pulping of bagasse with ethanol-water system. 1988 International Non-Wood Fiber Pulping and Papermaking Conference Proc., Beijing, PR China, p 281-290. Wells GL & Rose LM (1986) The art of chemical process design. Elsevier Science Publishers, Amsterdam. Westmoreland RA & Jefcoat IA (1991) Sulfur dioxide-ethanol-water pulping of hardwoods. Chem Eng Comm 104: 101-115. Williams W (1989) Organocell: cooking the books. Paper 211(10): 34, 37. Williamson PN (1987) Repaps Alcell process: how it works and what it offers. Pulp Paper Canada 88(12): 47-49. Williamson PN (1988) Repaps Alcell-process: ny demonstrationsanlaggning visar vagen till billigare massabruk. Svensk Papperstidning 91(7): 21-23. Wiltshire WA (1943) Acetic acid digestion of wood. Technical Section of the Paper Makers Association of Great Britain & Ireland 24: 347-353. Winkler A & Patt R (1988) Herstellung von zellstoffen aus vier verschiedenen holzarten nach dem ASAM-verfahren. Holz Roh Werkst 46: 341-345. Winner SR, Goyal GC, Pye EK & Lora JH (1991a) Pulping of agriculture residues by the Alcell process. In: Nonwood plant fiber pulping, progress report 20, Tappi Press, Atlanta, p 171-175. Winner SR, Goyal GC, Pye EK & Lora JH (1991b) Pulping of agriculture residues by the Alcell process. 1991 Pulping Conference Proc., Orlando, FL, 1: 435-439. Winner SR, Minogue LA & Lora JH (1997) Alcell pulping of annual fibers. 9th International Symposium on Wood and Pulping Chemistry Poster Presentations, Montreal, Canada, p 120:1-4. Wojtech B, Steppich W, Freudenberger D & Riedel K (1986) Hoechst AG, assignee. Verfahren zur extraktion von carbonsauren aus verdunnten wassrigen losungen. DE patent 34.36.348.A1. Worster HE (1974) The present and future of alkaline pulping. Pulp Paper Canada 75(10): 45-50. Wright GF & Hibbert H (1937) Studies on lignin and related compunds, XXVI, the properties of spruce lignin extracted with formic acid. J Am Chem Soc 59(1): 125-130. Wu LCF, Lora JH & Edwardson CF (1989) Alcell lignin: a new adhesive for waferboard. 23rd International Particleboard/Composite Materials Symposium, Pullman, WA, p 281. Xiao X, Fan P & Liang WZ (1993) The alkali-ethanol pulping of eucalyptus cilriodora. Seventh International Symposium on Wood and Pulping Chemistry Proc., Beijing, PR China, 3: 173-176. Xu Q, Lu F, Li X, Luo R & Jiang F (1990) A study on separation of acetic acid, water and formic acid system - refinement of acetic acid by esterification-azeotropic distillation. Chem J Chinese Univ 11(11): 1236-1239. Yang ST, White SA & Hsu ST (1991) Extraction of carboxylic acids with tertiary and quaternary amines: effect of pH. Ind Eng Chem Res 30(6): 1335-1342. Yasuda S (1988) Behavior of lignin in organic acid pulping, II, reaction of phenylcoumarans and 1,2-diaryl-1,3-propanediols with acetic acid. J Wood Chem Tech 8(2): 155-164. Yasuda S & Ito N (1987) Bahavior of lignin in organic acid pulping I, reaction of arylglycerol- aryl ethers with acetic acid. Mokuzai Gakkaishi 33(9): 708-715. Yasuda S, Abe Y & Hirokaga Y (1991) Behavior of lignin in organic acid pulping, part III, additive effects of potassium and sodium halides on delignification. Holzforschung 45(Suppl.): 79-82. Yiannos PN & Secrist RB (1968) Scott Paper Company, assignee. Process for removing lignin from lignin-polysaccharides. CA patent 801.727. Yoon SH & Labosky P (1998) Ethanol-kraft pulping and papermaking properties of aspen and spruce, part II, delignification kinetics, activation thermodynamics, and pulping productivity. Tappi J 81(4): 145-151. Yoon SH, Labosky P & Blankenhorn PR (1997) Ethanol-kraft pulping and papermaking properties of aspen and spruce. Tappi J 80(1): 203-210. Young RA (1985) Organic acid pulping of wood reduces chemical costs and energy requirements. Paper Trade J 169(11): 50-51. Young RA (1989) Ester pulping: a status report. Tappi J 72(4): 195-200. Young J (1992a) Commercial Organocell process comes online at Kelheim mill. Pulp Paper 66(9): 99-102. Young RA (1992b) Acetic acid based pulping. 1992 Solvent Pulping Symposium Notes, Boston, MA, p 61-66. Young RA (1994) Comparison of the properties of chemical cellulose pulps. Cellulose 1(2): 107-130. Young RA (1998) Acetic acid-based pulping. In: Young RA & Akhtar M (eds) Environmentally Friendly Technologies for the Pulp and Paper Industry. John Wiley & Sons, New York, NY, p 133-156. Young RA & Achmadi S (1983) Efficient utilization of woody biomass: a cellulose- particleboard-synfuels model. In: Cote WA (ed.) Biomass utilization. New York, NY, p 567-584. Young RA & Achmadi S (1992) Graft pulping in organosolv systems. J Wood Chem Tech 12(4): 485-510. Young RA & Baierl KW (1985) Ester pulping of wood: a revolutionary process. Southern Pulp Paper 48(12): 15-17. Young RA, & Davis JL (1986) Organic acid pulping of wood, part II, acetic acid pulping of aspen. Holzforschung 40(2): 99-108. Young RA & Davis JL (1991) Molecular weight characterization of lignins from acetic acid-based pulping processes. 6th International Symposium on Wood and Pulping Chemistry Proc., Melbourne, Australia, 1: 463-466. Young RA, Davis JL, Wiesmann EB & Baierl KW (1985) Organic acid pulping of wood, part I, an overview of applications. International Symposium on Wood and Pulping Chemistry, Vancouver, BC, p 169-172. Young RA, Davis JL, Wiesmann EB & Young TR (1987a) Ester pulping of wood. In: Kennedy JF (ed) Wood and cellulosics: industrial utilization, biotechnology, structure and properties. Chichester, England, John Wiley & Sons, p 367-373. Young RA, Fredman T, Keith T & Nelson J (1987b) Pulping of wood with organic acids and esters. Fourth International Symposium on Wood and Pulping Chemistry, Paris, France, 2: 185-187. Young RA, Keith T, Nelson J & Fredman T (1987c) Ester pulping. Solvent Pulping - Promises & Problems, Appleton, WI, p 28-34. Yuan Z, Ni Y & Heiningen A (1998a) An improved peracetic acid bleaching process. Appita 51(5): 377-380. (Paperbase 00522261). Yuan Z, Ni Y & Heiningen A (1998b) Peracetic acid and hydrogen peroxide brightening of a softwood kraft pulp delignified by methanol-assisted ozonation. Paperi ja Puu 80(4): 310-314. Zargarian K, Aravamuthan R & April GC (1988) Organosolv delignification of southern pine - an alternative pulping process. Chem Eng Tech 11(3): 195-199. Zarubin MY, Dejneko IP, Evtuguine DV & Robert A (1989) Delignification by oxygen in acetone-water media. Tappi J 72(11): 163-168. Zhang X, Ni Y & Heiningen ARP (1996) Basic engineering design data for ozone/solvent bleaching of Alcell-derived pulp. 1996 International Pulp Bleaching Conference Proc., p 601-612. (ABIPST 66:19418(AC)). Zhang XZ, Francis RC, Dutton DB & Hill RT (1998) The role of transition metal species in delignification with distilled peracetic acid. J Wood Chem Technol 18(3): 253-266. (Paperbase 00522184). Zheleznov A, Windmoller D, Korner S & Boddeker KW (1998) Dialytic transport of carboxylic acids through an anion exchange membrane. J Membrane Sci 139: 137-143. Ziegenfuss H & Maurer G (1994) Distribution of acetic acid between water and organic solutions of tri-n-octylamine. Fluid Phase Equilibria 102: 211-255. Zilbergleit MA & Glushko TV (1991) Products from the polymerization of furfural and hydroxymethyl furfural in acetic acid. Khim Drev (Riga) (1): 66-68. (ABIPST 62:13800). Zilbergleit MA, Reznikov VM & Yukhnovich ZK (1981) Belorussian Technological Institute, assignee. Pulp semifinished product. USSR patent 821.621. (CA 95:45056a). Zilbergleit MA, Korneichik TV & Reznikov VM (1987a) Wood delignification by aqueous acetic acid solutions, (6), physical properties of acetic acid lignins. Khim Drev (Riga) (6): 21-27. (ABIPC 59:228). Zilbergleit MA, Simkhovich BS & Reznikov VM (1987b) Wood delignification by aqueous acetic acid solutions, (7), kinetics of the process. Khim Drev (Riga) (6): 28-34. (ABIPC 59:489). Zilbergleit MA, Korneichik TV, Revyako MM, Lisova VS & Dolinskaya RM (1989a) Wood delignification by aqueous acetic acid solutions, 11., use properties of acetic acid lignins. Khim Drev (Riga) (6): 43-48. (ABIPST 62:4081). Zilbergleit MA, Simkhovich BS, Reznikov VM & Kirov SM (1989b) Delignification of hardwoods with aqueous solutions of acetic acid in the vapor phase. Bum Prom (12): 17- 18. (ABIPST 61:8279). Zilbergleit MA, Simkhovich BS, Korneichik TV, Beigelman AV, Burov AV, Galov LI & Mikhailov AI (1991a) Organosolv delignification of irradiated wood. 6th International Symposium on Wood and Pulping Chemistry Proc., Melbourne, Australia, 2: 219-223. Zilbergleit MA, Korneichik TV & Reznikov VM (1991b) Wood delignification by aqueous acetic acid solutions, (12), conversion of lignin macromolecules under acetic acid pulping conditions. Khim Drev (Riga) (2): 17-22. (ABIPST 63:314). Zilbergleit MA, Simkhovich BS, Korneichik TV, Kushner MA, Karalchuk TV & Glushko TV (1991c) Optimizing hardwood pulping with aqueous solutions of acetic acid. Bum Prom (12): 4. (ABIPST 64:6171). Zimmermann W & Tajana E (1996) Hydrolysis of kraft and organosolv pulps by bacterial and fungal hemicellulases. In: Srebotnik E & Messner K (eds) Biotechnology in pulp and paper industry. Facultas-Universitatsverlag, Vienna, p 135-138. Zimmermann M, Patt R, Kordsachia O & Hunter WD (1991) ASAM pulping of Douglas-fir followed by a chlorine-free bleaching sequence. Tappi J 74(11): 129-134. Zitzelsberger W (1990) Production of lignin from the Organocell pulping process. 1st European Workshop on Lignocellulosics and Pulp, Bergedorf, FR Germany, p 15-21. Zomers FHA, Tjeerdsma BF & Gosselink RJA (1993) Organosolv pulping and test paper characterization of fiber hemp. 1993 Pulping Conference Proc., Atlanta, GA, 2: 475- 484. Zomers FHA, Gosselink RJA, Dam JEG & Tjeerdsma BF (1995) Organosolv pulping and test paper characterization of fiber hemp. Tappi J 78(5): 149-155. Zule J, Likon M, Oblak-Rainer M, Moze A & Perdih A (1995) The characteristics of lignin from haloacetic acid pulping. The 8th International Symposium on Wood and Pulping Chemistry Proc., Helsinki, Finland, 2: 249-254.