EP0056409B1

EP0056409B1 - Pulping of lignocellulose with aqueous alcohol and alkaline earth metal salt catalyst

Info

Publication number: EP0056409B1
Application number: EP19810902305
Authority: EP
Inventors: Laszlo Paszner; Pei-Ching Chang
Original assignee: Bau und Forschungsgesellschaft Thermoform AG
Current assignee: Bau und Forschungsgesellschaft Thermoform AG
Priority date: 1980-07-25
Filing date: 1981-07-24
Publication date: 1990-06-27
Also published as: FI71781C; BR8108706A; CA1150012A; JPH0329917B2; FI820927L; WO1982000483A1; DE3177195D1; JPS57501239A; EP0056409A1; FI71781B; RU1830091C

Abstract

A high yield of a high quality pulp is obtained by cooking fragmented lignocellulose material at a temperature of 180 to 240`C with a mixture of methanol or ethanol and water containing 80 to 98 percent by volume of the alcohol, the mixture containing dissolved from 0.001 to 0.5 molar of an alkaline earth metal salt, and from zero to 0.005 normal/molar of (a) a strong mineral acid, or (b) a weak mineral acid or (c) a weak organic acid or (d) an acid reacting metal salt to aid rapid delignification. In addition, optionally, pressures substantially higher than developed in enclosed spaces by the vapors of the solvent mixture at the cooking temperature are used to further increase the delignification rate and suppressing carbohydrate degradation. Lignin is obtained in powder form by low temperature evaporation of the alcohol from the spent cooking liquor.

Description

The invention relates to a method for the digestion of lignocellulose material to recover a fibrous cellulose by dissolving the lignin and hemicellulose constituents of the lignocellulose material in an aqueous methyl or ethyl alcohol containing dissolved a chloride or nitrate of magnesium, calcium or barium or magnesium sulfate or a mixture thereof in an amount of from 0.001 to 1.0 molar, for not more than two hours at elevated temperatures above 130°C and pressures, the amount of cooking liquor being at least 4 parts by weight per part by weight of the lignocellulose material, and recovery of the cellulose which has remained undissolved.
Such a process is described in German Offenlegungsschrift No. 29 20 731. The main advantages of this known process over the conventional organosolv pulping processes which use mixtures of alcohol and water as the cooking liquor without the addition of an alkaline earth metal salt are the much more rapid delignification rate and the recovery of the lignin in powder form rather than as a dark-brown quasi-molten phase which is not easy to remove from the equipment and is of low commercial value.
The problem to be solved by the invention is to improve delignification specificity for all lignocellulose species and thereby to increase the yield of cellulose pulp, to reduce viscosity losses in cellulose due to carbohydrate degradation, improve the quality and stability of dissolved lignin in solution and increase the uniformity of digestion at reduced cooking times of air-dry softwood chips and increase temperature stability of lignocellulose materials during high temperature digestion.
These problems are solved by a method as stated at the beginning and further specified in the characterising portion of the sole claim.
Document DE-A-2920731, mentioned above, discloses alcohol:water cooking liquors ranging from "20:80" to "80:20" with "80:20" liquor being the highest exemplified. It is stated that a "70:30" (methanol) liquor is particularly effective (cf. sentence bridging pages 12 and 13). That process is carried out at autogenous pressure, i.e. in a sealed pressure vessel (cf. Example 1). Nothing in this document points towards an increase in the pressure.
The only document which describes the use of pressures above autogenous pressure in a process of the kind claimed is US―A―2022654, according to which the pressures can be up to 100 atm. or more above autogenous pressure, the maximum mentioned being 150 atm. The preferred range is however significantly lower. Aqueous solutions of organic solvents may be applied as cooking liquors; for instance mixtures with ethanol having a concentration of about 30% to 70% of said alcohol. In the relevant example 3 a "50:50" (ethyl alcohol) liquor is used for extraction at an overpressure of 40 atm, the reaction time being 6 hours; no concrete results are given in terms of pulp yield or viscosity values.
The preference for lower alcohol concentrations in the cooking liquor evinced by US-A-2022654 is also reflected in the other relevant literature. As early as 1931, Kleinert in his pioneer patent had discovered that alcohol-water mixtures with a water content of 20%-75% were useful for decomposing fibrous vegetable matter for the recovery of both the cellulose and the incrusting components. A "50:50" liquor was exemplified. The same document reported that the use of 96% alcohol produced only a dark brown residue (AT-A-124738). In his comprehensive article published in the same year he had shown that the percentage of wood digested steadily decreased with alcohol concentrations higher than 50% alcohol (cf. "Angew. Chemie," 11 Jg, 1931, page 789, left column, lines 3-5 and Abb. 1). A little later, in 1935, it was disclosed that cellulose could be produced from lignocellulose materials using 85%-95% aqueous alcohol. This elaborate process, including a predigestion step, was carried out at atmospheric pressure and involved extended cooking times of 4-6 hours; without, however, any specific yields being indicated (cf. US―A―2560638, column 3, lines 5-10, Claim 1 and examples 1-3). Finally, much more recently - i.e. in a review article on the subject appearing after the priority date of the application-in-suit - the use of a "50:50" liquor has been indicated to be optimal (cf. "Das Papier", 34 Jg., Heft 10A pages V9-V18, especially page V17, left column, second paragraph - published November 1980).
The prevailing opinion among experts even after a period of nearly 50 years was thus apparently still to run organosolv processes with cooking liquors of low alcohol content.
A closer analysis of documents US-A-2022654 and US―A―2560638 reveals, that the methods known therefrom can briefly stated, be characterised on the one hand by "high pressure/low alcohol content/one step digestion/extended reaction time" and on the other by "low pressure/high alcohol/multi- stage digestion/extended reaction time".
The skilled person considering in particular these divergent teachings would have had no incentive to select two specific parameters (high pressure, high alcohol content) and to integrate them into the existing system of DE-A-2920731, thus providing the "high pressure/high alcohol content/one step digestion/low reaction time" process claimed. With the modifications made it could not have been expected to arrive at an operable pulping process at all, let alone one with favourable results in the quantitative and/or qualitative respects.
This applies even at the crucial "80:20" liquor landmark, where the claimed subject-matter most closely approaches the disclosure of DE-A-2920731.
Therefore it was surprising that by the use of an aqueous methanol or ethanol of between 80 and 98 percent by volume of the alcohol in the presence of the alkaline earth metal salts and with the given pressure constraint much better results can be obtained than with the use of the lower range of alcohol contents of the cooking liquor and with pressures corresponding essentially to the autogenous pressures.
If in addition to the alkaline earth metal salts one or more acid reacting substances are preferentially added to the cooking liquor the effect of the additive is much greater than with the addition of either the alkaline earth metal salt or the acid reacting substances, so that the total amount of additives can be substantially reduced.
In the course of the digestion of the lignocellulose material organic acids such as formic or acetic acids are generated. These acids should be taken into account when defining the amount of acidic substance to be added to the cooking liquor, so that the pH of the reaction mixture be maintained preferably above 3.8 and below 7.0 regardless of the amount of added or autocatalytically generated acids. Such pH control is easily achieved by mild buffering of alkali metal ion systems such as occurs with technical grades of the alkali earth metal salts used in this invention or as afforded by standard buffering salts specified for this pH range. It is further a very important characteristic of such systems that the actual cooking pH changes only within relatively narrow limits between 3.8 to 5.6 or not at all depending on the wood species acidity and the degree of buffering produced by the alkali metal ions present or added to the cooking liquor.
In the process of the invention methanol is the preferred alcohol, but where methanol is not available in sufficient amount ethanol can be used as well. The alcohol content of the cooking liquor must be between 80 and 98 percent by volume, but the higher percentages within this range are relatively difficult to achieve because of the moisture contained in the lignocellulose starting material.
The preferred ratio of lignocellulose material to cooking liquor is 1:6 to 1:20 by weight.
At the high alcohol-water ratios and with an excess pressure of at least 5 bars over that developed by the vapors of the cooking liquor at the temperature used not only delignification is more complete, but carbohydrate degradation is suppressed.
The following table 1 shows the combinations of alkaline earth metal and acidic hydrolyzing substances which may be used in the process of the invention:

Example 1 (Comparative)

To investigate preliminarily the effectiveness of delignification specificity and yield of fibre attainable when using the novel largely methanol-water solvent extraction in the presence of alkali earth metal salts and auxiliary acid catalysts a number of cooks were carried out in laboratory-scale stainless steel pressure vessels having internal dimensions of 11 cm height and 4.5 cm diameter.
Wood chips in both air-dry and green conditions were conditioned to a uniform moisture content before the pulping trials. Batch quantities of commercial size chips were charged into the digester with ten times their weight of cooking liquor containing predetermined quantities of the salt catalysts. The volume ratio of methanol to water ranged between 90:10. The sealed stationary vessel was quickly brought to cooking temperature in a thermostatically controlled glycerine bath and the temperature held constant for the cooking interval required. The reported cooks are those which at the end of the stated period produced a free pulp when slurried in a disintegrator at slow stirred speed.
At the end of each cook the digester was rapidly cooled with cold water and the liquor decanted. After disintegration of the cooked chips in acetone or cooking solvent and final washing in water the pulp was air-dried to constant weight and yield, and Kappa number and TAPPI 0.5 per cent viscosity determined in an effort to characterize the pulp. The results are summarized in TABLE 2.
In TABLE 3 the effect of varying alcohol-water ratios and the compensating effect of increased temperature and prolonged cooking time are demonstrated on spruce wood. Pulping aspen and spruce wood at the high alcohol concentrations indicated in TABLE 3 shows that in the presence of 0.05 molar salt concentrations, with or without the secondary acid catalysts, free fibre separation is obtained within 15 to 60 min (including 11 minutes as heating-up time) and in spite of the relatively high Kappa number, fibre liberation was obtained at relatively high pulp yield. The pulps had viscosities between 20 to 52 m.Pa.s corresponding to a degree of polymerization of 1320 to 2200 (Rydholm, Pulping Processes, page 1120).
It can be observed that the system behaves according the laws of Arrhenius with respect to temperature in that cooking times to fibre separation and comparable Kappa number decrease with increasing temperature even for the unconventionally high cooking temperature of 220°C while viscosity is little or not at all affected even at this high temperature in the presence of 80 percent or higher alcohol volume in the cooking liquor the parameter most affected is residual lignin with only minor losses in pulp yield. An accelerated carbohydrate degradation is observed with alcohol:water ratios of 70:30 and lower, the carbohydrate stabilizing effect of high alcohol content of the digestion mixture is thus a truly surprising effect quite contrary to tendencies reported in the prior art. High alcohol content liquors further allow more thorough delignification within a given digestion time. The table indicates that a pulp yield in excess of 60 percent can be had from cooking spruce wood at a Kappa number of 45 and below a cooking time of 25 (minus 11) minutes.
The process also appears to be quite tolerant to extended cooking times wherein the parameter most affected is residual lignin content.
In a number of cooks (not reported) wherein the cooking time was not sufficient to allow fibre separation, the chips were found to be sufficiently soft so that a semi-mechanical pulp could be prepared on treatment at high speed in a blender. In certain of the cooks where "No fibre separation" (NFS) was reported earlier after a predetermined cooking time it was found that, on high-speed blending acceptable pulps could be produced. It is therefore to be understood that this invention is not limited to the length of cooking at which a free fibre state is reached but also includes cooks for only a sufficient time at which minimal delignification and hemicellulose removal took place to produce a semi-chemical pulp product of ultra high yields say about 80 to 90 percent by weight. Fully defiberized pulps can be had at 75 percent pulp yield.
The pulping liquor when subjected to vacuum distillation at low temperature yields a flocculated lignin precipitate. After recovery of the lignin by filtration or centrifuging a sugar solution having a solids concentration up to 25 per cent by weight is obtained. Charcoal filtration removes most of the yellow color due to the water soluble lignin depolymerisation products. The molecular weight distribution of the lignin shows one major and 2 to 3 minor peaks with the maximum being under 3800. Purification of the crude lignin is most effectively done by redissolution in acetone and spray drying in vacuum at low temperature to avoid melting and resinification. A dried solid filter cake is easily broken up into a free flowing tan- colored powder.
In conjunction with these tests summative carbohydrate analyses were also carried out for the original wood of spruce and aspen poplar and the pulps prepared therefrom. Findings of these investigations are summarized in TABLE 4. Sugar compositions of alpha-celluloses are those prepared from the pulps. The aspen pulp samples were found to be rich in xylan and spruce pulp samples rich in mannan with the other less important hemicelluloses being present in smaller amounts. Retention of these hemicelluloses explain the improvements in higher than usual yield had earlier with this process.
Analysis of the sugar wort showed (data not reported herewith) that the majority of dissolved sugars was present as monomers (about 30 to 50%) and the rest as low molecular-weight oligomers. Surprisingly no furfurals were detected in the residual liquors following the cooks done with the alkali earth metal salts as primary catalysts alone. In prior acid catalysed organosolv cooking degradation (dehydration) of the xylose and hexose sugars to furfurals is a simultaneous reaction with hydrolysis and delignification and was found to be prevalent at the higher temperatures (above 200°C). In solution these furfurals are very active and condense readily with the dislodged low molecular-weight lignin fragments to form alcohol insoluble products. The absence of furfurals in residual liquors of this invention assures complete solubility of the dissolved lignin and a high degree of sugar recovery as by-product. The sugar solutions are readily fermentable into ethanol, yeast and other fermentation products. The alkali earth metal catalysts do not interfere with such fermentation processes and can also be safely discharged in mill effluents.
Very similar results were obtained with other lignocellulose species. Sugarcane rind behaved like aspen poplar, jack pine, ponderosa pine and Western hemlock; and Douglas-fir behaved like spruce wood whereas birch and Eucalyptus species proved to be intermediate species and wheat straw was found to be a more difficult species than spruce requiring larger catalyst concentrations than spruce to yield pulps with equal degree of delignification. Numerous other secondary catalysts listed in TABLE 1 were also tested but their results not reported herein due to the large similarity in results obtainable on applying them. In these cases some adjustments in cooking conditions were necessary to compensate for the variation in acid strength.

Example 11 (present invention)

While the examples given before show a moderately satisfactory selectivity for delignification at thermodynamically defined conditions, allowing or causing an increase in internal pressures at least 5 bars higher than those normally found for enclosed liquids under free expansion conditions, or by deliberate application of pressure from a pressure intensifier or through compressed inert gases was found to influence the delignification and carbohydrate degradation rates at especially high alcohol water ratios and high temperatures by shifting the rate constant in a very favourable manner. In general it was observed, that in order to achieve the same degree of delignification at high alcohol water ratios especially over 85:15, higher temperatures were required. Thus the desired delignification rates could be maintained and cooking times could be held within reasonable limits. It was also found that as the system pressure increased so did the pulp viscosity, indicating the beneficial effects of pressure on delignification rates and on lower the sensitivity of the carbohydrates to increased thermal treatment which normally led to lower viscosities. It was also observed that the pressure effects were not linked to increased penetration into the wood matrix since when air-dry spruce chips were cooked with 90:10 or 95:5 alcohol:water solvent mixtures in the presence of 0.05 moles of CaCI₂ at 210°C under normal autogeneous pressure (35 atm and 39 atm, respectively) complete penetration of the chips is observed within the first 10 minutes of cooking yet poor fibre separation occurs even after prolonged cooking, up to 50 minutes. Under the same conditions but with added or internally generated overpressure fully cooked chips are obtained which show the same fibre liberation tendencies as chips cooked at lower alcohol concentration (under 80:20). While this in itself was a surprising effect, analysis of the resulting pulps showed a consistently higher pulp viscosity. In fact the pulp viscosity consistently increased with the level of pressure applied or generated. Some data on high pressure cooks are reproduced in TABLE 5. In comparison the previous test data provided in TABLE 5 wherein the increased selectivity of delignification and the lower carbohydrate degradation (higher pulp viscosity) and a significant reduction in cooking time with increased pressure is clearly evident. Thus the combined effect of high alcohol concentration and high pressure is an essential aspect of this invention in that it allows faster delignification of any wood species to low residual lignin content levels which earlier were not possible without considerable losses in cellulose viscosity. The pressure effect diminishes somewhat when solvent compositions lower than 80:20 alcohol:water content are used.
Table 5 demonstrates the advantages of alcohol concentration higher than 80% by volume and overpressure on delignification and carbohydrate retention at a cooking temperature of 210°C.
All cooks were done at a wood:liquor ratio of 1:10. Cooking times include 9 minutes for heating-up to temperature. In a similar series of cooks with 90:10 alcohol:water mixture, cooked at 210°C and 320 bar it was established that the ratio of lignin to carbohydrate removed can be as high as 9.48 on spruce wood and delignification could be pursued to a Kappa number of 14.5 at a residual pulp yield of 49% by weight. The viscosity dropped from an initial value of 55 mPa-s to 24 on cooking for 50 minutes under the above conditions. Thus the pulp properties generally increase with increased overpressure at the lower temperatures possible. Interestingly, the alpha-cellulose yield of the highly delignified pulp was still 43.2% by weight based on wood as 100, representing 88% of the total pulp mass. All pulps produced under these conditions were fully defiberized and produced no rejects on screening through a No. 6-cut standard laboratory flat pulp screen.

Claims

A method for the digestion of lignocellulose material to recover a fibrous cellulose by dissolving the lignin and hemicellulose constituents of the lignocellulose material in an aqueous methyl or ethyl alcohol containing dissolved a chloride or nitrate of magnesium, calcium or barium or magnesium sulfate or a mixture thereof in an amount of from 0.001 to 1.0 molar, for not more than two hours at elevated temperatures above 130°C and pressures, the amount of cooking liquor being at least 4 parts by weight per part by weight of the lignocellulose material, and recovery of the cellulose which has remained undissolved, characterized by the combined use of the following features:
1) the pressure during cooking is maintained at least 5 bars higher than that developed by the vapours of the cooking liquor at the cooking temperature; and

2) the aqueous methanol or ethanol contains between 80 and 98 percent by volume of the alcohol.