选育耐受复合抑制剂酿酒酵母提高乙醇产量

能够耐受纤维素预处理中抑制剂的酿酒酵母对高效、经济生产纤维素乙醇至关重要。利用诱变结合驯化工程选育了一株可耐受复合抑制剂(1.3g/L糠醛、5.3g/L乙酸及1.0g/L苯酚)的工业酿酒酵母YYJ003。在pH 4.0的含有抑制剂的培养基中,耐受菌株乙醇产率是原始菌株的7.8倍,糠醛转化速率提高了5倍。在pH 5.5的复合抑制剂条件下,YYJ003发酵时间(16h)比野生菌株发酵时间(22h)缩短6h。在pH 4.0的未脱毒的玉米秸秆水热法预处理水解液中YYJ003的乙醇产率达到0.50g/g(乙醇/葡萄糖),乙醇产速达到4.16g/(L·h),而对照菌株无乙醇产出。     

A strain of <Emphasis Type="Italic">Meyerozyma guilliermondii</Emphasis> isolated from sugarcane juice is able to grow and ferment pentoses in synthetic and bagasse hydrolysate media

The search for new microbial strains that are able to withstand inhibitors released from hemicellulosic hydrolysis and are also still able to convert sugars in ethanol/xylitol is highly desirable. A yeast strain isolated from sugarcane juice and identified as Meyerozyma guilliermondii was evaluated for the ability to grow and ferment pentoses in synthetic media and in sugarcane bagasse hydrolysate. The yeast grew in xylose, arabinose and glucose at the same rate at an initial medium pH of 5.5. At pH 4.5, the yeast grew more slowly in arabinose. There was no sugar exhaustion within 60 h. At higher xylose concentrations with a higher initial cell concentration, sugar was exhausted within 96 h at pH 4.5. An increase of 350 % in biomass was obtained in detoxified hydrolysates, whereas supplementation with 3 g/L yeast extract increased biomass production by approximately 40 %. Ethanol and xylitol were produced more significantly in supplemented hydrolysates regardless of detoxification. Xylose consumption was enhanced in supplemented hydrolysates and arabinose was consumed only when xylose and glucose were no longer available. Supplementation had a greater impact on ethanol yield and productivity than detoxification; however, the product yields obtained in the present study are still much lower when compared to other yeast species in bagasse hydrolysate. By the other hand, the fermentation of both xylose and arabinose and capability of withstanding inhibitors are important characteristics of the strain assayed.     

Strain improvement of thermotolerant Saccharomyces cerevisiae VS strain for better utilization of lignocellulosic substrates

AIM: Pentose-utilizing yeast development by protoplast fusion and sequential mutations and ethanol fermentation using lignocellulosic substrate. METHODS AND RESULTS: Protoplasts of thermotolerant Saccharomyces cerevisiae and mesophilic, xylose-utilizing Candida shehatae were fused by electrofusion. The fusants were selected based on their growth at 42 degrees C and ability to utilize xylose. The selected best fusant was mutated sequentially and 3 mutant fusants obtained were tested for their stability. The mutant fusant CP11 was found to be stable and used for lignocellulosic fermentation. The Prosopis juliflora wood material was hydrolysed with 1% sulphuric acid initially for 18 h at room temperature and then for 20 min at 121 degrees C. The acid hydrolysate was separated and used for detoxification by ethyl acetate and overliming. The hard cellulosic fraction was hydrolysed with Aspergillus niger crude cellulase enzyme for 18 h at 50 degrees C. The substrate (15% w/v) yielded 84 g l(-1) sugars, representing 80% (w/w) hydrolysis of carbohydrate content present in the lignocellulosic material. The acid and enzyme hydrolysates were then equally mixed and used for fermentation with the developed fusant yeast (CP11). The fusant yeast gave an ethanol yield of 0.459 +/- 0.012 g g(-1), productivity of 0.67 +/- 0.015 g l(-1) h(-1) and fermentation efficiency of 90%. CONCLUSIONS: Protoplast fusion followed by sequential mutations method gave a stable and good performing fusant with maximum utilization of reducing sugars in the media. SIGNIFICANCE AND IMPACT OF THE STUDY: This new method could be applied to develop fusants for better biotechnological applications.     

Bioconversion of water-hyacinth (Eichhornia crassipes) hemicellulose acid hydrolysate to motor fuel ethanol by xylose-fermenting yeast

Water-hyacinth (Eichhornia crassipes) hemicellulose acid hydrolysate has been utilized as a substrate for ethanol production using Pichia stipitis NRRL Y-7124. Hydrolysate fermentability was considerable improved by boiling, and overliming up to pH 10.0 with solid Ca(OH)(2) in combination with sodium sulfite. The percent total sugar utilized and ethanol yield (Y(p/s)) for the untreated hydrolysate were 20.15+/-0.17% and 0.19+/-0.003 g(p) g(s)(-1), respectively, compared with 76.0+/-0.32% and 0.35 g(p) g(s)(-1), respectively for the treated material. The fermentation was very effective at an aeration rate of 0.02 v/v/m, temperature 30+/-0.2 degrees C and pH 6.0+/-0.2. However, the volumetric productivity (Q(p)) was still considerably less than observed in a simulated synthetic hydrolysate medium with a sugar composition similar to the hemicellulose acid hydrolysate. L-Arabinose was not fermented but assimilated. The presence of acetic acid in the hydrolysate decreased the ethanol yield and productivity considerably.     

Ethanol production from eucalyptus wood hemicellulose hydrolysate by Pichia stipitis

Ethanol production was evaluated from eucalyptus wood hemicellulose acid hydrolysate using Pichia stipitis NRRL Y-7124. An initial lag phase characterized by flocculation and viability loss of the yeast inoculated was observed. Subsequently, cell regrowth occurred with sequential consumption of sugars and production of ethanol. Polyol formation was detected. Acetic acid present in the hydrolysate was an important inhibitor of the fermentation, reducing the rate and the yield. Its toxic effect was due essentially to its undissociated form. The fermentation was more effective at an oxygen transfer rate between 1.2 and 2.4 mmol/L h and an initial pH of 6.5. The hydrolysate used in the experiences had the following composition (expressed in grams per liter): xylose 30, arabinose 2.8, glucose 1.5, galactose 3.7, mannose 1.0, cellobiose 0.5, acetic acid 10, glucuronic acid 1.5, and galacturonic acid 1.0. The best values obtained were maximum ethanol concentration 12.6 g/L, fermentation time 75 h, fermentable sugar consumption 99% ethanol yield 0.35 g/g sugars consumed, and volumetric ethanol productivity 4 g/L day. (c) 1992 John Wiley & Sons, Inc.     

Evaluation of fermentative potential of <Emphasis Type="Italic">Kluyveromyces marxianus</Emphasis> ATCC 36907 in cellulosic and hemicellulosic sugarcane bagasse hydrolysates on xylitol and ethanol production

This paper evaluates the fermentative potential of Kluyveromyces marxianus grown in sugarcane bagasse cellulosic and hemicellulosic hydrolysates obtained by acid hydrolysis. Ethanol was obtained from a single glucose fermentation product, whereas xylose assimilation resulted in xylitol as the main product and ethanol as a by-product derived from the metabolism of this pentose. Fermentation performed in a simulated hydrolysate medium with a glucose concentration similar to that of the hydrolysate resulted in ethanol productivity (Qp?=?0.86 g L^?1 h^?1) that was tenfold higher than the one observed in the cellulosic hydrolysate. However, the use of hemicellulosic hydrolysate favored xylose assimilation in comparison with simulated medium with xylose and glucose concentrations similar to those found in this hydrolysate, without toxic compounds such as acetic acid and phenols. Under this condition, xylitol yield was 53.8 % higher in relation to simulated medium. Thus, the total removal of toxic compounds from the hydrolysate is not necessary to obtain bioproducts from lignocellulosic hydrolysates.     

Sugar recovery and fermentability of hemicellulose hydrolysates from steam-exploded softwoods containing bark

The hemicellulose sugar recovery and ethanol production obtained from SO2-catalyzed steam explosion of a mixed white fir (70%) and ponderosa pine (30%) feedstock containing bark (9% dry weight/dry weight) was assessed. More than 90% of the available hemicellulose sugars could be recovered in the hydrolysate obtained after steam explosion at 195 degrees C, 2.38 min, and 3.91% SO2, with 59% of the original hemicellulose sugars detected in a monomeric form. Despite this high sugar recovery, this hydrolysate showed low ethanol yield (64% of theoretical yield) when fermented with a spent sulfite liquor-adapted strain of Saccharomyces cerevisiae. In contrast, most hydrolysates prepared at higher steam explosion severity showed comparable or higher ethanol yields. Furthermore, the hydrolysates prepared from bark-free feedstock showed better fermentability (87% of theoretical yield) despite containing higher concentration of known inhibitors. The ethanol yield from the hydrolysate prepared from a bark-containing wood sample could be improved to 81% by an extra stage acid hydrolysis (121 degrees C for 1 h in 3% sulfuric acid). This extra stage acid hydrolysis and steam explosion at higher severity conditions seem to improve the fermentability of the hydrolysates by transforming certain inhibitory compounds present in the hydrolysates prepared from the bark-containing feedstock and thus lowering their inhibitory effect on the yeast used for the ethanol fermentation.     

Fermentation of lignocellulosic sugars to acetic acid by <Emphasis Type="Italic">Moorella thermoacetica</Emphasis>

A systematic study of bioconversion of lignocellulosic sugars to acetic acid by Moorella thermoacetica (strain ATCC 39073) was conducted. Four different water-soluble fractions (hydrolysates) obtained after steam pretreatment of lignocellulosic biomass were selected and fermented to acetic acid in batch fermentations. M. thermoacetica can effectively ferment xylose and glucose in hydrolysates from wheat straw, forest residues, switchgrass, and sugarcane straw to acetic acid. Xylose and glucose were completely utilized, with xylose being consumed first. M. thermoacetica consumed up to 62 % of arabinose, 49 % galactose and 66 % of mannose within 72 h of fermentation in the mixture of lignocellulosic sugars. The highest acetic acid yield was obtained from sugarcane straw hydrolysate, with 71 % of theoretical yield based on total sugars (17 g/L acetic acid from 24 g/L total sugars). The lowest acetic acid yield was observed in forest residues hydrolysate, with 39 % of theoretical yield based on total sugars (18 g/L acetic acid from 49 g/L total sugars). Process derived compounds from steam explosion pretreatment, including 5-hydroxymethylfurfural (0.4 g/L), furfural (0.1 g/L) and total phenolics (3 g/L), did not inhibit microbial growth and acetic acid production yield. This research identified two major factors that adversely affected acetic acid yield in all hydrolysates, especially in forest residues: (i) glucose to xylose ratio and (ii) incomplete consumption of arabinose, galactose and mannose. For efficient bioconversion of lignocellulosic sugars to acetic acid, it is imperative to have an appropriate balance of sugars in a hydrolysate. Hence, the choice of lignocellulosic biomass and steam pretreatment design are fundamental steps for the industrial application of this process.     

Ethanol production from wheat straw hemicellulose hydrolysate by Pichia stipitis

Ethanol production was evaluated from wheat straw (WS) hemicellulose acid hydrolysate using an adapted and parent strain of Pichia stipitis. NRRL Y-7124. The treatment by boiling and overliming with Ca(OH)(2) significantly improved the fermentability of the hydrolysate. Ethanol yield (Yp/s) and productivity (Qp av) were increased 2.4+/-0.10 and 5.7+/-0.24 folds, respectively, compared to neutralized hydrolysate. Adaptation of the yeast to the hydrolysate resulted further improvement in yield and productivity. The maximum yield was 0.41+/-0.01 g(p) g(s)(-1), equivalent to 80.4+/-0.55% theoretical conversion efficiency. Acetic acid, furfurals and lignins present in the hydrolysate were inhibitory to microbial growth and ethanol production. The addition of these inhibitory components individually or in various combinations at a concentrations similar to that found in hydrolysate to simulated medium resulted a reduction in ethanol yield (Yp/s) and productivity (Qp av). The hydrolysate used had the following composition (expressed in g x l(-1)): xylose 12.8+/-0.25; glucose 1.7+/-0.3; arabinose 2.6+/-0.21 and acetic acid 2.7+/-0.33.     

Cofactor Dependence in Furan Reduction by Saccharomyces cerevisiae in Fermentation of Acid-Hydrolyzed Lignocellulose

A decreased fermentation rate due to inhibition is a significant problem for economic conversion of acid-pretreated lignocellulose hydrolysates to ethanol, since the inhibition gives rise to a requirement for separate detoxification steps. Together with acetic acid, the sugar degradation products furfural and 5-hydroxymethyl furfural are the inhibiting compounds found at the highest concentrations in hydrolysates. These aldehydes have been shown to affect both the specific growth rate and the rate of fermentation by yeast. Two strains of Saccharomyces cerevisiae with different abilities to ferment inhibiting hydrolysates were evaluated in fermentations of a dilute acid hydrolysate from spruce, and the reducing activities for furfural and 5-hydroxymethyl furfural were determined. Crude cell extracts of a hydrolysate-tolerant strain (TMB3000) converted both furfural and 5-hydroxymethyl furfural to the corresponding alcohol at a rate that was severalfold higher than the rate observed for cell extracts of a less tolerant strain (CBS 8066), thereby confirming that there is a correlation between the fermentation rate in a lignocellulosic hydrolysate and the bioconversion capacity of a strain. The in vitro NADH-dependent furfural reduction capacity of TMB3000 was three times higher than that of CBS 8066 (1,200 mU/mg protein and 370 mU/mg protein, respectively) in fed-batch experiments. Furthermore, the inhibitor-tolerant strain TMB3000 displayed a previously unknown NADH-dependent reducing activity for 5-hydroxymethyl furfural (400 mU/mg protein during fed-batch fermentation of hydrolysates). No corresponding activity was found in strain CBS 8066 (<2 mU/mg). The ability to reduce 5-hydroxymethyl furfural is an important characteristic for the development of yeast strains with increased tolerance to lignocellulosic hydrolysates.     

Cofactor dependence in furan reduction by Saccharomyces cerevisiae in fermentation of acid-hydrolyzed lignocellulose

A decreased fermentation rate due to inhibition is a significant problem for economic conversion of acid-pretreated lignocellulose hydrolysates to ethanol, since the inhibition gives rise to a requirement for separate detoxification steps. Together with acetic acid, the sugar degradation products furfural and 5-hydroxymethyl furfural are the inhibiting compounds found at the highest concentrations in hydrolysates. These aldehydes have been shown to affect both the specific growth rate and the rate of fermentation by yeast. Two strains of Saccharomyces cerevisiae with different abilities to ferment inhibiting hydrolysates were evaluated in fermentations of a dilute acid hydrolysate from spruce, and the reducing activities for furfural and 5-hydroxymethyl furfural were determined. Crude cell extracts of a hydrolysate-tolerant strain (TMB3000) converted both furfural and 5-hydroxymethyl furfural to the corresponding alcohol at a rate that was severalfold higher than the rate observed for cell extracts of a less tolerant strain (CBS 8066), thereby confirming that there is a correlation between the fermentation rate in a lignocellulosic hydrolysate and the bioconversion capacity of a strain. The in vitro NADH-dependent furfural reduction capacity of TMB3000 was three times higher than that of CBS 8066 (1,200 mU/mg protein and 370 mU/mg protein, respectively) in fed-batch experiments. Furthermore, the inhibitor-tolerant strain TMB3000 displayed a previously unknown NADH-dependent reducing activity for 5-hydroxymethyl furfural (400 mU/mg protein during fed-batch fermentation of hydrolysates). No corresponding activity was found in strain CBS 8066 (<2 mU/mg). The ability to reduce 5-hydroxymethyl furfural is an important characteristic for the development of yeast strains with increased tolerance to lignocellulosic hydrolysates.     

Potential inhibitors from wet oxidation of wheat straw and their effect on ethanol production of Saccharomyces cerevisiae: wet oxidation and fermentation by yeast

Alkaline wet oxidation (WO) (using water, 6.5 g/L sodium carbonate and 12 bar oxygen at 195 degrees C) was used as pretreatment method for wheat straw (60 g/L), resulting in a hydrolysate and a cellulosic solid fraction. The hydrolysate consisted of soluble hemicellulose (8 g/L), low-molecular-weight carboxylic acids (3.9 g/L), phenols (0.27 g/L = 1.7 mM) and 2-furoic acid (0.007 g/L). The wet oxidized wheat straw hydrolysate caused no inhibition of ethanol production by Saccharomyces cerevisiae ATCC 96581. Nine phenols and 2-furoic acid, identified to be present in the hydrolysate, were each tested in concentrations of 50-100 times the concentration found in the hydrolysate for their effect on fermentation by yeast. At these high concentrations (10 mM), 4-hydroxybenzaldehyde, vanillin, 4-hydroxyacetophenone and acetovanillone caused a 53-67% decrease in the volumetric ethanol productivity in S. cerevisiae compared to controls with an ethanol productivity of 3.8 g/L. The phenol acids (4-hydroxy, vanillic and syringic acid), 2-furoic acid, syringaldehyde and acetosyringone were less inhibitory, causing a 5-16% decrease in ethanol productivity. By adding the same aromatic compounds to hydrolysate (10 mM), it was shown that syringaldehyde and acetovanillone interacted negatively with hydrolysate components on the ethanol productivity. Fermentation in WO hydrolysate, that had been concentrated 6 times by freeze-drying, lasted 4 hours longer than in regular hydrolysate; however, the ethanol yield was the same. The longer fermentation time could not be explained by an inhibitory action of phenols alone, but was more likely caused by inhibitory interactions of phenols with carboxylic acids, such as acetic and formic acid.     

Fermentation of newsprint prehydrolysate by an ethanologenic recombinant Escherichia coli

Summary Recombinant E. coli B (pLOI297) produced ethanol from a nutrient-supplemented, newsprint prehydrolysate medium, at about a 20% reduction in both yield and productivity compared to a synthetic softwood hemicellulose hydrolysate medium (lacking acetic acid). With pH controlled at 7, the sugar-to-ethanol conversion efficiency with the newsprint prehydrolysate was 74.5% of theoretical maximum. The final ethanol concentration was 14.6 g/L. Reduced ethanol yield was due to by-product formation, principally lactic acid. The specific rates of glucose, mannose and xylose utilization in the synthetic medium were 0.73, 0.42 and 0.22 g/g cell/h respectively. The ethanol yield from the pretreatment processing of newsprint is estimated at 85L per dry metric ton.     

Preparation of hydrolysates from tobacco stalks and ethanolic fermentation by Saccharomyces cerevisiae

Chipped tobacco stalks were subjected to steam pretreatment at 205 °C for either 5 or 10 min before enzymatic hydrolysis. Glucose (15.4–17.1 g/l) and xylose (4.5–5.0 g/l) were the most abundant monosaccharides in the hydrolysates. Mannose, galactose and arabinose were also detected. The hydrolysate produced by pretreatment for 10 min contained higher levels of all sugars than the 5 min-pretreated hydrolysate. The amounts of inhibitory compounds found in the hydrolysates were relatively low and increased with increasing pretreatment time. The hydrolysates were fermented with baker's yeast. Ethanol yield, maximum volumetric productivity and specific productivity were used as criteria of fermentability of the hydrolysates. The fermentation of the hydrolysates was only slightly inhibited compared to reference solutions having a similar composition of fermentable sugars. The ethanol yield in the hydrolysates was 0.38–0.39 g/g of initial fermentable sugars, whereas it was 0.42 g/g in the reference. The biomass yield was twofold lower in the hydrolysates than in the reference. The fermentation inhibition caused by the tobacco stalk hydrolysates was less than that caused by sugarcane bagasse hydrolysates obtained under the same hydrolysis conditions.     

Critical factors affecting ethanol production by immobilized Pichia stipitis using corn cob hemicellulosic hydrolysate

Fermentation of xylose from hydrolysate of acid-treated corn cob by Pichia stipitis is inhibited by acetic acid and lignin derivatives. In the present study, we have designed and implemented an immobilized cell culture for xylose to ethanol conversion from acid-treated corn cob hydrolysate without the removal of fermentation inhibitors. In this study, cultivations of suspended and immobilized Pichia were compared in terms of ethanol yield and productivity to investigate whether the cell immobilization could improve resistance to inhibitors. Cell immobilization clearly favored the fermentative metabolism in nondetoxified corn cob hydrolysate leading to an improvement of twofold ethanol productivity as compared to that achieved with suspension culture. Calcium alginate as an immobilization matrix was selected to immobilize Pichia cells. Concentrations of sodium alginate, calcium chloride, and fermentor agitation speed were optimized for ethanol production using statistical method. Statistical analysis showed that agitation speed had maximum influence on ethanol production by immobilized Pichia cells. In comparison to suspension culture, immobilization had a positive impact on the fermentative metabolism of Pichia, improving the ethanol yield from 0.40 to 0.43?g/g and productivity from 0.31 to 0.51?g/L/h for acid-treated corn cob hydrolysate.     

Ethanol recovery from corn fiber hydrolysate fermentations by pervaporation

Corn fiber, a byproduct of corn wet milling, is an attractive feedstock for biomass ethanol production. Corn fiber was hydrolyzed by dilute sulfuric acid and neutralized by one of two methods: conventional lime treatment or neutralization by strongly basic anion exchange. The anion exchange neutralized (AEN) hydrolysate contained substantially lower levels of the inhibiting compounds furfural, 5-hydroxymethylfurfural, and acetic acid compared to the lime neutralized hydrolysate. In batch fermentations the ethanol yields and final ethanol concentration of the two hydrolysates were similar at 0.32-0.43 g/g and 29-44 g/l, respectively. Sugar consumption in the AEN fermentations was superior. Coupling of a membrane pervaporation unit to a fed-batch fermentation of AEN hydrolysate maintained the ethanol concentration below 25 g/l with complete sugar utilization for approximately 5 days. A concentrated ethanol stream of 17 wt.% ethanol was produced by the pervaporation unit.     

Eucalyptus hydrolysate detoxification with activated charcoal adsorption or ion-exchange resins for xylitol production

Eucalyptus hemicellulosic hydrolysate used for xylitol production by Candida guilliermondii FTI20037 was previously treated either with ion-exchange resins or with activated charcoal adsorption combined with pH adjustment, in order that acetic acid, furfural and hydroxymethylfurfural could be removed. The best results for xylitol yield factor (0.76 g/g) and volumetric productivity (0.68 g/(l h) were attained when a three-fold concentrated hydrolysate was treated with ion-exchange resins. Using activated charcoal combined with pH adjustment for treating a three-fold concentrated hydrolysate resulted in a xylitol yield factor of 0.40 g/g and a volumetric productivity of 0.30 g/(l h). This same treatment applied to a six-fold concentrated hydrolysate resulted in a xylitol yield factor of 0.66 g/g and a volumetric productivity of 0.50 g/(l h).     

Comparative fermentability of enzymatic and acid hydrolysates of steam-pretreated aspenwood hemicellulose by Pichia stipitis CBS 5776

Summary Enzymatic hydrolysates of hemicellulose from steam-pretreated aspenwood were more fermentable than the acid hydrolysate after rotoevaporation or ethyl acetate extraction treatments to remove acetic acid and sugar- and lignin-degradation products prior to fermentation by Pichia stipitis CBS 5776. Total xylose and xylobiose utilization from 5.0% (w/v) ethyl acetate extracted enzymatic hydrolysate was observed with an ethanol yield of 0.47 g ethanol/g total available substrate and an ethanol production rate of 0.20 g·l^-1 per hour in 72 h batch fermentation.     

Effect of inhibitors formed during wheat straw pretreatment on ethanol fermentation by Pichia stipitis

The inhibitory effect of the main inhibitors (acetic acid, furfural and 5-hydroxymethylfurfural) formed during steam explosion of wheat straw was studied through ethanol fermentations of model substrates and hydrolysates from wheat straw by Pichia stipitis. Experimental results showed that an increase in acetic acid concentration led to a reduction in ethanol productivity and complete inhibition was observed at 3.5 g/L. Furfural produced a delay on sugar consumption rates with increasing concentration and HMF did not exert a significant effect. Fermentations of the whole slurry from steam exploded wheat straw were completely inhibited by a synergistic effect due to the presence of 1.5 g/L acetic acid, 0.15 g/L furfural and 0.05 g/L HMF together with solid fraction. When using only the solid fraction from steam explosion, hydrolysates presented 0.5 g/L of acetic acid, whose fermentations have submitted promising results, providing an ethanol yield of 0.45 g ethanol/g sugars and the final ethanol concentration reached was 12.2 g/L (10.9 g ethanol/100 g DM).     

Ethanol production from glucose byClostridium thermosaccharolyticum strains: Effect of pH and temperature

Summary The fermentation of glucose byClostridium thermosaccharolyticum strains IMG 2811^T, 6544 and 6564 was studied in batch culture in a complex medium at different temperatures in defined and free-floating pH conditions. All the strains ferment 5 g glucose.l^–1 completely. The yield of the fermentation products turned out to be independent of the incubation temperature for strain IMG 2811^T. Strain IMG 6544 produced at 60°C significantly more ethanol and less acetic acid, butyric acid, hydrogen gas and biomass than at lower temperatures. With strain IMG 6564, the opposite effect occurred: ethanol appeared to be the main fermentation product at 45°C; at 60°C less ethanol and more acetic acid, butyric acid and hydrogen gas was formed.Experiments, carried out with strain IMG 6564, at defined pH conditions (between 5.5 and 7) and different temperatures (45, 55 and 60°C) revealed no effect of the incubation temperature, but an important effect of the pH on the product formation. At pH 7, ethanol was the main fermentation product while minor amounts of hydrogen gas, acetic and butyric acid were produced. Lowering the pH gradually to 5.5 resulted in a decrease of ethanol and an increase of biomass, hydrogen gas, acetic, butyric and lactic acids. At pH higher than 7 no growth occurred. Similar conclusions could be drawn for strains IMG 2811^T and 6544.