Skip to main content

Technological improvement for ethanol production from lignocellulose


The process concepts studied in the project aimed at increasing the ethanol concentration (by high consistency hydrolysis and fermentation technology) and yield (by using tolerant yeast strains. The specific objective was to develop integrated high temperature (HT) conversion technology based on improved thermophilic enzymes and enhanced accessibility of the substrates. The proposed concept for high temperature and consistency hydrolysis technology involved a first stage prehydrolysis at high temperature, followed by a complete hydrolysis at lower temperature. The high number of cloned thermophilic enzymes during the project allowed, however, a more flexible approach for different process concepts. Enzymes needed for both a prehydrolysis (liquefaction) stage and a complete hydrolysis of various pretreated raw materials were identified. The new optimised thermophilic enzyme mixtures were evaluated in high temperature hydrolysis experiments. Hydrolysis temperature could be increased with technical steam pre-treated raw materials by about 10 XC. Operation at high consistency was also demonstrated. The concept has been preliminarily proven in experiments leading to high concentration of ethanol (4V5 %). The project also aimed at reduced water use in fermentation (high consistency, higher concentration ethanol process) and focused at high hydrolysis and fermentation yield. The results showed that a separate high temperature stage for liquefaction of the material at the start of the procedure can be feasible. A decrease of viscosity can be achieved using only endoglucanases. A more beneficial result may, however, be obtained, if a complete set of thermophilic hydrolytic enzymes is used already at the prehydrolysis stage. It was thus possible to replace a present commercial cellulase mixture in the first 24h stage of a two-step process with an artificial three component mixture of selected cellulase proteins with a high hydrolysis yield and without increasing the overall protein dosage. Thus, a high temperature prehydrolysis stage on spruce substrate using only selected cloned components of cellulase system could be carried out with at least comparable efficiency of total hydrolysis than with the present commercial mixtures. For an efficient total hydrolysis, the requirements for the composition of enzyme mixtures depend strongly on the substrates and the process configuration selected. Thus for example, the amount of beta-glucosidase can be reduced in case SSF is used. The new optimised thermophilic enzyme mixtures were evaluated in high temperature hydrolysis experiments. Hydrolysis temperature could be increased with technical steam pre-treated raw materials by about 10 C when compared to the present state-of-art industrial enzyme products. Operation in high dry matter conditions was also demonstrated, although the reactor design and mixing should be further optimised for high viscose fibre suspensions. Clearly more efficient hydrolysis per assayed FPU unit or CBHI protein amount used was obtained. The results provide a promising basis to produce and formulate improved enzyme products. These products can have high temperature stability in process conditions in the range of 55-60 C (with present industrial products at 45-50 C) and clearly improved specific activity, essentially decreasing the protein dosage required for efficient hydrolysis of lignocellulosics.
Nedalco has developed methods to adapt and select yeast strains on very toxic hydrolysates. The methods can be applied to any given toxic hydrolysate. The method has been published on two international symposia on fuels and chemicals in the USA together with the Technical University of Budapest and a paper was submitted antitled "Fermentation inhibitors from pretreated lignocellulosic materials: Problems and Solutions". The method can be used for customizing existing strains to new raw materials in combination with severe pretreatment conditions. Furthermore, hydrolysis techniques have been developed on pretreated Corn Stover preparations as provided by ENEA after steam pretreatment. The whole process of sugar hydrolysis and fermentation was demonstrated of these hydrolysates. More than 20 yeast strains were tested at VTT and ranked by their ability to produce ethanol in the presence of toxic compounds as well as by their ability to grow on pretreated materials. The best ethanol producers and the most tolerant strains were the industrial strains and genetically modified industrial strains, whereas the strains with laboratory strain background were not as tolerant to the toxic compounds. The xylose utilising strains were able to consume all xylose and convert it further to ethanol in aerobic conditions but the rate was lower than on glucose. All main lignocellulose derived sugars except arabinose in pure form were consumed completely aerobically by the most potential strains. All sugars were consumed more slowly on toxic pretreated materials. The best VTT strain tested was further mutagenised in order to improve the xylose utilisation rate, to improve the tolerance for inhibitors and to increase the ethanol yield. After several mutagenesis trials mutant strains were isolated which could clearly consume xylose better in aerobic and anaerobic conditions compared to the non-mutagenised strain. The rate of ethanol production of the mutants strains was improved around two-fold compared to the parent strain. Best strains produced up to 50% more ethanol. In fermenter experiments, xylose utilisation rate was around 35 % and ethanol production rate around 25% better compared to the host strain.
Steam pretreatment of softwood in one- and two-step configurations with the addition of SO2 or H2SO4 was optimized. The highest yield of hemicellulose sugars (measured as mannose) was around 85%, which was slightly lower than the target 90%, but the yield of glucose, i.e cellulose, was slightly higher than 90%. The overall yield of ethanol after SSF of the pretreated material was 80% of the theoretical based on available fermentable sugars in the raw material, which corresponds very well with the target. The maximum overall yields obtained for corn stover glucan and xylan were 89 % and 78 %, respectively. Continuous steam pretreatment was used for the steam explosion of corn stover. Both the solubilized total matter and sugar concentrations after the steam pretreatment were higher than obtained by the batch treatment. Steam pretreatment of willow resulted in glucose yields exceeding 90 % and xylose yields higher than 80 % after enzymatic hydrolysis both with and without the addition of an acid catalyst. Two step wet oxidation aimed at extraction of the majority of hemicellulose fraction at mild conditions during the first step and improving the digestibility of the cellulose by the subsequent, second stage performed at more severe conditions. A two-stage pre-treatment process was generally shown to improve significantly the sugar yields, especially the hemicellulose yield of woody materials (softwood and willow). Improvement of sugar yield from willow by a two stage wet-oxidation pretreatment was generally demonstrated. The highest hydrolysis yield (as glucose) of all wet oxidation experiments on soft wood was 61%, corresponding to a cellulose recovery of 83%. The highest total hemicellulose yield achieved for willow with additives was 87% in single step. The highest glucose yield was 58%. Acidic conditions of wet oxidation resulted in higher glucose yields. The highest total glucose yield was 80%, which was significantly higher than after alkaline pretreatment. For corn stover, the highest overall hemicellulose yield was 78% of the original hemicellulose content. The enzymatic conversion of cellulose after wet oxidation was between 40 and 77%, as compared to 18% conversion for the native, untreated corn stover. Experiments on corn stover were carried out also in pilot scale. No effect of the scale on the yields of hemicellulose sugars was observed, but the effect was significant on glucose yield. The highest overall glucose yield, 84% was achieved in optimal conditions in the pilot reactor, compared to 63% obtained in laboratory-scale. The final ethanol yield was about 70% in lab scale, while an average yield of 76% in pilot-scale was obtained.
Superior cellulases operating at high hydrolysis temperature were screened from various microbial strains. Two cellobiohydrolases, three endoglucanases, two xylanases and three beta-glucosidases were purified to homogeneity. The specific activities of the purified thermostable cellobiohydrolases against synthetic substrates and Avicel were generally higher than those of Trichoderma. Cellobiose inhibition of all purified enzymes was less severe than with Trichoderma CBHI. During the project thirteen genes (genomic copies) were cloned, sequenced and transferred into T.reesei for production. Eleven were expressed in T. reesei. The genes were transformed into a host strain, which has all the major cellulase genes deleted. In addition, CBDs (cellulose binding domains) were attached to some of the thermophilic enzymes, which naturally lack the CBD. The cloned enzymes were produced in T. reesei in laboratory scale and preliminarily characterized using the culture supernatants. Most of the enzymes were 10-15 ºC more thermostable than their T. reesei counterparts. The most interesting cellobiohydrolases were purified from crude culture filtrates using affinity chromatography. The specific activities of the purified enzymes against synthetic substrates and Avicel cellulose were generally higher than those of Trichoderma. Cellobiose inhibition of all purified enzymes was less severe than with Trichoderma CBHI. The thermal stability of the enzymes was about 10 °C higher than that of T. reesei CBHI. Based on kinetic measurements, the new cellobiohydrolases were clearly more efficient in hydrolysis than the state-of-art enzyme T. reesei CBHI. The kinetic constants of the beta-glucosidase preparations were determined. Substrate and product inhibition pattern analysis showed that the novel thermostable ß-glucosidase was less affected by end-product inhibition. However, similarly to the reference enzymes, the activity of the thermostable ß-glucosidase proved to be more influenced by glucose at 70°C than at 50°C. Thermophilic cellulase preparations were manufactured in pilot scale for designing the developmental cellulase preparation for hydrolysis experiments in the project. The T. reesei strains in the project were constructed to produce essentially monocomponent enzyme preparations. Three thermophilic cellulases, identified as most promising enzymes in their categories (CBH, EG and ß-glucosidase) were produced in T. reesei and mixed to compose a novel mixture of thermophilic cellulases. For the xylan containing substrates, thermophilic xylanase was added. A T. reesei strain was constructed to be used for evaluating the production of the cloned cellulases using various production conditions and raw materials. To test the utilization of various side fractions of the process, hydrolysed hemicellulose both from wet-oxidation and steam pretreatment using corn stover as raw material was examined. These hemicellulose hydrolysates could be used to replace some of the defined medium.