Factors influencing the fermentation of spruce hydrolysate by Dekkera bruxellensis and Saccharomyces cerevisiae

Abstract: Bioethanol and biogas are two possible alternatives to fossil fuel resources. The second generation fermentations involving lignocellulosic material is one of the latest fields for bioethanol science. In the current work two fermentation yeasts, Saccharomyces cerevisiae J672 and the alternative fermentation yeast Dekkera bruxellensis CBS 11269, were compared in an experimental design for their ability to ferment spruce lignocelluloses. The goal was to find differences between the growth and ethanol yield of the two species. The investigation was done with a multivariate data analysis tool to examine yeast growth and ethanol yield. Four quantitative factors, pH, temperature, concentration of hydrolysate, and initial cell density were used. The analysis also included one qualitative factor with two settings, aerobic and oxygen limited. The five factors were tested in a fractional factorial design created with MODDE software. The results for the experimental design with S. cerevisiae showed that the main significant factor for ethanol yield was oxygen. Oxygen was the most important factor for S. cerevisiae ethanol yield response, oxygen limitation had positive contribution and aerobic environment had negative contribution. The oxygen factor however was not important for growth. pH was a significant factor and high pH was shown to give better response, for both yeast growth and ethanol yield. Oxygen and temperature had negative contribution while the two factors initial cell concentration and concentration of hydrolysate did not show any significant contribution to the model for S. cerevisiae ethanol yield. D. bruxellensis was adapted to spruce hydrolysate before the final experiment. This was because the non adaptedD. bruxellensis CBS11269 did not grow well, in the first experiment, and did not build a model. The two models with S. cerevisiae J672 and spruce adapted D. bruxellensis CBS 11269 are therefore not directly comparable. The adapted D. bruxellensis model for ethanol yield was not complete. Analysis of the data showed a strong curvature within the model and further analysis is needed to complete the model. However the experiment resulted in a very good model for yeast growth. It showed that pH and initial cell concentration was positive for growth and that hydrolysate concentration had a strong negative contribution to growth. The temperature had a negative contribution and the oxygen factor was not significant for D. bruxellensis growth. When comparing the ethanol yields between both adapted and non adapted D. bruxellensis with S. cerevisiae, bothD. bruxellensis yeasts shows a higher or equal total yield. A significant statistical difference was shown between the centre samples with access to oxygen, where D. bruxellensis gave a higher ethanol yield.