In the production of 2nd generation ethanol, using Saccharomyces cerevisiae, the highest productivity obtained using C5/C6 fermenting yeast is in the co-fermentation phase, in which xylose and glucose are fermented simultaneously. Extending this phase in a fed-batch process increases the yield, rate and additionally reduces needed yeast amount for pitching. Extending this phase, as long as possible, would further enhance yield and economy of the process. To realise the concept a fermentation monitoring technique was developed and applied. Based on online measured refractive index an optimal residual sugar concentration could be maintained in the primary fermentor during the feed phase, requiring little knowledge of the nature of the substrate. The system was able to run stably for at least five fermentor volumes giving an ethanol yield >90% throughout the run. This was achieved with addition of only urea to the wheat straw hydrolysate and with an initial yeast pitch of 0.2 g/L total of finished broth. It has the potential to improve the fermentation technology used in fuel ethanol plants, which could help to meet the growing demand for more sustainable fuels.
An economically viable production of second-generation bioethanol by recombinant xylose-fermenting Saccharomyces cerevisiae requires higher xylose fermentation rates and improved glucose–xylose co-consumption. Moreover, xylose-fermenting S. cerevisiae recognises xylose as a non-fermentable rather than a fermentable carbon source, which might partly explain why xylose is not fermented into ethanol as efficiently as glucose. This study proposes propagating S. cerevisiae on non-fermentable carbon sources to enhance xylose metabolism during fermentation. When compared to yeast grown on sucrose, cells propagated on a mix of ethanol and glycerol in shake flasks showed up to 50% higher xylose utilisation rate (in a defined xylose medium) and a double maximum fermentation rate, together with an improved C5/C6 co-consumption (on an industrial softwood hydrolysate). Based on these results, an automated propagation protocol was developed, using a fed-batch approach and the respiratory quotient to guide the ethanol and glycerol-containing feed. This successfully produced 71.29 ± 0.91 g/L yeast with an average productivity of 1.03 ± 0.05 g/L/h. These empirical findings provide the basis for the design of a simple, yet effective yeast production strategy to be used in the second-generation bioethanol industry for increased fermentation efficiency.