Continuous Ethanol Production from D-Xylose by Candida shehatae M. A. Alexander and T. W. Chapman Department of Chemical Engineering, University of Wisconsin, Madison, Wisconsin 53706 T. W. Jeffries* Institute for Microbial and Biochemical Technology Forest Products laboratory ✝ Madison, Wisconsin 53705 Accepted for publication April 14, 1987 Much progress has been made since the discovery that some yeasts can ferment D-xylose to ethanol. 1 Initially, only a few yeasts were found that could produce ethanol from D-xylose. 2,3 Today, at least eight species of yeasts are known to produce significant amounts of ethanol from D-xylose. 2,4-6 Three of these species have been studied ex- tensively: Pachysolen tannophilus, Candida shehatae, and Pichia stipitis. Under oxygen-limited conditions C. shehatae and P. sti- pitis are capable of faster specific fermentation rates than is P. tannophilus. Specific fermentation rates in aerated cul- ture as high as 0.30 and 0.48 g/g/h have been reported for P. stipitis and C. shehatae, respectively. 7 A corresponding value of 0.12 g/g/h has been reported for P. tannophilus. 3 On the other hand, the specific fermentation rates of these three organisms in unaerated culture are not that different. Values of 0.06, 0.09, and 0.07 g/g/h have been reported for P. tannophilus, C. shehatae, and P. stipitis, re- spectively. 8 We chose to investigate the aerobic mode of ethanol production by C. shehatae because high fer- mentation rates are obtained with low aeration using this organism. The objectives of this work were to determine the effects of important process variables such as aeration rate, D-xylose concentration and ethanol concentration on the fer- mentation rate, and to obtain high volumetric ethanol pro- duction rates. Continuous methods were used because they allow collection of accurate rate data at constant environ- mental conditions. Such data facilitate the development of a kinetic model describing the effects of process variables. We were especially interested in fermentation performance at high volumetric production rates. High volumetric rates re- * To whom all correspondence should be addressed. ✝ Maintained in cooperation with the University of Wisconsin-Madison. The use of trade, firm, or corporation names in this publication is for the information and convenience of the reader. Such use does not constitute an official endorsement or approval by the U. S, Department of Agriculture of any product or service to the exclusion of others which may be suitable. quire high cell density, which is usually achieved using a cell immobilization or recycle technique. Two reports of continuous D-xylose fermentation using high cell density appear in the literature. Both studied anaer- obic D-xylose fermentation by cells entrapped in calcium alginate beads. Slininger and co-workers reported a specitic fermentation rate of 0.051 g/g/h with P. tannophilus, which was independent of cell density. 9 A maximum volu- metric ethanol production rate of 2.2 g/L/h was obtained with 43 g/L dry wt biomass. Linko and co-workers reported a maximum volumetric fermentation rate of 1.0 g/L/h with immobilized P. stipitis. 10 Immobilization via cell entrapment in a supporting matrix can introduce diffusional limitations. Slininger and co- workers found that the fermentation rate of entrapped cells was only 40% that obtained with free cell suspensions at the same biomass concentration. Furthermore, the lack of aera- tion in their bioreactor led to viability loss after five days. Diffusional problems can be circumvented by employing dynamic cell immobilization produced by membrane- assisted microfiltration. The use of microfiltration to retain cells within the fermenter eliminates the need for a support matrix, and therefore allows relatively unimpeded nutrient and oxygen transport. Thus, cell retention by microfiltration can give faster fermentation rates than cell entrapment and allows the use of aeration. For these reasons we chose this technique to obtain high cell density in aerated culture. Cell retention has been used to obtain extremely high volumetric fermentation rates with little effect on cell pro- ductivity in anaerobic glucose fermentation with Succha- romyces cerevisiae. An ethanol production rate of 95 g/L/h with complete substrate utilization has been ob- tained using 60 g/L cell density. 11 This rate represents a specific fermentation rate of 1.6 g/g/h which compares favorably with a specific rate of 1.7 g/g/h obtained in an- aerobic fermentation at much lower cell density. 12 Chemostat techniques have also been employed in the study of D-xylose fermentation. 13 Ethanol production by P. tannophilus under oxygen-limited conditions was examined Biotechnology and Bioengineering, Vol. 30, Pp. 685-691 (1987) 1987 John Wiley & Sons, Inc.