A 3(2) full-factorial design combined with response surface methodology was used to investigate the simultaneous effects of starting xylose concentration (S-0) and oxygen mass flowrate (q(O2)) on xylitol production from rice straw hydrolyzate. Fermentations were performed at 30 degreesC, using Debaryomyces hansenii NRRL Y-7426 as xylitol producer and varying S-0 between 50 and 150 g/l and qo(2) between 2.5 and 5.9 mg(O2)/s. At the lowest starting xylose concentration and an oxygen mass flowrate of 4.2 mg(O2)/s, volumetric productivity and xylitol yield reached maximum values (Q(p) = 0.70 g(P)/l h and Y-P/S = 0.73 g(P)/g(S), respectively), but xylitol concentration was quite low (35.9 g(P)/l). The results were in close agreement with the model prediction. The statistical model also allowed identifying the optimum operating conditions (S-0 = 71 g/l and q(O2) = 4.1 mg(O2)/s) able to simultaneously maximize volumetric productivity (0.53 g(P)/l h), xylitol yield (0.71 g(P)/g(S)) and final xylitol concentration (42.2 g(P)/l). (C) 2004 Elsevier Ltd. All rights reserved.