our previous work demonstrated that an acclimated mixed bacterial consortium was able to produce H(2) from sugar substrates. To reduce the medium cost for more commercially viable H(2) production, cassava starch was used as the feedstock to produce H(2) via dark fermentation. Three factors, namely, temperature, pH and starch concentration (C(starch)), were intensively examined for their effects on H(2) production activity The H(2) production kinetics was determined using a Monod-type kinetic model. The results show that mesophilic temperature (37 degrees C) is preferable for H(2) production with the H(2)-producing sludge used. The H(2) production efficiency and the composition of soluble metabolites were found to be highly sensitive to the change in pH, as pH 6.0 seemed to give the best overall H(2) production performance. In a non-pH-controlled culture (initial pH = 8.5), ethanol and butyrate were the major soluble metabolites, whereas the predominant metabolites switched to butyrate alone (accounting for 70-80% of total soluble microbial products) when the culture pH was controlled at a fixed level ranging from 5.5 to 7.0. Meanwhile, the maximum H(2) production rate occurred when the initial starch concentration was 24 g CODA. The dependence of H(2) production rate on starch concentration could be described by using Monod-type model and the predicted kinetic constants, namely, maximum H(2) production rate (upsilon(max),H(2)) and Monod constant (K(s)), were 1741 ml/h/l and 16.28 g COD/l, respectively Under the optimal conditions (37 degrees C, pH 6.0, C(starch) = 24 g COD/l), the H(2) production rate increased to 1119 ml/h/l, while a high H2 yield of 9.47 mmol H(2)/g starch was obtained. This performance appeared to be superior to that obtained from other starch-to-bioH(2) systems reported in the literature. (c) 2007 International Association for Hydrogen Energy. Published by Elsevier Ltd. All rights reserved.