In this work, H-2 was produced by anaerobic mixed microflora with phosphate-buffered medium containing starch or enzyme-treated starch hydrolyzate as the carbon substrate. The effect of pH on H-2-producing performance was examined for cultures converting raw starch or hydrolyzed starch into H-2. Response surface methodology was utilized to determine the best condition (41 degrees C, pH 5.2, 2.1% (v/v) enzyme dosage, 27h reaction time) for starch hydrolysis with concentrated crude amylase obtained from Bacillus subtilis ATCC 21332. The mixed culture was able to produce H-2 at an optimal pH of 7.0 irrespective of raw or hydrolyzed starch. Direct starch fermentation attained a highest maximum H-2 production rate (R-max), overall H-2 production rate (R-overall), and H-2 yield (Y-H2) of 25.6 ml/h, 88 ml/h/l, and 5.28 mmol H-2/9 starch (4.64mmol H2/9 COD), respectively. In contrast, using hydrolyzed starch as the substrate gave rise to much better H-2 producing performance, as the highest Rmax, Roverall, and YH2 values increased to 43.1 ml/h, 210ml/h/l, and 6.1 mmol H-2/9 COD, respectively. This clearly demonstrates the advantage of using hydrolyzed starch for fermentative H-2 production. The soluble metabolites consisted primarily of acetate (HAc), ethanol (EtOH), butyrate (HBu), and 2,3 butandiol. The amount of H-2 produced from raw and hydrolyzed starch (especially, raw starch) could be estimated from formation of HAc and HBu known to stoichiometrically correlate with H-2 production. (c) 2007 International Association for Hydrogen Energy. Published by Elsevier Ltd. All rights reserved.