Both batch experiments (initial pH value was 7.0 and 8.0, respectively) were conducted to convert soluble starch to hydrogen at 35 degrees C. Anaerobic mixed bacteria acclimated with soluble starch was used as inoculum. At initial pH = 8.0, the H-2 yield significantly increased from 106.4 to 274.0 ml/g starch with increasing iron concentration from 0 to 200 mg FeSO4/l. When iron concentration continued to increase from 200 to 4000 mg FeSO4/l, iron inhibition did not happen. On the contrary, hydrogen production was efficiently accelerated. Here, lag-phase time and end time all had about 22 h shortened. However, corresponding cumulative H2 volumes were adjacent to the maximum value (225 ml) required at [FeSO4] = 150 mg/l. As for pH 7.0 systems, though flocculation still appeared, superfluous iron (over 800 mg FeSO4/l) slightly inhibited hydrogen production. The difference in two strains of iron experiments resulted from the variance in initial solubility of iron under various initial pH values. The experiment results show that superfluous soluble iron shows slightly inhibitive influence on H-2 production, and that the optimum soluble iron concentration is 150 mg FeSO4/l from soluble starch (10.0 g/l). Furthermore, the effect of the starch concentration (5.0-40.0 g/l) on hydrogen production also was investigated under 150 mg FeSO4/l. The result shows that high starch concentration has no remarkable effect on H-2 production. The maximum cumulative hydrogen was 260.5 ml at starch concentration of 20 g/l. (c) 2006 International Association for Hydrogen Energy. Published by Elsevier Ltd. All rights reserved.