Modeling And Optimization of Hydrogen Production By The Photosynthetic Bacterium Rhodobacter capsulatus By The Methodology Of Design Of Experiments (DOE): Interaction Between Lactate Concentration And Light Luminosity

The methodology of the Design of Experiments was successfully tested for the modeling and the optimization of the hydrogen production by the photosynthetic bacterium Rhodobacter capsulatus. A D-optimal model based on the response surface methodology was established for six parameters. Five continuous parameters, implicated either as light, carbon, nitrogen sources or compounds of nitrogenase and one discrete parameter, the nature of buffer. The range of light intensity, concentrations of lactate, glutamate, iron, and molybdenum solutions varied between 8,000-30,000 lx, 20-120 mM, 7-50 mM, 40-200 mu M, and 1.5-50 mu M, respectively. The buffers were either the classical phosphate buffer (KHPO4/KH2PO4) or the borax buffer (Kolthof' buffer, KHPO4/Borax). The response concerning the volume of total hydrogen was modeled after transformation in its square root according to a polynomial model including simple, quadratic, and interaction effects. Two main parameters, in the model area, concerned the light intensity and the lactate concentration. The response was analyzed and discussed in terms of their variations in the 3D studied domains. (C) 2012 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of Canadian Hydrogen and Fuel Cell Association