Despite the multitude of studies on fermentative hydrogen production, the kinetic models which have already been developed or used to describe the process are limited. This is due to the fact that hydrogen production and metabolic products distribution is affected by many factors and up to now, the role of each is not well understood. So, there is a lack of models which incorporate important parameters such as pH, hydrogen partial pressure and regulation mechanisms like the ratio of NADH/NAD+, influencing hydrogen production and products’ stoichiometry.

The majority of the researchers have used simple models in order to describe their experimental data. For example many of them have used the modified Gompertz equation developed by Zwietering et al. (1990) to predict hydrogen evolution in batch tests, using different substrates and inocula (pure or mixed cultures) (Lay et al, 1999; Chen et al., 2002; Wu and Lin, 2004; Fang et al.,

2006) . However, this equation cannot be applied in continuous systems, and it cannot predict the concentrations of substrates utilized and those of metabolites produced along with hydrogen.

Recently, researchers have used more complicated models, such as modified versions of the IWA Anaerobic Digestion Model No.1 (ADM1) (Batstone et al., 2002). The latter is a widely applicable mathematical model, which was developed for describing the anaerobic digestion process. The application of ADM1 to non — methanogenic systems demands modifications, since the initial model structure uses constant-stoichiometry to describe product generation from carbohydrates fermentation as well as excludes lactate and ethanol — two important metabolic products — from its structure. Lin et al. (2007) used modified ADM1 to describe glucose metabolism and products distribution (butyrate, acetate and ethanol) by selected clostridium species in batch cultures. Rodriguez et al. (2006a; 2000b) proposed an initial model to mechanistically describe formation of products in anaerobic fermentations and the predictions of this model were integrated in ADM1 as a variable stoichiometry function. Penumathsa et al. (2008) modified ADM1 in order to apply it to continuous bio-hydrogen production systems using a variable stoichiometry approach derived from experimental information. The simulation results obtained, provided good predictions of the dynamics in a continuous bio-hydrogen production reactor fed with sucrose, over a wide range of influent substrate concentrations. However, the modified ADM1 cannot predict and simulate the distribution of products from glucose metabolism under different environmental conditions. So, the induction of a proper regulating mechanism to regulate the fractionation of monosaccharides depending on hydrogen partial pressure, temperature, pH, etc. should make the model more robust and reliable for describing continuous fermentative hydrogen production systems.