Modeling of underpotential deposition technique by a neural computation approach

The physical properties of metal monolayers and their influence on the surface properties of the substrate are of considerable interest to fundamental and applied physics. Electrooxidation of organic molecules as in the development of fuel cells is an important subject for studying of the effects of foreign metal adatoms. Monolayer deposition at underpotentials results from a strong interaction between foreign metal adatoms and substrate. In this study, a correlation between the differences of work functions of adsorbed metal and substrate metals (ΔΦ) and the shift in the oxidation peak potential (ΔE_p) and differences in current densities (Δj) with respect to monolayer formation was investigated by a neural network methodology for the electrooxidation reaction of D.glucose in alkaline medium. This paper presents mainly a subjective modeling of the experimental study by feed-forward neural networks.