▎ 摘　　要

Cyclic voltammogram (CV) curves of Pt/graphene on different synthesis conditions of graphene oxide (GO) such as ratio of sulfuric to phosphoric acid and amount of KMnO4 were experimentally investigated and theoretically modeled. Reduction of the GO and Pt (IV) complex ions happened simultaneously in water/propylene (V/V) solution by gamma radiation. Physicochemical properties of the catalyst were characterized using transmission electron microscopy (TEM), x-ray photo-electron spectroscopy (XPS), and Fourier-transform infrared (FTIR) techniques. Standard training algorithms of Bayesian regularization (BR), scaled conjugate gradient (SCG), and Levenberg-Marquardt (LM) with standard transfer functions of tan-sigmoid, satlins, satlin, and log-sigmoid in a hidden layer and a linear transfer function in the output layer were applied for the process modeling of methanol oxidation. The results showed that the catalyst properties were strongly influenced by the synthesis conditions of the substrate, and the artificial neural network (ANN) with BR algorithm, tan-sigmoid function, and 44 processing neurons created the best predictability of the methanol oxidation as fuel in acidic medium for the synthesized electrocatalyst samples.