The design of modified H-ZSM-5 catalyst for catalytic conversion of methanol to gasoline range hydrocarbons (MTG) was carried out using neuro-genetic approach. Four factors: SiO2/Al2O3 ratio in zeolite (denoted as Si/Al2), calcination temperature (Tc), reaction temperature (Tr) and nominal weight loadings (Lm) of three metals (Ag, Cu, Co) were modeled simultaneously by using combined experimental data and physicochemical descriptors of metals. The obtained optimum model was used as fitness function for hybrid genetic algorithm to find the optimum catalyst. The optimum catalyst was Cu/ H-ZSM-5 with Si/Al2 = 40, Tc = 490 8C, Tr = 340 8C, and Lm = 9.24 wt% with methanol conversion of 96.23%, and gasoline yield of 13% that resulted in 4.88% increase in gasoline yield compared to H-ZSM-5 catalyst at the same conditions. The higher experimental yield of gasoline for predicted catalyst along with the fact that this approach can be applied easily for a large number of catalysts, confirms that this approach can be used for design of heterogeneous catalysts. UV–Vis DR spectra confirmed the possibility of O2 Cu2+ O2 Cu2+ O2 chain structures in the channels of zeolite for Cu/HZSM-5 catalyst as active components in MTG reaction
