An intelligent approach to design and optimization of M-Mn/H-ZSM-5 (M: Ce, Cr, Fe, Ni) catalysts in conversion of methanol to propylene

•Design and optimization of the bimetallic catalysts (M-Mn/H-ZSM-5; M: Ce, Cr, Fe and Ni) was carried out in order to enhance the propylene selectivity.•An artificial neural network (ANN) model was developed using preparation conditions (wt. % of second metal loading, calcination temperature and calcination time) and the atomic descriptors of second metal.•The ANN model was linked to the Genetic algorithm and the optimum catalysts were found.•The maximum propylene selectivity was produced via Ce-Mn/H-ZSM-5 with the following catalyst preparation conditions: 2.46 wt. % of Ce loading, calcination temperature of 486 °C and calcination time of 4 h.•The best catalyst was characterized by, SEM, XDR, ICP, N2 adsorption, BET and TPD.

To enhance the propylene selectivity in catalytic conversion of methanol to propylene (MTP), the bimetallic catalysts were prepared by Mn/H-ZSM-5 with second metal of Ce, Cr, Fe and Ni. In order to design the bimetallic catalysts (M-Mn/H-ZSM-5; M: Ce, Cr, Fe and Ni) and to optimize the propylene selectivity, an artificial neural network (ANN) model was linked with genetic algorithm (GA). Investigation of the optimal catalyst preparation conditions (wt. % of second metal loading, calcination temperature and calcination time) and the atomic descriptors of second metal (electronegativity, melting enthalpy, atomic weight and ionization energy) were carried out by the ANN-GA model simultaneously. The model predicted that the maximum propylene selectivity was produced via Ce-Mn/H-ZSM-5 with the following catalyst preparation conditions: 2.46 wt. % of Ce loading, calcination temperature of 486 °C and calcination time of 4 h. The optimized propylene selectivity of model prediction and the experimental value were 54.3% and 54.8% respectively. The catalyst samples were characterized by XRD, FE-SEM, FT-IR, N2 adsorption/desorption, NH3-TPD and ICP-AES.

Graphical abstract Figure optionsDownload full-size imageDownload as PowerPoint slide