Oxidation of ethyl acetate by a high performance nanostructure (Ni, Mn)-Ag/ZSM-5 bimetallic catalysts and development of an artificial neural networks predictive modeling

The catalytic oxidation of ethyl acetate in low concentration was investigated over mono-metallic Ag/ZSM5 and bimetallic (Ni, Mn)-Ag/ZSM-5 catalysts. Catalytic studies were carried out in a catalytic fixed bed reactor under atmospheric pressure. The sequence of catalytic activity was as follows: Ni-Ag-ZSM-5 > Mn-Ag-ZSM-5 > Ag-ZSM-5 > H-ZSM-5. The catalysts were characterized by ICP-AES, X-ray diffraction (XRD), low temperature nitrogen adsorption, NH(3)-TPD, transmission electron microscopy (TEM), scanning electron microscopy (SEM) and diffuse reflectance UV-vis spectra (UV-vis). An artificial neural networks (ANN) model was developed to predict the performance of catalytic oxidation process over bimetallic Ni-Ag/ZSM-5 catalyst based on experimental data. For this purpose the standard feed forward back propagation algorithm was employed to train the model by using laboratory experimental data. A good agreement was resulted between experimental results and those obtained by ANN. Following order for variables effects on conversion yield of ethyl acetate was predicted by ANN model: reaction temperature (32.99%) > Ag loading (27.38%) > initial ethyl acetate concentration (23.58%) > Ni loading (16.05%).