An investigation of the oxidative dehydrogenation of propane kinetics over a vanadium-graphene catalyst aiming at minimizing of the COx species

In the current investigation, an application of the design of experiments (DOE) along with the artificial neural networks (ANN) in a kinetic study of oxidative dehydrogenation of propane (ODHP) reaction over a synthesized vanadium-graphene catalyst at 400-500 degrees C presented aiming at minimizing the COx production. In this venue, the main and side reactions' unknown and variable reaction kinetics network expressed through the power law equations and determined via non-linear regression analysis. The collected kinetic experimental data attributed to three operating factors including the temperature, feed molar ratio and total feed flowrate. The neural network-based optimum was compared with that of the experimental data. In addition, the predictions of the response surface methodology (RSM) and ANN models based upon the DOE information utilized to generate extra simulated data for further analysis. Then different data sets used to fit the power law kinetic rates for the main ODHP and side reactions. Propane to air ratio was found to be a critical parameter for optimization of the ODHP reaction. Distribution of the propane consumption rate as well as the propylene and ethylene formation rates to that of the COx investigated and optimizations were performed. It was revealed that, based on different number of data points utilized over the vanadium-graphene catalyst these ratios were higher than unity confirming that, the COx production minimized. Moreover, under such conditions, the ratio of ethylene to COx production rate was noticeably below unity indicating the COx formation was higher than that of the ethylene. Kinetic modeling results incorporating simulated data from the ANN and RSM models compared with those obtained from the experimental ones resulted in less than 10% deviations. Considering the complexity of the undertaken system, this comparison was rather satisfactory. (C) 2014 Elsevier B.V. All rights reserved.