The present study is dedicated to investigate the kinetic analysis for catalytic co-pyrolysis of palm kernel shell (PKS) and polyethylene waste (HDPE) mixtures with bifunctional HZSM-5 and mussel shell (MS) catalyst. Artificial neural network (ANN) modeling through 17 models based on the functions of reaction mechanism denoted as chemical reactions, diffusion reactions, nucleation and growth reactions, interfacial phase reactions and power law reactions was used in this study to achieve a suitable order of reaction promoting higher rate of accuracy based on the data achieved. It was found that the 2nd order, 3rd order, anti jander, jander and ginsling were selected as the suitable models out of 17 reaction mechanism kinetic models due to the giving a positive values produced. And also, it was observed that 3rd order reaction mechanism model provided higher activation energy (E A) and A values for all feedstock used. Comparison between experimental data and predicted data using Logsig-Tansig (LT) and Tansig-Tansig (TT) were carried out and it was observed that the predicted results from ANN showed similar trend as experimental data with minimal error of 0.67%, 6.61%, and 2.46% for PKS, HDPE, and mixture of PKS and HDPE with the presence of MS and HZSM-5 catalyst, respectively. From the kinetic analysis, E A and A value of PKS, HDPE, and of PKS and HDPE with the presence of bifunctional of HZSM-5/MS catalyst are 196.93 kJ/mol, 388.00 kJ/mol, 147.12 kJ/mol, 3.24×1013 s-1, 6.00×1026 s-1, and 1.72×10-1 s-1, respectively.
Read full abstract