Investigation of Silica-Supported Preyssler Nanoparticles as Nanocatalysts in Alkylation of Benzene With 1-Decene Using Artificial Intelligence Approach

Abstract

Silica-supported Preyssler nanoparticles were synthesized and tested in alkylation of benzene with 1-decene. Adaptive network based fuzzy inference system (ANFIS) was successfully applied for studying the operating parameters of this catalytic reaction. The reaction was carried out at a constant temperature of 80 °C for 2 h, while catalyst loading, catalyst weight percent, and benzene to 1-decene molar ratio (Bz/C10 ) were chosen as independent variables. Prediction of 1-decene conversion and linear alkylbenzene (LAB) production yield were performed by applying ANFIS method. The predictive ability and accuracy of ANFIS model were examined using unseen experimental data set and R2 was obtained to be 0.994 and 0.995 for 1-decene conversion and LAB production yield, respectively. Experimental results revealed that catalyst loading, Bz/C10 molar ratio, and catalyst weight percent have positive effect on 1-decene conversion, while increase in catalyst loading tends to decrease LAB production yield.