Modelling and Optimization of Biodiesel Production From Waste Cooking Oil Using Graphene Oxide Catalyst via DFT and PINN

This research focuses on the production of biodiesel from waste cooking oil using graphene oxide (GO) as heterogeneous catalyst employing experimental and kinetic modelling approaches. The modified Hummers method was used to synthesized the GO and subsequently it was characterized by SEM, HR‐TEM, XRD, BET, and FTIR to confirm its layered structure and oxygen‐containing functional groups. Process optimization was carried out using Response Surface Methodology with a Box–Behnken design (BBD), determined the optimal transesterification conditions, achieving a maximum biodiesel yield of 92.6% at a methanol‐to‐oil ratio of 5:1, catalyst loading of 6 wt%, reaction temperature of 52.5°C, and reaction time of 60 min. GC–MS analysis confirmed the high fatty acid methyl ester content (80%–90%). Density functional theory (DFT) calculations interpret the strong reactant adsorption on the GO surface and a significant reduction in the activation energy from 92.0 to 52.7 kJ /mol, showing the enhanced catalytic performance. Kinetic modelling favored the experimental results ( R 2 = 0.97), while the physics‐informed neural network (PINN) provides accurate yield prediction with minimal error. Overall, the results demonstrate that GO‐catalyzed transesterification, supported by multi‐scale modelling and artificial neural network provides a sustainable and the efficient route for biodiesel production from waste‐derived feedstocks.