Unravelling the adsorption mechanism of phenol on zinc oxide at various coverages via statistical physics, artificial neural network modeling and ab initio molecular dynamics

Abstract

A novel modeling approach based on ab initio molecular dynamics, artificial neural networks and statistical physics has been applied to analyze, characterize, and explain the phenol adsorption on zinc oxide. Phenol experimental adsorption studies were performed with zinc oxide synthesized via the precipitation method. Adsorption equilibrium data were obtained at 30 to 60 °C for the modeling analysis. Artificial neural networks and statistical physics calculations predicted a monolayer phenol adsorption process, which was endothermic and multimolecular. It was expected that hydrogen bonding and electrostatic forces were involved in the adsorption of this organic pollutant. Ab initio molecular dynamics confirmed these predictions thus concluding that phenol adsorption involved mixed vertical and tilted configurations depending on the adsorbent coverage. This study contributes with new theoretical findings to understand the adsorption of phenol molecules on zinc oxide surface.