Artificial neural network and molecular modeling for assessing the adsorption performance of a hybrid alginate-based magsorbent

Abstract The study reported herein describes a chemometric approach, based on an artificial neural network (ANN), to predict the adsorption performance of a new hybrid material. This composite adsorbent with magnetic properties (magsorbent) consists of a cross-linked alginate matrix that incorporates inorganic nanoparticles of cobalt ferrite. Instrumental physical–chemical techniques (FTIR, VSM, SEM, and EDX) were employed for material characterization. The produced magnetic-adsorbent was used to remove a persistent organic contaminant from aqueous solutions. Adsorption studies were carried out to assess the kinetics, isotherms, and thermodynamics parameters. The calculated free energy of adsorption ranged from 9.38 to 10.61 (kJ/mol) suggesting the retention mechanism based on ion-exchange. The good generalization capability of the ANN-model allowed establishing the optimal adsorption conditions by the model-based simulation. Under optimal conditions, the maximal removal efficiency of 96.54% was observed for an initial concentration of 110 mg/L of the pollutant. Details regarding the interactions at the molecular level were provided by computational chemistry methods. Molecular docking revealed that binding of the cationic organic pollutant to the cross-linked alginate matrix relied on hydrophobic and electrostatic interactions. Molecular dynamics simulation disclosed the behavior of the docked complex against the simulation time and in the presence of the explicit solvent.