Neural Network Modeling of Heavy Metal Sorption on Lignocellulosic Biomasses: Effect of Metallic Ion Properties and Sorbent Characteristics

This study reports the application of a neural network approach for modeling and analyzing the sorption performance of different lignocellulosic wastes, namely jacaranda fruit, plum kernels, and nutshell, for the removal of heavy metal ions (Pb2+, Cd2+, Ni2+, and Zn2+) from aqueous solutions. This artificial neural networks (ANNs) model was used to determine the relevance and importance of both sorbent and pollutant characteristics on the metal sorption kinetics and isotherms. Results of this study highlighted the role of acidic functional groups, lignin composition of tested biomasses, and the pollutant molecular weight in the sorption of heavy metals. The nutshell biomass showed the best sorption properties for heavy metal removal, where its monolayer sorption capacities ranged from 1.0 to 7.0 mg/g. In summary, this study highlights the capabilities of ANN-based models for analyzing and understanding complex but relevant sorption processes for environmental protection and wastewater treatment.