Predicting Fenton modification of solid waste vegetable oil industry for arsenic removal using artificial neural networks

Abstract

Arsenic pollution is a worldwide problem that severely affects the human health, since the ingestion of arsenic is through contaminated drinking water and food harvested and prepared with arsenical water. Anthropogenic activities contribute to increase the levels of the natural occurrence of inorganic and organic arsenic in the environment. Inorganic arsenic in groundwater is mainly present as nonionic As (III) and ionic As (V) forms, and its speciation depends on the redox conditions, pH, biological activity, and adsorption reactions in the aquatic environment. In this study a direct artificial neural network (ANNs) approach was developed to predict the removal of As (III) and As (V) from aqueous solution under various experimental conditions. The arsenic removal was carried out using a natural sorbent prepared by the modification of solid waste vegetable oil industry with Fenton reagent (FMSWVOI). The maximum arsenic removal by the FMSWVOI was achieved at Fe2+/H2O2=1:17 and 30min of contact time, with 81% As (III) removal at pH 2 and 75% As (V) removal at pH 5. The artificial neural network (ANNs) was found to interpolate the data with good accuracy. The simulations of the trained network were in close agreement with the actual values. It was also observed that the Fenton process is an effective method for the modification of the solid waste vegetable oil industry to remove arsenic (III) and (V) from aqueous solution.