Artificial Neural Network Modeling of Biosorptive Removal of Arsenic(V) by a Low-cost Biomass

The presence of arsenic in drinking water has been recognized as a serious community health problem because of their toxic nature and therefore, its removal is highly essential. This paper deals with batch biosorption study for the removal of pentavalent arsenic ions from aqueous solutions using finely ground (250 μm) Azadirachta indica (neem) bark powder (AiBP) as a low-cost biosorbent. Employing the batch experimental setup, the effect of operational variables such as initial concentration of As(V), pH, biosorbent dose, contact time, temperature and agitation speed on the As(V) removal process were studied. Under optimized batch conditions, the AiBP could remove up to 86.6% of As(V) from contaminated water. The biosorbent dose had the most significant impact on the biosorption process. The artificial neural network (ANN) model developed from batch experimental data sets, provided reasonable predictive performance (R^2 = 0.951; 0.967) of arsenic biosorption. The study on equilibrium biosorption of batch operation revealed that Freundlich isotherm model gave the best fit to experimental data. The nature of biosorption of As(V) by AiBP was physisorption as inferred from the D–R isotherm model. The biosorption is pseudo second–order, exothermic and spontaneous.