Neural Network Modeling and Sorption of As III with Zinc Oxide Nanoparticle Bounded on Activated Silica using Ocimum Sanctum.

A three-layered Artificial Neural Network (ANN) model was more advanced to foretell the removal efficiency of As (III) ions from aqueous solution with Ocimum Sanctum. Batch experiments resulted into standardization of optimum conditions of adsorbent dosage (3 g), As (III) concentration (0.007N) volume (20 ml)at pH 6 and agitation speed of 250 rpm. A time of thirty minutes was found sufficient to achieve the equilibrium. The ANN model was premeditated to forecast adsorption efficiency of Zinc oxide nanoparticle ingrained on activated silica using Ocimum Sanctum (ZnO-NPs-AS-Os) by combining back propagation (BP) with prime component analysis using neuron (R2011a) solution. A tangent sigmoid axon was used as transfer function for input to hidden layer whereas a linear purelin function is used at output layer. The Levenberg?Marquardt Algorithm (LMA) was applied to give minimum Mean Squared Error (MSE) for training, testing and cross validation. Comparison between the model results and experimental data gives a high degree of correlation (R2 = 0.9846) indicating that the model is able to predict the sorption efficiency with reasonable accuracy.