Modelling the solid–liquid adsorption processes using artificial neural networks trained by pseudo second order kinetics

Abstract

A three-layer feed forward neural network was constructed and tested to analyze the second order kinetics of solid–liquid adsorption process. The pseudo second order kinetics of auramine O onto activated carbon was used to train the artificial neural network (ANN) to model the sorption system for various operating conditions. The operating variables studied are the contact time, initial dye concentration, agitation speed, temperature, initial solution pH and activated carbon mass. The studied operating variables were used as the input to the constructed neural network to predict the dye uptake by pseudo second order kinetics at any time as the output or the target. The dye uptake predicted by ANN trained by pseudo second order kinetics was found to be precise in representing the experimental kinetics of auramine O uptake by activated carbon. The constructed network was also found to be precise in predicting the sorption kinetics of auramine O by activated carbon for the new input data which are kept unaware of the trained neural network showing its applicability to determine the dye uptake rate for any operating conditions under interest. The ANN was also trained using pseudo second order kinetics of sorption of divalent metal ions onto peat particles and also using the second order kinetics of cadmium ions onto tree fern particles. The ANN and pseudo second order kinetics compliment each other to model the studied sorption systems for a wide range of operating conditions.