Abstract
The ability of olive pips in the removal of lead ions from polluted water was investigated using three-layer artificial neural network and 114 batch experiments. The experiments revealed that the highest removal (99%) was achieved at pH 5, agitation time- 60 min, biosorbent dose- 0.3 g/100 mL of liquid phase, speed- 200 rpm at room temperature and 50 mg/L initial concentration. The transfer functions in the artificial network model were linear and tangent sigmoid for output and hidden layers, respectively where eight neurons presented a good description for results of batch experiments with coefficient of correlation >0.997. The analysis of sensitivity proved that the initial pH of solution had more influence in the present sorption process with relative importance of 34%. The outputs of artificial network were integrated with the solute transport equation for description the propagation of lead ions through fixed bed column packed with olive pips. For breakthrough curves, a satisfactory concurrence was observed between outputs of model and measured concentrations because Willmott's index was >0.97.