Removal of Methylene Blue, Malachite Green and Rhodamine B in a Ternary System by Pistachio Hull; Application of Wavelet Neural Network Modeling and Doehlert Design

Abstract

Most of previous papers in the field of dye removal used one dye or dyes with nearly separate spectra that simplifies dyes concentration determination by Beer's law at different λmax. In many real situations, dyes with highly overlapped spectra exist and their concentrations can be determined by multivariate analysis methods. In this study, principal component-wavelet neural network (PC-WNN) was used for concentration determination of rhodamine B (RB), methylene blue (MB) and malachite green (MG) in ternary solutions. Biosorption of dyes by pistachio hull was selected as model. The effects of operating parameters were studied by Doehlert experimental design. The maximum uptake capacity of pistachio hull for cationic dyes was 0.682 mmol/g. It was found that the overall biosorption data was described by the pseudo second-order kinetic model. Nine isotherm models were nonlinearly fitted to experimental data and Redlich-Peterson and Radke-Prausnitz isotherm models had the best fit for biosorption equilibrium data.