Determination of single and multi-component adsorption isotherms using nonlinear error functions and spreadsheet optimization technique

The present study investigated and determined the ability of existing predictive and correlative models to calculate equilibrium sorption data of Methylene Blue, Rhodamine B, and Malachite Green Oxalate onto a commercially available activated carbon Filtrasorb 400 and an agricultural byproduct made from acid activation of almond shells. Single, bi-solute and tri-solute system experimental data were obtained by conducting batch experiments in their aqueous solutions. Freundlich, Langmuir, Redlich-Peterson, and Toth isotherms coupled with five nonlinear error measures were evaluated in single-solute systems for experimental data derived from two analytical methods, namely the UV spectrophotometer and the TOC analysis. For the multi-solute systems, nonlinearly determined single-component parameters of the Freundlich, the Langmuir and the Redlich-Peterson isotherms coupled with their multi-component forms and interaction factor were used to correlate multi-component experimental data.; The empirical extended Freundlich model produced the closest fit to the binary data in each case. In the case of ternary system, the extended Redlich-Peterson model provided a very close fit to the experimental data of each component. Over all, optimization coupled with nonlinear error regression analysis provided a simple, robust and computationally effective technique of producing optimal fitting parameters for sorption models from multi-component equilibrium data and single-component optimal parameters.