Linear and Non-linear Regression Methods to Study the Adsorption of Cr(VI) from an Aqueous Solution Using Pomegranate Peel

Abstract

In this study, the previous data of the adsorption of Chromium (VI) ions on the pomegranate peels (MPGP) as adsorbents were used. linear and non-linear equations for Langmuir, Freundlich, Dubinin-Radushkevich, Tempkin, Redlich-Peterson, Sips, and Toth adsorption isotherm models were applied. The five error functions ERRSQ, HYBRD, MPSD, ARE, and EABS were analyzed for non-linear isotherm equations and in each case a set of isotherm parameters were determined. The algorithms for the simulation of linear and non-linear isotherm models using error functions were achieved with the aid of Microsoft Excel solver Add-Ins for minimizing the respective error function across the concentration range studied. Linear and non-linear methods show comparable data for two-parameter models, also high R2 values were obtained for Langmuir, Freundlich, and Temkin models while the Dubinin-Radushkevich model reveals low R2 values, on the contrary, three-parameter models show different data between linear and non-linear methods with high R2 value for all. Which indicates that the error does not obey the Gaussian distribution.