ABSTRACTA comparison of linear least-squares method and a trial and error non-linear method of three widely used isotherms, Langmuir, Freundlich, and Temkin, were examined. The batch sorption model, based on a pseudo-second-order mechanism, was applied to predict the rate constant of sorption, the equilibrium capacity and the initial sorption rate with the effects of the initial solution pH (4.5) and adsorbent dose. The calculated adsorption capacity (qe) was qe = 97.6 mg/g. Langmuir isotherm parameters obtained from the Langmuir linear equations were used to fit the adsorption equilibrium data. The adsorption capacity of the adsorbent, obtained from the Langmuir model, was up to 1.78 mol/g−1. In order to optimise the adsorption-isotherm model, correlation coefficient (R2) of 0.991 and 0.841 error functions were employed to facilitate the evaluation of fitting accuracy.