Analysis of two-parameter and three-parameter isotherms by nonlinear regression for the treatment of textile effluent using immobilized Trametes versicolor: comparison of various error functions

The ability of the immobilized Trametes versicolor, as biosorbent, to degrade the textile effluent, in terms of chemical oxygen demand and color, was studied under batch conditions. The removal efficiencies were calculated for the effect of initial concentrations, biomass dosages, shaking, and nutrients. Equilibrium data were analyzed using, two-parameter isotherms: Langmuir, Freundlich, BET, Flory-Huggins, Temkin, Jovanovic, Harkins-Jura and Halsey. Three-parameter isotherms: Redlich-Peterson, Sips, Khan, Toth, Langmuir-Freundlich, Fritz-Schlunder, Koble-Corrigan, Hill, Brouers-Sotolongo, and Radke Prausnitz, using nonlinear regression. The model parameters were optimized using Microsoft Excel Solver Add-Ins function. The best isotherm was identified by error analysis, using nine different error functions: residual root mean square error, normalized standard deviation, average relative error, sum of the squares of the error, sum of the absolute errors, hybrid fractional error function, residual variance, Marquardt's percent standard deviation, and chi-square test ((2)).