Modeling, optimization, and adsorptive studies of bromocresol green dye removal using acid functionalized corn cob

Abstract

Genetic algorithm (GA) assisted optimization was used in the adsorptive removal of bromocresol green (BCG) from solution. The adsorbent was acid-functionalized corn cob (AFCC). The properties of the adsorbent were investigated via instrumental analysis involving Fourier Transform Infra-Red (FTIR) and Scanning electron microscopy (SEM). Non-linear modeling involving various degrees of isotherm models were used in the isotherm study. Adaptive neuro-fuzzy inference systems (ANFIS), response surface methodology (RSM), and artificial neural network (ANN) were used to model the BCG removal. The result of the instrumental analysis showed that the properties of the AFCC were enhanced after the acid carbonization process with a surface area of 903.7 m²/g. The modeling and predictive adeptness of the ANFIS, RSM, and ANN was very significant with correlation coefficient (R²) of 0.9984, 0.9865, and 0.9979 with root mean square error (RMSE) of 0.00308, 0.00898, and 0.00351, respectively. Validation of the models’ optimization indicated maximum adsorption capacities of 38.04, 34.41, and 41.94 mg/g for RSM-GA, ANN-GA, and ANFIS-GA, respectively. Freundlich, Khan, and Marczewski-Jaroniec isotherms best described the adsorption isotherm for two-term, three-term, and four-term isotherm modeling respectively. Calculated values of Gibbs free energy change (∆G_max = -7.55 KJ/mol), enthalpy change (∆H = 35.84 KJ/mol), and entropy change (∆S = 130.20 Jmol⁻¹K⁻¹) indicated the adsorption process was spontaneous, endothermic and with increased randomness respectively. The study showed that the low-cost AFCC obtained from agro-waste has desirable adsorbent properties for the treatment of BCG polluted wastewater.