Multivariate modeling via artificial neural network applied to enhance methylene blue sorption using graphene-like carbon material prepared from edible sugar

摘要

Graphene-like carbon (GLC) material was facile synthesized from edible sugar by a thermal dehydration method, used an adsorbent for methylene blue (MB) removal. The purity and physicochemical properties, including surface morphology, textural property, surface elemental composition and nanostructure of as synthesized GLC was investigated by microscopy and spectroscopy techniques. These results confirmed the formation of nano size (50-100) aggregated plate GLC by showing high specific surface area, 674.593 m(2)/g and 0.278 cm(3)/g pore volume. The adsorptive removal of MB onto the GLC increased with increases of the dosages of adsorbent and the pH of the solution; however, as the initial concentration of MB was increased, its removal efficiency was decreased. From batch studies, initial concentration of 10 mg/L, pH of 8 and dosage of 4.0 g/L were found to be the optimum experimental conditions for maximum amount of MB removal. The optimized isotherm parameters were evaluated by a differential evaluation optimization (DEO) approach suggesting that the Langmuir model better describe the MB adsorption. This result indicates the adsorption process is a monolayer adsorption on homogeneous surface. The kinetic study demonstrated that the adsorption of dye onto GLC followed the pseudo second-order kinetic. Further, the adsorption process variables were optimized using multivariate modeling via artificial neural network (ANN). The maximum adsorption capacity (qm) of GLC for MB is around 20 mg/g. This may attributed to the high surface area of GLC and due to multiple adsorption mechanisms, including pore filling, and electrostatic interactions between MB and GLC. The overall results demonstrate the suitability of GLC for organic dye, MB removal from water. In addition, the present study confirms the viability and quantifiability of the use of GLC by comparison with other graphene/carbon based adsorbents for MB removal. (C) 2018 Elsevier B.V. All rights reserved.