Application of artificial neural networks and response surface methodology for dye removal using a novel adsorbent

摘要

Experiments were conducted to study the efficiency of green carbon prepared from natural source for removal of a basic dye Methylene Blue (MB). The effect of various process parameters such as temperature, initial pH, contact time, adsorbent dosage and initial dye concentration of the solution were studied by running batch experiments in Erlenmeyer flasks. Modeling equation was developed to study the effects of all the parameters on dye removal by Response Surface Methodology to optimize the process parameters. ANOVA analysis was also studied to know the interaction effect of dye and adsorbent. This experiment revealed that the adsorbent exhibited high adsorption capacities and this adsorption capacity was affected by the changes in temperature, initial pH, contact time, adsorbent dosage and initial dye concentration of the solution. The results obtained were also modeled by using Artificial Neural Networks (ANN). High values of correlation coefficients indicated the best fit of experimental results with that of values obtained from modeling. From these studies, it may be concluded that green carbon adsorbent prepared is efficient and economical for Methylene blue removal from aqueous solutions.