Optimization of methylene blue adsorption using agricultural biomass waste: a comparative study of batch performance via response surface methodology and artificial neural network

Abstract

Methylene Blue Dye Adsorption by Peanut shell. This study aims to investigate the factors influencing the adsorption and to evaluate the potential of waste peanut shells as an effective biosorbent for wastewater treatment. The Methylene Blue (MB) dye was adsorbed out of an aqueous media using a batch technique and a cheap adsorbent made from peanut shells. X-ray diffractometer (XRD), Fourier transformed inferred spectroscopy (FTIR) and Scanning electron microscopy (SEM) analysis were used to ascertain the peanut shell's structure and properties. The effects of different parameters on dye removal performance were examined using a batch system. The outcomes of modeling investigations demonstrated that the best representation of adsorption kinetics and isotherm data was provided by pseudo-second-order kinetics and Langmuir isotherms. Response surface methodology (RSM) based on Box–Behnken design revealed a highly efficient quadratic correlation for optimizing five parameters affecting dye removal, with R2 equal to 97.6% with difference less than 0.2 between R2 adjusted and predicted 95.7% and 91.4% respectively. To estimate the percentage of dye removal, Levenberg–Marquardt (LM) was employed as the training procedure for a feed-forward back propagation neural network (FFBP-NN). The network design 5-25-30-1 was found to have the best layers and neurons after evaluating several layers and neurons.