Jackfruit seed as low-cost adsorbent for removal of malachite green: artificial neural network and random forest approaches

Abstract

This study focused on the adsorption characteristics of jackfruit seed as an adsorbent for the removal of malachite green (MG) dye from aqueous solution via batch adsorption experiment. The performance of the adsorbent was investigated through several parameters, namely contact time, pH, dye concentration, ionic strength and temperature. Adsorption of MG by jackfruit seed required 2 h to reach equilibrium and was found to be sensitive to the pH medium, but relatively unaffected by ionic strength. The Langmuir, Freundlich and Sips isotherm models were used to describe the adsorption process, while the Lagergren first-order, pseudo-second-order and Weber–Morris intraparticle diffusion models were used to predict the adsorption mechanism. The Freundlich model was found to best fit the experimental data, and the maximum monolayer capacity (from Langmuir model) was determined as 66 mg g−1. The kinetics mechanism followed pseudo-second-order model and intraparticle diffusion was found not to be the rate-determining step. In addition, machine learning models such as artificial neural network and random forest were used as predictive models for the adsorption process where high R values of 0.966 and 0.981, respectively, and low RMSE of 0.048 and 0.038 were obtained. This study concludes that jackfruit seed has the potential to be used as a low-cost adsorbent for the removal of toxic MG dye.