Artificial neural network-assisted modelling of methylene blue dye adsorption onto ZIF-8 functionalized fishtail palm seed biomass: Isotherm and performance analysis

This study investigates the adsorption performance of acid-activated fishtail palm seed biomass functionalized with ZIF-8 (ATFPS-ZIF-8) for the efficient removal of methylene blue (MB) dye from aqueous solutions. FTIR characterization confirmed the presence of functional groups responsible for enhanced adsorption activity. Batch adsorption experiments were conducted to evaluate the effects of initial dye concentration, adsorbent dosage, pH, contact time, and temperature on MB dye removal. The highest dye removal efficiency of approximately 95% was achieved at pH 5, adsorbent dosage of 2.5 g/L, and 60 min contact time, whereas adsorption efficiency decreased with increasing initial dye concentration. A slight decline in adsorption performance with increasing temperature indicated the exothermic nature of the adsorption process. Isotherm analysis demonstrated that the adsorption followed the Langmuir model with a maximum monolayer adsorption capacity of 122.69 mg/g, suggesting homogeneous monolayer adsorption. Furthermore, an artificial neural network (ANN) model was developed to predict adsorption performance, exhibiting excellent agreement with experimental results with correlation coefficient values greater than 0.99 and low prediction error. The findings highlight the synergistic effect of fishtail palm seed biomass and ZIF-8 in producing a sustainable, low-cost, and highly efficient adsorbent for dye contaminated wastewater treatment.