Mechanistic insights and ANN modeling of eriochrome black and basic orange dye adsorption using surface activated algal-manila tamarind seed biomass

Abstract

Acid activation for adsorbent modification aims at enhancing the surface characteristics and adsorption nature. Artificial Neural Network (ANN) enables the prediction and precise modeling of adsorption processes for enhanced pollutant removal through optimization. The current research explores the mechanistic insights and ANN modeling for eriochrome black (EB) and basic orange (BO) dye adsorption using acid activated algal-manila tamarind seed biomass. Characterization analysis of Brunauer-Emmett–Teller (BET) analysis and X-Ray Photoelectron Spectroscopy (XPS) analysis confirmed the adequate surface properties and elements for the dye remediation. BET and XPS analysis revealed the porous diameter to be 2.868 nm with carbon composition of 79.28 %. Batch adsorptive analysis revealed the optimal parameters for dye adsorption to be 1.25 g/L dose, 40 min process time, pH of 5 and 8 for eriochrome black and basic orange dye. Sips model proved to be the best fitting isotherm with maximum adsorption of 253.6 mg/g of eriochrome black and 282.4 mg/g basic orange dye adsorption was observed. ANN Back propagation algorithm with Levenberg-Marquandt (LM) algorithm was observed to be optimal for the ANN model development. The mathematical predictive modeling of adsorption using artificial neural network revealed the correlation coefficient to be 0.9974 for BO dye and 0.9902 for EB dye adsorption using acid activated mixed biomass. The acid activated algal-plant seed biomass can be reused up to 7 consecutive adsorption cycles.