Optimization and performance prediction of carbon dioxide adsorption on chitosan/activated carbon/epichlorohydrin composite materials using Box–Behnken design and artificial neural network approaches

Abstract

Spent coffee grounds (SCGs) can be used as biomass to synthesize activated carbon (AC) through physical carbonization and chemical activation. Epichlorohydrin (EP) was used to create the chitosan (CS) and AC biopolymer composites via emulsion crosslinking. The main goal of this research is to boost the efficiency of CS/AC/EP composite materials for carbon dioxide (CO 2 ) capture by adsorption. The impact of CS content, AC concentration, and EP quantity on CO 2 removal was studied applying the Box–Behnken design (BBD)-based response surface methodology (RSM) and artificial neural network (ANN)-based artificial intelligence (AI) models. The conditions for the adsorption process were optimized to forecast the maximum CO 2 adsorption utilizing BBD and ANN approaches. Optimal process parameters of 15.11 g CS content, 38.95 %w/w AC concentration, and 7.16 g EP quantity resulted in a CO 2 adsorbed of approximately 7.62 cm 3 /g. The coefficient of determination (R 2 ) for the BBD model was 0.9995, while the correlation coefficient (R) for the ANN model was 0.9992. The CO 2 adsorption efficiency of adsorbents is enhanced by increasing the amounts of AC and EP. This study provides a technique for predicting and improving CO 2 capture through the development of porous polymer composite beads (CBs) with a high CO 2 adsorption capacity.