Incorporation of surfactant-intercalated graphene oxide into zeolitic imidazolate framework-9 for effective heavy metal and dye removal from wastewater: RSM optimization and ANN modeling

Wastewater pollution from toxic inorganic and organic contaminants poses serious environmental and health threats and must be removed. In response, graphene oxide/zeolitic imidazolate frameworks composites have gained attention for combining useful features of both materials in wastewater treatment. In this study, a novel cetyltrimethylammonium bromide/graphene oxide/zeolitic imidazolate framework-9 (CTAB@GO/ZIF-9) nanocomposite was synthesized and employed for the adsorption of lead (Pb²⁺) and methyl orange (MO) from aqueous media. The adsorption process was optimized using response surface methodology (RSM), and artificial neural networks (ANN) were applied to improve the prediction accuracy. RSM identified adsorbent mass, pollutant concentration, and temperature as significant parameters affecting adsorption capacity. ANN models outperformed RSM in prediction, achieving R² values of 0.975 for Pb²⁺ and 0.991 for MO. The isotherm study showed that Pb²⁺ adsorption followed the Freundlich (R² = 0.907) and Redlich–Peterson (R² = 0.908) models, while MO fitted the Langmuir (R² = 0.995) and Redlich–Peterson (R² = 0.997) models, with maximum adsorption capacities of 1139.7 mg/g and 1195 mg/g, respectively. The kinetic findings indicated that the Avrami model best described both adsorption processes, with R² values of 0.999 for Pb²⁺ and 0.995 for MO, suggesting a multi-step mechanism. The thermodynamic analysis confirmed the spontaneity of adsorption: Pb²⁺ removal was endothermic and entropy-driven, while MO adsorption was exothermic with decreased entropy. Possible adsorption mechanisms involved coordination interactions for Pb²⁺ and electrostatic and π–π interactions for MO. These results demonstrate the high efficiency of CTAB@GO/ZIF-9 as a versatile adsorbent for various pollutants in wastewater treatment.