Green energy-compatible cadmium (II) biosorption from wastewater using Codium decorticatum: Environmental impact, adsorption dynamics, and neural network modeling

Abstract

Cadmium (Cd) contamination in water is a critical environmental challenge due to its toxicity and persistence, necessitating sustainable and cost-effective remediation strategies. This study presents a novel approach by utilizing Codium decorticatum derived biosorbent (GS-CD) for Cd (II) removal, integrating both experimental biosorption studies and artificial neural network (ANN) modeling to enhance predictive accuracy and process optimization. The biosorbent’s structural and chemical properties were extensively characterized using proximate analysis, SEM, EDX, FTIR, and XRD. Batch adsorption experiments demonstrated a remarkable 98.38 % removal efficiency under optimal conditions: pH 5, biosorbent dosage of 2.5 g/L, contact time of 40 min, and temperature of 303 K at an initial Cd (II) concentration of 20 mg/L. Isotherm modeling confirmed monolayer adsorption with a Langmuir adsorption capacity of 138.26 mg/g (R² = 0.9709), while kinetic studies indicated a pseudo-second-order mechanism (R² = 0.9399), suggesting chemisorption as the dominant process. The thermodynamic analysis further revealed that the adsorption process is spontaneous and exothermic, favouring lower temperatures. ANN modeling provided precise predictions, optimizing biosorption conditions beyond conventional approaches and achieving excellent agreement with experimental data (R > 0.99). The study hypothesizes that ANN-based predictions can optimize biosorption conditions, improving efficiency and scalability for sustainable heavy metal remediation. GS-CD retained over 67 % efficiency after seven cycles, demonstrating durability despite active site depletion. These findings highlight GS-CD as an effective, sustainable solution for heavy metal remediation.