Eco-Engineered Biopolymer-Clay Composite for Phosphate IonRemoval: Synergistic Insights from Statistical and AI Modeling.

This research aims to synthesize a novel hydrogel bio-composite based on natural clay, sodium alginate (Na-AL), and iota-carrageenan as adsorbents to remove phosphate ions from aqueous solutions. The adsorbents were characterized by a variety of techniques, such as Fourier-transform infrared (FTIR) spectroscopy, scanning electron microscopy coupled with energy dispersive X-rays (SEM-EDX), and the determination of point zero charge (PZC). This research investigated how the adsorption process is influenced by parameters such as adsorbent dose, contact time, solution pH, and temperature. In this study, we used four isotherms and four kinetic models to investigate phosphate ion removal on the prepared bio-composite. The results showed that the second-order kinetic (PSO) model is the best model for describing the adsorption process. The findings demonstrate that the R values are highly significant in both the Langmuir and Freundlich models (very close to 1). This suggests that Langmuir and Freundlich models, with a diversity of adsorption sites, promote the adsorption of phosphate ions. The maximum adsorbed amounts of phosphate ions by the bio-composite used were 140.84 mg/g for HPO ions and 105.26 mg/g for HPO ions from the batch system. The positive ∆H° confirms the endothermic and physical nature of adsorption, in agreement with experimental results. Negative ∆G° values indicate spontaneity, while the positive ∆S° reflects increased disorder at the solid-liquid interface during phosphate uptake. The main parameters, including adsorbent dosage (mg), contact time (min), and initial concentration (mg/L), were tuned using the Box-Behnken design of the response surface methodology (BBD-RSM) to achieve the optimum conditions. The reliability of the constructed models is demonstrated by their high correlation coefficients (R). An R value of 0.9714 suggests that the model explains 97.14% of the variability in adsorption efficiency (%), which reflects its strong predictive capability and reliability. Finally, the adsorption behavior of phosphate ions on the prepared bio-composite beads was analyzed using an artificial neural network (ANN) to predict the process efficiency. The ANN model accurately predicted the adsorption of phosphate ions onto the bio-composite, with a strong correlation (R = 0.974) between the predicted and experimental results.