Experimental and Artificial Intelligence Approaches for the Analysis of Parameters Affecting Zinc Ion Adsorption in Environmental Remediation

Pollution caused by heavy metals in industrial wastewater has become a serious environmental concern, posing significant risk to both human health and ecosystems. Industrial effluents often contain pollutants form natural and anthropogenic sources such as volcanic activity, industrial processes, and vehicle emissions. Various treatment methods have been explored, but many are energy-intensive and costly. Among them adsorption stands out for its high selectivity, ease of operation, and reusability of adsorbents. In this study, a bio-based adsorbent derived from wheat straw was used to remove Zn (II) ion from wastewater. The removal efficiency was examined under varying concentrations, temperatures, and pH levels. Statistical analysis and artificial intelligence (AI) techniques were employed to evaluate the results, with Ker Person’s correlation used to validate the data. The Langmuir isotherm model showed a strong linear relationship (correlation coefficient 0.98 ), confirming effective adsorption. A Random Forest regression model further enhanced prediction accuracy, yielding an R ² value of 0.995 and a mean square error (MSE) of 8.384. The parity and 3D surface plots demonstrated excellent agreement between predicted and experimental outcomes. Overall, wheat straw proved to be a cost-effective and sustainable adsorbent for Zn(II) removal, offering promising potential for wastewater treatment.