Heavy Metal Pollution Remediation Using Green Coconut Shell: Eco-Friendly Natural Adsorbent with Random Forest Machine Learning Analysis

Abstract

The ecosystem has been deeply disrupted by societal advancements, leading to significant environmental degradation with water pollution being one of the most evident impacts. This study focuses on addressing heavy metal pollution in water bodies, a critical environmental challenge that threatens both aquatic ecosystems and human health. Wastewater from numerous industries including electroplating, leather processing, tannery operations, textile manufacturing, pigment production, paint manufacturing, wood processing, petroleum refining, and photographic film production contain substantial quantities of heavy metals in their effluents. The conventional methods for treating heavy metals are expensive, making adsorption a more viable and sustainable alternative for removing heavy metals from wastewater. In this comprehensive study, green coconut shells were utilized as a bio-adsorbent for heavy metal removal using the column adsorption method. The research demonstrates that optimal conditions of 30 mL/min flow rate and 5 cm bed height are most effective for heavy metal removal. The methodology proved capable of effectively managing real samples from electroplating industries, demonstrating practical applicability. Additionally, a Random Forest regression model was developed to predict heavy metal removal efficiency based on operational parameters (bed height, flow rate, and metal ion type), achieving an R² score 0.893 and providing insights into feature importance, with metal ion type being the most significant predictor followed by flow rate and bed height. This machine learning approach enhances process optimization and provides a predictive framework for scaling up bio-adsorption systems.

Major findings: The multiple element breakthrough curves revealed the following order of adsorption capacity: Pb²⁺ > Zn²⁺ > Ni²⁺ > Cd²⁺ > Cu²⁺.