Experimental and Machine Learning Modelling of Ni(II) Ion Adsorption onto Guar Gum: Artificial Neural Network (ANN)...

Area of Science:

• Environmental Chemistry
• Materials Science
• Biopolymer Engineering

Background:

• Heavy metal contamination, particularly Ni(II) ions, poses significant risks to aquatic ecosystems and human health.
• Developing efficient and sustainable adsorbents is crucial for effective water remediation.
• Guar gum, a natural biopolymer, offers potential as a low-cost, eco-friendly adsorbent material.

Purpose of the Study:

• To develop and evaluate a guar gum-based adsorbent for Ni(II) ion removal from aqueous solutions.
• To optimize adsorption parameters using experimental design and machine learning.
• To investigate the adsorption mechanism and thermodynamic properties.

Main Methods:

• Characterization of the guar gum adsorbent using FTIR, SEM, XRD, TGA, and BET analyses.
• Batch adsorption experiments to determine optimal pH, dosage, contact time, temperature, and initial Ni(II) concentration.
• Application of artificial neural network (ANN) and k-nearest neighbor (KNN) models for predicting Ni(II) removal efficiency.

Main Results:

• Maximum Ni(II) removal efficiency of 97% achieved at pH 6.0, 1.0 g L⁻¹, 60 min contact time, and 50 mg L⁻¹ initial concentration.
• Adsorption kinetics followed pseudo-second-order, and equilibrium data fitted the Langmuir isotherm model.
• Thermodynamic analysis indicated a spontaneous, endothermic, and physisorption process.
• ANN model (R² = 0.97) demonstrated higher prediction accuracy than KNN (R² = 0.95).

Conclusions:

• The developed guar gum-based adsorbent is highly effective for Ni(II) ion removal.
• The combined experimental and machine learning approach provides a robust framework for adsorbent optimization.
• This study highlights the potential of biopolymer-based materials for sustainable heavy metal remediation.