Investigating desalination mechanism of saltwater using superhydrophobic membranes and artificial intelligence

Desalination plays a vital role in addressing global water scarcity, and this study focuses on its efficiency when superhydrophobic membranes are employed. This research explores the performance of coated materials such as carbon felt, muslin cloth, and coconut shell fibers. These materials, when treated with superhydrophobic coatings, exhibited high water contact angles (WCA) of 171.7°, 167.46°, and 159.21°, respectively, confirming their hydrophobic properties. Fourier-transform infrared spectroscopy (FTIR) validated the presence of hydrophobic functional groups on the surfaces, indicating effective surface modification. This study also investigated the impact of temperature on WCA and water flux. Despite a decrease in WCA with rising temperature, the coated materials retained their hydrophobicity. The dual-layer coating UED+LC1 achieved the highest permeate flux across all tested temperatures, reaching a peak of 15.18 L/m²·h at 90°C. Surface characterization using confocal laser scanning microscopy provided insights into roughness, texture, and the distribution of micro- and nanostructures critical for water repellency. In addition, machine learning models were employed to predict desalination performance metrics such as water flux and production efficiency. Ridge Regression outperformed other models, achieving an R² value of 0.9996 and a Mean Squared Error (MSE) of 0.0094, surpassing Support Vector Regression (SVR) and Decision Tree models. These findings highlight the potential of AI to optimize desalination processes. This research shows that superhydrophobic coatings and AI-driven predictive modeling can significantly improve desalination efficiency, providing a promising answer to worldwide water crisis issues and instilling optimism about the future of water treatment.

Author

• Paranjpe, Nikhil

Identifier

hdl:10057/29170

OAI identifier oai:identifier

oai:soar.wichita.edu:10057/29170