A Review on Phase Change Materials for Sustainability Applications by Leveraging Machine Learning

Abstract

Phase change materials (PCMs) have been envisioned for thermal energy storage (TES) and thermal management applications (TMAs), such as supplemental cooling for air-cooled condensers in power plants (to obviate water usage), electronics cooling (to reduce the environmental footprint of data centers), and buildings. In recent reports, machine learning (ML) techniques have been deployed to improve the sustainability, performance, resilience, robustness, and reliability of TES platforms that use PCMs by leveraging the Cold Finger Technique (CFT) to avoid supercooling (since supercooling can degrade the effectiveness and reliability of TES). Recent studies have shown that reliability of PCMs can be enhanced using additives, such as nucleators and gelling agents, including for organic (paraffin wax) and inorganic (e.g., salt hydrates and eutectics) PCMs. Additionally, material compatibility studies for PCMs with different metals and alloys have also garnered significant attention. Long-term studies for demonstrating the material stability and reliability of candidate PCMs will be summarized in this review book chapter.