AI-crafted coating cools buildings by 36°F, slashes AC use, bills

Researchers from the University of Texas at Austin, Shanghai Jiao Tong University, the National University of Singapore, and Umea University have developed advanced thermal meta-emitters using machine learning to create over 1,500 novel materials capable of precise heat emission. By automating the design process and exploring complex three-dimensional structures, these materials achieve superior cooling performance previously unattainable through traditional trial-and-error methods. Testing showed that coating a model building with one such material reduced roof temperatures by 5 to 20°C (37 to 68°F) under direct sunlight, potentially saving about 15,800 kilowatt-hours annually in hot climates—significantly cutting air conditioning energy use and costs. Beyond energy savings, the researchers have designed seven classes of meta-emitters with diverse applications, including mitigating urban heat island effects by reflecting sunlight and emitting heat at specific wavelengths, and managing spacecraft temperatures in space. These materials also hold promise for everyday uses such as cooling fabrics, outdoor gear, and vehicle surfaces