The Elements of Innovation Discovered
The Elements of Innovation Discovered
Metal Tech News - July 7, 2025
A team of scientists is leveraging machine learning to overcome the painstaking process of developing coatings that can lower the temperature of homes, commercial buildings, and urban landscapes.
The researchers tested their meta-emitter materials by coating a roof exposed to the sun.
The meta-emitter roof (center) was significantly cooler than the white (left) and gray (right) roofs.
With some assistance from machine learning and AI, an international group of scientists has developed materials that can lower the temperature of homes, commercial buildings, and urban landscapes without the costs and energy consumption associated with traditional cooling systems.
Known in scientific circles as three-dimensional thermal meta-emitters, these specialized materials leverage the properties of various minerals and metals to selectively emit heat, making them ideal for enhancing energy efficiency through more precise cooling and heating.
Researchers from the University of Texas at Austin, Shanghai Jiao Tong University, National University of Singapore, and Umea University in Sweden leveraged a new machine learning-based approach to develop more than 1,500 complex thermal meta-emitters.
"Our machine learning framework represents a significant leap forward in the design of thermal meta-emitters," said Yuebing Zheng, professor in the Cockrell School of Engineering at the University of Texas and co-leader of the study published in Nature. "By automating the process and expanding the design space, we can create materials with superior performance that were previously unimaginable."
To demonstrate the effectiveness of these AI-designed materials, the team conducted real-world testing under the sun.
To accomplish this, they painted the roof of one building with meta-emitter material and adjacent buildings with white and grey commercial paints, after four hours of midday sun, the meta-emitter-coated roof temperature was 5 degrees Celsius (9 degrees Fahrenheit) cooler than the white roof and 20 degrees Celsius (36 degrees Fahrenheit) cooler than the grey.
The researchers estimate that this level of cooling could save the equivalent of 15,800 kilowatts per year to keep a 100-unit apartment building cool in a hot climate like Rio de Janeiro or Bangkok.
The researchers tested their meta-emitter materials by coating a roof exposed to the sun.
The meta-emitter roof (center) was significantly cooler than the white (left) and gray (right) roofs.
The scientific team says the cooling potential of the thermal meta-emitters they have developed goes far beyond just lowering energy consumption and power bills associated with cooling homes and offices.
Materials developed to reflect specific wavelengths of sunlight and heat could help reduce the temperature in cities by mitigating the urban heat island effect, where large metropolitan areas have higher temperatures due to a lack of vegetation and high levels of concrete.
These thermal meta-emitters could make everyday living a little cooler. The researchers say these materials could be integrated into textiles and fabrics to improve cooling technologies embedded in clothing and outdoor equipment. And cars with exterior and interior coatings could be much more comfortable to get into after sitting in the afternoon sun.
The same type of materials that could make living on Earth a bit cooler could also be used to manage the temperature of spacecraft by reflecting solar radiation and emitting heat efficiently.
The cooling potential of thermal meta-emitters has been held back by the painstakingly slow traditional methods of developing them, according to the international team of material scientists.
"Traditionally, designing these materials has been slow and labor-intensive, relying on trial-and-error methods," said Zheng. "This approach often leads to suboptimal designs and limits the ability to create materials with the necessary properties to be effective."
Machine learning has helped overcome the complexities that come with the 3D hierarchical structure behind efficient meta-emitters.
"Machine learning may not be the solution to everything, but the unique spectral requirements of thermal management make it particularly suitable for designing high-performance thermal emitters," said Kan Yao, a co-author of the thermal meta-emitters study and a research fellow in Zheng's group.
Key materials evaluated for their thermal properties include:
• Silicon carbide and aluminum nitride: These dielectric materials, which do not conduct electricity but can store electrical energy, were selected for their ability to convert thermal energy into electricity.
• Various metals: Gold, tungsten, rhodium, tantalum, molybdenum, niobium, chromium, and platinum were considered as metal covers for the thermal meta-emitters.
• Titanium nitride and aluminum zinc oxide: These materials were chosen for their suitability in high-temperature applications.
• Hafnium dioxide and tungsten: Multilayers of these materials were used to create metamaterial emitter structures.
• Platinum and alumina: A multilayer structure made from these materials was developed as a metamaterial for thermal meta-emitters.
• Liquid metal and aluminum nitride: This combination was used to create a "thermal interface material" that is optimal for cooling applications, including data centers.
The most effective material or combination of materials varies depending on the specific application and desired characteristics of the thermal meta-emitter.
With the computing power of machine learning, the international team of scientists has already developed over 1,500 advanced materials that could lead to a cooler, more energy-efficient future on Earth and beyond.
With more than 17 years of covering mining, Shane is renowned for his insights and in-depth analysis of mining, mineral exploration, and technology metals.
Reader Comments(0)