A new cooling technology developed by a team at The University of Texas at Austin could significantly impact heat management in electronic devices, ranging from semiconductors to large data centers. The researchers have created a "thermal interface material" that efficiently removes heat from high-powered electronics, potentially reducing the need for extensive cooling systems. This innovative material, composed of liquid metal and aluminum nitride, surpasses current commercial materials in conducting heat.
Guihua Yu, a professor at the Cockrell School of Engineering’s Walker Department of Mechanical Engineering and Texas Materials Institute, highlighted the growing power consumption of cooling infrastructure in energy-intensive data centers. He stated, “The power consumption of cooling infrastructure for energy-intensive data centers and other large electronic systems is skyrocketing. That trend isn’t dissipating anytime soon, so it’s critical to develop new ways, like the material we’ve created, for efficient and sustainable cooling of devices operating at kilowatt levels and even higher power.”
Cooling accounts for about 40% of data center energy usage or 8 terawatt-hours annually. The researchers estimate their technology could reduce this requirement by 13%, translating to a 5% reduction in overall data center energy usage if implemented industry-wide. This advancement also supports significant growth in processing power.
Published in Nature Nanotechnology, this discovery is part of ongoing efforts to harness thermal interface materials' potential. These materials aim to dissipate heat generated by electronic devices more effectively than current solutions.
Kai Wu, lead author in Yu’s lab, commented on the breakthrough: “This breakthrough brings us closer to achieving the ideal performance predicted by theory, enabling more sustainable cooling solutions for high-power electronics.” Wu added that their material could facilitate sustainable cooling across various applications from data centers to aerospace.
The demand for data centers is expected to rise due to artificial intelligence's rapid growth and technological proliferation. Goldman Sachs projected earlier this year that data center power demand would increase by 160% by 2030. AI alone may contribute an additional 200 terawatt-hours per year between 2023 and 2030.
The new material was developed using mechanochemistry—a process allowing precise mixing of liquid metal and aluminum nitride—creating gradient interfaces that enhance heat transfer efficiency. Initial tests on small lab-scale devices have been conducted successfully; efforts are underway to scale up production and prepare samples for testing with industry partners.
The research team includes Chuxen Lei from UT’s Materials Science and Engineering program alongside collaborators Zhengli Dou, Shibo Deng, Die Wu, Bin Zhang, Runlai Li, Yongzheng Zhang and Quiang Fu from Sichuan University as well as Haobo Yang from Huazhong University of Science and Technology.