Pusan National University Team Develops High-Performance Composite Material Using AI to Cool Electronic Devices

Thermal Conductivity of High Heat Dissipation Material Doubled
Potential Applications in Electronics, Automotive, and Aerospace Industries

A research team at Pusan National University has developed a polymer composite material that can dramatically reduce the heat generated by electronic devices by integrating artificial intelligence (AI) and three-dimensional imaging technology.

The new material more than doubles the heat dissipation performance compared to existing materials, presenting new possibilities for implementing high-efficiency cooling systems in the electronics, automotive, and aerospace industries.

The research team, led by Professor Kim Chaebin and Professor Lee Jaegun of the Department of Applied Chemical Engineering at Pusan National University, and Professor Ahn Hyoseong of the Department of Petrochemical Materials Engineering at Chonnam National University, applied a data-driven engineering approach to precisely analyze the pathways through which heat is transferred within polymer materials, enabling AI to autonomously identify the optimal structure.

From the left, Professor Kim Chaebin, Professor Lee Jaegun, Professor An Hyoseong, student Na Chaeseong, student Shin Sangsu / Provided by Pusan National University

As a result, the thermal conductivity of the composite material, which combines alumina (Al₂O₃) microparticles and silicone rubber (PDMS), reached 6.89 W/m·K-more than twice as high as conventional materials.

The key lies in the AI's ability to search for the optimal combination. Using a Bayesian Optimization algorithm, the team simulated and experimented with hundreds of combinations of particle sizes and mixing ratios, ultimately identifying that particles sized 90μm, 20μm, 3μm, and 0.6μm formed the most densely packed structure. This configuration created efficient pathways for rapid heat transfer, thereby maximizing heat dissipation efficiency.

The researchers utilized 3D X-ray CT (computed tomography) technology to visually examine the internal structure of the material. This allowed them to precisely visualize the connectivity between particles, pore distribution, and interfacial areas, enabling them to track the actual pathways of heat transfer.

The analysis revealed that thermal conductivity can be accurately predicted by three factors: particle volume, curvature, and the interfacial area between the filler and resin.

Professor Kim Chaebin stated, "This study represents the first case in which the random structure of a composite material was interpreted using data-driven methods and AI autonomously designed the optimal structure. It is expected to be widely applicable to high heat dissipation systems such as electric vehicle batteries, satellites, and aircraft."

Professors Kim Chaebin and Lee Jaegun of Pusan National University and Professor Ahn Hyoseong of Chonnam National University served as co-corresponding authors. Na Chaeseong, a master's student at Pusan National University, and Shin Sangsu, a combined master's and doctoral student at Pusan National University, were listed as co-first authors. This research was supported by the Ministry of Science and ICT and the National Research Foundation of Korea.

This content was produced with the assistance of AI translation services.