Advances in generative artificial intelligence (AI), high-performance CPUs and GPUs, and advanced semiconductor packaging technologies have made internal heat generation in electronic devices a critical technical challenge. In particular, AI semiconductors generate enormous heat during intensive computation, which can degrade performance and shorten device lifetime. This has increased the need for materials that maintain electrical insulation while efficiently dissipating heat.

Against this backdrop, Jeonbuk National University (JBNU) Department of Organic Materials and Textile Engineering Professor Seong-ryun Kim's team and Dr. Heon-su Lee's team at the Jeonbuk Branch of the Korea Institute of Science and Technology (KIST) have jointly developed an insulating thermally conductive composite material suitable for AI semiconductors and high-performance electronic devices, drawing attention.

The research teams proposed a differentiated approach by combining plasma-based hexagonal boron nitride (hBN) surface modification with a segregated-structure composite design. Unlike conventional methods that add large amounts of thermally conductive fillers, they precisely designed heat-transfer pathways within the composite, enabling an efficient thermal conduction network even at low filler loadings.

In particular, they enhanced hBN surface properties through a plasma process and created a segregated structure by controlling polymer particle size to maximize heat flow. They also established an integrated analysis platform that combines 3D micro-CT analysis, the 'Excluded-volume-informed Foygel model', and finite element analysis to elucidate heat-transfer mechanisms and predict thermal performance.

Results confirmed excellent thermal dissipation performance under actual GPU and CPU operating conditions, demonstrating the material's practical potential as an insulating thermal interface material (TIM) for AI semiconductors.

Professor Seong-ryun Kim said, 'It would not be an exaggeration to say that competitiveness in AI semiconductors depends on heat-control technology,' and added, 'This research, which integrates interfacial control, structural design, and theoretical modeling, will provide an important foundation for developing next-generation thermal management materials.'

First author Gyun-young Yu, an integrated master's–doctoral degree student, said, 'Precisely designing the internal heat-transfer structure of the composite is key,' and added, 'I hope this study will contribute to the structural design and theoretical modeling of insulating thermal-dissipation materials.'

The study, titled 'Structure-aware thermal percolation modeling of plasma-functionalized hBN composites with optimized segregated networks,' was published in the international materials engineering journal Composites Part B: Engineering (Impact Factor 14.2, top 0.3% in JCR).

Meanwhile, the research was conducted with support from the Ministry of Trade, Industry and Energy and the Korea Institute for Advancement of Technology under the 'K-Carbon Flagship Technology Development Program.'