Thermal Interface Materials with Both High Through-Plane Thermal Conductivity and Excellent Elastic Compliance

Junwei Li,Yuexing Zhang,Ting Liang,Xue Bai,Yunsong Pang,Xiangliang Zeng,Qinghua Hu,Wendian Tu,Zhenqiang Ye,Guoping Du,Rong Sun,Xiaoliang Zeng
DOI: https://doi.org/10.1021/acs.chemmater.1c03275
IF: 10.508
2021-01-01
Chemistry of Materials
Abstract:With the development of microprocessors toward higher power, larger chip, and higher frequency, heat dissipation is one of the central issues. Thermal interface materials (TIMs), which are used between the chip and the heat spreader and between the heat spreader and the heat sink, play an increasingly important role in microprocessor cooling. Currently, most of the research has primarily dealt with understanding the thermal conductivity of TIMs. For thermal design, elastic compliance is also important because excellent elastic compliance can reduce thermal contact resistance and relieve the warpage failure caused by stress concentration. However, high thermal conductivity and excellent elastic compliance are usually mutually exclusive in TIMs. Herein, we report a TIM made from vertically oriented graphite and polybutadiene that shows high through-plane thermal conductivity of 64.90 W.m(-1).K-1, excellent elastic compliance with only 93 kPa stress at 50% compressive strain similar to soft biological tissues, and outstanding compression resilience performance (storage modulus 220 kPa and mechanical loss factor 0.226). These excellent properties result from the vertical orientation of graphite films in polybutadiene, strong interfacial strength between graphite films and polybutadiene, and the minimized negative impact of graphite on the intrinsic mechanical properties of polybutadiene by means of cross-stacking techniques. The optimal TIM is applied in CPU microprocessor cooling and exhibits superior heat dissipation capability, by up to 158 degrees C reduction of chip temperature comparing with polybutadiene. This work provides a high-performance TIM to meet specific requirements for highperformance computing, such as GPU, AI computing, and cloud computing.
What problem does this paper attempt to address?