高功率芯片、5G基站以及电动汽车这些现代科技的心脏，都在发烫。散热，成了制约其性能与寿命的瓶颈。铜/金刚石复合材料，因兼具有超高的理论热导率以及可调控的热膨胀系数，被寄予厚望。然而，现实却令人沮丧：实际导热性能往往不到理论值的一半。问题出在哪？答案藏在那看不见的原子界面里。铜与金刚石“性格不合”，它们的原子振动方式（即“声子”）严重不匹配，导致热量在穿越界面时遭遇巨大阻力；同时，两者化学亲和力弱，界面结合松散，进一步阻碍热流。为改善这一状况，工程师常在中间插入一层“媒人”材料（如Cr、Mo、Ti、W、TiC和WC等）。但哪种“媒人”最称职？温度、压力、晶体方向如何影响效果？这些问题长期缺乏原子尺度的清晰解析。
Fig. 1 | Schematic diagram of NEP framework and performance of machine learning.
近日，武汉大学集成电路学院刘胜院士、吴改副研究员与沈威副教授团队在npj Computational Materials发表突破性工作，构建了一套高精度机器学习原子力场（Neuroevolution Potential, NEP），实现了多种中间层材料（Cr、Mo、Ti、W、TiC、WC）以及不同温度、应变和晶体取下铜/金刚石异质结构的大规模非平衡分子动力学模拟。这项研究如同给界面装上了“原子级热成像仪”，首次系统揭示了热阻背后的微观机制。
Fig. 2 | Cu/diamond heterostructure models with different interlayer materials and the corresponding predictions of interfacial heat transfer.
作者发现，一个优秀的中间层，关键在于能否充当“声子桥”，即有效缩小铜与金刚石高效传热声子之间的频率差距。钛（Ti）和碳化钛（TiC）表现最佳，因其不仅频率匹配度高，还能提供多条高效声子通道，使界面热阻降低高达42%。更有趣的是，升温反而有助于界面导热，因为温度升高能增强铜与金刚石声子态密度的重叠，从而“润滑”了热流；而沿热流方向施加轻微压缩，不仅能提升声子耦合，还能强化界面化学键，双管齐下降低热阻。尤为引人注目的是，尽管铜和金刚石本身导热没有明显的各向异性，但其界面热阻却明显受到晶体取向影响，这可能源于界面处的非弹性声子散射，该发现为未来界面热管理的取向工程研究提供了新视角。
Fig. 4 | The vibration spectra of the Cu/diamond heterostructures with different interlayer materials.
这项工作不仅为高性能铜/金刚石热管理材料的理论设计提供了清晰的原子级“操作手册”，其使用的机器学习力场方法更具备广泛适用性，可推广至其他异质界面体系，为微电子、能源转换、先进制造等领域的交叉研究注入新思路与新见解。
Fig. 7 | Schematic diagram of uniaxial strain and predictions of the effect of uniaxial strain on thermal transport.
原来，那颗默默承载现代科技的“心脏”，并非天生神力，而是靠界面处无数声子的“默契配合”。而今，科学家终于听懂了它们的“语言”，热管理的下一场革命，或许就从这纳米尺度的对话开始。该文近期发表于npj Computational Materials 11,359(2025)，英文标题与摘要如下，点击左下角“阅读原文”可以自由获取论文PDF。
A comprehensive exploration of thermal transport at Cu/diamond interfaces via machine learning potentials
Zhanpeng Sun #, Hutao Shi #, Yilong Zhu, Rui Li, Xiang Sun, Qijun Wang, Zijun Qi, Lijie Li, Sheng Liu, Wei Shen *, Gai Wu *
The fundamental thermal limitation of pure copper impedes progress in high-power devices, which is becoming more critical with advances in power electronics. The Cu/diamond composite becomes a promising candidate for thermal management due to its excellent theoretical thermal conductivity and customizable coefficient of thermal expansion (CTE). Actually, the thermal conductivity of Cu/diamond composite is much lower than its theoretical value, for which a key bottleneck is interfacial thermal transport at the Cu/diamond interface. However, many atomic-level microscopic mechanisms of heat transport at Cu/diamond interfaces remain poorly understood at present. Especially when different interlayer materials are involved, theoretical studies become extremely complex and challenging. In this work, a machine learning potential for comprehensive simulations of thermal transport at Cu/diamond interfaces has been successfully constructed. The effects of key factors, such as interlayer material, temperature, strain, and crystal orientation, on heat transport at Cu/diamond interfaces have been studied. Furthermore, the underlying mechanisms are thoroughly analyzed and discussed. Finally, the insightful strategies are proposed to optimize and enhance the thermal properties of Cu/diamond interfaces. These advancements can lay a foundation and pave the way for further investigations into interfacial thermal transport at Cu/diamond interfaces as well as in other structures containing interlayer materials.