Unveiling thermal transport mechanisms in alumina laser crystal using machine learning potential
Yan Dai, Zhongwei Zhang, Jiamin Quan, Wudi Wang, Chenbo Zhang, Jun Xu, and Jie Chen
Phys. Rev. Research 8, 033238 (2026) - Published 27 August, 2026
Heat dissipation capability of laser gain media is critical for ensuring the performance and stability of high-power laser systems. However, the thermal transport mechanisms in these materials, particularly in rare-earth ion-doped laser crystals under high-power heating conditions, remain largely elusive. In this study, we employ molecular dynamics simulations combined with a neuroevolution machine learning potential to investigate the effects of temperature and Ce doping on thermal transport in alumina. Our simulations reveal a significant reduction in thermal conductivity under uniform elevated temperatures and finite Ce doping concentrations, providing quantitative insight into thermal transport in Ce-doped under conditions relevant to laser operation. Simulation results demonstrate that elevated temperature causes a broad-spectrum suppression of thermal transport, while Ce doping predominantly affects specific low-frequency acoustic phonon modes and a few optical modes. Further analysis uncovers a mode-selective phonon-dopant scattering mechanism, where resonant and polarization-dependent coupling leads to the selective suppression of distinct phonon modes. This mechanism is in sharp contrast with the conventional understanding of phonon-dopant scattering observed in typical crystals. These findings provide valuable insights into the atomic-level mechanisms governing heat dissipation in alumina laser crystal, thus paving the way for the design of laser materials with optimized thermal properties for high-power applications.


