One-defect one-potential strategy for accurate machine learning prediction of phonons in defect-containing supercells

Atomic vibrations play a critical role in phonon-assisted electronic transitions at defects in solids. However, accurate phonon calculations in defect-laden systems are often hindered by the high computational cost of large-supercell first-principles calculations. Recently, foundation models, such as universal machine learning interatomic potentials (MLIPs), have emerged as a promising alternative for rapid phonon calculations, but the quantitatively low accuracy restricts its fundamental applicability for high-level defect-phonon calculations, such as nonradiative carrier capture rates. In this paper, we propose a ``one-defect, one-potential'' strategy in which an MLIP is trained on a limited set of perturbed supercells. We demonstrate that this strategy yields phonons with accuracy comparable to density functional theory (DFT), regardless of the supercell size. The predicted accuracy of defect phonons is validated by phonon frequencies, Huang-Rhys factors, and phonon dispersions. Further calculations of photoluminescence spectra and nonradiative capture rates based on this defect-specific model also show good agreement with DFT results, meanwhile reducing the computational expenses by more than an order of magnitude. Our approach provides a practical pathway for studying defect phonons in a ${10}^{4}$-atom large supercell with high accuracy and efficiency.