Research Paper#Computational Materials Science, Crystal Structure Prediction, Machine Learning🔬 ResearchAnalyzed: Jan 3, 2026 08:37
SSCHA-based Evolutionary Crystal Structure Prediction with Quantum Nuclear Motion
Published:Dec 31, 2025 13:17
•1 min read
•ArXiv
Analysis
This paper addresses the challenge of accurate crystal structure prediction (CSP) at finite temperatures, particularly for systems with light atoms where quantum anharmonic effects are significant. It integrates machine-learned interatomic potentials (MLIPs) with the stochastic self-consistent harmonic approximation (SSCHA) to enable evolutionary CSP on the quantum anharmonic free-energy landscape. The study compares two MLIP approaches (active-learning and universal) using LaH10 as a test case, demonstrating the importance of including quantum anharmonicity for accurate stability rankings, especially at high temperatures. This work extends the applicability of CSP to systems where quantum nuclear motion and anharmonicity are dominant, which is a significant advancement.
Key Takeaways
- •Integrates MLIPs with SSCHA for finite-temperature CSP.
- •Compares active-learning and universal MLIP approaches.
- •Highlights the importance of quantum anharmonicity for accurate stability rankings.
- •Extends CSP to systems where quantum nuclear motion and anharmonicity dominate.
Reference
“Including quantum anharmonicity simplifies the free-energy landscape and is essential for correct stability rankings, that is especially important for high-temperature phases that could be missed in classical 0 K CSP.”