About this Abstract |
| Meeting |
MS&T25: Materials Science & Technology
|
| Symposium
|
Thermodynamics of Materials in Extreme Environments
|
| Presentation Title |
Inferring Structure from Raman Spectroscopy and Connecting It to the Macroscopic Behavior of Molten ThCl4 |
| Author(s) |
Vyacheslav Bryantsev, Luke Gibson, Rajni Chahal |
| On-Site Speaker (Planned) |
Vyacheslav Bryantsev |
| Abstract Scope |
Understanding the local structure of molten salts is critical for developing predictive models. In the case of molten ThCl4, the local structure is not directly accessible and must be inferred from Raman spectra. Therefore, it is essential to develop computational tools capable of modeling spectroscopic observables. We employ machine learning-accelerated molecular dynamics simulations to uncover the structure and thermophysical behavior of molten ThCl4. Benchmarking against experimental densities and Raman spectroscopy, we construct machine learning interatomic potentials to assess structural changes across different DFT methods. Simulated Raman spectra highlight key experimental features and link local coordination environments of Th4+ ions to specific vibrational signatures. Knowing key structural features, we can now relate them to various thermophysical properties such as density, viscosity, ionic diffusivity, and heat capacity. This work provides a direct connection between atomic-scale structure and macroscopic behavior, facilitating the design and evaluation of next-generation molten salt reactor systems. |