EUVIMEDEUVIMED entdecken
Crossref

Machine-Learned Potential–Driven Insights into Li₂S–P₂S₅ Glass Electrolytes and Li₆PS₅Cl Interfacial Chemistry for Solid-State Batteries

Byungju Lee, Chihun Kim, Sojeong Yang

ECS Meeting Abstracts · 07.07.2026 · MA2026-01 (6) · 700-700

DOI: 10.1149/ma2026-016700mtgabs ↗

Kurz gefasst

Fachartikel in der Zeitschrift ECS Meeting Abstracts, erschienen am 07.07.2026. Verfasst von Byungju Lee und 2 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Machine-learned interatomic potentials (MLPs) provide an avenue for exploring complex structural motifs and chemical reactions in solid-state battery materials that are challenging to access via purely first-principles approaches. In this presentation, I introduce our recent studies employing MLP-based simulations to investigate two key classes of sulfide solid-state electrolytes (SSEs): Li₂S–P₂S₅ glass for composition and density optimization, and Li₆PS₅Cl argyrodite for interfacial and surface reactivity. For the Li₂S–P₂S₅ glass system , we develop an MLP-enabled framework that systematically maps the relationships among composition, densification behavior, structural motifs, and ionic transport. By performing extensive sampling of glass structures across a wide composition range, we identify optimal glass formulations and density regimes that enhance Li-ion mobility and thermodynamic stability. This approach allows efficient navigation of the broad design space inherent to multicomponent amorphous SSEs. In complement, we investigate Li metal | Li₆PS₅Cl interfaces and the surface chemistry of Li₆PS₅Cl under oxygen and moisture exposure. MLP-accelerated molecular dynamics reveals the initial reaction pathways, interphase formation, and structural evolution at the Li–solid interface, providing insights into stability and ion-transport implications. Additionally, we analyze degradation reactions triggered by O₂/H₂O on Li₆PS₅Cl surfaces, elucidating mechanisms relevant to processing conditions and long-term performance. Together, these results demonstrate how machine-learned potentials enable multiscale understanding—from bulk amorphous structure to interfacial chemistry—crucial for the design of robust and high-conductivity sulfide solid-state electrolytes.

Zitieren

APA

Lee, B., Kim, C., & Yang, S. (2026). Machine-Learned Potential–Driven Insights into Li₂S–P₂S₅ Glass Electrolytes and Li₆PS₅Cl Interfacial Chemistry for Solid-State Batteries. ECS Meeting Abstracts, MA2026-01(6), 700-700. https://doi.org/10.1149/ma2026-016700mtgabs

Vancouver

Lee B, Kim C, Yang S. Machine-Learned Potential–Driven Insights into Li₂S–P₂S₅ Glass Electrolytes and Li₆PS₅Cl Interfacial Chemistry for Solid-State Batteries. ECS Meeting Abstracts. 2026;MA2026-01(6):700-700. doi:10.1149/ma2026-016700mtgabs

BibTeX
@article{Lee2026,
  author = {Lee, Byungju and Kim, Chihun and Yang, Sojeong},
  title = {Machine-Learned Potential–Driven Insights into Li₂S–P₂S₅ Glass Electrolytes and Li₆PS₅Cl Interfacial Chemistry for Solid-State Batteries},
  journal = {ECS Meeting Abstracts},
  year = {2026},
  volume = {MA2026-01},
  number = {6},
  pages = {700--700},
  doi = {10.1149/ma2026-016700mtgabs}
}
Kostenlos starten →