Direct discovery · Materials science · Chemistry
AI and cloud HPC narrow battery-electrolyte candidates for synthesis
Solid-electrolyte candidates narrowed from a large computed search and followed by synthesis and conductivity characterization.
Summary
The JACS paper reports a Microsoft and PNNL workflow combining machine-learning models, physics-based filters and cloud high-performance computing to screen more than 32 million candidate materials. It narrowed the search to solid-electrolyte candidates, followed by synthesis and conductivity characterization of NaxLi3-xYCl6 compositions.
AI role
Combined machine-learning filters, physics-based screening, and cloud HPC to triage battery-electrolyte candidates.
Narrative role
This is a good scale event: AI and HPC make a search space visible before human synthesis and characterization.
Caveat
The discovered compositions are research candidates, not proven battery technologies.