| About this Abstract |
| Meeting |
2023 TMS Annual Meeting & Exhibition
|
| Symposium
|
Computational Discovery and Design of Materials
|
| Presentation Title |
Machine Learning Assisted Discovery of Composite Solid-state Electrolytes in Context of Li-ion Batteries |
| Author(s) |
Hasan Muhammad Sayeed, Taylor D. Sparks |
| On-Site Speaker (Planned) |
Hasan Muhammad Sayeed |
| Abstract Scope |
Solid-state lithium-ion batteries (SSLB) are considered next-generation energy storage devices for superior energy density and safety compared to their counterparts. Solid-state electrolytes (SSE) are the critical component of SSLBs. There are different types of SSEs, among which composite solid-state electrolytes (CSSEs) combines organic polymers and inorganic ceramics and can provide the advantages of all the single-phase electrolytes while solving their shortcomings. To use CSSEs in SSLBs, electrolyte materials must satisfy multiple requirements such as high ionic but low electronic conductivity, structural and electrochemical stability of interfaces and so on at once. We used Bayesian Optimization to search through vast potential combination space of CSSEs while optimizing for desirable properties of SSLBs. We generated random compositions at each iteration and predicted ionic and electronic conductivity. High performing samples were synthesized and characterized for validation. This data was then fed back into the model as training data and the process was repeated. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Energy Conversion and Storage, Computational Materials Science & Engineering, Machine Learning |