About this Abstract |
| Meeting |
MS&T25: Materials Science & Technology
|
| Symposium
|
Advances in Refractory High Entropy Alloys and Ceramics
|
| Presentation Title |
ULTERA Data Ecosystem for Compositionally Complex Materials |
| Author(s) |
Zi-Kui Liu, Adam Krajewski |
| On-Site Speaker (Planned) |
Zi-Kui Liu |
| Abstract Scope |
Scientific understanding of advanced alloys requires both knowledge and support of reliable experimental datasets. In recent years, digital literature access is putting more emphasis on data extent and reusability. To address this challenge in the domain of compositionally complex materials (CCMs), we have been developing the ULTERA Data Ecosystem (ultera.org) under ARPA-E ULTIMATE program. ULTERA has been designed to include elaborate custom-built processing pipelines allowing automatic curation (pyqalloy.ultera.org) and homogenization of literature data, while making contributing new data simple (spreadsheets) and governable, with everything handled on GitHub akin to open-source software development. Such extensible structure has allowed us to grow data extracted from over 800 HEA literature publications while continuously deploying new ML property models. In this presentation, we also showcase how we use inverse-design (10.20517/JMI.2021.05) to propose new alloys and compositional graphs constructed by our new algorithm (10.1038/s44335-024-00012-2) to explore complex HEA design spaces, including navigating eutectic manifolds. |