About this Abstract |
| Meeting |
2025 TMS Annual Meeting & Exhibition
|
| Symposium
|
Computational Thermodynamics and Kinetics
|
| Presentation Title |
Predicting Synthesis Outcomes With DFT Calculations and Literature Mining |
| Author(s) |
Anubhav Jain |
| On-Site Speaker (Planned) |
Anubhav Jain |
| Abstract Scope |
This talk will discuss the current ability to predict synthesis outcomes using density functional theory (DFT) calculations and machine learning. I will first describe our collaborative attempt to create a large corpus of inorganic materials synthesis recipes, including target phases and impurity phases that are observed during synthesis. Next, I will compare these reported observations to predictions from phase stability diagrams from density functional theory calculations. For example, I will report the extent to which impurities seen in synthesis are those expected from the DFT phase diagrams. I will also discuss whether finite temperature corrections to the DFT data improves agreement or not. Finally, I will provide a perspective on future pathways in this field. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Computational Materials Science & Engineering, Machine Learning, Powder Materials |