About this Abstract |
Meeting |
2020 TMS Annual Meeting & Exhibition
|
Symposium
|
Computational Discovery and Design of Emerging Materials
|
Presentation Title |
Density Functional Theory and Machine Learning Guided Prediction of Thermal Properties of Rare-earth Disilicates |
Author(s) |
Mukil Ayyasamy, Prasanna Balachandran |
On-Site Speaker (Planned) |
Mukil Ayyasamy |
Abstract Scope |
Application of rare-earth disilicates in extreme environments, such as the environmental barrier coatings (EBC) in jet engines, require knowledge of their coefficient of thermal expansion (CTE). Till date, CTE data have been determined experimentally for only a small fraction of the vast search space. The present work explores a computational approach, enabled by density functional theory (DFT) calculations and machine learning (ML) methods, to accelerate the prediction of CTE in previously unexplored compounds. DFT calculations were performed to construct the descriptors for ML. Ensemble ML models were used to establish a relationship between the descriptors from DFT and CTE (taken from published experiments). The trained models were then used to rapidly predict the CTE for previously unexplored compounds. One of the key outcomes is the identification of novel disilicates with CTE in the range (3-5.5×10-6 K-1) that are suitable for EBC applications. These compounds are recommended for experimental validation. |
Proceedings Inclusion? |
Planned: Supplemental Proceedings volume |