About this Abstract |
| Meeting |
MS&T25: Materials Science & Technology
|
| Symposium
|
Advances in Refractory High Entropy Alloys and Ceramics
|
| Presentation Title |
AI-Driven Discovery of Compositionally Complex Alloys for Enhanced Mechanical Performance |
| Author(s) |
Bernard Gaskey, Janith Wanni, Mikayla Obrist, Avanish Mishra, Nithin Mathew, Saryu Fensin |
| On-Site Speaker (Planned) |
Bernard Gaskey |
| Abstract Scope |
High entropy alloys (HEAs) break the traditional alloy design paradigm by taking advantage of composition spaces that have not been comprehensively studied. Design of HEAs or other compositionally complex alloys is more challenging than traditional alloys because of the additional degrees of freedom created by additional principal components. To overcome this hurdle, we employ an AI-based approach to accelerate the discovery of new compositions. Additional CALPHAD-based screening predicts solidification and phase behavior to eliminate compositions which form problematic intermetallic structures. This information feeds back into the AI model to iteratively improve predictions. Here, we apply these tools to the design of new refractory HEAs. We evaluate the dynamic mechanical response of a prototype alloy and link its performance to microstructural features identified computationally. By using this combined computational approach alongside high-throughput experiments, new alloys optimized for specific properties can be designed, prototyped, and tested orders of magnitude faster than before. |