| About this Abstract |
| Meeting |
2022 TMS Annual Meeting & Exhibition
|
| Symposium
|
Advanced Materials for Energy Conversion and Storage 2022
|
| Presentation Title |
Machine Learning Enables Discovery of Ternary Alloy Catalysts for Oxygen Reduction |
| Author(s) |
Youngtae Park, Hyuck Mo Lee |
| On-Site Speaker (Planned) |
Hyuck Mo Lee |
| Abstract Scope |
Pt is utilized as an electrocatalyst to accelerate the slow oxygen reduction reaction (ORR) rate in the cathode of proton exchange membrane fuel cells (PEMFCs). However, the high cost of Pt prevents PEMFCs from being commercially viable. Despite the fact that numerous studies attempt to lower the amount of Pt by alloying, most catalyst designs are confined to bimetallic alloys. Using a graph-based convolutional neural network (GCNN), we significantly extend the material design space from binary to ternary for ORR catalysts for PEMFCs. We have collected our own database of 9,267 surface binding energies composed of five key adsorbates (H, O, OH, OOH, and CO). With a mean absolute error of only 0.164 eV, the GCNN model has outstanding predictive performance. As a consequence, we discovered 10 promising ORR ternary alloy catalysts with lower costs and greater catalytic activity than Pt. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Energy Conversion and Storage, Modeling and Simulation, Computational Materials Science & Engineering |