About this Abstract |
| Meeting |
MS&T24: Materials Science & Technology
|
| Symposium
|
Energy Materials for Sustainable Development
|
| Presentation Title |
Machine Learning Assisted Discovery of Perovskite Oxides for Protonic Ceramic Electrochemical Cell Electrolytes |
| Author(s) |
Ximei Zhai, Xiayan Han, Feng Luo, Jianhua Tong |
| On-Site Speaker (Planned) |
Jianhua Tong |
| Abstract Scope |
Protonic ceramic electrochemical cells can operate at intermediate temperatures (400-650oC) because of the low activation energy of protons transporting through the protonic ceramics, garnering extensive advantages over their counterparts (oxygen-ion conducting solid oxide fuel cells and polymer electrolyte membrane fuel cells). However, the discovery of new electrolyte materials with high proton conductivity at even lower operating temperatures is still facing challenge. We searched more 3000 conductivity data for known proton-conducting oxides and defined around 40 features for training machine learning models for screening new perovskite compositions with high proton conductivity. The excellent prediction accuracy of the established machine learning model was successful verified by state-of-the-art electrolyte materials of BaCe0.7Zr0.1Y0.1Yb0.1O3. |