About this Abstract |
| Meeting |
MS&T24: Materials Science & Technology
|
| Symposium
|
Advanced Ceramics for Environmental Remediation
|
| Presentation Title |
Machine Learning Assisted Discovery of Perovskite Oxides for Thermochemical CO2 Decomposition |
| Author(s) |
Ximei Zhai, Xiaoyan Han, Zeyu Zhao, Feng Luo, Jianhua Tong |
| On-Site Speaker (Planned) |
Jianhua Tong |
| Abstract Scope |
With the increasing global population and decreasing storage of fossil energy sources, the capture and recycling of greenhouse gas, CO2, back into the energy lifecycle is intensively pursued. Among versatile CO2 utilization routes, the direct thermochemical decomposition of CO2 into synthesis gas (CO + H2) with the existence of water is one of the most promising ones. The two-step thermochemical redox cycles of nonstoichiometric perovskite oxide particles provide the simplest process, which doesn’t require complex manufacturing and operation. However, discovering new thermochemical oxide materials is confronting a significant challenge. In this work, we developed a new machine learning assistant method to predict the formation of perovskite oxide structure for the compositions with multiple dopants. Because the machine learning training materials are confined to the known thermochemical perovskite oxides, the new perovskite oxide materials have been discovered. The discovered materials have been demonstrated to be capable of decomposing CO2. |