Machine learning-driven predictive design of catalytic oxygen carriers for chemical looping processes
Abstract Chemical looping is an innovative technology for conversion of fossil fuels to clean energy with integrated carbon dioxide capture. The success of this technology depends on the performance of oxygen carriers which circulate between the reactors. However, traditional trial-and-error methods...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2025-07-01
|
| Series: | Discover Energy |
| Online Access: | https://doi.org/10.1007/s43937-025-00081-9 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Abstract Chemical looping is an innovative technology for conversion of fossil fuels to clean energy with integrated carbon dioxide capture. The success of this technology depends on the performance of oxygen carriers which circulate between the reactors. However, traditional trial-and-error methods for oxygen carrier design are time-consuming and resource-intensive. In recent years, machine learning techniques have rapidly evolved and been applied to material discovery. These techniques can efficiently screen a vast number of materials to find the most promising candidates based on their predicted properties. This perspective summarizes recent advances in utilizing machine learning for catalyst development and explores its application in the design of catalytic oxygen carriers for chemical looping processes. By integrating machine learning with high-throughput simulations, we can gain deeper insights into the active sites, oxygen vacancies, and redox reaction mechanisms of catalytic oxygen carriers, which are crucial for the development of these high-performance chemical looping materials. Additionally, we discuss the key challenges and opportunities of this approach, highlighting its potential for accelerating the application of chemical looping technologies. Graphical abstract |
|---|---|
| ISSN: | 2730-7719 |