Machine Learning and Theoretical Prediction of Highly Spin‐Polarized Cr<sub>2</sub>CO<i><sub>x</sub></i> MXene with Enhanced Curie Temperature
Jianhui Yang, Fei shi, Chengji J. Zhou, Shaozheng Zhang, Qiao Sui, Liang Chen
Abstract
Abstract 2D magnetic materials with high spin polarization and Curie temperature are highly desirable for ultrathin spintronic devices. This study utilizes first‐principles methods to systematically investigate 225 O adsorption configurations, demonstrating that Cr 2 CO x MXene consistently maintains a long‐range‐ordered ferromagnetic arrangement with high spin polarization, irrespective of the O adsorption configuration. Most configurations also display Curie temperature ( T C ) exceeding room temperature, with the possibility of further enhancement by reducing O coverage. Machine learning models are developed to accurately predict O adsorption configurations, exchange interaction energies, and T C . A novel approach of stripping F and OH groups to create Cr 2 CO x on Cr‐based MXene surfaces is proposed to address the difficulty in achieving long‐range‐ordered magnetic structures by manipulating surface adsorbates in MXene. This approach enhances the ability to control the magnetic properties of MXenes and paves the way for their application in ultrathin spintronic devices.