Novel low-energy La-Co-Pb ternary compounds involving a pair of immiscible Co-Pb elements are efficiently predicted by integrating machine learning (ML) with ab initio calculations.
Significance and Impact
Immiscible pair of elements are relatively unexplored but promising for novel quantum materials discovery due to well defined reduced dimensionality. Our ML approach is applicable to all crystalline systems for dramatic acceleration (~100 times) in materials discovery.
- Effectively integrated ML tools and databases with ab initio calculations and experimental validation.
- ML enables efficient exploration of vast chemical composition and crystal structure space for discovery in complex ternary systems.
- The ML-guided approach predicts stable La3CoPb and La18Co28Pb3 and other 56 low-energy La-Co-Pb ternary compounds.
Wang, R. H., Xia, W. Y., Slade, T. J., Fan, X. Y., Dong, H. F., Ho, K. M., Canfield, P. C., & Wang, C.-Z. (2022). Machine learning guided discovery of ternary compounds involving La and immiscible Co and Pb elements. npj Computational Materials, 8, 258. https://doi.org/10.1038/s41524-022-00950-0