High-throughput design of Co-based magnetic Heusler compounds

High-throughput (HTP) density functional theory (DFT) calculations are carried out on Co-based Heusler alloys, i.e., Co2YZ and X2CoZ (where X, Y, and Z range from lithium (Li) to bismuth (Bi) excluding inert gas and 4f rare-earth elements) in both cubic and tetragonal structures of the regular and inverse Heusler phases. After evaluating the thermodynamic stabilities based on the formation energy (& DBLBOND;Ef) and the distance to the convex hull (& DBLBOND;EHD), we validate the HTP calculations with 65 experimentally known cases and further predict 158 novel compounds with & DBLBOND;EHD smaller than 50 meV/atom, in which 117 compounds exhibit finite magnetization larger than 1 & mu;B/f.u.. The novel Co-based Heuslers predicted by our HTP work also includes some all -d metal compounds. In contrast to the conventional Heuslers with dominating p-d hybridization, the all -d-metal Heuslers consisting of mainly d-d hybridization cannot be properly interpreted by Burch's rule. The Slater- Pauling rule, on the other hand, is found to well describe the relationship between the intrinsic magnetic properties, including the total spin moment and Curie temperature, and the number of valence electrons. Furthermore, magneto-crystalline anisotropy is calculated for stable tetragonal compounds, providing useful information for potential applications in the fields of permanent magnets and magnetic recording materials.

Automated summary

In this study, high-throughput density functional theory calculations were performed on Co-based Heusler alloys. The calculations validated with 65 experimentally known cases and predicted 158 novel compounds. All-d metal Heuslers were found to have different hybridization characteristics compared to conventional Heuslers, and the Slater-Pauling rule was found to accurately describe the relationship between magnetic properties and valence electrons. In addition, magneto-crystalline anisotropy was calculated for stable tetragonal compounds.