Combining quasiparticle self-consistent and machine-learned to assess half-metallicity in Co- and Ni-based Heuslers
Phys. Rev. Materials 10, 094411 – Published 24 September, 2026
DOI: https://doi.org/10.1103/6y6t-xkx4
Abstract
Half-metallic Heusler compounds are of significant interest for spintronics. For device fabrication, compounds that can be epitaxially grown on III-V semiconductors are particularly attractive. We present a first-principles investigation of four Co-based and two Ni-based Heusler compounds that are lattice-matched to InAs. The results of density functional theory (DFT) using semilocal and hybrid functionals are compared with quasiparticle self-consistent (). We also assess DFT with machine-learned Hubbard corrections, as proposed in M. Yu et al. [npj Comput. Mater. 6, 180 (2020)]. Here, the values that yield the closest agreement with the band structure are determined using an updated Bayesian optimization (BO) objective function that considers the atomic magnetic moments in addition to the band structure. We find that (BO) can adequately reproduce the key features in most cases. Our results reveal a strong method dependence of the degree of spin polarization at the Fermi level and, in some cases, even the dominant spin channel (majority or minority). Of the materials studied here, and are the most likely to be half-metals.