Ab initio overestimation of the topological region in Eu-based compounds

This is the authors' abstract. We don't add key points for this paper.

An underestimation of the fundamental band gap values by the density functional theory within the local density approximation and associated approaches is a well-known challenge of ab initio electronic structure computations. Motivated by recent optical experiments [D. Santos-Cottin et al., arXiv:2301.08014], we have revisited first-principles results obtained earlier for EuCd₂As₂ and extended the computational studies to the whole class of systems EuCd₂4pt0exX₂ (X=P, As, Sb, Bi), to EuIn₂4pt0exX₂ (X=P, As, Sb), and to nonmagnetic AEIn₂As₂ (AE=Ca, Sr, Ba) employing a hybrid functional method. We find that our approach provides the magnitude of the energy gap for EuCd₂As₂ in agreement with the experimental value. Actually, our results indicate that EuSn₂As₂, BaIn₂As₂, EuCd₂Bi₂ and EuCd₂SbBi are robust topological insulators, while all other compounds are topologically trivial semiconductors. The trivial band gaps of EuCd₂P₂, EuCd₂As₂, and EuCd₂Sb₂ are in the range of 1.38--1.48 eV, 0.72--0.79 eV, and 0.46--0.49 eV, respectively. The topologically trivial Eu-based systems are antiferromagnetic semiconductors with a strong red shift of the energy gap in a magnetic field caused by the exchange coupling of the band states to spins localized on the $4f$-shell of Eu ions. Additionally, the EuIn₂X₂ (X=P, As) compounds show altermagnetic exchange-induced band spin-splitting, particularly noticeable in the case of states derived from $5d$-Eu orbitals.