Quantum Anomalous Hall Effect in $d$-Electron Kagome Systems: Chern Insulating States from Transverse Spin-Orbit Coupling

Abstract

The possibility of quantum anomalous Hall effect (QAHE) in two-dimensional kagome systems with $d$-orbital electrons is studied within a multi-orbital tight-binding model. We concentrate on the case of isotropic Slater-Koster integrals which is realized in a recently discovered class of metal-organic frameworks TM$_3$C$_6$O$_6$ with transition metals (TM) in the beginning of the 3$d$ series. Furthermore, in the absence of exchange-type spin-orbit coupling, only isotropic Slater-Koster integrals give a perfect flatband in addition to the two dispersive bands hosting relativistic (Dirac) and quadratic band crossing points at high symmetry spots in the Brillouin zone. A quantized topological invariant requires a flux-creating spin-orbit coupling, giving Chern number (per spin sector) $C=1$ not only from the familiar Dirac points at the six corners of the Brillouin zone, but also from the quadratic band crossing point at the center $\Gamma$. In the case of isotropic Slater-Koster integrals the on-site spin-orbit coupling (SOC) is ineffective to create the QAHE and it is only the transfer or exchange-type SOC which can lead to a QAHE. Surprisingly, this QAHE comes from the nontrivial effective flux induced by the \textit{transverse} part of the spin-orbit coupling, exhibited by electrons in the $d$-orbital state with $m_l=0$ ($d_{z^2}$ orbital), in stark contrast to the more familiar form of QAHE due to the $d$-orbitals with $m_l \neq 0$, driven by the Ising part of spin-orbit coupling. The $C=1$ Chern plateau (per spin sector) due to Dirac point extends over a smaller region of Fermi energy than that due to quadratic band crossing. Our result hints at the promising potential of kagome $d$-electron systems as a platform for dissipationless electronics by virtue of its unique QAHE.