Revisiting quadratic band crossing: from interaction-driven instability to intrinsic topology

Abstract

The realization of robust quantum anomalous Hall (QAH) phases at elevated temperatures remains a central challenge in condensed matter physics. While quadratic band crossing points (QBCP) provide a promising route towards QAH states, existing proposals are largely confined to idealized models or hindered by interaction-driven competing orders. Here, we demonstrate that these limitations are not intrinsic to QBCP but arise from their specific implementation. We propose a general mechanism where band inversion between a symmetry-protected orbital doublet (e.g. $d_{xz},d_{yz}$) and an isolated orbital (e.g. $d_{z^2}$)-generically generates a QBCP with opposite curvature. This crossing is directly gapped at the single-particle level by intrinsic atomic spin-orbit coupling, while the underlying band inversion naturally shields the resulting topological gap against other interaction-driven instabilities. We further suggest monolayer compounds $MNX_2$ ($M$= Ni, Pd, Pt; $N$= Nb, Ta; $X$= S, Se, Te) as a realistic material class that intrinsically realizes this mechanism. These findings provide a concrete pathway toward robust QAH phases in correlated materials.