Correlated quantum criticality and topological magnetoelectric response in Bi2Se3

arXiv:2611.03670 · cond-mat.str-el · Submitted 2026-10-12 · Read on arXiv

cond-mat.str-el

Submitted: 2026-10-12

Updated: 2026-10-12

Comments: 20 pages, 8 Figures, Submitted to npj Quantum Materials for consideration as a research article in the Special Collection \textit{Intertwined Orders in Quantum Materials--- In Memory of Daniel Khomskii}

Code: https://github.com/joselado/quantum-lattice

License: http://creativecommons.org/licenses/by/4.0/

The gist: A defining feature of quantum materials is that multiple orders govern low-energy physics and are not independent.

Terminology

Abstract

A defining feature of quantum materials is that multiple orders govern low-energy physics and are not independent. We study Bi 2 Se 3 by combining DFT with a renormalized correlated honeycomb model in which frustrated magnetic order and helical topological surface states are coupled via one control parameter. The surface states map to a honeycomb Hubbard model with frustrated J 1 -- J 2 exchange. Exact diagonalization up to 20 sites shows the Néel structure factor and valence-bond-solid order parameter cross at J 2/J 1 about 0.22, agreeing with QMC/DMRG, and a dynamical spin structure factor broadens near the crossover, as expected for a spinon continuum. Band inversion of the renormalized tight-binding model occurs on the same critical locus, producing jumps in the Chern number and Z 2 invariant mirrored by the skyrmion charge Q. This gives a microscopic link: Q selects the axion angle and thus the topological magnetoelectric coupling. We distinguish axion coupling, independent of skyrmion helicity, from the helicity-dependent magnetoelectric multipole coupling, whose toroidal component scales as γ and vanishes for Néel-type textures. We derive the Q-topological magnetoelectric energy and estimate about 10-20, J per skyrmion per cycle for few-tesla fields. The cycle is thermodynamically bounded: constant- θ cycles extract no net work, and efficiency is limited by axionic entropy change.

Related papers