Spin polarization of edge states and magnetosubband structure in quantum wires

TL;DR

This paper presents a numerical method to analyze spin-resolved edge states and subband structures in quantum wires under magnetic fields, revealing significant effects of exchange and correlation interactions on their properties.

Contribution

It introduces a Green's function based self-consistent approach incorporating spin interactions within density functional theory for quantum wire analysis.

Findings

Spin-resolved subband structures are significantly affected by exchange and correlation.

The method allows direct calculation of eigenstates and wave vectors for magnetotransport.

New features in magnetosubbands emerge due to many-body interactions.

Abstract

We provide a quantitative description of the structure of edge states in split-gate quantum wires in the integer quantum Hall regime. We develop an effective numerical approach based on the Green's function technique for the self-consistent solution of Schrodinger equation where electron- and spin interactions are included within the density functional theory in the local spin density approximation. The major advantage of this technique is that it can be directly incorporated into magnetotransport calculations, because it provides the self-consistent eigenstates and wave vectors at a given energy, not at a given wavevector (as conventional methods do). We use the developed method to calculate the subband structure and propagating states in the quantum wires in perpendicular magnetic field starting with a geometrical layout of the wire. We discuss how the spin-resolved subband structure, the current densities, the confining potentials, as well as the spin polarization of the electron and current densities evolve when an applied magnetic field varies. We demonstrate that the exchange and correlation interactions dramatically affect the magnetosubbands in quantum wires bringing qualitatively new features in comparison to a widely used model of spinless electrons in Hartree approximation.