Localized Wannier function based tight-binding models for two-dimensional allotropes of bismuth

TL;DR

This paper develops localized Wannier function-based tight-binding models for 2D bismuth allotropes, accurately capturing their topological electronic structures and symmetries, which are crucial for understanding and designing topological devices.

Contribution

The authors introduce a new parameterization method using Wannier functions from first-principles calculations to model 2D bismuth allotropes, overcoming limitations of previous semi-empirical approaches.

Findings

Successfully reproduces band structures and topological features

Models applicable to various 2D bismuth structures

Provides a foundation for studying transport and electromagnetic responses

Abstract

With its monoelemental composition, various crystalline forms and an inherently strong spin-orbit coupling, bismuth has been regarded as an ideal prototype material to expand our understanding of topological electronic structures. In particular, two-dimensional bismuth thin films have attracted a growing interest due to potential applications in topological transistors and spintronics. This calls for an effective physical model to give an accurate interpretation of the novel topological phenomena shown by two-dimensional bismuth. However, the conventional semi-empirical approach of adapting bulk bismuth hoppings fails to capture the topological features of two-dimensional bismuth allotropes because the electronic band topology is heavily influenced by crystalline symmetries as well as atom spacings. Here we provide a new parameterization using localized Wannier functions derived from the Bloch states in first-principles calculations. We construct new tight-binding models for three types of two-dimensional bismuth allotropes: a Bi (111) bilayer, bismuthene and a Bi(110) bilayer. We demonstrate that our tight-binding models can successfully reproduce the band structures, symmetries and topological features of these two-dimensional allotropes. We anticipate that these models can be extended to other similar two-dimensional topological structures such as antimonene and arsenene. Moreover, these models can serve as a starting point for investigating the electron/spin transport and electromagnetic response in low-dimensional topological devices.