Self-doping instability of the Wigner-Mott insulator

TL;DR

This paper proposes a microscopic theory for the 2D metal-insulator transition driven by self-doping in a Wigner-Mott insulator, explaining experimental phenomena like mass enhancement and resistivity changes.

Contribution

It introduces a two-band Hubbard model capturing vacancy-interstitial pair excitations, revealing a self-doping mechanism for the transition in clean 2D electron systems.

Findings

Identifies a critical carrier concentration for the transition.

Explains large effective mass and resistivity drop.

Provides a phase diagram consistent with experiments.

Abstract

We present a theory describing the mechanism for the two-dimensional (2D) metal-insulator transition (MIT) in absence of disorder. A two-band Hubbard model is introduced, describing vacancy-interstitial pair excitations within the Wigner crystal. Kinetic energy gained by delocalizing such excitations is found to lead to an instability of the insulator to self-doping above a critical carrier concentration $n=n_c$, mapping the problem to a density-driven Mott MIT. This mechanism provides a natural microscopic picture of several puzzling experimental features, including the large effective mass enhancement, the large resistivity drop, and the large positive magneto-resistance on the metallic side of the transition. We also present a global phase diagram for the clean 2D electron gas as a function of $n$ and parallel magnetic field $B_{\shortparallel}$, which agrees well with experimental findings in ultra clean samples.