Two-dimensional metal-insulator-transition as a potential fluctuation driven semiclassical transport phenomenon

TL;DR

This paper presents a classical percolation model explaining the 2D metal-insulator transition in semiconductor heterostructures as a fluctuation-driven phenomenon caused by long-range Coulomb disorder, matching experimental observations.

Contribution

It introduces a semiclassical, percolation-based framework for understanding the 2D MIT, emphasizing the role of potential fluctuations and inhomogeneous carrier localization.

Findings

Qualitative agreement with experimental 2D MIT behavior

Identification of a crossover in temperature dependence of resistivity

Demonstration of the role of long-range disorder in transport phenomena

Abstract

We theoretically consider the carrier density tuned (apparent) two-dimensional (2D) metal-insulator-transition (MIT) in semiconductor heterostructure-based 2D carrier systems as arising from a classical percolation phenomenon in the inhomogeneous density landscape created by the long-range potential fluctuations induced by random quenched charged impurities in the environment. The long-range Coulomb disorder inherent in semiconductors produces strong potential fluctuations in the 2D system where a fraction of the carriers gets trapped or classically localized, leading to a mixed 2-component semiclassical transport behavior at intermediate densities where a fraction of the carriers is mobile and another fraction immobile. At high carrier density, all the carriers are essentially mobile whereas at low carrier density all the carriers are essentially trapped since there is no possible percolating transport path through the lake-and-mountain inhomogeneous potential landscape. The low-density situation with no percolation would mimic an insulator whereas the high-density situation with allowed percolating paths through the lake-and-mountain energy landscape would mimic a metal with the system manifesting an apparent MIT in between. We calculate the transport properties as a function of carrier density, impurity density, impurity location, and temperature using a 2-component (trapped and mobile carriers) effective medium theory. Our theoretically calculated transport properties are in good qualitative agreement with the experimentally observed 2D MIT phenomenology in 2D electron and hole systems. We find a high (low) density metallic (insulating) temperature-dependence of the 2D resistivity, and an intermediate-density crossover behavior which could be identified with the experimentally observed 2D MIT.