Steady-state dc transport through an Anderson impurity coupled to leads with spin-orbit coupling

TL;DR

This paper investigates how spin-orbit coupling affects electron transport through an Anderson impurity, revealing a crossover from three-peak to two-peak conductance profiles and explaining recent experimental findings.

Contribution

The study provides a comprehensive analysis of dc transport in an Anderson impurity with spin-orbit coupling, using an interpolative perturbative approach validated across regimes.

Findings

Universal zero bias peak with width proportional to Kondo temperature

Finite bias peaks around bias voltage equal to Coulomb interaction U

Temperature induces a crossover from three-peak to two-peak conductance

Abstract

We study the steady-state dc transport characteristics of a system comprised of an interacting quantum dot, modelled as an Anderson impurity, coupled to two metallic non-interacting leads with Rashba SOC, using an interpolative perturbative approach (IPA). The single-particle spectra, current and differential conductance are obtained in weak and strong coupling regimes over a wide range of SOC and bias values. Extensive benchmarking of the IPA validates the method in the linear as well as non-linear response regime. The universal, zero bias ($V_{sd}=0$) peak with a width proportional to the Kondo scale ($T_K$) and two non-universal finite bias peaks around $V_{sd}=\pm U$ in the zero temperature differential conductance show a clear separation with increasing $U$ or increasing SOC. In the strong coupling regime, increasing temperature induces melting of the zero bias peak leading to a crossover from a three-peak conductance to a two-peak conductance. Recent experiments find the emergence of a two-peak structure by increasing SOC at a fixed temperature. Our results appear to provide a qualitative explanation of these observations as a SOC tuned crossover from weak/intermediate to strong coupling, and a simultaneous crossover from low $T/T_K$ to high $T/T_K$ ratio. We also reproduce the experimentally observed temperature dependence of the zero bias conductance.