Generic quantized zero-bias conductance peaks in superconductor-semiconductor hybrid structures

TL;DR

This paper presents a theoretical model showing that quantized zero-bias conductance peaks are common in superconductor-semiconductor hybrids and can occur without Majorana zero modes, challenging their use as definitive signatures.

Contribution

The study introduces a zero-dimensional random matrix model demonstrating that quantized conductance peaks are generic and not exclusive evidence of Majorana modes in superconductor-semiconductor systems.

Findings

Quantized conductance peaks are ubiquitous in the model.

These peaks can occur without Majorana zero modes.

Stable quantized peaks occupy a large parameter space.

Abstract

We show theoretically that quantized zero-bias conductance peaks should be ubiquitous in superconductor-semiconductor hybrids by employing a zero-dimensional random matrix model with continuous tuning parameters. We demonstrate that a normal metal-superconductor (NS) junction conductance spectra can be generically obtained in this model replicating all features seen in recent experimental results. The theoretical quantized conductance peaks, which explicitly do not arise from spatially isolated Majorana zero modes, are easily found by preparing a contour plot of conductance over several independent tuning parameters, mimicking the effect of Zeeman splitting and voltages on gates near the junction. This suggests that, even stable apparently quantized conductance peaks need not correspond to isolated Majorana modes; rather, the $\textit{a priori}$ expectation should be that such quantized peaks generically occupy a significant fraction of the high-dimensional tuning parameter space that characterizes the NS tunneling experiments.