Drastic effect of weak interaction near special points in semiclassical multiterminal superconducting nanostructures

TL;DR

This paper investigates how weak interactions near special phase points in multiterminal superconducting nanostructures drastically alter the spectrum, potentially inducing a gap in previously gapless phases, thus challenging topological protection.

Contribution

It develops a universal quantum action model for weak interactions near special points and analyzes their effect on the spectral phases, revealing the collapse of topological protection.

Findings

Weak interactions squeeze the gapless phase, causing gapped phases to contact.

Logarithmic divergence of corrections suggests possible gap opening.

Interaction effects challenge the topological protection of the spectrum.

Abstract

A generic semiclassical superconducting nanostructure connected to multiple superconducting terminals hosts a quasi-continuous spectrum of Andreev states. The spectrum is sensitive to the superconducting phases of the terminals. It can be either gapped or gapless depending on the point in the multi-dimensional parametric space of these phases. Special points in this space correspond to setting some terminals to the phase 0 and the rest to the phase of $\pi$. For a generic nanostructure, three distinct spectra come together in the vicinity of a special point: two gapped phases of different topology and a gapless phase separating the two by virtue of topological protection. In this paper, we show that a weak interaction manifesting as quantum fluctuations of superconducting phases drastically changes the spectrum in a narrow vicinity of a special point. We develop an interaction model and derive a universal generic quantum action that describes this situation. The action is complicated by incorporating a non-local in time matrix order parameter, and its full analysis is beyond the scope of the present paper. Here, we identify and address two limits: the semiclassical one and the quantum one, concentrating on the first-order interaction correction in the last case. In both cases, we find that the interaction squeezes the domain of the gapless phase in the narrow vicinity of the point so the gapped phases tend to contact each other immediately, defying the topological protection. We identify the domains of strong coupling where the perturbation theory does not work. In the gapless phase, we find the logarithmic divergence of the first-order corrections. This leads us to an interesting hypothesis: weak interaction might induce an exponentially small gap in the formerly gapless phase.