Do epitaxial semiconductor-superconductor hybrid devices really support topological superconductivity?

by Chris Reeg (Uni Basel)

Wednesday 7 Nov 2018, 10:00 → 11:00 Europe/Stockholm

NOrdita South Seminar room (Nordita South/Cond Mat)

Abstract: In recent years there has been significant emphasis placed on the development of hybrid devices consisting of a semiconducting nanowire coupled to a thin layer of epitaxially grown superconducting aluminum, and these devices are widely regarded as the ideal platform for studying topological superconductivity due to the very strong and uniform contact between the two materials. In this talk, I will discuss a microscopic treatment of the proximity effect in such a device and show that this approach exhibits significant deviations from more frequently utilized phenomenological theories. Most importantly, the finite thickness of the superconducting film introduces a large energy scale to the system that is determined by the quantized subband spacing. In the limit where a sizable superconducting gap is induced in the semiconductor, we show that all material parameters of the semiconductor become significantly renormalized due to hybridization with the superconductor and that the semiconducting bands undergo a large effective chemical potential shift that is comparable to the large subband spacing of the superconductor. In this limit, we argue it is unlikely that the hybrid system can support topological superconductivity. Additionally, in order to explain numerous recent experimental observations of zero-bias conductance peaks in these epitaxial devices, we present an analytical theory describing how one might generically observe topologically trivial zero-energy Andreev bound states in such a device.