Revision of the quasiclassical boundary resistance of the metal/superconductor interface

TL;DR

This paper revises the quasiclassical boundary resistance model for metal/superconductor interfaces, improving its consistency with experimental data by refining the boundary conditions and considering interface effects.

Contribution

It introduces an improved multi-channel derivation of boundary conditions that corrects the original Landauer conductance-based transparency, aligning theory more closely with experiments.

Findings

Revised boundary conditions reduce theory-experiment discrepancy

Interface roughness and wave function mismatch affect transparency

Original K-L boundary transparency linked to Landauer conductance

Abstract

Transparency of the metal/superconductor interface is discussed in a frame of the Kupryanov-Lukichev (K-L) quasiclassical boundary conditions. It is shown that the original K-L boundary transparency is expressed via the single-channel Landauer conductance. It is a source of dramatic discrepancy between the theory and experiments on the superconducting proximity effect. The improved multi-channel derivation is proposed for the boundary conditions, which rehabilitates consistency of the quasiclassical approach to the boundary transparency problem, and eliminates at least an order of magnitude of discrepancy between the theory and the experiment. The influence of the interface roughness and wave function symmetry mismatch at the boundary on the interface transparency is discussed. Warning is made to theories that use the quasiclassical boundary conditions with the interface transparencies close to unity.