The theory of planar ballistic SNS junctions at T = 0 T=0

Edouard B. Sonin · 2025

Read The theory of planar ballistic SNS junctions at T = 0 T=0 by Edouard B. Sonin (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-21583.txt.

Introduction

Edouard B Sonin presents a theory of layered superconducting junctions held at absolute zero. An SNS junction sandwiches a normal metal between two superconductors, and the current it carries depends on the phase gap between the sides. His 2025 letter derives that current-phase rule for flat junctions of any normal-layer thickness. The model uses a step-like pairing profile, and the result is exact when the Fermi energy dwarfs the gap. Phase slopes inside the leads are kept, which older theory dropped.

At short lengths the new rule differs sharply from the old one. Recent computer results and lab work on nanowire junctions match the new form. The analysis also fixes a clash with charge conservation that haunted the step model. Sonin offers the closed form as a reference for device physics.

Catalog record

CREATOREdouard B. Sonin
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE28 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-21583

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

The theory of planar ballistic SNS junctions at T = 0 T=0 Edouard B. Sonin Abstract The paper presents the theory of planar ballistic SNS junctions at T = 0 T=0 for any normal layer thickness L L taking into account phase gradients in superconducting layers. The current-phase relation was derived in the model of the steplike pairing potential model analytically and is exact in the limit of large ratio of the Fermi energy to the superconducting gap. Obtained current-phase relation is essentially different from that in the widely accepted theory neglecting phase gradients, especially in the limit L → 0 L\to 0 (short junction). The analysis resolves the problem with the charge conservation law in the steplike pairing potential model. Figure 1: The phase variation across the SNS junction. (a) The vacuum current produced by the phase \theta_{0} , which is called the vacuum phase. The current is confined to the normal layer, there is no current in superconducting layers. (b) The condensate current produced by the phase gradient \nabla\varphi in the superconducting layers. The phase \theta_{s}=L\nabla\varphi is the superfluid phase. In all layers the electric current is equal to env_{s} . (c) The superposition of the condensate and the vacuum current. The phase \theta is the Josephson phase. Introduction –The ballistic SNS junction has already been investigated a half-century. The pioneer papers Kulik 1969 ; Ishii 1970 ; Bardeen and Johnson 1972 and many subsequent ones used the self-consistent field method de Gennes 1966 . In this method an effective pairing potential is introduced, which transforms the second-quantization Hamiltonian with the electron interaction into an effective Hamiltonian quadratic in creation and annihilation electron operators. The effective Hamiltonian can be diagonalized by the Bogolyubov – Valatin transformation. The effective Hamiltonian is not gauge invariant, and the theory using it violates the charge conservation law. The charge conservation law is restored if one solves the Bogolyubov – de Gennes equations together with the self-consistency equation for the pairing potential. Starting from the original paper of Andreev 1964 , instead of solving the self-consistency equation, it was assumed that there is a gap \Delta of constant modulus \Delta_{0}=|\Delta| in the superconducting layers and zero gap inside the normal layer. The effective masses and Fermi energies were assumed to be the same in all layers. Further we call it the steplike pairing potential model. In the past it was assumed that not only the absolute value \Delta_{0} but also phases were constant in superconducting leads. Constant phases in two leads, however, were different. So, inevitable in current states phase gradients in leads were ignored Kulik 1969 ; Ishii 1970 ; Bardeen and Johnson 1972 as shown in Fig. 1 (a). At this phase profile the charge conservation law is violated since the current flows only inside the normal layer. But it was believed that the charge conservation law can be restored by so small phase gradients in leads that this cannot affect calculations ignoring the gradients. This suggestion is true for a weak link, inside which the phase varies much faster than in the leads. In Ref. Sonin 2021 it was demonstrated that if transverse cross-sections of all layers are the same, the SNS junction is not a weak link at zero temperature, and phase gradients in the leads do affect the current in the normal layer. Thus, one should determine currents in the normal and superconducting layers self-consistently. Such junctions are called planar junctions. We shall use this name even for 1D wires when the plane cross-section becomes a point. In Refs. Sonin 2021 ; Sonin 2024 the gap profile with the constant gradient \nabla\varphi in leads was considered: \Delta=\left\{\begin{array}[]{cc}\Delta_{0}e^{i\theta_{0}+i\nabla\varphi x}&x>L/2\\ 0&-L/2<x<L/2\\ \Delta_{0}e^{i\nabla\varphi x}&x<-L/2\end{array}\right.. (1) The strict conservation law was replaced by a softer condition that, at least, the total currents deep in all layers are the same. The Bogolyubov – de Gennes equations for the gap profile Eq. ( 1 ) can be exactly solved analytically for any \theta_{0} and \nabla\varphi if the gap \Delta_{0} is much smaller than the Fermi energy (see Sec. 9.5 in Ref. Sonin 2016 ). Among these solutions there is one, which does not violate the charge conservation law. It corresponds to the current state with \theta_{0}=0 and with the current J equal to the same current J_{s} as in a uniform superconductor: J=J_{s}=env_{s}=J_{0}{\theta_{s}\over\pi}, (2) where J_{0}={\pi en\hbar\over 2mL}, (3) e is the electron charge, m is the electron mass, n is the electron density, v_{s}={\hbar\over 2m}\nabla\varphi is the superfluid velocity, and \theta_{s}=L\nabla\varphi . In the 1D case J_{0}={e\hbar k_{f}\over mL}, (4) where k_{f} is the 1D Fermi wave number. The phase \theta_{s} and J_{s} were called the superfluid phase and the condensate current respectively Sonin 2021 . The phase profile for the condensate current shown in Fig. 1 (b) does not differ from that in a uniform superconductor. Since the condensate current is the only current in superconducting leads, the linear relation Eq. ( 2 ) between J_{s} and \theta_{s} automatically yields the linear current-phase relation J=J_{0}{\theta\over\pi} (5) between the total current J and the total Josephson phase \theta equal to \theta_{s} . The state with the condensate current can be obtained by the Galilean transformation of the ground state without currents. Galilean invariance of the ballistic SNS junction despite broken translational invariance has been already noticed by Bardeen and Johnson 1972 and confirmed in Ref. Sonin 2021 in the steplike pairing potential model. In Ref. Sonin 2024 the Galilean invariance was demonstrated for an arbitrary gap profile under the conditions that the ratio of \Delta_{0} to the Fermi energy is small and the Andreev reflection is the predominant mechanism of scattering. This means that the linear current-phase relation Eq. ( 5 ) is valid beyond the steplike pairing potential model for junctions with any normal layer thickness L . This is confirmed by numerical calculations by Riedel et al. 1996 . They did not used the steplike pairing potential model and solved the Bogolyubov – de Gennes equations together with the integral self-consistency equation. They obtained that although the pairing potential amplitude smoothly varied across interfaces between layers, the phase gradient remained strictly constant along the whole junction as in Fig. 1 (b). The solution of the Bogolyubov – de Gennes equations for the phase profile shown in Fig. 1 (a) is mathematically correct. But the current state corresponding to this profile cannot be realized physically if all Andreev level are unoccupied. The current in this state was called vacuum current J_{v} Sonin 2021 . However, if some Andreev level are at least partially occupied there is a current produced by quasiparticles at Andreev levels. It was called excitation current J_{q} Sonin 2021 . At zero temperature the excitation current appears at the critical current determined by the Landau criterion that the energy of the lowest Andreev level reaches 0. The current-phase relation must be derived from the condition that the total current flowing only in the normal layer vanishes: J_{v}+J_{q}=0 . The phase profile for the current state with nonzero currents J_{v} and J_{q} is shown in Fig. 1 (c). It is possible to solve the Bogolyubov – de Gennes equations also for this current state with phases \theta_{s} and \theta_{0} both

Read the complete text