Superconducting Interfaces: When Quantum Interference Skews the Current

Author: Denis Avetisyan

New research reveals that extended superconducting interfaces exhibit an intrinsic asymmetry in electron flow due to quantum interference effects.

The study demonstrates that asymmetry and interferometry reveal a relationship between normalized energy <span class="katex-eq" data-katex-display="false">G(E)</span> and extracted amplitude <span class="katex-eq" data-katex-display="false">\mathcal{A}(E)</span>, exhibiting a fan-diagram-like phase-space domain corresponding to propagating modes as a function of normalized length <span class="katex-eq" data-katex-display="false">L/\lambda_{osc}</span>. — The study demonstrates that asymmetry and interferometry reveal a relationship between normalized energy $G(E)$ and extracted amplitude $\mathcal{A}(E)$ , exhibiting a fan-diagram-like phase-space domain corresponding to propagating modes as a function of normalized length $L/\lambda_{osc}$ .

The study demonstrates bias-dependent conductance arising from interference in Andreev reflected quasiparticles at extended superconducting interfaces, providing a spectroscopic probe of interface properties.

The widely assumed symmetry between particle and hole transport in superconductors can be unexpectedly broken at interfaces. In the work ‘Quantum Interference Breaks Bias Symmetry at Extended Superconducting Interfaces’, we demonstrate that spatially extended superconducting interfaces exhibit an intrinsic interferometric effect, leading to measurable bias-asymmetry in conductance via unequal phase accumulation of quasiparticles. This asymmetry, revealed through a tight-binding scattering formalism, arises from normal-state interference yet is sharply governed by the superconducting gap, acting as a spectroscopic probe of interface physics. Could this interferometric effect offer a new avenue for characterizing nonlocal phenomena and pairing mechanisms in complex hybrid and topological systems?

The Illusion of Sharpness: Beyond Idealized Interfaces

Conventional analyses of superconducting junctions routinely employ the simplification of infinitely thin interfaces, a tactic that dramatically eases mathematical treatment but introduces a significant disconnect from physical reality. This idealization allows for tractable calculations of quantities like the critical current and Josephson frequency, yet it inherently neglects the complex spatial variations of the superconducting order parameter and the associated changes in electronic properties that arise within a finite-width interface. While computationally convenient, this approach fails to capture phenomena such as interface transparency effects, the formation of bound states, and the modification of the density of states-all critical aspects of junction behavior in actual devices. Consequently, understanding and predicting the performance of advanced superconducting circuits necessitates moving beyond this simplification and embracing more realistic models that account for the extended nature of the interface itself.

The assumption of atomically sharp interfaces in superconducting junctions, while mathematically convenient, diverges significantly from physical reality. Actual interfaces exhibit a finite, spatially extended structure due to material intermixing, roughness, and the inherent limitations of atomic-scale precision in fabrication. This extended nature introduces a range of complex phenomena absent in idealized models; for instance, the gradual variation of the superconducting order parameter across the interface impacts the dynamics of Andreev reflection and the formation of midgap states. Consequently, the simplistic treatment of the interface as a two-dimensional boundary condition proves inadequate, necessitating more sophisticated theoretical approaches that account for the three-dimensional, spatially varying characteristics of these junctions to accurately predict and understand their behavior, particularly in advanced quantum devices.

The spatially extended nature of superconducting junctions profoundly reshapes the behavior of quasiparticles – the elementary excitations carrying both energy and momentum – within the interface region. Unlike the predictions of simplified, idealized models, these quasiparticles experience altered energy levels and scattering probabilities due to the broadened potential landscape. This necessitates a move beyond traditional perturbation theory and the development of a more comprehensive theoretical framework, potentially leveraging non-equilibrium Green’s functions or kinetic equations, to accurately describe their dynamics. Such a framework must account for the spatial distribution of the superconducting order parameter and the resulting modifications to the quasiparticle density of states, ultimately influencing the junction’s critical current and overall performance. Understanding these effects is crucial for designing next-generation superconducting devices that move beyond the limitations imposed by overly simplistic assumptions.

Andreev Reflection and the Sensitivity of Phase

Andreev reflection describes the process where an electron incident on a superconducting interface is retroreflected as a hole, and simultaneously a Cooper pair forms within the superconductor. This occurs because, at the interface, momentum and energy must be conserved; the incoming electron transfers its charge and momentum to create the hole, while the remaining energy contributes to the formation of the Cooper pair. Consequently, Andreev reflection isn’t simply a reflection of charge, but a conversion of quasiparticles – electrons becoming holes – and is fundamentally different from ordinary reflection. This process is crucial for understanding low-temperature transport phenomena in superconductor-normal metal structures, as it allows current to flow without the need for direct electron transport through the superconducting barrier, leading to sub-gap conductance.

The extended interface between a normal metal and a superconductor introduces phase sensitivity to the transport process due to the spatial extent of the induced superconducting wavefunction. This means the electron and hole wavefunctions participating in Andreev reflection are not simply reflected at a point, but rather evolve over a finite distance within the interface region. The accumulated phase difference between these wavefunctions, determined by the interface length and the wave vectors of the incident electron and reflected hole, directly impacts the probability of Andreev reflection and thus the overall conductance. Any variations in the interface potential or geometry can alter this phase accumulation, leading to observable changes in the transport characteristics.

Phase-sensitive transport at the superconducting interface, resulting from Andreev reflection, produces interference effects characterized by an oscillation length $\lambda_{osc}(E) = 2\pi / |k_e(E) - k_h(E)|$ . This length is directly determined by the wave-vector mismatch between the incident electrons $k_e(E)$ and the reflected holes $k_h(E)$ as a function of energy, E. The oscillation length dictates the periodicity of constructive and destructive interference patterns observed in the current-voltage characteristics of the device, providing a measurable parameter linked to the properties of the interface and the superconducting energy gap. Variations in the energy dependence of $k_e(E)$ and $k_h(E)$ will directly alter $\lambda_{osc}(E)$ , influencing the observed interference behavior.

Constructing a Model: Simulating Extended Interfaces

The Bogoliubov-de Gennes (BdG) Hamiltonian provides a mean-field description of superconducting systems by extending the standard Schrödinger equation to include particle-hole excitations. This formulation allows for the treatment of Cooper pairs as fundamental degrees of freedom, essential for modeling superconducting behavior. In this work, the BdG Hamiltonian is discretized using a tight-binding approach, representing the system’s electronic structure on a lattice. This simplification reduces the computational complexity while retaining the key physics, enabling the calculation of energy levels and wavefunctions for both single-particle and quasiparticle excitations; the resulting $H_{BdG}$ matrix describes the system in a computationally tractable form.

The Scattering-Matrix (S-matrix) formulation provides an efficient method for calculating transport properties in mesoscopic systems by focusing on incoming and outgoing electron waves rather than solving for the full wavefunction within the device. This approach is particularly advantageous for systems with well-defined leads, as it directly yields quantities like conductance and transmission probability without requiring knowledge of the wavefunction’s behavior in the device interior. The Kwant package is a Python-based software designed for numerically solving the $S$ -matrix formalism, employing a nearest-neighbor tight-binding discretization and efficient algorithms to handle large systems and complex geometries. Its implementation allows for the calculation of various transport characteristics, including Landauer conductance, noise, and the effects of disorder, with computational scaling favorable for many realistic device sizes.

Dynes broadening is implemented to model the effects of imperfections and quasiparticle lifetime within the superconducting system. This technique introduces a finite imaginary part, Γ, to the superconducting energy gap, effectively smearing out the sharp Van Hove singularities in the density of states. The broadened gap, described by $\Delta(\omega) = \sqrt{\epsilon^2 + \Delta_0^2} - \epsilon$ where $\Delta_0$ is the ideal gap, accounts for processes that limit the coherence of quasiparticles, such as impurity scattering and phonon interactions. By including Dynes broadening, the computational model more accurately reflects realistic material properties and provides more reliable predictions of transport behavior in the presence of disorder.

Symmetrization of the normalized spectral function <span class="katex-eq" data-katex-display="false">G(E)</span> allows for accurate fitting with the Breit-Wigner-Kaon (BTK) model, revealing a dependence of the fitted parameters <span class="katex-eq" data-katex-display="false">Z_{fit}</span>, <span class="katex-eq" data-katex-display="false">\Delta_{fit}</span>, and <span class="katex-eq" data-katex-display="false">\Gamma_{fit}</span> on the fitting parameter <span class="katex-eq" data-katex-display="false">LL</span>. — Symmetrization of the normalized spectral function $G(E)$ allows for accurate fitting with the Breit-Wigner-Kaon (BTK) model, revealing a dependence of the fitted parameters $Z_{fit}$ , $\Delta_{fit}$ , and $\Gamma_{fit}$ on the fitting parameter $LL$ .

Revealing the Signature: Bias Asymmetry and Oscillation Length

Calculations reveal a striking asymmetry in the conductance of the superconducting interface, deviating significantly from behavior expected in idealized, infinitesimally thin junctions. This asymmetry isn’t a result of material imperfections, but an inherent consequence of the interface extending spatially – the superconducting properties aren’t confined to a single plane. Instead of a symmetrical response to applied voltage, the conductance exhibits a preferential direction, meaning current flows more easily in one direction than the other. This effect is particularly pronounced because the extended interface allows for a more complex interplay of electron waves, leading to constructive and destructive interference patterns sensitive to the voltage polarity, ultimately manifesting as this bias-dependent asymmetry. Understanding this asymmetry is crucial, as it provides a direct probe of the interface’s physical characteristics and distinguishes it from the simplified models often used in superconducting device analysis.

This observed asymmetry in conductance isn’t merely a quirk of the system, but a direct consequence of the interference patterns established within the extended superconducting interface; the oscillation length, $\lambda_{osc}$ , fundamentally defines the spatial extent of these patterns. Critically, this length scale is directly measurable through experimentation, offering a powerful diagnostic tool for characterizing the material’s electronic properties. Specifically, determining $\lambda_{osc}$ allows for the extraction of key parameters such as the Fermi velocity – the speed at which electrons propagate – as well as the carrier density and the band curvature of the material, providing insights into its fundamental electronic structure and potential for technological applications.

Analysis of the conductance reveals a predictable relationship between bias asymmetry and the ratio of junction length to oscillation length, $L/λ_{osc}$ . This manifests as a universal periodic curve, offering a robust diagnostic for characterizing the extended superconducting interface. Critically, attempts to model these spectra using the standard Blonder-Tinkham-Klapwijk (BTK) theory consistently overestimate both the energy gap $\Delta_{fit}$ and the level of broadening $\Gamma_{fit}$ by roughly 12%. This systematic deviation isn’t a limitation of the experimental setup, but rather a clear signature indicating the presence of the extended interface, effectively differentiating it from the simplified assumptions of idealized, infinitesimally thin junctions used in conventional models.

The study of extended superconducting interfaces reveals a system where seemingly isolated components-the interface itself, the applied bias, and the resulting conductance-are deeply interconnected. This holistic behavior echoes a fundamental principle articulated by Aristotle: “The whole is greater than the sum of its parts.” The observed bias-asymmetry isn’t simply a consequence of individual material properties, but an emergent phenomenon arising from the interplay of these elements. Just as a living organism functions not through isolated organs, but through their coordinated activity, the superconducting interface demonstrates that understanding the complete system is crucial to unlocking its properties and potential as a spectroscopic probe of pairing physics.

Beyond Symmetry: Future Directions

The demonstration of inherent interferometric effects at extended superconducting interfaces offers more than a novel spectroscopic technique; it reveals the limitations of treating such junctions as simple, localized points of contact. The observed bias asymmetry isn’t merely a nuisance to be corrected, but a signature of the interface’s extended nature – a reminder that the whole is demonstrably more than the sum of its constituent parts. Future work must address the challenge of disentangling these interferometric contributions from other sources of asymmetry, particularly those arising from intrinsic material properties or subtle variations in interface fabrication. A complete theoretical treatment will likely require moving beyond simplified models and embracing a fully self-consistent approach, incorporating realistic interface roughness and spatial variations in the superconducting order parameter.

The connection between interface structure and spectroscopic signatures also presents a compelling avenue for exploration. Can the precise details of the interference pattern be used to reconstruct the interfacial topography with nanometer resolution? Conversely, how do deviations from ideal interface smoothness affect the observed quasiparticle dynamics? There’s a trade-off here, of course; increasing complexity in the model invariably introduces additional parameters, potentially obscuring the very phenomena one seeks to understand.

Ultimately, this work highlights a recurring theme in condensed matter physics: the tendency for seemingly subtle geometric effects to exert a profound influence on macroscopic observables. The exploration of these extended interfaces, and the symmetries they break, promises to reveal not only the intricacies of superconductivity, but also fundamental principles governing the behavior of quantum systems at boundaries.

Original article: https://arxiv.org/pdf/2603.14329.pdf

Contact the author: https://www.linkedin.com/in/avetisyan/

The Illusion of Sharpness: Beyond Idealized Interfaces

Andreev Reflection and the Sensitivity of Phase

Constructing a Model: Simulating Extended Interfaces

Revealing the Signature: Bias Asymmetry and Oscillation Length

Beyond Symmetry: Future Directions

See also: