🔬 mesoscale physics

Floquet-Multiple Andreev Reflections

This paper demonstrates that voltage-biased three-terminal Josephson junctions on ballistic two-dimensional normal conductors exhibit characteristic finite-bias conductance and noise resonances arising from Floquet-multiple Andreev reflections driven by intrinsic time-periodic phase evolution.

Original authors: Régis Mélin, Romain Danneau, Morteza Kayyalha

Published 2026-05-05

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Régis Mélin, Romain Danneau, Morteza Kayyalha

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a superconductor as a super-highway where electrons travel in perfect pairs, like dancers holding hands. Usually, if you put a voltage on this highway, the dancers get stuck or scatter. But in this paper, the authors look at a special, three-way intersection of these super-highways (a "three-terminal Josephson junction") where something magical happens: the electrons start dancing to a new, rhythmic beat.

Here is the breakdown of their discovery using everyday analogies:

1. The Rhythm of the Highway (Floquet Theory)

Think of the voltage applied to the superconductors as a conductor waving a baton. Because the voltage is constant but the electrons are moving, the "phase" (the timing of the electron dance) changes periodically, like a clock ticking. In physics, this is called a Floquet drive. It's like the highway has a built-in metronome that forces the electrons to move in a repeating, time-based pattern, creating new "Floquet states" (new ways the electrons can exist).

2. The Bouncing Ball (Andreev Reflections)

Now, imagine a ball (an electron) rolling down a hill toward a wall (the superconductor). Instead of bouncing back as a ball, it turns into a "hole" (a missing electron) and bounces back the other way. This is called Andreev reflection.
In a normal junction, this happens once or twice. But in this complex three-way intersection, the ball bounces back and forth between the three different superconducting walls many times before finally escaping. This is called Multiple Andreev Reflection (MAR). It's like a pinball machine where the ball gets trapped in a loop, picking up energy and changing partners with every bounce.

3. The New Discovery: "Floquet-MAR"

The authors combined these two ideas. They found that when you have this rhythmic "metronome" (Floquet) driving the system while the electrons are bouncing around like pinballs (MAR), something special happens.

They call this Floquet-Multiple Andreev Reflection (Floquet-MAR).

The Quartet (The Group Dance): Usually, electrons move in pairs (charge 2e). But in this setup, the authors show that the system can move four electrons at once (charge 4e). They call this a "quartet." It's like four dancers linking arms and moving as a single unit, a feat that requires the specific rhythm of the three-way intersection.
The Octet and Beyond: They also found even larger groups (six, eight, or more electrons) moving together, which they call "octets" and higher-order multiplets.

4. The "Resonance" (The Sweet Spot)

The paper claims that if you tune the voltage and the "electrochemical potential" (which you can think of as the crowd density of electrons in the middle of the highway) to just the right numbers, these group dances become incredibly efficient.

They call these efficient moments resonances.

The Analogy: Imagine pushing a child on a swing. If you push at the wrong time, nothing happens. If you push at the exact right rhythm (resonance), the swing goes very high with very little effort.
The Result: The authors show that at these specific "sweet spots," the electrical conductance (how easily current flows) and the electrical noise (random fluctuations) spike in a very specific, predictable pattern. These spikes are the "fingerprints" of the Floquet-MAR process.

5. How They Proved It

The researchers didn't just guess this; they used a complex mathematical toolkit (Keldysh Green's functions) to map out the paths the electrons take.

They visualized these paths as "Andreev tubes" (tunnels where the electrons travel).
They calculated that when you measure the current's sensitivity to changes in electron density, you see distinct peaks.
They also calculated the Fano factor (a measure of how "noisy" the current is). They found that the noise is directly proportional to the size of the electron group. If 4 electrons move together, the noise is 4 times higher than if 1 moved alone. This proves the electrons are moving in coordinated, quantum-mechanical groups, not just randomly.

Summary

In simple terms, the paper describes a new way to make electrons dance in synchronized groups of four, six, or eight inside a superconducting wire. By applying a specific voltage rhythm, the electrons get trapped in a loop where they bounce back and forth, locking into a new, collective state. The authors provide a mathematical map showing exactly where to look (specific voltage settings) to see these "group dances" happening, proving that this complex quantum phenomenon is real and measurable.

Technical Summary: Floquet-Multiple Andreev Reflections in Ballistic Multiterminal Josephson Junctions

Problem Statement
The paper addresses the interplay between Floquet physics and Multiple Andreev Reflections (MARs) in superconducting weak links. While MARs in two-terminal and multiterminal Josephson junctions have been extensively studied, and Floquet theory has been applied to driven quantum systems (including voltage-biased Josephson junctions), the specific combination of these phenomena in multiterminal devices remains largely unexplored. In voltage-biased multiterminal junctions, the Josephson relation induces time-periodic evolution of superconducting phase differences, creating an intrinsic Floquet drive. The authors aim to develop a microscopic transport description for these systems to identify characteristic spectral features arising from the dressing of Floquet processes by MARs, termed "Floquet-MARs."

Methodology
The authors employ a theoretical framework combining three elements: a microscopic picture based on Andreev-tube trajectories, a spectroscopic theory of Floquet-MAR resonances in nonlinear conductance, and a quantum noise diagnostic.

System Model: The study focuses on three-terminal Josephson junctions (3TJJ) formed on a ballistic two-dimensional (2D) normal conductor. The superconducting leads ( $S_L, S_R, S_B$ ) are connected to a rectangular ballistic 2D conductor. The system is treated in the large-system limit ( $L_x, L_y \to \infty$ ), where the single-particle level spacing is negligible, allowing the normal region to be modeled as having a continuum spectrum.
Hamiltonian: The superconducting leads are described by standard BCS Hamiltonians, and the 2D conductor by tight-binding Hamiltonians (hexagonal or square lattices). The coupling is modeled via standard interfacial hopping amplitudes.
Approximation: Calculations utilize a "large-gap approximation," which captures the low-energy Floquet-MAR resonance structure without describing the full finite-gap MAR ladder of conventional junctions. This approximation is noted to capture the essential resonance structure at low bias and small interface transparency.
Formalism: The authors use Keldysh Green's functions to calculate the current susceptibility $\chi_I = \partial I(eV, \mu_N)/\partial \mu_N$ and quantum noise cross-correlations. The electrochemical potential $\mu_N$ of the normal region serves as an independent spectroscopic control parameter alongside the bias voltages ( $V_L, V_R$ ) and magnetic flux.
Diagrammatic Approach: The transport is analyzed using diagrammatic expansions involving "snake diagrams" that represent higher-order multipair processes (quartets, octets, etc.). The averaging over Fermi wavelength oscillations in the multichannel ballistic continuum is performed to extract the resonant features.

Key Contributions and Results

Definition of Floquet-MARs: The authors define Floquet-MAR resonances as finite-bias resonant features in differential current and quantum noise susceptibilities. These arise from Floquet processes dressed by MAR-like energy shifts. Unlike conventional MARs which connect subgap states to continuum quasiparticle states, these processes remain confined to the low-energy sector.
Spectral Signatures of Multipair Processes:
- Q-Quartets (Dimers): The elementary quartet process (involving two Cooper pairs from the grounded lead $S_B$ recombining into $S_L$ and $S_R$ ) yields four Floquet-MAR resonances in the spectral current susceptibility at energies $\Omega = \pm eV$ and $\mp eV$ (relative to $\mu_N$ ). The current susceptibility is shown to be proportional to the sum of equilibrium spectral currents at these Floquet replica energies.
- O-Octets and Q'-Quartets (Higher-order): The paper extends the analysis to higher-order "snake diagrams." O-octets (involving six Cooper pairs) and Q'-quartets yield more complex spectra. Specifically, O-octets produce resonances at $\mu_N = 0, \pm 2eV$ , while Q'-quartets produce resonances at $\mu_N = \pm eV, \pm 3eV$ .
Current Susceptibility Maps: The authors present calculated maps of the dimensionless current susceptibility $\partial i'(eV, \mu_N)/\partial \mu_N$ as a function of $\log(eV)$ and $\log(\mu_N)$ . These maps reveal distinct resonance lines corresponding to the Floquet-MAR multiplets (Q, O, and Q' channels). The results indicate that disorder averaging or multichannel averaging is necessary for these finite-bias resonances to appear; in a fully ballistic single-channel limit without averaging, the resonances vanish.
Quantum Noise and Fano Factor: The paper calculates the quantum noise cross-correlations between contacts. A key result is the derivation of the Fano factor $F_N = \partial S_{L,R}/\partial \mu_N / \partial I_L/\partial \mu_N = 2N$ , where $N$ is the order of the snake diagram (e.g., $N=1$ for quartets, $N=2$ for octets). This linear relationship demonstrates that the quantum noise cross-correlations are proportional to the dissipative current, highlighting the granularity and quantum coherence of the Floquet-MAR resonances. The resulting noise is orders of magnitude larger than Landau-Zener-Stückelberg noise in superconducting quantum dots.

Significance and Claims
The paper claims to provide a microscopic picture that opens a route toward implementing and probing Floquet-MAR physics in ballistic multiterminal Josephson junctions. The primary significance lies in:

Identifying a new class of nonequilibrium states (Floquet-MARs) that are distinct from standard MARs and conventional Floquet states.
Proposing that multiterminal Josephson junctions can serve as a controllable platform for studying nonequilibrium Floquet-Andreev spectra.
Demonstrating that the electrochemical potential $\mu_N$ acts as a spectroscopic control parameter, allowing the resolution of the energy spectrum of intermediate multipair states (quartets, octets, etc.) via current susceptibility spectroscopy.
Establishing a direct link between the order of the multipair process and the Fano factor of the quantum noise, providing a diagnostic tool for correlated multipair transport.

The authors maintain a modest tone, noting that their findings "open up the possibility" to engineer Floquet states in Josephson junctions based on 2D conductors, rather than claiming immediate experimental realization. They emphasize that the theoretical framework relies on the large-gap approximation and specific geometric limits (ballistic continuum), which define the scope of the predicted phenomena.