Superconducting Diode Effect

• The superconducting diode effect is a nonreciprocal phenomenon characterized by unequal critical currents due to broken inversion and time-reversal symmetry.
• Intrinsic mechanisms, such as finite-momentum pairing, lead to direction-dependent depairing thresholds and distinctive current-phase relations in superconductors.
• Extrinsic influences like vortex-mediated rectification and geometric chirality enhance diode performance, offering tunable and robust applications in superconducting devices.

The superconducting diode effect (SDE) is the nonreciprocity of dissipationless superconducting transport: the critical current for one current direction differs from that for the opposite direction, so that a device can remain superconducting in one bias polarity and become resistive in the other. In the standard notation, this is expressed as $|I_{c+}| \neq |I_{c-}|$, with a commonly used efficiency measure

$\eta=\frac{I_{c+}-|I_{c-}|}{I_{c+}+|I_{c-}|}.$

Since the experimental discovery by Ando et al. in noncentrosymmetric $[\mathrm{Nb}/\mathrm{V}/\mathrm{Ta}]_n$ superlattices, the subject has expanded into a broad research area spanning intrinsic finite-momentum pairing, vortex-mediated rectification, Josephson-junction physics, magnetic and topological heterostructures, and more recent zero-field, ultrafast, and quantum-dynamical variants (Ma et al., 17 Feb 2025, Nadeem et al., 2023).

1. Definition, symmetry framework, and diagnostic quantities

In its canonical form, SDE denotes a direction-dependent critical supercurrent, $I_{c+}\neq |I_{c-}|$, or equivalently a nonzero

$\Delta I_c=|I_{c+}|-|I_{c-}|.$

The effect can be identified either from direct-current transport near the superconducting transition or from asymmetric current-phase relations and nonreciprocal superfluid inductance deep in the superconducting phase (Nadeem et al., 2023). In the fluctuation regime, the transport language often overlaps with magnetochiral anisotropy, for which the resistance can be written as

$$R(I)=R_0\left[1+\beta B^2+\gamma(\boldsymbol{B}\times \boldsymbol{r})\cdot \boldsymbol{I}\right],$$

making explicit the odd-in-current, odd-in-field nonreciprocity (Nadeem et al., 2023).

The general symmetry principle emphasized in the reviews is that SDE usually requires simultaneous breaking of inversion symmetry and time-reversal symmetry. Broken inversion symmetry can be supplied by a noncentrosymmetric crystal, a Rashba or Ising interface, a polar heterostructure, or engineered geometry; time-reversal symmetry can be broken by magnetic field, magnetization, or, in some unconventional states, by the superconducting order itself (Ma et al., 17 Feb 2025). A frequently used phenomenological expression for the corresponding free-energy asymmetry is

$$F_L \sim f(\alpha,h)\,(\mathbf n_h\times \hat{\mathbf z})\cdot \mathbf v_s,$$

which encodes the fact that opposite superflow directions become inequivalent once spin-orbit coupling and Zeeman coupling are both active (Ma et al., 17 Feb 2025).

The same literature also stresses an important caveat. Several zero-field, $B$-even diode effects in polar systems do not fit the standard $B$-odd magnetochiral pattern and remain microscopically unresolved. The review identifies the zero-field responses reported in NbSe$_2$/Nb$\eta=\frac{I_{c+}-|I_{c-}|}{I_{c+}+|I_{c-}|}.$0Br$\eta=\frac{I_{c+}-|I_{c-}|}{I_{c+}+|I_{c-}|}.$1/NbSe$\eta=\frac{I_{c+}-|I_{c-}|}{I_{c+}+|I_{c-}|}.$2 and strained PbTaSe$\eta=\frac{I_{c+}-|I_{c-}|}{I_{c+}+|I_{c-}|}.$3 as challenges to the conventional symmetry-based understanding (Ma et al., 17 Feb 2025). This makes SDE a diagnostic of symmetry breaking, but not yet a universally interpreted one.

2. Intrinsic nonreciprocity and finite-momentum superconductivity

The intrinsic theory most directly associated with SDE is finite-momentum pairing. In that language the superconducting order parameter takes the form

$\eta=\frac{I_{c+}-|I_{c-}|}{I_{c+}+|I_{c-}|}.$4

so that the condensate already carries a preferred momentum $\eta=\frac{I_{c+}-|I_{c-}|}{I_{c+}+|I_{c-}|}.$5. Current applied parallel and antiparallel to $\eta=\frac{I_{c+}-|I_{c-}|}{I_{c+}+|I_{c-}|}.$6 then destabilizes the condensate differently, producing unequal depairing thresholds (Nadeem et al., 2023).

A microscopic formulation of this idea was developed for bulk noncentrosymmetric superconductors in the “Intrinsic Superconducting Diode Effect” theory. There the diode effect is identified with the nonreciprocity of the depairing current in a helical superconducting state. Near $\eta=\frac{I_{c+}-|I_{c-}|}{I_{c+}+|I_{c-}|}.$7, the nonreciprocal part scales as $\eta=\frac{I_{c+}-|I_{c-}|}{I_{c+}+|I_{c-}|}.$8, whereas at low temperature the effect is strongly enhanced and can even reverse sign as magnetic field changes, owing to the nonreciprocity of the Landau critical momentum and changes in the nature of helical superconductivity (Daido et al., 2021).

A complementary microscopic picture was obtained for quasi-one-dimensional superconductors, where the decisive low-energy ingredient is an inequality of the helical-band Fermi velocities. In that approach the critical current is defined by quasiparticle gap closing, and the diode efficiency is directly tied to the asymmetry of the critical condensate momentum. Quantum confinement is not merely a quantitative correction: it is the mechanism that generates $\eta=\frac{I_{c+}-|I_{c-}|}{I_{c+}+|I_{c-}|}.$9 and thereby enables intrinsic nonreciprocity in a Rashba wire (Picoli et al., 2023).

More recent theories broaden this intrinsic landscape. In $[\mathrm{Nb}/\mathrm{V}/\mathrm{Ta}]_n$0-wave altermagnets, very large intrinsic diode efficiencies appear in broad finite-momentum pairing regimes, and “perfect diode efficiency of $[\mathrm{Nb}/\mathrm{V}/\mathrm{Ta}]_n$1” can occur in the presence of magnetic field. The largest response is traced to competition between multiple zero-momentum and finite-momentum superconducting states near a topological nodal-to-nodeless transition of the altermagnetic Fermi surfaces (Chakraborty et al., 2024). In fully proximitized quantum spin Hall systems, the SDE can be externally controlled in wide wells and can reach $[\mathrm{Nb}/\mathrm{V}/\mathrm{Ta}]_n$2; in narrow wells an intrinsic field-free SDE can arise from edge reconstruction and asymmetric interedge tunneling. That work also makes a more restrictive point: inversion and time-reversal breaking are necessary but not sufficient, because the actual microscopic requirements are an asymmetric zero-bias superconducting gap structure in momentum space and a momentum-dependent spin texture (Fracassi et al., 2 Dec 2025).

Class	Representative mechanism	Representative papers
Bulk intrinsic SDE	Helical finite-momentum depairing asymmetry	(Daido et al., 2021)
Quasi-1D intrinsic SDE	Unequal helical-band Fermi velocities	(Picoli et al., 2023)
Altermagnetic SDE	Competition of BCS and FF-like states	(Chakraborty et al., 2024)
QSH intrinsic/tunable SDE	Edge spin texture and interedge tunneling	(Fracassi et al., 2 Dec 2025)

3. Geometric chirality, vortices, and extrinsic rectification

A large fraction of experimentally strong SDE is vortex-mediated rather than purely intrinsic. The most systematic cautionary statement is the claim that thin-film superconductors can exhibit a “ubiquitous” diode effect through the combination of asymmetric vortex edge or surface barriers and universal Meissner screening currents. In conventional V and Nb thin films, substantial nonreciprocity was observed under out-of-plane magnetic fields as small as $[\mathrm{Nb}/\mathrm{V}/\mathrm{Ta}]_n$3 Oe, and interfacing with EuS yielded a nonvolatile effect with efficiency reaching $[\mathrm{Nb}/\mathrm{V}/\mathrm{Ta}]_n$4. The paper’s main conclusion is that such Meissner-screening-induced SDE must be eliminated carefully before one interprets diode behavior as evidence for exotic finite-momentum pairing (Hou et al., 2022).

This vortex route can be deliberately engineered. In amorphous MoGe patterned with conformal nanohole arrays, the pinning landscape breaks in-plane inversion symmetry and produces direction-dependent vortex flow, hot-spot nucleation, and a current-driven superconducting-to-normal transition. The resulting rectification is in the millivolt range and is “three orders of magnitude larger” than a nominal flux-quantum diode signal (Lyu et al., 2021). A related extrinsic design uses a central superconducting strip flanked by two current-biased side wires. Numerical solutions of coupled time-dependent Ginzburg–Landau and heat-diffusion equations show that the optimum inhomogeneous field profile can facilitate vortex and antivortex entry in one current polarity while suppressing it in the other, realizing a superconducting half-wave rectifier with efficiencies “surpassing $[\mathrm{Nb}/\mathrm{V}/\mathrm{Ta}]_n$5” (Cadorim et al., 2024).

Geometry alone can also generate nonreciprocity in otherwise conventional superconductors. In a three-dimensional mesoscopic wedge, polarity-dependent Abrikosov vortex nucleation produces unequal critical currents; the effect is strongest at intermediate $[\mathrm{Nb}/\mathrm{V}/\mathrm{Ta}]_n$6 and low applied field, with efficiencies remaining below roughly $[\mathrm{Nb}/\mathrm{V}/\mathrm{Ta}]_n$7 in the parameter range studied (Aguirre et al., 2 Jun 2025). In a hollow superconducting helix, time-dependent Ginzburg–Landau simulations predict a diode effect arising solely from geometric chirality. For dirty-limit Nb at $[\mathrm{Nb}/\mathrm{V}/\mathrm{Ta}]_n$8, the helical geometry makes the magnetic-field-induced Meissner currents chiral and spatially asymmetric, giving efficiencies “over $[\mathrm{Nb}/\mathrm{V}/\mathrm{Ta}]_n$9” and a chirality-controlled crossover from screening-dominated to vortex-dominated nonreciprocity (Deenen et al., 17 Dec 2025).

The conceptual implication is straightforward. Not every SDE originates in an intrinsically nonreciprocal condensate; in many mesoscopic structures, direction-dependent vortex entry, screening-current superposition, and thermal feedback are sufficient.

4. Josephson diodes, multiterminal phase control, and topological hybrids

In Josephson systems the relevant object is the current-phase relation. A conventional junction obeys

$I_{c+}\neq |I_{c-}|$0

whereas a diode-capable junction generally requires a more general form such as

$I_{c+}\neq |I_{c-}|$1

so that the positive and negative critical currents are no longer equal in magnitude (Nadeem et al., 2023). An anomalous phase shift $I_{c+}\neq |I_{c-}|$2 alone does not guarantee a diode effect; the asymmetry requires a genuinely nonantisymmetric CPR.

Several microscopic platforms realize this principle. In QSHI-based Josephson junctions under out-of-plane magnetic field, finite Cooper-pair momentum shifts the helical-edge CPR differently on the two edges. A finite total diode response requires transport-inequivalent edges, and the low-temperature quality factor reaches a universal maximum $I_{c+}\neq |I_{c-}|$3 at $I_{c+}\neq |I_{c-}|$4; in the parity-conserving $I_{c+}\neq |I_{c-}|$5-periodic regime the effect is further enhanced (Scharf et al., 2024). In proximitized InSb/Al nanowire junctions, dual electrostatic control shows that the perpendicular-field diode effect is governed largely by the proximitized leads rather than only by the weak link. In the high-super-gate regime the efficiency is sinusoidal in angle, peaks near $I_{c+}\neq |I_{c-}|$6, and reaches about $I_{c+}\neq |I_{c-}|$7, which the paper interprets as a fingerprint of the spin-orbit-field direction in the proximitized leads (Mazur et al., 2022).

The multiterminal limit provides an even more stripped-down realization. A double quantum dot coupled to three superconducting leads exhibits a superconducting diode effect because the two dots form an Andreev molecule through crossed Andreev reflection in the common lead. The critical current into one branch becomes direction dependent whenever the second phase bias satisfies $I_{c+}\neq |I_{c-}|$8. In the noninteracting limit, Dirac cones appear in the $I_{c+}\neq |I_{c-}|$9-$\Delta I_c=|I_{c+}|-|I_{c-}|.$0 Andreev spectrum when the dot levels are tuned to the Fermi level, and the diode efficiency can approach $\Delta I_c=|I_{c+}|-|I_{c-}|.$1 near those Dirac points; the analogous single-dot device shows no SDE (Takeuchi et al., 2024). In S/F/TI hybrids, self-consistent quasiclassical calculations show that the linearized and nonlinear Usadel approaches differ mainly in absolute current scale, while the normalized diode quality factor remains comparatively robust (Karabassov et al., 2023).

5. Zero-field, ultrafast, quantum, and strictly two-dimensional forms

Zero-field SDE is a central experimental target because it removes the need for external magnetic bias and, in some cases, points to internal symmetry breaking. One route is magnetic multilayers. In noncentrosymmetric $\Delta I_c=|I_{c+}|-|I_{c-}|.$2, the remanent Co magnetization provides time-reversal breaking while the stacking sequence supplies inversion breaking, producing a field-free and nonvolatile diode effect whose polarity is controlled by magnetization direction (Narita et al., 2022).

A distinct zero-field regime is realized in twisted cuprate Josephson junctions. Under current training and microwave irradiation, these devices do not rectify between a superconducting branch and an ordinary quasiparticle-resistive branch; instead they switch between a zero-voltage superconducting state and quantized Shapiro-step states with

$\Delta I_c=|I_{c+}|-|I_{c-}|.$3

This “quantum superconducting diode” achieves perfect efficiency, $\Delta I_c=|I_{c+}|-|I_{c-}|.$4, operates at $\Delta I_c=|I_{c+}|-|I_{c-}|.$5, and is reported up to $\Delta I_c=|I_{c+}|-|I_{c-}|.$6 K, above liquid-nitrogen temperature (Wang et al., 29 Sep 2025).

The temporal limit has also shifted. In centrosymmetric NbN films biased with a quasi-DC supercurrent, picosecond current pulses aligned with the bias drive the strip across the depairing threshold and encounter a resistive impedance, whereas opposite-polarity pulses reduce the total current and see an inductive response dominated by kinetic inductance. This ultrafast nonreciprocal transport is demonstrated through rectification of a $\Delta I_c=|I_{c+}|-|I_{c-}|.$7 GHz bipolar signal with dissipation of a few fJ per cycle. The authors emphasize that this is functionally a superconducting diode, but not a canonical equilibrium supercurrent diode in the symmetry-based sense (Wang et al., 19 May 2026).

In the strict two-dimensional limit, the organizing concept itself changes. In Bi$\Delta I_c=|I_{c+}|-|I_{c-}|.$8Te$\Delta I_c=|I_{c+}|-|I_{c-}|.$9/FeSe$$R(I)=R_0\left[1+\beta B^2+\gamma(\boldsymbol{B}\times \boldsymbol{r})\cdot \boldsymbol{I}\right],$$0Te$$R(I)=R_0\left[1+\beta B^2+\gamma(\boldsymbol{B}\times \boldsymbol{r})\cdot \boldsymbol{I}\right],$$1, where superconductivity is established to be genuinely two-dimensional by BKT and Tinkham signatures, the diode effect appears not as a conventional direction-dependent critical current but as nonreciprocal vortex creep and nonreciprocal vortex-flow instability. At $$R(I)=R_0\left[1+\beta B^2+\gamma(\boldsymbol{B}\times \boldsymbol{r})\cdot \boldsymbol{I}\right],$$2 K and $$R(I)=R_0\left[1+\beta B^2+\gamma(\boldsymbol{B}\times \boldsymbol{r})\cdot \boldsymbol{I}\right],$$3 T, the efficiency at the abrupt voltage jump is about $$R(I)=R_0\left[1+\beta B^2+\gamma(\boldsymbol{B}\times \boldsymbol{r})\cdot \boldsymbol{I}\right],$$4, and the central claim is that in the 2D limit SDE should be understood as nonreciprocal dissipation of mobile vortices rather than only as a directional depairing threshold (Nagahama et al., 10 Oct 2025).

6. Materials landscape, applications, and unresolved problems

The present materials landscape spans noncentrosymmetric superlattices, Rashba and Ising interfaces, topological semimetal Josephson junctions, QSH heterostructures, thin films and vortex diodes, moiré graphene, kagome superconductors, cuprates, magnetic heterostructures, and geometrically chiral mesostructures (Ma et al., 17 Feb 2025, Nadeem et al., 2023). Across this landscape, SDE serves two rather different functions. One is technological: rectification, nonvolatile memory, logic, neuromorphic devices, quantum circuits, multilevel integrated cryogenic architectures, and potentially high-frequency superconducting signal processing. The other is spectroscopic: SDE has become an experimental probe of inversion breaking, time-reversal breaking, finite-momentum pairing, anomalous CPRs, and, in a narrower set of cases, candidate chiral or topological superconductivity (Ma et al., 17 Feb 2025).

Several problems remain open. The first is interpretive: strong diode behavior can arise from intrinsic helical pairing, from Josephson-phase engineering, or from entirely extrinsic vortex and screening-current physics. The second is the unresolved status of $$R(I)=R_0\left[1+\beta B^2+\gamma(\boldsymbol{B}\times \boldsymbol{r})\cdot \boldsymbol{I}\right],$$5-even zero-field diode effects in polar materials, which the 2025 review explicitly identifies as a challenge to current theory (Ma et al., 17 Feb 2025). The third is quantitative: many theories are controlled only near $$R(I)=R_0\left[1+\beta B^2+\gamma(\boldsymbol{B}\times \boldsymbol{r})\cdot \boldsymbol{I}\right],$$6, in the dirty limit, or without self-consistent electrodynamics, while several experiments operate in regimes where disorder, self-fields, pinning, thermal runaway, and geometry are decisive. A plausible implication is that the mature theory of SDE will remain mechanism-specific rather than universal.

The field has therefore moved beyond the question of whether superconducting diodes exist. The central questions are now how to classify them, how to distinguish intrinsic from extrinsic nonreciprocity, and how to optimize zero-field, high-temperature, and high-frequency operation without losing microscopic interpretability.