Pith. sign in

REVIEW 3 major objections 3 minor 50 references

A diffusive Josephson junction with a Rashba metal layer and an exchange field produces a spin Josephson diode effect without any external magnetic field, by combining a phi0-shifted first-harmonic spin current with a second-harmonic spin c

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-04 17:04 UTC pith:4QSQ7U47

load-bearing objection The qualitative SJDE mechanism is plausible, but the second-harmonic amplitude that drives the diode effect is not computed reliably; referee time is warranted but the numbers are not yet trustworthy. the 3 major comments →

arxiv 2608.01991 v1 pith:4QSQ7U47 submitted 2026-08-03 cond-mat.supr-con cond-mat.mes-hall

Spin Josephson diode effect induced by higher-harmonic spin Josephson currents in a diffusive Josephson junction

Shin-ichi Hikino This is my paper
classification cond-mat.supr-con cond-mat.mes-hall
keywords spin Josephson diode effectspin Josephson currentRashba spin-orbit interactionUsadel equationquasiclassical Green's functionphi0 junctionnonreciprocal spin transportspin-triplet correlations
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper tries to establish that spin supercurrent can act as a diode—flowing more easily in one direction than the other—without an external magnetic field. The proposed system is a diffusive Josephson junction whose middle layer is a normal metal with Rashba spin-orbit coupling, flanked by ferromagnetic layers and s-wave superconductors, with an exchange field present. Using a perturbative quasiclassical Green's-function analysis near the superconducting transition temperature, the author derives the first- and second-harmonic spin Josephson currents and shows that both acquire a phi0 phase shift and cosine terms from the combined Rashba/exchange interaction. The coexistence of those two harmonics breaks the antisymmetry I(θ) = -I(-θ) in the spin current-phase relation, so forward and backward critical spin currents become unequal. A sympathetic reader would care because this offers a route to nonreciprocal spin transport that needs neither an applied magnetic field nor suppression of the spin-singlet superconducting component.

Core claim

Within the quasiclassical Green's function framework, the y-component of the spin Josephson current in a diffusive S/F/RM/F/S junction is shown to be the sum of a first harmonic J_y^(1) ∝ [sin(θ+φ0) - ξαR cos(θ+φ0)] exp[-(df+dm)/ξ] and a second harmonic J_y^(2) ∝ (D_L^x)^2/(ξαR) [sin(2θ+2φ0) - 3 cos(2θ+2φ0)] exp[-2(df+dm)/ξ], with φ0 = 2 αR dm. The φ0 shift and the cosine terms appear even with zero external magnetic field, because the Rashba spin-orbit interaction breaks inversion symmetry while the exchange field breaks time-reversal symmetry. The second-harmonic term is not put in by hand; it is generated by the nonlinear coupling of first-order anomalous Green's functions in the Usadel e

What carries the argument

The carrying mechanism is the nonlinear Usadel equation solved perturbatively: the anomalous Green's function is expanded as f^(1) + f^(2), where f^(1) obeys a linearized equation and f^(2) is driven by products of f^(1) and its conjugate. The covariant derivative in the Rashba metal includes a spin-orbit gauge term that gives spin-dependent phase accumulation, leading to φ0 = 2αR dm; the exchange field converts singlet correlations into triplet correlations. The spin current is evaluated with the SU(2) covariant formula, which yields the y-component as the only one with a genuine 2θ dependence. The essential identity is that both harmonics are φ0-shifted with sine-plus-cosine form, and thei

Load-bearing premise

The load-bearing premise is that the second-harmonic anomalous Green's function can be computed with homogeneous boundary conditions—so only the particular solution of the second-order Usadel equation contributes—and that no second-harmonic Cooper pairs are injected at the interfaces; if that condition fails, the second-harmonic amplitude and phase, and hence the diode efficiency, change.

What would settle it

Solve the second-order Usadel equation (Eq. 10) for f^(2) in the Rashba metal with the full physical boundary conditions at x = L_f and x = L_m instead of homogeneous ones, and compare the resulting 2θ-component of J_y with Eq. (42); a substantial boundary correction would undermine the predicted efficiency. Experimentally, measure the spin current-phase relation of an S/F/RM/F/S junction and look for unequal forward and backward critical spin currents at zero applied magnetic field, with the sign of the asymmetry reversing under θ → θ + π.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • The spin Josephson diode effect appears in a diffusive junction with no external magnetic field, using only Rashba spin-orbit coupling plus an exchange field.
  • The diode works without suppressing spin-singlet correlations, so the material constraints are less severe than in charge JDE setups that rely on a thick ferromagnet to kill the singlet component.
  • The efficiency ηs is controlled by the Rashba-metal thickness and the spin-orbit strength; thinner Rashba layers and moderate αR give higher efficiency because the second harmonic is exponentially suppressed with thickness and inversely proportional to αR.
  • Only the y-component of the spin current shows the diode effect in this magnetization geometry; the x and z components remain first-harmonic and reciprocal.
  • The estimate near Tc is conservative; the author expects larger superconducting gaps at lower temperatures to increase both the spin current and the higher-harmonic contribution, raising the efficiency.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the mechanism holds, any junction in which Rashba spin-orbit interaction and an exchange field coexist should show some degree of spin-diode behavior; a quantitative prediction is that the diode direction reverses when the phase θ is shifted by π, which could be tested in phase-biased devices.
  • Because the second-harmonic term is inversely proportional to αR while the first-harmonic phase shift grows with αR, the model predicts a non-monotonic optimal spin-orbit strength for the diode; identifying that optimum in a material scan would provide a sharp test of this mechanism.
  • The argument suggests that interface engineering that deliberately injects second-harmonic pairs (relaxing the homogeneous boundary condition) could tune or enhance the diode efficiency; the paper's assumption can be tested by solving the boundary-value problem with full interface conditions.
  • If the magnetization is rotated so that the exchange field acquires a y-component, the x and z spin-current components should develop second-harmonic terms as well; this implies the diode-active spin direction can be selected by magnetization orientation, which the paper hints at but does not calculate.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

3 major / 3 minor

Summary. The paper studies a diffusive S/F/F/RM/F/S Josephson junction and, within the quasiclassical Usadel framework near T_c, derives analytic expressions for the first- and second-harmonic spin Josephson currents (Eqs. (41) and (42)). It finds a φ0=2α_R d_m phase shift induced by the combined Rashba spin-orbit interaction and exchange field, and argues that the coexistence of the φ0-shifted first- and second-harmonic currents produces a spin Josephson diode effect (SJDE) without an external magnetic field and without suppressing spin-singlet correlations. Numerical results for the SHSJC and diode efficiency η_s are presented in Figs. 2-8.

Significance. If established, the proposed mechanism is a useful alternative to charge-JDE schemes: it does not require spin-singlet suppression, gives closed-form current-phase relations, and makes parameter-free predictions for the dependence of η_s on d_m and α_R. The symmetry reasoning (φ0 shift plus second harmonic implies I(θ) ≠ -I(-θ)) is standard and sound, and the paper is commendably explicit about its approximations. The quantitative central claim, however, rests on the unverified treatment of the second-order boundary-value problem and on a purely imaginary spin-current expression; these issues must be resolved before the mechanism can be considered demonstrated.

major comments (3)
  1. [Section 2.3, Eq. (10) and Eqs. (29)-(30)] The statement 'only the particular solution contributes, while the homogeneous solution vanishes' is not valid for the inhomogeneous linear ODE (10). The general solution also contains homogeneous solutions of order (Δ/ω)^3, which are generally needed to satisfy the homogeneous boundary conditions at x=L_f and L_m. The particular solution (29)-(30) does not vanish at these interfaces, so it cannot satisfy f^(2)=0 by itself. Consequently, the amplitude and phase of the SHSJC in Eq. (42), and hence η_s in Eqs. (43)-(44) and Figs. 7-8, are not established. Please solve the boundary-value problem for f^(2) or prove explicitly that the particular solution obeys the interface conditions.
  2. [Section 3, Eqs. (41)-(42) and Discussion] J_y^(1),(2) are explicitly purely imaginary (factor i). The paper calls this representation-dependent and refers to 'appropriate real combinations' of Green's functions, but no real observable is constructed. A physical spin current must be Hermitian; defining J_s in Eq. (43) and η_s in Eq. (44) with an imaginary spin current is not meaningful without a concrete prescription (e.g., summing over all Matsubara frequencies, taking the real part at the operator level, or using a different spin-current definition). Please provide the real observable and verify that the diode effect persists for that quantity.
  3. [Eq. (42) and Section 4.1] The SHSJC amplitude is proportional to 1/(ξα_R), which diverges as α_R→0. Since the perturbative treatment requires f^(2) to be a small correction to f^(1), and f^(1) remains finite for α_R→0, the 1/α_R scaling is inconsistent with a regular limit. The paper should specify the regime of validity of this approximation and verify that f^(2)≪f^(1) for the parameters used in Figs. 7-8.
minor comments (3)
  1. [Section 2.2, Eqs. (25)-(27)] The RM-region functions are written with x-d_f, whereas RM is defined as L_f<x<L_m with L_f=d_L+d_f. If d_L is approximated to zero in the final expressions, this should be stated; otherwise the coordinate origin is inconsistent.
  2. [Section 2.1, Eq. (8)-(10)] The bookkeeping of the harmonic decomposition is not shown explicitly: how the 'second-order' component f^(2) gives rise to a 4π-periodic current is explained only after the fact. A short derivation or a table of terms at each order would improve readability.
  3. [Discussion] The statement that 'significantly larger efficiencies are expected at lower temperatures' is speculative and not supported by a calculation. Either provide a numerical estimate or soften the claim.

Circularity Check

0 steps flagged

No significant circularity: the φ0 shift and harmonic spin currents emerge from explicit Usadel-equation solutions, not from fitted inputs or self-citation; the homogeneous-boundary-condition gap is a correctness concern, not circularity.

full rationale

The paper's central claims are derived analytically from the nonlinear Usadel equation (Eqs. 1, 7) via a perturbative expansion in Δ/ω_n (Eqs. 8–10). The first-order Green's functions are obtained by solving Eq. (9) with explicit boundary conditions (Eqs. 12–19), giving the RM solutions (25)–(27). The second-order Green's functions are obtained from the inhomogeneous Eq. (10), with source terms built from products of first-order functions. The φ0 shift, φ0 = 2α_R d_m, emerges from the spin-dependent phase factors exp(±i2α_R x) in the RM general solution (Eq. 24); it is not injected as a fit or an ansatz. Equations (41) and (42) for the first- and second-harmonic spin Josephson currents are obtained by direct substitution into the SU(2) covariant spin-current formula (Eqs. 33–34). No parameter is fitted to data, and the SJDE efficiency (Eqs. 43–44) is a definition applied to the derived current–phase relation, not a fitted prediction. Self-citations [26,33,47] are used for context on the charge Josephson diode effect and for technical details of boundary-condition matching; the essential equations are restated in the paper, so these citations are not load-bearing for the present derivation. The assertion in Sec. 2.3 that 'only the particular solution contributes, while the homogeneous solution vanishes' is a mathematical gap that could affect the quantitative validity of Eq. (42), but it is not a circular step: the paper does not define the second-harmonic current as the particular solution, nor does it use the final diode efficiency to determine any input. That concern belongs under correctness risk, not circularity.

Axiom & Free-Parameter Ledger

0 free parameters · 6 axioms · 0 invented entities

The central calculation is a perturbative derivation from the Usadel equation with no fitted constants and no new physical entities. The main extras are standard approximations (rigid boundary conditions, n=0 Matsubara, geometric limits) plus one ad hoc modeling choice: homogeneous boundary conditions for the second-order Green's function, which determines the SHSJC and hence the diode effect.

axioms (6)
  • domain assumption Quasiclassical Usadel equation with the covariant derivative applies in the diffusive limit (Eqs. (1)-(2)).
    Standard theory for diffusive superconducting heterostructures, cited to Ref. [45]; enters at the start of Section 2.
  • domain assumption Normalization condition g^2 = 1 and rigid boundary conditions at the S electrodes (Eqs. (12)-(19), (22)).
    Standard quasiclassical boundary conditions; rigid boundary condition assumes ideally transparent interfaces, acknowledged in the Discussion.
  • domain assumption Only the n=0 Matsubara frequency is retained (near Tc, Delta/omega_n << 1).
    Section 2.1 states this simplification, standard for near-Tc perturbative treatments.
  • domain assumption Geometric and parameter limits: d_L(R)/xi << 1, d_f/xi >> 1, d_m/xi >> 1, and xi alpha_R << 1 with alpha_R != 0.
    Used to obtain the simplified Green's functions in Eqs. (25)-(30) and the final current expressions.
  • ad hoc to paper Homogeneous boundary conditions on the second-order anomalous Green's function: no second-harmonic injection at interfaces and vanishing homogeneous solution.
    Section 2.3 imposes this to isolate the bulk-induced SHSJC; it is a modeling choice specific to this paper, asserted without proof.
  • domain assumption Spin current defined via Tokatly's SU(2) covariant formula, Eq. (31).
    Standard in the field, cited to Ref. [48]; under this definition J_y is purely imaginary, which the paper argues is a representation artifact.

pith-pipeline@v1.3.0-daily-deepseek · 18694 in / 17251 out tokens · 178860 ms · 2026-08-04T17:04:53.721587+00:00 · methodology

0 comments
read the original abstract

We theoretically investigate the spin Josephson diode effect (SJDE) in a diffusive Josephson junction with a Rashba metal layer under a ferromagnetic exchange field. Within the quasiclassical Green's function framework, we derive analytical expressions for the first- and second-harmonic spin Josephson currents. The interplay between Rashba spin-orbit interaction and the exchange field breaks inversion and time-reversal symmetries, generating additional cosine terms in the spin current-phase relations and a finite phi0 phase shift. This phase shift induces an intrinsic asymmetry between forward and backward spin currents, leading to the SJDE without an external magnetic field. Numerical results show that the efficiency decreases with increasing metal thickness due to suppression of the second-harmonic component, while its dependence on spin-orbit interaction strength reflects competing effects between phase shift enhancement and harmonic suppression. These findings demonstrate that the interplay between harmonic components provides a mechanism for nonreciprocal spin transport without requiring suppression of spin-singlet correlations.

Figures

Figures reproduced from arXiv: 2608.01991 by Shin-ichi Hikino.

Figure 1
Figure 1. Figure 1: The schematic diagram of the SL/FL/F/RM/FR/SR junction is shown, where SL (SR) denotes the spin-singlet superconductor on the left (right), FL (FR) the thin ferromagnetic metal on the left (right), F the thick ferromagnetic metal, and normal metal with the RSOI (RM). The thick ferromagnet F serves as a barrier that suppresses the contribution of spin-singlet Cooper pairs. The total thicknesses of different… view at source ↗
Figure 2
Figure 2. Figure 2: First-harmonic spin Josephson current (FHSJC) as a function of the thickness of the Rashba metal (dm) for ˜α = ξαR = 0.1, 0.3, and 0.5. Panels (a)–(c) correspond to θ = 0, π/3, and π/2, respectively. For weak Rashba spin-orbit interaction (˜α = 0.1), the current exhibits a monotonic decay with increasing dm. As ˜α increases, a damped oscillatory behavior develops, and even for ˜α = 0.3, a sign change can b… view at source ↗
Figure 3
Figure 3. Figure 3: Second-harmonic spin Josephson current (SHSJC) as a function of the thickness of the Rashba metal (dm) for ˜α = ξαR = 0.1, 0.3, and 0.5. Panels (a)–(c) correspond to θ = 0, π/3, and π/2, respectively. The SHSJC is plotted on a logarithmic scale in terms of its absolute value. For all values of ˜α, the SHSJC exhibits a damped oscillatory behavior as a function of dm. The logarithmic representation highlight… view at source ↗
Figure 4
Figure 4. Figure 4: The first- and second-harmonic spin Josephson currents, J (1) s and J (2) s , as functions of the dimensionless Rashba spin–orbit interaction strength ˜α for θ = 0, π/3, and π/2. Panels (a) and (b) show the first- and second-harmonic spin Josephson currents, respectively. Both components change sign as ˜α is varied, while the second-harmonic component exhibits a more pronounced dependence. The pronounced ˜… view at source ↗
Figure 5
Figure 5. Figure 5: The spin current–phase relation (SCPR) of the first-harmonic spin Josephson current (FHSJC) for α˜ = 0.1, 0.3, and 0.5. The FHSJC is normalized by its maximum value. A sizable phase shift appears in the SCPR depending on ˜α. The φ0-phase shift provides a necessary ingredient for the SJDE, although an additional second-harmonic contribution is required to generate nonreciprocity. superconducting phase diffe… view at source ↗
Figure 6
Figure 6. Figure 6: The spin current–phase relation (SCPR) of the second-harmonic spin Josephson current (SHSJC) for α˜ = 0.1, 0.3, and 0.5. The SHSJC is normalized by its maximum value. A sizable phase shift appears in the SCPR depending on ˜α. The coexistence of the second-harmonic contribution and the φ0-phase shift constitutes the microscopic origin of the SJDE in the present junction [PITH_FULL_IMAGE:figures/full_fig_p0… view at source ↗
Figure 7
Figure 7. Figure 7: The efficiency of the spin Josephson diode effect (ηs) as a function of the thickness of the Rashba metal (dm) for ˜α = ξαR = 0.1, 0.3, and 0.5. ηs decreases monotonically with increasing dm and is enhanced as ˜α decreases. Here, ξ = p ℏD/(2πkBT). The finite value of ηs directly demonstrates the emergence of the SJDE in the present junction. components breaks the antisymmetry of the SCPR and plays a fundam… view at source ↗
Figure 8
Figure 8. Figure 8: The efficiency of the spin Josephson diode effect (ηs) as a function of the dimensionless Rashba spin-orbit interaction strength (˜α = ξαR) for dm/ξ = 2, 2.5, and 3. ηs increases with decreasing ˜α and decreases as dm increases. Here, ξ = p ℏD/(2πkBT). The strong dependence of ηs on ˜α highlights the essential role of the Rashba spin-orbit interaction in controlling the efficiency of the SJDE. The physical… view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

50 extracted references

  1. [1]

    Lett.1251

    Josephson B D 1962Phys. Lett.1251

  2. [2]

    de Gennes P G 1964Rev. Mod. Phys.36225

  3. [3]

    Likharev K K 1979Rev. Mod. Phys.51101

  4. [4]

    Golubov A A, Kupriyanov M Yu and Il’ichev E 2004Rev. Mod. Phys.76411

  5. [5]

    Buzdin A I 2005Rev. Mod. Phys.77935

  6. [6]

    Bergeret F S, Volkov A F and Efetov K B 2005Rev. Mod. Phys.771321

  7. [7]

    Chen C -Z, He J J, Ali M N, Lee G -H, Fong K C and Law K T 2018Phys. Rev. B98075430

  8. [8]

    Nanotechnol.1739

    Baumgartner C, Fuchs L, Costa A, Reinhardt S, Gronin S, Gardner G C, Lindemann T, Manfra M J, Faria Junior P E, Kochan D, Fabian J, Paradiso N and Strunk C 2022Nat. Nanotechnol.1739

  9. [9]

    Phys.18 1228

    Pal B, Chakraborty A, Sivakumar P K, Davydova M, Gopi A K, Pandeya A K, Krieger J A, Zhang Y, Date M, Ju S, Yuan N, Schr¨ oter N B M, Fu L and Parkin S S P 2022Nat. Phys.18 1228

  10. [10]

    Souto R S, Leijnse M and Schrade C 2022Phys. Rev. Lett.129267702

  11. [11]

    Wu H, Wang Y, Xu Y, Sivakumar P K, Pasco C, Filippozzi U, Parkin S S P, Zeng Y -J, McQueen T and Ali M N 2022Nature604653

  12. [12]

    Zhang Y, Gu Y, Li P, Hu J and Jiang K 2022Phys. Rev. X12041013

  13. [13]

    Ciaccia C, Haller R, Drachmann A C C, Lindemann T, Manfra M J, Schrade C and Sch¨ onenberger C 2023Phys. Rev. Research5033131

  14. [14]

    Tian W, Zhang H, Zhang D and Shao W 2023Appl. Phys. Lett.123112601

  15. [15]

    Commun.143078

    Gupta M, Graziano G V, Pendharkar M, Dong J T, Dempsey C P, Palmstrøm C and Pribiag V S 2023Nat. Commun.143078

  16. [16]

    Greco A, Pichard Q and Giazotto F 2023Appl. Phys. Lett.123092601

  17. [17]

    Chiles J, Arnault E G, Chen C C, Larson T F Q, Zhao L, Watanabe K, Taniguchi T, Amet F and Finkelstein G 2023Nano Lett.235257

  18. [18]

    Wei Y J, Wang J J and Wang J 2023Phys. Rev. B108054521

  19. [19]

    Hu J X, Sun Z T, Xie Y M and Law K T 2023Phys. Rev. Lett.130266003

  20. [20]

    Commun.151120

    Kim J K, Jeon K R, Sivakumar P K, Jeon J, Koerner C, Woltersdorf G and Parkin S S P 2024 Nat. Commun.151120

    2024

  21. [21]

    Li S, Deng Y, Hu D, Zhu C, Yang Z, Tian W, Wang X, Yue M, Wu Q, Liu Z and Wang X R 2024ACS Nano1831076 16 IOP PublishingJournalvv(yyyy) aaaaaa Authoret al

  22. [22]

    Zhang F, Rashid A S, Ahari M T, de Coster G J, Taniguchi T, Watanabe K, Gilbert M J, Samarth N and Kayyalha M 2024Phys. Rev. Applied21034011

  23. [23]

    Coraiola M, Svetogorov A E, Haxell D Z, Sabonis D, Hinderling M, Kate S C T, Cheah E, Krizek F, Schott R, Wegscheider W, Cuevas J C, Belzig W and Nichele F 2024ACS Nano18 9221

  24. [24]

    Huang H, de Picoli T and V¨ ayrynena J I 2024Appl. Phys. Lett.125032602

  25. [25]

    Fracassi S, Traverso S, Ziani N T, Carrega M, Heun S and Sassetti M 2024Appl. Phys. Lett. 124242601

  26. [26]

    Hikino S 2025J. Phys. Soc. Jpn.94094701

  27. [27]

    Buzdin A I 2008Phys. Rev. Lett.101107005

  28. [28]

    Goldobin E, Koelle D, Kleiner R and Mints R G 2011Phys. Rev. Lett.107227001

  29. [29]

    Lett.11057005

    Bergeret F S and Tokatly I V 2015Europhys. Lett.11057005

  30. [30]

    Phys.12568

    Szombati D B, Nadj-Perge S, Car D, Plissard S R, Bakkers E P A M and Kouwenhoven L P 2016Nat. Phys.12568

  31. [31]

    Mayer W, Dartiailh M C, Yuan J, Wickramasinghe K S, Rossi E and Shabani J 2020Nat Commun11212

  32. [32]

    Nanotechnol.15656

    Strambini E, Iorio A, Durante O, Citro R, Sanz-Fernandez C, Guarcello C, Tokatly I V, Braggio A, Rocci M, Ligato N, Zannier V, Sorba L, Bergeret F S and Giazotto F 2020Nat. Nanotechnol.15656

  33. [33]

    Hikino S 2022Physica C5921354003

  34. [34]

    Ryazanov V V, Oboznov V A, Rusanov A Yu, Veretennikov A V, Golubov A A and Aarts J 2001Phys. Rev. Lett.862427

  35. [35]

    Kontos T, Aprili M, Lesueur J and Grison X 2001Phys. Rev. Lett.86304; Kontos T, Aprili M, Lesueur J, Genˆ et F, Stephanidis B and Boursier R 2002Phys. Rev. Lett.89137007

  36. [36]

    Bergeret F S, Volkov A F and Efetov K B 2001Phys. Rev. Lett.864096

  37. [37]

    Volkov A F, Bergeret F S and Efetov K B 2003Phys. Rev. Lett.90117006

  38. [38]

    Bergeret F S, Volkov A F and Efetov K B 2003Phys. Rev. B68064513

  39. [39]

    Houzet M and Buzdin A I 2007Phys. Rev. B76060504(R)

  40. [40]

    Volkov A F and Efetov K B 2010Phys. Rev. B81144522

  41. [41]

    Robinson J W A, Witt J D S and Blamire M G 2010Science32959

  42. [42]

    and Birge N O 2010Phys

    Khaire T S, Khasawneh M A, Pratt W P, Jr. and Birge N O 2010Phys. Rev. Lett.104 137002; Klose C, Khaire T S, Wang Y, Pratt W P, Jr., Birge N O, McMorran B J, Ginley T P, Borchers J A, Kirby B J, Maranville B B and Unguris J 2012Phys. Rev. Lett.108127002

  43. [43]

    Schulz N L, Nikoliˇ c D and Eschrig M 2025Phys. Rev. B112104515

  44. [44]

    Sun C, Tjernshaugen J B and Linder J 2025Phys. Rev. B112064504

  45. [45]

    Richard C, Houzet M and Meyer J S 2013Phys. Rev. Lett.110217004

  46. [46]

    Demler E A, Arnold G B and Beasley M R 1997Phys. Rev. B5515174

  47. [47]

    Hikino S and Yunoki S 2015Phys. Rev. B92024512

  48. [48]

    Tokatly I V 2017Phys. Rev. B96060502(R)

  49. [49]

    Tserkovnyak Y, Brataas A, Bauer G E W and Halperin B I 2005Rev. Mod. Phys.771375

  50. [50]

    Sinova J, Valenzuela S O, Wunderlich J, Back C H and Jungwirth T 2010Rev. Mod. Phys.87 1213 17