{"id":"4edac98c-7153-4c76-96fa-47197250a104","arxiv_id":"2411.15807","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Spin-splitting fields can more than double the Josephson critical current in long superconductor-normal-superconductor junctions and lower the voltage needed for π-transitions.","lead":"A theoretical study shows that spin-splitting fields in the superconducting leads of a Josephson junction can increase the critical supercurrent by over 100% in long junctions at low temperatures, for both parallel and antiparallel alignments. The same setup offers voltage-controlled π-transitions at lower voltages and sharp supercurrent drops under tiny temperature biases, steps toward superconducting transistor devices.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The >100% enhancement in Fig. 2(e) may be an artifact of the inelastic broadening δ/Δ0=0.01, which equals the Thouless energy of the L=10ξ junction; no δ-dependence is shown.","rationale":"The paper is a standard Usadel/Riccati calculation with clearly stated parameters, and the short-junction benchmark against Ref. [28] provides genuine independent support for the numerical implementation. I agree with the reader's CONDITIONAL verdict and with the identification of Fig. 2(e) as the central claim. However, the reader's stated weakest assumption—fixed Δ near the Clogston limit—does not appear to be the decisive issue. At T/Tc=0.01 the BCS gap is essentially Δ0, and for a purely Zeeman-split superconductor at low temperature the self-consistent order parameter remains approximately Δ0 up to the first-order Clogston transition at h_s/Δ0≈0.707. The paper's own caveat about self-consistency is explicitly tied to T/Tc=0.5, not to the T/Tc=0.01 panels. The more unexamined numerical control is δ/Δ0=0.01, which coincides with the Thouless energy of the L=10ξ junction. This means the zero-field reference current in the denominator of the enhancement ratio is governed by an arbitrary inelastic broadening that is not small compared with the relevant spectral scale, and the quoted enhancement can be sensitive to that choice. The concrete δ-sweep proposed above would settle whether the >100% enhancement is physical or an artifact. Since this is a numerical sensitivity check rather than a proven failure, the appropriate verdict remains CONDITIONAL, unchanged from the reader's assessment.","tokens_in":12414,"tokens_out":15915,"duration_ms":158468,"concrete_test":"Recompute Fig. 2(e) (L=10ξ, T/Tc=0.01, P and AP) for δ/Δ0 in {0.001, 0.003, 0.01, 0.03} with all other parameters fixed, and record I_c(h_s/Δ0=0.7)/I_c(h_s=0). If the ratio drops below 2 (or varies by more than 50%) when δ is reduced below E_Th/Δ0=0.01, the >100% claim is not robust. As a second check, run L=20ξ at δ/Δ0=0.01: if the enhancement disappears, the L=10ξ result is tied to δ≈E_Th rather than to the long-junction physics.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim (Sec. III A, Fig. 2(e)) rests on a numerical calculation with δ/Δ0=0.01 added to quasiparticle energies 'to model inelastic scattering.' For a diffusive junction with L=10ξ and ξ=sqrt(D/Δ0), the Thouless energy is E_Th/Δ0=(ξ/L)^2=0.01, so δ is equal to, not small compared with, the dominant low-energy scale of the long junction. In Eq. (13) the integrand J_s(ε) is concentrated at ε≈E_Th for L=10ξ, and setting δ≈E_Th changes the low-energy Andreev spectrum. At h_s=0 the zero-field supercurrent is controlled by a decay length sqrt(D/δ)=10ξ=L in the normal metal, so the denominator of the enhancement ratio is itself set by the artificial broadening; at h_s=0.7 the balance changes because one spin channel sees an effective gap Δ0−h_s. No δ-dependence is reported, and no argument is given that δ=0.01 is small for this L. If the zero-field current is suppressed more than the h_s=0.7 current by δ, the quoted >100% enhancement is inflated. This is a more direct threat than the non-self-consistent Δ: at T/Tc=0.01 and h_s below the Clogston limit, fixed Δ is a reasonable approximation, whereas δ/E_Th=1 is a parameter-regime problem.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript theoretically studies Josephson transport in a diffusive SNS junction with spin-split superconducting electrodes. Using the Usadel equation in the Riccati parametrization, the authors compute the equilibrium and non-equilibrium supercurrent for parallel (P) and antiparallel (AP) spin-splitting alignments. The central equilibrium result is that for long junctions (L=10ξ) at low temperature (T/Tc=0.01) the critical supercurrent increases by more than 100% when the spin-splitting field approaches h_s/Δ0=0.7, for both P and AP configurations, in contrast to the short-junction limit where only AP shows enhancement and where the calculation recovers Ref. [28]. For non-equilibrium situations, the paper demonstrates a voltage-induced π-transition at lower voltages when spin-splitting is increased, and sharp supercurrent suppression under temperature bias. The paper is clearly written and the formalism is standard.","tokens_in":12571,"tokens_out":8181,"duration_ms":70842,"significance":"The predicted long-junction enhancement is a potentially interesting effect for superconducting spintronics and extends the short-junction result of Bergeret et al. to a regime where the effect is larger and configuration-independent. The use of the established Usadel/Riccati framework and the reproduction of the short-junction limit are strengths. However, the central claim rests on a numerical calculation with an inelastic broadening δ/Δ0=0.01 that coincides with the Thouless energy of the long junction, and the paper does not provide a convergence check with respect to δ. The lack of the current-phase relation data and the use of a fixed BCS gap at low temperature are additional points that need to be addressed before the quantitative enhancement claim is fully supported.","major_comments":[{"comment":"The inelastic broadening δ/Δ0=0.01 is equal to the Thouless energy E_Th/Δ0=(ξ/L)^2=0.01 for the L=10ξ junction. In this long-junction limit the spectral supercurrent at h_s=0 is concentrated at ε≈E_Th, so the artificial broadening δ is not a small parameter but rather sets the decay length sqrt(D/δ)=L of the proximity correlations. Since the zero-field supercurrent appears in the denominator of the enhancement ratio, the >100% enhancement in Fig. 2(e) may be overestimated if δ suppresses the h_s=0 current more than the h_s=0.7 current. The manuscript does not report any δ-dependence. Please provide a convergence study with δ/Δ0=10^-3 and 10^-4 (or an analytical argument for why δ=0.01 is small) and confirm that the enhancement persists. In addition, please verify that the supercurrent is conserved along the x direction with finite δ, since the equilibrium result is evaluated at x=L/2.","section":"Sec. III (after Eq. 13) and Fig. 2(e)"},{"comment":"The authors state that 'We have verified numerically that the supercurrent-phase relation I(Δφ) approaches the generic sinusoidal dependence when the spin-splitting is included for both P and AP configurations (not shown here)' and thereafter define the critical current as I(Δφ=π/2). Because the enhancement ratio is a quantitative claim (over 100%), it is important to display the current-phase relation for the relevant parameter regime (e.g., L=10ξ, h_s/Δ0=0.7, T/Tc=0.01) or to provide a quantitative bound on the deviation from the sine law over the parameter ranges used in Figs. 2, 5, and 6.","section":"Sec. III A, paragraph before Eq. (13)"}],"minor_comments":[{"comment":"There is a typo 'configuations' in the abstract; please correct it to 'configurations'.","section":"Abstract and Introduction"},{"comment":"The sentence 'This only has practical consequence for the plots in the present manuscript where we have set T/Tc=0.5, causing us to consider a maximum value of h_s/Δ0≃0.5 in that case and thus not including selfconsistency' is difficult to parse; please clarify explicitly which panels are computed with a self-consistent Δ and which are not.","section":"Sec. III A"},{"comment":"The numerical calculations set ζ_N=5, but the value of ζ_S used in the boundary condition is not specified; please state the values of all interface parameters used.","section":"Eq. (10) and Fig. 2 caption"},{"comment":"The density of states deviation from unity is shown on a scale (0.985 to 1) that makes it difficult to see the claimed enhancement of the proximity effect; please plot 1-DOS or use an enlarged inset.","section":"Fig. 3(d)"},{"comment":"The claim that spin accumulation at the center of the N wire produces the same supercurrent modulation as electric voltage is not supported by any figure; either provide the corresponding data or state it as an analytical consequence of the distribution function in Eq. (16).","section":"Sec. III B"},{"comment":"The paper does not mention the numerical method used to solve the Riccati equations (e.g., finite-difference scheme, iteration tolerance); a brief reproducibility note would be useful.","section":"Sec. II"}],"recommendation":"major_revision","confidential_remarks":"The main technical concern is the δ broadening issue, which is a parameter-regime problem rather than a mere presentational flaw. If the authors can demonstrate convergence in δ and provide the current-phase relation data, the central enhancement claim would be on solid ground. The non-self-consistent Δ issue is less concerning at T/Tc=0.01 given h_s/Δ0<0.7, but a brief justification would help. The novelty is incremental but sufficient for a specialized condensed matter theory journal; I would not recommend rejection based on the current evidence."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. The paper finds a robust >100% enhancement of the Josephson critical current in long diffusive SNS junctions when the superconducting leads are spin-split, in both parallel and antiparallel configurations, at low temperature. It also shows that spin-splitting lowers the gate voltage needed for the π-transition. Both are genuinely new compared with the short-junction SIS result of Ref. [28]. The framework is standard Usadel/Riccati, and the short-junction limit reproduces that benchmark, which is good evidence the numerics are done right.\n\nThe main soft spot is one the authors do not address. They add inelastic broadening δ/Δ0=0.01 to all quasiparticle energies. For the long junction L=10ξ, the Thouless energy is (ξ/L)^2 Δ0 = 0.01 Δ0, so δ equals the relevant low-energy scale, not a small perturbation. The zero-field supercurrent in that limit is carried by states at ε~E_Th, so a broadening of that size can suppress the denominator of the enhancement ratio more than the numerator. No δ-dependence is shown. This is a concrete numerical-regime problem that could inflate the headline >100% number, and the referee should ask for it before the quantitative claim is taken at face value. It does not obviously kill the qualitative trend, but it needs to be checked.\n\nThe non-self-consistent Δ at T/Tc=0.01 is less concerning: the authors explicitly restrict self-consistency caveats to T/Tc=0.5, and below the Clogston limit a fixed gap is a reasonable first cut. The sinusoidal current-phase relation is asserted rather than shown, and the claimed equivalence between spin accumulation and voltage is not derived; both are minor. The paper is clearly written, and the figures are informative.\n\nWho this is for: people working on superconducting spintronics, Josephson transistors, and voltage-tunable π-junctions. It deserves a serious referee; the framework is sound and the effect, if it survives the broadening check, is worth publishing. I'd ask the authors for a δ-sweep and, if feasible, a self-consistent low-T check before acceptance.","headline":"Predicts a robust >100% supercurrent enhancement in long spin-split SNS junctions, but the headline number is computed with inelastic broadening equal to the Thouless energy, and no δ-dependence is shown.","tokens_in":13244,"tokens_out":2990,"would_cite":true,"duration_ms":26962,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A spin-splitting field more than doubles the critical supercurrent of a long SNS Josephson junction at low temperature, in both parallel and antiparallel configurations.","keywords":["spin-split superconductors","Josephson junction","supercurrent enhancement","pi transition","Usadel equation","superconducting transistor","spin accumulation","temperature bias"],"falsifier":"A self-consistent calculation that lets the order parameter adjust to the spin-splitting field at $T/T_c = 0.01$ and $h_s/\\Delta_0 = 0.7$ would falsify the central claim if the more-than-100% enhancement disappears, as would an experimental measurement of the critical current of a long spin-split SNS junction versus in-plane field at dilution temperatures that shows no such rise before the Clogston limit.","tokens_in":12059,"feed_emoji":"⚡","tokens_out":12573,"duration_ms":97836,"temperature":0.7,"pith_summary":"This paper studies a superconductor/normal-metal/superconductor junction whose superconducting leads are spin-split by a magnetic exchange field. It argues that in long junctions at low temperature the critical supercurrent grows strongly with the spin-splitting field, increasing by more than 100% for both parallel and antiparallel lead configurations, and that voltage, spin-accumulation, or temperature-bias inputs can control the supercurrent like a transistor. The interest is that a magnetic field alone would amplify the supercurrent, while the out-of-equilibrium controls would give a lower-voltage sign switch and a sharp thermal response.","feed_headline":"Spin-splitting field more than doubles long-junction supercurrent","feed_subtitle":"The >100% rise works for both lead alignments; voltage or temperature bias adds transistor control.","key_machinery":"The machinery is the Usadel equation for the diffusive normal metal, solved with a Riccati parametrization of the quasiclassical Green's function and coupled to the superconducting leads through interface boundary conditions. The spin-split leads are treated as reservoirs with an analytical Green's function that depends on the field $h_s$, and the supercurrent is extracted from the Keldysh component of the Green's function at the mid-point of the junction. The effect that carries the enhancement is a spin-splitting-induced increase of the low-energy spectral supercurrent in long junctions, reflecting a deeper superconducting proximity effect in the normal metal.","core_discovery":"The central numerical result is that for a long junction with $L = 10\\xi$ at $T/T_c = 0.01$, the critical supercurrent increases by more than 100% as the spin-splitting field is raised to $h_s/\\Delta_0 \\approx 0.7$, and this happens for both parallel and antiparallel alignments of the spin-splitting fields. The authors contrast this with an earlier tunneling-limit result for superconductor/ferromagnet structures, where the enhancement is limited to antiparallel alignment and is smaller. The same framework yields two additional results: applying a transverse voltage to the normal metal produces the familiar supercurrent suppression and pi-transition, with the transition voltage lowered by increasing the spin-splitting or the junction length, and applying a small temperature difference to a long junction suppresses the supercurrent sharply, with roughly a 50% drop at $\\Delta T/\\Delta_0 \\approx 0.05$. The authors interpret the equilibrium enhancement as a strengthening of the proximity effect at low energies in the normal metal, visible in the local density of states.","pith_inferences":["Beyond the paper: if the enhancement mechanism is the low-energy proximity strengthening, the same spin-split leads should boost supercurrents in other long coherent junctions, such as multiterminal devices, which could be checked with existing magnetic-insulator/superconductor technology.","Beyond the paper: the equivalence of voltage and spin accumulation at the junction center implies a transistor variant controlled by pure spin injection, which would avoid charge current in the control line.","Beyond the paper: the position and sharpness of the temperature-bias supercurrent jump could serve as a calibrated probe of the spin-splitting field in the leads."],"forward_implications":["A long SNS junction with spin-split leads can serve as a field-controlled supercurrent amplifier in equilibrium, with the critical current increasing by more than a factor of two near $h_s/\\Delta_0 \\approx 0.7$.","The enhancement is insensitive to whether the two leads' spin-splitting fields are parallel or antiparallel, removing the need to switch or stabilize the relative magnetization orientation.","Adding spin-splitting to a voltage-controlled SNS transistor lowers the voltage at which the supercurrent reverses sign, making the pi-transition more energy-efficient.","A temperature bias of a few hundredths of the gap in a long junction can cut the supercurrent by about half in a sharp jump, suggesting a sensitive thermal switch.","Applying a spin accumulation to the normal metal reproduces the voltage-control behavior at the center of the junction, so the transistor can be operated by a pure spin signal."],"supporting_citations":[{"why":"provides the earlier superconductor/ferromagnet junction result that the paper extends and contrasts with, predicting antiparallel-only enhancement in the tunneling limit.","marker":"[28]"},{"why":"supplies the Usadel diffusion equation that forms the basis of the transport calculation.","marker":"[29]"},{"why":"provides the Riccati parametrization of the Green's function used for the numerical solution.","marker":"[30]"},{"why":"gives the boundary conditions connecting the superconducting and normal-metal Green's functions at the interfaces.","marker":"[33]"},{"why":"defines the voltage-controlled SNS transistor model whose nonequilibrium distribution function is adopted.","marker":"[34]"},{"why":"the experimental demonstration of voltage-controlled supercurrent reversal that the computed pi-transition is compared with.","marker":"[13]"},{"why":"sets the Clogston-Chandrasekhar limit that restricts the spin-splitting field to below 0.7 times the gap in the calculations.","marker":"[35]"},{"why":"demonstrates spin-splitting in EuS/Al heterostructures, supporting the experimental feasibility of the parameter regime.","marker":"[42]"}],"fun_headline_variants":["Spin splitting boosts long-junction supercurrent by 100%+","Both alignments: spin-split field doubles supercurrent in long junctions","Long-junction supercurrent doubles with spin splitting—no alignment penalty","Voltage and temperature bias turn spin-split junction into sharp switch"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The headline enhancement is computed with the superconducting gap held at its zero-field value even when the spin-splitting field reaches about $0.7\\Delta_0$ at $T/T_c = 0.01$, so the central result assumes the field does not itself weaken superconductivity.","fun_headline_variants_meta":{"raw":{"variants":["Spin splitting boosts long-junction supercurrent by 100%+","Both alignments: spin-split field doubles supercurrent in long junctions","Long-junction supercurrent doubles with spin splitting—no alignment penalty","Voltage and temperature bias turn spin-split junction into sharp switch"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00078,"raw_usage":{"total_tokens":3443,"prompt_tokens":937,"completion_tokens":2506,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":553,"completion_tokens_details":{"reasoning_tokens":2430}},"tokens_in":553,"tokens_out":2506,"duration_ms":15777,"temperature":1.0,"reasoning_tokens":2430,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T13:52:29.906585+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A self-consistent calculation that lets the order parameter adjust to the spin-splitting field at $T/T_c = 0.01$ and $h_s/\\Delta_0 = 0.7$ would falsify the central claim if the more-than-100% enhancement disappears, as would an experimental measurement of the critical current of a long spin-split SNS junction versus in-plane field at dilution temperatures that shows no such rise before the Clogston limit.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"provides the earlier superconductor/ferromagnet junction result that the paper extends and contrasts with, predicting antiparallel-only enhancement in the tunneling limit."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"supplies the Usadel diffusion equation that forms the basis of the transport calculation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"gives the boundary conditions connecting the superconducting and normal-metal Green's functions at the interfaces."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"defines the voltage-controlled SNS transistor model whose nonequilibrium distribution function is adopted."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"the experimental demonstration of voltage-controlled supercurrent reversal that the computed pi-transition is compared with."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"sets the Clogston-Chandrasekhar limit that restricts the spin-splitting field to below 0.7 times the gap in the calculations."},{"cited_title":"Strambini, V","cited_arxiv_id":null,"evidence_quote":"demonstrates spin-splitting in EuS/Al heterostructures, supporting the experimental feasibility of the parameter regime."}],"review_version":1}