{"id":"d577a47c-5a24-451e-8919-3ae063a626fd","arxiv_id":"2504.20366","paper_version":3,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"In 1D SF1F2S Josephson junctions, a barrier at the F1/F2 interface creates critical-current resonance peaks at Q_i d_i = (n_i + 1/2)π, attributed to zero-spin-projection triplet pairs, with accumulated phase setting the 0 or π ground state.","lead":"A model calculation shows that adding an insulating barrier between two ferromagnetic layers in a superconducting junction produces sharp periodic spikes in the maximum supercurrent. The spike positions are set by a quantized relation between each layer's magnetic exchange field and thickness, offering a way to switch between two superconducting phase states for cryogenic memory.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Triplet-filtering mechanism is asserted, not computed, and the collinear resonance conditions coincide with zero m=0 triplet amplitude in the paper's own single-pass formulas.","rationale":"The reader's weakest assumption is the same as mine: the triplet-filtering mechanism is imported from pair-decomposition formulas rather than derived from the numerical BdG solution, so a direct computation of pair amplitudes is needed. I agree with the CONDITIONAL verdict because the numerical resonance-peak claim is likely correct within the stated model and the mechanism, while questionable, is not directly falsified by anything in the paper. My stress-test sharpens the reader's concern by pointing out an internal tension: in the parallel and antiparallel collinear cases, the paper's own single-pass formulas (Eqs. 16 and 18) give zero m=0 triplet amplitude and maximal singlet amplitude exactly at the claimed resonance conditions. This does not prove the mechanism wrong, because multiple Andreev/barrier reflections may alter the decomposition, but it does show that the mechanism is not a straightforward consequence of the cited formulas. A direct spatial pair-amplitude calculation from the BdG eigenvectors is the minimal check that would settle whether the triplet interpretation or a barrier-phase-shift explanation for the sin-like oscillation is correct. I do not recommend REJECT because the numerical peak positions and the 0/π assignments are reported as computed quantities, and the requested calculation could confirm the mechanism. I also do not recommend ACCEPT because the paper's headline physics claim currently rests on an unverified attribution. The reader already requires such a check, so the verdict remains CONDITIONAL; my analysis strengthens the reason for that condition without changing the outcome.","tokens_in":20632,"tokens_out":11864,"duration_ms":133688,"concrete_test":"Use the BdG determinant solution: at a resonance peak in the parallel configuration (e.g., h1/EF=h2/EF=0.05, kF d1=kF d2=10π, Z=3, T=0), obtain the eigenvector X of Λ for the Andreev level and construct the real-space pair amplitudes inside F1 and F2 from the BdG spinors: singlet F_s(x)=⟨u↑(x)v↓(x)-u↓(x)v↑(x)⟩, zero-spin triplet F_t0(x)=⟨u↑(x)v↓(x)+u↓(x)v↑(x)⟩, and equal-spin triplets F_t↑↑=⟨u↑v↑⟩, F_t↓↓=⟨u↓v↓⟩. Compare these amplitudes at the same parameters with Z=0 and Z=3, and also repeat for the antiparallel equal-layer resonance. If F_t0 is not the dominant transmitted component across the barrier/resonance, or if F_s is not suppressed relative to the Z=0 case, the triplet resonant-tunneling mechanism and the φt=Q1d1±Q2d2 phase rule require revision, and the paper should be reframed as a numerical observation of resonance peaks without the triplet-filtering interpretation.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central numerical result—resonance peaks of Ic at Q_i d_i = (n_i+1/2)π—may be correct, but the central physical claim (Abstract, Secs. 3.1-3.3) that a spin-independent δ-barrier suppresses singlet pairs and transmits zero-spin-projection triplet pairs (↑↓+↓↑) is not derived from the BdG solution. The paper imports the single-pass pair decomposition of Refs. [5,6] (Eqs. 13-18) and never computes singlet/triplet pair amplitudes in the presence of the barrier. This matters because the imported formulas are in tension with the claimed mechanism in the collinear cases: for θ=0, Eq. (16) gives, after F1 and F2, a singlet amplitude cos(Q1d1+Q2d2) and an m=0 triplet amplitude sin(Q1d1+Q2d2). At the claimed resonances Q_i d_i=(n_i+1/2)π, the sum is (n1+n2+1)π, so the singlet is maximal (cos=±1) while the m=0 triplet is zero (sin=0). For antiparallel with equal layers, Eq. (18) gives triplet amplitude sin(Q1d1-Q2d2)=0 at the same resonances while the singlet amplitude is cos(0)=1. Thus, at the level of the paper's own pair-evolution formulas, the resonance positions for collinear configurations coincide with maximal singlet amplitude, not maximal zero-spin triplet amplitude. Multiple Andreev and barrier reflections could change this conclusion, but the paper does not show how. If the interfacial scattering converts pairs differently than the assumed sin(Q_i d_i) amplitudes, both the triplet attribution and the 0/π phase rule φt=Q1d1±Q2d2 would be unsupported even if the peak positions are numerically correct.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript studies a one-dimensional SF1F2S Josephson junction with a delta-function potential barrier at the F1/F2 interface. The authors numerically solve the Bogoliubov–de Gennes equations in the short-junction limit, obtain the Andreev bound-state energies from the characteristic equation det(Lambda)=0, and compute the current-phase relation and critical current from the thermodynamic potential. The central numerical observation is that, for a sufficiently strong barrier (Z=3) and low temperature, the critical current exhibits periodic peaks whenever Q1d1=(n1+1/2)pi and Q2d2=(n2+1/2)pi, in parallel, perpendicular, and antiparallel magnetization configurations. The paper interprets these peaks as quantized resonant tunneling of spin-triplet pairs (up-down + down-up), with a total phase phi_t=Q1d1 +/- Q2d2 determining whether the junction is in a 0- or pi-state, and proposes that for equal ferromagnetic layers the barrier can act as a magnetization-controlled 0/pi switch.","tokens_in":20944,"tokens_out":10909,"duration_ms":116347,"significance":"The numerical procedure is standard, and the reported peak positions are not fitted to the analytical sin/cos decomposition; they follow directly from the BdG characteristic equation. If the mechanism claim were established, the simple quantization condition and the magnetization-controlled 0/pi switching would be interesting for superconducting spintronics. However, the paper's physical interpretation is not derived from the BdG solution and, as detailed below, is contradicted at face value by the paper's own pair-evolution formulas. The numerical observation of resonances may survive, but the abstract's and conclusions' attribution to spin-triplet tunneling requires substantial additional support.","major_comments":[{"comment":"The central mechanistic claim, that the barrier transmits the m=0 triplet pair (up-down + down-up) and that the peaks occur where its amplitude sin(Q_i d_i) is maximal, is contradicted by the pair-evolution formulas used in the paper. Inserting Q1d1=(n1+1/2)pi and Q2d2=(n2+1/2)pi into Eq. (16) gives a triplet amplitude sin(Q1d1+Q2d2)=0 and a singlet amplitude cos(Q1d1+Q2d2)=+/-1, i.e., the opposite of the claimed mechanism. For the antiparallel case, Eq. (18) gives sin(Q1d1-Q2d2)=0 at the same resonance conditions. Since the paper never computes singlet/triplet pair amplitudes from the BdG solution in the presence of the barrier, the load-bearing attribution of the resonance peaks to (up-down + down-up) tunneling is unsupported and, at the level of the paper's own formulas, incorrect. The authors should either compute the anomalous Green's function components or revise the interpretation accordingly.","section":"Sec. 3.1, Eqs. (16) and (18)"},{"comment":"For perpendicular magnetizations, Eq. (17) shows that at the resonance values Q1d1=(n1+1/2)pi and Q2d2=(n2+1/2)pi the singlet amplitude cos(Q1d1)cos(Q2d2) and the opposite-spin triplet amplitude cos(Q1d1)sin(Q2d2) both vanish; the only surviving single-pass pair amplitude in the F2 basis is the equal-spin triplet i sin(Q1d1). The text nonetheless claims that the barrier filters out singlet pairs and that the remaining pairs are (up-down + down-up) in F1 transforming into (up-up - down-down) in F2. This is inconsistent with Eq. (17) and with the abstract's statement that pairs with zero spin projection are selected. The origin of the second-harmonic current at the resonance peaks therefore needs a direct derivation from the BdG solution rather than the imported single-pass formula.","section":"Sec. 3.2, Eq. (17)"},{"comment":"The proposed spin-filter action is not explained: U(r)=V delta(x) is spin-independent and therefore cannot by itself distinguish singlet from triplet pairs. Any spin selectivity must arise from the spin-dependent Fermi wave vectors combined with the boundary conditions, and this needs to be shown explicitly. Furthermore, the 0/pi phase rule phi_t=Q1d1 +/- Q2d2 does not discriminate between the singlet and triplet interpretations: the singlet amplitudes in Eqs. (16) and (18) oscillate as cos(Q1d1 +/- Q2d2), whose sign produces exactly the same alternation of 0- and pi-states claimed for the triplet phase. A direct computation of the spin-resolved anomalous pair amplitudes from the BdG eigenfunctions at the F1/F2 interface (or an equivalent Green's-function calculation) is required before the abstract and conclusions can claim that the resonance peaks are caused by resonant tunneling of spin-triplet pairs.","section":"Abstract and Sec. 3.1 (first paragraphs)"}],"minor_comments":[{"comment":"The sentence 'When phi_t is an even number...' is ambiguous; since phi_t is a phase, the text should state 'an even/odd multiple of pi' or 'phi_t/pi is an even/odd integer.'","section":"Sec. 3.1, paragraph defining phi_t"},{"comment":"The short-junction condition is stated as kF d1, kF d2 << 1000; it would be more precise to express the criterion as d1, d2 << xi_S = hbar v_F / Delta, which for EF=1000 Delta means kF d << 2000.","section":"Sec. 2, short-junction criterion"},{"comment":"The text says 'at h2/EF = 1.0, the current amplitude is diminished,' but the context and Fig. 5(d) indicate that this should be h2/EF = 0.10.","section":"Sec. 3.2, Fig. 5(d) discussion"},{"comment":"There are minor typographical errors: 'Shaa nxi' in the affiliation should be 'Shaanxi,' and 'Kümel' in Ref. [63] should likely be 'Kümmel.'","section":"Author affiliations and Ref. [63]"}],"recommendation":"major_revision","confidential_remarks":"The manuscript builds heavily on the authors' earlier framework (their Ref. [58]), and the referee process should ensure that the claimed novelty with respect to the interference resonances of Ref. [59] is clearly established. The main concern is not the numerical peak positions but the unsupported and internally inconsistent spin-triplet mechanism; a direct pair-amplitude computation would be needed to make the central claim convincing."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Couple of things you should know before reading this one. The numerical core is fine: the authors solve 1D BdG in the short-junction limit, and their reported critical-current peaks at Q_i d_i = (n_i+1/2)π are consistent across parallel, perpendicular, and antiparallel configurations. That gives a clean, potentially useful design rule for 0/π switching in spin-valve Josephson junctions. But the central physical claim—that the barrier filters out singlets and lets m=0 triplets tunnel resonantly—is not computed anywhere. It's imported from the single-pass pair decomposition of Eschrig, and the paper's own equations are in tension with it: for collinear configurations, at the claimed resonances the singlet amplitude is maximal and the m=0 triplet amplitude is zero in those formulas. So the mechanism could be wrong even if the peak positions are right.\n\nWhat's actually new: the explicit quantization condition for arbitrary exchange fields and thicknesses, and the 0/π selection for equal F1/F2 layers (π in parallel, 0 in antiparallel). This goes beyond their earlier paper [58], which only treated identical layers. The work is honest about what is inferred vs computed; the abstract says 'it can be inferred.'\n\nThe soft spots, in proportion. The major one is the mechanism. The paper never computes singlet/triplet pair amplitudes in the presence of the barrier, so the attribution of the resonances to (↑↓+↓↑) tunneling is an unverified interpretation. Additionally, the stress-test note is correct: if you take Eqs. (16) and (18) at face value, the resonance condition makes the singlet term ±1 and the m=0 triplet term 0. Multiple Andreev reflections and barrier scattering could reverse that, but the authors would need to show it. The minor point: no numerical convergence details or artifacts, which is a bit sloppy but not disqualifying.\n\nThe citation pattern looks fine: [58] is their own previous work and the comparison is legitimate; [59] is the Nikolić et al. geometric resonance paper, and the distinction is real (1D vs 3D, different dependence on d1,d2).\n\nWho this is for: people working on superconducting spintronics and cryogenic memory who want a simple analytic condition for peak positions. The result deserves a serious referee: the numerical finding is checkable and the design rule could be useful even if the proposed mechanism is wrong. But I would not take the triplet-filtering story at face value. If I were the referee, I'd ask the authors to compute the pair amplitudes in the barrier region, or at least to present a scattering argument that shows why their imported sin(Q_i d_i) weighting survives the barrier.","headline":"Solid numerical finding on quantized Ic resonances in SF1F2S junctions under a barrier; the triplet-filtering mechanism is asserted rather than computed and sits in tension with the paper's own pair-evolution formulas.","tokens_in":21546,"tokens_out":3518,"would_cite":true,"duration_ms":31705,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["74.50.+r","74.45.+c"],"model":"deepseek-v4-flash","headline":"A potential barrier at the F1/F2 interface makes the Josephson critical current of an SF1F2S junction peak at the quantization conditions Q1d1=(n1+1/2)π and Q2d2=(n2+1/2)π, through resonant tunneling of zero-spin-projection spin-triplet…","keywords":["Josephson effect","spin-triplet pairs","resonant tunneling","ferromagnetic bilayer","0-pi transition","Andreev bound states","critical current","superconducting spintronics"],"falsifier":"Use the same numerics to extract spin-resolved singlet and triplet pair amplitudes inside the ferromagnets at $Z=3$: the paper's mechanism requires the $(\\uparrow\\downarrow+\\downarrow\\uparrow)$ amplitude to peak at $Q_1d_1=(n_1+1/2)\\pi$ and $Q_2d_2=(n_2+1/2)\\pi$ while the singlet amplitude is suppressed there. If the critical-current peaks survive but the pair decomposition does not show that pattern, the resonant-triplet attribution fails.","tokens_in":20397,"feed_emoji":"🧲","tokens_out":10498,"duration_ms":89865,"temperature":0.7,"pith_summary":"The paper studies a one-dimensional superconductor/ferromagnet/ferromagnet/superconductor (SF$_1$F$_2$S) Josephson junction with a potential barrier at the interface between the two ferromagnetic layers. It claims that at low temperature the barrier turns the junction into a quantized resonant-tunneling device: the critical current develops sharp periodic peaks whenever $Q_1d_1=(n_1+1/2)\\pi$ and $Q_2d_2=(n_2+1/2)\\pi$, where $Q_i=2h_i/(\\hbar v_F)$ is the center-of-mass momentum that Cooper pairs acquire in the $i$-th ferromagnet. The paper attributes these peaks to resonant tunneling of spin-triplet pairs with zero spin projection $(\\uparrow\\downarrow+\\downarrow\\uparrow)$, which the barrier lets through while suppressing ordinary spin-singlet pairs. The total phase accumulated by those triplet pairs, $\\varphi_t=Q_1d_1\\pm Q_2d_2$, then fixes whether the junction sits in a 0-state or a $\\pi$-state, so the magnetization arrangement and barrier together act as a phase switch. This matters because controllable 0-$\\pi$ switching in ferromagnetic Josephson junctions is a route toward superconducting spintronics and cryogenic memory.","feed_headline":"Barrier creates quantized spin-triplet current peaks","feed_subtitle":"Resonance peaks at Qd=(n+1/2)π let a barrier choose between 0- and π-states, a lever for superconducting memory.","key_machinery":"The load-bearing object is the pair-decomposition identity that tracks how a Cooper pair evolves through the two ferromagnets, $(\\uparrow\\downarrow)e^{iQ_1d_1}-(\\downarrow\\uparrow)e^{-iQ_1d_1}\\to(\\uparrow\\downarrow-\\downarrow\\uparrow)\\cos(Q_1d_1\\pm Q_2d_2)+i(\\uparrow\\downarrow+\\downarrow\\uparrow)\\sin(Q_1d_1\\pm Q_2d_2)$, with signs and prefactors set by the relative magnetization angle. The paper reads this identity as a filter: the barrier is claimed to block the spin-singlet term $(\\uparrow\\downarrow-\\downarrow\\uparrow)$ while transmitting the zero-spin-projection triplet term $(\\uparrow\\downarrow+\\downarrow\\uparrow)$, whose magnitude is controlled by $\\sin(Q_1d_1)$ in F$_1$ and $\\sin(Q_2d_2)$ in F$_2$. That product is maximized exactly when $Q_id_i=(n_i+1/2)\\pi$, and the sum or difference of the two layer phases, $\\varphi_t=Q_1d_1\\pm Q_2d_2$, is what selects the 0- or $\\pi$-state. The numerical engine is a solution of the Bogoliubov–de Gennes equations with continuity and barrier boundary conditions, giving two Andreev bound-state energies from $\\det\\Lambda=0$, from which the Josephson current is computed through the thermodynamic potential.","core_discovery":"The central discovery is that adding a spin-independent $\\delta$-function barrier at the F$_1$/F$_2$ interface qualitatively changes the Josephson transport of an SF$_1$F$_2$S junction. Without the barrier, the critical current oscillates smoothly with exchange fields and layer thicknesses because spin-singlet pairs $(\\uparrow\\downarrow-\\downarrow\\uparrow)$ pass through the ferromagnets and accumulate a phase $Q_1d_1\\pm Q_2d_2$, producing the familiar 0-$\\pi$ oscillations. With a sufficiently strong barrier ($Z=3$) at low temperature, the singlet channel is suppressed and the current is carried by zero-spin-projection triplet pairs $(\\uparrow\\downarrow+\\downarrow\\uparrow)$, whose amplitude in each layer is $\\sin(Q_i d_i)$. The current then shows resonance peaks exactly at $Q_1d_1=(n_1+1/2)\\pi$ and $Q_2d_2=(n_2+1/2)\\pi$, with the total triplet phase $\\varphi_t=Q_1d_1+Q_2d_2$ in parallel magnetizations and $Q_1d_1-Q_2d_2$ in antiparallel magnetizations fixing the ground state. In perpendicular configurations the same quantization conditions hold but the surviving current is dominated by the second harmonic $I_2\\sin(2\\varphi)$; when both layers are identical the barrier suppresses the 0-state current in parallel alignment while preserving it in antiparallel alignment.","pith_inferences":["Beyond the paper: if the barrier is a genuine spin-singlet filter, the observed peak positions give an in situ measurement of the Cooper-pair center-of-mass momentum $Q_i=2h_i/(\\hbar v_F)$ in each ferromagnetic layer, so a thickness scan of the critical current could map the exchange splitting directly.","Beyond the paper: the predicted magnetization-controlled 0- and $\\pi$-state selection in identical layers suggests a Josephson phase switch operated by rotating one magnetization by $180^\\circ$, which could be tested in nanopillar spin-valve junctions with switchable magnetic layers.","Beyond the paper: replacing the idealized $\\delta$-barrier with a finite-width or spin-dependent interface, or moving to a quasi-one-dimensional wire, would test whether the resonance positions and the 0/$\\pi$ rule survive beyond the paper's minimal model."],"forward_implications":["At sufficiently strong barrier and low temperature in perpendicular magnetization configurations, the first harmonic is filtered out and the current-phase relation becomes dominated by $I_2\\sin(2\\varphi)$, giving a response with period $\\pi$.","When the two ferromagnetic layers are identical, parallel alignment yields resonance peaks with $\\varphi_t=(2n_1+1)\\pi$, locking the junction into the $\\pi$-state, while antiparallel alignment gives $\\varphi_t=0$ and keeps the 0-state.","The quantization condition $Q_id_i=(n_i+1/2)\\pi$ locates the current maxima in parallel, perpendicular, and antiparallel configurations alike, so the resonance condition is robust to the relative magnetization direction.","Raising the temperature suppresses the resonant tunneling peaks and restores the smooth 0-$\\pi$ oscillation pattern, so the effect is confined to low temperatures."],"supporting_citations":[{"why":"Supplies the pair-decomposition identity in Eqs. (13)-(18) that the paper uses to identify the sin(Q_i d_i) triplet amplitude.","marker":"[5]"},{"why":"The companion review that gives the same Cooper-pair phase accumulation in ferromagnets and the triplet pair forms used in the argument.","marker":"[6]"},{"why":"Provides the Bogoliubov-de Gennes solution framework and the critical-current formulation on which the numerical results are built.","marker":"[58]"},{"why":"Establishes the 0-pi transition mechanism for spin-singlet pairs in SFS junctions that the barrier is claimed to suppress.","marker":"[2]"},{"why":"Gives the odd-frequency triplet pairing background invoked to explain why zero-spin-projection triplet pairs survive the barrier.","marker":"[3]"},{"why":"Source of the superharmonic second-harmonic current mechanism used for the perpendicular configuration.","marker":"[39]"},{"why":"The alternative geometric-interference mechanism in perpendicular bilayers that this paper distinguishes from its own resonance effect.","marker":"[59]"},{"why":"Predicted exchange-field enhancement of the antiparallel critical current that motivates the question of where current maxima occur.","marker":"[51]"}],"fun_headline_variants":["Barrier triggers quantized triplet resonance peaks in Josephson junctions","Quantized resonant tunneling produces spin-triplet current peaks in JJs","Barrier-induced triplet resonance tunes 0-π states in Josephson junctions","Quantized triplet resonance peaks select 0-π state in ferromagnetic bilayer junctions"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the barrier acts as a clean spin-singlet filter, transmitting zero-spin-projection triplet pairs with amplitude $\\sin(Q_id_i)$; this filtering mechanism is imported from standard pair-decomposition formulas rather than being derived from the numerical wave functions.","fun_headline_variants_meta":{"raw":{"variants":["Barrier triggers quantized triplet resonance peaks in Josephson junctions","Quantized resonant tunneling produces spin-triplet current peaks in JJs","Barrier-induced triplet resonance tunes 0-π states in Josephson junctions","Quantized triplet resonance peaks select 0-π state in ferromagnetic bilayer junctions"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000871,"raw_usage":{"total_tokens":3932,"prompt_tokens":1265,"completion_tokens":2667,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":881,"completion_tokens_details":{"reasoning_tokens":2588}},"tokens_in":881,"tokens_out":2667,"duration_ms":19401,"temperature":1.0,"reasoning_tokens":2588,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T05:31:23.461736+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Use the same numerics to extract spin-resolved singlet and triplet pair amplitudes inside the ferromagnets at $Z=3$: the paper's mechanism requires the $(\\uparrow\\downarrow+\\downarrow\\uparrow)$ amplitude to peak at $Q_1d_1=(n_1+1/2)\\pi$ and $Q_2d_2=(n_2+1/2)\\pi$ while the singlet amplitude is suppressed there. If the critical-current peaks survive but the pair decomposition does not show that pattern, the resonant-triplet attribution fails.","supporting_citations":[{"cited_title":"Josephson current through a ferromagnetic bilayer: Beyond the quasiclassical approximation","cited_arxiv_id":"1912.04447","evidence_quote":"Provides the Bogoliubov-de Gennes solution framework and the critical-current formulation on which the numerical results are built."},{"cited_title":"Proximity effects in superconductor-ferromagnet heterostructures","cited_arxiv_id":"cond-mat/0505583","evidence_quote":"Establishes the 0-pi transition mechanism for spin-singlet pairs in SFS junctions that the barrier is claimed to suppress."},{"cited_title":"Odd Triplet Superconductivity and Related Phenomena in Superconductor-Ferromagnet Structures","cited_arxiv_id":"cond-mat/0506047","evidence_quote":"Gives the odd-frequency triplet pairing background invoked to explain why zero-spin-projection triplet pairs survive the barrier."},{"cited_title":"Long-Range Superharmonic Josephson Current","cited_arxiv_id":"1101.5416","evidence_quote":"Source of the superharmonic second-harmonic current mechanism used for the perpendicular configuration."},{"cited_title":"Interference phenomena in Josephson junctions with ferromagnetic bilayers: Spin-triplet correlations and resonances","cited_arxiv_id":"2206.05770","evidence_quote":"The alternative geometric-interference mechanism in perpendicular bilayers that this paper distinguishes from its own resonance effect."},{"cited_title":"Enhancement of the Josephson current by an exchange field in superconductor-ferromagnet structures","cited_arxiv_id":"cond-mat/0102012","evidence_quote":"Predicted exchange-field enhancement of the antiparallel critical current that motivates the question of where current maxima occur."}],"review_version":1}