{"id":"981e5a6d-ec52-40cd-8aa6-c97efd157953","arxiv_id":"2501.06845","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":9,"one_line_summary":"NiFeCr Josephson junctions show a 0-π transition at 2.3 nm and a 0.36 nm decay length, but critical current is two orders lower than NiFe and switching energy is not reduced.","lead":"This paper tests a chromium-doped nickel-iron alloy as the magnetic layer in superconducting memory switches. It finds the alloy reduces magnetization but also suppresses the switch current and does not lower switching energy, so it is a dead end for current cryogenic memory.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Quantitative claims in the abstract rest on Eq. (4), whose applicability to NiFeCr is explicitly uncertain; this weakens d0π and the switching-energy estimate but not the qualitative rejection.","rationale":"The reader's CONDITIONAL verdict is well calibrated. My independent stress-test finds no reason to soften or harden it. The qualitative conclusion—that NiFeCr suppresses critical current and therefore is unsuitable for high-Ic π-junction memory—is supported by direct IcRN data in the measured thickness range and by the large measured Ms reduction. The load-bearing weakness is the quantitative layer: the abstract quotes d0π=2.30 nm and ξF1=0.36 nm, and the conclusion compares switching energies, all of which depend on Eq. (4). The authors themselves flag the regime ambiguity for NiFeCr, and the treatment of near-transition points from the second run is under-specified. That is exactly the kind of model/systematic uncertainty a conditional acceptance should require the authors to quantify. The proposed refit is a straightforward check that settles whether the numbers in the abstract are stable.","tokens_in":11262,"tokens_out":6932,"duration_ms":72999,"concrete_test":"Refit the NiFeCr IcRN(dF) data twice: once with the diffusive-limit form (single ξ controlling both decay and oscillation, plus V0 and d0π) and once including the two second-run points that were excluded or downweighted; compare the fitted d0π, the π-state maximum, and the resulting Esw ratio. If d0π moves outside 2.30 ± 0.15 nm, or if the fitted π-state peak changes by more than a factor of 3, the abstract's quantitative claims are not robust to model choice.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Most load-bearing concern: The abstract's quantitative numbers—d0π=2.30 nm, ξF1=0.36 nm, and by extension the 'nearly two orders of magnitude' suppression and the switching-energy comparison—are products of Eq. (4), a four-parameter damped-sine fit whose physical-regime applicability for NiFeCr is explicitly uncertain in Sec. IV. The authors state it is not clear whether their junctions are diffusive or intermediate, and Eq. (4) is introduced as a model for both rather than derived from either. The same section also notes that a couple of second-sputtering-run points near the transition were exceptions to the good fit, yet the fitting/exclusion procedure is not described. Because d0π feeds directly into the switching-energy estimate Esw ∝ Ms^2 tF^2, a different valid functional form or a different handling of those points could change d0π by a few tenths of a nanometer and alter the 'switching energy is hardly changed' conclusion, and the magnitude of the π-state maximum is partly extrapolated through ξF1. Direct I-V data do robustly show that NiFeCr carries much less supercurrent in the measured thickness range, so the qualitative unsuitability claim survives; what is at risk is the quantitative precision claimed in the abstract and Table I.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports a comparative study of ferromagnetic Josephson junctions with NiFeCr (Ni73Fe18Cr9) and NiFe (Ni82Fe18) barrier layers, motivated by the search for a soft-magnetic material with low switching field and energy for cryogenic memory applications. The authors characterize the magnetization, coercivity, and Josephson critical current of both systems, and fit the measured IcRN versus ferromagnetic-layer thickness to an exponentially decaying sinusoidal function (Eq. 4). From these fits they extract a 0–π transition thickness d0π = 2.30 nm for NiFeCr versus 1.49 nm for NiFe, and a decay length ξF1 = 0.36 nm versus 0.67 nm. The central qualitative conclusion is that Cr doping reduces the magnetization by roughly one-third but suppresses the critical current by nearly two orders of magnitude in the π state, making NiFeCr unattractive for memory applications requiring large critical currents. The manuscript also reports switching-field measurements showing a few-mT coercivity for the thinnest NiFeCr films.","tokens_in":11506,"tokens_out":3449,"duration_ms":35609,"significance":"If the quantitative claims held, this would be a useful negative result for the superconducting-memory community: it identifies a specific alloy that initially appears promising—because Cr doping reduces Ms without destroying soft magnetic behavior—but that fails in the Josephson-junction context because of the short decay length and low critical current. The qualitative finding is directly visible in the raw IcRN data: NiFeCr junctions carry substantially smaller supercurrents than NiFe junctions across the measured thickness range, and the suppression is not a fitting artifact. The paper also provides useful complementary data on magnetization, dead-layer thickness, and coercivity. The main value is therefore the experimental comparison, not the extracted parameters. The extracted d0π and ξF1 values and the switching-energy comparison are model-dependent, because Eq. (4) is an assumed fitting form whose physical regime for NiFeCr is explicitly uncertain. The manuscript would be strengthened by clearly separating the robust qualitative conclusions from the model-dependent quantitative parameters.","major_comments":[{"comment":"The quantitative parameters in Table I and in the abstract—d0π = 2.30 nm and ξF1 = 0.36 nm for NiFeCr—are obtained by fitting the data to the exponentially damped sinusoid of Eq. (4). The paper states in §IV that it is not clear whether the NiFeCr junctions fall into the diffusive or intermediate limit, and that Eq. (4) is introduced only as an approximate model for both, rather than being derived from either limit. The functional form directly determines the extracted d0π and ξF1 values, so the reported precision of ±0.01 nm overstates what the data establish. Please quantify the model dependence by, for example, fitting the data with alternative physically motivated forms (e.g., independent decay and oscillation lengths without the absolute-value sine, or a full diffusive-limit Usadel expression) and reporting how d0π and ξF1 shift, or by presenting these parameters with a model-dependent systematic uncertainty.","section":"§IV, Eq. (4)"},{"comment":"The text reports that the fit is good 'with the exception of a couple of NiFeCr junctions fabricated in a second sputtering run with thicknesses very close to the 0−π transition', but the paper does not describe how those points were treated in the fit, how many points were excluded, or what criterion was used for exclusion. Since those points lie precisely in the thickness range that determines the reported d0π = 2.30 ± 0.01 nm, the selection rule matters. Please state the exclusion criterion, show the fit with and without those points, or justify their removal with a quantitative outlier test.","section":"§IV, Fig. 4(a)"},{"comment":"The quoted uncertainties (0.01–0.03 nm) are only the statistical errors from the nonlinear least-squares fits. They do not include systematic contributions from the 0.5 nm nominal thickness steps, the ±0.05 nm dead-layer correction determined from the magnetization intercept, or run-to-run variations between sputtering runs. The abstract's two-significant-digit values are therefore presented with false precision. Please report systematic uncertainties arising from thickness calibration, dead-layer uncertainty, and run-to-run reproducibility, and propagate them into the switching-energy estimate in the conclusion.","section":"Table I"},{"comment":"The claim that 'the NiFeCr thickness required to achieve the π state increased almost inversely proportionately to the decrease in magnetization, so that the switching energy is hardly changed' relies on the model-dependent d0π value and on the substitution Esw ∝ Ms^2 tF^2. The ratio Ms(NiFe)/Ms(NiFeCr) ≈ 1.61 and the ratio d0π(NiFeCr)/d0π(NiFe) ≈ 1.54 give a switching-energy ratio near unity only for the specific fitted d0π values. A 0.2 nm shift in d0π(NiFeCr)—well within the plausible systematic uncertainty given the thin-film thickness control—changes the switching-energy ratio by roughly 20%. The conclusion should either carry an explicit caveat that this estimate is contingent on the Eq. (4) fit, or be rephrased to distinguish the measured quantities from the derived estimate.","section":"§V, Conclusion"}],"minor_comments":[{"comment":"The caption's 'msat/Area' should be typeset consistently (e.g., as msat/Area with subscripts), and the text contains a typo: 'remanance' should be 'remanence'.","section":"§III A, Fig. 2 caption"},{"comment":"The RSJ expression V = sign(I)RN ℜ{sqrt(I^2 − Ic^2)} would be clearer if the sign convention for Ic (positive critical current) and the branch of the square root were stated explicitly.","section":"§III B, Eq. (1)"},{"comment":"The caption uses 'dNiFe' in panel (b) but the text refers to 'dNiFe'; please use a single notation for the ferromagnetic-layer thickness, e.g., dF, in both panels and in Table I.","section":"Fig. 4 caption"},{"comment":"The sentence 'the figure in panel (b) was published previously in Ref. 40' should clarify whether the NiFe data are reproduced from that reference or remeasured for this work, since Fig. 4(b) appears to include a previously published data set.","section":"§IV, second paragraph"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a useful, honest experimental paper from the Birge group, and it delivers a clear negative result for NiFeCr as a cryogenic-memory material. The 0–π transition at 2.3 nm, the rapid 0.36 nm decay length, and the roughly two-order suppression of Ic in the π-state are the first transport measurements on NiFeCr Josephson junctions, extending the group's systematic alloy comparison (NiFe, NiFeMo, NiFeCu, NiFeNb). The magnetic characterization is also careful: Ms = 579 ± 16 kA/m for NiFeCr versus 935 ± 17 for NiFe, with dead-layer estimates, and coercivity versus thickness data. That's real added value for material-selection tables.\n\nThe central qualitative claim is robust. The IcRN(dF) data show a clear minimum near 2.3 nm for NiFeCr and a much faster decay than NiFe; this does not depend on the specific fit form. The conclusion that NiFeCr is not a good replacement for Permalloy in π-junction memory follows directly.\n\nWhere I part ways with the abstract's precision: the quantitative numbers d0π = 2.30 ± 0.01 nm and ξF1 = 0.36 ± 0.01 nm come from fitting Eq. (4), a damped sinusoid whose regime of validity for NiFeCr the authors explicitly say is unclear. The reported uncertainties are pure fit statistics; they don't include model choice or the 0.5 nm thickness step calibration. The switching-energy conclusion (Esw ∝ Ms^2 tF^2) is an estimate, not a measurement, and depends on d0π. So the abstract overstates certainty, but the paper itself is more careful and acknowledges the regime caveat. The treatment of the couple of second-run points near the transition is vague—the authors mention exceptions to the fit but don't describe how they were handled in the fitting. That's a legitimate referee request.\n\nThe stress-test note makes a fair point but doesn't overturn the paper. A different functional form might shift d0π by a few tenths of a nanometer, which could change the 'switching energy hardly changed' conclusion quantitatively, but not the qualitative rejection of NiFeCr for memory applications.\n\nWho's this for? Researchers working on ferromagnetic Josephson junctions, particularly those doing material screening for superconducting memory. The paper is well-written, self-critical, and the data appear to be taken carefully. I'd send it to peer review; a competent referee should ask for model-uncertainty quantification and clarification of the fitting exclusion, but this is not a desk-reject.","headline":"First transport data on NiFeCr Josephson junctions deliver a robust negative result for cryogenic memory; the abstract overstates precision, but the paper is honest and deserves peer review.","tokens_in":12143,"tokens_out":2984,"would_cite":true,"duration_ms":25968,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["74.50.+r","85.25.Cp"],"model":"deepseek-v4-flash","headline":"Adding 9% chromium to Permalloy suppresses the Josephson critical current by nearly two orders of magnitude in the π state, making NiFeCr unsuitable as the soft magnetic layer in cryogenic memory junctions.","keywords":["Josephson junctions","π-junctions","ferromagnetic Josephson junctions","NiFeCr","Permalloy","cryogenic memory","critical current","0-π transition"],"falsifier":"Measure the critical current–resistance product for NiFeCr junctions with ferromagnet thickness from 3.7 nm up to roughly 6 nm: Eq. (4) predicts a continued exponential decay and a second 0–$\\pi$ transition whose position is fixed by $\\xi_{F2} = 0.67$ nm. If the second lobe appears at a thickness incompatible with that prediction, or the decay is not exponential, the reported transition thickness and decay length are model-dependent rather than intrinsic material properties.","tokens_in":10976,"feed_emoji":"🧲","tokens_out":11029,"duration_ms":91300,"temperature":0.7,"pith_summary":"The paper tests whether Cr-doped Permalloy (Ni73Fe18Cr9) can replace standard Permalloy (Ni80Fe20) as the soft magnetic layer in ferromagnetic Josephson junctions for cryogenic memory. It finds that the alloy lowers the saturation magnetization to about 62% of Permalloy's value, as intended, but at the cost of a supercurrent that is nearly two orders of magnitude smaller in the π state and a decay length of only 0.36 nm, less than half that of NiFe. The 0–π transition moves from 1.49 nm in NiFe to 2.30 nm in NiFeCr, so the ferromagnet thickness needed for a π junction grows almost in proportion to the magnetization drop, leaving the switching energy essentially unchanged. The authors conclude that NiFeCr does not meet the requirements for memory applications that demand π-junctions with large critical current density.","feed_headline":"NiFeCr suppresses π-junction supercurrent by two orders of magnitude","feed_subtitle":"The alloy's 0.36 nm decay length and high resistivity rule it out for cryogenic memory.","key_machinery":"The analysis is carried by an exponentially damped sinusoidal fit to the critical-current–resistance product, $I_c R_N = V_0 \\exp(-d_F/\\xi_{F1}) |\\sin((d_F - d_{0\\pi})/\\xi_{F2})|$, where $d_F$ is the ferromagnet thickness. The function is an approximate interpolation between the diffusive-limit and intermediate-limit theories of supercurrent through a ferromagnetic layer, and the fit extracts the decay length $\\xi_{F1}$, the oscillation length $\\xi_{F2}$, and the first 0–$\\pi$ transition thickness $d_{0\\pi}$. Those three numbers carry the paper's quantitative claim that NiFeCr suppresses the supercurrent too strongly for memory applications.","core_discovery":"On the paper's own terms, the central discovery is a negative result for a proposed material substitution: adding 9% Cr to Permalloy suppresses the Josephson critical current by nearly two orders of magnitude in the π state, and the NiFeCr thickness required to reach the π state increases almost inversely with the decrease in magnetization, so the switching energy is hardly changed. The quantitative signatures are the fitted values of the critical-current–resistance product versus ferromagnet thickness: a 0–π transition at $d_{0\\pi} = 2.30 \\pm 0.01$ nm, a decay length $\\xi_{F1} = 0.36 \\pm 0.01$ nm, and an oscillation length $\\xi_{F2} = 0.67 \\pm 0.03$ nm, compared with 1.49 nm, 0.67 nm, and 0.85 nm for NiFe. The paper reads these numbers as evidence that NiFeCr is not a viable replacement for Permalloy in cryogenic memory junctions, despite its promising low switching fields of a few millitesla.","pith_inferences":["If the same trade-off holds for other doped Permalloys, the search for a cryogenic-memory soft magnet should prioritize alloys whose exchange stiffness and mean free path survive doping, not merely those with lower saturation magnetization.","The sub-nanometer decay length could be dominated by spin-flip scattering from Cr impurities; a testable extension would be to measure NiFeCr junctions with different Cr concentrations and see whether $\\xi_{F1}$ tracks the Cr content.","The single-lobe data leave the fitted functional form unverified beyond 3.7 nm; mapping a second 0–π oscillation would either confirm the assumed fit or reveal that the 2.3 nm and 0.36 nm values are artifacts of the chosen functional form."],"forward_implications":["NiFeCr should not be adopted as the soft magnetic layer in cryogenic memory circuits that need π-junctions with large critical current density.","The short 0.36 nm decay length means NiFeCr junctions must be kept extremely thin, tightening fabrication tolerances compared with NiFe.","Because the π-state thickness rose from 1.49 nm to 2.30 nm while magnetization fell, the magnetic switching energy, which scales as $M_s^2 t_F^2$, is barely reduced by Cr doping.","The close similarity between NiFeCr and NiFeMo behavior near the 0–π transition suggests that alloying-induced softening generally comes with a suppressed supercurrent."],"supporting_citations":[{"why":"supplies the NiFe dataset that is fit with the same Eq. (4) and provides the direct comparison for the NiFeCr results.","marker":"[40]"},{"why":"provides the NiFeMo dataset, which shows behavior near the 0–π transition very similar to NiFeCr and is compared in Table I.","marker":"[42]"},{"why":"reports the structural and magnetic properties of NiFeCr films that motivated the choice of 9% Cr.","marker":"[61]"},{"why":"documents the small and rapidly decaying critical current in CuNi alloy, the precedent for alloy doping suppressing Ic.","marker":"[44]"},{"why":"gives the diffusive-limit theory of the oscillatory-decaying IcRN that Eq. (4) approximates.","marker":"[5]"},{"why":"gives the intermediate-limit formula derived from the Eilenberger equation that Eq. (4) also approximates.","marker":"[65]"}],"fun_headline_variants":["NiFeCr π-junction: 100x current loss, no energy gain","Cr-doped Permalloy: 0.36 nm decay length kills π-junction","Switching field win wiped out by low critical current in NiFeCr","Permalloy with Cr: magnetization down, supercurrent down harder"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The numbers 2.3 nm and 0.36 nm come from fitting the data to one assumed functional form, Eq. (4), and the paper itself notes that NiFeCr junctions may not lie cleanly in either the diffusive or intermediate transport regime; if another valid function fits the data, those values change.","fun_headline_variants_meta":{"raw":{"variants":["NiFeCr π-junction: 100x current loss, no energy gain","Cr-doped Permalloy: 0.36 nm decay length kills π-junction","Switching field win wiped out by low critical current in NiFeCr","Permalloy with Cr: magnetization down, supercurrent down harder"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00023,"raw_usage":{"total_tokens":1510,"prompt_tokens":1001,"completion_tokens":509,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":617,"completion_tokens_details":{"reasoning_tokens":426}},"tokens_in":617,"tokens_out":509,"duration_ms":5518,"temperature":1.0,"reasoning_tokens":426,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T20:50:39.201361+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the critical current–resistance product for NiFeCr junctions with ferromagnet thickness from 3.7 nm up to roughly 6 nm: Eq. (4) predicts a continued exponential decay and a second 0–$\\pi$ transition whose position is fixed by $\\xi_{F2} = 0.67$ nm. If the second lobe appears at a thickness incompatible with that prediction, or the decay is not exponential, the reported transition thickness and decay length are model-dependent rather than intrinsic material properties.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"supplies the NiFe dataset that is fit with the same Eq. (4) and provides the direct comparison for the NiFeCr results."},{"cited_title":"Devonport , author A","cited_arxiv_id":null,"evidence_quote":"reports the structural and magnetic properties of NiFeCr films that motivated the choice of 9% Cr."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"documents the small and rapidly decaying critical current in CuNi alloy, the precedent for alloy doping suppressing Ic."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"gives the diffusive-limit theory of the oscillatory-decaying IcRN that Eq. (4) approximates."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"gives the intermediate-limit formula derived from the Eilenberger equation that Eq. (4) also approximates."}],"review_version":1}