{"id":"005a7c78-5504-4315-9fc9-f23a2c11d562","arxiv_id":"2506.01330","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":0.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Ultrapure niobium is a type-I superconductor within 0.2 K of Tc, turning type-II at lower temperatures.","lead":"This paper reports old measurements showing ultrapure niobium behaves as a type-I superconductor within a narrow range near its critical temperature. It is an English translation of a 1974 Russian study, released because recent work has revived interest in pure niobium.","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Type-I claim rests on κ0 = 0.702, only 0.7% below the 1/√2 threshold, from an unquantified dirty-limit extrapolation; a small systematic shift would flip the classification.","rationale":"The reader's weakest assumption identified the same load-bearing point: the linear Goodman extrapolation that yields κ0 = 0.702 is fragile because it is an unquantified fit to four points lying very close to the type-I/II boundary. My read agrees but sharpens the technical basis: (i) the 0.7% margin is comparable to plausible systematic errors in κ; (ii) the Goodman relation is a dirty-limit result, while the purest wires are in the clean limit relative to ξ0; and (iii) the massive-single-crystal evidence (Hc1 ≈ Hc2 near Tc) is not discriminating because a type-II sample with κ just above 1/√2 would produce the same near-coincidence. The paper contains independent checks that could support the conclusion—e.g., the direct κ(t=1) values for Nb8 and Nb9 are also below threshold—but without error analysis the status remains conditional. No change to the reader's verdict is needed; the concern is real and the conditionality is appropriate.","tokens_in":9248,"tokens_out":12170,"duration_ms":131818,"concrete_test":"Reanalyze the κ versus √(ργ) data from Table II and Fig. 6 with a nonlinear fit (e.g., adding a quadratic term) and with uncertainty bands propagated from the scatter of the κ1(T) and κ2(T) curves near t = 1 and from the ±10 mK temperature accuracy. If the extrapolated intercept at ρ = 0 moves above 1/√2, or if the confidence interval for the intercept includes 1/√2, the type-I classification of ultrapure niobium near Tc is not statistically established.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central conclusion that ultrapure niobium is type-I near Tc rests on the fitted intercept κ0 = 0.702 in Eq. (3), which is only 0.7% below the type-I/II threshold 1/√2 ≈ 0.7071. The fit uses four wire samples (Nb6–Nb9, Table II), with the two purest points (κ = 0.705 and 0.703) already within 1% of the boundary. No error bars are given; a 1% systematic overestimate of κ in the purest wires—e.g., due to surface scattering or to the difficulty of defining Hc1 and Hc2 when κ is near the boundary—would move κ0 above 1/√2 and overturn Conclusion 1. The Goodman relation (Eq. 3) is a dirty-limit formula, but Table III shows the wire mean free paths are orders of magnitude larger than ξ0 (l/ξ0 > 1300 for α > 3×10^4), so the linear-in-√ρ extrapolation is applied outside its domain of validity. The single-crystal observation that Hc1 approximates Hc2 near Tc (Fig. 3) does not independently prove type-I, because a type-II sample with κ just above 1/√2 would show an equally small Hc2−Hc1. The paper's own caveat that surface effects may raise κ in wires actually cuts in the opposite direction—if κ is inflated, the true κ0 is lower—so the decisive issue is the magnitude of systematic errors in the κ values and the functional form of the extrapolation, both of which are unquantified.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports measurements of critical magnetic fields, magnetization curves, and critical currents on high-purity niobium single crystals and wires with residual resistance ratios up to 61,000. The authors claim that ultrapure niobium behaves as a type-I superconductor within about 0.2 K below the critical temperature, based on the near-coincidence of Hc1 and Hc2 in magnetization curves of massive single crystals and on a linear extrapolation of the Ginzburg-Landau parameter κ(t=1) to zero residual resistivity using Goodman's relation, yielding κ0 = 0.702 < 1/√2. The paper also discusses surface-related hysteresis, anisotropy of the upper critical field, and critical currents in longitudinal fields interpreted via force-free current distributions. The manuscript is an English translation of a 1974 Russian publication.","tokens_in":9602,"tokens_out":7950,"duration_ms":73260,"significance":"If the central claim is correct, the paper provides strong experimental support for the intrinsic type-I character of clean niobium near Tc, a question that has recently attracted renewed attention (e.g., Ref. [6]). The reported data on extremely pure samples (α up to 61,000) are unique and include comparisons with band-structure calculations and de Haas-van Alphen data. The paper also contains useful observations on surface hysteresis and force-free critical currents. However, the quantitative foundation of the type-I claim is the extrapolated value κ0 = 0.702, which lies only 0.7% below the type-I/II threshold, and the paper provides no uncertainty analysis.","major_comments":[{"comment":"The central quantitative claim that κ0 = 0.702 < 1/√2 rests on a linear fit to only four wire samples (Nb6–Nb9) with no quoted error bars on κ(t=1). The two purest samples have κ(t=1) = 0.705 and 0.703, within 0.5% of the type-I/II boundary 1/√2 ≈ 0.7071. A systematic error of order 1% in the measured or extrapolated κ—from surface effects, from the difficulty of defining Hc1 and Hc2 near the boundary, or from the extrapolation of κ1(1) and κ2(1) to t = 1—would move the intercept above the threshold and overturn Conclusion 1. The authors should provide an uncertainty estimate or a sensitivity analysis demonstrating that κ0 is robustly below 1/√2.","section":"Section 4, Eq. (3), Table II"},{"comment":"Goodman's relation κ = κ0 + kρ√γ is a dirty-limit result, but the wire samples are far from the dirty limit: the mean free path is orders of magnitude larger than the coherence length (Table III gives l/ξ0 > 1300 for α > 3×10^4 single crystals, and even the α = 9,000 wire has l ≫ ξ0). Applying a linear-in-√ρ extrapolation outside its domain of validity introduces an unquantified model error. The authors should justify the use of this relation for such pure samples, or show that a clean-limit extrapolation (e.g., of the form κ = κ0 + const/l) gives an intercept below 1/√2.","section":"Section 4, Eq. (3)"},{"comment":"The observation that Hc1 = Hc2 near Tc does not by itself establish type-I behavior, because a type-II superconductor with κ just above 1/√2 would exhibit a nearly vanishing Hc2 − Hc1 in the same temperature range. The paper acknowledges that κ cannot be determined accurately from the single-crystal magnetization curves; hence the type-I classification for these samples rests entirely on the wire-sample extrapolation. If the extrapolation is uncertain, Conclusion 1 is not independently supported.","section":"Section 3, Figure 3"}],"minor_comments":[{"comment":"The notation for the type-I/II boundary is inconsistent (e.g., '1/√2', '1/√ 2', '1/√2'); it should be unified.","section":"Throughout"},{"comment":"The unit 'Qe' should be 'Oe' in '1000 Qe'.","section":"Figure 3 caption"},{"comment":"'Mattheis [17]' should be spelled 'Mattheiss [17]'.","section":"Conclusion 3"},{"comment":"The surname is 'Goodman', not 'Goodmann'.","section":"Reference [16]"},{"comment":"The manuscript does not explicitly mark itself as a translation from Fiz. metal. metalloved., 37, 63 (1974); a footnote would inform readers.","section":"Title page"},{"comment":"The sentence 'If ρ → 0, κ → κ0 = 0.702 < 1/√2 (see Figure 6)' appears to refer to a plot that is not shown; Figure 6 displays κ1 and κ2 versus temperature, not the extrapolation to zero residual resistivity.","section":"Section 4, text near Figure 6"},{"comment":"'Pulse setup allowed single sawtooth current pulses of up to 500 A with a duration of from 0.3 to 10 msec' should read 'with durations from 0.3 to 10 msec'.","section":"Section 6"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a translation of a 1974 Russian-language paper, and the data are therefore historical. The recent revival of interest in clean niobium makes the translation timely, but the paper does not contain new measurements or analysis beyond the original. The authors should be transparent about this in the title or abstract. If the journal is open to historical data papers, this is acceptable; otherwise, the lack of new content may be an issue for scope. The self-citation [5] is appropriate. I would be reluctant to accept the type-I conclusion without an uncertainty analysis, but the raw data themselves are of value."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a translation of a 1974 Russian paper, so the headline is: the results are old, the preprint adds no new data or analysis. But the data itself is still the cleanest niobium dataset I know of (α up to 61,000), so the translation has real archival value if you work on niobium or the type-I/II crossover.\n\nWhat the paper does well: it is honest about its own limitations. The authors test surface effects (oxidation, etching), they report the extreme sensitivity of the perfect crystals to mechanical shock, and they compare their superconducting parameters to independent band-structure and dHvA data. The force-free current analysis for the wires is a nice self-contained piece of physics. The English translation looks faithful and the equations are standard.\n\nThe soft spots are exactly where the stress-test points. The type-I classification near Tc rests on κ0 = 0.702 from Eq. (3), which is 0.7% below 1/√2, and that number comes from a linear extrapolation of four wire samples. No error bars are given. Goodman's relation is a dirty-limit formula, but the wires have l/ξ0 > 1300, so the linear-in-√ρ extrapolation is outside its intended domain. The observation that Hc1/Hc2 → 1 near Tc is suggestive but not conclusive: a type-II superconductor with κ just above 0.7071 would show the same near-coincidence. The paper's own caveat that surface effects may raise κ in wires actually helps the type-I case if you believe it, but it also means the extrapolated κ0 is not a clean measurement. So the evidence is plausible but not quantitatively airtight.\n\nWho is this for? People who care about why niobium is an apparently type-II elemental superconductor, and who want the raw high-purity data. The paper deserves a serious referee if submitted as a historical data paper with a clear statement that it is an old measurement. As a new claim, it would need a quantitative error budget on κ.\n\nRecommendation: engage with it, but treat Conclusion 1 as a strong suggestion, not a proof.","headline":"A transparent 1974 translation with unique ultrapure niobium data, but the type-I claim rests on a thin extrapolation that the paper itself does not quantify.","tokens_in":10153,"tokens_out":2342,"would_cite":true,"duration_ms":27057,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["74.25.Ha","74.25.Wx","74.70.Ad"],"model":"deepseek-v4-flash","headline":"This paper reports that ultrapure niobium is a type-I superconductor near its critical temperature, with impurity effects pushing it to type-II behavior.","keywords":["niobium","type-I superconductor","Ginzburg-Landau parameter","residual resistivity","critical magnetic field","magnetization hysteresis","force-free current","clean-limit superconductivity"],"falsifier":"Take a niobium single crystal with residual resistance ratio above 100,000, measure its reversible magnetization at $T_c - T = 0.05$ K, and check whether $H_{c2}$ is strictly larger than $H_{c1}$; alternatively, on a wire with $\\alpha$ above 3,000, determine $\\kappa_1(1)$ by an independent method that does not rely on the linear extrapolation and compare it with $0.707$. Either observation at or above the boundary would falsify the type-I claim.","tokens_in":9050,"feed_emoji":"🧲","tokens_out":10043,"duration_ms":106976,"temperature":0.7,"pith_summary":"Niobium has long been treated as the one pure metal that is a type-II superconductor. This paper argues that its real, impurity-free behavior is type-I in a narrow temperature band just below the critical temperature: for single crystals with residual resistance ratios above about 30,000, the upper and lower critical fields coincide within $T_c - T \\le 0.2$ K. The argument is carried by measuring the Ginzburg-Landau parameter $\\kappa$ on wires of varying purity and extrapolating it to zero residual resistivity; the extrapolated value $\\kappa_0 = 0.702$ falls below the type-I/type-II boundary $1/\\sqrt{2} \\approx 0.707$. If that extrapolation is right, earlier reports of intrinsic type-II niobium were seeing impurity or surface effects, and the cleanest niobium is a superconductor of the first kind near $T_c$, turning type-II only as the temperature drops.","feed_headline":"Ultrapure niobium is type-I superconductor just below its Tc","feed_subtitle":"Extrapolating to zero impurity puts its Ginzburg-Landau parameter at 0.702, below the type-II threshold of 0.707.","key_machinery":"The machine that carries the argument is the Ginzburg-Landau parameter $\\kappa$, computed for each wire sample from the thermodynamic critical field via $H_c^2/8\\pi = -\\int_0^{H_{c2}} M\\,dH$ and from the ratios $\\kappa_1 = H_{c2}/\\sqrt{2}H_c$ and $\\kappa_2^2 = \\frac{1}{2}[1 + \\frac{1}{4\\pi \\cdot 1.16}(dM/dH)_{H_{c2}}]$. At $t = T/T_c = 1$ the two parameters agree, $\\kappa_1(1) = \\kappa_2(1) = \\kappa$, and $\\kappa$ is plotted against the residual resistivity combined with the Sommerfeld constant through $\\kappa = \\kappa_0 + k \\rho \\sqrt{\\gamma}$. The zero-resistivity intercept $\\kappa_0 = 0.702$ is the load-bearing number: it lies below $1/\\sqrt{2}$, the boundary separating type-I from type-II superconductors in Ginzburg-Landau theory. The temperature dependence then moves $\\kappa_1$ and $\\kappa_2$ upward as $T$ falls, which is why the material crosses into the type-II regime.","core_discovery":"The central claim is that ultrapure niobium is a type-I superconductor near its critical temperature. This is established from magnetization curves of single crystals with residual resistance ratios up to 61,000: as the temperature approaches $T_c$, the measured second critical field $H_{c2}$ drops toward the first critical field $H_{c1}$, and in the range $T_c - T \\le 0.2$ K the curves show $H_{c2} = H_{c1}$, which is the signature of a type-I superconductor. Because the massive crystals' magnetization is strongly hysteretic, the precise values of $H_c$, $\\kappa_1$ and $\\kappa_2$ are extracted from polycrystalline wires, whose demagnetizing factor is zero and whose hysteresis from surface currents is reduced. For wires with residual resistance ratios from 91 to 15,000, the value $\\kappa = \\kappa_1(1) = \\kappa_2(1)$ falls with increasing purity and follows the linear dependence $\\kappa = \\kappa_0 + k \\rho \\sqrt{\\gamma}$ on residual resistivity; extrapolating to $\\rho \\to 0$ gives $\\kappa_0 = 0.702 < 1/\\sqrt{2}$. The paper concludes that pure niobium is intrinsically type-I at $T_c$ and becomes type-II at lower temperatures because of the temperature growth of $\\kappa_1$ and $\\kappa_2$.","pith_inferences":["A direct check of the paper's extrapolation would be to measure $\\kappa_1(1)$ on a single crystal with residual resistance ratio above $10^5$; if the value came out at or above $0.707$, the intrinsic type-I classification would need revision, and the paper itself notes that surface-to-volume effects may raise $\\kappa$ in wires, making a crystal measurement the cleaner test.","Because the fitted slope $k$ is 49% larger than the spherical-Fermi-surface value, the impurity contribution to $\\kappa$ is enhanced by niobium's anisotropic Fermi surface; this suggests that theoretical estimates of $\\kappa_0$ for anisotropic clean superconductors should be made with band-structure averages rather than free-electron models.","The same protocol — purification to residual resistance ratio above $10^4$, then $\\kappa$ extrapolation to zero impurity — could decide the intrinsic type-I/type-II status of other marginal elemental superconductors whose classification has historically been blurred by surface and impurity effects."],"forward_implications":["In as-grown, high-purity niobium single crystals within 0.2 K of $T_c$, the magnetization transition should have the first-order character of a type-I superconductor rather than the reversible second-order form of a type-II superconductor.","The same material is type-II at lower temperatures, so standard mixed-state applications such as high-field magnets operating well below $T_c$ are not called into question; only the near-$T_c$ regime is reclassified.","Earlier observations of type-II behavior in nominally pure niobium with residual resistance ratios of a few thousand were seeing the impurity-driven value of $\\kappa$, not the intrinsic clean-limit value.","The wire critical-current data in longitudinal fields are quantitatively consistent with a force-free spiral current distribution, meaning flux capture is essentially absent in the cleanest wires and force-free states can be realized in niobium."],"supporting_citations":[{"why":"Supplies the earlier high-purity niobium data (alpha = 1,400) that the paper's cleaner samples supersede.","marker":"[1]"},{"why":"Records the prior claim of intrinsic type-II behavior at alpha = 2,000 that the present extrapolation overturns.","marker":"[2]"},{"why":"Gives the linear relation between kappa and rho*sqrt(gamma) used for the zero-resistivity extrapolation.","marker":"[16]"},{"why":"Provides the band-structure Fermi-surface values used as the comparison target for the normalized upper critical field and mean Fermi velocity.","marker":"[17]"},{"why":"Supplies earlier bulk upper-critical-field data whose absolute values the purest samples run about two percent below.","marker":"[22]"},{"why":"Gives the weak- and strong-coupling formulas connecting the initial Hc2 slope to the mean Fermi velocity, and the stationary force-free current model used in the wire-current comparison.","marker":"[23]"},{"why":"Introduces the longitudinal force-free current model with spiral current lines used to explain the measured critical currents.","marker":"[30]"},{"why":"Establishes the Hc1/Hc2 = 0.52 threshold that determines whether current destroys superconductivity through a mixed state or directly.","marker":"[31]"}],"fun_headline_variants":["Ultrapure niobium is type-I superconductor near Tc","Niobium's type-I nature revealed only with ultrapurity","Extrapolating to zero impurity: niobium is type-I","Ultrapure niobium: type-I near Tc, type-II at low T","Ultrapure niobium: type-I, not type-II, at Tc"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole classification depends on a linear extrapolation of the Ginzburg-Landau parameter $\\kappa(\\rho\\sqrt{\\gamma})$ measured on four wire samples down to zero residual resistivity, and the extrapolated value $\\kappa_0 = 0.702$ sits so close to the type-I boundary $1/\\sqrt{2} = 0.707$ that modest systematic errors in $H_c$, $H_{c2}$, or the wire surface contribution could move the material to the type-II side.","fun_headline_variants_meta":{"raw":{"variants":["Ultrapure niobium is type-I superconductor near Tc","Niobium's type-I nature revealed only with ultrapurity","Extrapolating to zero impurity: niobium is type-I","Ultrapure niobium: type-I near Tc, type-II at low T","Ultrapure niobium: type-I, not type-II, at Tc"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.002061,"raw_usage":{"total_tokens":8031,"prompt_tokens":962,"completion_tokens":7069,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":578,"completion_tokens_details":{"reasoning_tokens":6974}},"tokens_in":578,"tokens_out":7069,"duration_ms":53391,"temperature":1.0,"reasoning_tokens":6974,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T11:43:44.396788+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a niobium single crystal with residual resistance ratio above 100,000, measure its reversible magnetization at $T_c - T = 0.05$ K, and check whether $H_{c2}$ is strictly larger than $H_{c1}$; alternatively, on a wire with $\\alpha$ above 3,000, determine $\\kappa_1(1)$ by an independent method that does not rely on the linear extrapolation and compare it with $0.707$. Either observation at or above the boundary would falsify the type-I claim.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the earlier high-purity niobium data (alpha = 1,400) that the paper's cleaner samples supersede."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Records the prior claim of intrinsic type-II behavior at alpha = 2,000 that the present extrapolation overturns."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the linear relation between kappa and rho*sqrt(gamma) used for the zero-resistivity extrapolation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the band-structure Fermi-surface values used as the comparison target for the normalized upper critical field and mean Fermi velocity."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies earlier bulk upper-critical-field data whose absolute values the purest samples run about two percent below."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the weak- and strong-coupling formulas connecting the initial Hc2 slope to the mean Fermi velocity, and the stationary force-free current model used in the wire-current comparison."},{"cited_title":"Bergeron, Simple model for longitudinal force-free current flow in superconductors of the second kind, J","cited_arxiv_id":null,"evidence_quote":"Introduces the longitudinal force-free current model with spiral current lines used to explain the measured critical currents."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the Hc1/Hc2 = 0.52 threshold that determines whether current destroys superconductivity through a mixed state or directly."}],"review_version":1}