{"id":"eb267d5d-1662-4713-8e03-cd1048e383f2","arxiv_id":"1908.01645","paper_version":5,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Measurement of the polar-phase gap in 3He in nafen shows the predicted T^3 temperature law, providing the first experimental evidence for a Dirac nodal line and supporting a generalized Anderson theorem.","lead":"This paper reports NMR measurements of superfluid helium-3 confined in a nanostructured material, showing that the energy gap in the polar phase shrinks with temperature as T cubed, the signature of a Dirac nodal line. The result supports the idea that columnar non-magnetic defects leave the paired state nearly unchanged, a generalization of the Anderson theorem.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The T^3 evidence may be partly circular: the normalized shift uses an extrapolated ω(0), and no free-exponent fit with uncertainties is shown; without that, the nodal-line signature is not established.","rationale":"The theoretical part of the paper is internally consistent: the supplementary BCS derivation leading to 1 - Δ(T)/Δ(0) = a(T/Tc)^3 is self-contained, and the exponent T^3 is the expected signature of a Dirac nodal line. The main weakness is empirical rather than theoretical. The reported 'observation' of the T^3 law is not demonstrated in a way that rules out alternative exponents or a normalization artifact, because ω(0) is an extrapolated quantity and the extrapolation procedure is not described. The reader's weakest assumption concerns orientational disorder in nafen and the applicability of the Fomin-Anderson theorem; that is a real issue, but it is secondary to the existence of the nodal line, and the authors explicitly acknowledge the disorder. The more load-bearing question is whether the raw data actually force a cubic power law. A free-exponent fit with error bars would settle this cleanly, which is why the verdict should remain conditional: accept with the condition that the raw-data analysis be reported.","tokens_in":7903,"tokens_out":12339,"duration_ms":142701,"concrete_test":"Obtain the raw NMR shift data shown in Fig. 3(b) for T < 0.4Tc and fit Δω(T) = A + B T^n with A, B, and n all free, using the experimental noise level for weights. Report n with a 95% confidence interval and the corresponding a. Repeat with fixed n = 2, n = 3, and n = 4, and compare fits using an information criterion or F-test. If the extracted n is consistent with 3 and not with 2 or 4, the nodal-line signature is supported; if n is pulled toward 3 only because ω(0) was fit with a T^3 law, the central evidence fails.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim rests on Fig. 3(c), where the relative NMR shift is plotted against (T/Tc)^3 and fitted with prefactor a = 0.38. The y-axis is normalized by ω(0), which the caption says was 'determined by extrapolation of data in panel (b).' If that extrapolation used a fit containing the same T^3 functional form, then the apparent linearity in (T/Tc)^3 is not independent evidence: the exponent and prefactor are imposed by the normalization procedure. The paper gives no error bars, does not state the extrapolation method, and does not compare the raw shift against competing low-T forms such as T^2 (Weyl nodes) or T^4. The theoretical value a = 0.57 in the supplementary material is for a clean bulk polar phase, while the measured a = 0.38 is only 'comparable' after invoking an unmeasured strong-coupling correction. Thus the experimental evidence for the Dirac nodal line is only as strong as a free-exponent fit of the raw data.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports NMR measurements of the frequency shift of superfluid 3He confined in nafen-243 and uses the low-temperature shift to infer the temperature dependence of the quasiparticle gap of the polar phase. The authors argue that the normalized shift follows 1 - Delta^2(T)/Delta^2(0) proportional to (T/Tc)^3 with prefactor a = 0.38, matching the expected T^3 law for a Dirac nodal line and comparable with the weak-coupling BCS value a = 0.57 derived in the Supplementary Material. From this they conclude that the Dirac nodal line exists in the polar phase and that columnar non-magnetic disorder does not destroy the gap, supporting Fomin's extension of the Anderson theorem. The Supplementary Material contains a self-contained BCS derivation of the prefactor, including Delta(0)/Tc = 2.46.","tokens_in":8104,"tokens_out":9306,"duration_ms":86242,"significance":"If the claims hold, this would be the first experimental evidence for the Dirac nodal line in the polar phase of superfluid 3He and an important demonstration of disorder-robust unconventional pairing, with implications for topological flat bands and Bogoliubov Fermi surfaces. The theoretical scaffold is a genuine strength: the Supplementary BCS derivation is parameter-free once Delta(0)/Tc is fixed, gives the explicit value a = 0.57, and the comparison with the measured a = 0.38 via a 15% strong-coupling enhancement of Delta(0)/Tc is a falsifiable quantitative statement. However, the experimental evidence as presented lacks the procedural and statistical detail needed to test the central T^3 claim, so the significance is conditional on the requested revisions.","major_comments":[{"comment":"The normalization of the y-axis in Fig. 3(c) uses omega(0), which the caption states was \"determined by extrapolation of data in panel (b),\" but neither the extrapolation function nor its uncertainty is given. If that extrapolation already assumes the cubic form in Eq. (3), then the linearity of the normalized plot is imposed by construction rather than independently tested. Please state the extrapolation procedure, plot the raw shift omega(T) - omega_L against (T/Tc)^n for n = 2, 3, and 4 over the full measured range, fit the raw data with a free exponent, and report the fitted exponent and its uncertainty.","section":"Experimental, Fig. 3(c)"},{"comment":"No error bars are shown anywhere in Fig. 3, and the stated prefactor a = 0.38 is reported without an uncertainty. The comparison with the theoretical a = 0.57 hinges on a strong-coupling correction to Delta(0)/Tc of about 15%, so without estimates of the uncertainties in Tc, omega(0), and the fitted slope it is impossible to judge whether the discrepancy is physically meaningful or instead indicates a breakdown of the clean-limit BCS description. Please add at least standard errors from the fits and a sensitivity analysis of the omega(0) extrapolation.","section":"Experimental, Eq. (3) and Fig. 3(c)"},{"comment":"The authors explicitly acknowledge that real nafen has orientational disorder of the strands and that this \"somewhat violate[s] the Anderson theorem,\" yet the support for the Fomin-Anderson theorem is based on a comparison of the measured prefactor with the clean-limit value a = 0.57. The manuscript should quantify the expected effect of strand disorder on the T^3 prefactor (or on Delta(0)/Tc), or otherwise justify why the clean-limit comparison remains the relevant baseline. Without this, the conclusion that the Anderson theorem protects the polar phase in nafen is not fully supported.","section":"Experimental and Conclusion"}],"minor_comments":[{"comment":"The right-hand side \"2a T^3/T_c^2\" is dimensionally inconsistent with Eq. (1); it should read \"2a (T/Tc)^3\" or equivalently \"2a T^3/T_c^3.\"","section":"Eq. (3)"},{"comment":"The displayed result \"a T^3/T_c\" should be \"a (T/Tc)^3\"; as written the equation is dimensionally incorrect.","section":"Supplementary Material, Eq. (21)"},{"comment":"The statement that the x integration \"has been extended to infinity\" deserves one explanatory sentence: the exponential factor makes the extension harmless at low T, since contributions with x^2 + xi^2 much larger than (T/Delta(0))^2 are exponentially suppressed.","section":"Supplementary Material, after Eq. (13)"},{"comment":"Several typographical errors remain, including \"demonstartes\" in the Conclusion, \"againts\" in the final paragraph, \"supefluid\" in the Conclusion, and \"intergrals\" in the Supplementary Material.","section":"General"},{"comment":"The caption introduces omega(0) without defining it as the zero-temperature extrapolated frequency shift; please define all symbols in the caption.","section":"Fig. 3 caption"}],"recommendation":"major_revision","confidential_remarks":"This is a short experimental letter with a credible but incompletely documented central analysis. The main risk is not the theory but the normalization and extrapolation of the NMR shift, which could impose the very T^3 law the paper claims to measure. The requested analyses—raw data fits with a free exponent, uncertainties, and a sensitivity study of omega(0)—are feasible and within the manuscript's scope. I therefore recommend major revision rather than rejection. The citation pattern appears appropriate, and the complementary theoretical derivation is a useful contribution."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nWhat's new here: the first measurement of the low-temperature gap law in the polar phase of 3He confined in nafen, with a T^3 dependence and prefactor a = 0.38 compared to the BCS clean-limit value a = 0.57. The supplementary theory is a self-contained derivation of the T^3 law and that prefactor, so the comparison is genuine and not a fit. The connection to Fomin's Anderson-theorem extension is prior theory, which the authors cite rather than claim as new. The real contribution is experimental: a clean system (4He preplated, non-magnetic, columnar confinement) and a plausible line-node signature.\n\nWhat the paper does well: the gap-equation derivation is parameter-free (alpha = 1.23 gives Delta(0)/Tc = 2.46 and a = 0.57), so the measured a = 0.38 is a real test. The 15% larger Delta(0)/Tc is consistent with strong coupling, as the authors note from 3He-B. They are also honest about the real material's orientational disorder and the small Tc suppression. For an experimental letter, it is compact and readable, and the citation pattern looks appropriate.\n\nSoft spots, in proportion: the reported data are thin. No error bars appear on the NMR shifts, and the normalization uses omega(0) extrapolated from the same data set. If that extrapolation effectively fixes the functional form, the T^3 display is not fully independent evidence. I checked the stress-test note about circularity; it is a fair request to see the raw shifts and a free-exponent fit. But it is not a structural flaw, because the extrapolation yields one number and the low-T relative shift could still show curvature if a non-T^3 law were present. Still, the paper should show that analysis. Also, there is no disorder-free control, so the Anderson-theorem claim is inferred rather than demonstrated; the authors acknowledge this. The discrepancy between 0.38 and 0.57 is attributed to strong coupling without a calculation, which is plausible but hand-waved.\n\nBottom line: this is a solid experimental letter with one clear theory comparison. The nodal-line interpretation is likely right, but the evidence as presented is not airtight because of missing uncertainties and the normalization procedure. The paper deserves a serious referee, who should ask for raw data, a free-exponent fit, and error propagation.\n\nRecommendation: send to peer review, not desk reject.","headline":"First plausible experimental sighting of the polar-phase nodal line, but the T^3 evidence needs raw data and error bars before it fully lands.","tokens_in":8655,"tokens_out":2715,"would_cite":true,"duration_ms":27989,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The polar phase of superfluid helium-3 has a Dirac nodal line in its quasiparticle spectrum, and the line survives columnar disorder.","keywords":["polar phase","superfluid helium-3","Dirac nodal line","Anderson theorem","columnar disorder","NMR frequency shift","topological superfluid","nafen confinement"],"falsifier":"Measure the low-temperature NMR frequency shift of the polar phase in nafen samples with deliberately increased strand misalignment: if the $T^3$ law, or its prefactor $a$, moves substantially outside the strong-coupling window $a = 0.38$–$0.57$ as disorder grows, then the conservation of $p_z$ and the Anderson-type protection are not what is being observed.","tokens_in":7701,"feed_emoji":"🌀","tokens_out":11090,"duration_ms":105799,"temperature":0.7,"pith_summary":"This paper reports the first experimental evidence that the polar phase of superfluid $^3$He hosts a Dirac nodal line, a circle of zero gap in the quasiparticle spectrum. The evidence is the temperature dependence of the NMR frequency shift: at $T < 0.4T_c$ the relative shift follows $1 - \\Delta^2(T)/\\Delta^2(0) = 2a(T/T_c)^3$, with measured $a = 0.38$ close to the weak-coupling BCS value $a = 0.57$. A line node gives a density of states linear in energy, $N(\\omega) \\propto \\omega$, so the cubic temperature law is the distinctive signature of a nodal line rather than a full gap or point nodes. The result matters because it shows that strong scattering from the parallel nanoscale columns of the confining nafen material does not destroy the gap or its angular structure, in line with an extension of the Anderson theorem to this topological superfluid.","feed_headline":"NMR finds the predicted Dirac nodal line in helium-3","feed_subtitle":"Measured gap obeys the predicted T-cubed law despite disorder, supporting the Anderson theorem.","key_machinery":"The load-bearing object is the Dirac nodal line: with gap function $\\Delta(T)\\cos\\mu$, the gap vanishes at $\\mu = \\pi/2$, forming a circle on the Fermi surface around which the Berry phase winds by $\\pi$. That line produces $N(\\omega) \\propto \\omega$, which is what generates the cubic temperature dependence of the gap and of the NMR shift. The second mechanism is the columnar-confinement model of nafen: ideal impurities are infinitely long, straight, parallel, non-magnetic strands that scatter specularly and therefore conserve the quasiparticle momentum $p_z$ along the strand axis, so the polar phase splits into independent two-dimensional superconductors with gap $\\Delta(p_z)$, each fully gapped and each protected by the Anderson theorem against non-magnetic scattering. The measured observable is the NMR frequency shift $\\omega(T) - \\omega_L = \\Omega_P^2(T)/(2\\omega_L)$, with $\\Omega_P \\propto \\Delta(T)$, which turns the gap's $T^3$ law into a directly measurable relative shift.","core_discovery":"The central claim is that the polar phase of superfluid $^3$He, stabilized by confinement in the aligned nanoscale strands of the nafen material, has a topological Dirac nodal line in its Bogoliubov quasiparticle spectrum, and that this line is robust against the columnar defects of the confining material. The measured low-temperature NMR frequency shift follows $1 - \\Delta^2(T)/\\Delta^2(0) = 2a(T/T_c)^3$ with $a = 0.38$, in reasonable agreement with the weak-coupling BCS prediction $a = 0.57$; because the Leggett frequency entering the shift is proportional to $\\Delta(T)$, this is the $T^3$ law expected when the density of states is linear in energy, $N(\\omega) \\propto \\omega$. The same data imply $\\Delta(0)/T_c$ is about 15% larger than the weak-coupling value, a deviation the authors attribute to strong-coupling effects, as known for other superfluid phases of $^3$He. On this evidence, the columnar defects act as an ensemble of independent two-dimensional subsystems, leaving the gap unmodified as predicted by the columnar-defect extension of the Anderson theorem.","pith_inferences":["A testable extension: varying the distribution of strand tilts in nafen and tracking the prefactor $a$ would map how the Anderson-protected regime breaks down as $p_z$ conservation is lost, giving a quantitative test of the mechanism beyond the single material studied.","If the mechanism is general, the same $T^3$ signature should appear in other columnar-confined unconventional superconductors, and possibly in ultracold atomic gases with engineered 1D disorder, where the polar-phase analogue can be tuned by hand.","An alternative reading of the measured $a = 0.38$ versus weak-coupling $0.57$ is that residual disorder renormalises the effective gap, which could be checked by comparing the same NMR measurement in a cleaner or dirtier nafen sample.","A design consequence the authors do not state: aligning disorder along a symmetry axis may be a practical route to protecting topological superconductivity, which could inform the fabrication of superconducting devices from disordered materials."],"forward_implications":["The low-energy density of states in the polar phase should be linear in energy, so thermodynamic probes such as heat capacity should show the same $T^3$ scaling as the gap.","Under a superflow in the plane of the nodal line, the line should evolve into a Bogoliubov Fermi surface with two touching pseudo-Weyl points, and the gap at $T=0$ should be suppressed as $v_s^3/(3c^3)$.","A surface cut normal to the strands should host a topological flat band of fermionic quasiparticles, a direct consequence of the bulk-boundary correspondence for Dirac lines.","The disorder-protection mechanism implies that other nodal superconductors may be made robust to impurities by aligning the scatterers along a symmetry direction, not just by avoiding magnetic scattering.","Deviations from the ideal columnar model, such as strand misalignment or magnetic surface layers, should show up as a small suppression of $T_c$ and a measurable change in the prefactor $a$."],"supporting_citations":[{"why":"Provides the extension of the Anderson theorem to the polar phase: columnar non-magnetic defects conserve $p_z$, so each 2D subsystem is fully gapped.","marker":"[11]"},{"why":"Supplies the original Anderson theorem that non-magnetic impurities do not change the gap or $T_c$ of a conventional s-wave superconductor, the base of the columnar-defect argument.","marker":"[12]"},{"why":"Gives the phase-diagram data in nafen-243 showing the polar phase with small $T_c$ suppression, which the present experiment extends to the gap's temperature dependence.","marker":"[13]"},{"why":"Shows how magnetic scattering changes the superfluid phases in nafen, motivating the 4He preplating that makes the strand surfaces non-magnetic.","marker":"[14]"},{"why":"Defines the Bogoliubov Fermi surface that the paper argues forms from the nodal line under superflow.","marker":"[21]"},{"why":"Predicts the cubic suppression of the gap with superfluid velocity at $T=0$, a consequence the paper connects to the observed node line.","marker":"[22]"}],"fun_headline_variants":["Helium-3 polar phase exhibits predicted Dirac nodal line","Columnar defects fail to break helium-3 nodal line","Anderson theorem protects nodal line in helium-3","T^3 gap law reveals robust nodal line in helium-3","Nodal line in superfluid helium-3 immune to disorder"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the real nafen material is close enough to the ideal model of perfectly straight, parallel, smooth, non-magnetic columns that quasiparticle momentum along the column direction is conserved; the paper itself notes that the actual orientational disorder of the strands somewhat violates this, producing the small $T_c$ suppression.","fun_headline_variants_meta":{"raw":{"variants":["Helium-3 polar phase exhibits predicted Dirac nodal line","Columnar defects fail to break helium-3 nodal line","Anderson theorem protects nodal line in helium-3","T^3 gap law reveals robust nodal line in helium-3","Nodal line in superfluid helium-3 immune to disorder"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000835,"raw_usage":{"total_tokens":3657,"prompt_tokens":970,"completion_tokens":2687,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":586,"completion_tokens_details":{"reasoning_tokens":2605}},"tokens_in":586,"tokens_out":2687,"duration_ms":19655,"temperature":1.0,"reasoning_tokens":2605,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:06:43.021320+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the low-temperature NMR frequency shift of the polar phase in nafen samples with deliberately increased strand misalignment: if the $T^3$ law, or its prefactor $a$, moves substantially outside the strong-coupling window $a = 0.38$–$0.57$ as disorder grows, then the conservation of $p_z$ and the Anderson-type protection are not what is being observed.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the extension of the Anderson theorem to the polar phase: columnar non-magnetic defects conserve $p_z$, so each 2D subsystem is fully gapped."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the original Anderson theorem that non-magnetic impurities do not change the gap or $T_c$ of a conventional s-wave superconductor, the base of the columnar-defect argument."},{"cited_title":"Dmitriev, A.A","cited_arxiv_id":null,"evidence_quote":"Gives the phase-diagram data in nafen-243 showing the polar phase with small $T_c$ suppression, which the present experiment extends to the gap's temperature dependence."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows how magnetic scattering changes the superfluid phases in nafen, motivating the 4He preplating that makes the strand surfaces non-magnetic."},{"cited_title":"Agterberg, P.M.R","cited_arxiv_id":null,"evidence_quote":"Defines the Bogoliubov Fermi surface that the paper argues forms from the nodal line under superflow."},{"cited_title":"Muzikar and D","cited_arxiv_id":null,"evidence_quote":"Predicts the cubic suppression of the gap with superfluid velocity at $T=0$, a consequence the paper connects to the observed node line."}],"review_version":1}