{"id":"f080bf46-1058-4ecc-ad89-0e9abe3ad02f","arxiv_id":"2602.17000","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"A microscopic framework predicts power attenuation in superconducting rectangular waveguides from 100 GHz to THz, including TLS oxide losses and a predicted Higgs-mode peak in attenuation near the gap frequency.","lead":"This paper builds a microscopic framework to predict how much signal power is lost when millimeter-wave and terahertz signals travel through superconducting rectangular waveguides, covering materials from dirty to clean limits and including losses from surface oxides. It also predicts a new detectable peak in power loss near the superconducting gap frequency, caused by collective Higgs-mode dynamics, which could serve as a practical signature of this elusive mode.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (14) misnormalizes Rs by a factor ℓ/ξ0: clean-limit attenuation values in Figs. 5–6 are off by 10×.","rationale":"The reader's weakest_assumption emphasized local electrodynamics in clean-limit Nb and the omission of the nonlinear integrals from Ref. [23]. While those are legitimate concerns, a more concrete and checkable flaw exists in the linear-response core: Eq. (14) is internally inconsistent with Eq. (29) by a factor ℓ/ξ0. Because Eq. (14) feeds all of the linear-response attenuation calculations in Sec. III, the numerical values reported in Figs. 5–7 and the quantitative comparison with experiment are suspect. The error is a prefactor error, not a conceptual failure, so the paper can be corrected by fixing the normalization and recomputing the affected figures. The Higgs-peak prediction is not directly invalidated, but its linear baselines are affected. This warrants a CONDITIONAL verdict: the paper should be accepted only after the normalization error is corrected and the numerical results are re-evaluated.","tokens_in":22838,"tokens_out":33685,"duration_ms":245962,"concrete_test":"Analytically or numerically evaluate the dirty-limit, low-frequency limit of Eq. (14) using σ from Eq. (3) and λ from Eq. (2), and compare with the standard local formula Rs=(1/2)μ0²ω²λ³σ1 and with Eq. (29) at u=0. For any ℓ/ξ0≠1, the ratio (standard value)/(Eq. 14 value) should be exactly ℓ/ξ0 if the misnormalization is confirmed; if it is 1/(ℓ/ξ0), Eq. (14) is correct and my reduction is wrong. Then recompute the α curves in Figs. 5 and 6 with the corrected prefactor and check whether the stated agreement with Ref. [14] (5×10⁻⁴ dB/cm at WR10, 4–5 K) still holds.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The normalization used for the linear-response surface resistance, Eq. (14), contains 1/sqrt(ℓ/ξ0), but the dirty-limit low-frequency reduction gives sqrt(ℓ/ξ0). Reducing Eq. (14) with σ2≃1/(μ0ωλ²), λ0,dirty²=(ℓ/ξ0)λ0², and μ0σnλ0,dirty²=ℏ/(πΔ0) yields Rs/R0 = (1/2π)(ℓ/ξ0)^(−1/2)(ℏω/Δ0)²(σ1/σn). The standard local result Rs=(1/2)μ0²ω²λ³σ1, combined with R0=μ0Δ0λ0/ℏ, gives instead Rs/R0 = (1/2π)(ℓ/ξ0)^(1/2)(ℏω/Δ0)²(σ1/σn). The paper's own Eq. (29), used in the nonlinear section, has the correct sqrt(ℓ/ξ0) prefactor in its linear term. Thus Eq. (14) is internally inconsistent with Eq. (29) by a factor 1/(ℓ/ξ0). Concretely, every linear-response attenuation value for ℓ/ξ0=10 in Figs. 5–7 is underestimated by a factor of 10; for ℓ/ξ0=0.3 it is overestimated by ≈3.3. This affects the quantitative attenuation constants, the claimed comparison with Nakajima et al.'s 5×10⁻⁴ dB/cm, and the TLS-cross-over temperatures, even if the qualitative 'cleaner is better' conclusion survives. The Higgs-peak calculation in Sec. IV uses Eq. (29), so it is not directly affected, but the linear baselines in Fig. 10 inherit the wrong normalization.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a quasiclassical/Eilenberger linear-response framework for the power attenuation constant α of rectangular superconducting waveguides, valid for arbitrary mean free path within local electrodynamics, and derives an analytic TLS-loss formula. It evaluates NbN, Nb3Sn, and Nb waveguides over standard WR15–WR1 sizes, and then uses the author’s recent Keldysh–Usadel nonlinear-response theory [23] to predict a Higgs-mode peak in α near f ≃ Δ/h. Appendices validate the low-frequency σ2 approximation and document when σ1/σ2 is not small.","tokens_in":23248,"tokens_out":34573,"duration_ms":256751,"significance":"If the numerical results are correct, the paper provides a useful design tool for low-loss millimeter/THz superconducting waveguides and a concrete, falsifiable prediction of a Higgs-mode signature in attenuation. The linear-response core is standard Eilenberger theory with clear equations; the TLS attenuation is an elegant analytic result; and the Higgs prediction is based on a published microscopic calculation. However, several normalization and self-consistency issues in the surface-resistance formulas and in the nonlinear section must be resolved before the quantitative claims can be trusted.","major_comments":[{"comment":"Eq. (12) is dimensionally inconsistent as written: R_s = sqrt(μ0ω/2) sqrt((|σ|−σ2)/|σ|) has units Ω^{1/2} m^{−1/2}, not Ω. The exact local result is R_s = sqrt(μ0ω/(2σ2)) sqrt((|σ|−σ2)/|σ|). Eq. (14) is not generally equivalent to this exact expression; it reduces to the standard result only for σ1/σ2 ≪ 1. Since Figs. 5–7 extend to f ≲ 2Δ/h and Appendix B shows σ1/σ2 is not small there, the reported attenuation near the gap is an uncontrolled approximation unless the exact surface impedance is used.","section":"Sec. II C, Eqs. (12) and (14)"},{"comment":"Eq. (29) is internally inconsistent with Eq. (14). In the low-frequency limit Eq. (14) reduces to R_s/R0 = (1/2π)(ℏω/Δ0)^2 (σ1/σn) (λ/λ0,dirty)^3 / sqrt(ℓ/ξ0), using λ0,dirty² = (ξ0/ℓ)λ0² from the paper’s own dirty-limit definition. Eq. (29) instead has sqrt(ℓ/ξ0) in place of 1/sqrt(ℓ/ξ0), i.e. it is larger by a factor ℓ/ξ0. This changes the linear baselines and the Higgs-peak contrast in Fig. 10 by a factor 10 for ℓ/ξ0=10 and by 0.1 for ℓ/ξ0=0.1. The nonlinear section needs re-evaluation with the corrected prefactor.","section":"Sec. IV A, Eq. (29)"},{"comment":"The stated assumption λ ≫ ξ, which justifies local electrodynamics, is violated for clean-limit Nb. The paper notes this and labels clean Nb curves as qualitative, but the design recommendation for high-purity Nb in Sec. V and the comparison with Nakajima et al. rely on those numbers. A nonlocal/anomalous-skin-effect calculation is needed for quantitative clean-Nb attenuation; otherwise the claims should be restricted to the local regime or the clean-Nb values presented only as indicative.","section":"Sec. II A and Sec. III B"},{"comment":"The explicit forms of I^{qqq}_{1H}, I^{Higgs}_{1H}, and I^{Eliash}_{1H} are omitted, with a pointer to Eq. (84) of Ref. [23]. Because the Higgs peak is a central new claim, the manuscript should include or reproduce these kernels, at least in an appendix. As written, the nonlinear results cannot be verified from the manuscript alone, especially given the normalization inconsistency in Eq. (29).","section":"Sec. IV A, Eqs. (25)–(29)"}],"minor_comments":[{"comment":"The filling-factor interpretation after Eq. (22) is helpful, but the appearance of ε′_r in the denominator of the prefactor should be checked: the final αTLS expression already contains 1/(ε′_r b), so the text should ensure the effective filling factor is stated without double counting.","section":"Eq. (22)"},{"comment":"The text uses both B0 and Bac for the drive amplitude; the figure caption should define the quantity plotted and the relation Bac = μ0 sqrt(⟨H_∥²⟩) explicitly.","section":"Fig. 10"},{"comment":"Typos and formatting artifacts such as 'superc onducting' in the title should be corrected.","section":"Title/Abstract"},{"comment":"The comparison of σ2 with σ2,low is useful; stating the low-frequency expansion in Eq. (10) more explicitly would help readers apply it.","section":"Appendix A"}],"recommendation":"major_revision","confidential_remarks":"The proposed 10× error in Eq. (14) from the stress-test note appears to be based on an inverted dirty-limit relation; with the paper’s own definitions λ0,dirty² = (ξ0/ℓ)λ0², Eq. (14) actually reduces to the standard low-frequency prefactor. The real load-bearing problem is Eq. (29), which differs from Eq. (14) by ℓ/ξ0. The exact surface-impedance formula in Eq. (12) is also dimensionally wrong as printed. These issues are fixable but must be corrected before the quantitative conclusions, especially the Higgs-peak attenuation, can be considered reliable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Checked the stress-test note: it does not hold up. The note assumes λ0,dirty² = (ℓ/ξ0)λ0², but the paper's own Eq. (2) gives λ0,dirty² = (ξ0/ℓ)λ0², the standard dirty-limit result. With the correct relation, Eq. (14) reduces to the standard local surface-resistance normalization. The actual problem is Eq. (29), which has sqrt(ℓ/ξ0) where 1/sqrt(ℓ/ξ0) is required. That typo changes the linear baseline in Fig. 10 for Ta (ℓ/ξ0 = 0.1) by a factor of 10, and would also make clean samples look worse than dirty ones, contradicting the paper's own Fig. 4. So the stress-test's quantitative claim against Figs. 5–6 is unfounded, but there is a genuine internal inconsistency that needs fixing.\n\nThe linear-response core is solid. Applying the Eilenberger conductivity to rectangular waveguides across mean free paths is a useful, clearly presented contribution, and the appendices showing where the low-frequency approximations break down are valuable. The TLS derivation is transparent and the elliptic-integral result is compact and plausible.\n\nThe soft spot is the nonlinear/Higgs section. The central new prediction—a Higgs-mode peak in α—relies on the author's earlier Ref. [23] for the explicit forms of I^qqq, I^Higgs, and I^Eliash. Those integrals are not written out here, so a referee cannot verify the peak without consulting that paper. That is a paper-level gap, not a demonstrated error, but it needs to be addressed.\n\nThe clean-limit Nb results are also outside the strict local regime λ ≫ ξ, and the author says so; they are flagged as qualitative. Good. The material parameters are single-point values with no error bars, so the TLS crossover temperatures are order-of-magnitude estimates, which the text mostly acknowledges.\n\nBottom line: this deserves a serious referee. The linear-response and TLS parts are useful and likely citable; the Higgs section needs either the full expressions or a standalone derivation, and Eq. (29) needs a one-line fix. Send it to review with a request for those revisions.","headline":"Useful waveguide-attenuation framework with a real typo in Eq. (29) and a Higgs claim that depends on an earlier paper; the stress-test note about Eq. (14) is wrong.","tokens_in":23744,"tokens_out":19046,"would_cite":true,"duration_ms":129035,"reading_group":"maybe","serious_thinker":"no","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper develops a microscopic framework that predicts the attenuation of superconducting rectangular waveguides from 100 GHz into the terahertz band for arbitrary mean free path, and identifies a Higgs-mode peak in the attenuation near","keywords":["superconducting waveguides","attenuation constant","complex conductivity","Eilenberger theory","two-level systems","Higgs mode","terahertz","surface resistance"],"falsifier":"Measure the attenuation of a Ta WR5 rectangular waveguide near 160 GHz at T/Tc ≈ 0.2 with an applied ac field of order B0/Bc ≈ 0.02: the theory predicts a drive-amplitude-dependent peak in attenuation at f ≈ Δ/h; a clean measurement showing no such peak would falsify the Higgs-mode claim.","tokens_in":22657,"feed_emoji":"📡","tokens_out":6462,"duration_ms":56450,"temperature":0.7,"pith_summary":"This paper sets out to give a first-principles way to compute the attenuation constant of superconducting rectangular waveguides in the 100 GHz–THz band, valid for any electronic mean free path from dirty to clean. The intended payoff is a design tool: which material purity and waveguide size give ultralow transmission loss for cryogenic astronomy and quantum hardware. The paper also derives a compact analytic formula for loss from two-level systems in native oxide layers, and argues this loss only matters below roughly one-tenth of the critical temperature. Under strong drive, the same framework produces a peak in the attenuation at f ≈ Δ/h, which the paper identifies as a measurable hallmark of the Higgs mode, a collective oscillation of the superconducting gap.","feed_headline":"Superconducting waveguide loss reveals a Higgs-mode peak","feed_subtitle":"Microscopic theory computes attenuation for any mean free path and shows the Higgs mode can be read in transmission at f ≈ Δ/h.","key_machinery":"The load-bearing object is the impurity-scattering-renormalized complex conductivity σ(ℓ,T,ω) from the Eilenberger/Keldysh–Usadel theory, replacing the dirty-limit Mattis–Bardeen expressions; it is the only place microscopic superconductivity enters. It feeds the surface impedance Zs = sqrt(−iμ0ω/σ), then the perturbative TE10 attenuation formula. Two supporting pieces are the TLS attenuation formula that reduces the surface-oxide loss to complete elliptic integrals, and the third-order nonlinear response coefficient whose Higgs term produces the attenuation peak near f = Δ/h.","core_discovery":"The central claim is that the attenuation of a superconducting rectangular waveguide can be obtained from the microscopic complex conductivity σ(ℓ,T,ω) computed within the Eilenberger formalism, which interpolates between the Mattis–Bardeen dirty limit and the clean limit, so no mean-free-path assumption is needed. Feeding that conductivity into the standard surface-impedance formula and the TE10-mode attenuation formula gives the attenuation constant for materials such as Nb, NbN, and Nb3Sn across standard waveguide sizes from WR15 to WR1. The paper further claims that, in the strong-excitation regime, the amplitude dependence of the dissipative conductivity—computed to third order in the f","pith_inferences":["If the predicted Higgs peak survives nonperturbative checks, a simple transmitted-power measurement on a short waveguide could become a practical Higgs-mode detector, complementing optical-pump terahertz experiments.","The clean-limit Nb curves are the least trustworthy quantitative predictions; extending the framework to nonlocal electrodynamics would either confirm or correct the design rule that pure Nb gives ultra-low attenuation approaching 700 GHz.","For two-gap superconductors like MgB2, the single-gap assumption may need generalization; the smaller π-band gap (~900 GHz) is likely the relevant loss scale, so the framework's predictions would need adaptation rather than direct application.","The TLS formula's dependence on filling factor and loss tangent suggests that surface-oxide engineering, already pursued for cavities, becomes directly relevant to millimeter-wave interconnect loss at millikelvin temperatures."],"forward_implications":["For f ≳ 0.5Δ/h, low attenuation requires clean material with ℓ ≳ ξ0; high-purity Nb should sustain very low loss up to its gap frequency near 720 GHz.","Below T/Tc ≈ 0.1–0.2, native-oxide two-level systems, not quasiparticles, set the loss floor; above that, quasiparticle dissipation dominates.","A drive-dependent peak in attenuation at f ≈ Δ/h appears at accessible frequencies (~160 GHz in a WR5 Ta guide, ~730 GHz in a WR1 Nb3Sn guide), giving a microwave-domain signature of the Higgs mode distinct from third-harmonic generation.","The commonly used low-frequency relation σ2 ≈ 1/(μ0ωλ²) and the dirty-limit Mattis–Bardeen formulas are insufficient near the gap frequency; the general expression must be used there.","A simple exponential scaling lets measured attenuation at one temperature be extrapolated to other temperatures once the gap ratio is known."],"fun_headline_variants":["Microscopic theory finds Higgs-mode peak in waveguide loss","Tiny Higgs signature seen in superconducting waveguide loss","Clean superconductors minimize waveguide loss; Higgs peak appears","Attenuation framework exposes Higgs mode in superconducting waveguides","Higgs mode leaves its mark on superconducting waveguide loss"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The calculation assumes the superconductor responds locally to the electromagnetic field (penetration depth much longer than coherence length), an assumption the paper acknowledges breaks down for clean niobium, where the predicted numbers are therefore only qualitative.","fun_headline_variants_meta":{"raw":{"variants":["Microscopic theory finds Higgs-mode peak in waveguide loss","Tiny Higgs signature seen in superconducting waveguide loss","Clean superconductors minimize waveguide loss; Higgs peak appears","Attenuation framework exposes Higgs mode in superconducting waveguides","Higgs mode leaves its mark on superconducting waveguide loss"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000602,"raw_usage":{"total_tokens":2714,"prompt_tokens":879,"completion_tokens":1835,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":623,"completion_tokens_details":{"reasoning_tokens":1758}},"tokens_in":623,"tokens_out":1835,"duration_ms":11342,"temperature":1.0,"reasoning_tokens":1758,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T22:20:44.692289+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the attenuation of a Ta WR5 rectangular waveguide near 160 GHz at T/Tc ≈ 0.2 with an applied ac field of order B0/Bc ≈ 0.02: the theory predicts a drive-amplitude-dependent peak in attenuation at f ≈ Δ/h; a clean measurement showing no such peak would falsify the Higgs-mode claim.","supporting_citations":[],"review_version":1}