{"id":"e8b78e40-81ac-4265-8a50-753d643ccfef","arxiv_id":"2509.07669","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Tantalum superconducting resonators show a persistent, non-equilibrium quasiparticle density at millikelvin temperatures that is about three times lower than in niobium nitride at the same normalized temperature.","lead":"Researchers measured microwave losses in thin-film tantalum resonators at millikelvin temperatures and found losses higher than thermal theory predicts, which they attribute to stray quasiparticles. The result suggests tantalum suffers about a third of the quasiparticle density of niobium nitride at comparable operating temperatures, which matters for choosing materials for future quantum computers.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Residual-loss subtraction is not yet falsifiable: Qi fit uncertainty and TLS model parameters are unreported, and δ_other in Eq. 6 is silently set to zero before Eqs. 7-8 are used to infer persistent quasiparticles.","rationale":"I read the manuscript in good faith and credit the substantial experimental apparatus: TEM/XRD/EELS structural work, single-photon microwave spectroscopy, and the careful statement that the device is overcoupled below ~0.5 K are all useful contributions. The central argument, however, depends on a subtraction: measured internal loss minus a two-level-system model equals quasiparticle loss. The reader's weakest_assumption identifies exactly this dependence, and I agree with that diagnosis. My stress-test sharpens it in three ways. First, the paper's own discussion of the coupling-limited regime raises the question of whether Qi is even well-determined at the millikelvin temperatures where the non-equilibrium quasiparticle signal is claimed; without fit uncertainties, a small residual could be noise. Second, the TLS parameters are not reported for this device, so the subtraction is not reproducible. Third, δ_other is written in Eq. 6 but never quantified, and the NbN benchmark is attributed to a theoretical paper with a nonphysical normalized temperature. None of these are accusations of misconduct; they are requests for the data and error propagation needed to make the claim falsifiable. If the suggested reanalysis shows that the residual survives with meaningful uncertainties, the paper's central conclusion would be much stronger. If it does not, the conclusion should be softened to a materials comparison of loss without assigning the residual to quasiparticles. Since the reader already recommended CONDITIONAL and my concern is the same load-bearing point, I keep the verdict as UNCHANGED rather than escalating or downgrading it.","tokens_in":10605,"tokens_out":8431,"duration_ms":82300,"concrete_test":"Reanalyze the raw complex transmission data for the 40 nm resonator: fit each resonance with a full complex-circle model that returns a covariance matrix for Qi, Qc, and fr at every temperature, and fit the TLS parameters (1/Q0_TLS, nc, β) on this device from the power sweep at 77 mK. Then recompute δqp,measured and its propagated uncertainty. If the residual below 0.5 K is within 2σ of zero, or if the inferred n_qp changes by more than a factor of two when TLS parameters are varied within their fit range, the claimed persistent quasiparticle signal is not established.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim requires that the entire gap between Qi,measured and Qi,theory (Fig. 6a) is a steady-state quasiparticle population. This is obtained through Eq. 7: δqp,measured = 1/Qi,measured − 1/QTLS,derived, with δ_other from Eq. 6 set to zero without discussion. Three quantities are needed but not reported. First, fit uncertainties on Qi are absent, especially below ~0.5 K where the paper itself states Ql is nearly constant and consistent with Qc, i.e. the coupling-limited regime, so Qi is only weakly constrained by the resonance lineshape. Second, the TLS parameters (1/Q0_TLS, nc, β) used in Eq. 9 are imported from prior work [9] or otherwise not stated, and no values or error bars are given for the present 40 nm film. Third, no bound is placed on δ_other (magnetic, radiation, or interface losses). Without these, the residual δqp could be a fitting artifact rather than a real quasiparticle population. Separately, the material benchmark in the abstract and conclusions cites [24] for a 'measured' NbN quasiparticle density, but [24] is an analytical theory paper, and the text's 'T/Tc = 200' is nonphysical, so the quantitative one-third comparison is not currently traceable to data.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports fabrication and cryogenic microwave characterization of α-tantalum coplanar waveguide resonators on silicon with a niobium seed layer, including TEM, XRD, and EELS structural analysis. The authors measure the internal quality factor Qi as a function of temperature in the single-photon regime, compare it with a theoretical model combining two-level-system (TLS) loss and Mattis–Bardeen quasiparticle loss, and attribute the residual loss to persistent non-equilibrium quasiparticles at millikelvin temperatures. They extract a quasiparticle density from the residual loss and claim that α-Ta has a quasiparticle density roughly one-third that of NbN at equivalent normalized temperatures, based on a comparison with prior work.","tokens_in":10904,"tokens_out":5185,"duration_ms":45283,"significance":"If the residual-loss attribution is validated, the paper would provide a useful material-level study of quasiparticle loss in α-Ta resonators and a practical comparison platform for superconducting quantum circuits. The structural characterization is careful, the resonator quality factors are competitive, and the temperature- and power-dependent data set is valuable. The paper's main quantitative claims, however, depend on several unreported parameters and on a subtraction procedure that is not currently falsifiable. The material benchmarking claim against NbN is not traceable to the cited source. These issues are central to the abstract and conclusions, so the manuscript needs substantial revision before the claims can be accepted.","major_comments":[{"comment":"The central extraction of nqp,measured is not reproducible as reported. The kinetic inductance fraction α appears in Eq. (5) and is carried through Eq. (8), but no value or measurement of α is given anywhere in the manuscript. Likewise, the TLS parameters 1/Q0_TLS, nc, and β that determine QTLS,derived in Eq. (7) are said to be 'obtained [9]' but no values or uncertainties are reported for the 40 nm device. Since nqp,measured scales as 1/α and depends directly on the subtracted TLS loss, the claimed absolute quasiparticle density and the factor-of-three comparison with NbN cannot be checked without these inputs.","section":"§2.2, Eqs. (5) and (7)–(9)"},{"comment":"The residual-loss attribution silently sets δ_other to zero. Equation (6) defines total loss as δTLS + δqp + δ_other, but Eq. (7) equates δqp,measured directly to 1/Qi,measured − 1/QTLS,derived. No bound is placed on magnetic, radiation, interface, or TLS-model-mismatch losses, and no uncertainty is reported for Qi. The paper itself notes (Fig. 4(e)) that for T ≲ 0.5 K the system is coupling limited and Qi is only weakly constrained by the resonance lineshape; in that regime the inferred residual loss is particularly sensitive to fitting systematics. Without uncertainty propagation or an explicit upper bound on δ_other, the conclusion that the residual is a persistent quasiparticle population is not yet falsifiable.","section":"§2.2, Eqs. (6)–(8)"},{"comment":"The claimed agreement between experiment and Mattis–Bardeen theory is partly circular. The 'measured' nqp values are generated by inverting Eq. (8), which is an algebraic rearrangement of the same loss formula (Eq. 5) used to produce the theoretical curves, and the conductivity components come from Eqs. (1)–(2) with the same nqp–T relation. Therefore the red circles and blue lines in Figs. 6(c,d) are not independent measurements of the same relation; they test internal consistency of the inversion rather than validating the electrodynamic model. An independent validation would require comparing the measured complex conductivity or Qi to a model with independently determined nqp.","section":"Fig. 6(c,d) and Eqs. (8)–(12)"},{"comment":"The quantitative benchmark against NbN is not traceable. The text states that Ta has nqp = 0.3×10^3 µm^-3 at T/Tc = 200 and that NbN has 1×10^3 µm^-3, citing [24]; however, [24] is an analytical theory paper (Fischer and Catelani) and does not report a measured NbN quasiparticle density. The value T/Tc = 200 is also physically impossible, suggesting a typographical error, and no uncertainty or temperature is given for either number. This comparison should either be removed or replaced with a direct, referenced experimental comparison.","section":"Abstract and §2.2 (final paragraph)"}],"minor_comments":[{"comment":"The stated measurement temperature range is inconsistent: the Introduction says 0.77–1 K, while the experimental section and Fig. 4 use 77 mK to 1 K; please clarify whether 0.77 K is a typo or the actual base temperature used for the analysis.","section":"Introduction and §2"},{"comment":"The phrase 'beak a significant number' should read 'break a significant number.'","section":"§2.1, page 7"},{"comment":"Equation (4) writes '1.76×K_B×Tc'; this should use the conventional notation 1.76 k_B T_c for consistency with the rest of the text.","section":"Eq. (4)"},{"comment":"The sentence 'the values of 1/Q0_TLS, nc, and β are obtained [9]' should state explicitly that the values are taken from the prior work and should list them in a table or appendix, since they are central to the subtraction.","section":"§2.2"},{"comment":"The caption and text do not report the units of nqp on the axes of Fig. 6 or the calibration used to define ⟨nph⟩ ∼ 1; please add this information so that the quantitative claims can be interpreted.","section":"Figure 6 and §2.2"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is within the journal's scope, but the central claims currently exceed the evidence presented. In particular, the NbN benchmark appears to attribute a measured quasiparticle density to a theory paper, and the residual-loss extraction omits essential parameters and uncertainties. I recommend asking the authors to provide the missing α and TLS parameters, include error bars and a δ_other bound, and re-frame the Fig. 6(c,d) comparison as a consistency check rather than an independent validation. If these issues cannot be addressed with the existing data, the quantitative claims should be substantially weakened."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: the paper's raw resonator data and film characterization are solid, and the Ta-vs-NbN normalized-temperature comparison is a sensible thing to do, but the central claim—persistent non-equilibrium quasiparticles at millikelvin temperatures and a threefold lower n_qp in Ta—is not yet supported because the loss subtraction hides all the parameters that matter.\n\nWhat's genuinely useful here: the STEM/TEM and XRD characterization of the α-Ta/Nb/Si stack is careful, and the temperature sweeps in the single-photon regime look like real work. The choice to compare materials at matched T/Tc is a legitimate way to make the benchmark fair, and the raw Qi data appear plausible.\n\nThe soft spots are not cosmetic. The paper's Eq. 6 defines δi = δ_TLS + δ_qp + δ_other, but Eq. 7 silently sets δ_other = 0 when computing δ_qp,measured. That is a load-bearing assumption with no justification. The TLS model parameters (1/Q0_TLS, nc, β) are taken from previous work [9] and not reported here; α, the kinetic inductance fraction used in Eqs. 5, 8, 11, and 12, is never given. Without α, n_qp cannot actually be computed from the measured Qi, so the numbers in Fig. 6 and the abstract are not reproducible. There are also no error bars on Qi, and the text itself says that below ~0.5 K Ql ≈ Qc, i.e., the resonator is coupling-limited, which means Qi is weakly constrained precisely in the temperature range where the non-equilibrium claim is strongest.\n\nTwo clear errors that need to be flagged: 'T/Tc = 200' is nonphysical (at 77 mK and Tc ≈ 4.06 K, T/Tc ≈ 0.02), so either the value or the notation is wrong; and reference [24] is an analytical theory paper, not a measurement, so the 'measured value of 1×10^3 (µm^-3) for NbN' is not traceable. The authors likely meant their own prior NbN work [8].\n\nThe comparison between Ta and NbN might well survive a reanalysis—tantalum resonators do tend to have lower loss—but as written, the central quantitative claim is not yet established. This is fixable, not fatal. The paper would benefit from reporting all fit parameters, α, and error bars, and ideally from an independent cross-check of n_qp.\n\nVerdict: send to peer review, because the experimental work is worth refereeing, but expect major revision and re-review. I would not cite the one-third number until it is properly supported.","headline":"Solid raw data on Ta resonators, but the central quasiparticle-density claim lacks the parameters and error bars needed to support it.","tokens_in":11443,"tokens_out":3261,"would_cite":false,"duration_ms":27201,"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":"α-tantalum superconducting resonators show measurable non-equilibrium quasiparticle loss at millikelvin temperatures, and the quasiparticle density is roughly one-third that of NbN at equivalent normalized temperatures.","keywords":["superconducting microwave resonators","alpha-tantalum","quasiparticles","internal quality factor","single-photon regime","Mattis-Bardeen theory","two-level systems","cryogenic microwave spectroscopy"],"falsifier":"Perform the same microwave loss measurement on a second resonator from the same film while deliberately varying the quasiparticle population with a known pair-breaking source, such as a small heater or an above-gap photon pulse; if the extracted $n_\\mathrm{qp}$ does not track the injected quasiparticle rate, the attribution of residual loss to non-equilibrium quasiparticles fails. Alternatively, characterise the two-level-system loss on the identical 40 nm film and geometry; if the TLS fit does not account for the full zero-quasiparticle loss, the reported densities are overestimated.","tokens_in":10403,"feed_emoji":"❄️","tokens_out":6167,"duration_ms":56194,"temperature":0.7,"pith_summary":"The paper reports single-photon microwave measurements of α-tantalum coplanar waveguide resonators on silicon, at temperatures from 77 mK to 1 K. It aims to establish that non-equilibrium quasiparticles persist at millikelvin temperatures and that their density can be read out from the suppression of the internal quality factor relative to theoretical expectation. Using Mattis–Bardeen complex-conductivity theory together with a two-level-system loss model, the authors extract quasiparticle densities and find that α-Ta carries about one-third the quasiparticle density of NbN at equivalent normalized temperatures. Since non-equilibrium quasiparticles are a leading coherence limiter in superconducting qubits, this materials comparison at thermodynamically matched operating points matters directly for scalable quantum circuits.","feed_headline":"Tantalum circuits keep quasiparticle loss to a third of NbN","feed_subtitle":"Single-photon microwave readouts show persistent quasiparticles at millikelvin and put α-Ta ahead of NbN at matched temperature","key_machinery":"The carrying mechanism is on-chip microwave spectroscopy of quarter-wavelength coplanar waveguide resonators operated in the single-photon regime. The temperature- and power-dependent internal quality factor $Q_i$ is decomposed into two-level-system loss, quasiparticle loss, and residual loss, using the TLS model of Eq. (9) and the quasiparticle loss of Eq. (5). Quasiparticle dissipation is modelled through the Mattis–Bardeen complex conductivity $\\sigma = \\sigma_1 - j\\sigma_2$, which converts low-temperature loss into a quasiparticle density $n_\\mathrm{qp} \\propto \\sigma_1 \\propto e^{-\\Delta/k_B T}$. The material benchmark compares different superconductors at equal $T/T_c$, which is what allows the α-Ta versus NbN comparison to be stated independently of each film's critical temperature.","core_discovery":"In the paper's own terms, the central discovery is that high-Q α-Ta microwave resonators host a finite quasiparticle population even at millikelvin temperatures. This shows up as a persistent gap between the measured internal quality factor and the value predicted from thermal quasiparticles plus two-level-system loss, across the whole 0.77–1 K range. When the residual loss is converted to a quasiparticle density through the Mattis–Bardeen relations, the density stays finite at low temperature instead of falling to zero as the thermal formula predicts. The material benchmark is placed on a thermodynamic footing by comparing at the same fractional distance from each film's transition temperature, $T/T_c$: α-Ta reaches a quasiparticle density around $0.3\\times 10^3\\ \\mu\\mathrm{m}^{-3}$, about one-third of the $1\\times 10^3\\ \\mu\\mathrm{m}^{-3}$ reported for NbN, and the normalized conductivity traces agree with Mattis–Bardeen theory over the measured range.","pith_inferences":["A direct test at the same absolute temperature (rather than the same $T/T_c$) could show whether Ta's advantage persists at typical qubit operating points near 10–20 mK, where both densities are very low but the Ta benefit may change or vanish.","If non-equilibrium quasiparticles are indeed the residual loss source, adding quasiparticle traps or a gap-engineered layer to the Ta film should raise $Q_i$ toward the TLS-limited value; that is an implied design route not tested in the paper.","The method could be extended to other low-loss films such as aluminium or niobium and connected to qubit coherence measurements, since $Q_i$ suppression and qubit $T_1$ degradation share the same quasiparticle mechanism.","The normalized comparison suggests that a material's critical temperature alone is not the decisive figure of merit; the density of non-equilibrium quasiparticles at the fractional operating temperature matters, which reframes how new superconducting materials are screened."],"forward_implications":["If the reported density holds, α-Ta resonators should exhibit lower microwave dissipation than NbN at the same $T/T_c$, making them preferable for qubit readout resonators and kinetic-inductance detectors.","The persistent quasiparticle floor sets a limit on $Q_i$ at millikelvin temperatures, so further coherence gains will require quasiparticle trapping or mitigation rather than only surface preparation.","The normalized-temperature protocol gives a quantitative route for comparing quasiparticle densities across any superconducting material, not just Ta and NbN.","The extraction procedure of Eqs. (7)–(8) turns a standard resonator loss measurement into an on-chip quasiparticle sensor in the single-photon regime.","Because lower quasiparticle density directly reduces dissipation and charge noise, the comparison supports choosing α-Ta over NbN for coherence-limited circuit architectures."],"supporting_citations":[{"why":"Supplies the resonator design, fabrication route, and the two-level-system fit parameters used to separate TLS loss from quasiparticle loss.","marker":"[9]"},{"why":"Provides the NbN CPW resonator measurements used as the material benchmark for the threefold quasiparticle-density comparison.","marker":"[8]"},{"why":"Gives the Mattis–Bardeen theory used to compute the complex conductivity of the Ta film.","marker":"[38]"},{"why":"Provides the surface-impedance and resonator-loss formalism used to convert conductivity into $Q_i$ and quasiparticle density.","marker":"[40]"},{"why":"Supplies the thermodynamic equivalence relation for temperature change and external pair breaking used in the Mattis–Bardeen expressions.","marker":"[23]"},{"why":"Provides the two-level-system loss model fitted in Eq. (9).","marker":"[37]"},{"why":"Supplies the analytical framework for non-equilibrium quasiparticle distributions used in interpreting the low-temperature excess.","marker":"[24]"}],"fun_headline_variants":["α-Ta cuts quasiparticle density to one-third of NbN at same T/Tc","Tantalum resonators: quasiparticle density a third of NbN at matched T/Tc","Single-photon microwaves reveal α-Ta's quasiparticle population is 1/3 of NbN","Persistent quasiparticles in α-Ta, but 3x fewer than NbN at same T/Tc","α-Ta on-chip sensing: quasiparticle loss at millikelvin, but 1/3 of NbN"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that every bit of loss left over after accounting for temperature-activated quasiparticles and the two-level-system model comes from non-equilibrium quasiparticles, and that the two-level-system model taken from an earlier tantalum device still describes this 40 nm film with no other loss channels contributing.","fun_headline_variants_meta":{"raw":{"variants":["α-Ta cuts quasiparticle density to one-third of NbN at same T/Tc","Tantalum resonators: quasiparticle density a third of NbN at matched T/Tc","Single-photon microwaves reveal α-Ta's quasiparticle population is 1/3 of NbN","Persistent quasiparticles in α-Ta, but 3x fewer than NbN at same T/Tc","α-Ta on-chip sensing: quasiparticle loss at millikelvin, but 1/3 of NbN"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001055,"raw_usage":{"total_tokens":4432,"prompt_tokens":954,"completion_tokens":3478,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":570,"completion_tokens_details":{"reasoning_tokens":3344}},"tokens_in":570,"tokens_out":3478,"duration_ms":23839,"temperature":1.0,"reasoning_tokens":3344,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T16:11:16.834820+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Perform the same microwave loss measurement on a second resonator from the same film while deliberately varying the quasiparticle population with a known pair-breaking source, such as a small heater or an above-gap photon pulse; if the extracted $n_\\mathrm{qp}$ does not track the injected quasiparticle rate, the attribution of residual loss to non-equilibrium quasiparticles fails. Alternatively, characterise the two-level-system loss on the identical 40 nm film and geometry; if the TLS fit does not account for the full zero-quasiparticle loss, the reported densities are overestimated.","supporting_citations":[{"cited_title":"Engineering high-q superconducting tantalum microwave coplanar waveguide resonators for compact coherent quantum circuit,","cited_arxiv_id":null,"evidence_quote":"Supplies the resonator design, fabrication route, and the two-level-system fit parameters used to separate TLS loss from quasiparticle loss."},{"cited_title":"Characterizing niobium nitride-based superconducting coplanar waveguide resonators for microwave hybrid circuit quantum electrodynamics,","cited_arxiv_id":null,"evidence_quote":"Provides the NbN CPW resonator measurements used as the material benchmark for the threefold quasiparticle-density comparison."},{"cited_title":"Theory of the anomalous skin effect in normal and superconducting metals,","cited_arxiv_id":null,"evidence_quote":"Gives the Mattis–Bardeen theory used to compute the complex conductivity of the Ta film."},{"cited_title":"Gao,The physics of superconducting microwave resonators","cited_arxiv_id":null,"evidence_quote":"Provides the surface-impedance and resonator-loss formalism used to convert conductivity into $Q_i$ and quasiparticle density."},{"cited_title":"Equivalence of the effects on the complex conductivity of superconductor due to temperature change and external pair breaking,","cited_arxiv_id":null,"evidence_quote":"Supplies the thermodynamic equivalence relation for temperature change and external pair breaking used in the Mattis–Bardeen expressions."},{"cited_title":"Two level system loss in superconducting microwave resonators,","cited_arxiv_id":null,"evidence_quote":"Provides the two-level-system loss model fitted in Eq. (9)."},{"cited_title":"Nonequilibrium quasiparticle distribution in superconducting resonators: An analytical approach,","cited_arxiv_id":null,"evidence_quote":"Supplies the analytical framework for non-equilibrium quasiparticle distributions used in interpreting the low-temperature excess."}],"review_version":2}