{"id":"c8ccf2cc-2777-4774-9611-0b86ca88f129","arxiv_id":"2608.07127","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Doubly clamped NbTiN nanostrings show tensile stress up to 0.81 GPa, Young's modulus of 181 GPa, and intrinsic mechanical quality factors up to 850.","lead":"Researchers fabricated tiny, tightly stretched NbTiN wires and measured their mechanical properties from room temperature down to 13 K. The material carries high internal tension and a moderate mechanical quality factor, making it a candidate for quantum electromechanical devices that need superconductors robust to magnetic fields.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Length-averaged stress fit, not material stress, underlies the key 0.81 GPa and >2×-vs-Al claims; Appendix B shows strong length dependence, so the central advantage is not established.","rationale":"The reader identified the same weakest assumption, and I agree it is the most load-bearing issue. I considered whether a more serious concern exists in the α extraction (Eq. 4) or the Q_intr model, but the former is plausibly correct for uniaxial strings and the latter is secondary and likely conservative. The paper is honest and includes the needed appendix; the problem is that the headline and Table I do not carry the caveat. Therefore the verdict remains CONDITIONAL, not REJECT.","tokens_in":14450,"tokens_out":11396,"duration_ms":108471,"concrete_test":"Using the open Zenodo processed data, repeat the stress extraction at each temperature separately for the 70 µm and 90 µm strings (for which higher harmonics are available) with Eq. (B1), and recompute the Table I comparison using σ_90µm(13 K) in place of the length-averaged 0.81 GPa. If σ_90µm(13 K) is below ≈0.7 GPa, the 'more than two times higher tensile stress than Al' claim fails for long-string dissipation-dilution geometries; if σ_90µm(13 K) remains above 0.7 GPa, the headline is robust and only the abstract's wording needs to be clarified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that NbTiN is a high-tensile-stress platform rests on σ=(0.56±0.03) GPa at RT and σ=0.81 GPa at 13 K obtained by fitting Eq. (2) to the fundamental frequency as a function of 1/l. This fit assumes a single, length-independent stress. Appendix B explicitly contradicts that premise: per-string fits with Eq. (B1) give σ_90µm=(0.310±0.001) GPa and σ_70µm=(0.377±0.004) GPa at RT, with a single-harmonic estimate of 0.499 GPa for 20 µm strings. The quoted values are thus fit averages, not the stress in any measured geometry. Since dissipation-dilution devices typically use long strings, and Al has σ=0.35 GPa, the paper's claim of 'more than two times higher tensile stress' than Al is not supported for the relevant long-string regime if the same length dependence persists at 13 K. The abstract and Table I present 0.81 GPa without this caveat, although the main text does flag the issue. This is a presentation/correctness concern about the headline metric, not a hidden internal inconsistency. The same fixed-σ assumption biases the Q_intr extraction from Eq. (5), since λ(l) is mis-modeled when σ varies with l (here likely making Q_intr an underestimate), but the stress comparison is the load-bearing issue.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper reports fabrication and optical characterization of doubly clamped Nb0.7Ti0.3N nanostring resonators from room temperature down to 13 K. From the length dependence of the fundamental resonance frequency the authors extract a tensile stress of (0.56±0.03) GPa at 293 K and 0.81 GPa at 13 K; from higher-harmonic frequency differences they obtain a Young's modulus of (183±11) GPa at room temperature, with a temperature average of 181 GPa; and from the length dependence of the mechanical quality factor they extract an intrinsic quality factor with a maximum of (0.85±0.03)×10^3 near 140 K and 0.5×10^3 at 13 K. They also infer a thermal expansion coefficient for NbTiN from the temperature dependence of the stress. The central claim is that NbTiN is a promising high-tensile-stress, superconducting, magnetic-field-tolerant alternative to aluminum for cavity electro- and nanomechanics. The analysis uses standard thin-beam and dissipation-dilution models, and the high-stress approximation is cross-checked against the full model; processed data are deposited on Zenodo. The main caveat, acknowledged in the main text but not in the abstract or Table I, is that the quoted stress values come from a fit that assumes length-independent stress, while Appendix B shows a pronounced length dependence.","tokens_in":14719,"tokens_out":8025,"duration_ms":69255,"significance":"If the headline values were material properties, this would be a useful contribution: NbTiN combines a high tensile stress and a Young's modulus comparable to other nitrides with superconductivity and demonstrated magnetic-field robustness of NbTiN microwave resonators. The paper has clear strengths: standard data reduction, an explicit cross-check of the high-stress approximation (0.56 vs 0.55 GPa), a stress-independent Young's modulus extraction via Eq. (3), and open processed data. However, the central quantitative advantage over aluminum rests on a length-averaged stress parameter rather than a geometry-specific material stress. For long strings, which are the geometry relevant for dissipation dilution, the room-temperature per-length stresses in Appendix B are near or below the cited aluminum value of 0.35 GPa. The intrinsic-quality-factor extraction is also affected by the same length-dependence issue. These concerns do not invalidate the fabrication and characterization effort, but they require revision of the headline claims and of the quantitative comparison with aluminum.","major_comments":[{"comment":"The headline stress values, σ=(0.56±0.03) GPa at room temperature and σ=0.81 GPa at 13 K, are obtained by fitting Eq. (2) to Ω_m^(1)(l) under the assumption of a single length-independent tensile stress. Appendix B shows that this premise is violated for the measured length range: at room temperature the per-string fits with Eq. (B1) give σ_90µm=(0.310±0.001) GPa and σ_70µm=(0.377±0.004) GPa, and even the single-harmonic estimate for the 20 µm string is 0.499 GPa. I agree with the authors' own caveat in the main text that the quoted values are averages over lengths, but that caveat does not appear in the abstract or in Table I, where 0.81 GPa is presented without qualification. Because dissipation-diluted devices typically use long strings, and because the cited Al stress is 0.35 GPa, the claim of a 'more than two times higher tensile stress' than Al is not established for the geometry that matters. The authors should report per-length stresses at each temperature (at least for the 70 and 90 µm strings) and rephrase the 'up to 0.81 GPa' and 'more than two times' claims accordingly, or explicitly show that the length dependence does not change the comparison at 13 K.","section":"§2, Fig. 2(b), Appendix B"},{"comment":"The intrinsic quality factor is extracted by fitting Q_m(l) with Eq. (5), using λ=(h/l)√(E/(12σ)) and a single stress value at each temperature. If σ is length dependent as shown in Appendix B, this fit is mis-specified: at long l the actual stress is lower than the average, so the actual λ is larger than the assumed λ, and the fitted Q_intr is biased. The direction of the bias is likely to make the reported Q_intr an underestimate, but the point is that the reported values—including the maximum (0.85±0.03)×10^3 at 140 K and the 13 K value of 0.5×10^3 in Table I—are not robust to this modeling choice. The fit should be repeated with per-length stresses, or a sensitivity analysis should be provided.","section":"§4, Eq. (5)"},{"comment":"Table I presents the 'this work' row as σ=0.81 GPa, E=181 GPa, Q_intr=0.5×10^3 at 142 nm thickness and 13 K without identifying that σ is a length-averaged fit value. The subsequent conclusion that NbTiN 'clearly outperform[s] aluminum' because of 'more than two times higher tensile stress' therefore overstates what the data establish. At minimum, the table should identify the stress as length-averaged and give the corresponding long-string per-length value at 13 K, and the conclusion should compare NbTiN to Al using the geometry-specific values relevant to the intended device.","section":"§6, Table I and Conclusion"}],"minor_comments":[{"comment":"There are two typos in the opening paragraphs: 'is has supported' should be 'it has supported', and 'mircrowave transmission lines' should be 'microwave transmission lines'.","section":"Introduction"},{"comment":"The caption reads 'An decrease of the loss rate'; this should be 'A decrease of the loss rate'.","section":"Fig. 4 caption"},{"comment":"The Nb70Ti30N50 row appears to be missing the thickness and temperature entries; please format the table so that every row has entries for σ, E, Q_intr, h, and T.","section":"Table I"},{"comment":"The 20 µm stress is a single-harmonic estimate and the text already notes that it likely underestimates the true value; labeling it as a lower bound in Fig. B1(b) would make the discontinuity between 50 and 70 µm less likely to be misread as a physical effect.","section":"Appendix B, Fig. B1(b)"},{"comment":"The Young's modulus determination uses the nominal design length, but the uncertainty in l from lithography is not propagated into E or σ; a brief statement on this uncertainty would improve the reproducibility of the quoted values.","section":"Eq. (3)"},{"comment":"The thermal expansion coefficient α_NbTiN is derived from Δσ/EΔT using the length-averaged stress values; the extraction should carry the same caveat as the stress values, since a length-dependent stress contribution to Δσ is not excluded.","section":"§3, Eq. (4)"}],"recommendation":"major_revision","confidential_remarks":"This is a materials characterization study that fits an applied nanomechanics or applied physics journal. The main technical issue is not hidden inconsistency but the presentation of a length-averaged fit parameter as a material property in the abstract and Table I. I have no concerns about citation practice or novelty disclosure; the revision should focus on the stress-length caveat and its consequences for the intrinsic-Q extraction."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First, what you should know: this is the first real mechanical characterization of NbTiN in a clamped nanostring geometry, and it is done carefully. The paper reports a Young's modulus of 181 GPa, intrinsic Q around 850, and a room-temperature tensile stress that averages 0.56 GPa; the 13 K value is 0.81 GPa. That last number is the headline, and it is where the abstract overreaches.\n\nThe fabrication and measurement are solid. They use standard beam equations, cross-check the high-stress approximation against the full model, give error bars, and deposit processed data on Zenodo. They cite the relevant length-dependent stress literature (Bückle et al.) and are transparent in the main text that the 0.56 GPa/0.81 GPa numbers are averages over string lengths, not the stress of any measured string. Appendix B shows per-string fits: 90 µm strings give 0.31 GPa and 70 µm strings 0.38 GPa at room temperature. Shorter strings likely carry more stress, but only the fundamental mode was measured for them, so those estimates are crude.\n\nThis creates a real problem for the abstract and Table I, which present 0.81 GPa as the material's tensile stress and then claim it is more than two times higher than aluminum's 0.35 GPa. For dissipation-dilution devices, which usually use long strings, the relevant comparison is the long-string stress, and there the advantage over Al shrinks to near parity at room temperature. We don't know the 13 K values for long strings; they may be higher, but the paper doesn't show it. That should be fixed in revision—either report length-resolved stresses in the abstract and table, or clearly state that the quoted value is a length-averaged fit parameter.\n\nThe other soft spot is minor: the thermal expansion coefficient extraction assumes all stress change with temperature comes from CTE mismatch with a constant Young's modulus. That's a stated assumption, not verified. The Q_intr extraction also uses a single length-independent stress, which will bias the strain parameter and hence Q_intr a bit; not a big deal for the conclusions.\n\nWho should read this: anyone working on superconducting cavity electromechanics looking for alternatives to aluminum. The paper gives useful numbers and an honest discussion of limitations. It deserves a serious referee. My recommendation: send it to review, but ask the authors to present the tensile stress values length-resolved or at least caveat the headline number in the abstract.","headline":"First careful mechanical characterization of NbTiN nanostrings, but the headline stress is a length-averaged fit, so the >2x-over-Al claim needs qualification.","tokens_in":15332,"tokens_out":2717,"would_cite":true,"duration_ms":24357,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["07.10.Cm","62.20.Dc","85.25.-j"],"model":"deepseek-v4-flash","headline":"Niobium titanium nitride nanostrings carry 0.81 GPa of tensile stress and could replace aluminum in cavity electromechanics.","keywords":["NbTiN","nanomechanical resonators","tensile stress","dissipation dilution","superconducting circuit integration","cavity electromechanics","Young's modulus","mechanical quality factor"],"falsifier":"Measure the tensile stress of 90-µm-long NbTiN strings at room and low temperature using multiple harmonics, then build equally long NbTiN and aluminum strings with identical geometry and compare their dissipation-diluted quality factors; if the NbTiN advantage does not follow from the per-length stress, the central material comparison collapses.","tokens_in":14227,"feed_emoji":"🔩","tokens_out":6437,"duration_ms":51676,"temperature":0.7,"pith_summary":"This paper claims that niobium titanium nitride (NbTiN) can serve as a high-tensile-stress, superconducting material for nanomechanical resonators, and backs the claim with measurements of doubly clamped NbTiN nanostrings from room temperature down to 13 K. The authors report a tensile stress that reaches 0.81 GPa at low temperature, a Young's modulus of about 181 GPa, and an intrinsic mechanical quality factor up to roughly 850. If correct, this gives cavity electromechanics a conductive resonator material that is stiffer and more highly stressed than the usual aluminum, while remaining compatible with superconducting microwave circuits and resilient to magnetic fields. The paper positions NbTiN as an alternative platform for dissipation-diluted resonators whose quality factor exceeds intrinsic material limits.","feed_headline":"Superconducting NbTiN strings hit 0.81 GPa of tensile stress","feed_subtitle":"A magnetic-field-tolerant superconductor with twice aluminum's stress and stiffness could lift cavity electromechanics.","key_machinery":"The load-bearing object is the doubly clamped NbTiN nanostring, a $w=500$ nm wide, $h=142$ nm thick beam with length $l$ between 20 and 90 µm, whose flexural resonance frequencies are read out optically. The argument runs through three standard relations: the high-tensile-stress beam formula $\\Omega_m^{(n)} \\approx (n\\pi/l)\\sqrt{\\sigma/\\rho}$, which converts the $1/l$ dependence of the fundamental frequency into a stress value; the stress-independent formula for Young's modulus from the squared-frequency difference of harmonics $n$ and $m$; and the dissipation-dilution relation $Q_m^{(1)} = Q_{\\rm intr}/(\\pi^2\\lambda^2+2\\lambda)$ with strain parameter $\\lambda=(h/l)\\sqrt{E/12\\sigma}$, which converts the measured length dependence of $Q$ into an intrinsic quality factor. The thermal-expansion coefficient of NbTiN is then derived from the temperature change of stress via the substrate-expansion mismatch formula.","core_discovery":"The central claim is that NbTiN thin films, already used for high-quality superconducting microwave resonators, also have the mechanical properties needed for stressed nanomechanical strings: a room-temperature tensile stress of $(0.56\\pm0.03)$ GPa rising to $0.81$ GPa at 13 K, a Young's modulus of $(183\\pm11)$ GPa at room temperature (181 GPa averaged over temperature), and an intrinsic quality factor that peaks at $(0.85\\pm0.03)\\times10^3$ around 140 K. The tensile stress is extracted from the inverse-length scaling of the fundamental resonance frequency under the high-tensile-stress approximation, the Young's modulus from the frequency differences of higher harmonics, and the intrinsic quality factor from the length dependence of the measured $Q$ through the dissipation-dilution model. The authors argue that these properties make NbTiN a promising platform for cavity electro- and nanomechanics, outperforming aluminum in tensile stress and stiffness while keeping superconductivity and magnetic-field robustness.","pith_inferences":["If the relevant stress for a dissipation-diluted device is the stress of a long string, the per-length fits in Appendix B put room-temperature stress near 0.31 GPa for 90 µm strings rather than 0.56 GPa, which would bring NbTiN close to parity with aluminum in long-string geometry. This is an inference from the paper's own appendix, not the authors' headline claim.","The quoted 0.81 GPa low-temperature stress carries the same caveat; at 13 K the per-length stress for long strings is still below the global fit value, so the factor-of-two advantage over aluminum may not survive in the device geometry that dissipation dilution actually uses.","A direct test of the thermoelastic-damping hypothesis would be to vary film thickness and measure $Q_{\\rm intr}$ against the thermal-expansion coefficient; the paper points to this correlation but does not establish the mechanism.","The same characterization could be repeated on nitrogen-rich insulating NbTiN, which the paper notes can be deposited, to separate bulk, surface, and electronic loss channels."],"forward_implications":["NbTiN strings carry more than twice the tensile stress and more than twice the Young's modulus of aluminum, so dissipation dilution should yield higher mechanical quality factors than comparable aluminum strings.","Because the same sputtered film already gives superconducting microwave resonators with internal quality factors up to $0.2\\times10^6$ at single-photon level and survives fields up to about 130 mT, NbTiN strings can be integrated into cavity electromechanics without a separate conductor.","The measured intrinsic $Q$ of about 850, while below that of Si$_3$N$_4$ or SiC, is high for a conducting film of this thickness, and the paper's scaling argument predicts thinner films improve $Q$ through a larger strain parameter.","The normal-conducting optical measurements leave open the mechanical behavior in the superconducting state; a cavity-electromechanical readout would probe that regime without optical heating."],"supporting_citations":[{"why":"Provides the same-batch NbTiN film's superconducting transition temperature (16.3 K), microwave resonator quality factors, and magnetic-field robustness that anchor the claim of a superconducting, field-tolerant platform.","marker":"Ref. 47"},{"why":"Documents the universal length dependence of tensile stress in nanomechanical strings; the paper uses it to interpret the per-length stress fits in Appendix B and to bound the validity of the global stress value.","marker":"Ref. 53"},{"why":"Supplies the harmonic-frequency-difference method (Eq. 3) used to determine Young's modulus independently of tensile stress.","marker":"Ref. 54"},{"why":"Gives the dissipation-dilution relation (Eq. 5) used to extract the intrinsic quality factor from the measured length dependence.","marker":"Ref. 66"},{"why":"Establishes surface loss as a dominant intrinsic damping mechanism in thin SiN resonators, the basis for attributing NbTiN losses to surface-dominated behavior at this thickness.","marker":"Ref. 25"},{"why":"Provides the thermoelastic damping model invoked to explain the non-monotonic temperature dependence of the intrinsic quality factor.","marker":"Ref. 67"},{"why":"Supplies the aluminum comparison values (tensile stress 0.35 GPa, Young's modulus 69 GPa) used in Table I to argue NbTiN outperforms aluminum.","marker":"Ref. 39"},{"why":"Reports NbTiN membrane resonators used for Casimir force sensing, the prior evidence that NbTiN membranes work mechanically and superconduct at low temperature.","marker":"Ref. 50"}],"fun_headline_variants":["NbTiN strings: superconductive and stress-strong for nanomechanics","0.81 GPa stress in superconducting NbTiN nanomechanical strings","High-stress NbTiN: a promising platform for cavity electromechanics","Superconducting NbTiN resonators reach 0.81 GPa tensile stress"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The headline stress values come from a fit that assumes one tensile stress value for all string lengths, while the paper's own per-length analysis shows stress is lower in the long strings that dissipation-diluted resonators would actually use.","fun_headline_variants_meta":{"raw":{"variants":["NbTiN strings: superconductive and stress-strong for nanomechanics","0.81 GPa stress in superconducting NbTiN nanomechanical strings","High-stress NbTiN: a promising platform for cavity electromechanics","Superconducting NbTiN resonators reach 0.81 GPa tensile stress"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001089,"raw_usage":{"total_tokens":4557,"prompt_tokens":959,"completion_tokens":3598,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":575,"completion_tokens_details":{"reasoning_tokens":3513}},"tokens_in":575,"tokens_out":3598,"duration_ms":21923,"temperature":1.0,"reasoning_tokens":3513,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T14:19:32.066705+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the tensile stress of 90-µm-long NbTiN strings at room and low temperature using multiple harmonics, then build equally long NbTiN and aluminum strings with identical geometry and compare their dissipation-diluted quality factors; if the NbTiN advantage does not follow from the per-length stress, the central material comparison collapses.","supporting_citations":[],"review_version":1}