{"id":"2fff0e65-3bdf-4e07-b551-f670b332df31","arxiv_id":"2411.15048","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":8,"one_line_summary":"Switching current distributions in Nb/Al-AlOx/Nb junctions under 100-110 GHz irradiation show double-peak structures consistent with the strong-driving resonant escape model, indicating plasma frequencies around 140 GHz.","lead":"Scientists measured how niobium Josephson junctions switch when hit with 100-110 GHz radiation, observing a double-peak switching distribution that matches the resonant escape model. The results support the idea that high-frequency phase qubits could run at warmer temperatures, although only classical behavior was measured.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claimed common plasma frequency of 145.5 GHz from the Eq. (8) power-branch fits conflicts with the independent spectroscopic value 138.3 GHz from Fig. 3; this unaddressed discrepancy is the main load-bearing weakness.","rationale":"The reader's weakest assumption concerns the validity of the strong-driving model, Eq. (8), in the thermal regime and the freedom of its fit parameters. I agree that model validity matters, but the most concrete and load-bearing weakness is internal: the two independent determinations of the plasma frequency disagree by about 7 GHz, and the critical currents used in the two analyses differ by about 6%. These are not merely philosophical concerns about fit freedom; they are quantitative inconsistencies inside the manuscript that directly affect the central claim of a common ωp = 145.5 GHz and Q ≈ 90. The strong-driving limit condition itself is satisfied for either plasma-frequency value, so I do not see that as the decisive issue. The qualitative observation of double-peak structures and the broader conclusion of plasma frequencies around 140 GHz remain credible, and the model fits do provide supporting evidence, but the quantitative model claim needs a controlled refit and disclosure of residuals before acceptance. This does not move the verdict from the reader's CONDITIONAL assessment, so I recommend no verdict change, only a concrete verification step.","tokens_in":7794,"tokens_out":9085,"duration_ms":100867,"concrete_test":"Refit the Fig. 4 power-dependence data with Eq. (8) (or its harmonic limit Eq. (10)) with ωp fixed to 138.3 GHz, once with Ic = 242 μA and once with Ic = 258 μA, letting Q and k−1C−2 float for each of the three frequencies; then compare the residual sum of squares with those of the reported 145.5 GHz fits. If the 138.3 GHz fits are not significantly worse, the claimed unique common ωp is unsupported. In addition, report the raw peak-position data and error bars for all fit branches so the comparison is quantitative.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim rests on a common value ωp/2π = 145.5 GHz obtained by solving Eq. (8) in Fig. 4. However, the independent spectroscopic determination in Fig. 3, using Eq. (3), gives ωp/2π = 138.3(7) GHz, a discrepancy of about 7 GHz, and the undriven escape-rate fit in Fig. 2 gives ωp/2π = 0.1(2) THz, i.e., no real constraint. Additionally, the critical current used for the power-branch fits, Ic = 242(1) μA from Fig. 2, differs from the Ic = 258(3) μA used in the spectroscopic fit of Fig. 3. The paper does not discuss either inconsistency. Because the Fig. 4 fits allow Q and k−1C−2 to float separately for each frequency, and the power axis is in arbitrary units, the common ωp = 145.5 GHz may be absorbing model error or the Ic mismatch rather than representing the true plasma frequency. The qualitative double-peak observation survives, but the specific claim of good description by the strong-driving model with common ωp = 145.5 GHz and Q ≈ 90 is not yet secured.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports switching-current measurements of Nb/Al-AlOx/Nb Josephson junctions under millimeter-wave irradiation at 100–110 GHz. At these frequencies the measured switching-current histograms show double-peak structures, which the authors interpret as resonant escape of the Josephson phase from a metastable well. A spectroscopic scan of the resonant-peak position versus frequency is used to extract a plasma frequency fp=138.3(7) GHz, and power-dependent switching-current branches at three frequencies are fitted with the strong-driving model of Eqs. (8)–(10), using a common plasma frequency ωp/2π=145.5 GHz and quality factors Q≈90. The paper concludes that junctions with plasma frequencies around 140 GHz have been fabricated and that phase qubits operating near 100 GHz are feasible.","tokens_in":8093,"tokens_out":8524,"duration_ms":83024,"significance":"The qualitative observation of clear double-peak switching-current structures at millimeter-wave frequencies is a credible and potentially important experimental result, and it extends earlier microwave-frequency resonant-escape studies to a regime relevant for higher-temperature phase qubits. The use of the published strong-driving model (Fistul et al., ref. [6]) is appropriate, and the data set is systematic: three irradiation frequencies, power sweeps, and a frequency scan of the resonant peak. If the quantitative parameter inconsistencies are resolved, the work would provide a useful rapid characterization method for high-plasma-frequency junctions. The paper does not provide code or machine-checked proofs, but the experimental protocol is clearly described. The main weakness is that the central quantitative claim rests on fits with several free parameters and on a common plasma frequency that is not reconciled with the independently extracted spectroscopic value.","major_comments":[{"comment":"The power-branch fits in Fig. 4 yield a common plasma frequency ωp/2π=145.5 GHz, whereas the independent spectroscopic fit in Fig. 3 gives ωp/2π=138.3(7) GHz. This ≈7 GHz (5%) discrepancy is not discussed, yet the abstract and summary claim that the strong-driving model 'well explains' the data and that 'these measurements yield similar results.' Because the common ωp is a central output of the fits and the spectroscopic value is the physical plasma frequency, the authors should either reconcile the difference through a quantitative uncertainty analysis (e.g., fit covariance, systematic shifts from power-dependent peak positions, or model approximations) or soften the quantitative claim. As written, the specific assertion of agreement with Eqs. (8)–(10) is not fully supported.","section":"Sec. IV, Fig. 2 and Fig. 3"},{"comment":"The undriven escape-rate fit gives Ic=242(1) μA (Fig. 2), and this value is used as a fixed parameter for the Fig. 4 fits, while the spectroscopic fit of Eq. (3) gives Ic=258(3) μA (Fig. 3). The 16 μA difference between the two determinations is not mentioned. Since the small-oscillation frequency ω0(I) in Eq. (3) and the normalized shift ⟨δIsw⟩ depend on Ic, using Ic=242 μA may bias the extracted ωp and Q. The authors should justify the choice of one Ic value, include the uncertainty in Ic in the fit propagation, or fit Ic as a constrained parameter.","section":"Sec. IV, Fig. 2 and Fig. 3"},{"comment":"The Fig. 4 fits use a common ωp but allow Q and k^{-1}C^{-2} to be independent for each frequency, and the mm-wave power axis is in arbitrary units. With these free parameters, the visual agreement shown in Fig. 4 does not by itself establish that the common value ωp/2π=145.5 GHz is uniquely determined; model error or the Ic inconsistency could be absorbed by the per-frequency parameters. The reported values k^{-1}C^{-2}=0.82, 0.115, and 0.6 (arbitrary units) differ by almost an order of magnitude across the three frequencies, which is unexplained if k and C are junction properties. Please report residuals, parameter uncertainties, and a sensitivity analysis (e.g., fixing Q to a common value or using the spectroscopic ωp) to demonstrate the robustness of the common-ωp claim.","section":"Sec. IV, Fig. 4"},{"comment":"The text states that the red curves in Fig. 4 are fits to 'solutions of Eq. (8)', but the reported fit parameter k^{-1}C^{-2} appears explicitly only in the harmonic-approximation form Eq. (10), not in the multilevel Eq. (8). This ambiguity is important for reproducibility: if Eq. (10) was actually solved, the text and figure caption should say so; if the full Eq. (8) was used, the definition of C^{-2}, the summation range over n,m, and the truncation of the level sum should be provided.","section":"Sec. IV, Eqs. (8) and (10)"}],"minor_comments":[{"comment":"The red fits in Fig. 4 are shown without residuals, error bars, or a goodness-of-fit measure, so the reader cannot independently assess how well the model reproduces the data.","section":"Sec. IV, Fig. 4"},{"comment":"The double-Lorentzian fitting procedure used to extract the peak positions is not described; please state the fit function, any shared parameters, and the uncertainties on the peak positions, since those uncertainties propagate into the spectroscopic plasma frequency.","section":"Sec. IV, Fig. 3"},{"comment":"Please define all symbols in Eq. (8), including the summation range over n,m and the normalization of the matrix elements f_nm; currently k^{-1}C^{-2} appears only in Eq. (10).","section":"Sec. II, Eq. (8)"},{"comment":"The schematic and text omit the current-divider ratio and the cut-off frequencies of the low-pass filters; stating these values would improve reproducibility of the noise environment.","section":"Sec. III, Fig. 1"},{"comment":"There are minor typographical issues, including 'T op' in the Fig. 2 caption and 'V A100' in Sec. III A, that should be corrected.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The qualitative double-peak observation is solid and well worth publishing, but the quantitative claim of agreement with the strong-driving model needs to be made more robust. The referee report should emphasize the unreconciled plasma-frequency discrepancy (145.5 vs 138.3 GHz), the critical-current mismatch (242 vs 258 μA), and the high number of free parameters combined with an arbitrary power scale. There is no indication of misconduct; the issues are incomplete analysis and overstatement of the quantitative agreement."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe short version: this paper has a genuinely new measurement—switching current distributions in Nb/Al-AlOx/Nb junctions driven at 100-110 GHz—and the double-peak structures are credible. The quantitative claim that the strong-driving model fits the power dependence with a common plasma frequency is less solid than the text suggests, because the fit's ωp and the independently measured spectroscopic value don't agree.\n\nWhat's new and good: it's the first resonant-escape study in this frequency range using high-Jc Nb junctions, aimed at phase qubits around 100 GHz. The measurement setup is carefully described, the statistics are decent, and the qualitative evolution of the double peaks with power at three frequencies matches the Fistul-Wallraff-Ustinov strong-driving model. That model is published and has prior experimental support, so using it is reasonable. The observation of a resonant branch in the W-band is a legitimate step forward.\n\nThe soft spots are quantitative. The spectroscopic fit in Fig. 3 gives ωp/2π = 138.3(7) GHz, while the Fig. 4 fits to Eq. (8) use a common ωp/2π = 145.5 GHz; the discrepancy is about 5% and isn't mentioned. The same fits use Ic = 242 μA from the thermal escape fit, whereas the spectroscopic fit uses Ic = 258 μA. Power is in arbitrary units, and Q and k^{-1}C^{-2} are free per frequency, so the fit has enough freedom to absorb model error. No error bars or fitting degeneracy are discussed. These issues don't invalidate the central qualitative result—junctions with plasma frequencies around 140 GHz do show resonant escape—but they mean the specific values of ωp and Q are not firmly established.\n\nThe paper doesn't provide raw data or analysis code; for an experiment paper that's a minor but reasonable request.\n\nI'd send this to a serious referee. The experiment is new, the phenomenon is real, and the problems are fixable with clarification, error analysis, and a direct discussion of the ωp discrepancy. For my own citing purposes, it's relevant only if I were working on mm-wave qubit characterization.\n\nRecommendation: engage with it; request the clarifications above.","headline":"Real double-peak mm-wave data, but the quantitative fit to the strong-driving model has an unaddressed 7 GHz plasma-frequency discrepancy.","tokens_in":8732,"tokens_out":2938,"would_cite":false,"duration_ms":28282,"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":"Under 100–110 GHz irradiation, the switching-current distribution of Nb/Al-AlOx/Nb Josephson junctions splits into two peaks, a signature of resonant escape of the Josephson phase that the strong-driving model reproduces quantitatively.","keywords":["Josephson tunnel junctions","resonant escape","switching current distribution","millimeter-wave irradiation","plasma frequency","phase qubits","strong-driving model","Nb/Al-AlOx/Nb"],"falsifier":"An independent measurement of the plasma frequency that disagrees with the fitted 145.5 GHz by more than the stated uncertainty, for example direct microwave spectroscopy of the same junction as a qubit at base temperature, would falsify the identification; likewise, if only the primary peak appeared at higher drive power, or if fitting each frequency required a different plasma frequency rather than one common value, the strong-driving model's claim would fail.","tokens_in":7624,"feed_emoji":"⚡","tokens_out":7231,"duration_ms":66619,"temperature":0.7,"pith_summary":"This paper tries to establish that the switching statistics of a current-biased Nb/Al-AlOx/Nb Josephson junction under millimeter-wave irradiation carry a clear fingerprint of resonant phase escape: as the radiation power increases, the single peak in the switching-current distribution splits into a primary peak and a resonant peak at lower current. The authors show that the power dependence of these two branches at 106–110 GHz is quantitatively captured by the strong-driving model of effective barrier suppression, with a single shared plasma frequency ωp/2π = 145.5 GHz and quality factors Q ≈ 90. This matters because junctions with plasma frequencies near 140 GHz have been difficult to fabricate and characterize, and operation of phase qubits near 100 GHz would relax cooling requirements well beyond the current millikelvin standard. The measurements are performed at 4.2 K in the thermally activated regime, so the work is a feasibility proof for mm-wave phase qubits rather than a demonstration of quantum-coherent operation.","feed_headline":"100-GHz drive splits Josephson switching into two peaks","feed_subtitle":"Double-peak statistics fit a strong-driving model with plasma frequency 145.5 GHz, pointing to phase qubits near 100 GHz.","key_machinery":"The central object is the Josephson phase difference δ, treated as the coordinate of a fictitious particle in the tilted-washboard potential of the resistively and capacitively shunted junction model. Escape over the current-reduced barrier is enhanced by radiation when the drive frequency matches the small-oscillation frequency ω0(I), and the analysis rests on the strong-driving limit (ω/ωp)^5 ≫ ℏωp/EJ (about 170 here), where the escape is governed by a radiation-induced suppression of the effective potential barrier. The load-bearing identity is Eq. (8), a transcendental equation for the average switching-current shift ⟨δIsw(P)⟩ whose multiple solutions produce the coexisting primary and resonant peaks; in the harmonic approximation the sum reduces to a single resonant term (Eq. (10)) with the phase matrix element of the harmonic oscillator.","core_discovery":"The central claim is that the double-peak structures observed in the switching-current distributions are not noise artifacts but resonant escape of the Josephson phase from a metastable well: the drive at 100–110 GHz enhances the escape rate when the small-oscillation frequency ω0(I) = ωp(1 − (I/Ic)²)^{1/4} matches the irradiation frequency, producing a second, power-dependent peak at lower current. Fitting the resonant-peak position versus frequency gives ωp/2π = 138.3(7) GHz, close to the 140 GHz design target. The full power dependence at three frequencies is fitted by the strong-driving model of Eq. (8), using Ic = 242 μA and T = 4.2 K, with a common plasma frequency of 145.5 GHz and effective quality factors of about 90. The authors take this agreement as evidence that low-loss Nb/Al-AlOx/Nb junctions with plasma frequencies around 140 GHz can be fabricated, and that the same switching-current measurement provides a rapid characterization route for mm-wave qubit development.","pith_inferences":["One extension the authors leave implicit is that the harmonic approximation keeps only the 0→1 transition, so the fit cannot test the anharmonic part of the well; repeating the measurement at lower frequencies or higher power could expose deviation from Eq. (10) and thus map the level structure directly.","The arbitrary power axis means the extracted coupling coefficient k is not an absolute microwave amplitude; an independent calibration of the on-chip field would turn Eq. (8) into a quantitative mm-wave power meter.","If the quality factor at 4.2 K is indeed quasiparticle-limited at Q ≈ 90, cooling the same junctions below 1 K should raise Q substantially; a straightforward test is to repeat the double-peak spectroscopy at 100–200 mK and check whether the resonant-peak width narrows as expected.","The technique could be transferred to other high-gap materials such as niobium nitride, where even larger plasma frequencies would push the qubit operating frequency further and potentially enable operation at liquid-nitrogen temperatures, though this goes well beyond what the data demonstrate."],"forward_implications":["If the strong-driving model is right, the position of the resonant peak as a function of irradiation frequency is a direct measure of the junction plasma frequency, so switching-statistics measurements can be used as a fast, non-destructive diagnostic for high-current-density junction fabrication.","Junctions with plasma frequencies around 140 GHz are within reach of the standard Nb/Al-AlOx/Nb trilayer process, which removes a material barrier to phase qubits operating near 100 GHz.","Operating a phase qubit near 100 GHz raises the temperature scale T0 = hf/kB to about 4.8 K, suggesting that qubit operation at 1 K or above is not ruled out by level-spacing requirements.","The agreement of the fitted quality factors Q ≈ 90 at 4.2 K with quasiparticle-limited damping implies the same junctions should be substantially less dissipative at millikelvin temperatures, where coherent mm-wave qubit operation would be attempted.","The multivalued solutions of Eq. (8) explain the coexistence of primary and resonant peaks, giving a predictive handle on the microwave power levels needed to resonantly switch a junction in applications."],"supporting_citations":[{"why":"Supplies the strong-driving model of Eq. (8) that the paper fits to the mm-wave power dependence.","marker":"[6]"},{"why":"Provides the RCSJ-based thermal escape-rate framework, Eq. (7), used to fit the undriven switching distribution and extract Ic.","marker":"[4]"},{"why":"Establishes resonant enhancement of escape through multiphoton transitions and the dependence of the resonant-peak position on the small-oscillation frequency.","marker":"[5]"},{"why":"Provides experimental precedent for microwave-induced thermal escape and multi-peaked switching-current distributions above the crossover temperature.","marker":"[15]"},{"why":"Supplies the harmonic-approximation form of the strong-driving model and its application to resonant phase escape in intrinsic Josephson junctions.","marker":"[16]"},{"why":"Gives the relation between switching-current statistics and escape rates used to transform histograms into escape-rate data.","marker":"[9]"},{"why":"Provides the harmonic-oscillator phase matrix element used in the harmonic approximation of Eq. (10).","marker":"[17]"},{"why":"Demonstrates the current-biased junction as a phase qubit and the resonant escape measurement that the present work extends to mm-wave frequencies.","marker":"[3]"}],"fun_headline_variants":["Double-peak switching reveals resonant escape at 100 GHz","Microwave drive creates twin escape peaks in Josephson junctions","Resonant escape: key to 100-GHz phase qubits","Josephson junctions show resonant escape under mm-wave drive"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that radiation suppresses the escape barrier in the way the paper's strong-driving model describes, with a known plasma frequency and damping; if that simplified model is wrong, the inferred plasma frequency and quality factor do not follow from the data, especially since the microwave power axis is in arbitrary units and the coupling is fit rather than independently measured.","fun_headline_variants_meta":{"raw":{"variants":["Double-peak switching reveals resonant escape at 100 GHz","Microwave drive creates twin escape peaks in Josephson junctions","Resonant escape: key to 100-GHz phase qubits","Josephson junctions show resonant escape under mm-wave drive"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000163,"raw_usage":{"total_tokens":1240,"prompt_tokens":938,"completion_tokens":302,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":554,"completion_tokens_details":{"reasoning_tokens":233}},"tokens_in":554,"tokens_out":302,"duration_ms":3172,"temperature":1.0,"reasoning_tokens":233,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T14:32:39.417691+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"An independent measurement of the plasma frequency that disagrees with the fitted 145.5 GHz by more than the stated uncertainty, for example direct microwave spectroscopy of the same junction as a qubit at base temperature, would falsify the identification; likewise, if only the primary peak appeared at higher drive power, or if fitting each frequency required a different plasma frequency rather than one common value, the strong-driving model's claim would fail.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the strong-driving model of Eq. (8) that the paper fits to the mm-wave power dependence."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the RCSJ-based thermal escape-rate framework, Eq. (7), used to fit the undriven switching distribution and extract Ic."},{"cited_title":"Wallraff, T","cited_arxiv_id":null,"evidence_quote":"Establishes resonant enhancement of escape through multiphoton transitions and the dependence of the resonant-peak position on the small-oscillation frequency."},{"cited_title":"Grønbech-Jensen, M","cited_arxiv_id":null,"evidence_quote":"Provides experimental precedent for microwave-induced thermal escape and multi-peaked switching-current distributions above the crossover temperature."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the harmonic-approximation form of the strong-driving model and its application to resonant phase escape in intrinsic Josephson junctions."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the relation between switching-current statistics and escape rates used to transform histograms into escape-rate data."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the harmonic-oscillator phase matrix element used in the harmonic approximation of Eq. (10)."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Demonstrates the current-biased junction as a phase qubit and the resonant escape measurement that the present work extends to mm-wave frequencies."}],"review_version":1}