{"id":"77cee836-c56f-4ea4-8489-08b0a8353222","arxiv_id":"2507.09339","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A granular-aluminum superinductor galvanically couples a flux qubit to a resonator, achieving g/ω_r ≃ 0.13 and entering the perturbative ultrastrong coupling regime.","lead":"This paper reports a superconducting circuit in which a granular-aluminum superinductor galvanically couples a flux qubit to a resonator, reaching a coupling fraction g/ω_r of about 0.13. The result places the device in the ultrastrong coupling regime while using the superinductor as a linear, high-inductance coupler rather than a Josephson junction.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Independent L_c estimate and Bloch-Siegert shift do not quantitatively support the fitted g/ω_r≈0.13; the claimed USC validation rests on an unevaluated internal inconsistency.","rationale":"The reader's conditional verdict identifies exactly the load-bearing weak point: the paper's independent validation of the USC claim is Eq. (3), but the numerical inputs used in that equation are not mutually consistent. Recomputing Eq. (3) with the fitted persistent current yields g/ω_r≈0.08, below the USC threshold, so the statement that the L_c estimate is 'compatible with g/ω≳0.1' is not supported by the fitted parameters. Separately, the 23 MHz Bloch-Siegert shift is in tension with the fitted g=0.578 GHz: the same formula quoted in the paper predicts about 33 MHz. A legitimate explanation may exist, for example a distinction between bare and renormalized persistent current or higher-order corrections beyond the simple BS formula, but the manuscript does not provide it. The additional weakness that Eq. (3) is derived in an unpublished manuscript (reference 33) makes this quantitative check more, not less, important. The experimental work and fabrication details are otherwise presented in a reproducible manner, but the central quantitative claim—entering the USC regime—depends on a coupling strength that is not consistently corroborated by the two independent estimates given. A constrained refit would settle the issue; absent that, the CONDITIONAL verdict is appropriate. I therefore keep the reader's verdict unchanged.","tokens_in":11715,"tokens_out":5704,"duration_ms":67438,"concrete_test":"Re-fit the measured single-tone and two-tone spectra with g constrained to the value obtained from Eq. (3) using the fitted I_p=(11.619±0.004) nA and L_c=(0.74±0.14) nH, while allowing the other parameters to vary. Report the change in goodness of fit relative to the unconstrained fit that returns g/2π=0.578 GHz. If the constrained fit is rejected at more than 3σ, the independent inductance estimate contradicts the coupling required for USC. If the constrained fit is not rejected, compute the Bloch-Siegert shift from the newly fitted g via ω_BS=g²/(ω_r+ω_q) and compare it with the measured 23 MHz to determine which of the two estimates is consistent with the data.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that the independently measured coupler inductance L_c=(0.74±0.14) nH validates the USC value g/ω_r≈0.13 depends on Eq. (3) being evaluated with the same persistent current that appears in the fitted Hamiltonian. The paper's own numbers break this consistency. For the design current I_p≈19.6 nA, Eq. (3) gives g/2π≈0.61 GHz. Scaling linearly with I_p to the fitted value I_p=(11.619±0.004) nA gives g/2π≈0.36 GHz, i.e. g/ω_r≈0.08 with ω_r/2π=4.463 GHz, below the claimed USC threshold g/ω_r>0.1. The paper instead compares the fitted g=0.578 GHz to a prediction made at a current that is not the fitted one. A second internal inconsistency appears in the stated Bloch-Siegert shift: using the paper's own formula ω_BS=g²/(ω_r+ω_q) with the fitted g=0.578 GHz, ω_r/2π=4.463 GHz, and Δ/h=5.707 GHz at the sweet spot gives ω_BS≈33 MHz, not the measured 23 MHz. The observed 23 MHz instead implies g/2π≈0.48 GHz and g/ω_r≈0.11. The manuscript does not explain why the full QRM fit and the Bloch-Siegert estimate differ, nor why Eq. (3) at the fitted I_p falls below the USC boundary. Since the headline coupling fraction depends on a fitted g that is not supported by the two independent checks presented, the USC claim is not yet settled.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports a flux qubit galvanically coupled to a lumped-element LC resonator through a granular aluminum (grAl) superinductor. Single-tone and two-tone spectroscopy are measured as a function of flux, and the observed transitions are fitted to the quantum Rabi model, yielding I_p = (11.619±0.004) nA, Δ/h = (5.707±0.002) GHz, ω_r/2π = (4.463±0.001) GHz, and g/2π = (0.578±0.001) GHz, i.e., g/ω_r ≈ 0.13, which is claimed to enter the perturbative ultrastrong coupling (USC) regime. The coupler inductance is independently estimated from room-temperature and 4 K resistance measurements via the Mattis-Bardeen formula, giving L_c = (0.74±0.14) nH. A Bloch-Siegert shift of 23 MHz is reported for the ω_01 transition at the sweet spot. The paper claims that both the inductance estimate and the Bloch-Siegert shift are consistent with the fitted coupling, thereby validating the USC regime.","tokens_in":12051,"tokens_out":6800,"duration_ms":74280,"significance":"If the central claim holds, the work demonstrates a promising alternative to Josephson-junction-based USC couplers: a linear superinductor made of granular aluminum can provide large coupling while keeping the qubit persistent current and loop area small, potentially improving coherence. The direct spectral fit to multiple transitions over a range of flux biases is a robust method for extracting the QRM parameters, and the fabrication and material characterization are detailed and reproducible. However, the two independent checks presented in support of the USC claim are internally inconsistent under the manuscript's own equations: Eq. (3) evaluated with the fitted persistent current gives a coupling below the USC threshold, and the Bloch-Siegert formula with the fitted parameters does not reproduce the reported 23 MHz shift. These quantitative tensions are load-bearing because the abstract and conclusions use them to claim validation of the USC regime. The central claim may still be correct, but it is not yet settled by the evidence presented.","major_comments":[{"comment":"Equation (3) is linear in the persistent current I_p, and the predicted g/2π ≈ 0.61 GHz is evaluated with the design value I_p ≈ 19.6 nA. The QRM fit, however, returns I_p = (11.619±0.004) nA. Re-evaluating Eq. (3) with the fitted I_p and L_c = 0.74 nH gives g/2π ≈ 0.61 × (11.619/19.6) ≈ 0.36 GHz, corresponding to g/ω_r ≈ 0.08, which is below the USC threshold g/ω_r > 0.1 used in the paper. Therefore the statement that the independent L_c estimate is 'consistent with' the USC claim is not supported by the manuscript's own equations. The authors should re-evaluate Eq. (3) at the fitted current and discuss the resulting discrepancy, or justify why the design current should be used instead of the fitted one.","section":"Main text, Section 'Using Eq. (3)' (p. 2-3)"},{"comment":"The paper reports ω_BS = 23 MHz for the ω_01 transition at the sweet spot and gives the formula ω_BS = g²/(ω_r + ω_q). Substituting the fitted parameters g/2π = 0.578 GHz, ω_r/2π = 4.463 GHz, and Δ/h = 5.707 GHz yields ω_BS ≈ 33 MHz, not 23 MHz. Conversely, the observed 23 MHz implies, through the same formula, g/2π ≈ 0.48 GHz, about 17% below the fitted g/2π = 0.578 GHz. The manuscript states that these are 'consistent' without quantifying or explaining the discrepancy. The authors should clarify whether the analytic formula is expected to be accurate at g/ω_r ≈ 0.13, how the measured shift was extracted from the data, and why the full QRM fit and the Bloch-Siegert estimate differ.","section":"Main text, Bloch-Siegert shift (p. 3)"},{"comment":"The abstract states that the independently measured L_c = (0.74±0.14) nH is 'compatible with g/ω ≳ 0.1.' This claim depends on evaluating Eq. (3) with the design persistent current rather than the fitted current; with the fitted I_p, Eq. (3) gives g/ω_r ≈ 0.08, below the USC boundary. The abstract's compatibility claim therefore needs to be revised or the analysis underlying it must be corrected.","section":"Abstract"}],"minor_comments":[{"comment":"The caption reads 'transmission magnitude |S_21| vs. flux bias (Φ_ext)' with inconsistent spacing; please add units for Φ_ext and ensure consistent formatting of subscripts.","section":"Fig. 2 and Fig. 3 captions"},{"comment":"The term 'ˆp2 + 1 γ ˆp2 3' contains an orphan '1' and unclear grouping; the expression should be typeset as (1/γ) p̂_3² or with explicit parentheses.","section":"Eq. (15)"},{"comment":"The quantity J_c = (0.66±0.03) µA/µm² is introduced without definition; please define it (presumably a critical current density).","section":"Appendix B"},{"comment":"The formula for ξ_R² contains a stray 'r' and unclear notation such as 'ω_2^4 = 1/L_eff C_tot'; the expression should be simplified and all symbols defined.","section":"Footnote 34"},{"comment":"The abstract uses 'g/ω' while the main text uses 'g/ω_r'; please use one notation consistently.","section":"Notation"},{"comment":"Equation (3) is central to the validation, but its derivation is deferred to a paper 'in preparation.' Since the consistency of the inductance estimate with the fitted coupling is a key point, the derivation should be summarized or the reference should be replaced by an available source.","section":"Reference [33]"}],"recommendation":"major_revision","confidential_remarks":"The authors are clearly capable and the direct QRM fit to multiple transitions is a strong piece of evidence; the manuscript is not fundamentally flawed. However, the two independent checks—the inductance-based estimate and the Bloch-Siegert shift—are not quantitatively consistent with the fitted g under the paper's own equations, and one of them (Eq. (3) at the fitted I_p) falls below the USC threshold. This is exactly the kind of load-bearing inconsistency that must be resolved before publication. I would encourage the editor to ask for a revised version that addresses the discrepancy, rather than rejecting the paper outright, because the central spectral fit may well be correct and the issue may be fixable by a more careful error budget or a corrected analysis."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"What you should know: this is the first demonstration I've seen of a superinductor—granular aluminum—used as the galvanic coupling element between a flux qubit and a resonator, with spectroscopy showing a transition spectrum that fits the Quantum Rabi model with g/omega_r ~ 0.13. That's a genuinely new device idea, and the fabrication and measurement work looks careful: calibration data for the grAl, raw transmission in the appendix, and a two-tone spectrum that shows the qubit-resonator anticrossing.\n\nWhat's less solid is the quantitative validation. The paper claims the independently measured L_c=(0.74±0.14) nH is compatible with g/omega >= 0.1. That check uses Eq. (3) with the design persistent current I_p≈19.6 nA, which gives g/2π≈0.61 GHz. But the QRM fit returns I_p=(11.62±0.01) nA. Plugging that into the same equation gives g/2π≈0.36 GHz, i.e. g/omega_r≈0.08, below the USC threshold. That's a factor of 1.6 in g, and it's not within the error bars. The paper doesn't flag this.\n\nThe Bloch-Siegert shift is similarly inconsistent. The paper reports 23 MHz at the sweet spot and says this is consistent with the fitted g=0.578 GHz. Using their own formula omega_BS=g^2/(omega_r+omega_q) with omega_r/2π=4.463 GHz and Delta/h=5.707 GHz gives about 33 MHz. The measured 23 MHz would correspond to g/2π≈0.48 GHz, g/omega_r≈0.11. That's still USC, but it's 17% below the fitted value, and the manuscript states they are consistent without explaining the difference.\n\nThere's also a reproducibility issue: Eq. (3), the key link between L_c and g, is deferred to an unpublished manuscript (ref. 33). Appendix A derives a simpler limiting case, but the general form with xi_R is not derived here. That makes it hard for a reader to assess whether the linear scaling with I_p is even the right model for this circuit.\n\nSo where does this leave the central claim? The QRM fit itself is a reasonable reading of the spectroscopy, and the device clearly shows strong coupling features. But the two independent checks the paper presents to support the USC classification don't hold up quantitatively. The truth might be g/omega_r≈0.11 from the BS shift, which still enters USC, or it might be ~0.08 if the inductance estimate is right, which doesn't. The paper needs to reconcile these numbers before the headline is solid.\n\nWho is this for? People working on USC circuit QED and superinductor materials. The demonstration is novel enough to deserve peer review, but the referees should ask for a consistent cross-check among I_p, L_c, g, and the Bloch-Siegert shift—and ideally a full derivation of the coupling formula.","headline":"A novel and promising superinductor coupling scheme, but the paper's own quantitative checks (inductance estimate and Bloch-Siegert shift) don't line up with the fitted g, so the USC claim needs a consistent re-analysis.","tokens_in":12635,"tokens_out":7071,"would_cite":false,"duration_ms":80181,"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":"A granular-aluminum superinductor galvanically couples a flux qubit to a resonator with g/ω_r ≈ 0.13, placing the system in the perturbative ultrastrong coupling regime.","keywords":["ultrastrong coupling","superinductor","granular aluminum","flux qubit","Bloch-Siegert shift","quantum Rabi model","kinetic inductance","superconducting circuit"],"falsifier":"A direct time-domain measurement of vacuum Rabi oscillations (or a resolved vacuum Rabi splitting) at the sweet spot that returns a coupling rate g/2π below about 0.45 GHz would put g/ω_r under 0.1 and would falsify the claim that the device is in the USC regime.","tokens_in":11510,"feed_emoji":"⚡","tokens_out":10539,"duration_ms":105083,"temperature":0.7,"pith_summary":"The paper claims that a superinductor, a large-inductance, low-loss circuit element, can replace the Josephson junction usually used to couple a flux qubit to a resonator and still push the system into the ultrastrong coupling regime. Spectroscopy of the coupled circuit fits the Quantum Rabi model with g/ω_r ≈ 0.13, above the 0.1 threshold, and shows a 23 MHz Bloch-Siegert shift, the hallmark of counter-rotating terms. An independent measurement of the coupler inductance from low-temperature resistance is reported as compatible with the strong coupling. If correct, this relaxes the design constraints for ultrastrong coupling circuits, allowing smaller qubit loops and lower persistent currents, which should improve coherence.","feed_headline":"Superinductor pushes qubit-resonator pair into ultrastrong coupling","feed_subtitle":"A granular-aluminum wire replaces the Josephson junction, pointing toward lower-loss circuits with smaller qubit loops.","key_machinery":"The central object is the granular-aluminum (grAl) superinductor used as the shared coupling inductor L_c of a three-junction flux qubit and a lumped-element LC resonator. Granular aluminum is a disordered superconductor whose kinetic inductance provides a large surface inductance in a compact wire, giving a linear, low-loss coupler without a Josephson junction. The coupling strength, g ≈ ξ_R (L_eff I_p I_rms,R)/ħ, is derived from the circuit, and the resulting Quantum Rabi Hamiltonian is used to fit the spectra; the large L_c also renormalizes the resonator frequency, ω_r ≈ [(L_R + L_c)/C_R]^{-1/2}. The Bloch-Siegert shift, ω_BS = $g^{2}$/(ω_r + ω_q), serves as an independent check of the coupling strength.","core_discovery":"The central experimental result is a transmission spectrum of a galvanically coupled flux-qubit–resonator system fitted to the Quantum Rabi Hamiltonian with parameters I_p = (11.619 ± 0.004) nA, Δ/h = (5.707 ± 0.002) GHz, ω_r/2π = (4.463 ± 0.001) GHz, and g/2π = (0.578 ± 0.001) GHz, giving a coupling fraction g/ω_r ≈ 0.13. The difference between the QRM and Jaynes-Cummings spectra at the sweet spot is 23 MHz, attributed to the Bloch-Siegert shift, and this shift independently yields g/2π ≈ 0.48 GHz, consistent with the fit. The coupling inductor is a granular-aluminum wire with an independently estimated inductance L_c = (0.74 ± 0.14) nH obtained from low-temperature resistance and the Mattis-Bardeen formula, a value reported as compatible with g/ω ≳ 0.1. The paper concludes that superinductors can bring qubit-resonator systems into the USC regime while keeping persistent currents low and qubit loops small.","pith_inferences":["The same galvanic superinductor coupling scheme could be applied to other qubit types, potentially relaxing impedance-matching constraints for USC in transmon-like circuits.","Because the superinductor is linear, the approach may extend to ultrastrong coupling between two resonators or to multi-qubit USC networks without adding junction nonlinearities.","A time-domain measurement of vacuum Rabi oscillations in this device would directly verify the fitted coupling rate and also probe USC corrections to the decay dynamics.","The close agreement between the Bloch-Siegert shift and the QRM fit suggests that this shift could be used as a precision calibration tool for coupling strengths in other superconducting circuits."],"forward_implications":["Ultrastrong coupling can be reached with a linear superinductor instead of a shared Josephson junction, avoiding junction losses and stray nonlinearities.","Design constraints are relaxed: smaller qubit loops and lower persistent currents become compatible with g/ω_r > 0.1, which should translate into longer coherence times.","The coupling fraction can be pushed to g/ω_r > 0.3 by increasing the superinductance, while keeping the qubit loop small.","The fabrication flow is modular, so the superinductor material can be replaced by other high-kinetic-inductance materials such as nitrides.","The Bloch-Siegert shift of 23 MHz provides a clear spectral signature of the counter-rotating terms in the perturbative USC regime."],"supporting_citations":[{"why":"Established the Bloch-Siegert shift as the spectroscopic signature of ultrastrong coupling; the 23 MHz shift here is compared to this effect.","marker":"[6]"},{"why":"Prior realization of USC in a flux-qubit-resonator circuit with a shared Josephson junction; the baseline the superinductor approach is contrasted with.","marker":"[5]"},{"why":"Prior USC experiment in a superconducting circuit, used as a comparison point for reaching g/ω_r ≳ 0.1.","marker":"[7]"},{"why":"Defines the USC threshold g/ω_r ≳ 0.1 used to assess the regime.","marker":"[10]"},{"why":"Granular aluminum as a high-kinetic-inductance superinductor material; the coupler is fabricated from it.","marker":"[27]"},{"why":"Mattis-Bardeen formula (L_k = 0.18 ħ R_4K/(k_B T_c)) used to estimate the coupler inductance L_c from low-temperature resistance.","marker":"[36]"},{"why":"The Quantum Rabi Hamiltonian that the measured spectrum is fitted to.","marker":"[32]"},{"why":"Jaynes-Cummings model used to compute the Bloch-Siegert shift as the difference from the QRM spectrum.","marker":"[38]"},{"why":"Derivation of the general coupling formula (Eq. 3) used to predict g from the circuit parameters.","marker":"[33]"}],"fun_headline_variants":["Superinductor drives flux qubit-resonator into ultrastrong coupling","Granular aluminum superinductor yields USC in superconducting circuit","Ultrastrong coupling via superinductor with low persistent current","Superinductor-based USC: coupling fraction 0.13 in flux qubit","Small-loop flux qubit reaches ultrastrong coupling via superinductor"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The USC claim stands or falls on the spectral fit's value of g (g/ω_r ≈ 0.13); using the independently measured L_c with the fitted persistent current gives a sub-threshold coupling, so the fit, not the inductance measurement, carries the claim.","fun_headline_variants_meta":{"raw":{"variants":["Superinductor drives flux qubit-resonator into ultrastrong coupling","Granular aluminum superinductor yields USC in superconducting circuit","Ultrastrong coupling via superinductor with low persistent current","Superinductor-based USC: coupling fraction 0.13 in flux qubit","Small-loop flux qubit reaches ultrastrong coupling via superinductor"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000222,"raw_usage":{"total_tokens":1458,"prompt_tokens":952,"completion_tokens":506,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":568,"completion_tokens_details":{"reasoning_tokens":411}},"tokens_in":568,"tokens_out":506,"duration_ms":5888,"temperature":1.0,"reasoning_tokens":411,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T18:00:04.431570+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A direct time-domain measurement of vacuum Rabi oscillations (or a resolved vacuum Rabi splitting) at the sweet spot that returns a coupling rate g/2π below about 0.45 GHz would put g/ω_r under 0.1 and would falsify the claim that the device is in the USC regime.","supporting_citations":[{"cited_title":"Stassi , author M","cited_arxiv_id":null,"evidence_quote":"Established the Bloch-Siegert shift as the spectroscopic signature of ultrastrong coupling; the 23 MHz shift here is compared to this effect."},{"cited_title":"Wallraff , author D","cited_arxiv_id":null,"evidence_quote":"Prior realization of USC in a flux-qubit-resonator circuit with a shared Josephson junction; the baseline the superinductor approach is contrasted with."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Prior USC experiment in a superconducting circuit, used as a comparison point for reaching g/ω_r ≳ 0.1."},{"cited_title":"Niemczyk , author F","cited_arxiv_id":null,"evidence_quote":"Defines the USC threshold g/ω_r ≳ 0.1 used to assess the regime."},{"cited_title":"Torras-Coloma , author L","cited_arxiv_id":null,"evidence_quote":"Granular aluminum as a high-kinetic-inductance superinductor material; the coupler is fabricated from it."},{"cited_title":"Torras-Coloma et al., in preparation NoStop","cited_arxiv_id":null,"evidence_quote":"Mattis-Bardeen formula (L_k = 0.18 ħ R_4K/(k_B T_c)) used to estimate the coupler inductance L_c from low-temperature resistance."},{"cited_title":"Gupta , author P","cited_arxiv_id":null,"evidence_quote":"The Quantum Rabi Hamiltonian that the measured spectrum is fitted to."},{"cited_title":"Magazz \\`u , author P","cited_arxiv_id":null,"evidence_quote":"Jaynes-Cummings model used to compute the Bloch-Siegert shift as the difference from the QRM spectrum."},{"cited_title":"u ller , author L. Ding , author A. P. \\ Veps \\","cited_arxiv_id":null,"evidence_quote":"Derivation of the general coupling formula (Eq. 3) used to predict g from the circuit parameters."}],"review_version":1}