{"id":"135d77ae-612e-47e7-a860-1d743ff47dc4","arxiv_id":"2508.11731","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":8.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"First demonstration of direct optical interferometric readout and feedback cooling of a magnetically levitated superconducting microsphere, with about 1 nm/√Hz displacement sensitivity at 3 K.","lead":"Researchers used laser light to measure the position of a 6 microgram superconducting sphere floating in a magnetic trap at 3 K, reaching a sensitivity of about one nanometer per square-root hertz. They then used that optical signal to feedback-cool the sphere's jittering motion, a step toward quantum experiments with heavier objects.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Abstract's 'better than 1 nm/√Hz' is not supported by the reported uncertainty: 955 ± 148 pm/√Hz puts the 1σ upper bound above the threshold.","rationale":"The reader's weakest_assumption focuses on the calibration (Eq. S2 and constant suppression). I do not find that assumption decisive: the probe-tone and piezo-mirror calibrations agree within uncertainty, and the uniform-field approximation for a 50 µm sphere in the external coil field is likely good to a few percent. The calibration is also internally consistent with the feedback-cooling result. The concern I substantiate is the abstract's statistical overstatement: 955(148) pm/√Hz does not support an unqualified 'better than 1 nm/√Hz' at 1σ. This is a wording/correctness issue in the central claim, not an invalidation of the experiment. Since the reader already reached CONDITIONAL on this basis, my read does not change the verdict; it sharpens the reason and proposes a concrete re-analysis. No ad hominem or manufactured concern is intended.","tokens_in":18892,"tokens_out":14135,"duration_ms":168867,"concrete_test":"Recompute the noise floor from the Zenodo raw time series by averaging the calibrated one-sided displacement PSD over the same frequency bins, and form a 95% confidence interval on the mean that propagates the calibration uncertainties (suppression ratio 7.95 ± 1.17 and piezo amplitude ±10%). If the 95% upper bound exceeds 1 nm/√Hz—as expected from the 0.3σ margin—the abstract must be revised; also verify that the suppression factor measured at λ/8 is unchanged when the piezo amplitude is reduced to ~10 nm, to confirm the calibration scale.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's headline claim is a sub-nm/√Hz displacement resolution. The measured calibrated one-sided noise floor is √2 S_zz = 955(148) pm/√Hz. Compared with the 1 nm/√Hz threshold, the 45 pm/√Hz margin is only 0.3σ, so the data do not establish 'better than 1 nm/√Hz' at a standard confidence level. This is not a calibration failure: the two calibrations agree (suppression ratio 7.95 ± 1.17 vs 7.5 ± 0.75), and the uniform-field approximation in Eq. S2 is plausible given the coil/particle geometry. The remaining load-bearing issue is the wording of the abstract, which converts a point estimate into an unqualified threshold claim. The same uncertainty propagates to the 8 nm cooled amplitude, but that claim is not challenged at the same level. The central experimental demonstration stands; the headline should be tempered to '≈1 nm/√Hz' or carry the error bar.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports interferometric readout of the axial motion of a 6 µg magnetically levitated superconducting PbSn microsphere in an anti-Helmholtz trap at 3 K. A Mach-Zehnder-like interferometer with balanced homodyne detection and AOM phase tracking is used; the signal is calibrated by two independent methods (piezo-driven mirror and probe-tone force response) and is used for feedback cooling. The quoted calibrated one-sided displacement noise floor is sqrt(2) S_zz = 955(148) pm/sqrt(Hz), about two orders of magnitude above the 11 pm/sqrt(Hz) shot-noise limit, attributed to surface roughness and rotation. Feedback cooling reaches sqrt(<z^2>) ≈ 8 nm. The paper also reports levitation lifetimes under optical illumination, a thermal decoherence rate, and a projected cavity-enhanced route toward ground-state cooling.","tokens_in":19107,"tokens_out":11118,"duration_ms":121338,"significance":"The demonstration of direct optical interferometric readout of a levitated superconductor is significant: it addresses a longstanding concern that optical probing would quench superconductivity, and it provides a calibrated displacement measurement at the nm/sqrt(Hz) level with feedback cooling to ~8 nm amplitude. Strengths include the use of two independent calibration methods that agree within uncertainty, the quantification of the noise floor, the availability of the data on Zenodo, and the explicit statement of the quantum cooperativity and ground-state cooling requirements. If the headline claim is appropriately qualified, the work is a solid experimental step toward quantum experiments with microgram levitated masses.","major_comments":[{"comment":"The abstract states 'achieving a resolution better than 1 nm/√Hz', but the main text reports a calibrated one-sided noise floor of √2 S_zz = 955(148) pm/√Hz near the mechanical frequency. At 1σ the upper bound is 1103 pm/√Hz, which exceeds 1 nm/√Hz. The data therefore do not establish 'better than 1 nm/√Hz' at the stated uncertainty; the concluding 'around 1 nm/√Hz' is supported. Please temper the abstract to '≈1 nm/√Hz' or quote the value with its uncertainty.","section":"Abstract and main text, noise-floor statement"},{"comment":"The probe-tone calibration assumes Δz = B_ext/(dB_trap/dz), i.e., a uniform external calibration field over the 100 µm sphere and a linear trap gradient. Any violation of these assumptions enters directly into the absolute displacement scale, and hence into the 955 pm/√Hz floor and the 8 nm cooled amplitude. The agreement between the two calibration methods is reassuring, but the uniform-field and linear-gradient approximations should be justified quantitatively from the coil geometry and trap profile, and the residual systematic uncertainty should be stated.","section":"Supplement, Eq. (S2)"},{"comment":"The reported value √2 S_zz = 955(148) pm/√Hz lacks a definition of the estimator: no bandwidth or smoothing is specified, no number of independent spectra, and no explicit propagation of the calibration uncertainties (suppression ratio 7.95±1.17 vs 7.5±0.75) into the 148 pm/√Hz error bar. This information is needed to assess the statistical significance of the comparison with the 1 nm/√Hz threshold and should be added to the supplement.","section":"Main text, noise-floor statement"}],"minor_comments":[{"comment":"Reference [31] is the self-reference 'Supplementary material, .' and is incomplete. The supplement also contains an unresolved cross-reference 'described in Sec. .'.","section":"References"},{"comment":"The expression 'n <1 ⇒ 4 g² ncav/(κ Γ_th) > 1/(9η-1)>0' is confusing: the condition should be η>1/9 so that the denominator is positive; the trailing '>0' is either redundant or a typo.","section":"Main text, ground-state condition"},{"comment":"The phrase 'The resolution exceeds the shot-noise limit' is ambiguous because a larger displacement noise floor is worse, not better. Consider wording such as 'the noise floor is a factor of ~87 above the shot-noise limit'.","section":"Abstract"},{"comment":"The statement that readout and cooling protocols are 'carried out over much shorter time periods' should be reconciled with the main text's 100 s run-time limit and the green region in Fig. S11.","section":"Supplement, Levitation time"}],"recommendation":"minor_revision","confidential_remarks":"The experimental core is sound and the required changes are textual and clarificatory: qualify the headline threshold claim, add the requested calibration and noise-floor details, and fix the incomplete references. I would not require new measurements. The paper is suitable for publication in a good journal after these revisions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Jannek Hansen et al. have done the first direct optical interferometric readout of a magnetically levitated superconducting microsphere. That is the headline result, and it holds up. The displacement noise floor of 955(148) pm/√Hz near the mechanical frequency is calibrated two independent ways — a piezo-driven mirror and a probe-tone force model — and the two agree within uncertainty (suppression ratios 7.95±1.17 vs 7.5±0.75). The data are on Zenodo. The feedback cooling of the axial motion to about 8 nm is a natural consequence and is well executed. The supplementary material is honest about the causes of noise (particle roughness, rotations) and about the levitation lifetime limits under optical illumination. The path to quantum ground state is explicitly a projection, and the reported quantum cooperativity Cq ≈ 10⁻²¹ makes clear that cavity enhancement is needed; that is not hidden.\n\nThe one real defect is the abstract's phrase 'achieving a resolution better than 1 nm/√Hz'. The point estimate 955 pm/√Hz is indeed below 1 nm/√Hz, but with a 148 pm/√Hz standard uncertainty the 1σ upper bound is 1103 pm/√Hz. So the data do not establish sub-nm resolution at a standard confidence level. The authors' own summary says 'around 1 nm/√Hz', which is exactly the right wording. The abstract should say '≈1 nm/√Hz' or include the uncertainty. This is a statistical wording issue, not a flaw in the measurement: the two calibrations agree, the noise floor is reproducible, and the central claim — first optical interferometric readout — does not depend on crossing the 1 nm/√Hz threshold.\n\nThe softer spots are minor. The calibration model in Eq. S2 assumes a uniform external field over the sphere and a linear trap gradient; that is plausible given the geometry and is partially supported by the agreement between the two methods. The cooled amplitude of 8 nm inherits the same calibration uncertainty, but nothing in the paper's conclusions hinges on that precise number. The 'shot-noise limit of 11 pm/√Hz' is an idealization; the demonstrated sensitivity is two orders of magnitude worse, so the headline is modest in absolute terms, but it is still a first for this platform.\n\nWho benefits: anyone working on levitated optomechanics, macroscopic quantum mechanics, or magnetic levitation sensors. This is a serious experimental paper with deposited data and a clear, falsifiable central claim. I would send it to a good referee and expect it to be accepted after the abstract is tempered. My own verdict: cite it, bring it to reading group, and don't let the overconfident abstract ruin an otherwise solid result.","headline":"First direct optical readout of a levitated superconducting microsphere: solid experiment, only real blemish is the abstract's 'better than 1 nm/√Hz' overstatement.","tokens_in":19619,"tokens_out":2476,"would_cite":true,"duration_ms":24550,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The authors demonstrate direct optical interferometric readout of a magnetically levitated superconducting microsphere with sub-nanometer-per-root-hertz resolution, and use the signal to feedback-cool its axial motion to about 8 nm.","keywords":["levitated optomechanics","superconducting microsphere","optical interferometry","feedback cooling","magnetic levitation","displacement sensing","cryogenic","balanced homodyne detection"],"falsifier":"Repeat the measurement with a polished sphere of the same mass: if the noise floor does not move substantially toward the $11\\,\\mathrm{pm}/\\sqrt{\\mathrm{Hz}}$ shot-noise limit, the surface-roughness explanation is not the whole story. Independently, calibrate the displacement signal with a method that does not assume a uniform probe field—e.g., a calibrated radiation-pressure force or an auxiliary SQUID readout—and check whether it agrees with the probe-tone calibration.","tokens_in":18790,"feed_emoji":"🧲","tokens_out":11676,"duration_ms":117087,"temperature":0.7,"pith_summary":"This paper reports direct optical interferometric readout of a magnetically levitated superconducting microsphere. At 3 K, a 6 µg lead-tin sphere in an anti-Helmholtz trap serves as one mirror of a balanced homodyne interferometer; the calibrated displacement noise floor is $955(148)\\,\\mathrm{pm}/\\sqrt{\\mathrm{Hz}}$ near the mechanical frequency, better than $1\\,\\mathrm{nm}/\\sqrt{\\mathrm{Hz}}$, and the same signal feedback-cools the axial motion to about $8\\,\\mathrm{nm}$. The authors argue this removes a key bottleneck for levitated superconducting masses, since optical readout was considered incompatible with superconductors because single photons can break Cooper pairs; at the low photon flux used ($10^7$ photons/s) the particle stays superconducting for hundreds of seconds. The measured floor is about two orders above the $11\\,\\mathrm{pm}/\\sqrt{\\mathrm{Hz}}$ photon shot-noise limit, with the excess attributed to surface roughness, rotations, and drifts of the commercial sphere. If the levitator is placed in an optical cavity, the paper's calculation says ground-state cooling of a $6\\,\\mu\\mathrm{g}$ mass would need only about $0.9\\,\\mathrm{pW}$ of input light.","feed_headline":"Optical readout cools a levitated superconductor to 8 nm","feed_subtitle":"Sub-nanometer resolved motion points the way to quantum experiments with microgram masses.","key_machinery":"The load-bearing mechanism is a Mach-Zehnder-like balanced homodyne interferometer that uses the levitated microsphere as the signal-arm reflector. A closed-loop phase lock drives an acousto-optic modulator to track the frequency difference between the arms, compensating Doppler shifts and phase fluctuations from the rough particle; this linearises the detector difference signal for displacements of several $\\mu\\mathrm{m}$ and keeps it proportional to axial position. Calibration is done two ways—reflection from a piezo-mounted mirror at known amplitude, and an oscillating magnetic probe tone whose harmonic-oscillator response is fitted—and the calibrated signal is bandpass-filtered, phase-de","core_discovery":"The central experimental claim is that the axial motion of a $6\\,\\mu\\mathrm{g}$ superconducting PbSn microsphere levitated in an anti-Helmholtz trap at 3 K can be measured optically with a calibrated one-sided displacement noise floor of $\\sqrt{2 S_{zz}} = 955(148)\\,\\mathrm{pm}/\\sqrt{\\mathrm{Hz}}$ near the mechanical frequency, and that this signal is clean enough to feedback-cool the mode to $\\sqrt{\\langle z^2 \\rangle_{\\min}} \\approx 8\\,\\mathrm{nm}$. The authors note this is about two orders of magnitude above the $11\\,\\mathrm{pm}/\\sqrt{\\mathrm{Hz}}$ shot-noise limit set by $10^7$ collected photons/s, and they trace the excess to the particle's surface roughness ($\\sigma_r = 50\\,\\mathrm{nm}","pith_inferences":["If the dominant excess noise comes from the sphere rotating under the beam, locking the probe to a fixed scattering region—e.g., by tracking the particle's rotation or using structured illumination—could cut the floor without needing smoother spheres; the paper does not test this.","The abrupt drop in levitation lifetime above about $2 \\times 10^7$ photons/s hints at quasiparticle or two-photon processes; probing at wavelengths below the superconducting gap energy could extend lifetimes or permit higher readout power.","The same phase-tracked interferometer should work on any reflective levitated object, so the readout could transfer to optically levitated or hybrid particles where photon-pair-breaking is not a constraint.","A third calibration method that avoids the uniform-field assumption would directly test the most fragile step in the absolute displacement scale; the two existing methods agreeing within uncertainty is encouraging but not a proof of that assumption."],"forward_implications":["Direct optical readout of a levitated superconductor works at low photon flux ($10^7$ photons/s) without immediate quenching, so the readout bottleneck that previously favoured SQUIDs is removed.","At the achieved resolution, interferometric feedback cooling brings the axial mode to about $8\\,\\mathrm{nm}$, and the ring-up data give a thermal decoherence rate $\\Gamma_{\\mathrm{th}} = 6.4 \\times 10^{12}\\,\\mathrm{Hz}$ at a trap frequency of 160 Hz.","If the surface-roughness, rotation, and drift noise are suppressed, the same interferometer could approach its $11\\,\\mathrm{pm}/\\sqrt{\\mathrm{Hz}}$ shot-noise limit, a roughly 100-fold sensitivity gain.","With an optical cavity of finesse $10^5$ at $\\lambda = 1.55\\,\\mu\\mathrm{m}$ and detection efficiency $\\eta = 0.75$, the paper's calculation puts ground-state cooling of a $6\\,\\mu\\mathrm{g}$ sphere at only $7 \\times 10^6$ photons/s, or $0.9\\,\\mathrm{pW}$.","Combined with the high mechanical quality factors already reported for levitated superconducting microspheres (up to $2.6 \\times 10^7$), this points toward quantum experiments with microgram-scale masses."],"supporting_citations":[{"why":"Supplies the quadrupole-trap frequency relation $f_i = \\sqrt{3/(8\\pi^2 \\mu_0 \\rho)}\\, b_i$ for levitated superconducting microspheres.","marker":"[20]"},{"why":"Demonstrated high-Q levitation and control of superconducting microspheres, the platform this work extends with optical readout.","marker":"[22]"},{"why":"Provides the closed-loop optical phase-measurement technique used to linearize the interferometer via AOM frequency tracking.","marker":"[33]"},{"why":"Supplies the quantum-noise formalism and the shot-noise limit $S_{zz} = \\lambda^2/(64\\pi^2 n_{\\mathrm{det}})$ quoted as 11 pm/√Hz.","marker":"[34]"},{"why":"Supplies the feedback-cooling and noise-squashing analysis used to interpret the cooled spectra.","marker":"[35]"},{"why":"Supplies the cavity-optomechanics framework used to project ground-state cooling requirements.","marker":"[36]"},{"why":"Supplies the probe-tone calibration method and the oscillator response equation used to convert volts to displacement.","marker":"[61]"},{"why":"Provides the quantum-cooperativity definition and the feedback-control framework used in the sensitivity analysis.","marker":"[12]"}],"fun_headline_variants":["Levitated superconductor cooled to 8 nm by light","Optical probe cools 6 µg microsphere to 8 nm","Sub-nm motion sensing tames a levitated superconductor","Feedback cooling of a levitated sphere to 8 nm"],"cache_read_input_tokens":2816,"weakest_assumption_plain":"The absolute displacement scale rests on the calibration assumption that the external probe field is uniform across the $100\\,\\mu\\mathrm{m}$ sphere and the trap gradient is linear, and that the interferometer lock's suppression factor is constant across the measurement band; if any of these fails, the quoted noise floor and cooled amplitude shift systematically.","fun_headline_variants_meta":{"raw":{"variants":["Levitated superconductor cooled to 8 nm by light","Optical probe cools 6 µg microsphere to 8 nm","Sub-nm motion sensing tames a levitated superconductor","Feedback cooling of a levitated sphere to 8 nm"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000153,"raw_usage":{"total_tokens":1000,"prompt_tokens":659,"completion_tokens":341,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":403,"completion_tokens_details":{"reasoning_tokens":281}},"tokens_in":403,"tokens_out":341,"duration_ms":4783,"temperature":1.0,"reasoning_tokens":281,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T19:56:26.406335+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Repeat the measurement with a polished sphere of the same mass: if the noise floor does not move substantially toward the $11\\,\\mathrm{pm}/\\sqrt{\\mathrm{Hz}}$ shot-noise limit, the surface-roughness explanation is not the whole story. Independently, calibrate the displacement signal with a method that does not assume a uniform probe field—e.g., a calibrated radiation-pressure force or an auxiliary SQUID readout—and check whether it agrees with the probe-tone calibration.","supporting_citations":[{"cited_title":"File and R","cited_arxiv_id":null,"evidence_quote":"Supplies the quadrupole-trap frequency relation $f_i = \\sqrt{3/(8\\pi^2 \\mu_0 \\rho)}\\, b_i$ for levitated superconducting microspheres."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Demonstrated high-Q levitation and control of superconducting microspheres, the platform this work extends with optical readout."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the closed-loop optical phase-measurement technique used to linearize the interferometer via AOM frequency tracking."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the quantum-noise formalism and the shot-noise limit $S_{zz} = \\lambda^2/(64\\pi^2 n_{\\mathrm{det}})$ quoted as 11 pm/√Hz."},{"cited_title":"Poggio, C","cited_arxiv_id":null,"evidence_quote":"Supplies the cavity-optomechanics framework used to project ground-state cooling requirements."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the probe-tone calibration method and the oscillator response equation used to convert volts to displacement."}],"review_version":1}