{"id":"ab8ffa35-3c83-4595-8424-329f55e625e3","arxiv_id":"1909.01608","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"A proposed high-frequency nanomechanical phonon-counting experiment could test spontaneous collapse models down to a collapse rate of 1e-12 per second, covering both Adler's and Bassi et al.'s predicted bounds.","lead":"The paper designs a tabletop experiment using a tiny vibrating crystal cooled to near absolute zero, with light pulses that count single vibrations, to detect the faint heating predicted by spontaneous collapse theories. If the device parameters can be achieved, it could for the first time test the predicted lower bounds on the collapse rate and possibly reveal the physical origin of the collapse process.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Table I's λc=1e-12 s^-1 rests on simultaneously assuming a 10x higher g0 and ~50x narrower optical linewidth than any demonstrated device; recalculating with demonstrated parameters pushes the minimum testable rate far above the Bassi lower bound.","rationale":"The reader's weakest-assumption note accurately identifies the joint g0/κ0 extrapolation as the critical hinge. I checked the internal arithmetic of Table I: the individual λc,i entries are consistent with the quoted noise rates and Dη=561, and the master-equation results are internally coherent. The strongest claim, however, is not that this device exists; it is that a specific, not-yet-demonstrated parameter point would enable a test at λc=1e-12 s^-1. The noise budget is balanced: optomechanical phonons and dark counts contribute comparable equivalent rates, so the conclusion is not robust to the assumed device parameters. A recalculation with the demonstrated 575 MHz linewidth and ~1 MHz coupling degrades the resolved-sideband suppression and conversion efficiency simultaneously, which would raise λc,min well above the level needed to cover the lower edge of Bassi's bound. I do not see a separate internal inconsistency that would overturn the paper's logic: the apparent κp,ex versus κp factor in the Rom formula in the main text appears to be a typo, since the supplemental calculation and Table I use ar n_p κp = ηpΓ; if anything, the corrected formula would lower the optomechanical noise estimate. Nor do I regard the absence of a formal statistical-confidence treatment as decisive, because one can integrate longer and the paper's stated time-to-one-event is not the same as a full exclusion. Thus the appropriate verdict remains CONDITIONAL, unchanged from the reader.","tokens_in":22260,"tokens_out":28614,"duration_ms":303320,"concrete_test":"Recompute Table I with the current state-of-the-art parameters from Ref. [38] (κp,0=κs,0=2π×575 MHz, g0/2π≈1.15 MHz), holding all other assumptions — Ω, Γ, T, ηp=0.01, filter and detector parameters — fixed, using the paper's master-equation solver. If the resulting λc,min exceeds the Bassi lower edge (≈10^-12 s^-1) by more than an order of magnitude, then the claimed coverage of the Bassi and Adler bounds is contingent on the simultaneous theoretical improvements, not on the demonstrated device class. If the numerical code is not made available, an equivalent estimate follows from ηom≈(4g0/κ)^2 in the bad-cavity limit and pcounterrot∝(κ/Ω)^2.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The headline minimum testable collapse rate is set by the sum of two comparable noise terms: measurement-induced optomechanical phonons (λc,om = 3.5×10^-13 s^-1) and coincidence dark counts (λc,det = 6.7×10^-13 s^-1). The optomechanical term is obtained from a master-equation calculation whose inputs are g0/2π = 11.5 MHz and κp,0 = κs,0 = 2π×9.2 MHz — a tenfold increase in g0 and a roughly fiftyfold reduction in optical linewidth relative to the demonstrated device [38], both taken from theoretical designs [39,40]. The conversion efficiency ηom = 0.32 and the spurious-phonon probabilities (which scale as (Ω/κ)^2) depend on these ratios, and the paper assumes both improvements are simultaneously realizable in a single device. If either falls short, the signal efficiency drops while the measurement-induced phonon background rises; with the demonstrated κ/2π≈575 MHz and g0/2π≈1.15 MHz, (Ω/κ)^2 is degraded by more than three orders of magnitude and ηom drops by a comparable factor, pushing λc,min well above the 10^-12 s^-1 lower edge of Bassi et al.'s bound. The paper explicitly calls these parameters 'challenging but plausible' and offers no measured data or error budget around them; the central claim is therefore conditional on an unquantified joint extrapolation.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript proposes a phonon-counting optomechanical protocol to test spontaneous collapse models, in particular Continuous Spontaneous Localization (CSL), using a gigahertz nanomechanical resonator at millikelvin temperatures. In the proposed three-mode optomechanical system, a weak probe mode and a signal mode are coupled by the mechanical resonator; a collapse-induced phonon enables anti-Stokes scattering of a probe photon into the signal mode. Signal photons are spectrally filtered, nonlinearly downconverted into photon pairs, and detected in coincidence to suppress dark counts. The authors estimate the CSL-induced phonon flux via a geometrical factor D=5.1×10^5, compute the thermal phonon flux analytically, solve a Born-Markov master equation for measurement-induced phonon noise, estimate photoabsorptive heating, and evaluate coincidence dark counts. With the assumed device parameters (g0/2π=11.5 MHz, κp,0=κs,0=2π·9.2 MHz, Ω/2π=5.3 GHz, T=10 mK), they report a minimum testable collapse rate λc=1.0×10^−12 s^−1, which would cover both Bassi et al.'s and Adler's proposed lower bounds; with an array of N~10^4 devices they estimate a measurement time of about two days. An alternative quadratic-coupling readout is also analyzed.","tokens_in":22555,"tokens_out":12497,"duration_ms":126433,"significance":"If the projected device parameters are realized, this is an important and timely proposal. It directly targets the main limitation of previous mechanical collapse tests—thermal noise—by combining a high-frequency resonator, cryogenic cooling, and quantum measurement techniques, and it would, for the first time, close the gap between measured upper bounds and proposed lower bounds for CSL. The noise budget is carefully constructed: the master-equation calculations of measurement-induced phonons, the explicit comparison of all noise sources in Table I, and the coincidence-counting scheme for dark-count suppression are genuine strengths. The protocol is also falsifiable in the sense that it specifies a concrete device, a concrete detection chain, and a concrete expected signal rate. However, the headline sensitivity is conditional on a joint extrapolation of optomechanical parameters that has not been demonstrated in a single device, so the contribution is best understood as a design study with a clear roadmap rather than a demonstrated experimental sensitivity.","major_comments":[{"comment":"The central claim that the protocol can test Bassi et al.'s and Adler's lower bounds rests on the simultaneous assumption of g0/2π=11.5 MHz and κp,0=κs,0=2π·9.2 MHz, a tenfold increase in single-photon coupling and a roughly fiftyfold reduction in optical linewidth relative to the demonstrated device of Ref. [38], with both values taken from theoretical predictions [39,40]. These parameters enter the conversion efficiency ηom and the measurement-induced phonon background, which scales as (Ω/κ)^2. With the demonstrated values (g0/2π≈1.15 MHz, κ/2π≈575 MHz), the resolved-sideband suppression is degraded by more than three orders of magnitude and ηom drops correspondingly, pushing the minimum testable collapse rate well above the Bassi et al. lower edge. The manuscript states that these improvements are 'challenging but plausible' but provides no sensitivity analysis or error budget. I request a quantitative analysis of λc,min as a function of g0 and κ (and, where relevant, detector dark counts), together with a discussion of whether the two predicted improvements are compatible in a single device.","section":"Table I; 'Feasibility and alternative parameter regimes'"},{"comment":"The absorption-heating estimate relies on an unproven model assumption: 'any discrete photon absorption event is expected to create a fixed amount of heat' once the interval between events exceeds the THz dissipation time. The numerical input n_abs,1=1.9×10^−9 is obtained by extrapolating from Refs. [17,34] with n_abs=10 at n_cav=1; no derivation or uncertainty is given. This term is small in the main protocol (Table I), so it does not by itself change the headline λc,min, but the paper's broader claim that optical absorption noise can be brought below the collapse lower bounds depends on this estimate. In the quadratic-coupling section the same model yields n_abs≈5 and a noise rate ≈3 s^−1, about seven orders of magnitude above the Bassi flux, and the proposed pulsed regime is only sketched. The authors should either derive this model or explicitly label it as an assumption, and quantify the impact of its uncertainty on both the main protocol and the quadratic-coupling alternative.","section":"Methods, 'Optical absorption heating'; Supplemental Note 2, 'Noise phonons due to photoabsorptive heating'"}],"minor_comments":[{"comment":"Reference [41] is given as 'See Supplemental Material at []' with an empty URL; the placeholder should be completed with a DOI or a permanent link, since the Supplemental Material is an integral part of the noise budget.","section":"Reference list; main text Ref. [41]"},{"comment":"The filter-transmission formula pf(Ω)=ηf[1+(4/κf)^2 sin^2(πΔ/ωfsr)]^−1 is dimensionally inconsistent as written if κf is a linewidth; the term (4/κf)^2 would carry units of time squared. Please clarify whether κf denotes a finesse or a normalized linewidth, and verify that the quoted pf=3.5×10^−10 follows from the stated cavity parameters.","section":"Supplemental Note 2, Eq. (10)"},{"comment":"Both references are listed as 'Calculations for other experiments to be published elsewhere', which is not a citable reference; these should be replaced by actual publications or removed.","section":"References [80] and Supplemental Ref. [26]"},{"comment":"The notation λc is used both for the CSL collapse rate and for the minimum testable collapse rate (e.g., 'λc=∑iλc,i' in the Table I caption). Please introduce an explicit symbol such as λc,min to avoid ambiguity.","section":"Table I and surrounding text"},{"comment":"The dark-count parameters Rd=3.5 s^−1 and τc=30 ps are attributed to a private communication with PhotonSpot; since coincidence dark counts contribute a substantial share of the noise budget (λc,det=6.7×10^−13/N), a publicly available datasheet or published characterization would make the proposal more robust.","section":"Detector parameters, Ref. [47]"},{"comment":"There are several typographical and typesetting issues: 'The later approach' should read 'the latter approach' (Introduction); the Fig. 3 caption begins with the garbled 'Ea) b) c)'; and the symbol 'greaterorsimilar' appears in the Methods section where a typeset ≳ is intended.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The paper is a well-structured proposal whose main result is explicitly conditional on a joint extrapolation of optomechanical parameters. In my view, the central physics and noise-budget analysis are sound, but the abstract and summary overstate the capability by presenting λc=1.0×10^−12 s^−1 as a near-term testable sensitivity rather than a target for a specific, not-yet-demonstrated device. Requiring a sensitivity analysis and a clearer statement of the parameter conditions would substantially strengthen the manuscript and would make the claim appropriately conditional. I would not reject the paper; the proposed revision is feasible within the manuscript's scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nRead Forstner et al. The genuinely new thing is combining high-frequency (5.3 GHz) nanomechanical resonators, near-ground-state operation, and phonon counting with coincidence dark-count suppression into a CSL test. The paper works out a self-consistent noise budget: thermal phonons are exponentially suppressed at 10 mK, and Table I gives the two dominant backgrounds as measurement-induced optomechanical phonons (3.5e-13 s^-1 in lambda_c units) and coincidence dark counts (6.7e-13 s^-1). The master-equation treatment of the three-mode optomechanics is appropriate, and the quadratic-coupling alternative is a useful bonus. If the device reaches the assumed parameters, lambda_c,min = 1.0e-12 s^-1, enough to close the gap to Bassi et al.'s lower edge and, for colored collapse models, potentially reach Adler's region. That is a real step in capability, not a repackaging of older proposals.\n\nThe soft spot is exactly the one the stress-test note names. The headline sensitivity depends on two simultaneous improvements over current devices: g0/2pi = 11.5 MHz (tenfold) and intrinsic optical linewidths of 2pi x 9.2 MHz (~fiftyfold narrower than the demonstrated loaded linewidth). Both are taken from theoretical predictions, not measurements, and the paper has no error budget around them. Because the measurement-induced phonon background scales like (Omega/kappa)^2 and the conversion efficiency also degrades with lower g0/kappa, failing either assumption moves lambda_c,min substantially above the Bassi bound. The paper says the parameters are 'challenging but plausible'; that is honest, but it means the central claim is conditional. Two smaller caveats: the dark-count rate is a private communication, and the absorption-heating estimate is extrapolated into a regime where its power-law relation is expected to break down; the paper acknowledges the latter. No code is shipped with the numerics.\n\nI would send this to a serious referee. It is a careful design study that could shape the next generation of collapse tests, and the authors are clear about what remains to be demonstrated. The referee should ask for a sensitivity plot over g0 and kappa and for a more quantitative feasibility argument than 'challenging but plausible.' Overall: worth engaging, not yet a conclusive test.","headline":"A careful, well-scoped design study for a CSL test, but the headline reach into Bassi/Adler territory hinges on two unmeasured optomechanical improvements that the paper itself flags as challenging.","tokens_in":23154,"tokens_out":3417,"would_cite":true,"duration_ms":35132,"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":"The paper proposes a phonon-counting optomechanical protocol that could test spontaneous-collapse models down to a collapse rate of $\\lambda_c = 10^{-12}\\,\\mathrm{s}^{-1}$, enough to cover the proposed lower bounds on continuous…","keywords":["spontaneous wavefunction collapse","continuous spontaneous localization","optomechanics","phonon counting","nanomechanical resonator","collapse rate bounds","coincidence detection","quantum measurement problem"],"falsifier":"Build the proposed photonic-phononic crystal and measure its optical linewidth and single-photon coupling at 10 mK. If $\\kappa_{p,0}=\\kappa_{s,0}$ is not near $2\\pi\\times9.2\\,\\mathrm{MHz}$ or $g_0/2\\pi$ is not near $11.5\\,\\mathrm{MHz}$, the predicted $\\lambda_c=1.0\\times10^{-12}\\,\\mathrm{s}^{-1}$ is not reachable. If the parameters are met, run the array of $10^4$ resonators for two days: a coincidence rate significantly above the predicted total background of $5.6\\times10^{-10}\\,\\mathrm{s}^{-1}$ would indicate collapse-induced phonons, while a null result at that background would confirm the projected bound.","tokens_in":21975,"feed_emoji":"⚛️","tokens_out":10414,"duration_ms":84705,"temperature":0.7,"pith_summary":"Spontaneous collapse theories modify quantum mechanics with a stochastic nonlinear term, predicting that isolated mechanical oscillators should heat up at a rate set by a collapse parameter $\\lambda_c$. The best existing tests cannot resolve the tiny predicted heating because thermal noise overwhelms the signal and the oscillators are too large and slow. This paper proposes a gigahertz nanomechanical resonator, cooled near its ground state in a dilution refrigerator, whose collapse-induced phonons are read out one at a time through an optomechanical Raman transition. With noise from heat, probe light, measurement backaction, and detector dark counts suppressed, the protocol reaches a minimum testable collapse rate of $\\lambda_c = 1.0\\times10^{-12}\\,\\mathrm{s}^{-1}$, which would cover both proposed lower bounds on $\\lambda_c$ from latent-image formation and human-eye collapse. If realised, this would be the first experiment able to decisively test whether continuous spontaneous localization resolves the measurement problem.","feed_headline":"A phonon counter could finally test wavefunction collapse","feed_subtitle":"Gigahertz resonator plus coincidence counting reaches collapse rates of 10^-12 s^-1.","key_machinery":"The central mechanism is a three-mode optomechanical phonon counter: a probe mode at $\\omega_p$, a signal mode at $\\omega_s=\\omega_p+\\Omega$, coupled by a gigahertz mechanical resonator. A collapse-induced phonon converts a probe photon to a signal photon via anti-Stokes Raman scattering, and the signal photon is filtered, downconverted to a pair, and counted in coincidence. The argument is carried by the master equation for the three modes, which yields the conversion efficiency $\\eta_{\\mathrm{om}}=0.32$, and by the estimate that measurement-induced phonons are suppressed by $(\\Omega/\\kappa_p)^2$ while dark counts are suppressed as $R_{d,1}R_{d,2}\\tau_c$. This combination reduces every noise channel below the predicted collapse phonon flux for $\\lambda_c\\ge10^{-12}\\,\\mathrm{s}^{-1}$.","core_discovery":"At the level of a single phonon, continuous spontaneous localization (CSL) predicts that a mechanical resonator is heated by a collapse noise field at rate $\\dot n_c = \\lambda_c D$, where $D$ is a geometry factor; for the proposed 136 fg photonic-phononic crystal $D = 5.1\\times10^5$. The scheme counts these phonons by driving a probe optical mode and detecting anti-Stokes photons emitted into a signal mode at frequency $\\omega_s=\\omega_p+\\Omega$, with $\\Omega/2\\pi=5.3\\,\\mathrm{GHz}$. After filtering, each signal photon is nonlinearly downconverted to a pair and registered only upon coincidence, suppressing dark counts to $3.7\\times10^{-10}\\,\\mathrm{s}^{-1}$. The authors show, by solving the master equation, that thermal phonons, measurement-induced phonons, absorption heating, and leaked probe photons each contribute less than one coincidence per month, giving a total background of $5.6\\times10^{-10}\\,\\mathrm{s}^{-1}$ and hence a minimum testable collapse rate $\\lambda_c = 1.0\\times10^{-12}\\,\\mathrm{s}^{-1}$. That rate is sufficient to test both the latent-image bound and the human-eye bound, and, because the resonator frequency lies near the conjectured cutoff $\\Omega_{\\mathrm{csl}}/2\\pi\\sim10^{10}$–$10^{11}\\,\\mathrm{Hz}$, the same apparatus could distinguish a physical origin of collapse from white-noise models.","pith_inferences":["The quadratic-coupling variant described in the paper suggests a nearer-term path: zero-point quadratic couplings near $2\\pi\\cdot245\\,\\mathrm{Hz}$ already exceed the $2\\pi\\cdot28\\,\\mathrm{Hz}$ needed for the human-eye bound, so the decisive milestone is demonstrating suppression of linear coupling to $g/2\\pi\\lesssim0.1\\,\\mathrm{Hz}$.","The coincidence-downconversion readout is a generic low-dark-count phonon detector; the same trick could be applied to searches for rare events such as dark-matter or axion interactions, where detector dark counts are the limiting background.","If the required linewidth and coupling improvements are not reached, the minimum testable rate degrades through the conversion efficiency and the measurement-induced phonon background; quantifying that trade-off as $\\lambda_c \\propto \\kappa^2/(g_0^2)$ would show which device parameter to push first.","Because the signal rate scales linearly with the geometry factor $D$ while the background does not, resonators engineered for larger $D$ (denser, heavier, or shaped to maximise the CSL decoherence operator) would extend the same protocol to smaller $\\lambda_c$, at the cost of higher thermal occupation."],"forward_implications":["If the protocol is built, it would close the gap between current experimental upper bounds on $\\lambda_c$ and the proposed lower bounds, giving the first conclusive test of CSL as a resolution of the measurement problem.","The thermal background at 10 mK for a 5.3 GHz oscillator is exponentially suppressed, so collapse-induced heating would appear above background; existing lower-frequency experiments cannot reach this regime.","With an array of $N\\sim10^4$ resonators, the time to probe either proposed bound drops from more than 57 years to about two days, and the dark-count limit becomes negligible.","At frequencies near the conjectured cutoff, the protocol could distinguish coloured-noise collapse from white-noise CSL, and could support or rule out a cosmological origin for the collapse field.","The same setup is predicted to bound classical-channel gravity models roughly an order of magnitude more strongly than previous experiments."],"supporting_citations":[{"why":"Supplies the demonstrated device parameters: 5.3 GHz resonance, 108 mHz damping, 136 fg effective mass, used throughout the protocol.","marker":"[38]"},{"why":"Provides the theoretical scattering-limited intrinsic optical linewidths $\\kappa_{p,0}=\\kappa_{s,0}=2\\pi\\cdot9.2\\,\\mathrm{MHz}$ assumed for both modes.","marker":"[39]"},{"why":"Predicts the tenfold-improved single-photon optomechanical coupling $g_0/2\\pi=11.5\\,\\mathrm{MHz}$ that sets the conversion efficiency.","marker":"[40]"},{"why":"Defines the human-eye lower bound on $\\lambda_c$ that the proposed sensitivity is designed to cover.","marker":"[20]"},{"why":"Defines the latent-image lower bound on $\\lambda_c$, the larger bound the scheme also covers.","marker":"[21]"},{"why":"Supplemental material containing the geometry factor $D$, efficiency calculations, and noise-rate derivations reported in the text.","marker":"[41]"},{"why":"Supported the filter-cavity parameters used to suppress leaked probe photons to $p_f = 3.5\\times10^{-10}$.","marker":"[45]"},{"why":"Demonstrated near-unit single-photon-to-pair downconversion efficiency $\\eta_\\chi = 0.95$ used in the coincidence readout.","marker":"[46]"},{"why":"Provides the commercial single-photon counter dark-count rate and timing jitter used to estimate coincidence dark counts.","marker":"[47]"},{"why":"Supplies the photothermal absorption model relating absorbed photons to mechanical phonon occupancy, bounding absorption heating.","marker":"[33]"}],"fun_headline_variants":["Gigahertz resonator counts phonons to test wavefunction collapse","Phonon coincidence scheme reaches collapse rate 10^-12 s^-1","Nanomechanical probe targets collapse origin with single phonons","Proposed test: phonon counter vs wavefunction collapse","Single-phonon counting could reveal collapse mechanism"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole projected sensitivity rests on two device improvements that have not yet been shown in the lab: a tenfold stronger photon–phonon coupling and a fiftyfold narrower optical resonance; if real devices deliver less, the claimed coverage of the collapse bounds fails.","fun_headline_variants_meta":{"raw":{"variants":["Gigahertz resonator counts phonons to test wavefunction collapse","Phonon coincidence scheme reaches collapse rate 10^-12 s^-1","Nanomechanical probe targets collapse origin with single phonons","Proposed test: phonon counter vs wavefunction collapse","Single-phonon counting could reveal collapse mechanism"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000221,"raw_usage":{"total_tokens":1456,"prompt_tokens":957,"completion_tokens":499,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":573,"completion_tokens_details":{"reasoning_tokens":415}},"tokens_in":573,"tokens_out":499,"duration_ms":5143,"temperature":1.0,"reasoning_tokens":415,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T05:12:30.381244+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Build the proposed photonic-phononic crystal and measure its optical linewidth and single-photon coupling at 10 mK. If $\\kappa_{p,0}=\\kappa_{s,0}$ is not near $2\\pi\\times9.2\\,\\mathrm{MHz}$ or $g_0/2\\pi$ is not near $11.5\\,\\mathrm{MHz}$, the predicted $\\lambda_c=1.0\\times10^{-12}\\,\\mathrm{s}^{-1}$ is not reachable. If the parameters are met, run the array of $10^4$ resonators for two days: a coincidence rate significantly above the predicted total background of $5.6\\times10^{-10}\\,\\mathrm{s}^{-1}$ would indicate collapse-induced phonons, while a null result at that background would confirm the projected bound.","supporting_citations":[{"cited_title":"Penrose, General Relativity and Gravitation 28, 581 (1996)","cited_arxiv_id":null,"evidence_quote":"Defines the human-eye lower bound on $\\lambda_c$ that the proposed sensitivity is designed to cover."},{"cited_title":"Kafri, J","cited_arxiv_id":null,"evidence_quote":"Defines the latent-image lower bound on $\\lambda_c$, the larger bound the scheme also covers."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the photothermal absorption model relating absorbed photons to mechanical phonon occupancy, bounding absorption heating."}],"review_version":1}