{"id":"938b698c-f653-4b51-8d9f-b96f188e70e6","arxiv_id":"1908.00610","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Using an ergodic ensemble of many diffusing colloids, the authors reconstruct radiation pressure force fields with a claimed resolution of 0.3 fN at a 99.7% confidence level.","lead":"This paper measures the tiny push of light on microscopic beads floating in water, and claims to detect forces as small as 0.3 femtonewtons. It works by treating many beads' random jiggling as one huge statistical sample, which makes the average motion much clearer.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Equation (13) with the paper's own N=2e5, Δt=1/120 s, and sqrt(2kBTγ)=8.56 N/sqrt(Hz) gives about 0.63 fN, not the advertised 0.3 fN.","rationale":"The paper's method is plausible and the dual-beam ergodicity checks are a real strength; I am not arguing the approach is invalid. However, the headline number is the central claim, and it is contradicted by the authors' own formula and parameters. A conditional acceptance should require either a corrected resolution calculation or an explicit statement of the N that yields 0.3 fN. The ergodicity-in-forced-mode concern raised by the reader is also legitimate and would make the resolution worse, not better, so it does not rescue the 0.3 fN claim. I disagree with the reader's choice of weakest assumption only because the arithmetic error is the most immediately decisive issue.","tokens_in":17121,"tokens_out":9198,"duration_ms":91795,"concrete_test":"Recompute F_min from Eq. (13) with N=2e5, Δt=1/120 s, m=3 and the authors' quoted sqrt(2kBTγ)=8.56 N/sqrt(Hz); if the result is about 0.63 fN rather than 0.3 fN, the abstract's headline resolution must be revised or the specific N basis for 0.3 fN must be provided.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing concern is numerical: Eq. (13) with the paper's own stated values does not produce the advertised 0.3 fN. Using m=3, N=2e5, Δt=1/120 s, and the text's sqrt(2kBTγ)=8.56 N/sqrt(Hz) gives F_min = 3 * 8.56e-15 / sqrt(2e5/120) ≈ 6.3e-16 N = 0.63 fN. Recomputing with the main-text parameters (T=298.65 K, η=0.88 mPa·s, d=940 nm) gives about 0.59 fN. To reach 0.3 fN would require N≈8e5, about four times the stated typical N. The abstract's central quantitative claim is therefore internally inconsistent. This is not a consensus disagreement but a straightforward arithmetic check. A separate, related concern is that Eq. (13) uses the zero-force thermal variance, while in the forced single-beam configuration the displacement ensemble also contains deterministic variation of F(z), so the actual resolution may be worse than even the corrected thermal value.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes and demonstrates a method for measuring radiation-pressure force fields acting on an overdamped colloidal suspension by averaging single-frame displacements over a very large ensemble built from all particles and all times. The central estimator is Eq. (3), the ensemble-averaged overdamped Langevin equation, and the resolution is given by Eq. (13), a standard error-propagation formula. The authors carefully verify the white, stationary, and ergodic character of the thermal noise in a dual-beam force-compensated mode, using Allan variance, the ergodic parameter, and comparison with the theoretical long-time law 4Δ/3T. In single-beam mode they reconstruct the Gaussian radiation-pressure force profile and report a resolution of 0.3 fN at the 99.7% confidence level.","tokens_in":17329,"tokens_out":8047,"duration_ms":76093,"significance":"If the central claims hold, the method provides a simple, high-sensitivity force measurement technique using standard video microscopy, with the strong feature that the ergodicity and noise properties are experimentally characterized rather than assumed. The ergodic-parameter data match the theoretical prediction without adjustable parameters, and the diffusion coefficient is independently calibrated; these are genuine strengths. However, the headline resolution of 0.3 fN is numerically inconsistent with the stated experimental parameters, and the resolution formula is derived for a zero-force configuration even though it is applied to forced single-beam measurements. These issues bear directly on the central quantitative claim of the paper and must be corrected before the result can be accepted.","major_comments":[{"comment":"The headline resolution of 0.3 fN is inconsistent with the stated parameters. Using the manuscript's own values N=2×10^5, Δt=1/120 s, m=3, and sqrt(2kBTγ)=8.56 N/√Hz, Eq. (13) gives ⟨F⟩min = 8.56×10^-15 × 3 / sqrt(2×10^5/120) ≈ 6.3×10^-16 N = 0.63 fN, not 0.3 fN. To reach 0.3 fN one would need N≈8×10^5, about four times the 'typical' N stated in the main text. The abstract's central quantitative claim is therefore internally inconsistent with the experimental parameters and must be corrected.","section":"Abstract and 'Radiation pressure force measurement' (Eq. 13)"},{"comment":"Eq. (13) uses the zero-force thermal variance σ(Δy_n)=√(2kBTΔt/γ), as explicitly stated in the sentence preceding Eq. (12). In the forced single-beam configuration, the displacement ensemble includes the deterministic contribution F(z_n)Δt/γ, which varies with z_n across the Gaussian beam. The total displacement variance is Var(Δy_n) = 2kBTΔt/γ + Var(F(z)Δt/γ), which exceeds the thermal value because the z-distribution of particles spans a region comparable to the waist w0. Consequently Eq. (13) underestimates the resolution of the actual force-field measurement. The paper must either demonstrate that the estimator in Eq. (10) achieves the thermal limit (e.g., by subtracting the known z-dependent force from each displacement) or recompute the resolution using the actual sample variance, as already done for the error bars in Eq. (62). The blue 'thermal limit' surfaces in Fig. 4(a) and the statement that a ~0.5 fN force was 'actually measured' at 99.7% confidence need to be re-evaluated in light of this.","section":"Radiation pressure force measurement, Eqs. (11)-(13)"},{"comment":"The stationarity and ergodicity tests are performed exclusively in the dual-beam force-compensated mode, as the authors note that free diffusion is unreachable in single-beam mode. The paper does not explicitly justify that the concatenated displacement ensemble in the forced single-beam mode remains identically distributed and stationary. In particular, the z-dependent force and the evolving z-distribution under radiation pressure and convection could introduce non-stationarity or additional correlations. The authors should either present a stationarity check for the forced-mode data or state clearly that the same properties are assumed to hold, together with the physical justification.","section":"Thermal noise and stationarity; Ergodicity (Secs. III-IV)"}],"minor_comments":[{"comment":"The limit in Eq. (6) is written as lim_{T/Δ→∞} δy_i^2(Δ) = ⟨δy_i^2(Δ)⟩; this is confusing because the left-hand side is a single-trajectory time average while the right-hand side is an ensemble average of such time averages. The statement should specify how the limit is taken and that the equality holds in probability for an ergodic process.","section":"Ergodicity, Eq. (6)"},{"comment":"The caption contains a typographical error: 'All ratio ρ(Δ)=⟨δy2(Δ)⟩/⟨y2(Δ)⟩)' has mismatched parentheses and should read ρ(Δ)=⟨δy^2(Δ)⟩/⟨δy^2(Δ)⟩' or equivalent.","section":"Fig. 3 caption"},{"comment":"The phrase 'power spectrum densitiy' should be 'power spectral density'.","section":"Appendix B"},{"comment":"The sentence 'it is possible to collect displacement values acquired from differenti trajectories at differenti k times' contains typos; 'differenti' should be 'different'.","section":"Ergodicity section"},{"comment":"The notation ⟨exp(−2(z_n−z_0)^2/w_0^2)⟩ is used without a formal definition; defining α_n = exp(−2(z_n−z_0)^2/w_0^2) and α = ⟨α_n⟩ would improve readability and clarify the estimator.","section":"Eq. (10)"}],"recommendation":"major_revision","confidential_remarks":"The numerical discrepancy in the headline resolution (0.3 vs. 0.63 fN) is a simple arithmetic check that the authors should have caught; it undermines trust in the abstract. The deeper issue of whether Eq. (13) applies to the forced configuration is conceptually important and could affect the validity of the claimed confidence intervals. If the authors can correct the numbers and clarify the applicability of the thermal-limit formula, the paper could be publishable. I would not recommend rejection if these issues are properly addressed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe genuinely new thing here is treating a dispersion of many freely diffusing colloids as one big ergodic ensemble: gather every single-step displacement from every particle, verify the noise is white and stationary via Allan variance and the 4Δ/3T ergodic parameter, then reconstruct radiation-pressure force fields from the ensemble-averaged drift. That is a clean twist on the usual single-particle optical-tweezer approach, and the ergodicity checks are careful and parameter-free. The agreement with Mie theory for F0 vs intensity and the zone-by-zone profiles is convincing. This is a paper worth reading for anyone who works on weak optical forces or stochastic thermodynamics.\n\nThe soft spot is the numbers in the abstract. Eq. (13) gives F_min = sqrt(2kBTγ) m / sqrt(NΔt). With their stated sqrt(2kBTγ)=8.56 N/√Hz, m=3, N=2×10^5, Δt=1/120 s, the result is about 0.63 fN, not 0.3 fN. Four times more statistics would be required to hit 0.3 fN. The paper's own \"0.5 fN actually measured\" line is consistent with the corrected arithmetic, so the abstract simply overstates the resolution by a factor ~2. That is an internal inconsistency, not a subtle scientific disagreement, and it should be fixed before publication.\n\nA second, milder concern: ergodicity is verified in the dual-beam (force-compensated) configuration. In the forced single-beam mode, the mean displacement depends on z because the Gaussian profile varies across the field of view. The paper bins by layers for the profile reconstruction, which mitigates this, and the decoupling of y from z is well-supported. Still, the resolution formula (13) implicitly assumes identically distributed zero-mean noise; applying it to the global ensemble might be too optimistic. This is more of a methodological caveat than a fatal flaw.\n\nOverall: the central idea is sound, the measurements look careful, and the only real problem is the headline number. I would send it to peer review and let the referees police the arithmetic. It deserves to be published, with corrections.","headline":"A solid ergodic-ensemble method for force mapping whose headline 0.3 fN number does not survive the paper's own error formula; the correct value with the stated statistics is roughly 0.6 fN.","tokens_in":17888,"tokens_out":2898,"would_cite":false,"duration_ms":26938,"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 demonstrates that pooling every Brownian step from many diffusing colloids measures radiation-pressure forces down to 0.3 femtonewton.","keywords":["radiation pressure","optical force measurement","Brownian motion","ergodicity","colloidal particles","overdamped Langevin equation","Allan variance","femtonewton sensitivity"],"falsifier":"Take two runs at the same laser power with very different total displacement counts and check that the mean force is unchanged while the error bars shrink as $1/\\sqrt{N}$; alternatively, split a single-beam run into early and late time blocks and test whether the mean displacement per frame drifts, since a drift would mean Eq. (3) averages a non-stationary signal and the resolution is not thermal-limited.","tokens_in":16912,"feed_emoji":"🔬","tokens_out":5898,"duration_ms":55262,"temperature":0.7,"pith_summary":"The paper reports that the radiation pressure exerted by a laser on colloidal particles in water can be measured and mapped with a resolution of about $0.3$ fN at a $99.7\\%$ confidence interval. The method treats every recorded single-step displacement, from every particle trajectory and every time, as one draw from a single statistical ensemble, then reads the average force from the ensemble-averaged overdamped Langevin equation, $\\langle F \\rangle = \\gamma \\langle \\Delta y \\rangle / \\Delta t$. The resolution formula $\\langle F \\rangle_{\\min} = \\sqrt{2 k_B T \\gamma}\\, m / \\sqrt{N \\Delta t}$ shows that the force floor falls as $1/\\sqrt{N}$, so the number of collected displacements, not the instrument, sets the sensitivity. The authors verify the needed white, stationary, ergodic noise through Allan variance, displacement covariance, mean-square displacement, and the ergodic parameter before applying the average. If correct, the method makes sub-femtonewton force-field reconstruction in fluids accessible with ordinary video microscopy.","feed_headline":"Brownian ensemble measures light's push at 0.3 femtonewton","feed_subtitle":"Pooling every step from thousands of diffusing colloids reaches the thermal noise floor of force detection.","key_machinery":"The load-bearing object is the concatenated displacement ensemble: all single-frame steps $\\{\\Delta y_i(t_k)\\}$ recorded from all tracked particles, at all times, treated as one statistical sample. Its validity is established by three checks: the Allan variance follows $\\sigma(\\tau)=\\sqrt{2k_B T/(\\gamma \\tau)}$ over three decades, the displacement covariance stays near zero, and the ergodic parameter $\\epsilon(\\Delta)=\\sigma^2(\\delta y_i^2(\\Delta))/\\langle \\delta y_i^2(\\Delta) \\rangle^2$ follows the free-Brownian prediction $4\\Delta/(3T)$. These checks are what turn Brownian noise from a measurement limit into a resource: the noise averages to zero over the ensemble, the force is read from the mean displacement, and the standard error shrinks as $1/\\sqrt{N}$.","core_discovery":"On the paper's own terms, the discovery is that an overdamped colloidal dispersion can act as a quantitative force sensor whose resolution is set by the total number of recorded displacements rather than by the stiffness of any trap or the sensitivity of any instrument. The argument reduces to the ensemble-averaged overdamped Langevin equation, $\\langle F \\rangle = \\gamma \\langle \\Delta y \\rangle / \\Delta t$, together with the thermal-resolution formula $\\langle F \\rangle_{\\min} = \\sqrt{2 k_B T \\gamma}\\, m / \\sqrt{N \\Delta t}$. Pooling every one-frame displacement from many trajectories at many times makes $N$ large enough to reach $0.3$ fN at the $99.7\\%$ confidence level, and the paper verifies the necessary white, stationary, ergodic character of the noise. It then reconstructs the Gaussian radiation-pressure profile $F_y(z) = F_0 \\exp(-2(z-z_0)^2/w_0^2)$ and cross-checks the measured maximum force against Mie-scattering calculations.","pith_inferences":["A direct test of the core scaling would be to vary $N$ over an order of magnitude at fixed laser power and confirm that the resolution follows $1/\\sqrt{N}$ down to the predicted thermal floor; the data hint at this but do not sweep $N$ systematically.","If stationarity holds in the forced single-beam mode, the same concatenation procedure could be applied to force fields with strong spatial gradients, provided the data are analyzed in blocks to check that the ensemble mean displacement does not drift.","The ergodic verification in the dual-beam mode does not by itself prove ergodicity under radiation pressure, since the forced mode includes z-dependent convection; extending the Allan-variance test to single-beam data after removing the mean drift would close this gap.","Because the resolution floor scales as $\\sqrt{T \\gamma}$ for fixed ensemble size, smaller particles or lower-viscosity fluids could push the detectable force below $0.1$ fN, at the cost of faster diffusion and shorter usable exposure times."],"forward_implications":["Force resolution is set by $\\sqrt{2 k_B T \\gamma}\\, m / \\sqrt{N \\Delta t}$, so any increase in the total number of recorded displacements, whether from more particles, longer recordings, or higher frame rates, directly lowers the detectable force.","The method reconstructs the full spatial profile of a force field, not just its maximum, by binning displacements in layers along the transverse axis, and the same logic extends to three dimensions with volumetric tracking.","Because the approach requires only stationary white thermal noise, it applies to any weak force on colloids in fluids, not only optical radiation pressure, including hydrodynamic, magnetic, or Casimir-type forces.","The agreement between the ensemble-averaged forces and Mie-scattering calculations provides a quantitative check that the measured mean displacement is a genuine force signal rather than a tracking artifact."],"supporting_citations":[{"why":"Supplies the tracking algorithm used to extract particle positions and build the displacement ensembles.","marker":"[20]"},{"why":"Provides the Allan variance method and its noise-classification relation used to establish the white stationary character of the thermal noise.","marker":"[22]"},{"why":"Gives the mean-square displacement fitting procedure used to extract diffusion coefficients.","marker":"[23]"},{"why":"Defines the ergodic parameter and its $4\\Delta/(3T)$ law, which the paper uses to verify ergodicity.","marker":"[24]"},{"why":"Provides the Mie-scattering calculation of radiation pressure against which the measured force values are compared.","marker":"[25]"},{"why":"Supplies the tracking-error corrections used for the mean-square displacement and the ergodic parameter.","marker":"[42]"}],"fun_headline_variants":["Ergodic colloids measure forces at 0.3 fN","0.3 fN force resolution from pooling Brownian steps","Ergodic ensemble pushes force sensitivity to 0.3 fN","Colloidal noise maps light's force at 0.3 fN","Thermal noise becomes a 0.3 fN force probe"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The resolution claim requires that every displacement pooled from all trajectories and times, including those recorded under the single-beam forced illumination, is an independent sample from the same zero-mean white thermal noise; stationarity and ergodicity are verified explicitly only in the dual-beam force-compensated mode, not in the measuring mode.","fun_headline_variants_meta":{"raw":{"variants":["Ergodic colloids measure forces at 0.3 fN","0.3 fN force resolution from pooling Brownian steps","Ergodic ensemble pushes force sensitivity to 0.3 fN","Colloidal noise maps light's force at 0.3 fN","Thermal noise becomes a 0.3 fN force probe"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000816,"raw_usage":{"total_tokens":3548,"prompt_tokens":891,"completion_tokens":2657,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":507,"completion_tokens_details":{"reasoning_tokens":2564}},"tokens_in":507,"tokens_out":2657,"duration_ms":17272,"temperature":1.0,"reasoning_tokens":2564,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:45:48.815845+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take two runs at the same laser power with very different total displacement counts and check that the mean force is unchanged while the error bars shrink as $1/\\sqrt{N}$; alternatively, split a single-beam run into early and late time blocks and test whether the mean displacement per frame drifts, since a drift would mean Eq. (3) averages a non-stationary signal and the resolution is not thermal-limited.","supporting_citations":[{"cited_title":"Magallanes and E","cited_arxiv_id":null,"evidence_quote":"Supplies the tracking algorithm used to extract particle positions and build the displacement ensembles."},{"cited_title":"Tinevez, N","cited_arxiv_id":null,"evidence_quote":"Provides the Allan variance method and its noise-classification relation used to establish the white stationary character of the thermal noise."},{"cited_title":"The particles imaged on our camera therefore diffuse along the x axis within a constant laser intensity throughout the experiments","cited_arxiv_id":null,"evidence_quote":"Gives the mean-square displacement fitting procedure used to extract diffusion coefficients."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the ergodic parameter and its $4\\Delta/(3T)$ law, which the paper uses to verify ergodicity."},{"cited_title":"Michalet, Phys","cited_arxiv_id":null,"evidence_quote":"Provides the Mie-scattering calculation of radiation pressure against which the measured force values are compared."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the tracking-error corrections used for the mean-square displacement and the ergodic parameter."}],"review_version":1}