{"id":"5ea67735-4354-4755-b351-98628f18a3be","arxiv_id":"2607.06746","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"CFD-DDA coupling demonstrates non-uniform UV scattering inside airborne saliva droplets that cannot be neglected, yielding the Dbouk-Yurkin correction to the Chick-Watson inactivation law.","lead":"A coupled CFD-DDA model shows that UV light scatters highly non-uniformly inside spherical and evaporated irregular saliva droplets, so classical dose-based virus kill rates overstate inactivation. The work supplies a corrected inactivation law and a design tool for UV air purifiers.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.5","headline":"The square-root map H(\\psi)=\\sqrt\\psi H_\\xi=1 that turns DDA low-intensity volume fractions into effective dose is undervived and uncalibrated, yet every quantitative Ns prediction rests on it.","rationale":"The Reader correctly isolates the undervived square-root correction as the single weakest link that converts a solid methodological advance (first CFD–DDA coupling for evaporating irregular saliva droplets) into quantitative survival numbers. No other assumption—optical constants, Z extremes, wall-interaction laws—is as directly multiplicative on every reported Ns. Because the paper supplies neither a derivation of the map nor a sensitivity study, the qualitative claim that scattering cannot be neglected stands, while the specific Dbouk–Yurkin numbers remain provisional. The concrete recomputation test above would settle the issue in a few hours of post-processing of the already-computed DDA fields; if the curves collapse, the concern evaporates; if they diverge, the CONDITIONAL verdict is confirmed and further calibration is required before design use. No stronger objection is needed; the Reader’s diagnosis is already the load-bearing one.","tokens_in":27151,"tokens_out":672,"duration_ms":7019,"concrete_test":"Recompute the entire set of Ns(t) curves in figures 19–24 three ways: (i) H_eff = \\sqrt\\psi H_\\xi=1 (paper), (ii) H_eff = \\psi H_\\xi=1, (iii) H_eff = \\langle\\xi\\rangle H_\\xi=1 using the full DDA intensity histogram. If any pair of curves differs by more than a factor of two in final Ns/Ntot for the same Z and flow rate, the square-root map is load-bearing and the quantitative Dbouk–Yurkin predictions cannot be trusted without independent calibration.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central quantitative claim is the Dbouk–Yurkin law (section 12, eqs. 54–55): Ns(\\psi)=N0 exp(-Z H(\\psi)) with H(\\psi)=\\sqrt\\psi H_\\xi=1, where \\psi is the DDA-derived volume fraction of the droplet interior with normalized intensity \\xi below a chosen threshold (0.05, 0.25, …). The paper never derives or calibrates the square-root functional form; it is simply asserted after the DDA intensity maps (figures 9–16). Because virions are treated as immobilized (diffusion analysis, section 6) and the local field is highly non-uniform, any map from the intensity histogram to an effective fluence is model-dependent. Replacing \\sqrt\\psi by \\psi, by a volume-averaged \\langle\\xi\\rangle, or by a thresholded integral would change the predicted log-reduction by factors of order 2–10 for the reported \\psi values (10–40 %). All CFD survival curves (figures 19–24) therefore inherit an arbitrary scaling whose magnitude is unknown. The qualitative statement that scattering is non-uniform remains intact; the numerical Ns values used for design do not.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The manuscript introduces a coupled Euler–Lagrange CFD–DDA platform that computes UV irradiance and droplet trajectories/evaporation (CFD) together with three-dimensional light-scattering fields inside spherical and irregularly shaped saliva droplets (DDA). It shows that the interior UV intensity is highly non-uniform, proposes the “Dbouk–Yurkin law” Ns(ψ)=N0 exp(−Z H(ψ)) with H(ψ)=√ψ Hξ=1 as a correction to the classical Chick–Watson law, and uses the platform to quantify the influence of initial size distribution, droplet number, airflow rate and wall-interaction boundary conditions on the number of surviving coronavirus copies under a fixed set of environmental parameters.","tokens_in":27463,"tokens_out":1090,"duration_ms":16162,"significance":"If the non-uniform scattering maps and a properly justified effective-dose correction hold, the work would supply the first fully three-dimensional multiphysics framework for UV air-purifier design and would demonstrate that scattering cannot be neglected for either spherical or evaporated irregular droplets. The DDA intensity fields themselves (Figs. 9–16) and the systematic parametric CFD study of size distributions, loadings and wall laws constitute a genuine advance over prior CFD-only treatments that assumed uniform fluence. The platform is therefore potentially useful for engineering optimization once the mapping from intensity histograms to effective fluence is placed on a firmer footing.","major_comments":[{"comment":"Section 12, Eqs. (54)–(55): the central quantitative claim rests on the undervived square-root map H(ψ)=√ψ Hξ=1 that converts the DDA-derived volume fraction ψ of low-intensity regions into an effective dose. No derivation, averaging argument or independent calibration is supplied; replacing √ψ by ψ, by a volume-averaged ⟨ξ⟩ or by a thresholded integral changes the predicted log-reduction by factors of order 2–10 for the reported ψ values (10–40 %). All subsequent Ns curves therefore inherit an arbitrary scaling.","section":"Section 12, Eqs. (54)–(55)"},{"comment":"Section 9.2 and Figs. 13–16: the definition of ψ itself depends on an arbitrary intensity threshold (ξ<0.05, 0.25, 0.5, 0.75). Different thresholds produce systematically different ψ(Ds) curves, yet the manuscript never shows how the choice propagates into the final Ns predictions or justifies a preferred cutoff on physical grounds (e.g., relative to the virion absorption cross-section).","section":"Section 9.2, Figs. 13–16"},{"comment":"Section 13 (Figs. 19–24): the quantitative survival curves are presented under the classical Chick–Watson form with Hξ=1; the Dbouk–Yurkin correction is not applied consistently to the same data sets. Consequently it is unclear whether the reported effects of size distribution, loading and wall law survive once the scattering correction is inserted.","section":"Section 13, Figs. 19–24"},{"comment":"No experimental validation or even order-of-magnitude comparison is offered for either the DDA interior intensity fields or the predicted Ns values. Given that the platform is advanced as a design tool, at least a limited comparison against existing bulk-liquid or aerosol UV-susceptibility data (or a clear statement of the validation path) is required for the quantitative claims to be load-bearing.","section":"Sections 9–13"}],"minor_comments":[{"comment":"The refractive index n=1.60 adopted for dried saliva is stated without citation or sensitivity study; a short justification or range would strengthen the DDA results.","section":"Section 9.2"},{"comment":"Several figure captions repeat nearly identical long parameter lists; condensing them would improve readability.","section":"Figs. 5–7, 25–27"},{"comment":"Typographical inconsistencies appear (e.g., “PFD-a” vs “PDF-a”, “Dbouk-Yurkin” hyphenation, missing spaces before units).","section":"Throughout"},{"comment":"The spin-down time estimate (Eq. 37) is useful but the assumption that rotation remains negligible for the entire residence time could be checked against the local shear rates extracted from the CFD fields.","section":"Section 7.3"}],"recommendation":"major_revision","confidential_remarks":"The square-root map is the single most load-bearing and least justified element; if the authors can supply a derivation or replace it by a volume-averaged fluence the paper becomes much stronger. The novelty claim of “first CFD–DDA platform” appears accurate on the basis of the cited literature. Scope is appropriate for a fluids/aerosol journal provided the quantitative law is tightened."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The real advance here is methodological: they couple a standard Euler–Lagrange CFD solver (evaporation, two-way drag, wall stick/escape, view-factor UV lamp) to DDA scattering fields inside both spheres and irregular dehydrated saliva shapes. Prior UV-air-purifier CFD (Marshall 2024, Sankurantripati 2025) treated droplets as uniform spheres and ignored internal scattering. The 3-D intensity maps (Figs. 9–16) show clear non-uniformity with low-ξ pockets spaced 0.1–0.5 µm, and the diffusion analysis (section 6) makes a solid case that virions are essentially immobilized on the residence-time scale. That qualitative claim holds up.\n\nWhat does not hold up is the quantitative step that turns those maps into design numbers. The Dbouk–Yurkin law (eqs. 54–55) simply asserts H(ψ)=√ψ H_ξ=1. No derivation, no calibration against bulk or aerosol data, no comparison to volume-averaged ξ or a thresholded integral. For the reported ψ values (10–40 %) that choice moves log-reduction by factors of a few. Every Ns curve (Figs. 19–24) therefore inherits an arbitrary scaling. Optical constants (n=1.60, μ a=0.1 cm⁻¹) and the two literature Z extremes are also free parameters. Code and data are “available on request,” so the platform is not independently checkable.\n\nCitation pattern is fine; self-cites are mostly earlier CFD work by the first author and standard DDA references by the second. The CFD and DDA modules themselves look competent.\n\nThis is for people who design or model UVGI systems and want a multiscale framework that finally includes scattering and morphology. It is not yet a calibrated design tool. I would send it to peer review: the coupling and the non-uniformity maps deserve referee time, but the square-root map and the lack of any experimental benchmark need to be fixed or heavily caveated before the numbers can be trusted.","headline":"First CFD–DDA coupling for evaporating irregular saliva droplets is real; the named square-root inactivation law is not yet derived or calibrated.","tokens_in":28083,"tokens_out":552,"would_cite":false,"duration_ms":7274,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"UV light scattering inside airborne saliva droplets is highly non-uniform and must be counted when predicting virus inactivation.","keywords":["UV scattering","UVC air purifiers","virus disinfection","saliva aerosols","computational fluid dynamics","discrete dipole approximation","Chick-Watson law","airborne coronavirus"],"falsifier":"Measure inactivation of a known number of virions inside single, size-controlled evaporating saliva droplets under a calibrated UVC beam and compare the survival curve to the Dbouk–Yurkin prediction with and without the square-root dose correction.","tokens_in":28032,"feed_emoji":"🦠","tokens_out":932,"duration_ms":21691,"temperature":0.7,"pith_summary":"Existing models of ultraviolet disinfection of viruses carried in airborne saliva droplets treat the UV dose as if the light field inside each droplet were uniform. This paper couples a fluid-dynamics solver that tracks evaporating, polydisperse saliva droplets in airflow with a discrete-dipole solver that computes the internal UV intensity field, and shows that both spherical and evaporation-distorted droplets develop strong intensity contrasts that leave large fractions of their volume poorly illuminated. The authors therefore replace the classical Chick–Watson survival law with a corrected “Dbouk–Yurkin” form that multiplies the nominal dose by a scattering-derived factor derived from those intensity maps. They then quantify how initial droplet size distribution, airflow rate, and wall-interaction rules change the final number of surviving coronavirus copies under fixed temperature and humidity. A reader who designs or evaluates UV air purifiers cares because the work claims that every efficiency estimate built on uniform-dose kinetics is systematically incomplete.","feed_headline":"UV scattering inside saliva droplets shields viruses from disinfection","feed_subtitle":"A CFD-DDA platform and corrected inactivation law show why uniform-dose models overestimate air-purifier kill rates.","key_machinery":"The CFD–DDA platform (Euler–Lagrange CFD for droplet transport and evaporation coupled to the discrete-dipole approximation for the internal UV field) together with the Dbouk–Yurkin law Ns(ψ)=N0 exp(-Z H(ψ)), H(ψ)=√ψ × Hξ=1, which folds the DDA-derived volume fraction of poorly illuminated regions into an effective dose.","core_discovery":"Infected saliva droplets—whether still spherical or already irregular from evaporation—experience highly non-uniform UV light scattering; the resulting low-intensity pockets significantly increase the number of surviving virus copies relative to models that assume uniform illumination, so scattering cannot be neglected in any realistic three-dimensional inactivation calculation.","pith_inferences":["The square-root map from low-intensity volume fraction to effective dose is an uncalibrated ansatz; single-droplet optical or inactivation experiments could fix or refute it.","The same CFD–DDA coupling could be re-run for far-UVC (222 nm) or multi-lamp geometries where angular incidence and inter-droplet scattering become first-order.","Because 100 nm virions diffuse only micrometres inside viscous, gelling saliva on purifier transit timescales, the intensity map essentially freezes each virion’s survival probability at its initial location."],"forward_implications":["Air-purifier designs that only lengthen residence time will still leave virions protected if droplets do not rotate or mix internally.","Efficiency claims based on bulk-liquid or uniform-dose Chick–Watson kinetics will overestimate inactivation for real polydisperse saliva aerosols.","Evaporation-driven shape change and the associated rise in refractive index further shield embedded viruses and must be modeled.","Wall–droplet interaction rules (stick versus escape) and the initial size distribution strongly alter predicted survivor counts, so geometry and operating conditions matter quantitatively.","Engineering approaches that induce droplet spin or local mixing are required to overcome internal optical shielding."],"fun_headline_variants":["UV scatter in saliva droplets creates virus-safe pockets","Non-uniform UV inside saliva shields airborne viruses","Droplet scattering raises surviving virus counts vs uniform models","Evaporating saliva leaves low-UV zones that protect virions","CFD-DDA shows saliva scatter must be modeled for true kill rates"],"cache_read_input_tokens":16512,"weakest_assumption_plain":"The conversion of the poorly illuminated volume fraction into an effective UV dose relies on a square-root formula that is introduced without derivation or independent experimental calibration.","fun_headline_variants_meta":{"raw":{"variants":["UV scatter in saliva droplets creates virus-safe pockets","Non-uniform UV inside saliva shields airborne viruses","Droplet scattering raises surviving virus counts vs uniform models","Evaporating saliva leaves low-UV zones that protect virions","CFD-DDA shows saliva scatter must be modeled for true kill rates"]},"model":"grok-4.5","effort":"low","cost_usd":0.005614,"raw_usage":{"total_tokens":1545,"prompt_tokens":866,"num_sources_used":0,"completion_tokens":85,"cost_in_usd_ticks":56140000,"prompt_tokens_details":{"text_tokens":866,"audio_tokens":0,"image_tokens":0,"cached_tokens":128},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":594,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":866,"tokens_out":85,"duration_ms":21985,"temperature":1.0,"reasoning_tokens":594,"cache_read_input_tokens":128,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-10T22:11:15.326975+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Measure inactivation of a known number of virions inside single, size-controlled evaporating saliva droplets under a calibrated UVC beam and compare the survival curve to the Dbouk–Yurkin prediction with and without the square-root dose correction.","supporting_citations":[],"review_version":1}