{"id":"31f66ad7-84a1-4c88-881a-ebeb5105dd5d","arxiv_id":"2412.03196","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A one-month HIAF simulation projects branching-ratio upper limits of about 10^-9 (e+e-) and 10^-6 (pi pi) for eta decays to a dark scalar, with parameter sensitivities beyond current bounds.","lead":"This paper simulates a proposed experiment at a new Chinese accelerator facility (HIAF) that would search for a hypothetical 'dark scalar' particle appearing in rare decays of the η meson. It estimates how many signal events could be seen in one month and what limits the experiment could set on dark matter portal models.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"MC scale factor is internally inconsistent: 13M inelastic events x 0.76% x 1e5 gives ~1e10 eta, not 5.9e11; matching N_eta requires ~6e6, so projected limits may be off by sqrt(60).","rationale":"The reader's weakest assumption correctly identified the event-yield estimate as the linchpin of the projected sensitivities. I agree with that overall location, but I find a sharper and more internal problem: the stated Monte-Carlo scaling factor is inconsistent with the quoted N_eta. This is load-bearing because Eq. (4) uses the scaled background counts directly, and the mismatch, if real, would bias the upper limits by roughly sqrt(60). The paper deserves credit for using established simulation tools (GiBUU, Geant4/FairRoot) and for explicitly acknowledging that systematic uncertainties are not quantified. However, the scaling inconsistency means the numerical results cannot be independently reproduced from the text alone. Because this is a checkable arithmetic/code issue rather than a demonstrated physics flaw, the appropriate verdict remains CONDITIONAL: the central claim is plausible but requires the authors to supply the correct scale factor and confirm that the projected limits are not affected by the factor-60 discrepancy.","tokens_in":13236,"tokens_out":9226,"duration_ms":87883,"concrete_test":"Ask the authors to report the exact scale factor used to produce Figs. 14-17, and recompute the branching-ratio upper limits and sensitivity curves using scale factor = 5.9e11 / (0.0076 x 13e6) ~ 6e6 instead of 1e5. If any quoted upper limit changes by more than a factor of about 2, the central sensitivity projections are not robust. Also verify the 100 MHz assumption against L = 1e35 cm^-2 s^-1 and sigma = 0.7 mb: if the actual rate is 70 MHz, N_eta drops to about 4.1e11 and all quoted limits degrade correspondingly.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section IV states that 13 million inelastic p-A events were simulated; with the quoted 0.76% eta probability this yields approximately 1.0e5 eta events. The text then says both background distributions and eta samples were scaled up by 'a scale-up factor of around 10^5', giving about 1.0e10 eta, a factor of roughly 60 below the quoted N_eta = 5.9e11. Consistency requires a scale factor of 5.9e11 / (13e6 x 0.0076) = 6.0e6. Since Eq. (4) computes Br upper limit = 3 sqrt(N_bg) / (N_eta epsilon), a background histogram scaled by 1e5 instead of 6e6 has N_bg too small by a factor of about 60, so the upper limits would be underestimated by sqrt(60) ~ 7.7 in background-dominated bins. This would directly inflate the quoted sensitivities (sin^2 theta ~ 1e-1, g_u ~ 1e-6) by the same factor. In addition, the stated L = 1e35 cm^-2 s^-1 and sigma = 0.1 x A mb with A = 7 give a 70 MHz inelastic rate, below the assumed 100 MHz cap, so even N_eta appears internally optimistic by about 1.4. As written, the numerical projections cannot be reproduced without resolving this factor. If the actual code used 6e6 and the text contains a typo, the concern is minor; but the paper should state the correct scale factor explicitly.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents a Monte Carlo simulation study of a proposed fixed-target experiment at the HIAF accelerator (the 'Huizhou η factory') to search for a light dark scalar particle S produced in η → S π0 decays, with S decaying to either e+e− or π+π−. The η mesons are produced in 1.8 GeV p-7Li collisions simulated with GiBUU, and the detector response is modeled with a GEANT4-based package (ChnsRoot). The authors estimate a one-month η yield of 5.9×10^11, determine detection efficiencies around 40%, and compute projected 99% CL upper limits on Br(η→Sπ0) as a function of m_S, reaching ~10^-9 in the e+e− channel and ~10^-6 in the π+π− channel. From these limits they derive projected sensitivities to the Higgs-mixing parameter sin^2θ in the minimal scalar model and to the coupling g_u in the hadrophilic scalar model, benchmarking against REDTOP and existing experimental constraints.","tokens_in":13609,"tokens_out":14858,"duration_ms":121937,"significance":"If the quantitative projections are correct, the proposed experiment would be competitive with and in some mass regions superior to existing and planned searches for light scalar particles produced in η decays. The paper's strengths include a full simulation chain from production to detector response, explicit and falsifiable projections benchmarked against other experiments, and an honest statement of the main limitations (e.g., no quantitative treatment of systematic uncertainties). The qualitative conclusions are interesting and timely. However, the numerical projections are directly tied to a set of assumed beam, target, and scaling parameters, and one of these parameters — the Monte Carlo scale-up factor — is internally inconsistent, which directly affects the quoted upper limits. Until that inconsistency is resolved, the specific sensitivity numbers cannot be considered reliable.","major_comments":[{"comment":"The scale-up factor stated in Section IV is internally inconsistent with the quoted η yield. The text says that 13 million inelastic p-A events were simulated, the η probability is 0.76%, and 'both the background distributions and the number of produced η samples' were scaled up by a factor of 'around 10^5'. This yields approximately 1.0×10^10 η events, a factor of about 60 below N_η = 5.9×10^11 quoted in the same section. Consistency requires a scale factor of 5.9×10^11 / (13×10^6 × 0.0076) ≈ 6.0×10^6. Since Eq. (4) computes the upper limit as 3√N_bg / (N_η ε), using a background histogram scaled by 10^5 instead of 6×10^6 underestimates N_bg by a factor of about 60 and therefore underestimates the branching-ratio upper limits by √60 ≈ 7.7 in background-dominated bins. This directly affects the sensitivities quoted in Sections V.D and V.E (e.g., sin^2θ ~ 10^-1 and g_u ~ 10^-6). The authors must correct the scale factor or explicitly state the actual factor used; if the code used approximately 6×10^6, the text should be corrected accordingly.","section":"Section IV and Eq. (4)"},{"comment":"The derivation of N_η = 5.9×10^11 is not reproducible from the inputs stated in Section IV. The paper quotes a luminosity of 10^35 cm^-2 s^-1, a p-A η-production cross section of about 0.1×A mb (with A=7), a 100 MHz inelastic event-rate cap, and a 30% duty factor for one month, but it does not give the formula used to obtain N_η. Using the quoted luminosity and η-production cross section directly gives N_η ≈ 5×10^13, whereas using the 100 MHz cap together with the 0.76% η probability from GiBUU gives 5.9×10^11. The paper should present the explicit expression for N_η and clarify whether the quoted luminosity is an upper bound that is not actually achieved in a rate-limited running scenario. Without this, the numerical projections cannot be reproduced by an independent reader.","section":"Section IV"},{"comment":"Section VI states that 'the experimental uncertainties are not evaluated quantitatively' and asserts that systematic uncertainties are 'at the level of several percentages' without showing their impact on the results. Because Eq. (4) uses only the statistical Poisson fluctuation 3√N_bg, and because the background distributions in Figs. 14 and 15 contain large event counts, systematic uncertainties in the background normalization or shape would likely dominate the statistical uncertainties. A quantitative estimate of the leading systematic effects (e.g., background normalization, particle misidentification, mass-scale calibration) and their effect on the projected upper limits is needed to support the claimed 99% CL sensitivities.","section":"Section VI and Eq. (4)"}],"minor_comments":[{"comment":"The phrase 'the η probability in elastic scattering is around 0.76%' is presumably a typo, since η production is an inelastic process; it should read 'inelastic scattering' or 'p-7Li collisions'.","section":"Section IV"},{"comment":"The term 'hardrophilic' appears to be a typo for 'hadrophilic' in several places (e.g., Sections II.B and V.E and the figure captions).","section":"Throughout"},{"comment":"The paper should specify the exact event selection criteria (for example, the ±3σ windows around the η and π0 masses) and the bin width used in Figs. 14 and 15, as these affect the interpretation of N_bg in Eq. (4).","section":"Section V.C and Eq. (4)"},{"comment":"The formula 3√N_bg is a Gaussian approximation to the 99% CL upper limit on a Poisson signal in the presence of background; the authors should state this approximation and comment on its validity for bins with small N_bg.","section":"Eq. (4)"},{"comment":"The axis labels in Figs. 14 and 15 (especially the y-axis 'Events') are poorly rendered in the manuscript and should be made legible in the published version.","section":"Figures 14 and 15"}],"recommendation":"major_revision","confidential_remarks":"The topic fits the journal's scope and the simulation framework is a useful contribution. The main blocker is the internal inconsistency in the Monte Carlo scale-up factor, which directly propagates into the central sensitivity claims; this must be fixed and the numeric projections re-evaluated. If the code in fact used the correct factor and the text contains a typo, the paper may be acceptable after a careful revision. I would also encourage the authors to provide the explicit derivation of N_η and to include at least a simplified systematic-uncertainty estimate, since the projected background counts are large enough that systematic effects are likely to dominate."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a facility-specific sensitivity study for a proposed HIAF eta factory, projecting dark scalar limits in eta -> S pi0 decays. The new content is real: detector efficiencies, mass resolutions, and projected limits for this setup. But the central yield number has an arithmetic inconsistency that makes the quoted limits optimistic by about a factor of 8 as written.\n\nThe paper does honest work. It uses GiBUU for p+Li -> eta at 1.8 GeV, builds a ChnsRoot detector simulation, reports ~40% efficiencies, sub-2 MeV mass resolution for the scalar, and benchmarks against REDTOP and existing constraints. The authors openly state in Sec. VI that experimental uncertainties aren't quantified quantitatively. That's a credit.\n\nThe soft spot is load-bearing. Section IV says 13 million inelastic events were simulated, with a 0.76% eta probability. That yields about 1e5 eta events. The text says they scaled up by 'around 10^5', giving about 1e10 eta. But they quote N_eta = 5.9e11 for one month. Repeating the math, the scale factor should be ~6e6, not 1e5. If backgrounds were scaled by 1e5 while N_eta is taken as 5.9e11 in Eq. (4), then N_bg is too small by a factor of 60, and the upper limits are underestimated by sqrt(60) ~ 7.7. That directly undercuts the quoted sin^2(theta) ~ 1e-1 and g_u ~ 1e-6 sensitivities. This could be a typo—if the code actually used 6e6, the issue is minor—but as written the numbers do not reproduce.\n\nThere is also a smaller rate assumption issue: with L=1e35 cm^-2s^-1 and sigma = 0.1*A mb (A=7), the inelastic rate is about 70 MHz, not the 100 MHz cap the paper assumes for N_eta. So N_eta is optimistic by ~1.4 even after fixing the scale factor.\n\nThe upper limit formula (Eq. 4) is a 3-sigma approximation, not a proper 99% CL limit, and no systematics are included. That is acceptable for a projection if stated, but here the statement is loose.\n\nWho is this for? Anyone planning an eta factory or interpreting REDTOP projections will want to read it. As written, I would not trust the absolute limits until the scale factor and rate assumption are clarified. But the framework and the facility-specific numbers are worth refereeing. Send it to review, with the expectation that the authors fix the arithmetic and either justify the 100 MHz rate or recompute N_eta from the luminosity.","headline":"A facility-specific dark scalar sensitivity study for a proposed HIAF eta factory; the setup and simulations are solid, but a scale-up factor inconsistency makes the projected limits optimistic by roughly a factor of 8 as written.","tokens_in":14155,"tokens_out":4541,"would_cite":false,"duration_ms":37743,"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 argues that a one-month run at the proposed Huizhou eta factory could reach branching-ratio limits near $10^{-9}$ for dark scalars decaying to $e^+e^-$.","keywords":["dark scalar portal","eta meson rare decays","fixed-target experiment","HIAF","sensitivity projection","hadrophilic scalar model","minimal scalar model","silicon pixel spectrometer"],"falsifier":"Measure the actual $p + {}^7$Li $\\to \\eta X$ cross-section at 1.8 GeV and the real fixed-target luminosity and duty factor at the facility; if the resulting one-month $\\eta$ yield is below about $3\\times 10^{11}$, the projected branching-ratio limits would be correspondingly worse. Alternatively, check whether the simulated background shapes match data from an existing $\\eta$ experiment such as BESIII in the same final states.","tokens_in":13022,"feed_emoji":"🎯","tokens_out":9041,"duration_ms":74579,"temperature":0.7,"pith_summary":"This paper argues that a one-month fixed-target run at the proposed Huizhou eta factory could search for a light dark scalar particle in two rare eta decay channels, $\\eta \\to S\\pi^0 \\to e^+e^-\\pi^0$ and $\\eta \\to S\\pi^0 \\to \\pi^+\\pi^-\\pi^0$, with projected branching-ratio upper limits near $10^{-9}$ and below $10^{-6}$, respectively. The target is a narrow bump in the $e^+e^-$ or $\\pi^+\\pi^-$ invariant-mass spectrum sitting on a smooth background. The projected reach translates into sensitivities of $\\sin^2\\theta \\sim 10^{-1}$ for the minimal scalar (Higgs-mixing) model and $g_u \\sim 10^{-6}$ for the hadrophilic scalar model at 99% confidence. These numbers are obtained from a simulation of $\\eta$ production in $p+{}^7$Li collisions at 1.8 GeV, a compact silicon-pixel spectrometer design, and a scaling of the background to an assumed sample of $5.9\\times 10^{11}$ $\\eta$ events in one month.","feed_headline":"One-month HIAF run could reach 10^-9 dark-scalar branching fraction","feed_subtitle":"A fixed-target eta factory would set gu limits near 10^-6, beating current experiments in the sub-GeV range.","key_machinery":"The load-bearing mechanism is the bump-hunt: the dark scalar $S$ would appear as a narrow peak, with invariant-mass resolution below 2 MeV, in the $e^+e^-$ or $\\pi^+\\pi^-$ spectrum, on top of a smooth background from $\\eta\\to \\pi^0 e^+e^-$ and $\\eta\\to\\pi^+\\pi^-\\pi^0$. The sensitivity is set by the upper-limit formula $\\mathrm{Br}_{\\rm UL} = 3\\sqrt{N_{\\rm bg}}/(N_\\eta\\,\\epsilon)$, which converts the background counts in each mass bin into a branching-ratio limit using the total $\\eta$ yield $N_\\eta=5.9\\times 10^{11}$ and the bin-dependent detection efficiency $\\epsilon$. The translation from branching ratio to model parameters is done through two identities: $\\mathrm{Br}(\\eta\\to\\pi^0 S)\\simeq 1.8\\times 10^{-6}\\,\\lambda^{1/2}\\sin^2\\theta$ for the minimal scalar model, and the corresponding expression $\\mathrm{Br}(\\eta\\to\\pi^0 S) = c_{S\\pi^0\\eta}^2 g_u^2 B^2 \\lambda^{1/2}/(16\\pi m_\\eta \\Gamma_\\eta)$ for the hadrophilic model.","core_discovery":"The paper's central claim is that a fixed-target $\\eta$ factory can be a competitive dark-scalar hunter. Using an event generator for proton–lithium collisions at 1.8 GeV, a full detector simulation of a compact silicon-pixel-based spectrometer, and background-only Monte Carlo scaled by a factor of about $10^5$, the authors project that one month of running would set an upper limit on $\\mathrm{Br}(\\eta\\to S\\pi^0\\to e^+e^-\\pi^0)$ close to $10^{-9}$ for $S$ masses above the pion mass, and an upper limit below $10^{-6}$ on $\\mathrm{Br}(\\eta\\to S\\pi^0\\to \\pi^+\\pi^-\\pi^0)$. Interpreting these limits through the standard portal relations, they become sensitivities of $\\sin^2\\theta \\sim 10^{-1}$ and $g_u \\sim 10^{-6}$ at 99% CL. The paper presents these as conservative: it assumes one month at 30% duty factor, a 100 MHz inelastic event-rate cap, and detection efficiencies around 40%.","pith_inferences":["If the assumed $\\eta$ yield is not realized—say the luminosity or the $\\eta$ production cross-section is lower than assumed—every projected limit scales linearly with $N_\\eta$; the paper's reach figures should be read as an upper bound on what this facility could deliver, not a guaranteed performance.","The same bump-hunt strategy applies to other portals: replacing the $e^+e^-$ or $\\pi^+\\pi^-$ final state with $\\mu^+\\mu^-$ or $\\gamma\\gamma$ would let the same spectrometer search for vector and axion-like dark particles, since the background shape, not the model, determines the reach.","The low-mass sensitivity in the $e^+e^-$ channel is limited by the $\\pi^0$ Dalitz background dropping at the pion mass; a calorimeter with better energy resolution would push the reach to lower $S$ masses, a testable improvement in the detector design.","The projection uses a scale-up factor of about $10^5$ on a 13-million-event simulation; the statistical fluctuations of the background in the simulation should be checked against the scaled Poisson expectation, since bins with zero simulated events could underestimate $N_{\\rm bg}$."],"forward_implications":["The one-month experiment would set the best direct constraints on the hadrophilic scalar coupling $g_u$ below 1 GeV, surpassing existing bounds from MAMI, BESIII, KLOE, and E787/E949 in the mass region above the pion threshold.","The projected $\\sin^2\\theta \\sim 10^{-1}$ sensitivity in the minimal scalar model is comparable to the REDTOP proposal in the high-mass region, showing that a fixed-target $\\eta$ factory can compete with a dedicated $\\eta$/ $\\eta'$ production experiment.","Because the sensitivity is driven by $N_\\eta$, increasing the running time or event-rate capability directly improves the limits; the ideal one-year, 500 MHz scenario would improve them further.","The simulation demonstrates that the compact silicon-pixel spectrometer design achieves near-geometric acceptance (about 40%) and sub-2 MeV mass resolution, which are the key performance requirements for the bump-hunt."],"supporting_citations":[{"why":"Proposes the Huizhou super eta factory that this study builds on, providing the facility context for the one-month running scenario.","marker":"[57]"},{"why":"Supplies the parametrization of $\\mathrm{Br}(\\eta\\to\\pi^0 S)$ in terms of $\\sin^2\\theta$ and the hadrophilic coupling $g_u$, used to translate branching-ratio limits into model parameters.","marker":"[7]"},{"why":"Defines the hadrophilic scalar model and the $g_u$ coupling, and provides the SN1987A constraint compared in the sensitivity figure.","marker":"[13]"},{"why":"REDTOP projection, the baseline comparison for the sensitivity curves in both models.","marker":"[35]"},{"why":"Provides the measured $p$-$p$ $\\eta$ production cross-section near 1.8 GeV, extrapolated to $p$-$A$ by the $0.1\\times A$ mb assumption used in the event-yield estimate.","marker":"[69]"},{"why":"The transport-model event generator used to simulate $\\eta$ production and the inelastic background events.","marker":"[65]"},{"why":"The detector-simulation framework on which the ChnsRoot package is built, used to evaluate efficiencies and resolutions of the spectrometer.","marker":"[68]"}],"fun_headline_variants":["Eta factory simulation projects 10^-9 dark scalar limit","One-month HIAF run could probe dark scalar at 10^-9","HIAF eta factory: dark scalar sensitivity via rare decays","Fixed-target eta decays: new window to dark scalar at HIAF","Simulation shows HIAF eta factory can test dark scalar portal"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The projected reach assumes a one-month sample of $5.9\\times 10^{11}$ $\\eta$ mesons, which scales linearly with the assumed luminosity of $10^{35}\\,{\\rm cm}^{-2}{\\rm s}^{-1}$, the 100 MHz inelastic event-rate cap, the 30% duty factor, and the 0.76% $\\eta$-production probability from a $0.1\\times A$ mb cross-section; if any of these inputs is overestimated, every limit and sensitivity curve worsens in proportion.","fun_headline_variants_meta":{"raw":{"variants":["Eta factory simulation projects 10^-9 dark scalar limit","One-month HIAF run could probe dark scalar at 10^-9","HIAF eta factory: dark scalar sensitivity via rare decays","Fixed-target eta decays: new window to dark scalar at HIAF","Simulation shows HIAF eta factory can test dark scalar portal"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000638,"raw_usage":{"total_tokens":2968,"prompt_tokens":1001,"completion_tokens":1967,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":617,"completion_tokens_details":{"reasoning_tokens":1875}},"tokens_in":617,"tokens_out":1967,"duration_ms":15239,"temperature":1.0,"reasoning_tokens":1875,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T22:40:30.684896+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the actual $p + {}^7$Li $\\to \\eta X$ cross-section at 1.8 GeV and the real fixed-target luminosity and duty factor at the facility; if the resulting one-month $\\eta$ yield is below about $3\\times 10^{11}$, the projected branching-ratio limits would be correspondingly worse. Alternatively, check whether the simulated background shapes match data from an existing $\\eta$ experiment such as BESIII in the same final states.","supporting_citations":[{"cited_title":"The legacy of the experimental hadron physics programme at COSY","cited_arxiv_id":"1611.07250","evidence_quote":"Provides the measured $p$-$p$ $\\eta$ production cross-section near 1.8 GeV, extrapolated to $p$-$A$ by the $0.1\\times A$ mb assumption used in the event-yield estimate."},{"cited_title":"Al-Turany, D","cited_arxiv_id":null,"evidence_quote":"The detector-simulation framework on which the ChnsRoot package is built, used to evaluate efficiencies and resolutions of the spectrometer."}],"review_version":1}