{"id":"2b56de44-ebcd-4a40-a887-f4265de3b441","arxiv_id":"2607.02783","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"The neutron background flux at J-PARC is [1.45^{+0.22}_{-0.24} (stat.) ± 0.55 (sys.)] × 10^{-7} cm^{-2} s^{-1} POT^{-1} under an exponential spectrum assumption.","lead":"Researchers measured the flux of neutrons produced by a neutrino beam interacting with surrounding materials at Japan's J-PARC facility. This background number helps design cleaner detectors for rare neutrino-nucleus scattering studies.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.5","headline":"Flux definition φ=Aλ is model-dependent; the exponential spectrum is unconstrained by independent data and the recoil window only weakly constrains λ.","rationale":"The Reader correctly isolates the exponential-spectrum (and isotropic-direction) assumption as the weakest link and assigns CONDITIONAL with medium correctness risk. That assessment is accurate: the data-selection chain, calibrations, three-setup consistency and POT-normalized rates are solid, and no internal contradiction appears. The load-bearing issue is precisely that the flux is defined as the integral of an unconstrained ansatz whose parameters are only loosely fixed by a limited recoil window. The concrete re-fit test proposed above would quantify how much the quoted number moves under a plausible alternative spectrum; if it stays inside the present error band the result can be used more confidently, otherwise the model dependence must be enlarged or the claim re-phrased as a rate under the exponential hypothesis. No stronger objection (e.g., data-handling error or calibration failure) is evident, so the Reader’s verdict needs no change.","tokens_in":13203,"tokens_out":699,"duration_ms":6478,"concrete_test":"Re-fit the three LiqSci spectra of Fig. 6 after replacing the pure exponential with a two-component form (e.g. exp(-λ1) + f·exp(-λ2) or a power-law high-energy tail) while keeping the same quenching, resolution and PSD cuts; if the integrated flux φ changes by more than the quoted total uncertainty (~40 %), the headline number is model-dependent at the level claimed.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim quotes a single flux number φ=[1.45^{+0.22}_{-0.24}(stat.)±0.55(sys.)]×10^{-7} cm^{-2}s^{-1}POT^{-1}. By construction (Sec. 3.4.3, Eq. 1) this is the integral of an assumed pure exponential Γ(E_n)=A exp(-E_n/λ) from 0 to ∞, so φ≡Aλ. Neutrons are generated flat in 1–10000 MeV and re-weighted; the observed proton-recoil spectrum is only 0.98–11.60 MeVee (Fig. 6). Because of quenching, that window maps to a limited band of true E_n, leaving A and λ strongly correlated (Fig. 7) and the high-energy tail (which still contributes to the integral) essentially free. An isotropic angular distribution is likewise assumed (Sec. 3.4.4). The paper itself notes that production points, sand composition and energy loss are poorly known. Consequently the quoted flux is not a model-independent integral of the measured rate; it is the integral of a two-parameter ansatz whose shape is only weakly constrained by the data. The large PSD systematic (36 %) further softens the spectral constraint. The measurement remains useful under the stated model, but the numerical value cannot be treated as a spectrum-independent flux without additional validation.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The paper reports the first measurement of the accelerator-neutrino-induced neutron background flux at the J-PARC neutrino facility (NM building B2 floor). Using a plastic-scintillator veto, BGO crystals, and a BC-501A liquid scintillator with pulse-shape discrimination, the authors select 88 proton-recoil events in the 0.98–11.60 MeVee window from a neutrino-mode exposure of 2.972 × 10^20 POT. After beam-on/off subtraction, charged-particle veto, and PSD cuts calibrated at RCNP, they re-weight a Geant4 neutron sample with an assumed exponential spectrum Γ(E_n) = A exp(-E_n/λ) (Eq. 1) and extract the integrated flux φ = Aλ = [1.45^{+0.22}_{-0.24} (stat.) ± 0.55 (sys.)] × 10^{-7} cm^{-2} s^{-1} POT^{-1}. Three successive setups yield consistent rates; the dominant systematic is the PSD selection (36 %). The result is intended as input for future short-baseline NCQE measurements.","tokens_in":13564,"tokens_out":1199,"duration_ms":9494,"significance":"A site-specific neutron-flux measurement at J-PARC is of practical value for any short-baseline NCQE or nuclear-recoil experiment planned in the NM hall. The experimental chain—energy calibration with tagged γ sources, 65 MeV neutron-beam PSD calibration at RCNP, beam-on/off subtraction, and multi-setup consistency—is carefully documented and constitutes a solid first data point. The large PSD systematic and the model dependence of the quoted flux limit the precision, but the measurement remains useful under the stated assumptions and will help guide detector placement and shielding design.","major_comments":[{"comment":"Sec. 3.4.3, Eq. (1) and Sec. 5.2: the reported flux is defined as φ = Aλ, the integral of a pure exponential from 0 to ∞. Neutrons are generated flat in 1–10 000 MeV and re-weighted; the observed recoil window (0.98–11.60 MeVee) maps, after quenching, to only a limited band of true E_n. Consequently A and λ are strongly correlated (Fig. 7) and the high-energy tail that still contributes to the integral is essentially unconstrained. The paper should either (i) quote the flux only inside a well-defined energy window that is actually constrained by the data, or (ii) demonstrate that alternative spectral shapes (power-law, broken exponential, etc.) yield fluxes consistent within the quoted systematics. Without such a check the single number cannot be treated as a spectrum-independent flux.","section":null},{"comment":"Sec. 3.4.4: an isotropic angular distribution is assumed because production points and wall geometry are poorly known. Because the detectors are small and sit near walls, a strongly forward- or wall-directed component would change both the effective solid angle and the recoil spectrum. A brief sensitivity study (e.g., cos θ distributions weighted toward the beam axis or the nearest wall) should be added so that readers can judge how much of the 0.55 systematic already covers this uncertainty.","section":null},{"comment":"Sec. 5.3: the PSD systematic is quoted as 36 % and dominates the total error. The text states that the PSD cut boundaries were varied, but does not specify the variation range or the resulting change in selected event count. Given that the cut is taken from a 65 MeV mono-energetic beam while the in-situ spectrum is continuous and softer, a more transparent propagation (or an alternative PSD metric) is needed before the 36 % figure can be accepted as complete.","section":null}],"minor_comments":[{"comment":"Table 1 and abstract: total POT is written both as 2.972 × 10^20 and as 29.72 × 10^19; standardize to a single notation.","section":null},{"comment":"Fig. 3 caption: “root temperature” should be “room temperature”.","section":null},{"comment":"Throughout: “Liqid scintillator” / “fl ame” / “P ARC” spacing errors should be corrected.","section":null},{"comment":"Sec. 4.2: the pulse-separation limit of 0.4 µs is stated without a reference or measurement; a short justification would help.","section":null},{"comment":"Appendix A, Table A.1: σ_0 is negative; a brief remark that this is an empirical offset (not a physical resolution) would avoid confusion.","section":null}],"recommendation":"major_revision","confidential_remarks":"The measurement is useful and the experimental work is careful, but the central numerical claim is more model-dependent than the abstract suggests. I would not accept without the spectral-shape and angular-sensitivity checks requested above. Scope is appropriate for an instrumentation journal."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"This is the first direct measurement of accelerator-neutrino-induced neutron flux in the J-PARC NM hall. They took 2.97e20 POT, selected 88 proton-recoil events in a liquid scintillator via PSD (0.98–11.6 MeVee), subtracted beam-off, vetoed with plastic, and got a consistent rate across three setups. The number they quote is [1.45^{+0.22}_{-0.24}(stat) ± 0.55(sys)] × 10^{-7} cm^{-2} s^{-1} POT^{-1}.\n\nWhat they did well is the experimental chain. Energy calibration with 22Na/60Co, PSD cuts from a 65 MeV RCNP beam test, quenching factors from literature, Geant4 response, and χ^{2} re-weighting of a flat-generated neutron sample are all documented. The three periods give compatible rates, so the raw event yield looks real. For anyone designing a short-baseline NCQE recoil experiment in that hall, this is the number they will use until something better appears.\n\nThe soft spot is exactly the one the stress-test flags, and it is real but not fatal. They define flux as φ = Aλ where Γ(E_n) = A exp(-E_n/λ) is fitted to the recoil spectrum (Eq. 1). The observed window maps, after quenching, to a limited band of true neutron energy, so A and λ are strongly correlated and the high-energy tail that still contributes to the integral is essentially free. Isotropic directions are also assumed. They state both assumptions openly and the large PSD systematic (36 %) already softens the spectral constraint. This is ordinary spectral fitting for a background measurement, not circularity that invents the result; it just means the number is conditional on the exponential model. Future work with a broader energy lever arm or independent spectral data will tighten it.\n\nMath and citations look clean; no invented entities or load-bearing contradictions. The paper is for the small community planning NCQE or other low-threshold detectors at J-PARC (and anyone sharing that hall). It deserves a serious referee. I would cite the flux value with the model caveat when I need a background estimate for that site.","headline":"First usable site-specific neutron flux at J-PARC under an exponential model; solid data chain, but the quoted number is model-dependent by construction.","tokens_in":14135,"tokens_out":554,"would_cite":true,"duration_ms":5407,"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":"Neutron background flux from the J-PARC neutrino beam is measured at 1.45 × 10^{-7} cm^{-2} s^{-1} POT^{-1}.","keywords":["neutron flux","J-PARC","neutrino beam background","liquid scintillator","pulse-shape discrimination","NCQE","proton recoil","BGO"],"falsifier":"An independent measurement of the neutron energy spectrum (or angular distribution) at the same location that cannot be described by a single exponential would force a revision of the quoted flux.","tokens_in":14127,"feed_emoji":"⚛️","tokens_out":1003,"duration_ms":7864,"temperature":0.7,"pith_summary":"Neutrino experiments that search for nuclear-recoil signals from neutral-current quasi-elastic scattering are limited by accidental neutrons produced when the beam interacts in surrounding sand and concrete. This paper reports the first direct measurement of that neutron flux at the J-PARC neutrino facility. Using a liquid scintillator that identifies protons by pulse-shape discrimination, a plastic veto against charged particles, and 2.972 × 10^{20} protons on target, the collaboration selected 88 recoil-proton events between 0.98 and 11.60 MeV electron-equivalent. After efficiency and resolution corrections and a fit to Monte Carlo, they extract a flux of [1.45^{+0.22}_{-0.24} (stat.) ± 0.55 (sys.)] × 10^{-7} cm^{-2} s^{-1} POT^{-1} under an exponential energy spectrum. The number supplies a concrete background estimate for future short-baseline recoil detectors at the same site.","feed_headline":"J-PARC neutrino beam makes 1.45e-7 neutrons per cm^{2} s POT","feed_subtitle":"First direct flux of sand-and-concrete neutrons that mask nuclear-recoil signals","key_machinery":"Pulse-shape discrimination in a liquid scintillator, calibrated with a 65 MeV neutron beam, isolates proton recoils from gamma and heavy-ion backgrounds; the observed recoil spectrum is then fitted by re-weighting an isotropic Monte Carlo sample whose kinetic-energy distribution is forced to an exponential form Γ(E_n) = A exp(-E_n/λ).","core_discovery":"With 2.972 × 10^{20} protons on target in neutrino mode, 88 neutron-induced proton recoils were observed in a liquid scintillator between 0.98 and 11.60 MeVee after pulse-shape discrimination and charged-particle veto. Comparing the measured spectrum with a re-weighted Monte Carlo simulation yields a neutron flux of [1.45^{+0.22}_{-0.24} (stat.) ± 0.55 (sys.)] × 10^{-7} cm^{-2} s^{-1} POT^{-1} when the kinetic-energy spectrum is taken to be exponential.","pith_inferences":["If the true spectrum is harder or softer than the fitted exponential, the absolute flux (defined as the integral Aλ) will shift even if the recoil rate in the 1–12 MeVee window remains fixed.","A compact BGO array with the same liquid-scintillator tag could convert this background measurement into a simultaneous NCQE cross-section measurement once the neutron component is subtracted.","The large PSD systematic suggests that a dual-scintillator or capture-gated detector would reduce the dominant uncertainty for the next generation of flux measurements."],"forward_implications":["Future short-baseline NCQE experiments at J-PARC can now estimate their neutron-induced proton-recoil background rate per POT using the reported flux.","The measured flux is independent of beam power in the 650–800 kW range, so background scales simply with exposure.","The same apparatus can be used to map flux versus detector location and to compare neutrino versus antineutrino modes.","Systematic uncertainties are dominated by PSD efficiency (36 %), giving a clear target for improved calibration."],"fun_headline_variants":["J-PARC measures neutrino-induced neutron flux of 1.45e-7 cm^{-2} s^{-1} POT^{-1}","First neutron flux from J-PARC beam: 1.45e-7 cm^{-2} s^{-1} POT^{-1}","88 recoils give J-PARC neutron flux [1.45^{+0.22}_{-0.24}±0.55]e-7 cm^{-2}s^{-1}POT^{-1}","Sand-concrete neutron flux at J-PARC: 1.45e-7 cm^{-2} s^{-1} POT^{-1}","Neutrino beam neutrons clocked at 1.45e-7 per cm^{2} s POT at J-PARC"],"cache_read_input_tokens":128,"weakest_assumption_plain":"The neutron kinetic-energy spectrum is assumed to be a pure exponential whose two free parameters are fitted only to the observed recoil spectrum, and the neutrons are assumed to arrive isotropically.","fun_headline_variants_meta":{"raw":{"variants":["J-PARC measures neutrino-induced neutron flux of 1.45e-7 cm^{-2} s^{-1} POT^{-1}","First neutron flux from J-PARC beam: 1.45e-7 cm^{-2} s^{-1} POT^{-1}","88 recoils give J-PARC neutron flux [1.45^{+0.22}_{-0.24}±0.55]e-7 cm^{-2}s^{-1}POT^{-1}","Sand-concrete neutron flux at J-PARC: 1.45e-7 cm^{-2} s^{-1} POT^{-1}","Neutrino beam neutrons clocked at 1.45e-7 per cm^{2} s POT at J-PARC"]},"model":"grok-4.5","effort":"low","cost_usd":0.007822,"raw_usage":{"total_tokens":1979,"prompt_tokens":917,"num_sources_used":0,"completion_tokens":182,"cost_in_usd_ticks":78220000,"prompt_tokens_details":{"text_tokens":917,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":880,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":917,"tokens_out":182,"duration_ms":6858,"temperature":1.0,"reasoning_tokens":880,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-12T07:03:27.881341+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"An independent measurement of the neutron energy spectrum (or angular distribution) at the same location that cannot be described by a single exponential would force a revision of the quoted flux.","supporting_citations":[],"review_version":1}