{"id":"d89cdded-4fc0-42ea-85ca-4a45eeba8a13","arxiv_id":"2507.17955","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Nuclear gamma cascades in reactors produce millicharged particle pairs, giving the strongest constraints on millicharge for masses between 0.7 and 2 MeV.","lead":"This paper shows that gamma rays from nuclei inside nuclear reactors can convert into pairs of hypothetical millicharged particles, giving a new source that tightens limits on their electric charge. A smart generalist should care because it probes dark-sector particles in a mass window near 1 MeV, where other experiments have little sensitivity.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Internal inconsistency between Eq. (6) and Eq. (45) in the E1 χχ̄ differential rate makes the reactor MCP flux and resulting limits potentially mis-normalized.","rationale":"The reader identified the off-shell nuclear matrix element assumption as the weakest, but that assumption is well supported by the small qR expansion at MeV energies and is not the most load-bearing issue. The concrete algebraic inconsistency between the main-text and appendix formulas for the E1 pair-production rate is a verifiable, internal flaw that directly affects the calculated flux and hence the claimed exclusion limits. The central claim of the strongest constraints in 0.7–2 MeV depends on the normalization of this flux; an uncorrected factor in the rate would shift the limits by an amount comparable to the margin over competing constraints. The recommended verdict remains CONDITIONAL, because the issue is resolvable by an independent derivation and numerical check, and until then the numerical exclusion curve should not be taken at face value. The reader's call for a tabulated limit curve and uncertainty budget is still valid; this algebraic check is an additional necessary condition.","tokens_in":17817,"tokens_out":23124,"duration_ms":245894,"concrete_test":"Numerically integrate both expressions for mχ/ω ∈ {0.1, 0.25, 0.45} and compare them with a direct phase-space integration of the squared amplitude in Eq. (4), including the multipole expansion and kinematical limits of Appendix A. If the three results do not agree within 10%, identify which expression is correct and recompute the TEXONO-based limit on ε; also perform the same check for the M1 expressions Eq. (7) versus Eq. (46).","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 2's Eq. (6) and Appendix A's Eq. (45) both claim to give dr_E1/dEχ for the same E1 internal-pair-production process, but they disagree in the non-logarithmic terms. Using Eχ + Eχ̄ = ω, A = EχEχ̄ + mχ², B = |pχ||pχ̄|, and the identity (A+B)(A−B) = mχ²ω², Eq. (45) reduces to a bracket 20B + (Eχ²+Eχ̄²) ln[(A+B)/(A−B)] times αε²/(4πω³). Equation (6), after converting the prefactor from αε²/(2πω³) to αε²/(4πω³), gives 8B + (Eχ²+Eχ̄²) ln[(A+B)/(A−B)]. The two expressions differ by 12B, which is nonzero for all masses except exactly at threshold. This is not a cosmetic typo: the integrated pair-production rate differs by an O(1) factor that varies with mχ/ω. Because the derived limits scale as ε ∝ (flux)^{−1/4}, a factor 2–3 error in the flux moves ε by roughly 20–50%, comparable to the claimed margin over previous constraints in the 0.7–2 MeV window. The on/off-shell nuclear matrix element assumption flagged by the reader is standard and likely safe at these low momentum transfers; the algebraic mismatch between Eq. (6) and Eq. (45) is a more direct threat to the numerical claim and must be resolved before the limits can be trusted.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper argues that nuclear γ-ray cascades from neutron capture in reactors — specifically the 3.297 and 4.060 MeV E1 lines of 239U from 238U(n,γ) and the 2.223 MeV M1 line of 2H from p(n,γ) — are an overlooked and powerful source of millicharged particles, since every nuclear γ transition with ω > 2mχ can internally convert into a χχ̄ pair. Using the multipole-expansion formalism of Pitrou and Pospelov, the authors derive the differential pair-production ratios for E1 and M1 in Sec. 2 (Eqs. 3–8), with E2/E3 generalized in Appendix A, construct the MCP flux at the TEXONO detector (Eqs. 9–10), convolve it with the PAI atomic-ionization cross section (Eq. 11), and match the predicted electron-recoil spectrum to the TEXONO measurement to set upper limits on the millicharge ε. The resulting limits are claimed to be the strongest laboratory constraints in the 0.7–2 MeV mass range, surpassing SLAC, CONNIE, and Atucha-II. Sections 4 and 5 extend the method to geo-radioactivity and solar fluxes and to MeV-scale dark photons, with candid statements about which of those results are not competitive.","tokens_in":18063,"tokens_out":42671,"duration_ms":404099,"significance":"The central significance is that the new production channel hardens the reactor MCP spectrum and extends the reachable mass range from ~0.5 MeV to ω/2 ≈ 2 MeV; if correct, the resulting 0.7–2 MeV limit would be the strongest laboratory exclusion to date. The paper's strengths are analytic: the multipole ratios are a parameter-free QED input, the appendix supplies the E2/E3 generalization, and the ε ∝ (rate)^{1/4} scaling makes the limits fairly robust to input-normalization errors. I independently checked the two E1 derivations and found them consistent: with A = EχEχ̄ + mχ², B = |pχ||pχ̄|, A²−B² = mχ²ω², I0 = 4B, I−1 = ln[(A+B)/(A−B)] and I−2 = B/(mχ²ω²), Eq. (45) gives (αε²/4πω³)[8B + 2(Eχ²+Eχ̄²) ln(...)], which equals Eq. (6) after prefactor conversion; the 12B discrepancy claimed in an internal stress-test is a prefactor-conversion artifact. The on/off-shell matrix-element assumption is the standard internal-conversion long-wavelength approximation and is safe here (kR ≲ 10⁻²).","major_comments":[{"comment":"The limit-setting procedure is described only qualitatively: the limiting ε is found when the counting rate (12) is 'matched to the experimental rate in [30]', and the Fig. 3a caption adds that the predicted rate must not exceed the 2σ region of the TEXONO analysis. Because the central claim — the strongest constraints in the 0.7–2 MeV range — rests on the curve in Fig. 3b, and because a factor-of-2 shift in ε is comparable to the displayed margin over the other constraints at some masses, the procedure must be specified precisely for reproducibility: (i) the recoil-energy range and binning used; (ii) whether the curve is the envelope of per-bin 2σ upper limits or the result of a binned likelihood/χ² statistic; (iii) how the 2σ band of the background-subtracted TEXONO spectrum is defined (statistical only, or including systematic uncertainties); and (iv) how the mχ values are sampled. Please also state how the curve shifts if the comparison is done at the 1σ or 90% level.","section":"§3, Eq. (12), Fig. 3b"},{"comment":"The input data that set the flux normalization are documented only loosely. The neutron-capture yields Y_n are taken from a 2005 simulation of the TEXONO reactor environment, justified only by a footnote, and the numerical values of Y_n(H), Y_n(238U), and the ENSDF line intensities I_γ for the three adopted transitions are not stated in the text. Although ε ∝ (Y_n)^{−1/4}, a factor-of-2–3 error in a yield moves ε by 20–30%, which is non-negligible relative to the claimed exclusion margins at some masses. Please provide a table of the adopted inputs, clarify the role of the symbol 'P_Yn' in Eq. (9), and either estimate the resulting uncertainty on ε or demonstrate that bracketed variations of the yields leave the limit curve essentially unchanged.","section":"§2, Eq. (9)"}],"minor_comments":[{"comment":"Typo: '214Bo' should be '214Bi'; the 238U-chain gamma line above 1 MeV used in the text is from 214Bi.","section":"§4"},{"comment":"There is a factor-of-2 inconsistency in the solar flux normalization: Eq. (22) at r = mχ/me → 0 gives ε²/2, while the text states the probability is 'just ε²' and writes the total flux as 2ε² times the pp neutrino flux, which fixes the prefactor of Eq. (21). Please clarify whether the probability counts pairs or individual MCPs and correct the prefactor if needed.","section":"§4, Eqs. (21)-(22)"},{"comment":"Please double-check the prefactor 2/6 in Eq. (14): the 232Th chain has four beta decays (hence four geoneutrinos) per chain, and the 2.615 MeV line is emitted in the 36% 208Tl sub-branch, so one 208Pb de-excitation per six neutrino emissions appears to overestimate the geo-MCP flux by roughly a factor of two.","section":"§4, Eq. (14)"},{"comment":"The comparison plot shows only laboratory constraints. A sentence locating the new limits with respect to astrophysical bounds (for example, SN1987A energy-loss limits, which apply in a different, trapping-dependent ε window) would make the 'strongest constraint' claim unambiguous.","section":"§3, Fig. 3b"},{"comment":"Please justify the restriction to three transitions in the MCP flux sum; in particular, the 7.6 MeV 56Fe(n,γ) lines used in Sec. 5 are not included in the MCP flux, and the paper should state whether their contribution for the TEXONO geometry was checked.","section":"§3"},{"comment":"In the Discussion, 'loose sight' should read 'lose sight'.","section":"§6"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the journal's scope, and the QED derivation is sound. I directly verified the consistency of Eq. (6) with Eq. (45), so the algebraic inconsistency raised during internal review is not a real defect and should not block the paper. The two substantive requests concern reproducibility of the limit curve (precise matching procedure) and documentation of the input yields and intensities; both are addressable by the authors without new physics, so I would not support rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper is worth a serious referee. The core idea is that nuclear de-excitation gammas in a reactor, especially from 239U neutron-capture cascades, can internally produce millicharged pairs, extending kinematic reach in mχ from ~0.5 to ~2 MeV. Previous reactor analyses only considered Compton-like eγ→eχχ̄, which cuts off around 0.5 MeV. The new limits at 0.7-2 MeV are plausibly the strongest there, and there's a new solar MCP flux estimate.\n\nThe derivation is clean: multipole expansion gives parameter-free pair-production ratios, and the limits scale as ε^4, so factor-of-few uncertainties in neutron yields or detector response barely matter. The paper is transparent about assumptions, e.g., the on/off-shell nuclear matrix element equivalence, which is standard and safe at these momentum transfers.\n\nThe soft spots are real but not fatal. The TEXONO recast is a simplified 2-sigma matching with no full uncertainty budget; that should be tightened before publication. Only three gamma lines are used, and the solar/geo sections are explicitly exploratory. The dark photon part is subdominant but a nice extra.\n\nI checked the alleged inconsistency between Eq. (6) and Eq. (45) flagged in the stress test. It does not hold up. Using A = EχEχ̄ + mχ², B = |pχ||pχ̄|, the integrals give I0 = 4B, I_{-1} = ln[(A+B)/(A-B)], I_{-2} = B/(mχ²ω²). Substituting into Eq. (45) yields 8B + 2(Eχ²+Eχ̄²) ln(...) times αε²/(4πω³), which is exactly what Eq. (6) reduces to after converting its prefactor. The apparent mismatch came from miscalculating I_{-2}. So the central flux formula is consistent.\n\nRecommendation: send to a serious referee. The authors should add a table of the exclusion curve and an uncertainty estimate for the flux inputs, but the physics point is solid. The solar flux estimate is a useful prospectus for low-threshold detectors.","headline":"Genuinely new reactor MCP limits from nuclear de-excitation gammas; the Eq. (6)-vs-Eq. (45) mismatch in the stress test does not hold up, but the TEXONO recast would benefit from an uncertainty budget.","tokens_in":18704,"tokens_out":4966,"would_cite":true,"duration_ms":42862,"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":"Nuclear gamma cascades inside reactors produce millicharged particle pairs, letting existing low-threshold germanium data set the tightest millicharge limits for masses 0.7-2 MeV.","keywords":["millicharged particles","internal pair production","nuclear gamma decays","reactor searches","electron recoil","dark photon","solar millicharged flux","MeV-scale dark sectors"],"falsifier":"Measure the pair-production-to-photon ratio for the 3.297 and 4.060 MeV gamma lines of 239U by placing a low-threshold detector near a neutron beam on 238U; if the measured ratio falls several times below the E1 formula, the paper's ε constraints would weaken by roughly the fourth root of that deficit.","tokens_in":17523,"feed_emoji":"⚛️","tokens_out":8520,"duration_ms":83017,"temperature":0.7,"pith_summary":"The paper argues that every nuclear gamma transition with energy ω can, in principle, emit a millicharged particle pair instead of a photon whenever 2mχ<ω, and that this internal pair-production channel is the dominant reactor source of millicharged particles for masses above roughly 0.5 MeV. Previous reactor analyses considered only Compton-like production e+γ→e+χχ̄, which cuts off near mχ~0.5 MeV; the gamma cascades from neutron capture, notably the 3.297 and 4.060 MeV E1 lines of 239U and the 2.223 MeV M1 line of 2H, extend the reach to about half the gamma energy. Matching the predicted electron-recoil rate to the measured spectrum of a 500 g germanium detector placed 28 m from a 2.9 GW reactor yields the strongest existing constraints on the millicharge ε in the mass interval 0.7-2 MeV, surpassing beam-dump and other reactor searches. The same machinery gives a solar millicharged flux and suggests that future low-threshold dark matter detectors could probe ε below $10^{-6}$.","feed_headline":"Reactor gamma cascades yield strongest millicharge limits yet","feed_subtitle":"Nuclear gamma lines of 2.2-4.1 MeV become millicharged pairs, extending reactor limits to 0.7-2 MeV masses.","key_machinery":"The load-bearing object is the internal pair-production ratio r = σ(n+N→N'+χχ̄)/σ(n+N→N'+γ), computed by multipole expansion of the nuclear matrix element under the recoil-less approximation. For E1 and M1 transitions the differential ratios give the millicharged energy spectrum from each gamma line, and the flux is summed over the 3.297 and 4.060 MeV lines of 239U and the 2.223 MeV line of 2H using neutron-capture yields. On the detection side, the Photo Absorption Ionization model converts the flux into electron-recoil spectra, which are matched to the measured spectrum of the low-threshold germanium experiment.","core_discovery":"For a fermion χ with charge εe, any nuclear gamma transition of energy ω can internally convert to a χχ̄ pair, and the pair-production rate relative to photon emission is controlled by the multipolarity of the transition, with the E1 channel dominating near the kinematic endpoint. Using neutron-capture yields simulated for a natural-water reactor and gamma line intensities from nuclear databases, the authors compute the millicharged flux from the 239U and 2H transitions and fold it with the photo-absorption ionization cross section to predict electron recoils in a low-threshold germanium detector. Matching this rate to the measured recoil spectrum gives ε exclusions that are strongest in the 0.7-2 MeV mass range, and the paper also derives a solar flux from positron annihilation and D(p,γ)3He, plus constraints on MeV-mass dark photons decaying to e+e-. The central claim is that reactor gamma cascades, not just photon-electron scattering, are the right production mechanism and substantially widen the mass range that near-reactor experiments can exclude.","pith_inferences":["Beyond the paper, the same internal-pair-production argument applies to any intense gamma source, such as spallation neutron sources, medical isotope production, or fusion facilities, which could place comparable or better limits at different mass points.","Beyond the paper, the two 239U lines at 3.297 and 4.060 MeV should imprint a distinctive two-step structure in the recoil spectrum, so a shape analysis of the existing data could separate a millicharged signal from backgrounds even if the absolute flux is uncertain.","Beyond the paper, for mχ below the electron mass, hard reactor gammas convert to e+e- on uranium nuclei and the subsequent positron annihilation to χχ̄ scales as ε² rather than αε², so simulating photon conversion yields could improve sub-MeV limits beyond the range the paper stresses.","Beyond the paper, if future low-threshold detectors see a solar millicharged signal, disentangling it from neutrino coherent scattering will require independent knowledge of the solar millicharged energy distribution, which depends on thermalization in the Sun."],"forward_implications":["The 0.7-2 MeV mass window for millicharged fermions is now excluded down to ε values that no previous experiment reached, with the limit curve set by reactor gamma cascades rather than beam dumps.","Reactor searches no longer lose sensitivity above mχ~0.5 MeV; the available energy of the (n,γ) lines sets a new kinematic ceiling near 2 MeV per particle.","Because the ionization cross section grows as 1/T at low recoil energy, detectors with thresholds below the current 300 eV will strengthen the same limits without any new source.","The gamma-line source also produces MeV-scale dark photons, giving the first near-reactor constraints on visibly decaying A' in the 1.1-4 MeV range, though these are weaker than existing limits.","The solar millicharged flux estimate gives a concrete target for low-threshold dark matter detectors: ε around 10^-6 and below could be probed once propagation through the Sun is modeled better."],"supporting_citations":[{"why":"Supplies the pair-production ratio formalism, including the E1 and M1 differential rates used for the flux calculation.","marker":"[36]"},{"why":"Supplies the neutron-capture yields per fission and the reactor simulation used to normalize the millicharged flux.","marker":"[38]"},{"why":"Supplies the gamma line energies and intensities for the 239U and 2H transitions that set the kinematic reach.","marker":"[39]"},{"why":"Provides the measured electron-recoil spectrum and detector configuration against which the predicted rate is matched.","marker":"[30]"},{"why":"Provides the Skipper-CCD reactor limits used as comparison benchmarks for the derived ε constraint.","marker":"[31]"},{"why":"Provides the previous beam-dump limit that the new reactor constraint surpasses in the 0.7-2 MeV range.","marker":"[17]"},{"why":"Establishes that nuclear gamma emission can be accompanied by emission of a charged pair, the physical basis for the production mechanism.","marker":"[35]"},{"why":"Supplies the photo-absorption cross sections used by the Photo Absorption Ionization model for germanium.","marker":"[42]"}],"fun_headline_variants":["Reactor gamma cascades yield strongest millicharge limits yet","Gamma cascades set new millicharge bounds near 1 MeV","Nuclear decays tighten constraints on millicharged particles","Overlooked gamma cascades expose millicharged particle flux"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument rests on assuming that the nucleus emits a real photon and a millicharged pair with the same matrix element, so pair production follows the known gamma line intensities; if the virtual photon is suppressed, every derived limit weakens.","fun_headline_variants_meta":{"raw":{"variants":["Reactor gamma cascades yield strongest millicharge limits yet","Gamma cascades set new millicharge bounds near 1 MeV","Nuclear decays tighten constraints on millicharged particles","Overlooked gamma cascades expose millicharged particle flux"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00069,"raw_usage":{"total_tokens":3115,"prompt_tokens":926,"completion_tokens":2189,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":542,"completion_tokens_details":{"reasoning_tokens":2118}},"tokens_in":542,"tokens_out":2189,"duration_ms":18131,"temperature":1.0,"reasoning_tokens":2118,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T14:41:51.432003+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the pair-production-to-photon ratio for the 3.297 and 4.060 MeV gamma lines of 239U by placing a low-threshold detector near a neutron beam on 238U; if the measured ratio falls several times below the E1 formula, the paper's ε constraints would weaken by roughly the fourth root of that deficit.","supporting_citations":[{"cited_title":"Version available at http://www.nndc.bnl.gov/ensarchivals/","cited_arxiv_id":null,"evidence_quote":"Supplies the gamma line energies and intensities for the 239U and 2H transitions that set the kinematic reach."},{"cited_title":"Internal pair production associated with the emission of high-energy gamma rays,","cited_arxiv_id":null,"evidence_quote":"Establishes that nuclear gamma emission can be accompanied by emission of a charged pair, the physical basis for the production mechanism."},{"cited_title":"X-Ray Interactions: Photoabsorption, Scattering, Transmission, and Reflection at E = 50-30,000 eV, Z = 1-92,","cited_arxiv_id":null,"evidence_quote":"Supplies the photo-absorption cross sections used by the Photo Absorption Ionization model for germanium."}],"review_version":1}