{"id":"ac80da79-455c-4e44-a5a4-3399dfaa179c","arxiv_id":"1207.3971","paper_version":3,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Di-photon dark matter annihilation operators induce loop-level nucleus scattering whose destructive interference produces a non-standard recoil spectrum, and for p-wave annihilation existing direct-detection limits are competitive with gamma-ray-line searches.","lead":"Dark matter that annihilates into two photons must also scatter off nuclei through loop diagrams, and this paper computes that scattering in a model-independent effective-field-theory framework. The resulting recoil spectrum has a distinctive interference dip, and for p-wave annihilation current direct-detection limits already rival gamma-ray-line searches.","discovery_kind":"unclear","skeptic_critique":{"model":"selfhosted/deepseek-v4.1-flash","headline":"The reader's EFT-validity concern is real but the authors flag it themselves in Section 3; the sharper unresolved issue is that the p-wave comparison in Fig. 4 is drawn at M_* = 200 GeV while the benchmark line cross section needs M_* ~ 10 GeV, so the plotted reach is not the reach for the signal it","rationale":"The reader's weakest_assumption correctly identifies the M_* / top-threshold tension as the load-bearing soft spot, and I agree with that identification. The paper is an explicit EFT calculation; the operator algebra of Sections 2-4 and the conversion in Eq. (5.6) are internally consistent, and the coefficients are bounded by published XENON100 and CDMS-II limits rather than fitted, so there is no circularity. The authors themselves state in Section 3 that the benchmark line cross section needs M_* ~ 10 GeV (p-wave) to ~100 GeV (s-wave), below m_t, while Fig. 4 fixes M_* = 200 GeV; that is an acknowledged limitation, and per the review rules I should not present it as a new finding. What remains unresolved is that the acknowledgement is not propagated: Fig. 4 and the XENON1T projection are drawn at a scale the motivating signal does not supply, so the plotted reach is conditional. That is a quantitative-reach caveat, not a correctness error, and it does not change the qualitative claim of abstract:s3 that direct and indirect searches are comparable for p-wave. The reader's secondary items (no XENON1T uncertainty band, velocity enhancement of Appendix A not propagated, no shipped code) are correctly judged non-load-bearing. I therefore keep the reader's CONDITIONAL verdict and agree with the reader's diagnosis; the concrete test above would settle whether the scale dependence is large enough to require an M_* band on Fig. 4.","tokens_in":54731,"tokens_out":99223,"duration_ms":1967421,"concrete_test":"Recompute Fig. 4 (both panels) at M_* = 10 GeV and M_* = 100 GeV, keeping all other inputs fixed, and report whether the XENON100 curve still crosses the Fermi LAT line limit below m_chi ~ 40 GeV and whether the XENON1T curve still reaches the 130 GeV tentative-signal box. If the crossing mass moves by more than a factor of two, the quoted reach is scale-dependent and the figure should carry an M_* band.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim (abstract:s3) is a comparison: for p-wave annihilation, the XENON100 bound on the operator coefficient maps, via Eq. (5.6), onto a gamma-gamma annihilation cross section comparable to the Fermi LAT line limit. The mapping is not a pure number: it depends on M_* through the mixing logarithms in Eq. (5.3), and the authors fix M_* = 200 GeV in Fig. 4. Section 3 states that reproducing the benchmark 10^-27 cm^3/s line cross section requires M_* ~ 10 GeV for p-wave. At M_* = 10 GeV the top-quark matching of Eqs. (4.5)-(4.7) does not apply, and the coefficient of the Helm term in Eq. (5.5), which carries the ln(M_*^2/(200 GeV)^2) factor, changes sign and magnitude. Since the destructive interference between the Rayleigh and mixing terms is what produces the distinctive recoil spectrum and sets the event rate, the bound on C_chi and hence the reach in Fig. 4 is not the bound at the parameter point the motivating signal actually occupies. The authors acknowledge the M_* tension in Section 3 but do not propagate it into Fig. 4; the figure is drawn at a scale the signal does not supply. This is a quantitative-reach concern, not a correctness error in the operator algebra, and it does not by itself overturn the qualitative statement that direct and indirect searches are comparable for p-wave. It does mean the specific numbers in Fig. 4 and the XENON1T projection should be read as conditional on M_* = 200 GeV rather than as the reach for a 130 GeV line.","agreement_with_reader":"agree"},"referee_report":{"model":"selfhosted/deepseek-v4.1-flash","summary":"The manuscript studies the direct detection cross section induced by the same effective operators that generate dark matter annihilation into two photons. Working in an effective field theory with operators up to dimension 7, the authors compute the loop-induced dark matter-nucleus scattering cross section, including QED and QCD mixing, heavy-quark threshold corrections, and Rayleigh scattering off the entire nucleus. They find that the Rayleigh and mixing contributions interfere destructively, producing a distinctive non-standard recoil spectrum with a dip near threshold. Using XENON100 and CDMS-II limits, they bound the operator coefficients and translate these bounds into upper limits on the gamma-gamma annihilation cross section. For s-wave annihilation, direct detection is far less sensitive than indirect searches; for p-wave annihilation, the constraints are comparable, and XENON1T can probe the parameter space of the tentative 130 GeV line.","tokens_in":37118,"tokens_out":45388,"duration_ms":862348,"significance":"The paper provides a model-independent bridge between two search strategies, with an explicit and checkable operator matching chain and a novel recoil spectrum that serves as a falsifiable prediction. The correction to the Rayleigh form factor of Weiner and Yavin is a useful service to the community, and the qualitative conclusion that direct and indirect searches are competitive for p-wave annihilation is robust and motivates further experimental work. The derivation is parameter-free in the sense that the operator coefficients are free parameters of the EFT, not fitted to the data; the data enter only as published upper limits.","major_comments":[{"comment":"The bound on the gamma-gamma annihilation cross section depends on the new-physics scale M_* through the effective coupling f_0 in Eq. (5.3), and Fig. 4 is drawn at M_* = 200 GeV. Section 3 states that reproducing the benchmark p-wave cross section of 10^-27 cm^3/s requires M_* of about 10 GeV. At M_* = 10 GeV the logarithm in Eq. (5.5) is large and negative, changing the coefficient of the Helm term from 0.57 to roughly 0.03, which alters the interference pattern and hence the bound on C_chi. The authors should quantify this dependence and show the reach at the benchmark scale, or justify why 200 GeV is representative. This is load-bearing for the quantitative comparison in abstract:s3 and for the XENON1T projection.","section":"Fig. 4 and Eq. (5.6)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The matching chain of Eqs. (4.5)-(4.7) assumes that the new-physics scale M_* lies above the top-quark threshold, but the p-wave benchmark of Section 3 requires M_* of about 10 GeV, below m_t. The authors should clarify whether M_* in the logarithms is the true matching scale or the effective suppression scale, and whether the EFT is valid at the benchmark point. If the operator is generated below m_t, the top-quark matching is absent and the interference pattern changes, which affects both the recoil spectrum and the bound on C_chi.","section":"Section 4 vs Section 3","authors_acknowledge":null,"authors_acknowledge_p":null}],"minor_comments":[{"comment":"The numerical prefactors in the annihilation and direct detection columns appear to differ from Eqs. (3.2) and (5.1) by factors of pi^2; please verify the entries against the equations in the text.","section":"Table 1","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The caption reads 'The blue (green) line indicated the Rayleigh (mixing) contribution'; 'indicated' should be 'indicates'.","section":"Fig. 3 caption","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The numerical coefficient 1.69 for the Rayleigh term should be checked against the formula in Eq. (5.3); a brief derivation or intermediate step would help the reader reproduce it.","section":"Eq. (5.5)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The assumption of no additional dark matter-nucleus couplings, stated in abstract:s1 and Section 2, should be reiterated in the conclusions so that the scope of the bounds is clear.","section":"Section 6","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The caption could more clearly explain the mapping between the Majorana/Dirac limits and the real/complex scalar limits, since the same curves are used for both cases.","section":"Fig. 4 left panel","authors_acknowledge":null,"authors_acknowledge_p":null}],"recommendation":"minor_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"selfhosted/deepseek-v4.1-flash","letter":"This one is worth your time if you work on either side of the direct/indirect divide. The real content is not the operator list — that builds on Rajaraman-Tait-Whiteson — but the interference between the Rayleigh amplitude (two photons scattering off the whole nucleus, Z²) and the loop-induced mixing contributions (DM to quarks and gluons via QED/QCD running and heavy-quark thresholds). At zero momentum transfer the two are comparable and opposite in sign; at finite q the Rayleigh form factor dies much faster, so the recoil spectrum has a dip near threshold and peaks at large recoil. That is a distinctive, falsifiable shape, and they show it for XENON100 and CDMS-II. They also fix two errors in Weiner-Yavin's Rayleigh treatment (a missing factor of 2 in an exponent, a sine that should be sinh). That correction alone earns the paper a look. The algebra from (4.2) through (5.6) is explicit and I could re-derive it; nothing is fitted to the data being used as a bound, so the circularity worry does not land. Where I part company with the presentation is Fig. 4. The right panel plots the XENON100 bound on ⟨σv⟩γγ at fixed M_* = 200 GeV, but Section 3 itself says the benchmark 10⁻²⁷ cm³/s needs M_* ≈ 10 GeV for p-wave. The Helm coefficient in (5.5) carries 0.09 ln(M_*²/(200 GeV)²); at 10 GeV that term is about −0.54 against the 0.57, so the destructive interference nearly cancels and both the recoil shape and the reach move. The authors acknowledge the EFT-validity tension in Section 3 and note M_* is not literally the new-physics scale, but they do not carry it into the figure. So the qualitative claim — direct and indirect are comparable for p-wave — survives; the specific numbers and the XENON1T projection should be read as conditional on M_* = 200 GeV. Secondary: the XENON1T curve has no uncertainty band, and Appendix A's 5–30% velocity enhancement is quantified but not propagated into the p-wave comparison where it matters. Both minor. Audience is DM direct/indirect detection people and EFT practitioners; a serious referee should see it, and I would ask for one variant of Fig. 4 at the scale the benchmark actually requires, or an explicit statement that the reach is conditional.","headline":"A careful, explicit EFT calculation whose technical core holds up; the p-wave reach plot is drawn at a scale the motivating signal does not supply, and the authors flag but do not propagate that tension.","tokens_in":19283,"tokens_out":2045,"would_cite":true,"duration_ms":39193,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"pith_extraction":{"msc":[],"pacs":["95.35.+d"],"model":"selfhosted/deepseek-v4.1-flash","headline":"A dark matter particle that annihilates into two photons should also scatter off nuclei through a loop-induced coupling, and for p-wave annihilation the resulting xenon limit on the annihilation rate is already comparable to the gamma-ray-l","keywords":["dark matter direct detection","gamma-ray lines","effective field theory","loop-induced scattering","Rayleigh scattering","operator mixing","p-wave annihilation","xenon detectors"],"falsifier":"A tonne-scale xenon detector running with the stated exposure and finding no events at the rate and recoil shape predicted in Figures 3 and 4 for fermionic dark matter with a large di-photon cross section near 130 GeV would remove the claimed reach, while a recoil spectrum showing the predicted deficit near threshold followed by a rising tail would confirm the interference.","tokens_in":19531,"feed_emoji":"🔭","tokens_out":4042,"duration_ms":73660,"temperature":0.7,"pith_summary":"The paper asks whether a dark matter particle that annihilates into two photons — the process behind the tentative 130 GeV gamma-ray line — must also scatter off nuclei in a way that underground detectors can see. Working in an effective field theory in which the photon-bilinear operator is the only link between dark matter and ordinary matter, the authors compute the one-loop matching that turns that operator into couplings to quarks and gluons, and add the coherent scattering of the two photons off the entire nuclear charge, known as Rayleigh scattering. The two contributions carry opposite signs and fall off with momentum transfer at very different rates, so the predicted recoil spectrum is unlike a standard weakly interacting massive particle spectrum: suppressed near threshold and peaked at larger recoil. Translating current XENON100 and CDMS-II limits into bounds on the di-photon annihilation rate, they find that s-wave scenarios lie six to eight orders of magnitude beyond direct-search reach, while p-wave fermionic scenarios are already competitive with the Fermi LAT line limit for masses below roughly 40 GeV.","feed_headline":"Xenon limits rival gamma-ray lines for p-wave dark matter","feed_subtitle":"For fermionic dark matter below 40 GeV, underground detectors already constrain the same coupling that line searches probe.","key_machinery":"The load-bearing object is the effective dark matter–nucleus coupling $f_chi$ of Eq. (5.3), which sums three matrix elements: the Rayleigh amplitude $2sqrt{2}pi alpha Z^2 Q_0 F_{rm Ray}(bar q)$, the light-quark contribution proportional to $-3m_N A (tfrac{4}{9}f_u^{NT}+tfrac{1}{9}(f_d^{NT}+f_s^{NT})) L_N F_{rm Helm}(bar q)$, and the gluon contribution proportional to $-tfrac{2}{9}m_N A f_G^{NT}$ times a sum over heavy-quark logarithms $Delta_Q L_Q$. The two form factors are the crux: $F_{rm Ray}(bar q)$ is a numerically computed function normalised to 1 at zero momentum and falling steeply, while $F_{rm Helm}(bar q)$ is the standard Helm form factor, so the relative weight of the two terms shifts with recoil energy. Equation (5.6) then converts a bound on the scattering cross section at zero momentum into a bound on the annihilation rate, with the factor $v_{rm rel}^2$ appearing for fermionic dark matter because the annihilation is p-wave while the scattering is not. Recoil-energy-averaged form factors, computed with the XENON100 response function and the maximum gap statistical method, give the numbers plotted in Figure 4.","core_discovery":"The central claim is a two-way translation. If dark matter annihilates into two photons through a dimension-6 (scalar dark matter) or dimension-7 (fermionic dark matter) operator built from the electromagnetic field strength, then the same operator necessarily induces a spin-independent dark matter–nucleus cross section at one loop, and a bound on the operator coefficient from a direct detection experiment becomes a bound on $langle sigma v_{rm rel}rangle_{gammagamma}$. The authors show that this loop-induced cross section receives two contributions of comparable size but opposite sign: a Rayleigh term in which the two virtual photons scatter off the whole nucleus, scaling as $Z^2$ in the amplitude and $Z^4$ in the rate, and a term from QED mixing of the operator into quark mass operators plus QCD threshold corrections at the charm, bottom and top masses, scaling with the usual $A^2$. Because the Rayleigh form factor falls much faster with momentum transfer than the Helm form factor, the interference is destructive at zero momentum but leaves the mixing term dominant at the recoil energies a real detector records. The resulting event rate has a dip at the lower end of the search window and a rising tail, and the conversion in Eq. (5.6) shows that the $v_{rm rel}^2$ suppression of p-wave annihilation is exactly what lets a direct detection limit on the scattering cross section compete with an indirect limit on the annihilation cross section.","pith_inferences":["The interference-and-dip logic is not specific to the operators tabulated here; any loop-induced electromagnetic coupling of dark matter to the nucleus, such as an anapole or charge-radius term, would produce a similar competition between a coherent long-wavelength piece and a short-distance piece, so the recoil shape could serve as a general discriminant.","If the 130 GeV line is confirmed and a tonne-scale xenon detector sees no corresponding events, the p-wave operator interpretation is excluded, which would push model building toward the s-wave or resonant s-channel completions the effective theory explicitly does not cover.","The paper's own benchmark numbers imply a tension worth making explicit on a public page: matching $10^{-27}~{rm cm}^3{rm s}^{-1}$ needs a suppression scale near 10 GeV for p-wave annihilation, below the top mass, while the plotted limits fix $M_* = 200$ GeV, so the projected reach is better quoted as a band over $M_*$ than as a single curve.","The same matching chain that produces destructive interference here would flip sign or change magnitude if the new-physics scale sat below the top threshold, which suggests a broader scan over $M_*$ as a natural follow-up calculation."],"forward_implications":["For Majorana dark matter with p-wave di-photon annihilation, XENON1T with a 1.1 tonne fiducial mass and one year of data reaches the parameter region of the tentative 130 GeV line, so the scenario becomes testable by direct search alone.","A recoil spectrum with a deficit near the energy threshold and a rising tail at larger recoil would be a signature of two interfering contributions rather than a single contact interaction, and would distinguish this class of models from ordinary spin-independent scattering.","For Dirac dark matter the XENON100 and XENON1T bounds on the annihilation cross section tighten by a factor of two relative to the Majorana case, because the annihilation rate carries an extra factor of one half.","Because the signal sits at large recoil energies, raising the upper end of the primary scintillation signal window in a xenon detector would add sensitivity beyond the estimate quoted here.","Scalar dark matter annihilating in s-wave through the same operators stays six to eight orders of magnitude below direct detection sensitivity, so indirect searches retain sole constraining power in that case."],"supporting_citations":[{"why":"Supplies the Rayleigh-scattering matrix element and form factor for two photons scattering off the whole nucleus, the term that interferes with the mixing contribution.","marker":"[24]"},{"why":"Provides the classic tree-level nucleon matrix element of the gluon field-strength operator used to evaluate the threshold-induced contribution.","marker":"[30]"},{"why":"Supplies the two-loop matching correction $delta_t$ included in the top-quark threshold matching of the gluon operator.","marker":"[31]"},{"why":"The Fermi LAT line-search upper limit that the direct detection bound is compared against in Figure 4.","marker":"[2]"},{"why":"The tentative 130 GeV line analysis whose parameter region is shown as the target box in the projected-sensitivity plot.","marker":"[4]"},{"why":"The XENON100 limit on the spin-independent cross section that is converted into the bound on the operator coefficient and hence on the annihilation rate.","marker":"[36]"},{"why":"The maximum gap statistical method used to derive the limits from the XENON100 and CDMS-II data.","marker":"[38]"},{"why":"The lattice value of the strange-quark scalar form factor $f_s^{NT}$ adopted in the numerical evaluation of the mixing contribution.","marker":"[35]"},{"why":"The relative scintillation efficiency curve for xenon used to build the detector response function and the recoil-energy-averaged form factors.","marker":"[49]"}],"fun_headline_variants":["Loop-induced recoils test di-photon dark matter","Destructive interference shapes the dark matter recoil spectrum","Gamma-ray line coupling induces a nuclear recoil signal","Direct detection rivals line searches for p-wave dark matter","Two-photon annihilation leaves a loop-induced recoil signal"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The entire translation assumes that the di-photon operator is the only coupling between dark matter and nuclei and that the effective description with the new-physics scale above the top quark applies to both annihilation and scattering, even though the paper's own benchmark cross section would require a scale of order 10 GeV for p-wave annihilation while the plotted limits fix $M_* = 200$ GeV by hand.","fun_headline_variants_meta":{"raw":{"variants":["Loop-induced recoils test di-photon dark matter","Destructive interference shapes the dark matter recoil spectrum","Gamma-ray line coupling induces a nuclear recoil signal","Direct detection rivals line searches for p-wave dark matter","Two-photon annihilation leaves a loop-induced recoil signal"]},"model":"selfhosted/deepseek-v4.1-flash","effort":"low","attempts":1,"cost_usd":0.0,"raw_usage":{"total_tokens":1863,"prompt_tokens":945,"completion_tokens":918,"prompt_tokens_details":null,"completion_tokens_details":{"reasoning_tokens":854}},"tokens_in":945,"tokens_out":918,"duration_ms":20169,"image_pages":[],"temperature":1.0,"reasoning_budget":50,"reasoning_tokens":854,"transport_provider":"selfhosted","cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-11-10T14:06:16.099903+00:00","model_set":{"reader":"selfhosted/deepseek-v4.1-flash"},"falsifier":"A tonne-scale xenon detector running with the stated exposure and finding no events at the rate and recoil shape predicted in Figures 3 and 4 for fermionic dark matter with a large di-photon cross section near 130 GeV would remove the claimed reach, while a recoil spectrum showing the predicted deficit near threshold followed by a rising tail would confirm the interference.","supporting_citations":[{"cited_title":"How Dark Are Majorana WIMPs? Signals from MiDM and Rayleigh Dark Matter","cited_arxiv_id":"1206.2910","evidence_quote":"Supplies the Rayleigh-scattering matrix element and form factor for two photons scattering off the whole nucleus, the term that interferes with the mixing contribution."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the classic tree-level nucleon matrix element of the gluon field-strength operator used to evaluate the threshold-induced contribution."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the two-loop matching correction $delta_t$ included in the top-quark threshold matching of the gluon operator."},{"cited_title":"Fermi LAT Search for Dark Matter in Gamma-ray Lines and the Inclusive Photon Spectrum","cited_arxiv_id":"1205.2739","evidence_quote":"The Fermi LAT line-search upper limit that the direct detection bound is compared against in Figure 4."},{"cited_title":"A Tentative Gamma-Ray Line from Dark Matter Annihilation at the Fermi Large Area Telescope","cited_arxiv_id":"1204.2797","evidence_quote":"The tentative 130 GeV line analysis whose parameter region is shown as the target box in the projected-sensitivity plot."},{"cited_title":"Dark Matter Results from 225 Live Days of XENON100 Data","cited_arxiv_id":"1207.5988","evidence_quote":"The XENON100 limit on the spin-independent cross section that is converted into the bound on the operator coefficient and hence on the annihilation rate."},{"cited_title":"Yellin, Phys","cited_arxiv_id":null,"evidence_quote":"The maximum gap statistical method used to derive the limits from the XENON100 and CDMS-II data."},{"cited_title":"Nucleon strange quark content in 2+1-flavor QCD","cited_arxiv_id":"1012.1907","evidence_quote":"The lattice value of the strange-quark scalar form factor $f_s^{NT}$ adopted in the numerical evaluation of the mixing contribution."},{"cited_title":"Dark Matter Results from 100 Live Days of XENON100 Data","cited_arxiv_id":"1104.2549","evidence_quote":"The relative scintillation efficiency curve for xenon used to build the detector response function and the recoil-energy-averaged form factors."}],"review_version":1}