{"id":"3d965d0d-6f5c-4d46-b130-b25fea964ac0","arxiv_id":"2506.00115","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":8,"one_line_summary":"A disordered dielectric powder can act as a broadband dark-matter-to-photon conversion target, and the proposed DPHaSE experiment could probe QCD axions and dark photons in the 10 meV to 1 eV range.","lead":"This paper proposes a dark matter detector built from a container of ordinary dielectric powder, where axion or dark photon dark matter converts into photons at the surfaces of the powder grains. A smart generalist should read it because the design is cheap, broadband, and could, if it works, probe dark matter masses from 10 meV to 1 eV far beyond current experiments.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central reach assumes O(10^15)-particle powder behaves like O(100)-particle simulations; bandgap and localization suppression is explicitly acknowledged but not quantified, so the large-volume extrapolation is the load-bearing assumption.","rationale":"The paper's single-object and multiple-scattering derivations are coherent, and the small-volume 2D and 3D simulations do support the area-law scaling within their simulated regime. The reader's CONDITIONAL verdict is appropriate because the projected reach depends on an extrapolation from O(100) particles to O(10^15) particles. The authors themselves flag the two dominant risks, photonic bandgaps and Anderson localization, in Sec. III E and again in Sec. VI, and they propose transmission measurements as a calibration path. That is an honest limitation rather than a hidden error, and no internal inconsistency or by-construction circular step was found. The central claim remains plausible but not fully established at the experimental scale, so the verdict should remain CONDITIONAL rather than being upgraded to ACCEPT or downgraded to REJECT. The proposed scaling simulation is a direct, feasible check of whether the small-volume behavior survives at larger system sizes, and the experimental transmission measurement would ultimately settle whether localization or bandgap suppression degrades the effective volume.","tokens_in":48162,"tokens_out":13672,"duration_ms":149917,"concrete_test":"Run configuration-averaged 3D finite-element simulations at fixed f = 0.5, n = 1.77, and the same polydisperse radius distribution, for boxes with N ≈ 200, 400, 800, and 1600 spheres. Compute the normalized geometrical-regime power P/(N P0) at λ/⟨R⟩ = 0.3, 0.5, and 1.0. If the plateau value changes by more than about 20% with N, or if frequency gaps develop as the box grows, the small-volume calibration is not representative of the full powder; if the plateau is stable, the extrapolation gains support but still needs the experimental diffuse-transmission check proposed in Sec. IV B.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The weakest point is the leap from small simulated volumes to the full DPHaSE target. Equation (30) and Fig. 8 assume that the volume-averaged conversion power measured for up to 500 fibers (2D) and 189 spheres (3D) continues to scale linearly with particle number in a 10^3 cm^3 powder containing roughly 10^15 particles, with no Anderson localization or photonic-bandgap suppression in the search bands. This is not a technicality: in the geometrical regime the signal photons must diffuse out of the powder before detection, and Veff in Eqs. (36)-(40) assumes diffusive transport with a well-defined transport mean free path. If localization or a bandgap reduces the local density of optical states or the diffusion constant at any frequency in the 1-120 µm range, the photons are either not produced or cannot reach the photon collection chamber, and the projected sensitivity in Fig. 8 is overestimated. The authors explicitly concede in Sec. III E that they 'might have underestimated the significance' of bandgaps and localization, and in Sec. VI that 'there can be gaps in the sensitivity, in particular around λ ≃ ⟨R⟩.' The assertion in Sec. III E that interference effects 'remain irrelevant' for larger volumes is not derived or simulated; the 3D results are single configurations with visible box-mode oscillations at long wavelengths, so the smooth transition at λT ≃ 3⟨R⟩ used for G(f) is not ensemble-validated at the target scale.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a new broadband dark-matter-to-photon conversion target: a volume filled with disordered dielectric powder (spheres or aligned fibers), and an experimental design (DPHaSE) pairing this target with an SNSPD. The theoretical core is the derivation of single-scatterer conversion power for a dielectric fiber and a dielectric sphere, the area-law scaling of the frequency-integrated power with total interface area, and its extension to incoherent sums over many particles. The authors then present multiple-scattering T-matrix simulations for up to 500 fibers in 2D and COMSOL simulations for up to 189 spheres in 3D, from which they extract a filling-factor-dependent correction G(f) and a geometric-to-Rayleigh transition at lambda_T ~ 3<R>. A diffusion model for photon transport in the powder gives an effective volume Veff for a cavity-coupled sensor, and the projected sensitivity is presented in Fig. 8 for three wavelength bands, claiming up to five orders of magnitude improvement over existing dark-photon constraints and sensitivity to QCD axion couplings in the 10 meV-eV range.","tokens_in":48494,"tokens_out":2689,"duration_ms":34360,"significance":"The single-scatterer analysis and the 2D multi-fiber formalism in Secs. III A-III C and Appendix B are internally consistent and provide a clean demonstration that interface conversion in disordered media obeys an area-law scaling in the geometric regime. The proposal's appeal is genuine: commercially available powders, no resonant tuning, and a simple calibration strategy via optical transmission measurements. The paper is also honest about its main weaknesses: Sec. III E explicitly states that photonic bandgaps and Anderson localization might have been underestimated, and Sec. VI admits possible sensitivity gaps at lambda ~ <R>. However, the reach curves in Fig. 8 and Eq. (42) rely on assumptions that are not validated at the target scale: that O(100)-particle simulations remain representative for a 10^15-particle powder, that the simulation-calibrated G(f) and the immediate transition at lambda_T ~ 3<R> persist, and that diffusive transport with a well-defined transport mean free path holds in the entire search band. If these assumptions fail, the projected sensitivity is correspondingly overestimated.","major_comments":[{"comment":"The large-volume extrapolation from O(100) simulated particles to a 10^15-particle powder is load-bearing for the projected reach, and it is not demonstrated. Sec. III E concedes that photonic bandgaps and Anderson localization may be significant and that G(f) cannot be predicted at large N, and Sec. VI concedes that 'there can be gaps in the sensitivity, in particular around lambda ~ <R>'. Localization or a bandgap would suppress either the local density of optical states (reducing production) or the diffusion constant used in Veff in Eqs. (36)-(40), so the smooth reach curves in Fig. 8 would be overestimated. The authors should either quantify these effects (e.g., with larger-scale numerical studies or a localization criterion based on the Ioffe-Regel parameter) or explicitly reframe the central claim as a proof-of-principle whose broadband reach remains to be established by prototype measurements.","section":"Sec. III E; Sec. VI; Fig. 8"},{"comment":"The 3D numerical evidence for the smooth area-law claim rests on only three configurations, one per filling factor, with no ensemble averaging and no error bars. The paper notes that Fig. 6(c) shows noticeable fluctuations and box-mode oscillations, and in Sec. III E it asserts that interference effects 'remain irrelevant' for larger volumes, but this assertion is not derived or simulated. At minimum the authors should present multiple independent configurations for at least one filling factor and show that the configuration-averaged power converges to the incoherent-sum result in the geometric regime before using that regime to anchor the reach calculation.","section":"Sec. III D; Fig. 6(c)"},{"comment":"The headline reach depends on the simulation-calibrated G(f) and on the assumed transition at lambda_T ~ 3<R> with an immediate transition between the geometric and Rayleigh regimes. These are not parameter-free predictions: G(f) is extracted from the 3D COMSOL data in Fig. 6(c), and the 'immediate transition' is a modeling choice, not a measured or derived property. The manuscript should state how sensitive the reach curves in Fig. 8 are to these choices, e.g., by showing a band of reach curves for plausible variations of G(f) and lambda_T, or by presenting the calibration procedure that would determine them before the claimed sensitivity is used.","section":"Eq. (30); Sec. VI; Fig. 8"},{"comment":"The diffusion model assumes a well-defined transport mean free path l_s* and absorption length l_a throughout the search band. For band C in particular, the absorption length in Eq. (41) is based on bulk high-resistivity silicon and the paper acknowledges that cryogenic powder absorption lengths are largely unmeasured. Since Veff in Eqs. (36)-(40) is directly proportional to l_a, the projected reach in the far-IR band is sensitive to an assumed material property for which the paper cites no powder-specific measurement. This should be presented as an assumed parameter with a clear uncertainty estimate, not as a baseline sensitivity.","section":"Sec. IV A; Sec. V A"}],"minor_comments":[{"comment":"The text says band A covers Compton wavelengths 1-6 microns, but Table III lists the band A bandwidth as 1-4 microns; these values should be made consistent.","section":"Sec. V A; Table III"},{"comment":"In the paragraph before Eq. (42), the text refers to 'the blue and red lines', but the three reach curves in Fig. 8 are described as blue, magenta, and green in both the figure caption and the surrounding text; the color reference should be corrected.","section":"Sec. VI"},{"comment":"The same symbol P is used for the power of a single particle and for the power of many particles, despite a statement that the two are distinguished; in several places (e.g., Eq. (23) and surrounding text) the distinction is easy to miss. A different symbol for the summed power would improve readability.","section":"Sec. III A; Sec. III B"},{"comment":"The word 'repectively' appears in the sentence describing Eq. (31); this is a typo.","section":"Sec. IV B"},{"comment":"The abstract and conclusions state that the reach 'exceeds current constraints on dark photon dark matter by up to 5 orders of magnitude'; this wording should be conditioned on the Phase II assumptions and on the unresolved large-volume questions raised in Sec. III E, to avoid implying a demonstrated sensitivity rather than a projection.","section":"Sec. I; Sec. VII"}],"recommendation":"major_revision","confidential_remarks":"The paper is a strong proof-of-principle with a clean analytical core, but the projected sensitivity claims are currently supported by a combination of small-volume simulations and optimistic assumptions about large-volume behavior. The authors may be able to address this within the manuscript's scope by softening the reach claims, adding configuration-averaged 3D results, and quantifying the sensitivity to G(f), lambda_T, and the powder absorption lengths. If the central claim is reframed as a proposal with unproven broadband performance, the manuscript would be much more defensible."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. First, this is the first paper I have seen that proposes a disordered dielectric powder as a broadband conversion target, replacing interference-engineered stacks with randomness. Second, the load-bearing move is the large-volume extrapolation, and the authors know it: in Sec. III E they say they might have underestimated the significance of photonic bandgaps and Anderson localization, and in Sec. VI they concede there can be sensitivity gaps near lambda ~ <R>. Read the headline reach curves as projections, not predictions.\n\nWhat is actually new and solid: the single-fiber and single-sphere derivations in App. B are careful and internally consistent, and the disorder-averaged area-law scaling in Eq. (30) is a clean, parameter-free result in the geometric regime. The 2D multiple-scattering T-matrix work is well done, and the COMSOL 3D results match the semi-analytical formula's functional form for the simulated distributions. The calibration strategy—measuring diffuse reflectance and transmittance to fix G(f) and check for localization gaps—is genuinely clever, because light-scattering calibration sidesteps the DM couplings entirely. The proposal itself (powder, SNSPD, veto) is concrete and cheap relative to resonant haloscopes.\n\nWhere it is soft. The simulations cover up to 500 fibers in 2D and 189 spheres in 3D; the experiment is 10^15 particles. The claim in Sec. III E that interference effects remain irrelevant for larger volumes is asserted, not derived. The stress-test note is right that localization, ubiquitous in 2D and possible in 3D in the Mie-resonance regime, could either suppress photon production or trap signal photons in the powder; both kill the projected sensitivity in Fig. 8. Also, G(f) is calibrated from the authors' own COMSOL data with an assumed transition at lambda_T ~ 3<R>, so the headline curves are not independent of the simulation. Cryogenic powder absorption lengths are assumed at 10-100 m, while powder surface absorption can be far worse than bulk; the authors flag that too. These are acknowledged in text, which counts in their favor, but they are still the difference between a proof of concept and a detector proposal.\n\nWho gets value: anyone working on axion or dark photon detection in the 10 meV to 1 eV range, plus people in disordered optics who want a new application. It deserves a serious referee. The paper is internally coherent, honestly flagged, and potentially important; a referee should push hard on localization and bandgap physics and demand a prototype-scale calibration before the reach claims are taken at face value.","headline":"A genuinely new broadband haloscope idea with clean small-scale derivations, but the headline reach rests on the unvalidated leap from O(100) simulated particles to a 10^15-particle powder.","tokens_in":49046,"tokens_out":2116,"would_cite":true,"duration_ms":24769,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["95.35.+d","14.80.Va","42.25.Dd","85.25.Pb"],"model":"deepseek-v4-flash","headline":"This paper argues that a randomly packed dielectric powder converts axion and dark-photon dark matter into photons with broadband power set by total interface area, and that such a powder can serve as a practical detector in the 10 meV to…","keywords":["dark matter","axion","dark photon","dielectric haloscope","interface conversion","disordered media","SNSPD","broadband detector"],"falsifier":"Measure the diffuse transmission and reflectance of a centimeter-scale powder slab across wavelengths from $0.1\\langle R\\rangle$ to $10\\langle R\\rangle$ and compare the inferred transport lengths with the dark-matter conversion power of Eq. (30): if at $\\lambda\\simeq\\langle R\\rangle$ the transmission drops to near zero, or the calibrated conversion power falls below the surface-area scaling by more than the expected absorption loss, the large-volume extrapolation fails.","tokens_in":47943,"feed_emoji":"🔍","tokens_out":10068,"duration_ms":107419,"temperature":0.7,"pith_summary":"The paper proposes a dark-matter detector in which a volume filled with randomly packed dielectric powder, or aligned fibers, acts as the conversion target for light bosonic dark matter. Its central claim is that dark-matter-to-photon conversion happens at every dielectric-vacuum interface, so the broadband conversion power is set by the total interface area of the powder rather than by a finely tuned resonance. The authors derive this scaling for isolated spheres and fibers, verify it in small disordered 2D and 3D samples with multiple-scattering and finite-element calculations, and summarize the result in a compact formula for the total conversion power. They then design a concrete experiment, DPHaSE, coupling a powder shell to a superconducting nanowire single-photon detector, with projected sensitivity to QCD axion couplings and dark photon kinetic mixing in the 10 meV to 1 eV range.","feed_headline":"Powder haloscope broadens dark-matter search to meV-eV masses","feed_subtitle":"Conversion power scales with total powder surface area; projected reach beats dark-photon limits by up to 5 orders of magnitude.","key_machinery":"The central object is the dielectric-vacuum interface treated as a conversion surface: each interface radiates photons when driven by a dark-matter-induced current, and in a powder with many randomly placed particles of varied radii the incoherent sum of these emitters produces the broadband power of Eq. (30). The supporting machinery is a semi-analytical multiple-scattering calculation for 2D fiber bundles, which solves a linear system for the outgoing-wave coefficients, together with finite-element simulations for 3D sphere packs and a diffusion model of photon transport that connects the collected power to bulk powder parameters such as the transport scattering length $\\ell_s^*$, absorption length $\\bar\\ell_a$, and diffusion length $\\ell_d$. Because the ratio $\\langle\\sigma_{\\rm DM}\\rangle/\\langle\\sigma_s\\rangle$ is frequency-independent in the geometrical and Rayleigh regimes, ordinary light-scattering measurements on the same powder calibrate the dark-matter conversion rate.","core_discovery":"The paper claims that a disordered dielectric medium converts dark matter to photons broadband, with total power $\\langle P\\rangle = G(f) (E_0^2/2) f V_{\\rm eff} (\\langle\\sigma_{\\rm DM}\\rangle/\\langle V_{\\rm particle}\\rangle)$, where $G(f)$ is a filling-factor factor close to unity in the geometrical regime, $E_0$ is the electric field induced by the axion or dark photon background, and $\\langle\\sigma_{\\rm DM}\\rangle/\\langle V_{\\rm particle}\\rangle$ is an effective conversion cross-section per particle volume that, in the geometrical regime, is proportional to the specific surface area of the powder. At wavelengths short compared with the particle radius, averaging over a spread of radii smooths out the interference oscillations seen for a single particle and leaves a smooth broadband response; at long wavelengths the signal falls into a Rayleigh tail suppressed by the filling-factor-dependent factor $G(f)$, which must be calibrated optically.","pith_inferences":["If the surface-area scaling persists to experimental volumes, the same reasoning could be pushed to smaller mean radii or higher filling factors to extend the geometric regime to shorter wavelengths, limited mainly by material loss and by the onset of photonic bandgaps or Anderson localization.","The identification of the conversion cross-section with the light-scattering cross-section suggests that existing diffuse-reflectance data for powdered materials could be screened to rank candidate target dielectrics without new dark-matter-specific measurements.","A direct testable extension is a prototype that measures the single-photon rate from an intense source placed at the same dielectric interfaces and checks that the rate grows linearly with total surface area in the way Eq. (30) predicts."],"forward_implications":["A fixed volume of powder converts dark matter with scanning speed proportional to total dielectric-vacuum interface area, making specific surface area the figure of merit rather than resonant quality factor.","The proposed DPHaSE design can reach QCD axion-photon couplings down to about 10 meV with a 10 T magnet, and dark-photon kinetic mixing down to $\\varepsilon\\sim 5\\times 10^{-16}$ with a 1 cm$^2$ sensor.","The projected sensitivity exceeds current dark-photon constraints by up to five orders of magnitude in the 10 meV to 1 eV band.","Target preparation reduces to purchasing or sieving low-loss dielectric powder, avoiding the nanofabrication and tuning required for half-wave dielectric stacks.","Because dark-matter conversion and ordinary light scattering share the same powder, the sensitivity curve can be calibrated with transmission and reflectance measurements before and during the search."],"supporting_citations":[{"why":"Derives the area law for dielectric haloscopes that underlies the interface-area scaling.","marker":"[51]"},{"why":"Supplies the dielectric-haloscope scanning-speed framework, the half-wave stack and chirped stack analysis, and the result that random 1D layer spacings give spiky power.","marker":"[55]"},{"why":"Proposes the dish-antenna broadband conversion target whose surface-area scaling the powder generalizes to arbitrary geometries.","marker":"[84]"},{"why":"Provides the scattering-matrix multiple-scattering method used for the 2D fiber calculations.","marker":"[90]"},{"why":"Provides the finite-element numerical scheme used for the 3D sphere simulations.","marker":"[96]"},{"why":"Reports the SNSPD sensor performance whose dark-count rate anchors the background and reach projections.","marker":"[56]"},{"why":"Documents localization and photonic bandgap effects and their calibration by transmission measurements, the main caveat to the broadband claim.","marker":"[101]"},{"why":"Supplies the alumina absorption length used to estimate the effective volume and projected reach.","marker":"[114]"}],"fun_headline_variants":["Disordered dielectric powder haloscope for broad mass range","Powder-based detector turns dark matter into photons","Dark-matter-to-light conversion via disordered powder","Haloscope using powder targets meV-eV dark matter","Powder detector beats dark-photon limits by 5 orders"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the smooth, surface-area-limited conversion seen in simulations with a few hundred particles carries over unchanged to a powder with up to $10^{15}$ particles, with no photonic bandgap or Anderson-localization suppression in the search bands; the authors flag this uncertainty explicitly in Sec. III E.","fun_headline_variants_meta":{"raw":{"variants":["Disordered dielectric powder haloscope for broad mass range","Powder-based detector turns dark matter into photons","Dark-matter-to-light conversion via disordered powder","Haloscope using powder targets meV-eV dark matter","Powder detector beats dark-photon limits by 5 orders"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00103,"raw_usage":{"total_tokens":4380,"prompt_tokens":1025,"completion_tokens":3355,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":641,"completion_tokens_details":{"reasoning_tokens":3278}},"tokens_in":641,"tokens_out":3355,"duration_ms":29329,"temperature":1.0,"reasoning_tokens":3278,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T12:11:55.754776+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the diffuse transmission and reflectance of a centimeter-scale powder slab across wavelengths from $0.1\\langle R\\rangle$ to $10\\langle R\\rangle$ and compare the inferred transport lengths with the dark-matter conversion power of Eq. (30): if at $\\lambda\\simeq\\langle R\\rangle$ the transmission drops to near zero, or the calibrated conversion power falls below the surface-area scaling by more than the expected absorption loss, the large-volume extrapolation fails.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the alumina absorption length used to estimate the effective volume and projected reach."}],"review_version":1}