{"id":"a03933d2-d1d7-48e5-b6c2-3d4678892d80","arxiv_id":"1908.06986","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":8,"one_line_summary":"Adding a long-range interaction between a millicharged dark matter subcomponent and the dominant cold dark matter lets the dark sector act as a heat sink, reviving the millicharged explanation of the EDGES 21-cm anomaly.","lead":"This paper proposes that a small fraction of dark matter carrying electric charge, plus an extra long-range force linking it to the rest of the dark matter, can cool hydrogen gas in the early universe enough to explain the anomalous 21-cm absorption seen by EDGES. It maps out newly allowed masses and fractions and predicts signals for upcoming dark matter detectors.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The quantitative contours of Fig. 2, especially the high-mass and low-f_m extensions, rely on the untested assumption that non-Born mDM-atom scattering can only enhance the cooling rate; that sign is not guaranteed and should be computed.","rationale":"I agree with the reader's weakest_assumption. The paper's most distinctive quantitative claims, namely that mDM masses extend to a few hundred GeV and that fractions as small as 10^-8 can explain EDGES, rely on the cooling from mDM-neutral atom scattering. The authors explicitly note in App. B2 that the Born approximation fails in exactly the low-velocity regime where this process matters, and then invoke an unproven sign for the correction. This is a genuine correctness risk rather than a mere caveat, because the direction of the correction determines whether the plotted contours are conservative or over-optimistic. The CMB recast in App. C2 is also heuristic, but the authors themselves state that a full treatment is likely to relax the constraints, so it errs on the conservative side; the Born assumption, by contrast, is asserted to err by enlarging the allowed region, but without a supporting calculation. The central mechanism, adding a CDM heat bath coupled long-range to mDM, is plausible and the paper gives a clear analytic picture (Secs. III and IVB) that does not depend on the atom-scattering details. The direct detection predictions in Fig. 4 are illustrative and will be tested regardless. Therefore the appropriate verdict remains CONDITIONAL, as the reader set: the qualitative idea is sound, but the quantitative contours should be treated as indicative until the non-Born atom scattering is computed. No change to the reader's verdict is needed.","tokens_in":49375,"tokens_out":18099,"duration_ms":214401,"concrete_test":"Run a full partial-wave close-coupling calculation of mDM scattering on hydrogen and helium with the potential of Eq. B13, for the Fig. 3 benchmark and for representative (Q, m_m) points along the fm = 10^-8, 10^-6, and 10^-4 contours of Fig. 2. Compute sigma_T(v_rel) for v_rel between 10^-5 and 10^-7, compare with the Born expression, and recompute the Fig. 2 contours using the non-Born rates; if the non-Born rates are ever lower, the plotted viable parameter space should be revised.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing concern is the treatment of mDM scattering off neutral hydrogen and helium beyond the Born approximation. In Sec. IVD and App. B2, the authors compute the atom-mDM rates with Born form factors, note that the Born approximation breaks down when Z Q alpha_EM / v_rel log(...) >= 1 (Eq. B16), and then assert that the full quantum treatment, including resonances and bound states, 'will only enhance the mDM-baryon scattering rate, thereby enlarging the viable parameter space further.' This sign assumption is structurally important because Sec. IVD states that atom scattering changes the required Q by O(5) for mDM masses above 100 MeV, and the fm = 10^-8 contour in Fig. 2 sits at relatively large Q where the perturbative expansion is badly violated at the low velocities relevant near z ~ 17. The claimed extension to masses up to 200 GeV and fractions as low as 10^-8 is therefore quantitatively anchored to an untested approximation. The sign is not a theorem: low-energy scattering off a finite-range neutral-atom potential can saturate or even decrease relative to a v^-4 Born extrapolation (e.g., if no near-threshold resonance is present, or if inelastic ionization channels deplete the elastic channel). A full quantum computation could confirm the enhancement, but the paper does not provide one, and the statement that corrections can only increase the rate is an assumption rather than a derivation.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a dark sector in which a small millicharged dark matter (mDM) component is coupled to the dominant cold dark matter (CDM) component through a new long-range force. The CDM acts as a large heat sink, making baryon cooling at cosmic dawn much more efficient than in the standard mDM scenario. Using analytic scaling arguments and a numerical three-fluid treatment based on a published code (Ref. [59]), the authors show that the mDM mass range extends to roughly 10 MeV–200 GeV, the mDM fraction can be as small as 10^-8, and the scenario can explain the EDGES 99% CL absorption feature. They further derive a loop-induced CDM–electron scattering cross section that falls within the reach of near-future low-threshold direct detection experiments.","tokens_in":49714,"tokens_out":5223,"duration_ms":58896,"significance":"If established, this is a significant contribution: it revives millicharged DM as a viable explanation of the EDGES anomaly while opening a broad and testable parameter region, with distinctive signatures in beam-dump and low-threshold direct-detection experiments. The analytic parametric understanding of the heat-capacity mechanism is a clear strength, and the numerical framework is grounded in a published code. The authors are also transparent about the main limitations of their analysis, including the Born approximation for atom scattering and the heuristic nature of the CMB constraint recast. The central heat-capacity argument is robust. However, two load-bearing approximations currently control the quantitative contours in Fig. 2 and the direct-detection predictions in Fig. 4, so the quantitative claims should be strengthened before the paper reaches its final form.","major_comments":[{"comment":"The statement that beyond-Born corrections to mDM–neutral atom scattering 'will only enhance the mDM-baryon scattering rate, thereby enlarging the viable parameter space further' is an assumption, not a derivation. The Born self-consistency condition in Eq. (B16) is badly violated at the low relative velocities relevant near z ~ 17 for the large-Q, heavy-mDM region, and Sec. IVD states that atom scattering changes the required Q by O(5) for mDM masses above 100 MeV. Low-energy scattering off a finite-range neutral atom need not follow the v^-4 Born extrapolation: without a near-threshold resonance the rate can saturate, and inelastic channels can deplete the elastic channel. Since the fm = 10^-8 contour in Fig. 2 and the claimed extension to masses of ~200 GeV rely on this sign, a full quantum computation, or at least a concrete estimate of the size and sign of the correction, is needed before the quantitative claims can be taken as established. This does not undercut the heat-capacity argument itself, but it directly controls the outer contours of the claimed parameter space.","section":"Sec. IVD and App. B2"},{"comment":"The recast CMB constraint relies on the 'heuristic' 0.1 safety factor introduced after Eq. (C20). This bound fixes the maximal αmαC used for the Fig. 2 contours and, via Eq. (21), the CDM direct-detection cross sections shown in Fig. 4. Because no dedicated CMB analysis is provided, the quantitative location of the contours and the predicted CDM cross sections carry an unquantified systematic. The paper should either provide a dedicated CMB analysis or a sensitivity study showing how Fig. 2 and Fig. 4 shift when the 0.1 factor is varied over a reasonable range (for example, from 1 to 0.01). A statement that a full analysis would 'likely relax' the bound is useful context but does not quantify the impact on the results.","section":"App. C2, Eqs. (C17)–(C20), and Fig. 2"}],"minor_comments":[{"comment":"The definition of αmαC|max would be clearer if the heuristic 0.1 safety factor appeared explicitly in the equation itself rather than only in the surrounding text.","section":"Around Eq. (C20)"},{"comment":"The dashed green line is described as an extrapolation of the results of Ref. [47] to higher masses; stating the assumed energy-loss model and its uncertainty in the caption would improve reproducibility.","section":"Fig. 2, dashed green line"},{"comment":"The direct-detection cross section in Eq. (21) is a benchmark-dependent quantity: the numerical values in Fig. 4 assume a specific UV completion (m+ = 10 TeV in the left panel, or a 1% mass splitting) and the maximal αmαC allowed by the heuristic CMB recast. The text states these assumptions, but the abstract's 'testable predictions' could be read more strongly; a qualifier such as 'within the benchmark models considered' would be appropriate.","section":"Eq. (21) and abstract"}],"recommendation":"major_revision","confidential_remarks":"No additional confidential concerns beyond the major comments; the manuscript fits the scope of the journal, and the requested changes are quantitative and technical rather than conceptual."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing you should know: the central trick in this paper is real. The authors couple a tiny millicharged DM component to the dominant CDM bath through a long-range force, converting the CDM into a heat sink. That is the piece missing in earlier mDM explanations of EDGES, which were closed off by the heat capacity of the mDM fluid itself. With the CDM bath absorbing the heat, the mDM mass can go up to about 200 GeV and the fraction down to 1e-8. That is a genuine extension, not a cosmetic one.\n\nThe paper does several things well. The analytic parametric arguments in Secs. III and IVB explain why the heat-capacity bound changes, and they match the numerical treatment, which is built on a published code (Ref. [59]). The experimental section connects the parameter space to a broad set of beam dump, collider, and low-threshold direct detection searches, and the loop-induced CDM signal in Sec. V gives a concrete near-term target. The authors also flag their own approximations more than most papers do.\n\nThe soft spots are real but not fatal. The main one is the treatment of mDM scattering off neutral hydrogen and helium beyond the Born approximation (Sec. IVD and App. B2). They compute Born rates, note the breakdown condition, and assert that the full quantum treatment—resonances, bound states—'will only enhance' the rate. That sign is not guaranteed. Low-energy scattering can saturate or even suppress relative to a v^-4 Born extrapolation. Since this affects the required Q by O(5) for mDM masses above 100 MeV, the high-mass and very-low-fm contours in Fig. 2 are softer than the plot suggests. I don't think it breaks the paper, because the CDM heat-sink mechanism does not depend on that sign, but the contours should be read as indicative, not definitive.\n\nThe CMB recast in App. C2 also uses a heuristic 0.1 safety factor (Eq. C20); the authors admit this. And Q is fit to EDGES, so the 'predictions' for direct detection are partly downstream of the fit. Those are normal limitations for this kind of model-building paper, and they are disclosed.\n\nBottom line: this is a solid, honest theory paper with a new idea and a concrete experimental program attached. It deserves peer review. My main request to the authors would be to either compute the beyond-Born atom scattering or state plainly that the sign is an assumption. If they do that, the paper is publishable essentially as is.","headline":"The CDM heat-sink mechanism is a genuinely new idea that deserves referee time, but the high-mass contours rest on an unproven sign assumption about non-Born atom scattering.","tokens_in":50289,"tokens_out":2532,"would_cite":true,"duration_ms":26631,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A long-range force between a tiny millicharged dark-matter fraction and a cold dark-matter bath can cool the early-universe gas enough to explain the EDGES 21-cm absorption anomaly.","keywords":["millicharged dark matter","21-cm cosmology","EDGES anomaly","dark matter-baryon scattering","long-range dark force","cosmic dawn","direct detection","heat capacity"],"falsifier":"A full quantum-mechanical computation of mDM scattering off neutral hydrogen and helium at cosmic-dawn relative velocities ($v_{\\rm rel}\\sim 10^{-6}$) that yields a cooling rate below the Born-approximation estimate would falsify the paper's EDGES-fit contours; conversely, a null SENSEI-100g one-year search at the predicted loop-induced CDM-electron cross section in the 10 MeV-to-few-GeV CDM mass window would rule out the benchmark vector-portal realization.","tokens_in":49136,"feed_emoji":"📡","tokens_out":16800,"duration_ms":155020,"temperature":0.7,"pith_summary":"Millicharged dark matter by itself cannot easily cool the ordinary gas of hydrogen and helium at cosmic dawn, because cosmological bounds force its abundance to be tiny and its mass into a narrow window. This paper argues that giving that millicharged subcomponent a long-range force to the dominant cold dark matter (CDM) bath changes the picture: the much larger CDM number density acts as a heat sink, so the millicharged particles need only ferry heat from the gas to the CDM. The authors show that this three-bath setup can produce the at least roughly 1.7 K of extra gas cooling that the EDGES 21-cm absorption measurement at $z\\simeq 17$ would require, while extending the allowed millicharged mass range to $10$ MeV--a few hundred GeV and its dark-matter fraction down to $10^{-8}$. If correct, the scenario makes concrete predictions: CDM with mass between $10$ MeV and a few GeV, and a CDM-electron scattering cross section generated by a loop of millicharged particles that near-future low-threshold direct-detection experiments can reach.","feed_headline":"A cold dark-matter bath lets millicharged DM explain EDGES","feed_subtitle":"A tiny charged dark-matter fraction can cool the cosmic gas by dumping heat into a cold dark-matter bath.","key_machinery":"The load-bearing device is the replacement of the two-bath cooling problem with a three-bath one. The paper writes temperature and bulk-velocity evolution equations for baryons, mDM, and CDM, with transfer cross sections of the Rutherford form $\\sigma_T\\propto v_{\\rm rel}^{-4}$ for both mDM-baryon and mDM-CDM scattering; the mDM-baryon cross section is regulated by the plasma Debye mass, the mDM-CDM cross section by a sub-keV dark mediator. The key parametric identity is the heat-capacity bound: the cooling bath must stay cool enough to absorb the heat. In the standard mDM scenario this bound gives $m_m\\lesssim 67$ MeV at $f_m=0.4\\%$; with the CDM bath as heat sink the analytic bound becomes $m_C\\lesssim 18$ GeV, and the full numerical treatment places the viable CDM mass around a few GeV. A second ingredient is the loop-induced CDM-SM coupling: in a vector-portal realization, a loop of mDM generates mixing between the dark mediator and the photon, giving a CDM-electron cross section that depends on the millicharged mass splitting and sets the direct-detection target.","core_discovery":"The central claim is that a millicharged dark-matter (mDM) subcomponent that also interacts with the dominant cold dark-matter (CDM) bath through a new long-range force can cool baryons efficiently enough to explain the anomalous 21-cm absorption reported by EDGES, without violating CMB, BBN, or collider bounds. The mechanism is a three-fluid thermal system: Rutherford-like $\\sigma_T\\propto v_{\\rm rel}^{-4}$ scattering moves heat from baryons to mDM, and the same velocity-enhanced scattering moves it onward to a CDM bath whose much larger number density supplies the heat capacity. With the mDM-CDM coupling set below the CMB drag bound, the allowed mDM mass extends to $m_m\\lesssim 200$ GeV and the mDM fraction down to $f_m\\sim 10^{-8}$, while the CDM mass is bracketed by BBN at $m_C\\gtrsim 10$ MeV and by a heat-capacity bound near a few GeV. The CDM acquires an electron-scattering cross section through an mDM loop that, in the minimal vector-portal model, lies below current XENON-10 and SENSEI limits but within reach of future SENSEI and DAMIC exposures.","pith_inferences":["Beyond the paper: the heat-sink trick should generalize to any subdominant dark-matter component with a velocity-enhanced scattering rate to baryons, not only a millicharged one, as long as a dominant cold bath with large number density is present.","Beyond the paper: the loop-induced CDM cross section's sensitivity to the millicharged mass splitting means a positive CDM direct-detection signal would constrain the internal spectrum of the millicharged sector, turning a cosmology anomaly into a spectroscopy tool.","Beyond the paper: the paper saturates the mDM-CDM coupling at the CMB bound when presenting direct-detection targets; if that coupling is smaller by orders of magnitude, the predicted CDM cross section drops by up to two orders of magnitude, so a null direct-detection result would not rule out the cooling mechanism itself.","Beyond the paper: a future measurement of the full 21-cm absorption trough shape, not just its depth, could distinguish this three-bath cooling from an excess radio background or other new-physics explanations, because the cooling turns on sharply when the ionization fraction falls."],"forward_implications":["If the scenario is correct, the millicharged component can be as rare as $10^{-8}$ of the dark matter and as heavy as roughly 200 GeV, with charges $Q$ between $10^{-5}$ and 1 in a window above direct-detection limits and below collider bounds.","The dominant cold dark matter must weigh between about 10 MeV and a few GeV, because a lighter or heavier CDM cannot satisfy BBN and simultaneously absorb enough heat to produce the required baryon cooling.","The mDM loop that gives CDM its Standard-Model coupling produces a CDM-electron recoil cross section that near-future SENSEI and DAMIC exposures should be able to probe or exclude in minimal models.","Beam-dump and collider searches for millicharged particles, together with balloon- or satellite-borne single-electron-threshold detectors, can cover much of the allowed charge-mass plane, so the explanation is testable rather than merely consistent.","The same long-range mDM-CDM force induces CDM self-interactions, giving an independent observable consequence bounded by colliding-cluster and cluster-ellipticity constraints."],"supporting_citations":[{"why":"the EDGES sky-averaged 21-cm absorption measurement that defines the anomaly the paper seeks to explain.","marker":"[20]"},{"why":"supplies the two-bath heat-transfer and drag formalism that the paper generalizes to three fluids.","marker":"[1]"},{"why":"connects DM-baryon scattering to the 21-cm signal and computes mDM-atom scattering rates extended here.","marker":"[2]"},{"why":"derives the cosmological bounds on the standard mDM scenario that this setup is designed to evade.","marker":"[7]"},{"why":"computes the mDM-baryon transfer cross section and cooling used as the baseline for the EDGES anomaly.","marker":"[8]"},{"why":"quantifies the combined BBN, CMB, and $N_{\\rm eff}$ constraints on standard mDM that set the mass and fraction floors.","marker":"[10]"},{"why":"CMB power-spectrum constraints recast into the bound on the mDM-CDM coupling strength in Eq. (19).","marker":"[14–17]"},{"why":"numerical treatment of temperature, ionization, and bulk-velocity evolution used for the full parameter scans.","marker":"[59]"},{"why":"provides the direct-detection reach for mDM including atmospheric energy losses, defining the lower-charge boundary in Fig. 2.","marker":"[47]"},{"why":"XENON-10 electron-recoil limit that the loop-induced CDM cross section must beat, setting the direct-detection baseline.","marker":"[22]"}],"fun_headline_variants":["Millicharged DM cools cosmic gas via a dark bath","New dark force widens millicharged DM's cosmic range","Tiny charged DM + cold bath explains EDGES absorption","Dark bath revives millicharged DM for 21-cm signals"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The paper assumes that the breakdown of the Born approximation for millicharged particles scattering off neutral hydrogen and helium, where resonances and bound states can form, will only increase the low-velocity cooling rate and therefore enlarge the viable parameter space; if a full quantum treatment instead suppressed that rate, the EDGES-fit contours and the derived direct-detection predictions would shift.","fun_headline_variants_meta":{"raw":{"variants":["Millicharged DM cools cosmic gas via a dark bath","New dark force widens millicharged DM's cosmic range","Tiny charged DM + cold bath explains EDGES absorption","Dark bath revives millicharged DM for 21-cm signals"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00058,"raw_usage":{"total_tokens":2798,"prompt_tokens":1077,"completion_tokens":1721,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":693,"completion_tokens_details":{"reasoning_tokens":1646}},"tokens_in":693,"tokens_out":1721,"duration_ms":13551,"temperature":1.0,"reasoning_tokens":1646,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:29:28.790695+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A full quantum-mechanical computation of mDM scattering off neutral hydrogen and helium at cosmic-dawn relative velocities ($v_{\\rm rel}\\sim 10^{-6}$) that yields a cooling rate below the Born-approximation estimate would falsify the paper's EDGES-fit contours; conversely, a null SENSEI-100g one-year search at the predicted loop-induced CDM-electron cross section in the 10 MeV-to-few-GeV CDM mass window would rule out the benchmark vector-portal realization.","supporting_citations":[],"review_version":1}