{"id":"00140962-d1e3-4969-aabf-aae2b943bf6b","arxiv_id":"2505.13402","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"high","formal_verification":"none","parameter_count":5,"one_line_summary":"HydroX, a proposed hydrogen-doped liquid xenon TPC, projects sensitivity to sub-GeV dark matter through enhanced proton-recoil signals in a large xenon detector.","lead":"HydroX proposes adding hydrogen to liquid xenon dark matter detectors so that light dark matter particles can scatter off protons instead of heavy xenon nuclei. This could open a new search window for dark matter below 1 GeV, especially for spin-dependent interactions.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Sub-100 MeV reach rests on an unmeasured proton-recoil yield in LXe; the SRIM all-electronic-stopping assumption needs direct data before the projection is reliable.","rationale":"The reader's weakest_assumption identifies the same load-bearing spot: the unmeasured H-recoil yield in LXe. I agree. The paper is unusually explicit about this caveat ('There are no data measuring yields for H recoils in LXe'; 'projections should be taken as merely indicative'; 'S2-only sensitivities in particular should be viewed as speculative'). Those admissions reduce the overclaiming risk but do not supply the missing measurement. I considered two other candidate concerns. H2 solubility in LXe (Sec. 2.4) is also unmeasured, but it is backed by LUX helium/argon/krypton scaling and affects sensitivity mostly as a linear exposure factor. Accidental S2 backgrounds (Sec. 3.2) are acknowledged and would degrade sensitivity by roughly a factor of five in cross section without necessarily removing the mass reach. The H-yield assumption is more load-bearing because it sets the energy threshold, and the threshold is what makes the steeply rising low-mass WIMP spectrum accessible; it enters through the NEST/SRIM model in Fig. 3 and propagates to both analysis channels. The helium SRIM/NEST agreement and the alpha-recoil data are real but indirect support; they do not remove the need for direct proton-recoil data. The proposed neutron-scattering yield measurement would settle whether the factor-of-five enhancement is real. Until then, the concept is plausible but not demonstrated, so CONDITIONAL remains the right verdict and no change to the reader's recommendation is needed.","tokens_in":17508,"tokens_out":14639,"duration_ms":148340,"concrete_test":"Measure proton-recoil yields in LXe directly: scatter monoenergetic neutrons from a thin hydrogenous target inside a small LXe TPC, tag proton recoils in the 0.1-10 keV range, and extract absolute S1 and S2 yields. Compare these yields with the SRIM all-electronic-stopping curve in Fig. 1. If the measured yield is more than a factor of 2 below that curve at the energies that set the S2-only threshold in Fig. 3, recompute Fig. 5 with the measured yield and treat the <100 MeV S2-only reach as unsupported until confirmed.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central projection depends on the energy threshold for H recoils in LXe, and that threshold is taken from a SRIM-based model in which 'all of the electronic stopping goes into signal but none of the nuclear stopping does' (Sec. 2.2). The paper itself states 'There are no data measuring yields for H recoils in LXe'; the factor-of-five enhancement is supported only by a SRIM/NEST agreement for He recoils and an alpha-power-law extrapolation. For a keV-scale proton in xenon the reduced energy is about 0.15, near the nuclear-stopping maximum, so nuclear stopping is not negligible, and a dense proton track is subject to the same recombination and ionization-quenching effects that NEST already models for alpha particles. Because the low-mass WIMP spectrum rises steeply toward threshold, the S2-only 'well below 100 MeV' reach (Sec. 3.2) is very sensitive to this yield: a factor-of-two reduction in H yield roughly doubles the effective threshold and can push the mass reach above 100 MeV. This is compounded by the paper's own statement that S2-only sensitivities 'should be viewed as speculative' because accidental backgrounds are not modeled. The concern is not that the authors are wrong; it is that the headline reach currently rests on an unmeasured microphysics input that no amount of detector modeling can substitute for.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript proposes HydroX, a concept for doping a LZ-scale liquid xenon TPC with molecular hydrogen (1.1% mole fraction) to search for sub-GeV dark matter via proton recoils. The central physics argument (Sec. 2.2) is that a recoiling proton in LXe loses most of its energy to electronic stopping, producing roughly five times more detectable quanta per unit energy than a xenon recoil, and the authors incorporate a SRIM-based hydrogen yield into NEST together with the LZ background model from Ref. [88] to compute trigger efficiencies (Fig. 3) and 500-day WIMP sensitivity projections (Fig. 5). They report that a standard S1/S2 analysis reaches down to about 200 MeV in WIMP mass, while an S2-only analysis could reach well below 100 MeV, and that the spin-dependent proton channel would be uniquely sensitive in that mass range. The paper also discusses H2 solubility in LXe, S1/S2 quenching, tritium backgrounds, PMT permeation, and the cryogenic circulation challenges of introducing a non-condensible gas.","tokens_in":17729,"tokens_out":7333,"duration_ms":65224,"significance":"If the assumed hydrogen recoil yield and the ER-like S2/S1 partitioning are correct, HydroX would open a new and well-motivated window for sub-GeV dark matter searches using existing tonne-scale infrastructure, with particularly interesting spin-dependent proton sensitivity. The manuscript is valuable as a concept paper: it gives a clear physical motivation, identifies the R&D needed (H2 solubility, tritium removal, distillation, signal quenching), and is explicit about the absence of key measurements. The authors state plainly that there are no data on H recoil yields in LXe (Sec. 2.2), that H2 solubility in LXe is unmeasured (Sec. 2.4), and that S2-only sensitivities should be viewed as speculative because accidental backgrounds are not modeled (Sec. 3.2). These admissions are honest, but they mean that the quantitative reach in Fig. 5 is a scenario under optimistic microphysics assumptions rather than a robust prediction. No reproducible code or machine-checked derivations are provided, so the central value of the paper is the concept and the roadmap, not the precision of the sensitivity curves.","major_comments":[{"comment":"The factor-of-five enhancement in hydrogen recoil yield relative to xenon recoils is the load-bearing input for the energy thresholds in Fig. 3 and the sensitivity curves in Fig. 5. This enhancement is based entirely on a SRIM calculation in which all electronic stopping is assumed to produce signal and none of the nuclear stopping does, and the paper itself states 'There are no data measuring yields for H recoils in LXe.' For a few-keV proton in xenon the reduced energy is near the nuclear-stopping maximum, so this assumption is not a minor detail. Because the low-mass WIMP spectrum rises steeply toward threshold, a factor-of-two uncertainty in the H yield can substantially raise the effective threshold and plausibly move the S2-only reach above 100 MeV. The authors acknowledge in general terms that the sensitivity depends critically on the threshold (Sec. 3.2), but they do not propagate any uncertainty band from the yield model into Fig. 5. I request either a quantitative sensitivity band derived from plausible H-yield models or an explicit reframing of the mass-reach claims as conditional on a specific unverified microphysics model.","section":"Sec. 2.2, Fig. 1 and Fig. 3"},{"comment":"The 'well below 100 MeV' S2-only projection excludes accidental backgrounds, and the text states both that 'this analysis does not include accidental events' and that 'the S2-only sensitivities in particular should be viewed as speculative.' Since S2-only analyses abandon ER/NR discrimination, accidental coincidences are generally the limiting background in this mode, as the paper itself notes by citing the LZ observation of about five accidentals in the relevant energy range. This omission is load-bearing for the headline sub-100 MeV claim. The abstract and conclusion should not present this reach without the accidental-background caveat, and ideally the authors should add at least an order-of-magnitude estimate of the accidental rate for the proposed threshold and drift geometry.","section":"Sec. 3.2, Fig. 5"},{"comment":"The sensitivity model assumes that the S2/S1 partition of hydrogen recoils is similar to electronic recoils, based on high-energy alpha data, but the only low-energy measurement of a light-ion recoil in LXe (helium recoils, Ref. [40]) shows xenon-like discrimination. The authors acknowledge that if hydrogen recoils are instead xenon-recoil-like, the S1/S2 sensitivity improves by up to a factor of two at low mass but the dominant background becomes 8B neutrino-xenon scattering, which degrades the absolute cross-section sensitivity. This is a second unmeasured microphysics input with a direct effect on both threshold and background discrimination, and it should carry the same prominence in the caveats as the yield uncertainty.","section":"Sec. 2.2 and Sec. 3.2, Fig. 5"},{"comment":"The projected sensitivity assumes 0.95 kg of H2 in the fiducial volume (1.1% mole fraction), but the text states that 'no measurements of H2 solubility in LXe currently exist.' Since the sensitivity scales approximately linearly with the amount of H2 (Sec. 3.2), an overestimate of the achievable loading by a factor of two degrades the reach correspondingly. This should be listed explicitly as a major assumption of the sensitivity calculation, not only as a caveat in the solubility subsection.","section":"Sec. 2.4 and Sec. 3.2"}],"minor_comments":[{"comment":"The text in Sec. 2.2 refers to the 'yellow curve' in Fig. 1 for xenon recoils, while the figure caption says the blue curve; the color labeling should be made consistent.","section":"Fig. 1 caption vs. Sec. 2.2"},{"comment":"The phrase 'well below 100 MeV' is used repeatedly but never quantified; stating an approximate lower mass reach for the S2-only curve would help readers understand how close the projection is to the neutrino floor or to the assumed threshold.","section":"Sec. 3.2"},{"comment":"The notation ENR is typeset without a subscript in Eq. (1); using E_NR would be clearer, and the sentence 'ENR∼5 keV' could specify the electric field condition under which this threshold applies.","section":"Eq. (1)"},{"comment":"The right panel would benefit from a definition of the plotted quantity (e.g., spin-dependent proton cross section) in the caption, and from a note that the SD-neutron curve is shown for comparison at a different coupling.","section":"Fig. 5"}],"recommendation":"major_revision","confidential_remarks":"This is a well-written concept paper from a strong experimental group, and I do not see internal inconsistency or circularity: the sensitivity projections are derived from external models and measurements, not from fitting the target result. My concern is calibration of the presentation. The abstract and the discussion of Fig. 5 lead with a quantitative mass reach that depends on an unmeasured hydrogen recoil yield and on neglected accidental backgrounds; the authors' own caveats are present in the text but are not reflected in the summary statements. A major revision that adds uncertainty bands or explicitly demotes the quantitative reach to 'conditional on unverified microphysics' would make the paper's contribution clear without overclaiming. I do not recommend rejection; the concept and the technical feasibility discussion are genuinely useful."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"What you should know before reading: HydroX is a concept paper, not a measurement. The core idea—doping a tonne-scale liquid xenon TPC with hydrogen to make proton recoils the dark-matter target—is genuinely new, and the authors are commendably clear that the headline sensitivity depends on an unmeasured microphysics quantity: the quantity of light and charge produced by proton recoils in liquid xenon. Treat the sensitivity curves as an R&D roadmap, not as a projected limit.\n\nThe paper does a lot right. The kinematic argument is solid: a recoiling proton loses most of its energy to electronic stopping, so per unit energy it should produce roughly five times more detectable quanta than a xenon recoil. The SRIM calculation is cross-checked against a NEST prediction for helium recoils, which gives some confidence. The technical sections are also careful—Henry's law solubility estimates, distillation of H2 out of the circulation system, PMT permeability at cryogenic temperatures, and Garfield simulations of S2 gain. The background discussion is honest, including the tritium contamination challenge and the need for underground hydrogen. The paper explicitly labels the S2-only projections as speculative because accidental backgrounds are not modeled.\n\nThe soft spot is the one they flag: there are no data for H recoil yields in LXe. The SRIM-based estimate assumes all electronic stopping produces signal and none of the nuclear stopping does. For a keV-scale proton in xenon, the reduced energy sits near the nuclear-stopping maximum, so that assumption is not obviously safe. If the real yield is half what they assume, the effective threshold roughly doubles, and the sub-100 MeV reach could disappear entirely. The authors' own sensitivity scaling shows how sharp this is: a factor-of-two loss in S1 degrades the S1/S2 sensitivity by a factor of 3.5–4. The S1/S2 partition is also unknown; taking electron-recoil-like behavior as the baseline is reasonable but untested.\n\nBottom line: this is a serious, well-written concept paper that deserves a real referee. The physics case is novel, the caveats are stated, and the remaining unknowns are concrete R&D questions rather than hidden assumptions. I would send it to peer review, and I would bring it to group—the discussion of whether proton recoils really produce five times more quanta than xenon recoils is worth having. I'd also cite it if I were working on sub-GeV direct detection.","headline":"A genuinely new and honest concept paper whose headline sensitivity rests on an unmeasured proton-recoil yield in LXe; worth reading and refereeing, but the projections are an R&D roadmap, not a demonstrated reach.","tokens_in":18476,"tokens_out":3627,"would_cite":true,"duration_ms":32868,"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":"The paper projects that hydrogen doping expands a liquid xenon dark matter detector's reach to masses below 100 MeV, a window current xenon-only detectors cannot probe.","keywords":["HydroX","light dark matter","liquid xenon TPC","hydrogen doping","spin-dependent dark matter","low-mass WIMP search","proton recoil yield","sub-GeV direct detection"],"falsifier":"A calibration measurement of scintillation and ionization yields for single proton recoils in liquid xenon between about 1 and 10 keV would settle the matter: if the true yield is half the assumed five-times enhancement, the effective threshold rises and the claimed sub-100 MeV reach shifts to higher masses.","tokens_in":17264,"feed_emoji":"⚛️","tokens_out":5889,"duration_ms":56038,"temperature":0.7,"pith_summary":"HydroX proposes adding about one percent hydrogen to a liquid xenon time projection chamber, making protons the dark matter targets while the xenon acts as the sensor for light and charge signals. The paper argues that a recoiling proton produces roughly five times more detectable light and charge per unit energy than a recoiling xenon atom, because its energy goes mostly into electronic excitation rather than silent atomic collisions. That enhancement lowers the effective energy threshold enough to project sensitivity to dark matter masses below 100 MeV in an ionization-only (S2-only) analysis, and down to about 200 MeV with the standard combined light-and-ionization (S1/S2) analysis. It also gives an unmatched low-mass spin-dependent dark matter-proton channel. If the projection holds, this would open a mass window below 1 GeV that current tonne-scale xenon detectors cannot reach.","feed_headline":"Hydrogen-loaded xenon could see dark matter below 100 MeV","feed_subtitle":"One percent hydrogen makes proton recoils shine about five times brighter per keV, opening the sub-GeV window.","key_machinery":"The central mechanism is the conversion of proton kinetic energy into detectable xenon excitation and ionization. A recoiling proton transfers only about 1.5% of its energy per hard-sphere elastic collision to xenon atoms, so it survives many collisions below the xenon ionization threshold and keeps its energy in electronic stopping, which produces S1 and S2 signals. This is quantified as a fivefold gain in quanta per unit energy relative to a xenon recoil, computed with SRIM and encoded in NEST; that gain sets the low-energy thresholds that carry the whole sensitivity projection.","core_discovery":"The paper's central claim is that a liquid xenon TPC loaded with hydrogen at the one-percent level becomes a low-mass dark matter detector with qualitatively new reach. A dark matter particle scattering off a proton in the liquid produces a recoiling proton that, unlike a recoiling xenon atom, loses most of its energy to electronic excitation and ionization rather than to silent elastic collisions with surrounding xenon atoms. Using SRIM-based stopping calculations embedded in NEST, the authors estimate about five times more detectable quanta per unit energy for hydrogen recoils than for xenon recoils. Combined with the LZ detector model, a standard S1/S2 analysis is projected to be competitive down to 200 MeV dark matter mass, and an S2-only analysis extends access well below 100 MeV, with the spin-dependent dark matter-proton sensitivity unmatched at low masses. The authors explicitly state that the projections are indicative, pending measurements of hydrogen recoil yields, hydrogen solubility, and backgrounds.","pith_inferences":["A direct measurement of proton recoil yields in liquid xenon would not only test HydroX but also inform other proton-in-xenon signatures, such as neutron-induced backgrounds in existing xenon detectors.","The claimed hydrogen doping benefits for electron drift and diffusion could be tested in small TPCs independently of any dark matter search; if confirmed, they would improve multiple-scatter rejection for other rare-event physics.","The same kinematic argument suggests that hydrogen-doped argon or neon TPCs could also probe sub-GeV dark matter, although their scintillation and ionization properties differ from xenon."],"forward_implications":["With 1.1% hydrogen loading in an LZ-scale detector, a 500-day run would set spin-independent limits extending to dark matter masses below 100 MeV in the S2-only analysis.","The same run would provide the most sensitive spin-dependent dark matter-proton limits at low masses, a channel no current experiment covers.","Substituting deuterium for hydrogen gives comparable spin-dependent dark matter-neutron sensitivity, shifted upward in mass by the square root of two, with a factor-of-four better spin-independent sensitivity.","Because the low-mass signal spectrum rises steeply at threshold, actual sensitivity depends on exact detector threshold behavior and accidental backgrounds; the paper presents these numbers as indicative."],"supporting_citations":[{"why":"Supplies the SRIM calculation of hydrogen stopping powers that produces the fivefold signal enhancement.","marker":"[35]"},{"why":"Supplies the NEST liquid xenon excitation and ionization model into which the hydrogen recoil yields are inserted.","marker":"[28]"},{"why":"Provides the NEST v2.0 alpha-recoil power-law extrapolation used to model low-energy hydrogen recoils.","marker":"[30]"},{"why":"Provides the measured 50% S1 and S2 signal losses at 1.1% hydrogen mole fraction that set the signal model.","marker":"[50]"},{"why":"Supplies the LZ detector model, background model, and profile likelihood used for the sensitivity projections.","marker":"[88]"},{"why":"Supports the claim that proton-xenon collisions are soft, so most proton energy remains in electronic stopping.","marker":"[33]"},{"why":"Provides the measured helium recoil partitioning in liquid xenon used to bound the S2/S1 behavior of hydrogen recoils.","marker":"[40]"},{"why":"Defines the LZ detector geometry and operating conditions used as the case study for the projections.","marker":"[24]"},{"why":"Supplies the accidental background estimate that could reduce the S1/S2 sensitivity by about a factor of five.","marker":"[6]"}],"fun_headline_variants":["Hydrogen doping turns xenon TPC into sub-100 MeV dark matter probe","One percent hydrogen in liquid xenon amplifies proton recoils fivefold","HydroX: hydrogen-rich xenon targets light dark matter with enhanced signals","S2-only readout pushes hydrogen-xenon dark matter sensitivity below 100 MeV","Proton recoils shine brighter in hydrogen-loaded xenon, opening sub-GeV window"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The projected sensitivity collapses if the simulation-based assumption that hydrogen recoils in liquid xenon produce roughly five times more detectable light and charge than xenon recoils is wrong, because there are no measured data for hydrogen recoil yields in liquid xenon.","fun_headline_variants_meta":{"raw":{"variants":["Hydrogen doping turns xenon TPC into sub-100 MeV dark matter probe","One percent hydrogen in liquid xenon amplifies proton recoils fivefold","HydroX: hydrogen-rich xenon targets light dark matter with enhanced signals","S2-only readout pushes hydrogen-xenon dark matter sensitivity below 100 MeV","Proton recoils shine brighter in hydrogen-loaded xenon, opening sub-GeV window"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00018,"raw_usage":{"total_tokens":1279,"prompt_tokens":893,"completion_tokens":386,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":509,"completion_tokens_details":{"reasoning_tokens":282}},"tokens_in":509,"tokens_out":386,"duration_ms":4451,"temperature":1.0,"reasoning_tokens":282,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T20:13:45.681981+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A calibration measurement of scintillation and ionization yields for single proton recoils in liquid xenon between about 1 and 10 keV would settle the matter: if the true yield is half the assumed five-times enhancement, the effective threshold rises and the claimed sub-100 MeV reach shifts to higher masses.","supporting_citations":[{"cited_title":"An embedding of the Morse boundary in the Martin boundary","cited_arxiv_id":"2004.14624","evidence_quote":"Provides the measured 50% S1 and S2 signal losses at 1.1% hydrogen mole fraction that set the signal model."},{"cited_title":"First measurement of discrimination between helium and electron recoils in liquid xenon for low-mass dark matter searches","cited_arxiv_id":"2308.02430","evidence_quote":"Provides the measured helium recoil partitioning in liquid xenon used to bound the S2/S1 behavior of hydrogen recoils."}],"review_version":1}