Does the High Energy LUX-ZEPLIN Event Suggest Hyperfine Atomic Dark Matter?

Bhaskar Das; Prolay Chanda; Sagnik Mukherjee

REVIEW

Does the High Energy LUX-ZEPLIN Event Suggest Hyperfine Atomic Dark Matter?

Prolay Chanda, Bhaskar Das, Sagnik Mukherjee

High Energy Physics - Phenomenology · Submitted 29 Sep 2026

Pith's verdict: An honest, careful model-interpretation paper that gives the LZ 248 keV candidate a hyperfine-atomic-DM reading with falsifiable handles, but the whole claim sits on an untested sliver of the halo tail near the SHM cutoff.

T0 · AI review

How this review was made: An AI referee (deepseek/deepseek-v4-flash) read the paper on September 30, 2026 and answered a fixed list of questions about it. T0 is the starting level every paper gets. The levels describe how deeply Pith has checked the paper, never the authors.

T0 AI review An AI read the paper and answered a fixed list of questions.
T1 Audit Automatic checks that the references and files the paper relies on exist and match.
T2 Compute The paper's code or calculations re-run with the same results.
T3 Formal statement The claims written out in a formal math language a computer can read.
T4 Proof Those claims proved, and the proof checked by computer.

How Pith checks papers →

· 4 major objections · 5 minor · Ask about this paper · The compiled record · Read the full review →

Challenge this review Comment (0) Read on arXiv ↗

Cite ▾

arXiv ID2609.36814 (version 2)

Pith ID2C5TOOGA, Pith's permanent record of this paper and its review

CitationBhaskar Das, Prolay Chanda, Sagnik Mukherjee. "Does the High Energy LUX-ZEPLIN Event Suggest Hyperfine Atomic Dark Matter?." arXiv:2609.36814 (2026). https://pith.science/paper/2C5TOOGA

Cite this reviewBibTeX for Pith's review of this paper

Claim paper

Why Pith says this

Highlight any text to comment on it.

Editor's note

read the letter

Quick verdict: competent, honest model-interpretation paper, and the honesty matters more than the model. It applies the hyperfine-excitation mechanism from the atomic-dark-matter literature (Kaplan-Krnjaic-Rehermann-Wells, Tucker-Smith-Weiner) to the LZ 248 keV candidate, and it does not oversell: the ~21% spectral-placement claim is stated as placement, not significance, and the authors flag halo sensitivity in several places (b118, b173, b251).

What is useful: the endothermic kinematics put the signal at a sharp kinematic edge, producing two testable handles as a package — the Ar/Ge/Xe/W target hierarchy (closed / near-threshold / open) and a predicted annual modulation near 0.84 in the SHM (eq. 43). The spectral reconstruction is careful: isotope sums, a digitized efficiency curve, a 23 keV Gaussian response, Helm form factor. The MW+LMC extension (delta_hf ~ 510 keV) is a reasonable illustration of tail sensitivity, though it stays illustrative.

The soft spot — acknowledged by the authors and then lived with — is that everything rests on the extreme high-speed tail. vmin sits at roughly 750–800 km/s against the SHM cutoff vmax = 809 km/s, and fD,loc = 1 is assumed (App. B2). Their own stress tests make the fragility quantitative: Tsallis and Mao-E9 demand roughly 4x the coupling, pushing g5 to within ~15% of the 10% portal bound in eq. (24). A tail suppressed by another factor of ~2, or a local neutral fraction appreciably below one, closes the window. The cosmological estimate (xbar_D ~ 0.09–0.17) does not fix fD,loc, and b267 concedes the local neutral abundance is unestablished. So the interpretation is viable in the SHM, marginal in the paper's own stress-test halos, and dead if the real tail is poorer than the SHM. Credit for laying this out, but it means the claim is a fitted, contingent one: delta_hf = 340 keV is placed by hand near the xenon threshold, and g5 normalizes to one event.

Minor: the MW+LMC distribution underlying Fig. 6 is not given explicitly, so that figure is not reproducible from the paper alone.

Who it's for: anyone sorting through the pile of LZ-excess interpretations. Not the sharpest model in the pile, but among the more honest ones. Send it to referees; the referee should push on the halo-tail and fitted-input questions, which is exactly what the authors invite.

Referee report

Significance

The paper's strength is that it is an internally consistent, honestly scoped mechanism study rather than a discovery claim. I spot-checked the load-bearing arithmetic: eq. (3) with the benchmark parameters reproduces δhf ≈ 340 keV; eq. (15) gives vthr ≈ 750 km/s against vmax = 809.1 km/s; and eq. (40) reproduces δmax values of 131, 232, 396, and 533 keV for Ar, Ge, Xe, and W respectively. The recast is unusually careful for this genre: isotope sum, digitized efficiency, Gaussian smearing, and a 72-phase annual average are all documented (App. B). If the mechanism is correct, the target hierarchy and the near-maximal annual modulation are concrete falsifiable predictions usable by future W, Ge, and Ar detectors, and the atomic origin of the inelastic gap (rather than a free parameter) is a genuine conceptual virtue over plain endothermic dark matter.

Major objections (4)

[Sec. III and App. B2, with Eq. (52)] The local neutral fraction is set to fD,loc = 1 (App. B2) without a supporting argument, and the central rate scales linearly in it: Nsig ∝ fD,loc (g5 εγ)². The paper's own cosmology gives a residual ionized fraction x̄D ≈ 0.09–0.17 (eq. (52)), sentence b190 states that cosmological fD = 1 does not require a fully neutral local population, and b267 concedes that the local neutral abundance is unestablished. A reduction of fD,loc by a factor of order 1.4–2 raises the coupling needed for one event by 1.2–1.4, which, combined with the halo-dependent normalization of eq. (44), pushes the required g5 to or beyond the eq. (24) bound (g5 ≲ 9×10⁻³) — see the next comment. The paper should either derive a bracket on fD,loc from a self-consistent galactic treatment of the neutral and ionized components of the atomic sector, or state explicitly that the benchmark assumes effectively full local neutrality and quantify how the inferred g5 degrades as fD,loc drops. The authors acknowledge this: “A complete assessment must also establish the local neutral abundance and consistently in- clude the axial portal’s contribution to the atomic spec- trum, including longitudinal exchange.”
[Sec. IV.A, Eqs. (42)–(44), Fig. 5] The SHM-normalized coupling g5 = 1.9×10⁻³ (eq. (34)) is a lower bound: under the paper's own stress tests the required coupling product rises by factors of 4.0 (Tsallis) and 4.1 (Mao-E9), giving g5 ≈ 7.6–7.8×10⁻³, within roughly 13–16% of the eq. (24) consistency bound — before any additional suppression. Because the benchmark sits at vthr ≈ 750 km/s ≈ 0.93 vmax (eqs. (15)–(16) and B1–B2), the interpretation is viable only if the true local high-speed tail is at least as rich as the SHM, precisely the least-constrained part of the halo input. The authors flag this sensitivity (b118, b173, b251), but the Conclusions (Sec. VI) should present the margin-to-bound arithmetic explicitly: a tail suppression or neutral-fraction deficit of a factor ≈ 1.4, well within the stated halo uncertainty, closes the one-event window. This is the load-bearing fragility of the central claim and should be featured, not footnoted. The authors acknowledge this: “The SHM value ofg 5 should therefore not be interpreted as halo in- dependent.”
[Sec. IV.B, Fig. 6] The headline 'extended reach' example, MW+LMC at δhf = 510 keV, is normalized with g5 = 9.00×10⁻³ — coincident with the eq. (24) 10% portal-consistency bound — and at εγ = 8×10⁻⁴, the interpolated BaBar floor (eqs. (36) and (38)). The LMC extension is therefore demonstrated only at the very edge of the allowed portal parameter space: both constraints are saturated simultaneously. The paper should quantify the remaining margin of this example and state its sensitivity to the same halo-tail and neutral-fraction inputs that limit the SHM benchmark, rather than presenting Fig. 6 as an unqualified broadening of reach.
[Sec. III, Eqs. (29)–(34)] The one-event normalization Nsig = 1 is imposed with no reference to the background expectation in the 225–271 keV candidate band. The paper correctly disclaims any statistical claim (b250), but for the reader to judge what 'one signal event from this model' means against the LZ data, the expected background count in that band should at least be quoted from the LZ sideband; otherwise the normalization is a pure convention and the 21% placement fraction has no evidentiary weight.

Minor objections (5)

[Sec. III, before Eq. (29)] The sentence 'Our recast uses the exposure E = 2.84 tonne yr and the efficiency ϵNR(ER), to the public NR efficiency curve in D. S. Akerib et al.' is grammatically broken; it should read '...and the efficiency ϵNR(ER) taken from the public NR efficiency curve of Ref. [1]'.
[Fig. 6 caption] The caption sentence 'the shifts MW+LMC the spectrum toward lower recoil energies' is garbled and should be rewritten.
[Eq. (19)] The notation f⁴_eff = m⁴_X / 2(g5 εY cW)² is ambiguous: the factor 2 could be read as inside or outside the squared term. Use explicit brackets, e.g. m⁴_X / [2 (g5 εY cW)²].
[Sec. I] 'In the minimalistic realization for elastic WIMP-nucleon scattering' should read 'In the minimal realization'.
[Sec. V.A, Eq. (51)] The analytic residual-ionization estimate of eq. (52) is stated to carry up to an order-of-magnitude uncertainty (b223). Since this estimate is the only quantitative check on the neutrality of the local population that the direct-detection rate assumes, a numerical recombination treatment using the Barron et al. machinery cited as Ref. [61] would substantially strengthen the consistency claim; at minimum, the order-of-magnitude caveat should be repeated where fD,loc = 1 is set in App. B2.

How Pith checks papers → · Full audit of this paper →

The paper

The paper claims

A TeV-scale dark atom whose hyperfine spin flips on xenon can reproduce the LUX-ZEPLIN 248 keV nuclear-recoil candidate, with roughly 21 percent of its predicted yield landing in the observed band.

The reading

This paper proposes that the 248 keV nuclear-recoil candidate reported by the LUX-ZEPLIN experiment is the hyperfine excitation of a hydrogen-like dark atom, making it a potential first direct-detection signal of atomic dark matter. For a dark atom of mass 1 TeV with a hyperfine splitting near 340 keV, xenon scattering becomes kinematically accessible only through the extreme high-velocity tail of the Galactic halo, which explains why a single high-energy recoil appears while the low-energy elastic channel is suppressed. The axial-vector portal makes the singlet elastic matrix element vanish, so the leading signal is purely endothermic and predicts a sharp target hierarchy: closed on argon and germanium, near-threshold on xenon, and open on tungsten. About 21 percent of the central benchmark spectrum populates the 225 to 271 keV candidate band, and the signal carries a large annual modulation of roughly 0.84 in the standard halo model. The paper also shows that the Milky Way plus Large Magellanic Cloud halo distribution extends the kinematic reach to larger splittings, up to about 510 keV.

What carries the argument

The load-bearing object is the hyperfine singlet-triplet splitting of a hydrogen-like dark atom bound by a massless dark photon, together with the axial-vector portal that allows the transition. The singlet elastic matrix element vanishes, so the leading low-velocity population cannot produce ordinary recoils, while the endothermic hyperfine channel requires incident speeds above $v_{\rm min}(E_R)$ given by eq. (13). The combination of a 1 TeV atom mass and a 340 keV splitting places the xenon channel near its kinematic ceiling, which is what makes the signal live almost entirely in the halo's extreme high-velocity tail.

Extended reading notes

Core claim

The central claim is that the LZ high-energy event can be produced by hyperfine excitation of a TeV-scale hydrogen-like dark atom, with the required axial coupling small enough to respect the portal-consistency constraint and the xenon nuclear-recoil spectrum placing roughly 21 percent of the accepted one-event yield in the 225 to 271 keV candidate band for the standard halo model. The endothermic kinematics impose a threshold that selects only the fastest dark atoms, above roughly 760 km/s up to the 809 km/s SHM cutoff, and this same threshold produces the predicted target hierarchy. The authors explicitly state that the 21 percent fraction describes spectral placement and carries no claim of statistical significance, a caveat they repeat in the conclusions.

Load-bearing premise

The entire signal depends on the extreme high-velocity tail of the local dark matter halo being populated at speeds above roughly 760 km/s up to the 809 km/s cutoff, and on essentially all local dark matter ($f_{D,\rm loc}=1$) being in neutral dark atoms; if the true tail is depleted or the neutral fraction is lower, the coupling required to produce one event rises and can exceed the portal-consistency bound.

Editorial extensions

If this is right

Argon and germanium detectors should see no events in the 225-271 keV band, while a tungsten target at comparable exposure should show events, a fingerprint unique to this benchmark.
The signal should exhibit a large annual modulation of roughly 0.84 in the SHM, with the rate peaking near maximum Earth speed around June 2.
The required portal coupling $g_5 \simeq 1.9\times10^{-3}$ lies below the $9\times10^{-3}$ consistency bound, leaving a viable parameter window between the 1 percent portal shift and the BaBar constraint.
In the MW+LMC halo, splittings up to about 510 keV become accessible, placing the spectral peak near the candidate energy and broadening the viable parameter space.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Because the interpretation hinges on $f_{D,\rm loc}=1$ and the halo's high-speed tail, a measurement separating the cosmological neutral fraction from the local neutral fraction would directly test the model's normalization.
The 0.84 annual modulation is a clean observational discriminator: a time-resolved analysis of LZ high-energy events comparing phase and amplitude would settle whether the excess is halo-selected endothermic scattering rather than background.
If the same physics applies, other inelastic or excited-state signatures on heavy targets such as tungsten could be searched at future facilities, and the model predicts a near-zero argon and germanium high-energy rate.
The mechanism, if confirmed, would tie direct-detection signals to atomic-structure parameters (binding energy, Bohr radius, hyperfine gap) and remove the inelastic gap as a free parameter, a distinctive feature among LZ interpretations.

Original abstract

The LUX-ZEPLIN (LZ) Collaboration has reported a nuclear-recoil candidate at $248\pm23~({\rm stat})\pm23~({\rm sys})~{\rm keV}$ in an extended high energy search window. We investigate whether this event can arise from the hyperfine excitation of hydrogen-like atomic dark matter. For a TeV-scale dark atom, a hyperfine splitting near $340~{\rm keV}$ places xenon scattering close to the observed value by naturally selecting the high-velocity tail of the Galactic halo while suppressing the leading low-energy elastic response. The resulting endothermic kinematics predict a pronounced target hierarchy: the benchmark transition is inaccessible on Ar and Ge, lies close to threshold on Xe, and remains open on heavier targets such as W, which is specific to our model parameters. Because xenon probes the extreme high-speed tail, the signal exhibits a large annual modulation and a strong dependence on the assumed halo distribution. We also examine how the infall of the Large Magellanic Cloud enhances the high-speed tail of the Milky Way's local dark matter distribution, extending the kinematic reach to larger hyperfine splittings.

The compiled record

The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-09-30: 1 request, 5,163 input tokens.

What the abstract itself states, as probabilities: claims first 0.04 · states limitation 0.76 · human participants 0.01 · controlled comparison 0.10 · randomized assignment 0.01 · states quantitative result 0.92 · compares with prior results 0.35 · mentions code or data availability 0.02.

The abstract, sentence by sentence

arXiv listing abstract v1. Results and the main claim are underlined; show every role.

The LUX--ZEPLIN (LZ) Collaboration has reported a nuclear-recoil candidate at $248\pm23~({\rm stat})\pm23~({\rm sys})~{\rm keV}$ in an extended high-energy search window. We investigate whether this event can arise from the hyperfine excitation of hydrogen-like atomic dark matter. For a TeV-scale dark atom, a hyperfine splitting near $340~{\rm keV}$ places xenon scattering close to the observed value by naturally selecting the high-velocity tail of the Galactic halo while suppressing the leading low-energy elastic response. The resulting endothermic kinematics predict a pronounced target hierarchy: the benchmark transition is inaccessible on Ar and Ge, lies close to threshold on Xe, and remains open on heavier targets such as W, which is specific to our model parameters. Because xenon probes the extreme high-speed tail, the signal exhibits a large annual modulation and a strong dependence on the assumed halo distribution. We also examine how the infall of the Large Magellanic Cloud enhances the high-speed tail of the Milky Way's local dark matter distribution, extending the kinematic reach to larger hyperfine splittings.

background contribution method result interpretation main claim

What the classifier reports for each sentence. The number is the probability it gave the label.
#	sentence	role	the authors' own work	states the main conclusion	strength
s0	The LUX--ZEPLIN (LZ) Collaboration has reported a nuclear-recoil candidate at $248\pm23~({\rm stat})\pm23~({\rm sys})~{\rm keV}$ in an extended high-energy search window.	background 0.96	0.07	0.03	asserted 1.00
s1	We investigate whether this event can arise from the hyperfine excitation of hydrogen-like atomic dark matter.	contribution 0.99	0.64	0.05	hedged possibility 0.61
s2	For a TeV-scale dark atom, a hyperfine splitting near $340~{\rm keV}$ places xenon scattering close to the observed value by naturally selecting the high-velocity tail of the Galactic halo while suppressing the leading low-energy elastic response.	result 0.49	0.71	0.40	asserted 0.87
s3	The resulting endothermic kinematics predict a pronounced target hierarchy: the benchmark transition is inaccessible on Ar and Ge, lies close to threshold on Xe, and remains open on heavier targets such as W, which is specific to our model parameters.	result 0.85	0.81	0.17	asserted 0.99
s4	Because xenon probes the extreme high-speed tail, the signal exhibits a large annual modulation and a strong dependence on the assumed halo distribution.	result 0.61	0.73	0.09	asserted 1.00
s5	We also examine how the infall of the Large Magellanic Cloud enhances the high-speed tail of the Milky Way's local dark matter distribution, extending the kinematic reach to larger hyperfine splittings.	contribution 0.80	0.50	0.05	asserted 0.95

What kind of paper the classifier reports this is: theoretical result 0.83 · evidence from analysis of existing data 0.58.

What was examined, and what was not

sources read

arXiv listing abstract v1

not examined

full text
figures and tables
references and citing sentences
relations to other papers

levels run

P abstract

not run

S sections
T body sentences
C citations
R cross-paper relations

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. Every number in the measurements table is the authors', checked by code. Record as JSON

Figures

**Figure 1.** FIG. 1. Here we demonstrate the kinematic accessibility of en

**Figure 2.** FIG. 2. Reconstructed natural-xenon event spectra for

**Figure 4.** FIG. 4. Maximum endothermic splitting accessible to rep

**Figure 5.** FIG. 5. Seasonal variation of the accepted recoil rate in xenon

**Figure 6.** FIG. 6. Reconstructed recoil spectra for the SHM and

**Figure 7.** FIG. 7. The solid curve uses the residual-ionization estimate

Showing 6 of the paper's 7 figures. See all figures in the PDF ↗ Figures reproduced from arXiv:2609.36814 by the authors.

References

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

The Electron-Recoil Response Tail and the LZ High-Energy Event Background
hep-ex 2026-10 accept novelty 7.0 of 10

The published LZ electron-recoil response characterization does not determine the tail probability needed for the background estimate; tiny quantile shifts (0.14%) can change the ER contribution by a factor of 13.
Simulation-Informed High-Velocity Dark Matter in the Effective Theory of Direct Detection
hep-ph 2026-10 conditional novelty 6.0 of 10

A Milky Way dark matter velocity distribution combining Gaia stellar kinematics and a FIRE-2 Large Magellanic Cloud analog limits local dark matter speeds to under 900 km/s, shifting the inferred inelastic mass splitt...
Exothermic Dark Matter with a Dark Higgs: The LZ 248 keV Event and the Galactic Center Excess
hep-ph 2026-10 conditional novelty 6.0 of 10

Exothermic dark matter in a dark U(1) model with a dark Higgs can simultaneously explain the LZ 248 keV event and the Galactic Center gamma-ray excess.
Signature of GW from the Hidden Sector in light of the Recent LZ Event
hep-ph 2026-10 conditional novelty 5.0 of 10

A U(1)_X hidden sector model simultaneously explains the LZ 248 keV recoil event, provides the correct dark matter relic density, and predicts gravitational waves detectable by future interferometers.
One Event, Two Frontiers: From LZ to LHC
hep-ph 2026-10 conditional novelty 5.0 of 10

Most elastic dark matter interactions proposed to explain the LZ 248 keV event are excluded by LHC monojet data for dark matter masses below roughly a TeV, assuming EFT validity.

Discussion

Disagree with the review, or know something it missed? Leave a comment, highlight any passage above to comment on it, or challenge this review formally.