Pith. sign in

REVIEW 4 major objections 6 minor 28 references

The paper claims that a MMThGEM-Micromegas detector operated in low-pressure SF6 achieves the first negative-ion gas gain above 100,000, and uses it to reconstruct track direction and observe nuclear-recoil-like events in a cubic-metre-scal

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-02 23:44 UTC pith:F3OUF6GP

load-bearing objection A genuine milestone for negative-ion gas gain, but the 252Cf 'nuclear recoil identification' is not yet data-anchored; the paper overclaims in Section 5. the 4 major comments →

arxiv 2602.12658 v2 pith:F3OUF6GP submitted 2026-02-13 physics.ins-det

High Negative Ion Gain MMThGEM-Micromegas Detector for Directional Dark Matter Searches

A. G. McLean , S.Higashino , R. R. Marcelo Gregorio , K. Miuchi , N. J. C. Spooner This is my paper
classification physics.ins-det PACS 29.40.Cs29.40.Gx95.35.+d
keywords negative-ion TPCSF6MMThGEMMicromegasgas gaindirectional dark matternuclear recoiltrack reconstruction
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

Dark-matter searches that record the direction of nuclear recoils need large low-pressure gas volumes and a readout that can both amplify faint ionisation signals and image tracks. Negative-ion drift gases such as SF6 are attractive targets but have long been limited to gas gains far below those of electron-drift gases. This paper claims to break that limit: a coupled MMThGEM-Micromegas detector in 40 Torr of SF6 produces an effective gas gain of 1.22±0.08×10^5, the largest reported for a negative-ion gas, with an energy resolution around 1.4. The same detector reconstructs the direction of alpha-particle tracks and, after installation in a cubic-metre-scale SF6 vessel, records events whose energies and ranges fall in the simulated fluorine nuclear-recoil band. If these claims hold, negative-ion drift TPCs become a realistic path to the scale-up needed for unambiguous directional dark-matter detection below the neutrino fog.

Core claim

The central claim is that the two-stage amplification stack formed by a MMThGEM and a Micromegas can be operated stably in low-pressure SF6 and delivers an effective gas gain of 1.22±0.08×10^5, roughly two orders of magnitude above typical negative-ion gas gains and comparable with electron-drift gases such as CF4. This gain is measured on 32 individually instrumented strips via the 5.89 keV X-ray peak of 55Fe, assuming a W-value of 34 eV for SF6. With the same detector, 5.5 MeV alpha particles are reconstructed as 2D tracks whose axis angle and dE/dx sense (head-tail) match simulation, demonstrating directionality. Finally, in a cubic-metre-scale vessel filled with 40 Torr SF6 and exposed t

What carries the argument

The load-bearing device is the coupled MMThGEM-Micromegas: a thick GEM with two electrode planes and four intermediate mesh layers provides two stages of avalanche gain (roughly 10^4), and a Micromegas placed 1 mm below provides a third parallel-plate amplification stage while depositing charge on orthogonal x/y micro-strips with 250 µm pitch. This stack converts the drift charge from negative-ion SF6 into large, localised pulses on 32 readout strips, making both the high gain and the 2D imaging possible. The analysis additionally relies on a discrimination parameter eta = E/R^2 (energy over squared 2D range), computed from simulated nuclear- and electron-recoil bands, to separate recoil typ

Load-bearing premise

Everything rests on the assumption that the simulation chain used to predict electron- and nuclear-recoil band positions is accurate enough that the measured event distribution can be labelled as nuclear recoils without a direct gamma-ray calibration of the recoil-energy scale.

What would settle it

Expose the same detector in the cubic-metre vessel to a strong gamma-ray source and compare the measured electron-recoil band in the energy-versus-range plane with the simulated band; if the measured band is shifted by more than the quoted uncertainties, the simulated ER/NR discrimination cuts are not trustworthy and the nuclear-recoil count would need to be re-derived.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Negative-ion drift gases are no longer inherently limited to low gain, so detector designs for directional dark-matter searches can use SF6 without sacrificing low-energy recoil sensitivity.
  • The detector can resolve both the axis and the sense (head-tail) of alpha tracks, showing that directionality information survives the high-gain readout chain.
  • A cubic-metre-scale SF6 volume can be operated with this readout, and neutron-induced events populate the expected fluorine nuclear-recoil band, supporting the scalability of the approach.
  • The eta = E/R^2 selection cut, built from simulated recoil bands, rejects 99% of electron recoils at ln(eta) >= 9 and 98% at ln(eta) >= 8, providing a working though not yet data-calibrated electron-recoil/nuclear-recoil discriminant.
  • Identified hardware improvements — smaller MMThGEM hole pitch and removal of the resistive micromegas layer — should reduce track discontinuities and charge dissipation in future versions.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • A natural extension not tested in the paper: the same MMThGEM-Micromegas stack may provide comparable gain in other negative-ion gases or gas mixtures, which could broaden the choice of target for directional searches.
  • Because the electron-recoil band was simulated rather than measured with a gamma source, the quoted 99%/98% rejection rates should be treated as provisional; a direct gamma-source calibration would test whether the simulated bands are correctly placed.
  • The high gain appears to make minority negative-ion peaks (small leading charge clusters) more visible, which could become a new handle for event fiducialisation or head-tail identification if confirmed in a dedicated run.
  • If the gain and track reconstruction are reproduced with both x- and y-strip planes instrumented, full 3D readout becomes possible, enabling track-morphology cuts that the current 2D projection cannot support.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper presents a coupled MMThGEM-Micromegas negative-ion TPC operated in 40 Torr SF6. It reports an effective gas gain of 1.22 ± 0.08 × 10^5 from 55Fe X-rays, claims this is the largest negative-ion gas gain yet reported, and demonstrates 2D track reconstruction and directional sense using 241Am alpha particles. The detector is then installed in the cubic-metre-scale C/N-1.0 vessel, exposed to a 252Cf neutron source, and events are classified as nuclear-recoil-like using an E/R^2 parameter with cuts informed by SRIM/SREM simulations. The paper concludes that these measurements provide strong evidence that nuclear recoils were successfully observed in a large SF6 volume.

Significance. If the gain result is correct, it addresses a long-standing limitation of negative-ion drift TPCs and strengthens the case for scaling up SF6-based directional dark-matter detectors. The alpha-particle track reconstruction and the first operation of a cubic-metre-scale SF6 volume with this readout are useful milestones for the CYGNUS R&D program. The paper is direct and reports new hardware results rather than re-analysis. However, the most ambitious claim — observation of nuclear recoils — is currently supported only by simulation-internal discrimination bands with no data anchor, so the significance of that result is not yet established. The gain claim is plausible but needs more systematic detail to sustain the headline number.

major comments (4)
  1. [Section 5, Fig. 10] The ER/NR discrimination and the conclusion that the observed events are 'strong evidence' of nuclear recoils rely entirely on simulated NR and ER bands from SRIM/SREM with Lindhard quenching; the cuts ln(eta) ≥ 8 and ≥ 9 and the claimed 99%/98% ER rejection rates are simulation-internal. The paper itself acknowledges that diffusion, charge dissipation, and MMThGEM hole-pitch discretisation can distort E and R. A gamma-source calibration of the ER band is explicitly listed as future work. Please either provide a measured ER calibration from the same detector to anchor the ER band and validate the rejection rates, or perform a quantitative systematic study showing that realistic variations of the acknowledged detector effects do not move the simulated ER band across the cuts. Without this, the 'strong evidence' claim is not supported.
  2. [Section 3, Fig. 4] The record gain of 1.22 ± 0.08 × 10^5 is derived from a single 55Fe spectrum with a very broad resolution (FWHM/mean = 1.41 ± 0.07) and a position-containment cut requiring the central channel to be between 13 and 18. The 6.6% total uncertainty is quoted without a breakdown, and the capacitive-injection calibration is not described. Please provide the individual systematic contributions (charge calibration, W-value, cut efficiency, gain non-uniformity, source geometry) and show that the extracted gain is stable under reasonable variations of the central-channel cut. This is needed to sustain the 'largest NI gas gain ever reported' claim.
  3. [Section 5, energy/range reconstruction] The recoil energy is computed using the gain measured in the small test vessel, where a central-channel cut was applied, but the C/N-1.0 analysis does not state whether an equivalent containment cut is used. Charge lost outside the instrumented area (32 y-strips covering 7.85 mm × 10 cm) reduces both E and R, and because eta = E/R^2, these losses shift events in a correlated way in Fig. 10. Please state whether the central-cut is applied to the 252Cf data, quote the number of events passing it, and quantify how events with charge outside the instrumented strips move in the (E, R) plane.
  4. [Section 5, Figs. 10-11] The text says 'a significant portion' of events fall within the strict NR cut and 'most observed events' are consistent with fluorine recoils, but no event counts are given after applying the strict and lenient cuts. Without the number of events before and after each cut, and without an estimate of the expected 252Cf-induced recoil rate and possible backgrounds, the strength of the evidence cannot be quantitatively assessed. Please report these counts and the associated statistical uncertainties.
minor comments (6)
  1. [Section 2] Typo: 'solderd' should be 'soldered'.
  2. [Section 3] Typo: 'electron equivalant' should be 'electron equivalent'.
  3. [Section 5] The notation 'ER_r = 99%' is not defined; presumably it means the ER-rejection efficiency. Please define it in the text or figure caption.
  4. [Section 4, Fig. 6] The angular distributions show clear peaks, but no event counts, error bars, or a quantitative measure of angular resolution are given. Please state the number of events in each exposure and, if possible, the fitted peak widths.
  5. [Section 5] The activity of the 252Cf source is not given, although the 55Fe and 241Am source strengths are quoted. The source activity is useful for normalising the observed event rate.
  6. [Abstract/Section 2] The text says 'complete 2-dimensional directionality' but only the y-strip plane is instrumented (the x-strips are mentioned but not read out). The reconstruction is therefore in the y-z plane; please rephrase to avoid overstating the readout dimensionality.

Circularity Check

0 steps flagged

No significant circularity; NR classification is a model-dependent interpretation, not a reduction to the inputs.

full rationale

The paper's central claims rest on direct measurements. The record negative-ion gas gain is obtained from a Gaussian fit to a measured 55Fe spectrum using the known 5.89 keV X-ray energy and the literature W-value of SF6, not from any fitted quantity that is later renamed as a prediction. The directionality claim is based on measured alpha-particle track angles and Bragg-curve asymmetries, again direct detector data. In the 252Cf study, the measured charge and 2D range are converted to E and R using the independently measured gain and W-value, then compared to SRIM/SREM simulated NR and ER bands. The ER/NR discrimination cut (ln eta >= 8 or 9) is defined from the simulated distributions, but the 1210 measured events themselves are not used to set the cut; many observed events fall outside the NR selection, so the conclusion that some events are consistent with NRs is not forced by construction. The paper explicitly acknowledges the limitation that a gamma-ray source run is needed for 'a more explicit ER/NR discrimination study' and describes the NR observation as 'tentatively explored' in the introduction. That is an external validation gap, not a circular reduction. The one self-citation [15] is used only as ancillary context for an expected MMThGEM amplification factor, while the actual gain is measured directly in this work. No load-bearing step reduces to its own inputs by definition, so the circularity score is 0.

Axiom & Free-Parameter Ledger

5 free parameters · 4 axioms · 0 invented entities

The free parameters are analysis thresholds and cuts chosen from data or simulation to achieve the stated discrimination; they do not correspond to new physics entities. The axioms are standard literature values or simulation models applied without independent in-paper validation, which is acknowledged for the ER/NR discrimination.

free parameters (5)
  • ln(eta) strict cut = >= 9
    Chosen from simulated NR/ER distributions to achieve 99% ER rejection; applied to data to identify NRs.
  • ln(eta) lenient cut = >= 8
    Chosen from simulated NR/ER distributions to achieve 98% ER rejection; applied to data to identify NRs.
  • Lower energy threshold for strict cut = 10 keV_ee
    Strict cut includes events above 10 keV_ee; the choice of this threshold is not independently justified.
  • 40 mV charge threshold = 40 mV
    Points above 40 mV are used for track reconstruction and range estimation; no noise baseline is given.
  • Central channel cut 13-18 = 13 <= channel <= 18
    Applied to X-ray events to ensure charge is not lost outside the instrumented area; may bias gain estimate if gain varies with position.
axioms (4)
  • domain assumption W-value of SF6 is 34 eV (ref [21])
    Used in Section 3 to convert measured charge into electron-equivalent energy; value taken from literature, not remeasured here.
  • domain assumption SRIM/SREM simulations accurately model stopping and range for ions and electrons in 40 Torr SF6
    Used in Sections 4 and 5 for alpha dE/dx comparison and NR/ER band simulation; no benchmark against measured ranges is provided.
  • domain assumption Lindhard model correctly converts nuclear-recoil energy to electron-equivalent energy
    Used in Section 5 to compute keV_ee for simulated fluorine nuclear recoils; model taken from refs [24,25].
  • domain assumption The capacitive charge-injection calibration is linear and accurate at the operating gain
    Invoked in Section 3 without showing the calibration curve or its uncertainty; underpins the absolute gain and energy scales.

pith-pipeline@v1.3.0-alltime-deepseek · 7870 in / 13265 out tokens · 122105 ms · 2026-08-02T23:44:17.495823+00:00 · methodology

0 comments
read the original abstract

Low pressure gaseous Negative Ion Time Projection Chambers (NITPCs) have been used previously by the DRIFT experiment to search for a directional Dark Matter (DM) signature. The main challenge with using a Negative Ion Drift (NID) gas target is the significantly lower gas gains to which they are typically limited. Recently, a MMThGEM device has been successfully demonstrated as an excellent gain stage device in the NID gas SF$_6$; capable of producing gas gains comparable with the electron drift gas CF$_4$. The next major challenge is to extend this high gain capability to multi-dimensional readout for the purpose of particle track reconstruction. The MMThGEM is therefore ideal for coupling to a strip readout detector like a Micromegas to achieve a high gain multi-dimensional Negative Ion (NI) readout plane, which is potentially suitable for the scale up required by future searches proposed by the CYGNUS consortium. In this paper, the first high gain demonstration of such a MMThGEM-Micromegas detector in low pressure SF$_6$ is described. This includes detector characterisation in a small test vessel resulting in the largest NI gas gain ever reported, 1.22 $\pm$ 0.08 $\times$ 10$^5$ , and directionality with alpha particles. Finally, this gain characterisation and tracking capability is leveraged to measure the energy and range of events, and identify those consistent with Nuclear Recoils (NRs), in a large cubic metre scale volume of SF$_6$ for the first time.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

28 extracted references · 2 linked inside Pith

  1. [1]

    Bo et al., Phys

    Z. Bo et al., Phys. Rev. Lett. 133 (2024) 191001

    2024

  2. [2]

    Aprile et al., Phys

    E. Aprile et al., Phys. Rev. Lett. 133 (2024) 191002

    2024

  3. [3]

    D. S. Akerib et al., arXiv:2512.08065 (2025)

    Pith/arXiv arXiv 2025

  4. [4]

    C. A. J. O’Hare, Phys. Rev. Lett. 127 (2021) 251802

    2021

  5. [5]

    S. E. Vahsen et al., Annu. Rev. Nucl. Part. Sci. 71 (2012) 189

    2012

  6. [6]

    Morgan, Nucl

    B. Morgan, Nucl. Instrum. Methods. Phys. Res. A. 513 (2003) 226

    2003

  7. [7]

    Klinger and V

    J. Klinger and V. A. Kudryavtsev., Phys. Rev. Lett. 114 (2015) 151301

    2015

  8. [8]

    V. A. Kudryavtsev et al., J. Phys.: Conf. Ser. 203 (2010) 012039

    2010

  9. [9]

    J. H. Davis Phys. Rev. Lett. 113 (2014) 081302

    2014

  10. [10]

    DarkSide-20k Collaboration., Eur. Phys. J. C. 84 (2024) 24

    2024

  11. [11]

    J. B. R. Battat et al., Astropart. Phys. 91 (2017) 65

    2017

  12. [12]

    Shimada et al., Prog

    T. Shimada et al., Prog. Theor. Exp. Phys. 10 (2023) 103F01

    2023

  13. [13]

    S. E. Vahsen et al., arXiv:2008.12587 (2020)

    Pith/arXiv arXiv 2008

  14. [14]

    N. S. Phan et al., JINST 12 (2017) P02012

    2017

  15. [15]

    McLean et al

    A.G. McLean et al. JINST 19 (2024) P03001

    2024

  16. [16]

    F. D. Amaro et al., JINST 19 (2024) P06021

    2024

  17. [17]

    Bressler et al., Prog

    S. Bressler et al., Prog. Part. Nucl. Phys. 130 (2023) 104029

    2023

  18. [18]

    de Olivera and M

    R. de Olivera and M. Cortesi, JINST 13 (2018) P06019

    2018

  19. [19]

    Attie et al., Appl

    D. Attie et al., Appl. Sci. 11 (2021) 5362

    2021

  20. [20]

    Kosaka et al., J

    T. Kosaka et al., J. Phys. 2374 (2022) 012077 – 12 –

    2022

  21. [21]

    Lopes et al., J

    I. Lopes et al., J. Phys. D. 19 (1986) 107

    1986

  22. [22]

    Glaister, Math

    P. Glaister, Math. Gaz. 85 (2001) 104

    2001

  23. [23]

    The Stopping and Range of Electrons in Matter

    P. Hovongton and D. Drouin, "The Stopping and Range of Electrons in Matter" (2024) http://www.srim.org/SREM.htm

    2024

  24. [24]

    Lindhard et al., Mat

    J. Lindhard et al., Mat. Fys. Medd. Dan . Vid. Selsk . 33 (1963) 10

    1963

  25. [25]

    Sorensen, Phys

    P. Sorensen, Phys. Rev. D. 91 (2015) 083509

    2015

  26. [26]

    N. S. Phan et al., Astropart. Phys. 84 (2016) 82

    2016

  27. [27]

    R. J. Lafler, Thesis, University of New Mexico. (2019)

    2019

  28. [28]

    Eldridge, Thesis, University of Sheffield

    C. Eldridge, Thesis, University of Sheffield. (2021) – 13 –

    2021