Pith. sign in

REVIEW 2 major objections 5 minor 30 references

Calibration of liquid argon detector with $^{83m}Kr$ and $^{22}Na$ in different drift field

T0 review · 2 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

Pith's one-line read Krypton-83m, carried in by recirculating argon, calibrates the low-energy light response of a liquid argon TPC.

desk verdict Useful incremental LAr calibration data, but the volume-averaged 83mKr light yield is likely biased by position-dependent response and the paper quotes only statistical errors. read the letter →

arxiv 1909.02207 v1 pith:L7E6AS36 submitted 2019-09-05 physics.ins-det hep-ex

classification physics.ins-dethep-ex PACS 85.60.Ha14.60.Pq
keywords liquidargontimeprojectionchamber83mKrcalibrationlightyielddriftelectricfieldelectronrecoilnoble-liquiddetector
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper demonstrates that the radioactive isotope $^{83m}$Kr can be swept into a liquid argon (LAr) detector through the existing argon recirculation system and used as an internal, low-energy calibration source. The authors measure the detector's light yield at the 41.5 keV sum of the two $^{83m}$Kr transitions and compare it with the 511 keV line from $^{22}$Na. They report absolute light yields of 7.26$\pm$0.02 photoelectrons per keV for $^{83m}$Kr and 7.66$\pm$0.01 photoelectrons per keV for $^{22}$Na at zero drift field, and they track how both yields fall as the drift field rises to 200 V/cm. If the result stands, large LAr detectors can calibrate their central volume without deployment hardware, and field-dependent quenching must be folded into the energy response.

What carries the argument

The mechanism that carries the argument is the transport of $^{83m}$Kr atoms from a $^{83}$Rb-doped zeolite trap through the closed argon circulation loop, so the isotope is distributed throughout the active volume rather than collimated from outside. Inside the liquid, $^{83m}$Kr decays via two conversion-electron/x-ray transitions summing to 41.5 keV; because the 154 ns separation is much shorter than the argon triplet scintillation time, the two interactions merge into a single light pulse. Light yield is then inferred from Gaussian fits to the full-absorption peak, converted from photoelectrons to keV using a single-photoelectron calibration of the two immersed PMTs based on a PMT response function fitted to LED data (mean charges 0.53 pC and 0.44 pC, monitored to less than 2% variation). Field scans from 0 to 200 V/cm expose the recombination-driven decrease in scintillation yield, which the paper interprets with the PARIS recombination model.

What would settle it

One concrete check: recalibrate the same detector's light yield using an independent method, such as a collimated gamma source with a known emission rate and a Monte Carlo of the geometry, or a precision source of single photons with known quantum efficiency. If the resulting photoelectrons-per-keV differs from 7.26 at 41.5 keV by more than the quoted statistical uncertainty, the LED-based single-photoelectron scale is wrong. Alternatively, using a second PMT with a different photocathode or TPB configuration would reveal whether the field-quenching trend is a detector artifact or a genuine argon property.

Watch

Extended reading notes

Core claim

The paper's central finding is that $^{83m}$Kr introduced through the gas recirculation system behaves like a well-understood point-like source of 41.5 keV electronic recoils inside a liquid argon TPC: the two cascaded transitions at 32.1 keV and 9.4 keV fire within about 154 ns, well inside the 1.6 $\mu$s argon triplet time, so they appear as a single peak. Fitting that peak gives a light yield of 7.28$\pm$0.02 photoelectrons per keV at zero field, with 17.6% energy resolution, versus 7.66$\pm$0.01 photoelectrons per keV and 3.6% resolution for the 511 keV line from $^{22}$Na. Raising the drift field from 0 to 200 V/cm suppresses recombination and quenches the light yield by 8.4% for $^{83m}$Kr and 21.4% for $^{22}$Na; the weaker quenching at 41.5 keV indicates that recombination is more complete for the denser ionization column of the lower-energy electrons. The fitted half-life of 1.83$\pm$0.11 h after stopping the fill matches the known 1.83$\pm$0.02 h, confirming that the source decays cleanly away with no permanent contamination.

Load-bearing premise

The result assumes that the PMT single-photoelectron calibration, established with an LED, accurately gives the charge for the TPB-shifted 420 nm scintillation pulses recorded in the physics runs; if it does not, every reported light yield scales by a common factor that the paper does not estimate.

Editorial extensions

If this is right

  • Large liquid argon TPCs can use $^{83m}$Kr as a low-energy, volume-filling calibration source without any insertion mechanism, since it enters with the recirculating gas and leaves with the known 1.83 h half-life.
  • The measured absolute light yields give a direct energy-scale conversion for electronic recoils at 41.5 keV and 511 keV in a TPB-coated LAr detector with two immersed PMTs.
  • The field-induced quenching curves (8.4% at 41.5 keV, 21.4% at 511 keV, at 200 V/cm) provide data that recombination models such as PARIS must reproduce, and imply that the energy scale shifts with drift field.
  • The small 5% light-yield difference between 41.5 keV and 511 keV indicates a mild non-linearity in LAr scintillation response over this range, relevant for low-energy rare-event searches.

Reading between the lines

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

  • The reported light-yield ratio between 511 keV and 41.5 keV, with the higher-energy line brighter by about 5%, could be tested against models that treat recombination as a function of linear energy transfer; if the trend extends to tens of keV, calibration sources at multiple energies would be needed to map the nonlinearity.
  • Because the single-photoelectron calibration relies on LED pulses, a subtle systematic would arise if the LED optical pulse shape or PMT gain differs from that for TPB-shifted 128 nm scintillation events; comparing the 511 keV peak position with an independent, absolutely calibrated source would settle the scale.
  • If $^{83m}$Kr can be circulated through a full-scale detector without degrading argon purity, as this small detector suggests, it may also serve for continuous stability monitoring during long physics runs, not just calibration campaigns.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 5 minor

Summary. This paper reports calibration measurements of a small two-phase liquid argon TPC using 83mKr introduced via the gas recirculation system and 22Na as an external collimated source. The authors measure the scintillation light yield at 41.5 keV and 511 keV for drift fields of 0, 50, 100, 150, and 200 V/cm, report the field-induced quenching (8.4% for 83mKr and 21.4% for 22Na at 200 V/cm), and verify the 83mKr half-life (1.83±0.11 h) after halting source injection. The paper's central claims are that 83mKr provides a distributed low-energy calibration source for LAr detectors and that the absolute yields and field dependence are measured to ~0.01–0.02 pe/keV precision.

Significance. If the quantitative results are correct, the paper offers a simple and useful validation of 83mKr-based distributed calibration for liquid argon detectors, along with a concrete dataset on electric-field quenching of the scintillation yield. The half-life check is a nice internal consistency test, and the use of two independent sources with different energies is a reasonable approach. However, the paper as presented does not yet support the stated absolute precision because it quotes only statistical uncertainties and because the volume-distributed 83mKr sample may be subject to position-dependent light collection that is not characterized.

major comments (2)
  1. [Sec. 4.1, Figs. 6–7, Tables 1–2] The 83mKr full-absorption peak at 0 V/cm has µ=301.9 pe and σ=52.88 pe, giving a resolution of 17.5%, whereas photostatistics alone contributes 1/sqrt(301.9)≈5.8%. In contrast, the 22Na 511-keV peak has ~3.4% resolution versus a 1.6% photostatistics contribution. This large unexplained broadening for the volume-distributed 83mKr source, compared with the centrally-collimated 22Na source, is strong evidence of position-dependent light collection that is not addressed anywhere in the paper. Because the 83mKr events populate the entire active volume while 22Na samples a small central region, the quoted 5% difference between the 41.5-keV and 511-keV light yields is confounded by spatial geometry. The authors should either provide a position-uniformity map or position correction based on the PMT charge ratio, or restrict both calibrations to a matching fiducial volume; without this, the absolute 83mKr yields and the field-quenching comparison between the two sources in Section 5 are not reliable.
  2. [Sec. 3.2 and Sec. 4.1] The uncertainties reported for the light yields (e.g., 7.28±0.02 pe/keV) are only statistical fit errors. The absolute scale relies on the LED-based single-photoelectron calibration of Section 3.2, which yields mean SPE charges of 0.53 pC and 0.44 pC with a monitored variation below 2%. The paper does not estimate the systematic error in transferring this LED calibration to 128-nm scintillation light shifted by TPB to 420 nm, including possible differences in photoelectron collection, pulse shape, or gain. Therefore the claims of 0.01–0.02 pe/keV precision are not supported. A systematic error budget, or at least a conservative estimate based on the SPE calibration uncertainty and optical collection variations, is needed.
minor comments (5)
  1. [Abstract vs. Sec. 4.1] The abstract states 7.26±0.02 pe/keV for 83mKr, while Section 4.1 and Table 1 report 7.28±0.02; the value should be reconciled.
  2. [Table 2, 200 V/cm row] σp is listed as 1475.3 pe, which is far outside the range of the other rows and inconsistent with Fig. 10; it is presumably a typo for 147.5 pe.
  3. [Sec. 4.2 vs. Abstract] The Section 4.2 title says 'from 0 to 200V/cm' but the abstract says 'from 50 to 200V/cm'; the stated range should be consistent.
  4. [Sec. 5] The claim that the PARIS model 'provides a good description' of the data is not supported by any quantitative comparison in the paper; the authors should add a model curve or residuals, or qualify the statement.
  5. [Figures and references] There are minor typos: 'E_drfift' in the captions of Figs. 9–10, 'after stop filling 83mKr atom' in Section 4.1, and non-standard capitalization in several references (e.g., refs. [20] and [26]).

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: this is a direct experimental calibration with externally anchored energy and half-life standards.

full rationale

The paper reports measurements rather than a derivation, and no quantity is fitted to itself or renamed as a prediction. The light yields are obtained by fitting measured charge spectra in photoelectrons and dividing by the known decay energies (41.5 keV for 83mKr, 511 keV for 22Na), which come from external nuclear data, not from the detector response. The 83mKr half-life is fit from a decay curve and then compared with an external literature value of 1.83±0.02 h; it is not used as an input to the fit. The single-photoelectron calibration uses an external PMT response model (ref. [25]) and is monitored independently with an LED, so it does not assume the physics-run light yields. The field-quenching results are empirical ratios of measured peak positions, and the PARIS comparison is cited only as a consistency check, not as an input generating the quoted values. The only self-citation, ref. [23] for cooling, purification, and recirculation details, is descriptive and not load-bearing for any central claim. No circular step can be identified from the paper's own equations or argumentative structure.

Assumptions & free parameters 2 free parameters · 5 assumptions · 0 invented entities

This is an experimental calibration paper, so no new entities are introduced. The free parameters are the single-photoelectron charge calibrations that convert recorded charge to photoelectron number; all reported light yields scale with them. The axioms are domain assumptions about energy deposition, calibration transfer, background, and field uniformity that the paper invokes without independent validation.

free parameters (2)
  • SPE charge for PMT1 (Q1) = 0.5355 +/- 0.0041 pC
    Fitted from the LED single-photoelectron spectrum in Fig. 5. All light yields in photoelectrons per keV are derived by dividing the fitted charge peak by this calibration, so a bias in this value propagates linearly into the central results.
  • SPE charge for PMT2 (Q1) = 0.44 pC
    Second PMT calibration quoted in Section 3.2. Same propagation as PMT1, but no fit parameters are shown for this PMT.
assumptions (5)
  • domain assumption The 32.1 keV and 9.4 keV transitions of 83mKr deposit their full 41.5 keV energy in the active volume and appear as a single pulse because the 154 ns separation is shorter than the argon triplet decay time (~1.6 us).
    Section 4.1 assigns the 41.5 keV peak. If the two transitions are not fully contained or summed, the light yield per keV is underestimated.
  • domain assumption The LED-based single-photoelectron calibration applies to the TPB-shifted 420 nm scintillation pulses during the physics runs.
    Section 3.2 calibrates with an LED only; no in-situ check with scintillation light is reported, and the paper does not estimate a possible systematic offset.
  • domain assumption The 511 keV peak from 22Na is a clean full-absorption peak with negligible background, so the Gaussian-plus-exponential fit gives an unbiased mean.
    Section 4.1 states the background is negligible, but no background-subtracted 22Na spectrum is shown to support this assumption.
  • domain assumption The drift field in the active volume is uniform and equal to the nominal voltage divided by the active height for the stated 0-200 V/cm values.
    Section 2 describes a resistive divider and copper rings but provides no field map or uniformity measurement.
  • domain assumption The PMT response model of Bellamy et al. [25] describes the single-photoelectron spectra of both PMTs.
    Section 3.2 fits the LED spectra with this model for one PMT; the same model is assumed to hold for the second PMT without a shown fit.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Calibration of liquid argon detector with $^{83m}Kr$ and $^{22}Na$ in different drift field." pith.science (2026). https://pith.science/paper/L7E6AS36

@misc{pith2026190902207,
  author       = {Pith},
  title        = {Pith review of: Calibration of liquid argon detector with $^83mKr$ and $^22Na$ in different drift field},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/L7E6AS36}},
  note         = {Machine review of arXiv:1909.02207}
}
abstract

$^{83m}Kr$ and $^{22}Na$ have been used in calibrating a liquid argon (LAr) detector.$^{83m}Kr$ atoms are produced through the decay of $^{83}Rb$ and introduced into the LAr detector through the circulating purification system. The light yield reaches 7.26$\pm$0.02 photonelectrons/keV for 41.5keV from $^{83m}Kr$ and 7.66$\pm$0.01 photonelectrons/keV for the 511keV from $^{22}Na$, as a comparison. The light yield varies with the drift electric field from 50 to 200V/cm have been also reported. After stopping fill, the decay rate of $^{83m}Kr$ with a fitted half-life of 1.83$\pm$0.11 h, which is consistent with the reported value of 1.83$\pm$0.02 h.

Figures

Figures reproduced from arXiv: 1909.02207 by the authors.

Figure 1
Figure 1. Energy level diagram (in keV) for the 83Rb decay. 83Rb decays 75% of the time to the long-lived isomeric 83mKr level that is 41.5keV above the ground state, which subsequently decays in two steps, a 32.1keV transition, typically a conversion electron and asso￾ciated x-rays, followed by a similar 9.4keV transition [20]. TPB ESR Fused silica window Grid PTFE fused silica window HV1 Cathode (HV2) Fused silica window Fu… view at source ↗
Figure 2
Figure 2. Left panel:Three-dimensional diagram of the detector. [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 5
Figure 5. Example of the single photoelectron spectrum of a single [PITH_FULL_IMAGE:figures/full_fig_p003_5.png] view at source ↗
Figures from the paper (5 more)
Figure 6
Figure 6. Figure 6: Before 83mKr run, several background runs have been taken for a background subtraction. The decay of 83mKr to stable 83Kr is characterized by two closely-spaced se￾quential decays producing either IC electrons ,x-rays or γs. In this case the ionization is produced by p…
Figure 6
Figure 6. Figure 6: Scintillation spectrum of 22Na collimated at the central po￾sition, the data is not background subtracted because the background rate is negligible. p.e. 0 500 1000 1500 2000 counts (10pe/bin) 0 200 400 600 800 1000 1200 1400 1600 1800 Kr with background 83m background…
Figure 7
Figure 7. Figure 7: (top) The black and red curve indicate changes before and [PITH_FULL_IMAGE:figures/full_fig_p004_7.png]
Figure 9
Figure 9. Figure 9: The full-absorption peak of 83mKr at all drift fields, the fit results are listed in [PITH_FULL_IMAGE:figures/full_fig_p005_9.png]
Figure 10
Figure 10. Figure 10: The 511keV full-absorption peak of 22Na at all drift fields,the fit results are listed in [PITH_FULL_IMAGE:figures/full_fig_p005_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

30 extracted references · 25 canonical work pages

  1. [1]

    Ade, P. A. R. et al., Planck 2013 results. XVI. Cosmological parameters,Astron. Astrophys. 571, A16 (2014). 5

  2. [2]

    Aprile et al., Dark Matter Results from 225 Live Days of XENON100 Data, Phys

    E. Aprile et al., Dark Matter Results from 225 Live Days of XENON100 Data, Phys. Rev. Lett. 109, 181301 (2012)

  3. [3]

    Tan et al., Dark Matter Results from First 98.7 Days of Data from the PandaX-II Experiment, Phys

    A. Tan et al., Dark Matter Results from First 98.7 Days of Data from the PandaX-II Experiment, Phys. Rev. Lett. 117, 121303 (2016)

  4. [4]

    Agnes et al., Results from the first use of low radioactivity argon in a dark matter search, Phys

    P. Agnes et al., Results from the first use of low radioactivity argon in a dark matter search, Phys. Rev. D 93, 081101 (2016)

  5. [5]

    Zani, The W ArP Experiment: A Double-Phase Argon De- tector for Dark Matter Searches, Adv

    A. Zani, The W ArP Experiment: A Double-Phase Argon De- tector for Dark Matter Searches, Adv. High Energy Phys. 2014, 205107 (2014)

  6. [6]

    Kuzniak et al., DEAP-3600 Dark Matter Search, Nucl

    M. Kuzniak et al., DEAP-3600 Dark Matter Search, Nucl. Part. Phys. Proc. 273C275, 340 (2016)

  7. [7]

    Agnese et al., New Results from the Search for Low-Mass Weakly Interacting Massive Particles with the CDMS Low Ion- ization Threshold Experiment, Phys

    R. Agnese et al., New Results from the Search for Low-Mass Weakly Interacting Massive Particles with the CDMS Low Ion- ization Threshold Experiment, Phys. Rev. Lett. 116, 071301 (2016)

  8. [8]

    Lewin and P

    J. Lewin and P. Smith, Review of mathematics, numerical fac- tors, and corrections for dark matter experiments based on elas- tic nuclear recoil, Astropart. Phys. 6, 87 (1996)

Show all 30 references
  1. [9]

    Aprile et al

    E. Aprile et al. (The XENON Collaboration),arXiv:1805.12562v1 (2018)

  2. [10]

    D. S. Akerib et al. (The LUX Collaboration), Phys.Rev. Lett. 118, 021303 (2017)

  3. [11]

    Cui et al

    X. Cui et al. (The PandaX-II Collaboration), Phys.Rev. Lett. 119, 181302 (2017)

  4. [12]

    C. E. Aalseth et al. (The DarkSide Collaboration),Eur. Phys. J. Plus 133, 131 (2018)

  5. [13]

    et al., arXiv:1707.08145v2(2018)

    Aalseth, C.E., Acerbi, F., Agnes, P. et al., arXiv:1707.08145v2(2018)

  6. [14]

    Hongguang ZHANG et al, arXiv:1806.02229(2018)

  7. [15]

    Aalbers,L

    J. Aalbers,L. Baudis et al, arXiv:1606.07001v1

  8. [16]

    Venos et al., 83Krm radioactive source based on 83Rb trapped in cation-exchange paper or in zeolite, Appl

    D. Venos et al., 83Krm radioactive source based on 83Rb trapped in cation-exchange paper or in zeolite, Appl. Radiat. Isot. 63 (2005) 323; D. Venos et al., Precise energy of the weak 32- keV gamma transition observed in 83Krm decay, Nucl. Instrum. Meth. A 560 (2006) 352; S.C. ...

  9. [17]

    Agnese et al., First results from the DarkSide-50 dark matter experiment at Laboratori Nazionali del Gran Sasso, Phys

    R. Agnese et al., First results from the DarkSide-50 dark matter experiment at Laboratori Nazionali del Gran Sasso, Phys. Rev. B. 743, 456-466 (2015)

  10. [18]

    Aprile et al

    E. Aprile et al. ”First observation of two-neutrino double electron capture in 124Xe with XENON1T”, arXiv:1904.11002v1(2019)

  11. [19]

    Zbo˘ril et al, Ultra-stable implanted 83Rb /83mKr electron sources for the energy scale monitoring in the KATRIN experi- ment, arXiv:1212.5016v2

    M. Zbo˘ril et al, Ultra-stable implanted 83Rb /83mKr electron sources for the energy scale monitoring in the KATRIN experi- ment, arXiv:1212.5016v2

  12. [20]

    L. W. Kastens, S. B. Cahn, A. Manzur, and D. N. McKinsey, Calibration of a liquid xenon detector with 83Krm, Phys. rev. C 80, 045809 (2009)

  13. [21]

    3M Vikuiti, solutions.3m.com, 2012

  14. [22]

    Burton, B.A

    W.M. Burton, B.A. Powell, Applied Optics 12 (1973) 87C89

  15. [23]

    Pei-Xian Li, Meng-Yun Guan et al, Preliminary test results of LAr prototype detector, arXiv:1601.01075v1

  16. [24]

    Kubota, M

    S. Kubota, M. Hishida, A. Nohara, Nuclear Instruments and Methods 150 (1978) 561C564

  17. [25]

    Bellamy et al, 1994 Absolute calibration and monitoring of a spectrometric channel using a photomultiplier, Nucl

    E.H. Bellamy et al, 1994 Absolute calibration and monitoring of a spectrometric channel using a photomultiplier, Nucl. Inst. Meth. A 339 468

  18. [26]

    W. H. Lippincott et al, Calibration of liquid argon and neon de- tectors with 83Krm, Phys. rev. C.81,045803(2010)

  19. [27]

    http://physics.nist.gov/PhysRefData/Star/Text/ESTAR.html

    ESTAR C Stopping power and range tables for elec- trons, National Institute of Standards and Technology. http://physics.nist.gov/PhysRefData/Star/Text/ESTAR.html

  20. [28]

    Kubota, M

    S. Kubota, M. Hishida, M. Suzuki, and J. Ruan (Gen), Dynam- ical behavior of free electrons in the recombination process in liquid argon, krypton, and xenon, Phys. Rev. B 20, 3486 (1979)

  21. [29]

    Agnes et al

    P. Agnes et al. (DarkSide), J. Instrum. 12, P10015 (2017)

  22. [30]

    Agnes et al, Measurement of the liquid argon energy re- sponse to nuclear and electronic recoils, Phys

    P. Agnes et al, Measurement of the liquid argon energy re- sponse to nuclear and electronic recoils, Phys. ReV . D 97, 112005(2018). 6

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.