REVIEW 4 major objections 5 minor 2 cited by
Sub-keV Electron Recoil Calibration for Macroscopic Cryogenic Calorimeters using a Novel X-ray Fluorescence Source
T0 review · 4 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash
Pith's one-line read Two-stage X-ray fluorescence calibrates cryogenic calorimeters from 677 eV to 8 keV and shows the response is quadratic, not linear.
desk verdict A genuinely useful calibration hardware paper; the nonlinear response is real, but the claims of percent-level accuracy and the phonon-efficiency extraction outrun the statistics. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The enabling device is a two-stage X-ray fluorescence source. A $^{55}$Fe source irradiates an aluminum first target; the aluminum K$\alpha$ fluorescence near 1.5 keV, together with reflected primary radiation, illuminates a second target made of PTFE, glass, and copper, producing characteristic lines from 677 eV to 8.04 keV. The staged geometry exploits the rise of photo-ionisation cross-section at lower photon energies to boost faint low-energy fluorescence, while the 45-degree reflection geometry suppresses the rate of high-energy primary X-rays reaching the detector, and a thin aluminum foil blocks Auger electrons. On the analysis side, an optimal filter reconstructs pulse amplitudes from the transition-edge-sensor stream, and local Gaussian fits around each line convert those amplitudes into an amplitude-versus-energy calibration. A phenomenological resolution model, $\sigma(E)/E = \alpha/\sqrt{N_{\mathrm{at}}(E)} \oplus \beta \oplus \sigma_0/E$ with $N_{\mathrm{at}}(E) = E\,\eta/e_{\mathrm{ath}}$, is then used to separate phonon-statistics effects from position-dependent non-uniformities and baseline noise, yielding a collection-efficiency estimate.
What would settle it
Measure the F K$\alpha$ line emitted by the same PTFE target with a high-resolution spectrometer and compare it with 676.8 eV, then test whether the paper's quadratic curve still predicts the observed pulse amplitudes; alternatively, expose the calorimeter to a tunable monoenergetic source between 0.7 keV and 2 keV and check whether the quadratic calibration, rather than a line through the Mn K$\alpha$ point, predicts the reconstructed amplitudes.
Extended reading notes
Core claim
On its own terms, the paper's central result is that a gram-scale cryogenic calorimeter's reconstructed pulse amplitude is a nonlinear function of deposited electron-recoil energy, and that a multi-line X-ray fluorescence source can measure that function directly. From 18.7 days of exposure, the authors identify ten X-ray lines: F K$\alpha$ at 676.8 eV, Cu L$\alpha$ at 927.7 eV, Al K$\alpha$, Si K$\alpha$, K K$\alpha$, Ca K$\alpha$, Ca K$\beta$, Mn K$\alpha$, Mn K$\beta$, and Cu K$\alpha$. They fit their optimal-filter amplitudes against database energies and find that the quadratic curve $A[\mathrm{mV}] = -2.65\times10^{-6}\,E^2[\mathrm{eV}] + 0.105\,E[\mathrm{eV}]$ is favored over the linear relation through the Mn K$\alpha$ line and zero by $\Delta\chi^2 \approx 2730$ for one extra degree of freedom, which the authors read as excluding the linear hypothesis at far beyond 5$\sigma$. They conclude that a single-line linear extrapolation can misrepresent the energy scale by up to 18% at the baseline and that sub-keV calibration lines are necessary for experiments with thresholds below 100 eV.
Load-bearing premise
The load-bearing premise is that the tabulated X-ray line energies, such as F K$\alpha$ = 676.8 eV and Cu L$\alpha$ = 927.7 eV, are accurate for the chemical state of the actual PTFE, glass, and copper targets; if chemical shifts move those lines beyond the quoted precision, the sub-keV end of the quadratic calibration is biased, and the low F K$\alpha$ rate of about 6 counts per day also limits the statistical pull of the lowest-energy point.
Editorial extensions
If this is right
- Calibrating only with the 5.9 keV Mn K$\alpha$ line and a linear extrapolation can misplace the detector energy scale by up to 18% near the threshold.
- The two-stage XRF source provides usable calibration lines below 1 keV, with the F K$\alpha$ line at 676.8 eV and the Cu L$\alpha$ line at 927.7 eV, although accumulating roughly 1000 counts in the weakest line requires about 4000 hours with the current source.
- Heater test pulses do not reproduce the nonlinearity seen in X-ray pulses in this dataset, so heater-pulse linearization should not be treated as equivalent to particle calibration without further study.
- The resolution model separates baseline noise, position-dependent broadening, and phonon statistics, and estimates the athermal phonon collection efficiency at about half a percent for an assumed mean phonon energy of 1 meV, with a 95% upper limit of 1.45%.
- Extending the calibration to nuclear recoils in the same sub-keV range would make the technique directly applicable to dark-matter and CEvNS searches.
Reading between the lines
- If the same quadratic response is typical of gram-scale TES calorimeters, spectra from detectors calibrated only with $^{55}$Fe lines may need a low-energy reanalysis, since the single-line extrapolation error would reach 18% at baseline.
- A direct test of the chemical-shift systematic would be to measure the F K$\alpha$ and Cu L$\alpha$ energies emitted by the actual PTFE and copper targets with a high-resolution spectrometer, and to see whether the fitted quadratic shifts when target chemistry changes.
- The staged-source design principle should generalize to other line sets, potentially reaching even lower energies such as carbon or oxygen lines near 277 and 525 eV, if rates and background can be controlled.
- A useful next experiment would be to compare XRF calibrations with simultaneous monoenergetic synchrotron or radioactive-source lines in the 0.7–2 keV band to separate true detector nonlinearity from reconstruction artifacts of the optimal filter.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper describes a two-stage X-ray fluorescence (XRF) calibration source for cryogenic calorimeters, based on a 55Fe primary source illuminating aluminum and then secondary targets (PTFE, glass, copper), producing characteristic lines from 676.8 eV to 8.04 keV. The authors report 18.7 days of data from a 0.75 g CaWO4 TES detector, introduce the CryoLab analysis framework, and derive a quadratic calibration curve A [mV] = -2.65e-6 E^2 [eV] + 0.105 E [eV] over 0.68-8.0 keV. They claim a linear calibration anchored at Mn K-alpha and zero would introduce up to 18% error at the baseline. They also compare XRF pulses with artificial heater pulses, and fit a phenomenological resolution model with statistical, systematic, and baseline-noise terms to extract the athermal phonon collection efficiency eta, reporting eta = 0.0045(+0.0034/-0.0026) for an assumed mean phonon energy of 1 meV.
Significance. If the central calibration claim is robust, the two-stage XRF source is a genuinely useful instrumental development: it provides multiple mono-energetic lines down to 677 eV, a region where previous cryogenic calorimeter calibrations relied on linear extrapolation from higher-energy lines. The paper also publicly documents a complete analysis chain (CryoLab) and a reproducible fitting procedure with residual plots and p-values, which are strengths. However, the quantitative claims—especially percent-level calibration accuracy and the 18% baseline correction—are not yet supported by the statistical treatment as presented. The raw quadratic fit has chi2/dof = 103.1, and the ad hoc error rescaling by 10.2 without an identified physical source is a load-bearing gap. The resolution model for eta is a fit to the same data and is explicitly subject to unassessed systematic uncertainties. These issues are addressable, but they currently limit the strength of the conclusions.
major comments (4)
- [Section 5, Fig. 5] The quadratic calibration fit is statistically rejected by the quoted uncertainties: the paper reports chi2/dof = 103.1 with 8 degrees of freedom, and the errors are then rescaled by an ad hoc factor k=10.2 in footnote 3 to force chi2/dof = 1. No physical source of this 10x scatter is identified. Since the calibration claim is 'percent-level' and the peak-location uncertainties are quoted as 0.01%-0.2%, the unmodeled scatter is exactly the kind of systematic that the central claim must control. The paper should either provide a model or bound for this systematic, or explicitly weaken the statement that the calibration is known at the percent level.
- [Section 5, Table 2 and Fig. 5] The sub-keV part of the calibration is anchored by only two lines, F K-alpha at 676.8 eV (rate 6.03 +/- 0.65 cpd, about 113 total counts) and Cu L-alpha at 927.7 eV. The paper itself notes in Section 5 that chemical shifts and satellite lines are a possible systematic error affecting the true line energies. A shift of a few eV in either of these two lines directly changes the quadratic coefficient A2 and hence the extrapolated 18% baseline correction. The authors should propagate the line-energy uncertainties, or at least provide a sensitivity scan showing how A2 and the baseline error change under plausible chemical-shift assumptions.
- [Section 7, Eqs. (1)-(2)] The extraction of the athermal phonon collection efficiency eta is presented as a novel measurement, but it is a fit of the resolution model to the same XRF data used for calibration, with the mean phonon energy e_ath fixed to 1 meV and admittedly uncertain by O(50%). The paper also states in Section 7 that a 'careful assessment of the systematic uncertainties in the resolution sigma(E)' is beyond scope. The reported value eta = 0.0045(+0.0034/-0.0026) is therefore a proof-of-concept demonstration, not a measurement with quantified accuracy. The wording in the text ('evaluating', 'measuring', 'reported for the first time') should be softened accordingly.
- [Section 5, chi-square test] The statement that Delta-chi^2 is approximately 2730 and 'clearly excludes the linear hypothesis at a confidence level exceeding 5 sigma by far' relies on a test statistic in which the quadratic alternative itself is rejected by the data (chi2/dof = 103.1 before rescaling). The unmodeled scatter that inflates the chi2 values makes the Delta-chi^2 statistic unreliable as a significance measure. A proper comparison would require modeling the systematic scatter or using a robust test that does not depend on the quoted statistical errors alone.
minor comments (5)
- [Section 6] Typo: 'an oder of magnitude' should read 'an order of magnitude'.
- [Section 5, footnote 3] The ad hoc error rescaling factor is introduced only in a footnote; since it is central to the reported chi2/dof, it should be brought into the main text and discussed explicitly as a systematic-uncertainty term.
- [Table 2] The column header 'eV/V Resolution' is unclear; it appears to be the peak amplitude divided by energy (mV/eV), but the units as written are ambiguous. Please clarify the definition.
- [References] Reference [20] is a URL to an online X-ray database; the citation should include the retrieval date and version to allow reproducibility of the line energies.
- [Fig. 5] In the residual panel, the caption says 'errorbars scaled by a factor 10.2' but the printed chi2/dof in the top panel is the raw value; make the rescaling explicit in the figure caption to avoid confusion.
Circularity Check
No significant circularity: the calibration curve is an empirical fit to external X-ray line energies, and the phonon-collection-efficiency parameter is explicitly a fitted parameter, not a prediction.
full rationale
The central calibration result is an empirical quadratic fit of measured optimal-filter peak amplitudes (Table 2) to literature line energies taken from an external X-ray database (Table 1, Ref. [20]). The fit coefficients and the Δχ² test against a zero-anchored Mn Kα line are computed from these measured positions and stated external energies, so the nonlinearity claim is not equivalent to an input by construction. The resolution-model section is explicit that Eq. (1)-(2) parameters are obtained by a fit: 'The best-fit value for the ratio η/eath is 0.0045...', with χ²/dof = 1.105 and p = 0.36 reported. η is therefore a fitted parameter inferred from the resolution data, and the paper does not label it an externally validated prediction; the phrase 'novel approach to evaluating athermal phonon collection efficiency' describes a fit-based estimate rather than a predicted quantity. The ad hoc 10.2 error rescaling (footnote 3) and the deferred systematic assessment (Section 7) are statistical and accuracy limitations, not circular reductions. No load-bearing uniqueness theorem, self-citation chain, or ansatz smuggled in via citation forces the choice of model; cited prior work ([15], [25]) is used for motivation, algorithm description, or cross-checking, not to establish the calibration result. Thus no step satisfies the required 'equivalent by construction' condition for circularity.
Assumptions & free parameters
free parameters (6)
- Quadratic calibration coefficient A1 =
0.105 mV/eV
- Quadratic calibration coefficient A2 =
-2.65e-6 mV/eV^2
- Calibration error rescaling factor k =
10.2
- Athermal phonon efficiency to phonon-energy ratio eta/e_ath =
0.0045 +0.0025/-0.0012 per meV
- Non-uniformity constant beta =
0.013 +/- 0.0011
- Baseline resolution sigma0 =
9.9 eV
assumptions (5)
- domain assumption Mean athermal phonon energy e_ath = 1 meV
- domain assumption Poisson statistics for athermal phonon number fluctuations (alpha = 1)
- domain assumption X-ray database line energies are accurate (Table 1 from ref [20])
- ad hoc to paper Detector response is quadratic in energy over 0.68-8 keV
- domain assumption Pulse shape is independent of energy
Cite this review
Pith. "Pith review of Sub-keV Electron Recoil Calibration for Macroscopic Cryogenic Calorimeters using a Novel X-ray Fluorescence Source." pith.science (2026). https://pith.science/paper/7LVVQ74S
@misc{pith2026250517686,
author = {Pith},
title = {Pith review of: Sub-keV Electron Recoil Calibration for Macroscopic Cryogenic Calorimeters using a Novel X-ray Fluorescence Source},
year = {2026},
howpublished = {\url{https://pith.science/paper/7LVVQ74S}},
note = {Machine review of arXiv:2505.17686}
}
abstract
Percent-level calibration of cryogenic macro-calorimeters with energy thresholds below 100~eV are crucial for light Dark Matter (DM) searches and reactor neutrino studies based on coherent elastic neutrino-nucleus scattering (CEvNS). This paper presents a novel calibration source based on X-ray fluorescence (XRF) of light elements. It uses a $^{55}$Fe source to irradiate a two-staged target arrangement, emitting characteristic emission lines from 677\,eV to 6.5\,keV. We demonstrate the potential of this new XRF source to calibrate a 0.75 gram CaWO$_4$ crystal of the NUCLEUS and CRAB experiments. Additionally, we introduce CryoLab, an advanced analysis tool for cryogenic detector data, featuring robust methods for data processing, calibration, and high-level analysis, implemented in MATLAB and HDF5. We also present a phenomenological model for energy resolution, which incorporates statistical contributions, systematic effects, and baseline noise, enabling a novel approach to evaluating athermal phonon collection efficiency in macro-calorimeters based on transition edge sensors (TES).
Forward citations
Cited by 2 Pith papers
-
Prospect of the NUCLEUS Experiment at Chooz for Coherent Elastic Neutrino-Nucleus Scattering and New Physics Searches
Assuming the low-energy background can be eliminated, a 7-gram NUCLEUS detector at Chooz is projected to see coherent neutrino-nucleus scattering at 4.7σ in one year and to set competitive bounds on new neutrino interactions.
-
Observation of a low energy nuclear recoil peak in the neutron calibration data of an Al$_{2}$O$_{3}$ crystal in CRESST-III
CRESST-III observes a 1.11 keV nuclear-recoil peak in an Al2O3 crystal during AmBe neutron irradiation, attributed to 27Al neutron capture, providing a new low-energy calibration line for cryogenic detectors.
Reference graph
Works this paper leans on
-
[1]
Gram-scale cryogenic calorimeters for rare-event searches
Strauss, R., et al. : Gram-scale cryo- genic calorimeters for rare-event searches. Phys. Rev. D 96(2), 022009 (2017) arXiv:1704.04317 [physics.ins-det]. https: //doi.org/10.1103/PhysRevD.96.022009
work page Pith review arXiv 2017
-
[2]
Physical Review D 104(3) (2021)
Ren, R., et al.: Design and characteriza- tion of a phonon-mediated cryogenic par- ticle detector with an ev-scale threshold and 100 kev-scale dynamic range. Physical Review D 104(3) (2021). https://doi.org/10. 1103/physrevd.104.032010
work page 2021
-
[3]
Augier, C., et al. : First demonstration of 30 eVee ionization energy resolution 12 with Ricochet germanium cryogenic bolome- ters. Eur. Phys. J. C 84(2), 186 (2024) arXiv:2306.00166 [astro-ph.IM]. https://doi. org/10.1140/epjc/s10052-024-12433-1
arXiv 2024
-
[4]
Journal of Low Temperature Physics 209(3–4), 510–517 (2022)
Wen, O., Aralis, T., Basu Thakur, R., Bumble, B., Chang, Y.-Y., Ramanathan, K., Golwala, S.R.: Performance of a phonon-mediated detector using kids opti- mized for sub-gev dark matter. Journal of Low Temperature Physics 209(3–4), 510–517 (2022). https://doi.org/10.1007/ s10909-022-02764-2
work page 2022
-
[5]
Abdelhameed, A.H., et al.: First results from the CRESST-III low-mass dark matter pro- gram. Phys. Rev. D 100(10), 102002 (2019) arXiv:1904.00498 [astro-ph.CO]. https://doi. org/10.1103/PhysRevD.100.102002
arXiv 2019
-
[6]
https://arxiv.org/abs/2503.03683
Chang, C.L., et al.: First Limits on Light Dark Matter Interactions in a Low Thresh- old Two Channel Athermal Phonon Detector from the TESSERACT Collaboration (2025). https://arxiv.org/abs/2503.03683
arXiv 2025
-
[7]
Physical Review Letters 127(6) (2021)
Alkhatib, I., et al.: Light dark matter search with a high-resolution athermal phonon detector operated above ground. Physical Review Letters 127(6) (2021). https://doi. org/10.1103/physrevlett.127.061801
-
[8]
Angloher, G., et al.: Exploring CE νNS with NUCLEUS at the Chooz nuclear power plant. Eur. Phys. J. C 79(12), 1018 (2019) arXiv:1905.10258 [physics.ins-det]. https:// doi.org/10.1140/epjc/s10052-019-7454-4
arXiv 2019
Show all 30 references
-
[9]
In: 19th International Workshop on Low Temperature Detectors (2021)
Augier, C., et al.: Ricochet Progress and Sta- tus. In: 19th International Workshop on Low Temperature Detectors (2021)
2021
-
[10]
Springer Proc
Chaudhuri, M.: The Mitchell Insti- tute Neutrino Experiment at Reactor (MINER). Springer Proc. Phys. 277, 589–593 (2022). https://doi.org/10.1007/ 978-981-19-2354-8 107
2022
-
[11]
Jour- nal of Cosmology and Astroparticle Physics 2018(11), 016–016 (2018)
Billard, J., Johnston, J., Kavanagh, B.J.: Prospects for exploring new physics in coher- ent elastic neutrino-nucleus scattering. Jour- nal of Cosmology and Astroparticle Physics 2018(11), 016–016 (2018). https://doi.org/ 10.1088/1475-7516/2018/11/016
2018 doi
-
[12]
: The ν-cleus exper- iment: A gram-scale fiducial-volume cryogenic detector for the first detection of coherent neutrino-nucleus scatter- ing
Strauss, R., et al. : The ν-cleus exper- iment: A gram-scale fiducial-volume cryogenic detector for the first detection of coherent neutrino-nucleus scatter- ing. Eur. Phys. J. C 77, 506 (2017) arXiv:1704.04320 [physics.ins-det]. https: //doi.org/10.1140/epjc/s10052-017-5068-2
2017 arXiv
-
[13]
(ed.) Transition-Edge Sensors, pp
Irwin, K.D., Hilton, G.C.: In: Enss, C. (ed.) Transition-Edge Sensors, pp. 63–150. Springer, Berlin, Heidelberg (2005). https: //doi.org/10.1007/10933596 3. https://doi. org/10.1007/10933596 3
2005 doi
-
[14]
Abele, H., et al.: Observation of a nuclear recoil peak at the 100 ev scale induced by neutron capture. Phys. Rev. Lett. 130, 211802 (2023). https://doi.org/10.1103/ PhysRevLett.130.211802
2023
-
[15]
Astroparticle Physics 32(6), 318–324 (2010)
Lang, R.F.a.o.: Electron and gamma back- ground in cresst detectors. Astroparticle Physics 32(6), 318–324 (2010). https://doi. org/10.1016/j.astropartphys.2009.09.009
2010 doi
-
[16]
Cardani, L., Casali, N., Colantoni, I., Cru- ciani, A., Di Domizio, S., Martinez, M., Pettinacci, V., Pettinari, G., Vignati, M.: Final results of CALDER: kinetic induc- tance light detectors to search for rare events. Eur. Phys. J. C 81(7), 636 (2021) arXiv:2104.06850 [physic...
2021 arXiv
-
[17]
Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 354(2), 408–416 (1995)
Colling, P., Nucciotti, A., Bucci, C., Cooper, S., Ferger, P., Frank, M., Nagel, U., Pr¨ obst, F., Seidel, W.: Low-energy x-ray detection in cryogenic detectors with tungsten ther- mometers. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrome...
1995 doi
-
[18]
Alkhatib, I., et al.: Light dark matter search with a high-resolution athermal phonon detector operated above ground. Phys. Rev. Lett. 127, 061801 (2021). https://doi.org/10. 1103/PhysRevLett.127.061801
2021
-
[19]
Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 370(1), 208–210 (1996)
Leblanc, E., Coron, N., Leblanc, J., de San- ott, J., Nollez, G., Plagnard, J., Bouchard, J., de Marcillac, P.: Bolometer characteri- sation with a specially developed cryogenic source having more than five peaks in the 1–6 kev range. Nuclear Instruments and Methods in Physics...
1996 doi
-
[20]
Center for Advanced Radiation Sources, University of Chicago
X-ray DB (version 2024.11.0). Center for Advanced Radiation Sources, University of Chicago. https://xraydb.xrayabsorption. org/
2024
-
[21]
Accessed January
BlueFors: LD, Dilution Refrigerator Mea- surement System. Accessed January
-
[22]
https:// arxiv.org/abs/2501.04471
Wex, A., et al.: Decoupling Pulse Tube Vibra- tions from a Dry Dilution Refrigerator at milli-Kelvin Temperatures (2025). https:// arxiv.org/abs/2501.04471
2025
-
[23]
https://starcryo.com/lts-sensors/
STAR Cryoelectronics, low-Tc DC SQUID sensors. https://starcryo.com/lts-sensors/. [Online; accessed 10-January-2025] (2025)
2025
-
[24]
Angloher, G., et al.: Deep-underground dark matter search with a cosinus detector proto- type. Phys. Rev. D110, 043010 (2024). https: //doi.org/10.1103/PhysRevD.110.043010
2024 doi
-
[25]
Pr¨ obst, F., Frank, M., Cooper, S., Colling, P., Dummer, D., Ferger, P., Forster, G., Nuc- ciotti, A., Seidel, W., Stodolsky, L.: Model for cryogenic particle detectors with super- conducting phase transition thermometers. J. Low Temp. Phys. 100(1), 69–104 (1995). https://doi...
1995 doi
-
[26]
The MathWorks, Inc., Natick, Massachusetts, United States
The MathWorks, Inc.: MATLAB. The MathWorks, Inc., Natick, Massachusetts, United States. The MathWorks, Inc.. Version R2023a
-
[27]
The HDF Group: Hierarchical Data Format, Version 5. (1997). http://www.hdfgroup.org/ HDF5/
1997
-
[28]
Gatti, E., Manfredi, P.F.: Processing the Signals From Solid State Detectors in Ele- mentary Particle Physics. Riv. Nuovo Cim. 9N1, 1–146 (1986). https://doi.org/10.1007/ BF02822156
1986
-
[29]
https://arxiv
Hell, N., Brown, G.V., Eckart, M.E., Fairchild, A.J., Kilbourne, C.A., Leuteneg- ger, M.A., Porter, F.S., Witthoeft, M.C.: Frequently Used References For Atomic Data In X-ray Spectroscopy (2025). https://arxiv. org/abs/2506.17106 14 Appendix 0.07 0.08 0.09 0.1 10 20 30 40 50 6...
2025 arXiv
-
[2025]
https://bluefors.com/products/ dilution-refrigerator-measurement-systems/ ld-dilution-refrigerator-measurement-system/
Reviewed August 7, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.