Pith. sign in

REVIEW 4 major objections 6 minor 38 references

Bulk-Surface Event Discrimination in Point Contact Germanium Detectors at Near-Threshold Energies with Shape-Matching Pulse-Shape Methods

T0 review · 4 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

Pith's one-line read A cross-correlation and low-pass filter seeding step for pulse rise-time fits reduces surface-event leakage by roughly 70% at near-threshold energies and lowers the TEXONO analysis threshold by at least 10 eV_ee.

desk verdict Solid pulser-validated methods paper whose real-data threshold claim outruns its statistics—worth reviewing, needs major revisions. read the letter →

arxiv 2412.00089 v2 pith:S7UYFVHN submitted 2024-11-27 physics.ins-det hep-ex

classification physics.ins-dethep-ex PACS 29.40.Wk
keywords point-contactgermaniumdetectorspulse-shapediscriminationrise-timeanalysiscross-correlationlow-passfiltersurfaceeventleakageTEXONOthreshold
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

Point-contact germanium detectors are used in searches for dark matter and neutrinos because they can see sub-keV energy deposits, but pulses originating near the detector's passive surface are slower and are treated as background. At energies near the analysis threshold the rise-time spectra of bulk and surface events overlap, so misclassified surface events leak into the signal spectrum. This paper proposes seeding the pulse-rise-time fit with parameters obtained by cross-correlating the pulse against a step-function reference and low-pass filtering the pulse to estimate its amplitude and baseline. On pulser-generated pulses the new seeding reduces the false-fit peaks in both bulk and surface rise-time spectra by more than 70%, and on TEXONO reactor data it reduces the near-threshold surface leakage into the bulk spectrum by nearly 70%, allowing the analysis threshold to be lowered by at least 10 eV_ee, from 200 to 190 eV_ee, while cutting computation time roughly in half.

What carries the argument

The key machinery is a two-step seeding procedure for the four-parameter hyperbolic-tangent fit $A/2\tanh(s(t-t_0))+C$. First, the pulse is cross-correlated with a step-function reference that rises from $-n$ to $n$ (a hyperbolic tangent of infinite time constant), whose peak position yields the time offset $t_0$; the step is chosen because its correlation peak is not widened by the reference's own rise time. Second, a low-pass filter smooths the pulse so that amplitude $A$ and pedestal $C$ can be estimated from the asymptotic end segments. The resulting seeds let the nonlinear fit converge faster and more accurately, reducing rise-time fitting errors at low signal-to-noise ratios, where the smoothed-fit method frequently lands on false secondary peaks.

What would settle it

Compare the cross-correlation method and the smoothed-fit method on calibration-source data (for example 241Am or 137Cs) in the 0.18–0.4 keV_ee range, where the true bulk and surface spectral shapes are known independently; if the measured leakage reduction is substantially smaller than 70%, the pulser's single-time-constant surface sample overstates the real-world gain.

Watch

Extended reading notes

Core claim

The central claim is that the quality of the hyperbolic-tangent rise-time fit in pPCGe detectors is controlled by the initial parameter estimates ('seeding'), and that replacing the Savitzky–Golay smoothed-fit seeding with cross-correlation shape-matching plus low-pass filtering sharpens rise-time resolution at near-threshold energies. The paper argues that a step function is the optimal reference shape for cross-correlation because it has no finite rise time of its own, so the correlation peak width reflects only the signal; the correlation maximum gives the pulse time offset $t_0$, and a low-pass filter reduces noise when estimating amplitude $A$ and pedestal $C$. In pulser-generated samples at 0.3 keV_ee, the false-fit peaks in both bulk and surface rise-time spectra shrink by more than 70%, and the primary peaks become more concentrated. In 49 kg-days of TEXONO reactor data, the near-threshold (0.18–0.3 keV_ee) bulk spectrum is reduced by nearly 70% relative to the smoothed-fit method, consistent with suppression of surface leakage, and the full-spectrum analysis shows the lowest analysable bin shifting from 0.20 to 0.19 keV_ee, i.e., a threshold gain of at least 10 eV_ee. The improved seeding also lets the fit be run twice instead of four times, a ~50% computation-time saving.

Load-bearing premise

The load-bearing premise is that the pulser-generated surface events, which use one programmed time constant and so form a much narrower rise-time distribution than real surface events, faithfully represent the behaviour of true near-threshold surface events well enough that the roughly 70% leakage suppression measured on these simplified samples transfers to TEXONO reactor data.

Editorial extensions

If this is right

  • The analysis threshold for TEXONO pPCGe data can be lowered from 200 eV_ee to at least 190 eV_ee, extending the reach of low-energy rare-event searches.
  • Surface-event leakage into the bulk rise-time spectrum is suppressed by roughly 70% at near-threshold energies (0.3 keV_ee in pulser samples; 0.18–0.3 keV_ee in TEXONO data).
  • Rise-time analysis computation time drops by about 50% because the number of fits per pulse is reduced from four to two.
  • The method is robust across detectors: improvements were seen on both the TEXONO pPCGe detector and a local nPCGe test detector.

Reading between the lines

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

  • Because the improvement comes purely from better initial conditions for an existing nonlinear fit, the same cross-correlation seeding idea could benefit other experiments that fit sigmoidal or template pulse shapes at low signal-to-noise ratio.
  • The 10 eV_ee threshold gain is a software-only improvement, so it should compound with hardware noise reduction rather than substitute for it; the paper itself notes hardware advances are still needed.
  • The quantitative leakage-reduction claim rests on pulser-generated surface pulses that use a single time constant, while real surface events span a wider rise-time range; a direct test with calibration sources would show whether the 70% figure survives in situ.
  • Since the bulk-selection curve in the full analysis is optimised on pulser-generated bulk pulses, applying the method to other detectors will require re-optimisation with their own pulser settings.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper proposes a pulse-shape analysis method for p-type point-contact germanium detectors in which a cross-correlation shape-matching step and a low-pass filter are used to seed the four-parameter hyperbolic tangent rise-time fit. The method is developed on programmable pulser-generated bulk-like and surface-like pulses with known true time offsets and rise times, where it is reported to reduce the size of false-fit peaks by more than 70% at 0.3 keV_ee, improve the ROC curve for bulk-surface discrimination, and roughly halve the computation time. The method is then applied to TEXONO reactor data, where the authors report a nearly 70% reduction in the low-energy bulk spectrum and claim that the analysis threshold can be lowered by at least 10 eV_ee, from 200 to 190 eV_ee.

Significance. If the reported improvements are robust, the method would be a simple and portable enhancement to low-energy pulse-shape discrimination in pPCGe detectors, with direct relevance to dark-matter and coherent-elastic-neutrino-nucleus-scattering searches. The use of pulser pulses with known truth information is a genuine strength, as is the explicit ROC-curve comparison and the demonstration of consistency across two detector systems. The main limitation is that the headline quantitative claims—the 70% leakage reduction in real data and the 10 eV_ee threshold gain—rest on comparisons without independent ground truth and on a single low-statistics energy bin, respectively. The core algorithmic idea appears sound and testable, but the current manuscript overstates the strength of the evidence for those specific performance gains.

major comments (4)
  1. [Section V C, Table I] The claimed lowering of the analysis threshold by at least 10 eV_ee is not supported by the quoted numbers. The 0.19 keV_ee bin, which is the only new bin rendered 'analysable', has rates of 206 ± 125 cpkkd (smoothed fit) and 139 ± 53 cpkkd (cross-correlation). The difference of 67 cpkkd has a combined uncertainty of approximately 136 cpkkd, i.e., a significance of roughly 0.5σ. The paper should either present a quantitative threshold-setting procedure that accounts for this uncertainty or explicitly retract the 'at least 10 eV_ee' claim.
  2. [Section IV and Section V B] The pulser surface sample is generated with a single programmed time constant, and the authors acknowledge in Section IV that the resulting surface spectrum 'is much narrower than in data'. The quantitative 70% false-peak suppression measured on this simplified sample (Figure 10) therefore does not by itself establish that the method will suppress surface leakage by the same amount for real surface events, which have a range of rise times and pulse shapes. The manuscript should either obtain pulser surface samples spanning the observed rise-time distribution, or present the real-data result as a qualitative trend rather than a quantitative transfer of the 70% figure.
  3. [Section V C, Figure 13] In the TEXONO data comparison there is no independent ground truth: the 'reduction' is the difference between two algorithms applied to the same unlabeled events. The near-70% decrease in the low-energy bulk spectrum could in principle be caused by a systematic shift of genuine bulk events into the surface region or by a bias in the new t0 estimate at low signal-to-noise ratios, rather than by true removal of surface leakage. The authors should cross-check the result against an independent bulk-surface decomposition, for example the ratio method of reference [16], or demonstrate with injected pulser bulk pulses that genuine bulk events are not lost by the cross-correlation method.
  4. [Section III B] The choice of the time segments 3 to 5.5 microseconds after t0 and the 6 microsecond rise-time upper bound is justified only by the statement that this covers 99% of surface events, but no quantitative support or sensitivity study is provided. Because the same pulser sample is used both to motivate this choice and to validate the method, there is a mild circularity. The paper should discuss how the results depend on these ad hoc parameters, or at least provide a reference measurement for the claimed 99% coverage.
minor comments (6)
  1. [Section III A, Eq. (3)] The discrete cross-correlation sum in Eq. (3) uses the notation S_{j+\Delta t} without specifying how indices are handled at the array edges; please clarify the boundary treatment.
  2. [Section IV] The pulser samples were taken with an nPCGe detector at Academia Sinica while the TEXONO data are from a pPCGe detector at KSNL. The paper normalizes energies by pedestal RMS, but it should also discuss whether differences in detector geometry and electronics could affect the transferability of the pulse-shape conclusions.
  3. [Section V B, Figure 11] The ROC curve in Figure 11 is presented without confidence bands or a statement of the sample sizes used to construct it; adding these would make the improvement more quantitatively assessable.
  4. [Table I] The uncertainties quoted in Table I appear to be much larger than Poisson counting statistics alone (e.g., 206 ± 125 cpkkd at 0.19 keV_ee). The text should specify the statistical and systematic components of these uncertainties and how they were propagated.
  5. [Abstract and Section VI] The phrase 'at least 10 eV_ee' should be softened to reflect the large uncertainty in the 0.19 keV_ee bin, e.g., 'a possible reduction of the analysis threshold by about 10 eV_ee', until a more precise analysis is available.
  6. [General] There are several typographical and wording issues: 'starts to convolve with each other' in the abstract, 'receiver-operation-characteristic' for 'receiver operating characteristic', 'hallow' for 'hollow' in the Figure 15 caption, and 'T est' in the Section V C heading.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the pulser validation provides an external ground truth, and the TEXONO threshold gain is a comparative algorithm effect, not a fitted input relabeled as a prediction.

full rationale

The paper's central validation is self-contained against an external benchmark: pulser-generated pulses with programmed true time offsets and rise times are compared with the fitted outputs (Figs. 8-11), so the claimed ~70% false-peak suppression is an empirical result rather than an identity. The TEXONO comparison (Fig. 13, Table I) is an algorithm-to-algorithm difference on the same events; the interpretation that the reduced bulk-spectrum rate is surface-leakage removal is an inference, not a by-construction equivalence. The method's design choices (step-function reference, low-pass filter, 6 µs rise-time bound for segment placement) are motivated by pulse characteristics and are not fitted to the quantity being predicted. The acknowledged narrowness of the pulser surface sample (single time constant, Sec. IV) is a validity limitation, and the 0.19 keVee rate difference in Table I has large statistical uncertainty, but those are correctness risks rather than circular steps. No load-bearing self-citation chain or uniqueness import is present: citations [11] and [16] supply established background methods, not the new result.

Assumptions & free parameters 4 free parameters · 6 assumptions · 0 invented entities

The method rests on standard pulse-shape assumptions plus several chosen parameters. The central benchmark, pulser pulses with known true parameters, is external, but the time-segment and filter settings are hand-picked and the pulser surface model is simplified. No new physical entities are introduced.

free parameters (4)
  • Rise-time upper bound for segment selection = 6 microseconds
    Chosen in Section III B as an assumed 99% coverage of surface event rise times; determines the 3 to 5.5 microsecond averaging segments for amplitude and pedestal estimates.
  • Averaging time segments for amplitude and pedestal = 3 to 5.5 microseconds before and after t0
    Derived from the 6 microsecond upper bound and the 2.5 microsecond post-95% amplitude requirement; directly affects the A and C estimates that feed the final fit.
  • Low-pass filter cutoff and order = not specified
    The low-pass filter is described qualitatively in Section III B; the smoothed amplitude and pedestal estimates depend on the unstated filter configuration.
  • Cross-correlation integration window T = not specified
    Equation (2) defines T as the integration window width, but no value is reported, and this window sets the range over which the t0 scan is performed.
assumptions (6)
  • domain assumption The induced timing pulse is well described by A/2 * tanh(s(t - t0)) + C.
    Used throughout; Eq. (1). Standard for PCGe pulses but not exact for all surface events.
  • standard math The cross-correlation of two hyperbolic tangent-like signals is maximized when their centers coincide.
    Stated in Section III A; used to justify extracting t0 from a step-function reference.
  • domain assumption A step function is an optimal reference because it has zero rise time and is signal-independent.
    Section III A; argued from the peak-width behavior but not proven against other possible references.
  • domain assumption Low-frequency components contain the pulse shape while high-frequency components are noise.
    Section III B; the basis for applying a low-pass filter when estimating amplitude and pedestal values.
  • domain assumption Pulser-generated pulses with a single programmed time constant are representative of real bulk and surface events at lower energies.
    Section IV; the paper acknowledges that real surface events have a range of rise times, but relies on the pulser sample for quantitative performance claims.
  • ad hoc to paper The time segments 3 to 5.5 microseconds from t0 are in the asymptotic region for 99% of surface events.
    Section III B; depends on the 6 microsecond upper bound, which is itself an assumed value.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Bulk-Surface Event Discrimination in Point Contact Germanium Detectors at Near-Threshold Energies with Shape-Matching Pulse-Shape Methods." pith.science (2026). https://pith.science/paper/S7UYFVHN

@misc{pith2026241200089,
  author       = {Pith},
  title        = {Pith review of: Bulk-Surface Event Discrimination in Point Contact Germanium Detectors at Near-Threshold Energies with Shape-Matching Pulse-Shape Methods},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/S7UYFVHN}},
  note         = {Machine review of arXiv:2412.00089}
}
read the original abstract

The p-type point-contact germanium (pPCGe) detectors have been widely adopted in searches for low energy physics events such as neutrinos and dark matter. This is due to their enhanced capabilities of background rejection, sensitivity at energies as low as the sub-keV range and particularly fine energy resolution. Nonetheless, the pPCGe is subject to irregular behaviour caused by surface effects for events near the passivated surface. These surface events can, in general, be distinguished from events that occur in the germanium crystal bulk by its slower pulse rise time. Unfortunately, the rise-time spectra of bulk and surface events starts to convolve with each other at sub-keV energies. In this work, we propose a novel method based on cross-correlation shape-matching combined with a low-pass filter to constrain the initial parameter estimates of the signal pulse. This improvement at the lowest level leads to a 50% reduction in computation time and refinements in the rise-time resolution, which will, in the end, enhance the overall analysis. To evaluate the performance of the method, we simulate artificial pulses that resembles bulk and surface pulses by using a programmable pulse generator module (pulser). The pulser-generated pulses are then used to examine the pulse behaviours at near-threshold energies, suggesting a roughly 70% background-leakage reduction in the bulk spectrum. Finally, the method is tested on data collected from the TEXONO experiment, where the results are consistent with our observations in pulser and demonstrated the possibility of lowering the analysis threshold by at least 10eV.

Figures

Figures reproduced from arXiv: 2412.00089 by the authors.

Figure 1
Figure 1. FIG. 1. Schematic view of p-type (left) and n-type (right) [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. An example rise-time (log [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. Typical bulk (left) and surface (right) pulses of ener [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: FIG. 4. The cross-correlation profile for bulk (left) and sur [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 5
Figure 5. Figure 5: FIG. 5. An example of a noisy pulse (grey) at 0.25 keV [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 7
Figure 7. Figure 7: shows the rise-time distribution plotted against energy for pulser-generated pulses compared to real data. −10 0 10 20 30 40 50 Estimate (µs) 0 t 1 10 2 10 3 10 4 10 5 10 Events Smoothed Fit Cross-Correlation Pulser Bulk −10 0 10 20 30 40 50 Estimate (µs) 0 t 1 10 2 10…
Figure 9
Figure 9. Figure 9: FIG. 9. The normalised rise-time (log [PITH_FULL_IMAGE:figures/full_fig_p007_9.png]
Figure 11
Figure 11. Figure 11: FIG. 11. The receiver-operation-characteristic (ROC) [PITH_FULL_IMAGE:figures/full_fig_p008_11.png]
Figure 12
Figure 12. Figure 12: FIG. 12. Typical bulk (left) and surface (right) pulses of ener [PITH_FULL_IMAGE:figures/full_fig_p008_12.png]
Figure 13
Figure 13. Figure 13: FIG. 13. The rise-time (log [PITH_FULL_IMAGE:figures/full_fig_p009_13.png]
Figure 15
Figure 15. Figure 15: FIG. 15. The rise-time (log [PITH_FULL_IMAGE:figures/full_fig_p010_15.png]
Figure 16
Figure 16. Figure 16: FIG. 16. The measured energy spectrum (main bottom) of [PITH_FULL_IMAGE:figures/full_fig_p010_16.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references · 38 canonical work pages

  1. [16]

    H. B. Li et al. (TEXONO Collaboration), Limits on spin- independent couplings of wimp dark matter with ap-type point-contact germanium detector, Phys. Rev. Lett. 110, 261301 (2013). 13

  2. [1]

    Refined Rise-Time Resolution Figure 12 shows the fitting results of typical data pulses at 0.25 keV ee and 1.5 keV ee. It is clear that the cross- correlation method produces results that are congruous with the smoothed fit method in the higher energy re- gion, reassuring the reliability of the new algorithm. On the other hand, with the implementation of ...

  3. [2]

    cpkkd” stands for “counts per kg-keV-day

    Impact on Analysis Threshold The discussion on the analysis threshold is hitherto mainly based on qualitative observations. In order to es- tablish a rigorous quantitative assessment for the impact of the cross-correlation method on the analysis threshold, a full analysis on the energy spectrum is necessary. Com- pared to the basic treatments in the rise-...

  4. [3]

    Improved initial parameter estimation: At the low- est level of the rise-time analysis, the fidelity of the hyperbolic tangent fit is largely determined by the initial seeding of the parameters t0, A and C. The results in Figures 5 and 8 have clearly illustrated the substantial enhancements in t0 determination with the cross-correlation calculation and th...

  5. [4]

    Enhanced rise-time determination: Figure 12 eluci- dates the immediate improvements with the more accurate seeding for the fit parameters on deter- mining the rise time at the pulse level. The effects are most prominent at lower energies, where the cross-correlation method has manifested the capa- bilities of correcting the evidently inefficient fits, whi...

  6. [5]

    Resolving the bulk-surface convolution: The accu- rate determination of the pulse rise time largely in- hibits the spectral leakage between bulk and surface events. With the pulser-generated samples, it has been observed that the magnitude of the spectral leakage in the sub-keV region can be reduced by approximately 70% with the implementation of the cros...

  7. [6]

    Lowering analysis threshold: Finally, with the more constrained rise-time distributions and im- proved timing-pulse amplitude estimates, the cross- correlation method features substantial enhance- ment in both the efficiency for bulk selection and the signal retention efficiency for noise cleaning. Combined with the enhanced ability to discern be- tween b...

  8. [7]

    Reduced computation time: Apart from the obvi- ous achievements in analysis, the proposed cross- correlation approach also exhibits another rather desirable feature, that is, time efficiency. With the increased precision in parameter estimation and by obtaining the initial estimates for the parameter via a relatively simple calculation instead of a four- ...

Show all 38 references
  1. [8]

    P. Luke, F. Goulding, N. Madden, and R. Pehl, Low ca- pacitance large volume shaped-field germanium detector, IEEE Trans. Nucl. Sci. 36, 926 (1989)

  2. [9]

    P. S. Barbeau, J. I. Collar, and O. Tench, Large-mass ul- tralow noise germanium detectors: performance and ap- plications in neutrino and astroparticle physics, J. Cos- mol. Astropart. Phys. 2007 (09), 009

  3. [10]

    M. K. Singh et al. , Constraints on bosonic dark matter with low threshold germanium detector at kuo-sheng re- actor neutrino laboratory, Chin. J. Phys. 58, 63 (2019)

  4. [11]

    Soma et al., Characterization and performance of ger- manium detectors with sub-kev sensitivities for neutrino and dark matter experiments, Nucl

    A. Soma et al., Characterization and performance of ger- manium detectors with sub-kev sensitivities for neutrino and dark matter experiments, Nucl. Instr. Meth. Phys. Res. A 836, 67 (2016)

  5. [12]

    K. J. Kang et al., Introduction to the CDEX Experiment, Front. Phys. 8, 412 (2013)

  6. [13]

    Bonet et al

    H. Bonet et al. (CONUS Collaboration), Constraints on Elastic Neutrino Nucleus Scattering in the Fully Coher- ent Regime from the CONUS Experiment, Phys. Rev. Lett. 126, 041804 (2021)

  7. [14]

    C. E. Aalseth et al. (CoGeNT Collaboration), Results from a search for light-mass dark matter with a p-type point contact germanium detector, Phys. Rev. Lett. 106, 131301 (2011)

  8. [15]

    C. E. Aalseth et al. (CoGeNT Collaboration), Cogent: A search for low-mass dark matter using p-type point contact germanium detectors, Phys. Rev. D 88, 012002 (2013)

  9. [17]

    H. T. Wong, Low energy neutrino and dark matter physics with sub-kev germanium detectors, Int. J. Mod. Phys. D 20, 1463 (2011)

  10. [18]

    Li et al

    H. Li et al. , Differentiation of bulk and surface events in p-type point-contact germanium detectors for light wimp searches, Astropart. Phys. 56, 1 (2014)

  11. [19]

    Zhao et al

    W. Zhao et al. (CDEX Collaboration), First results on low-mass wimps from the cdex-1 experiment at the china jinping underground laboratory, Phys. Rev. D88, 052004 (2013)

  12. [20]

    Yue et al

    Q. Yue et al. (CDEX Collaboration), Limits on light weakly interacting massive particles from the cdex-1 ex- periment with a p-type point-contact germanium detec- tor at the china jinping underground laboratory, Phys. Rev. D 90, 091701 (2014)

  13. [21]

    Martin et al., Determining the drift time of charge car- riers in p-type point-contact hpge detectors, Nucl

    R. Martin et al., Determining the drift time of charge car- riers in p-type point-contact hpge detectors, Nucl. Instr. Meth. Phys. Res. A 678, 98 (2012)

  14. [22]

    Aguayo et al

    E. Aguayo et al. , Characteristics of signals originat- ing near the lithium-diffused n+ contact of high purity germanium p-type point contact detectors, Nucl. Instr. Meth. Phys. Res. A 701, 176 (2013)

  15. [23]

    Yang et al., Bulk and surface event identification in p- type germanium detectors, Nucl

    L. Yang et al., Bulk and surface event identification in p- type germanium detectors, Nucl. Instr. Meth. Phys. Res. A 886, 13 (2018)

  16. [24]

    P. A. Burns, L. J. Martin, and J. R. Moroney, Needle beam studies of hpge detectors for photon efficiency cal- ibration from 6 to 25 kev, Nucl. Instr. Meth. Phys. Res. A 286, 480 (1990)

  17. [25]

    U. Tamm, W. Michaelis, and P. Coussieu, A pulse shape discrimination circuit for lithium-drifted germa- nium diodes, Nucl. Instr. Meth. 48, 301 (1967)

  18. [26]

    J. L. Ma et al. , Study of inactive layer uniformity and charge collection efficiency of a p-type point-contact ger- manium detector, Appl. Radiat. Isot. 127, 130 (2017)

  19. [27]

    Savitzky and M

    A. Savitzky and M. J. E. Golay, Smoothing and Differen- tiation of Data by Simplified Least Squares Procedures, Anal. Chem. 36, 1627 (1964)

  20. [28]

    Lu, The Double-Lined Spectroscopic Binary Psi Ori- onis, Publ

    W. Lu, The Double-Lined Spectroscopic Binary Psi Ori- onis, Publ. Astron. Soc. Pac. 97, 428 (1985)

  21. [29]

    H. C. Spruit and G. Kanbach, Correlated X-ray and opti- cal variability in KV UMa, Astron. Astrophys. 391, 225 (2002)

  22. [30]

    Malavolta et al., Atmospheric stellar parameters from cross-correlation functions, Mon

    L. Malavolta et al., Atmospheric stellar parameters from cross-correlation functions, Mon. Not. R. Astron. Soc. 469, 3695 (2017)

  23. [31]

    Blake et al

    C. Blake et al. , On the cross-correlation of sub-mm sources and optically selected galaxies, Mon. Not. R. As- tron. Soc. 368, 732 (2006)

  24. [32]

    D. L. Lorimer, Binary and Millisecond Pulsars, Living Rev. Relativ. 1, 10 (1998)

  25. [33]

    Allen and J

    B. Allen and J. D. Romano, Detecting a stochastic background of gravitational radiation: Signal process- ing strategies and sensitivities, Phys. Rev. D 59, 102001 (1999)

  26. [34]

    Abbott et al

    B. Abbott et al. (LIGO Collaboration), Analysis of First LIGO Science Data for Stochastic Gravitational Waves, Phys. Rev. D 69, 122004 (2004)

  27. [35]

    Coleiro et al., Combining neutrino experimental light- curves for pointing to the next Galactic Core-Collapse Supernova, Eur

    A. Coleiro et al., Combining neutrino experimental light- curves for pointing to the next Galactic Core-Collapse Supernova, Eur. Phys. J. C 80, 856 (2020)

  28. [36]

    Bonet et al

    H. Bonet et al. (CONUS Collaboration), Pulse shape dis- crimination for the conus experiment in the kev and sub- kev regime, Eur. Phys. J. C 84, 139 (2024)

  29. [37]

    H. T. Wong, Taiwan EXperiment On NeutrinO — His- tory and Prospects, Int. J. Mod. Phys. A 33, 1830014 (2018)

  30. [38]

    L. T. Yang et al. (CDEX Collaboration), Limits on light wimps with a 1 kg-scale germanium detector at 160 ev physics threshold at the china jinping underground lab- oratory, Chin. Phys. C 42, 023002 (2018)

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.