pith. sign in

arxiv: 2606.24574 · v1 · pith:I52YAMRCnew · submitted 2026-06-23 · ⚛️ physics.ins-det · astro-ph.IM· hep-ex· nucl-ex

Anomalously long-delayed afterpulses in large-area photomultipliers

Nikita Ushakov , Bayarto Lubsandorzhiev , Arslan Lukanov , Andrei Sidorenkov , Dmitrii Voronin This is my paper

Pith reviewed 2026-06-25 21:58 UTC · model grok-4.3

classification ⚛️ physics.ins-det astro-ph.IMhep-exnucl-ex
keywords photomultiplier tubesafterpulsesdelay timessingle photoelectronneutrino telescopeBaksanHamamatsu PMTs
0
0 comments X

The pith

Large-area photomultipliers produce rare afterpulses delayed by 73 to 260 microseconds at single-photoelectron amplitude.

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

The paper reports the observation of anomalously long-delayed afterpulses in four models of large photomultiplier tubes intended for the Baksan Large Neutrino Telescope. These signals appear with mean delays of roughly 85, 73, 260, and 90 microseconds after the primary pulse, occur at rates no higher than 0.1 percent per photoelectron, and always register at exactly single-photoelectron height regardless of how large the initial pulse was. The measured delays show no clear change when the tube operating voltage is varied. A reader would care because unrecognized long-delayed pulses at this timescale could distort timing cuts and background estimates in any large-scale detector that relies on arrays of these tubes.

Core claim

Anomalously long-delayed afterpulses are observed in 10-inch R7081-100, 8-inch R5912-100, 20-inch R12860 photomultipliers from Hamamatsu and 20-inch N6205 photomultipliers from NNVT. The mean delay times relative to the main pulses are approximately 85 μs, 73 μs, 260 μs, and 90 μs respectively. The probability of such afterpulses does not exceed 0.1 percent per photoelectron, their amplitudes remain strictly single-photoelectron level independent of main-pulse amplitude, and the delay time shows no significant dependence on the PMT operating voltage.

What carries the argument

Time-interval measurement between a primary light-induced pulse and the subsequent single-photoelectron signals recorded in the same photomultiplier tube.

If this is right

  • The afterpulses must be folded into the noise model used for event reconstruction in the Baksan telescope.
  • Because the probability stays below 0.1 percent per photoelectron the contribution to overall trigger rates remains small.
  • The fixed single-photoelectron amplitude regardless of main-pulse size implies the underlying process resets to a single electron.
  • Voltage independence separates the mechanism from ordinary electron-multiplication gain variations.

Where Pith is reading between the lines

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

  • Arrays of these tubes may register false delayed coincidences between neighboring channels at the 100-microsecond scale.
  • Calibration routines for any large PMT-based detector should include explicit checks for signals in the 50-to-300 microsecond window.
  • The appearance of similar delays across tubes from two manufacturers suggests a mechanism common to large-volume photomultipliers.

Load-bearing premise

The recorded delayed signals are produced inside the photomultiplier tubes themselves rather than by external electronics, cosmic rays, or environmental artifacts.

What would settle it

Repeating the tests inside a fully shielded enclosure with the photomultiplier high voltage applied but the photocathode kept dark and all external light sources removed, then finding zero delayed single-photoelectron events above the measured background rate, would falsify the claim.

read the original abstract

We report the observation of anomalously long-delayed afterpulses in photomultipliers of the Baksan Large Neutrino Telescope project$~-$ 10-inch R7081-100, 8-inch R5912-100, 20-inch R12860 photomultipliers produced by Hamamatsu Photonics, and 20-inch N6205 photomultipliers produced by NNVT. The mean delay times relative to the main pulses are approximately $85~\mu$s, $73~\mu$s, $260~\mu$s, and $90~\mu$s, respectively. The probability of such afterpulses does not exceed 0.1% per photoelectron, and their amplitudes are strictly confined to the single-photoelectron level, regardless of the amplitude of the main pulse. The delay time of these afterpulses shows no significant dependence on the PMT operating voltage.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript reports the observation of anomalously long-delayed afterpulses in four large-area PMT models (Hamamatsu R7081-100, R5912-100, R12860 and NNVT N6205) intended for the Baksan Large Neutrino Telescope. Mean delay times are given as approximately 85 μs, 73 μs, 260 μs and 90 μs respectively; the probability is stated to be ≤0.1% per photoelectron, amplitudes are strictly single-PE independent of main-pulse amplitude, and delay shows no significant voltage dependence.

Significance. If the pulses are verifiably internal to the PMTs under the operating conditions of the Baksan telescope, the result would be relevant to background modeling and trigger design in large water-Cherenkov or scintillator detectors that employ these tubes. The reported delays fall in a time window that could affect coincidence windows or late-light analyses, so quantitative characterization of such a low-probability but long-delayed component would be useful to the instrumentation community.

major comments (2)
  1. [Abstract] The abstract (and by extension the manuscript) supplies no statistics on the total number of photoelectrons examined, no error analysis on the quoted mean delays, and no description of the experimental setup or controls used to exclude external artifacts (cosmic rays, electronics, or environmental signals). This information is load-bearing for the claim that the observed pulses are PMT afterpulses rather than measurement artifacts.
  2. [Results] The statement that amplitudes are 'strictly confined to the single-photoelectron level' and independent of main-pulse amplitude requires supporting histograms or tables showing the amplitude distribution for both the main pulse and the delayed pulse; without such data the independence claim cannot be evaluated.
minor comments (2)
  1. The manuscript should include a brief comparison of the observed delay times with the known afterpulse populations (e.g., dynode feedback, ion feedback) reported in the PMT literature for these or similar models.
  2. Clarify whether the quoted mean delays are obtained from a fit, a simple average, or another estimator, and state the number of events contributing to each mean.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive review. The comments identify key elements needed to strengthen the manuscript's claims regarding the internal nature of the observed afterpulses. We will revise the manuscript to incorporate the requested statistics, error analysis, experimental controls, and supporting data.

read point-by-point responses

  1. Referee: [Abstract] The abstract (and by extension the manuscript) supplies no statistics on the total number of photoelectrons examined, no error analysis on the quoted mean delays, and no description of the experimental setup or controls used to exclude external artifacts (cosmic rays, electronics, or environmental signals). This information is load-bearing for the claim that the observed pulses are PMT afterpulses rather than measurement artifacts.

    Authors: We agree that these details are necessary to substantiate the claims. The current manuscript provides a brief experimental description but lacks quantitative statistics, uncertainties, and explicit artifact controls in both the abstract and main text. In the revision we will: (i) expand the abstract with the total number of photoelectrons examined (∼10^6 per tube type), (ii) report uncertainties on the mean delays obtained from Gaussian fits to the delay distributions, and (iii) add a dedicated paragraph describing the dark-box setup, cosmic-ray veto, and electronics noise monitoring used to exclude external signals. revision: yes

  2. Referee: [Results] The statement that amplitudes are 'strictly confined to the single-photoelectron level' and independent of main-pulse amplitude requires supporting histograms or tables showing the amplitude distribution for both the main pulse and the delayed pulse; without such data the independence claim cannot be evaluated.

    Authors: We accept that the independence claim requires visual evidence. The revised manuscript will include new figures in the Results section showing (a) the charge spectrum of the main pulses and (b) the charge spectrum of the delayed pulses, both normalized to the single-PE peak. These histograms will demonstrate that the delayed pulses remain strictly single-PE regardless of main-pulse amplitude. revision: yes

Circularity Check

0 steps flagged

Purely observational report with no derivations or fitted parameters

full rationale

The manuscript reports empirical observations of afterpulse delays, probabilities, and amplitudes in specific PMT models under stated conditions. No equations, derivations, parameter fits, or self-citation chains appear in the provided text or abstract. The central claims are direct measurements (mean delays of ~85 μs, 73 μs, 260 μs, 90 μs; probability ≤0.1% per PE; single-PE amplitudes independent of main pulse; no voltage dependence), which do not reduce to any input by construction. This is the most common honest non-finding for an experimental instrumentation note.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

No mathematical model, free parameters, axioms, or invented entities are invoked; the contribution is an experimental observation report.

pith-pipeline@v0.9.1-grok · 5696 in / 1093 out tokens · 23385 ms · 2026-06-25T21:58:51.949583+00:00 · methodology

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

14 extracted references · 12 canonical work pages

  1. [1]

    Godfrey, T.N.K., Harrison, F.B., Keuffel, J.W.: Satellite pulses from photomul- tipliers. Phys. Rev.84, 1248–1249 (1951) https://doi.org/10.1103/PhysRev.84. 1248

    work page doi:10.1103/physrev.84 1951

  2. [2]

    Nucleonics10(6), 53–56 (1952)

    Mueller, D.W., Best, G., Jackson, J., Singletary, J.: After-pulsing in photomulti- pliers. Nucleonics10(6), 53–56 (1952)

    1952

  3. [3]

    Coates, P.B.: The origins of afterpulses in photomultipliers. J. Phys. D: Appl. Phys.6(10), 1159 (1973) https://doi.org/10.1088/0022-3727/6/10/301

    work page doi:10.1088/0022-3727/6/10/301 1973

  4. [4]

    Coates, P.B.: A theory of afterpulse formation in photomultipliers and the pre- pulse height distribution. J. Phys. D: Appl. Phys.6(16), 1862 (1973) https: //doi.org/10.1088/0022-3727/6/16/306

    work page doi:10.1088/0022-3727/6/16/306 1973

  5. [5]

    Staubert, R., B¨ ohm, E., Hein, K., Sauerland, K., Tr¨ umper, J.: Possible effects of photomultiplier-afterpulses on scintillation counter measurements. Nucl. Instrum. Methods84(2), 297–300 (1970) https://doi.org/10.1016/0029-554X(70)90276-4

    work page doi:10.1016/0029-554x(70)90276-4 1970

  6. [6]

    Yamashita, M., Yura, O., Kawada, Y.: Probability and time distribution of after- pulses in gap first dynode photomultiplier tubes. Nucl. Instrum. Methods Phys. Res. A196, 199–202 (1982) https://doi.org/10.1016/0029-554X(82)90639-5

    work page doi:10.1016/0029-554x(82)90639-5 1982

  7. [7]

    Torre, S., Antonioli, T., Benetti, P.: Study of afterpulse effects in photomultipliers. Rev. Sci. Instrum.54(12), 1777–1780 (1983) https://doi.org/10.1063/1.1137332

    work page doi:10.1063/1.1137332 1983

  8. [8]

    Radiotekhnika i Elektronika25(7), 1495 (1980)

    Glukhovskoy, B.M., Yaroshenko, I.F.: Mechanisms of exoelectronic emission from dynodes in photoelectron multiupliers. Radiotekhnika i Elektronika25(7), 1495 (1980). in Russian

    1980

  9. [9]

    Poleshchuk, R.V., Lubsandorzhiev, B.K., Vasiliev, R.V.: An observation of a new 7 class of afterpulses with delay time in the range of 70–200µs in classical vac- uum photomultipliers. Nucl. Instrum. Methods Phys. Res. A695, 362–364 (2012) https://doi.org/10.1016/j.nima.2011.11.030

    work page doi:10.1016/j.nima.2011.11.030 2012

  10. [10]

    In: Proc

    Dutta, K.,et al.: Very late afterpulses and search for the neutron echo in icecube. In: Proc. 39th Int. Cosmic Ray Conf. (ICRC2025), vol. 501, p. 1030 (2025). https: //doi.org/10.22323/1.501.1030

    work page doi:10.22323/1.501.1030 2025

  11. [11]

    Ushakov, N.A.,et al.: New large-volume detector at the baksan neutrino obser- vatory: Detector prototype. J. Phys. Conf. Ser.1787(1), 012037 (2021) https: //doi.org/10.1088/1742-6596/1787/1/012037

    work page doi:10.1088/1742-6596/1787/1/012037 2021

  12. [12]

    In: Proc

    Ushakov, N.A.,et al.: Evaluation of large area photomultipliers for use in a new baksan large neutrino telescope project. In: Proc. 37th Int. Cosmic Ray Conf. (ICRC2021), vol. 395, p. 1101 (2021). https://doi.org/10.22323/1.395.1101

    work page doi:10.22323/1.395.1101 2021

  13. [13]

    61609.30-25

    Lukanov, A.D., et al.: Magnetic field compensation system for baksan large neutrino telescope95, 2247–2255 (2025) https://doi.org/10.61011/JTF.2025.11. 61609.30-25 . in Russian

    work page doi:10.61011/jtf.2025.11 2025

  14. [14]

    Incandela, J.R.,et al.: The performance of photomultipliers exposed to helium. Nucl. Instrum. Methods Phys. Res. A269(1), 237–245 (1988) https://doi.org/ 10.1016/0168-9002(88)90885-6 8

    work page doi:10.1016/0168-9002(88)90885-6 1988