Pith. sign in

REVIEW 3 major objections 4 minor 19 references

Efficient and precise Cherenkov-based charged particle timing using SiPMs

T0 review · 3 major / 4 minor · reviewed 2026-08-03 · deepseek-v4-flash

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

Pith's one-line read A Monte Carlo simulation of a fused-silica Cherenkov radiator coupled to a silicon photomultiplier predicts total time resolution below 30 ps for radiator thicknesses of 1 mm or more, and reproduces existing beam-test results.

desk verdict Useful design scan for Cherenkov+SiPM timing, but the sub-30 ps headline is a projection set by unmeasured electronics and SPTR parameters, not an independent measurement. read the letter →

arxiv 2601.14768 v1 pith:OSVSUHJF submitted 2026-01-21 physics.ins-det hep-ex

M. N. Mazziotta , A. Di Mauro , M. Giliberti , A. Liguori , L. Lorusso , E. Nappi , N. Nicassio , G. Panzarini
show 2 more authors
R. Pillera G. Volpe
This is my paper · ORCID
classification physics.ins-dethep-ex
keywords Cherenkovradiationtime-of-flightsiliconphotomultipliertimeresolutionMonteCarlosimulationfusedsilicaphotoelectronstatisticsfront-endjitter
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

The paper argues that a thin fused-silica radiator coupled to a silicon photomultiplier (SiPM) can time charged particles to better than 30 picoseconds, provided the radiator is at least a millimetre thick. It builds a Monte Carlo simulation of Cherenkov photon production, wavelength-dependent detection, and electronic jitter, and shows that the dominant limit is the read-out electronics, not the physics of Cherenkov emission. The simulation reproduces existing beam-test data at 1 mm thickness, giving confidence that the predicted performance is reachable. If true, this makes SiPM-based time-of-flight detectors a practical route to very precise particle identification.

What carries the argument

The central objects are two scaling relations: the photon-arrival-time spread σ_t(d) = Δt_max/√12 with Δt_max ∝ d from Eq. (2), and the photoelectron-statistics jitter σ_pe ≈ σ_SPTR/√N_pe with N_pe ∝ d. These are combined in quadrature with a front-end noise term σ_ele = 50 ps/N_pe ⊕ 20 ps and simulated photon-by-photon in a Monte Carlo that assigns arrival times via Eq. (1). The work these relations do is to expose the design trade-off: thinner radiators reduce geometric spread but raise jitter; the optimum balances the two.

What would settle it

Measure the time resolution of a single SiPM coupled to a 1 mm fused-silica radiator using read-out electronics whose jitter is independently characterized; if the single-sensor resolution comes out above about 32 ps at typical N_pe, the assumed electronic jitter is too optimistic. Alternatively, measure the resolution as a function of light intensity to test whether σ_ele indeed scales as 1/N_pe.

Watch

Extended reading notes

Core claim

The central quantitative claim is that the total time resolution stays below 30 ps for fused-silica radiators of thickness d ≥ 1 mm, for the considered SiPM with 75-µm microcells and 50% peak photon detection efficiency. This emerges from a trade-off: the geometric spread of Cherenkov photon arrival times grows linearly with d, while the photoelectron-statistics term and the front-end jitter shrink as 1/√N_pe and 1/N_pe, respectively. At d = 1 mm, the simulation gives about 30 ps, consistent with the roughly 46 ps (≈32.5 ps per sensor) beam-test result quoted from the paper's references. The paper also shows that the highest-charge pixel dominates the time measurement, and that averaging ove

Load-bearing premise

The predicted sub-30 ps performance rests on an assumed electronic-jitter model (σ_ele = 50 ps/N_pe ⊕ 20 ps) and an assumed single-photon time resolution of 100 ps; these are not measured in the paper, and the authors explicitly note that the read-out electronics set the ultimate limit.

Editorial extensions

If this is right

  • For radiator thicknesses of 1 mm or more, the expected total time resolution is below 30 ps, setting a concrete target for real detector development.
  • The time resolution is driven by the highest-charge pixel; using finer-pitch SiPMs with smaller microcells lowers N_pe and worsens resolution at a given thickness.
  • Averaging over two or three high-charge channels improves resolution, but only when the radiator is thick enough to spread significant charge into neighboring pixels.
  • The number of photoelectrons in the brightest pixel saturates with radiator thickness (around 4 mm for 3-mm SiPMs, 2.5 mm for 2-mm, and 1.3 mm for 1-mm), so adding thickness beyond that point yields diminishing returns.
  • The simulation's 1-mm prediction of about 30 ps per sensor is consistent with the beam-test result of about 32.5 ps per sensor reported in the paper's references.

Reading between the lines

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

  • A direct experimental check would be to measure single-sensor resolution for a 1 mm fused-silica radiator with low-jitter read-out; if it exceeds about 32 ps at typical N_pe, the assumed electronic jitter model (50 ps/N_pe ⊕ 20 ps) is too optimistic.
  • Because optical reflections at the radiator, glue, and coating interfaces are neglected, real assemblies may show degraded timing; adding anti-reflection coatings or index-matched optical cement could recover some of the predicted performance.
  • The σ_ele ∝ 1/N_pe scaling, if verified, implies that improving photon detection efficiency or light collection is as valuable as reducing raw electronic noise when chasing sub-30 ps resolution.
  • If two such arrays were used as start and stop in a full time-of-flight system, the combined resolution would be about √2 times a single array's ~30 ps, i.e. roughly 42 ps; reaching sub-30 ps for the full system would require per-array resolution near 21 ps, which the current electronics model does not support.
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

3 major / 4 minor

Summary. The manuscript presents an analytical and Monte Carlo study of time resolution for a Cherenkov-based Time-of-Flight detector composed of a thin fused-silica radiator optically coupled to SiPM arrays. The authors derive simple scaling formulas for the photon arrival-time spread (Eqs. 1–3), define a SiPM intrinsic jitter term scaling as σ_SPTR/√N_pe (Eq. 4), and model electronic jitter as 50 ps/N_pe ⊕ 20 ps. A Monte Carlo simulation of Cherenkov photon production, SiPM PDE, and Gaussian jitter predicts a total time resolution below 30 ps for radiator thicknesses d ≥ 1 mm for the considered SiPM. The simulation results are compared with the authors' earlier beam-test values from Refs. [11,12], reporting ~46 ps for two sensors, corresponding to ~32.5 ps per sensor, which they argue is consistent with the simulated ~30 ps at 1 mm.

Significance. If the predicted sub-30 ps timing can be validated, the work would be a useful contribution to SiPM-based ToF R&D, particularly because it cleanly separates geometric, intrinsic-SiPM, and electronic contributions and provides a transparent Monte Carlo description. The analytic scalings and the explicit simulation recipe are strengths, as is the attempt to connect with existing beam-test data. However, the central quantitative claim is conditional on two unmeasured parameters — σ_SPTR = 100 ps and σ_ele = 50 ps/N_pe ⊕ 20 ps — and on the neglect of optical reflections and scattering. The manuscript itself acknowledges in Secs. 4 and 5 that exact electronic parameters and reflection effects are not included. The comparison with Refs. [11,12] is a qualitative consistency check using the same group's data, not an independent validation. The paper is therefore a helpful design-oriented study, but its headline number is a conditional projection rather than a demonstrated detector performance.

major comments (3)
  1. [Sec. 3, Eq. (4) and Fig. 3] The sub-30 ps result for d ≥ 1 mm is dominated by the assumed σ_SPTR = 100 ps and σ_ele = 50 ps/N_pe ⊕ 20 ps, neither of which is measured in this paper. At d = 1 mm, with N_pe ≈ 20 in the highest-hit pixel, the SPTR term contributes about 22 ps and the electronic floor about 20 ps, while the photon-arrival-spread term is only about 1–2 ps. If the unmeasured electronic floor is 50 ps instead of 20 ps, or σ_SPTR is 150 ps instead of 100 ps, the 1-mm prediction moves above 30 ps. Since the analytical model (Eq. 4) is essentially a propagation of these inputs, the central claim should either be supported by direct measurements of SPTR and front-end jitter for the actual readout chain, or reframed explicitly as a parameter-dependent projection with a sensitivity study included.
  2. [Sec. 3, Monte Carlo description] The timing model deliberately neglects photon reflections at radiator/glue/coating interfaces and optical scattering, as stated in Sec. 3 and again in Sec. 5. Reflections and interface-related delays can both increase the photon arrival-time spread and reduce the collected photoelectron yield. The claimed sub-30 ps resolution is therefore an upper-bound in timing performance under idealized optical coupling. Since the authors list reflections as future work, the current manuscript should either quantify the expected degradation using a reasonable range of interface parameters (e.g., refractive-index mismatches, anti-reflection coatings) or explicitly state that the prediction applies to a perfectly coupled, reflection-free configuration and should not be read as a realistic detector limit.
  3. [Sec. 4, comparison with Refs. [11,12]] The consistency check against the beam-test result of approximately 46 ps relies on the authors' own earlier measurements and assumes equal contributions from the two sensors to derive 32.5 ps per sensor. The simulated ~30 ps at 1 mm is then stated to be 'consistent' with this value, but this is not a fit or an independent validation. The manuscript itself notes in Sec. 4 that 'for a more rigorous comparison, the exact electronic parameters and the SiPM time resolution from the reference experiments should be incorporated.' Given that the agreement is sensitive to unmeasured parameters and to the equal-contribution assumption, the validation claim should be softened and the sensitivity of the comparison to the assumed parameters should be quantified.
minor comments (4)
  1. [Fig. 3 caption] Typo: 'SIPM' should be 'SiPM'.
  2. [Sec. 4] Typo: 'wavelenght' should be 'wavelength'. Also, in the list of simulated configurations, the notation '3 mm SiPMs', '2 mm SiPMs', '1 mm SiPMs' should define whether these are SiPM pixel/sensor side lengths or another geometrical quantity, and the microcell labels in Fig. 4 ('75 m', '50 m') are missing the μ symbol.
  3. [Sec. 3, electronic jitter] The notation σ_ele = 50 ps/N_pe ⊕ 20 ps should be explicitly defined: '⊕' presumably denotes quadrature sum (σ_ele = sqrt((50 ps/N_pe)^2 + (20 ps)^2)). The scaling arguments for σ_FE ∝ 1/N_pe and σ_TDC = LSB/√12 would also benefit from a brief justification or a reference, since these are central to the results.
  4. [Eqs. (2)–(3) and Fig. 3] Equation (2) uses a single wavelength λ, while the simulation uses a spectral range and PDE weighting. It would help to state more explicitly that the weighted average over the Cherenkov spectrum is used in Fig. 3 and in the Monte Carlo, rather than a monochromatic n(λ).

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the <30 ps timing prediction is a parameter-dependent Monte Carlo output, and the beam-test comparison is an acknowledged external consistency check rather than a fitted input.

full rationale

The paper's central claim is a Monte Carlo expectation for a SiPM-Cherenkov timing detector. The prediction is built from explicit physical and detector inputs: geometry (Eq. 1), a standard Gaussian statistical scaling (Eq. 4, sigma_pe = sigma_SPTR/sqrt(N_pe)), an assumed electronics jitter model (sigma_ele = 50 ps/N_pe xor 20 ps), and the PDE/refracive indices. These are inputs, not outputs that are then relabeled as predictions; the total resolution is the quadrature combination of these independent contributions. The paper does not fit sigma_SPTR or sigma_ele to the quoted beam-test data: it uses 'reference parameters' and then states the 1-mm result 'is consistent' with Refs. [11,12]. The authors explicitly acknowledge the comparison is not rigorous unless the exact electronic parameters and SiPM time resolution from those experiments are incorporated, and Sec. 5 lists electronic jitter and optical reflections as future work. This transparency confirms that the simulation is a conditional projection, not a circular derivation. Refs. [11,12] are self-citations, but they provide external beam-test data and are used only as a sanity check, not as the logical basis for the <30 ps result. No step in the derivation reduces by construction to its own input.

Assumptions & free parameters 3 free parameters · 6 assumptions · 0 invented entities

The central prediction is a Monte Carlo propagation of (i) standard Cherenkov kinematics, (ii) commercial SiPM PDE, and (iii) two hand-assigned jitter models. The first two are external and well-supported; the last two are chosen ad hoc and dominate the result.

free parameters (3)
  • σ_SPTR (SiPM single-photon time resolution) = 100 ps (RMS)
    Chosen as a reference parameter in Sec. 3; enters Eq. 4 (σ_pe ≈ σ_SPTR/√N_pe) and drives the quoted sub-30 ps resolutions.
  • σ_FE noise scaling constant = 50 ps/N_pe
    Introduced in Sec. 3 as the front-end electronic jitter scaling; no measurement or derivation is provided, and it dominates the total resolution for thin radiators.
  • σ_TDC / constant electronic jitter = 20 ps (LSB/√12)
    Added in quadrature in Sec. 3; the LSB value is not specified, and this constant sets the asymptotic resolution floor at large thickness.
assumptions (6)
  • standard math Cherenkov emission angle and spectrum follow cos θ_C = 1/(nβ) and the Frank–Tamm spectrum (PDG Eq. 34.44).
    Used throughout Secs. 3-4 to generate photons and compute arrival times; standard, well-established physics.
  • domain assumption Fused-silica refractive index dispersion n(λ) from Malitson/Tan.
    Cited in Sec. 3; determines chromatic dispersion and photon speeds. Material data are external, not measured here.
  • domain assumption SiPM PDE as a function of wavelength from the Hamamatsu datasheet [19] at nominal bias, 25°C.
    Central to converting photons to photoelectrons (N_pe); taken from a commercial datasheet, not characterized in this work.
  • ad hoc to paper Normal incidence, on-axis track, and no scattering or photon reflections in the timing model.
    Explicitly assumed in Sec. 3 ('neglecting secondary effects (e.g. scattering, and photon reflections)') and Sec. 4; real detectors have non-normal incidence and interface reflections that can add late photons and reduce N_pe.
  • ad hoc to paper SiPM intrinsic and electronic jitter are independent and Gaussian, with σ_SPTR=100 ps and σ_ele=50 ps/N_pe ⊕ 20 ps.
    Stated in Sec. 3; no measurement or justification for the Gaussian form or the numerical constants, and the authors concede this model sets the resolution limit.
  • standard math Number of Cherenkov photons per event follows a Poisson distribution.
    Used in Sec. 4 to generate event-by-event photon counts; standard stochastic model.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Efficient and precise Cherenkov-based charged particle timing using SiPMs." pith.science (2026). https://pith.science/paper/OSVSUHJF

@misc{pith2026260114768,
  author       = {Pith},
  title        = {Pith review of: Efficient and precise Cherenkov-based charged particle timing using SiPMs},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OSVSUHJF}},
  note         = {Machine review of arXiv:2601.14768}
}
read the original abstract

Dedicated R&D efforts are currently underway to couple a thin Cherenkov radiator to Silicon Photomultiplier (SiPM) arrays for precise charged particle Time-of-Flight (ToF) measurements. The prompt nature of Cherenkov radiation makes it an ideal candidate for achieving ultimate timing performance in a ToF detector. Using a thin radiator with a high refractive index, such as fused silica, enables the generation of a fast signal from charged particles that exceed the Cherenkov threshold. A crucial requirement for approaching the target time resolution is the optimization of both the radiator material and thickness, as well as the optical coupling to the SiPM arrays. In this work, we present the main factors that affect the time resolution and the expected performance achieved through a detailed Monte Carlo simulation and the comparison with beam test results.

Figures

Figures reproduced from arXiv: 2601.14768 by the authors.

Figure 1
Figure 1. Cherenkov momentum threshold as a function of the refractive index [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Conceptual sketch depicting the optical coupling of a thin Cherenkov [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Expected time resolution as a function of a fused silica radiator [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Top panel: collected photoelectrons in the three highest-hit pixels as [PITH_FULL_IMAGE:figures/full_fig_p003_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

19 extracted references · 7 canonical work pages

  1. [1]

    Credo, et al., in: 2004 IEEE Nuclear Science Symposium and Medical Imaging Conference, 1, 2004, pp

    T. Credo, et al., in: 2004 IEEE Nuclear Science Symposium and Medical Imaging Conference, 1, 2004, pp. 586–590. doi:10.1109/NSSMIC.2004.1462263

    arXiv 2004

  2. [2]

    Inami, et al., Nucl

    K. Inami, et al., Nucl. Instrum. Methods Phys. Res. A 560 (2006) 303–308. doi:10.1016/j.nima.2006.01.027

    work page doi:10.1016/j.nima.2006.01.027 2006

  3. [3]

    Krizan, J

    P. Krizan, J. Phys. Conf. Ser. 110 (2008) 092015. doi:10. 1088/1742-6596/110/9/092015

    2008

  4. [4]

    Va’vra, et al., Nucl

    J. Va’vra, et al., Nucl. Instrum. Methods Phys. Res. A 606 (2009) 404–410. doi:10.1016/j.nima.2009.04.053

    work page doi:10.1016/j.nima.2009.04.053 2009

  5. [5]

    M. G. Albrow, et al., JINST 7 (2012) P10027. doi:10. 1088/1748-0221/7/10/P10027.arXiv:1207.7248

    arXiv 2012

  6. [6]

    Gundacker, A

    S. Gundacker, A. Heering, Phys. Med. Biol. 65 (2020) 17TR01. doi:10.1088/1361-6560/ab7b2d

    doi:10.1088/1361-6560/ab7b2d 2020

  7. [7]

    Acerbi, S

    F. Acerbi, S. Gundacker, Nucl. Instrum. Methods Phys. Res. A 926 (2019) 16–35. doi:10.1016/j.nima.2018. 11.118

    work page doi:10.1016/j.nima.2018 2019

  8. [8]

    Carnesecchi, et al., Eur

    F. Carnesecchi, et al., Eur. Phys. J. Plus 138 (2023) 788. doi:10.1140/epjp/s13360-023-04397-0. arXiv:2305.17762

    arXiv 2023

Show all 19 references
  1. [9]

    Nicassio, et al., in: 2023 9th International Workshop on Advances in Sensors and Interfaces (IW ASI), 2023, pp

    N. Nicassio, et al., in: 2023 9th International Workshop on Advances in Sensors and Interfaces (IW ASI), 2023, pp. 199–204. doi:10.1109/IWASI58316.2023.10164558

    2023

  2. [10]

    M. N. Mazziotta, et al., JINST 20 (2025) C01001. doi:10.1088/1748-0221/20/01/C01001

    2025 doi

  3. [11]

    M. N. Mazziotta, et al., JINST 20 (2025) C05038. doi:10.1088/1748-0221/20/05/C05038

    2025 doi

  4. [12]

    M. N. Mazziotta, et al., Particles 8 (2025). doi:10.3390/ particles8040094

    2025

  5. [13]

    Brigida, et al., Nucl

    M. Brigida, et al., Nucl. Instrum. Methods Phys. Res. A 533 (2004) 322–343. doi:10.1016/j.nima.2004.05. 127

    2004 doi

  6. [14]

    Mazziotta, Nucl

    M. Mazziotta, Nucl. Instrum. Methods Phys. Res. A 584 (2008) 436–439

    2008

  7. [15]

    Riegler, P

    W. Riegler, P. Windischhofer, Nucl. Instrum. Methods Phys. Res. A 1003 (2021) 165265. doi:10.1016/j. nima.2021.165265.arXiv:2102.00091

    2021

  8. [16]

    Navas, et al

    S. Navas, et al. (Particle Data Group), Phys. Rev. D 110 (2024) 030001. doi:10.1103/PhysRevD.110.030001

    2024 doi

  9. [17]

    I. H. Malitson, Journal of the Optical Society of America (1917-1983) 55 (1965) 1205. doi:10.1364/JOSA.55. 001205

    1917 doi

  10. [18]

    Tan, Journal of Non-Crystalline Solids 223 (1998) 158–163

    C. Tan, Journal of Non-Crystalline Solids 223 (1998) 158–163. doi:https://doi.org/10.1016/ S0022-3093(97)00438-9. [19]https://www.hamamatsu.com/eu/en/product/ optical-sensors/mppc/mppc_mppc-array/ S13360-3075CS.html, 2025

    1998

  11. [20]

    Pillera, et al., Nucl

    R. Pillera, et al., Nucl. Instrum. Methods Phys. Res. A 1080 (2025) 170708. doi:10.1016/j.nima.2025. 170708. 4

    2025 doi

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.