REVIEW 2 major objections 5 minor 11 references
Gamma-ray and high-energy X-ray detection with large area scintillating crystals: a hands-on review
T0 review · 2 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash
Pith's one-line read A hands-on review argues that cerium-doped scintillating crystals, led by LaBr3:Ce, remain the practical gold standard for 0.1–2 MeV gamma spectroscopy, and that silicon photomultiplier arrays with temperature compensation can now match…
desk verdict A useful hands-on review with a solid central message, but the abstract's 'up to 2% at 662 keV' claim overstates the evidence in its own body and needs to be corrected. 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 load-bearing object is the scintillation detector chain: an inorganic crystal converts an incoming photon into visible or ultraviolet light, and a photodetector converts that light into a measurable pulse. The quantitative workhorse is the variance decomposition of energy resolution, $(\Delta E/E)^2 = \delta_{\rm scint}^2 + \delta_{\rm tr}^2 + \delta_{\rm stat}^2 + \delta_{\rm noise}^2$, with the statistical term $\delta_{\rm stat} = 2.355\sqrt{F/N_{\rm pe}}$; this separates crystal properties from readout properties and explains why bright, fast, linear crystals paired with high-photon-detection-efficiency photodetectors approach the 2–3% range. The second mechanism is the temperature dependence of SiPM breakdown voltage, $V_{\rm bd}(T) = V_{\rm bd}(T_0)(1+\beta(T-T_0))$, corrected online by a temperature sensor on the back of the array; the quoted SiPM-read resolutions assume this correction, which the paper reports reduces pulse-height drift from 41% to 5% over 10–30 °C. The third mechanism is SiPM array ganging, meaning the electrical interconnection of individual silicon photomultiplier cells, with parallel, series, hybrid, and a custom four-quadrant '4-1' circuit trading capacitance, operating voltage, and noise to set signal rise and fall times and achievable resolution.
What would settle it
Place a 1-inch LaBr3:Ce detector with SiPM array and online temperature correction in a climatic chamber, cycle the temperature between 10 and 30 °C while the detector is live, and measure the FWHM of the 662 keV photopeak at each temperature; if the width grows by more than the statistical uncertainty of the ~3% value, the assumption of full gain-drift compensation fails, and the SiPM-read resolutions are only valid at fixed temperature.
Extended reading notes
Core claim
The central claim is that LaBr3:Ce is the 'golden standard' for gamma-ray and high-energy X-ray spectroscopy in the 0.1–2 MeV range, delivering FWHM energy resolution of about 2.9% at 662 keV and 6.6% at 122 keV with a decay time near 30 ns, and that cerium-doped crystals bring resolutions up to about 2% at 662 keV within reach. The second claim is that silicon photomultiplier arrays, once their temperature-dependent gain drift is corrected online and the array is wired in a suitable ganging scheme, reproduce the energy resolution of photomultiplier tubes (for example, 2.6% for a 3-inch crystal with a custom SiPM module and 3.0% for a 1-inch detector with a commercial SiPM array, versus 3.0–3.5% for PMT-read detectors of similar size). Timing is also preserved: a small LaBr3:Ce(5%) crystal read by a SiPM gives a 10–90% rise time near 9 ns, and a 101±2 ps coincidence resolving time has been obtained with two small LaBr3:Ce detectors, enabling time-of-flight positron-emission tomography. The review also argues that co-doping LaBr3:Ce with Sr or Ba improves light output by about 25% and reduces resolution by about 14%, while the main remaining limitations are intrinsic radioactivity from 138La, 176Lu, and 227Ac, hygroscopicity, and incomplete understanding of why measured resolution stays above the intrinsic crystal limit.
Load-bearing premise
The quoted ~3% energy resolutions for SiPM-read LaBr3:Ce detectors assume that the temperature sensor on the back of the array, through its online correction, removes essentially all gain drift, even though the paper's own data show drift falling only from 41% to 5% over 10–30 °C, with that residual not folded into the reported resolutions.
Editorial extensions
If this is right
- If LaBr3:Ce with SiPM readout matches PMT resolution, compact gamma spectrometers can operate inside magnetic fields, at low voltage, and with reduced power, options previously closed to PMT-based systems.
- Temperature-compensated SiPM arrays make field-deployable and satellite gamma detectors practical across a 10–30 °C range, as demonstrated by a satellite gamma-ray observatory's in-orbit gain correction reducing non-uniformity from 17% to 0.6%.
- Timing near 100 ps coincidence resolving time supports time-of-flight positron-emission tomography with about 15 mm position resolution, and fast signal fall times enable pulse-shape separation of prompt and delayed X-rays in muonic-atom experiments.
- Co-doped LaBr3:Ce and non-hygroscopic crystals such as PrLuAG and Ce:GAGG are positioned as the next step, but their current resolutions or intrinsic backgrounds keep unmodified LaBr3:Ce as the best compromise.
- The resolution decomposition implies that once readout statistics improve, intrinsic crystal nonlinearity, not the photodetector, will limit further gains in energy resolution.
Reading between the lines
- If the residual 5% pulse-height drift after temperature correction is not negligible in rapidly varying thermal environments, the quoted ~3% SiPM-read resolutions could degrade outside laboratory conditions; testing the correction under a realistic field temperature profile would settle this.
- The same variance decomposition suggests a testable scaling law: as SiPM photon detection efficiency rises and dark noise falls, LaBr3:Ce energy resolution should approach the intrinsic crystal resolution set by non-proportionality, so improving crystal linearity through co-doping may matter more than further readout optimisation.
- The 4-1 ganging result implies a general design recipe—subdivide the SiPM array into smaller parallel-ganged quadrants, then sum with pole-zero compensation—that should extend to larger crystals and to other fast scintillators such as CeBr3 or LBC, reducing fall times by roughly a factor of two without sacrificing resolution.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. This manuscript is a hands-on review of inorganic scintillating crystal detectors for high-energy photons in the 0.1-2 MeV range, covering crystal properties (Table 1), readout options (PMTs, SiPM arrays, Si-APDs, SDDs), optical coupling, SiPM ganging schemes, temperature-gain correction, and detector system examples from the FAMU, G-NUMEN, and GECAM experiments. The central claim is that LaBr3:Ce, typically read out by SiPM arrays, is the current 'golden standard' for gamma-ray spectroscopy in this energy range, offering energy resolutions around 2.9-3% at 662 keV and fast timing, with SiPM readout approaching PMT performance. The abstract and conclusions additionally state that energy resolutions up to 2% at 662 keV are 'within reach'.
Significance. If the quantitative inconsistencies are resolved, this review will be a useful practical reference for practitioners building crystal-based detectors, particularly because it compiles published performance data (Tables 1, 3-6) and shares concrete hands-on experience from the FAMU program, including SiPM array ganging and temperature correction. The candid discussion of trade-offs (e.g., hygroscopicity, intrinsic background, SiPM gain drift) is valuable. The claim that SiPM readout can match PMT performance is supported by the cited best result of 2.6% at 662 keV (Table 6), so this part of the message is credible. However, the headline 'up to 2%' figure in the abstract and conclusions is not supported by the body of the paper and must be corrected or explicitly sourced.
major comments (2)
- [Abstract and Section 5 (Conclusions)] The claim 'Energy resolutions up to 2 % at 662 keV' is unsupported by the evidence presented in the manuscript itself. The best resolution cited for LaBr3:Ce in Table 6 is 2.6% at 662 keV (Vita et al. 2022), Table 4 lists 2.65% at 847 keV (Omer et al. 2013), and Section 3.2 states 'up to 2.5% at 662 keV' for co-doped LaBr3:Ce (Yang et al. 2012). Even applying the 14% resolution improvement attributed to Sr/Ba co-doping to the 2.9% base value yields 2.5%, not 2%. Since the abstract and conclusions are the most-quoted parts of a review, this internal inconsistency must be fixed: either trace the 2% value to a specific measurement/derivation or correct the claim to 'up to 2.5%' or '~2.5-3%'.
- [Section 2.3.2.1 and Figure 2] The temperature correction study shows that the pulse-height drift of the 662 keV photopeak is reduced from 41% to 5% over 10-30°C, and the text asserts that 'no major degradation of the 137Cs photopeak is observed' with the correction applied. However, the impact of the remaining 5% drift on the energy resolution values quoted elsewhere (e.g., Table 6) is not explicitly quantified or discussed. Since the review argues that SiPM readout is a practical alternative to PMTs, the authors should state the residual effect of the 5% pulse-height drift on the reported energy resolutions, or clarify that the Table 6 results were obtained under temperature-stabilized conditions.
minor comments (5)
- [Abstract] There are several typographical errors: 'experimenta devices' should be 'experimental devices', 'Development are under way' should be 'Development is under way', and 'CE:GAAG' should be 'Ce:GAGG' for consistency with Table 1 and the rest of the text.
- [Section 2.1 and Table 1] The crystal name appears inconsistently as 'PrLuAg' in the abstract and text but 'PrLuAG' in Table 1; please standardize. The same applies to 'Ce: GAAG' vs 'Ce:GAGG'.
- [Section 2.4] The acronym FADC is introduced as 'Flash Analog to Digital Converter', but later the text describes a 'Flash to Analog to Digital Converter (FADC)'; please align the terminology.
- [Introduction] The historical name 'Hofstader' is used in the text and reference list, but the standard spelling is 'Hofstadter'; please correct.
- [Section 3, Figure 4] The fit to the LaBr3:Ce resolution data is described as 'intended to guide the eye', but the statement 'The FWHM decreases linearly as a function of 1/√E' is only partially supported by the figure; a clearer caption indicating the fit function and its parameters would avoid ambiguity.
Circularity Check
No circularity: the review compiles independently published measured data and does not derive its claims from fitted inputs or self-citations.
full rationale
This is a hands-on review paper, not a derivation or modeling paper. Its central statements about LaBr3:Ce energy resolution, timing, and SiPM readout are supported by tables compiled from published measurements by many independent groups (e.g., Table 4: Omer et al. 2013, Quarati et al. 2007, Giaz et al. 2014; Table 6: Vita et al. 2022, Poleshchuck et al. 2021, He et al. 2023). No quantity is fitted to one data subset and then 'predicted' for another; the resolution decomposition in Section 3 is a standard error-budget formula, and the paper does not use it to manufacture a prediction. The FAMU-related self-citations (Bonesini et al. 2016, 2020, 2022, 2023a, 2023b) are reports of laboratory measurements and experimental apparatus, not unverified premises that the review's conclusions reduce to. The temperature-gain correction discussion in Section 2.3.2.1 reports measured drift reduction (41% to 5%) rather than using that correction as an assumed input to a derived result. The abstract and conclusions state 'energy resolutions up to 2% at 662 keV,' while the body's best cited values are about 2.5-2.6%, so the headline number appears unsupported by the paper's own evidence; however, that is an internal consistency or correctness concern, not a circularity, because the claim is not obtained by defining an input in terms of an output. No step in the paper reduces, by construction or by self-citation, to its own inputs. Accordingly, the circularity score is 0.
Assumptions & free parameters
assumptions (2)
- standard math Standard scintillation model: N = E/(βEgap) × QET × QEL, with β in the range 2-3, describes photon production in inorganic crystals.
- domain assumption The crystal properties compiled in Table 1 from published data and producers' datasheets are representative and accurate.
Cite this review
Pith. "Pith review of Gamma-ray and high-energy X-ray detection with large area scintillating crystals: a hands-on review." pith.science (2026). https://pith.science/paper/UD54PRO3
@misc{pith2026250506929,
author = {Pith},
title = {Pith review of: Gamma-ray and high-energy X-ray detection with large area scintillating crystals: a hands-on review},
year = {2026},
howpublished = {\url{https://pith.science/paper/UD54PRO3}},
note = {Machine review of arXiv:2505.06929}
}
read the original abstract
Detection of photons with scintillating inorganic crystals in the high-energy range (> 0.1 MeV) will be discussed, making a comparison with other available methods. Energy resolutions up to 2 % at 662 keV and fast decay time of the order of 20 ns are within reach, with the introduction of Ce-doped crystals instead of alkali halide ones. Development is underway for the production of non-hygroscopic scintillating crystals, such as PrLuAg and CE:GAAG. At the end of this review, examples of experimental devices based on scintillating inorganic crystals will be discussed. Practical hands-on experience is emphasized at the expense of a more comprehensive description of all available and possible options. Detectors' construction details and the consequences of the different choices will be discussed. Emphasis will be put on the LaBr3:Ce-based detectors that are the present ''golden standard'' in gamma ray spectroscopy. This review will focus on photon detection in the high-energy region: mainly 0.1-2 MeV, including both gamma rays and high-energy X-rays, even if many considerations may be applied to the detection of low-energy X-rays.
Figures
Figures from the paper (4 more)
Reference graph
Works this paper leans on
-
[1]
Ackerman, U. (2015). Time and energy resolution measurements of BaF 2, BC-418, LYSO and CeBr3 scintillators. Nucl. Instr. Meth. A786, 5–11. doi:10.1016/j.nima.2015.03.016 Adamczak, A. et al. (2012). Hyperfine spectroscopy of muonic hydrogen and the PSI Lamb Shift Experiment. Nucl. Instr. Meth. B281, 72–76. doi:10.1016/j.nimb.2012.04.001 Adamzack, A. et al....
-
[24]
doi:10.1016/j.nima.2016.06.044 Grimm, O. et al. (2003). A channel photomultiplier with a scintillator faceplate. Nucl. Instr. Meth. A513, 644–646. doi:10.1016/j.nima.2003.07.034 Gu, F. et al. (2023). Achieving significant performance recovery of SiPM’s irradiation damage with in-situ current annealing. Nucl. Instr. Meth. A1053, 168381. doi:10.1016/j.nima.2...
-
[33]
doi:10.1016/j.nima.2013.04.045 Cozzi, G. et al. (2017). A SiPM-based Detection Module for 2 inches LaBr3:Ce Readout for Nuclear Physics Application. IEEE NSS/MIC Symposium, Atlanta, USA , 1–3doi:10.1109/NSSMIC.2017. 8532888 Dinu, M. et al. (2010). Temperature and bias voltage dependence of the MPPC detectors. IEEE NSS-MIC Symposium , Piscateway, NJ doi:10...
arXiv 2017
-
[99]
doi:10.3390/condmat8040099 Bouchet, M. et al. (2007). SPIROC ( SiPM Integrated Read-Out Chip). IEEE NSS-MIC , 1857–1860doi:10. 1109/NSSMIC.2007.4436519 Brice, J. (1986). CRYSTAL GROWTH PROCESSES(Glasgow: Blackie Halsted Press) Burr, K. and Wang, G. (2007). Scintillation detection using 3 mm× 3 mm silicon photomultipliers. IEEE Trans. Nucl. Sci. Conf. Reco...
-
[185]
doi:10.1140/epja/s10050-020-00195-9 Pohl, R. (1938). Zusammen fassender bericht uber elektronenleitung und photochemische vorgang in alkalihalogeneid kristallen. Zeit. Physik 39, 36 Pohl, R. et al. (2010). The size of the proton. Nature 466, 213–216. doi:10.1038/nature09250 Poleshchuck, O. et al. (2021). Performances tests of a LaBr3:Ce detector coupled t...
arXiv 1938
-
[317]
doi:10.1016/j.nima.2004.08.034 Moszynski, M. et al. (2002). Intrinsic energy resolution of NaI (Tl). Nucl. Instr. Meth. A484, 259–269. doi:10.1016/S0168-9002(01)01964-7 Moszynski, M. et al. (2008). A comparative study with PMTs, Avalanche Photodiodes, Photodiodes and PIN photodiodes in Gamma Spectroscopy with LaBr3 Crystals . IEEE Trans. Nucl. Sci. Berkel...
arXiv 2002
-
[365]
doi:10.1016/s0168-9002(01)00694-5 McConnel, M. et al. (2000). Three-dimensional imaging and detection efficiency performance of orthogonal CZT strip detectors. Proceedings of SPIE 4141, 157–167. doi:10.1117/12.407576 Moon, R. (1948). Inorganic crystals for the detection of high energy particles and quanta. Phys. ReV .73, 1210 Moses, W. (2007). Recent advan...
-
[596]
doi:10.1016/j.nima.2013.08.005 Roentgen, W. (1896). On a new kind of rays. Science 3 (59), 227–230 Frontiers 24 M. Bonesini Gamma-ray and high-energy X-ray detection with scintillating crystals Roy, U. et al. (2019). Evaluation of cdzntese as a high-quality gamma-ray spectroscopic material with better compositional homogeneity and reduced defects. Scienti...
Show all 11 references
-
[988]
doi:10.1109/tns.2012.2233497 Zambelli, N. et al. (2020). CdZnTe-based X-ray spectrometer for Absolute Density Determination. IEEE Trans. Nucl. Sci. 67, 2273–2277. doi:10.1109/tns.2020.2996272 Zemach, A. (1956). Proton structure and the hyperfine shift in hydrogen. Phys. ReV .10...
2020
-
[1575]
doi:10.1063/1.1385342 Vita, D. D. et al. (2022). A 144-SiPM 3 inches LaBr3 readout module for PMTs replacement in Gamma Spectroscopy. Nucl. Instr. Meth. A1040, 167179. doi:10.1016/j.nima.2022.167179 Winston, R. (1970). Light Collection within the Framework of Geometric Optics....
2022
-
[7307]
doi:doi: 10.1038/s41598-019-43778-3 Scafe, R. et al. (2007). Si-APD readout for LaBr3:Ce scintillator. Nucl. Instr. Meth. A571, 355–357. doi:10.1016/j.nima.2006.10.108 Shah, K. et al. (2002). LaBr 3:Ce scintillators for gamma ray spectroscopy. LBNL-51793 Shim, H. S. et al. (20...
2007
Reviewed August 15, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.