{"id":"1dd915dd-0c60-4fe9-88ca-def572d3767e","arxiv_id":"2502.02423","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"After five years of operation in the CTAO LST-1 camera, removed PMTs showed no increase in afterpulsing, while stored spare tubes worsened, indicating operation offsets helium-driven aging.","lead":"After five years on the CTAO's first large telescope, the removed PMTs showed no increase in their afterpulsing rate, while similar stored tubes did. This suggests that normal camera operation, high voltage plus night-sky light, slowly cleans residual gas out of the tubes and offsets helium infiltration, keeping the telescope's low-energy sensitivity stable.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Used-vs-Spare comparison conflates operation with helium exposure: Spare tubes were stored ~9 years (undisclosed conditions, likely sea level) while Used tubes spent ~5 years at 2200 m; the lower helium partial pressure and shorter exposure alone could explain the absence of an increase, without…","rationale":"The reader's weakest_assumption correctly identifies the unrepresentative selection and the storage-history difference between Used and Spare tubes as the main threat to the conclusion. I agree with that concern but would sharpen it: even if the ten Used tubes were randomly removed, the Spare group is not a matched control for helium aging because the two groups differ in storage duration, storage environment, and operating altitude. At LST-1's ~2200 m site the external helium partial pressure is about 78% of sea level, so the Used tubes' integrated helium exposure over five years is roughly 43% of what the Spare tubes experienced over nine years at sea level. The paper does not quantify helium exposure at all, so the observed Spare-increase/Used-no-increase pattern cannot be uniquely attributed to operation-driven cleaning. This is a gap in causal attribution, not a refutation: the empirical result that ten used PMTs remain below the 2×10^-4 specification after five years is still useful. The claim of an operation-driven decrease in afterpulsing, however, would need either a matched control group or an exposure-corrected analysis. This is consistent with the reader's CONDITIONAL verdict, so I recommend no change to the verdict. The paper's measurements, arrival-time peak identification, and Gaussian fits are valuable; the concern is only about the control-group logic underlying the equilibrium interpretation.","tokens_in":5626,"tokens_out":5063,"duration_ms":54079,"concrete_test":"Obtain the storage locations and atmospheric-pressure history for the twelve Spare PMTs, plus the manufacturing batch/serial numbers and total high-voltage-on hours for all twenty-two tubes. Compute each tube's integrated helium exposure (partial pressure × time, with temperature scaling if available) and compare the 2014/15-to-2023 afterpulsing-rate change per unit exposure for the Used and Spare samples. If the per-exposure change for the Used tubes is not significantly below the Spare trend, the operation-driven self-cleaning effect is not demonstrated by this dataset.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central inference in Section 4 — that the absence of an afterpulsing increase in the ten Used PMTs demonstrates an operation-driven cleaning effect that offsets helium infiltration — depends on the Spare PMTs being an adequate control for helium aging. That control is not established. The Spare tubes sat unused since production in 2013 and were measured again after roughly nine years; the Used tubes were installed at LST-1 (La Palma, ~2200 m) and measured after roughly five years. Helium ingress into a PMT scales with external helium partial pressure and integration time. The Spare storage conditions are not reported, but if they were stored near sea level, their integrated helium exposure was approximately 9 × 1.0, while the Used tubes experienced approximately 5 × 0.78 (barometric pressure at 2200 m), i.e., roughly 43% of the Spare exposure. The paper does not correct for this difference, nor does it report the manufacturing batches of the two samples or how the ten Used tubes were selected for removal. Consequently, the observed contrast — Spare increased (second peak by a factor of 4.4), Used slightly decreased — is quantitatively consistent with the Used tubes having received a much lower helium dose, and the 'equilibrium' conclusion in Section 4 is underdetermined. A cleaner statement would be that the Used PMTs remain below specification after five years, without attributing the steady value specifically to operation-driven removal.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports afterpulsing-rate measurements of ten photomultiplier tubes removed from the CTAO LST-1 telescope after about five years of operation, compared with twelve unused spare PMTs and with 2014-2015 quality-control data. The spare PMTs show an increased afterpulsing rate, concentrated in the second arrival-time peak attributed to He+, while the used PMTs show no increase and remain below the 2e-4 specification. The authors interpret this as evidence that operation at high voltage with light exposure removes residual gas molecules, offsetting helium infiltration and keeping the afterpulsing rate stable over the telescope's lifetime.","tokens_in":5887,"tokens_out":6822,"duration_ms":63200,"significance":"If established, the result is important for CTAO operations and for the general understanding of PMT aging in IACTs: it provides direct field evidence on the balance between helium ingress and operation-induced cleaning, and it suggests that the stringent afterpulsing specification can be maintained over years. The paper's strengths are the direct measurement on PMTs from an operating telescope, the use of the original QC baseline, and the arrival-time decomposition that links the spare-tube increase to a specific ion species. However, the central attribution of the used-spare contrast to an operation-driven removal effect is not yet quantitatively supported, because the two samples differ in helium exposure, batch, and selection, and because no controlled high-voltage/light-aging experiment is presented.","major_comments":[{"comment":"The comparison that drives the conclusion is confounded by helium exposure. The spare tubes were produced in 2013 and measured after roughly nine years of storage, while the used tubes were installed around first light in December 2018 and measured after about five years at 2200 m altitude. Helium ingress scales with external partial pressure and time, yet the paper does not report the storage conditions of the spares, the manufacturing batches of either sample, or how the ten used PMTs were selected for removal. If the spares were stored near sea level, their integrated helium exposure would be approximately 9 yr x 1.0 atm versus 5 yr x 0.78 atm for the used tubes, roughly a factor of two higher. The observed contrast (factor-of-4.4 increase in the second peak for spares, no increase for used tubes) is therefore quantitatively consistent with a lower helium dose alone, and the equilibrium claim in Section 4 is underdetermined. Please report the relevant exposure parameters and selection criteria, or limit the conclusion to the observation that the used PMTs remain below specification.","section":"§2 and §4"},{"comment":"The central claim that the used PMTs show \"no increase\" rests on a visual comparison without error bars or a statistical test. The histograms overlap and the scatter plot has no uncertainty per point; the statement that the used rate \"slightly decreased\" has no confidence interval. With n=10, a paired comparison of the 2014-2015 and 2023 values should be reported with a confidence interval on the mean change, and a predefined equivalence bound or significance test should justify the phrase \"no increase.\" Without this, the reader cannot distinguish a real null result from a small sample that lacks power.","section":"§3.1 and Figure 1"},{"comment":"The mechanism that operation at high voltage with light exposure removes residual molecules is asserted but not demonstrated in this paper. No controlled experiment is shown in which a PMT is operated under high voltage and light and its afterpulsing rate is observed to decrease; the only evidence is the spare-used contrast, which is confounded as discussed above. The statement that \"residual molecules are removed from the vacuum during regular observations\" goes beyond what the data can show, and the open question about where the molecules go (photocathode trapping) is speculative. Moreover, a single before/after pair cannot establish that an equilibrium has been reached; at most it bounds the net change. Either present a controlled HV and light aging measurement or weaken the conclusion to \"the used PMTs did not increase above specification\" and present the removal mechanism as a hypothesis.","section":"§4 and Introduction"}],"minor_comments":[{"comment":"The color coding is inconsistent between the histograms (Spare red solid, Used blue dashed) and the scatter plot (Used red circles, Spare blue squares), which makes the figure hard to read.","section":"Figure 1"},{"comment":"The timeline should be stated unambiguously: the QC data are from 2014-2015, the used PMTs were installed around first light in December 2018, and the removal was in October 2023. Please clarify whether the \"before\" values of the used PMTs were taken at production QC or at installation, and whether the pairing in Figure 1 is per serial number.","section":"§2"},{"comment":"The Gaussian fits are performed only on the Spare-after distribution; the statements that the first peak increased by about 8% and the second by a factor of 4.4 should specify that these are based on binned rates rather than on fitted peak amplitudes, and should give uncertainties for those factors.","section":"§3.2"},{"comment":"Typos should be corrected, including \"eailer\" for \"earlier\" in Section 4, \"PMT sampleset\" and \"abscissaisinthelogarithmicscale\" in the Figure 2 caption, and \"IfNN\" for \"INFN\" in the acknowledgments.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should read this as a solid empirical datapoint with an interpretation that outruns the data. The authors pulled ten PMTs out of LST-1 after five years on the sky, measured their afterpulsing in the lab, and compared them to twelve spares that sat on a shelf. The used tubes are still below the 2e-4 spec; the spares have drifted up, with a helium-associated peak growing by a factor of 4.4. As far as I know this is the first public check of afterpulsing on field-returned IACT PMTs, and the result matters for the LST cameras' energy threshold over a 20-year lifetime.\n\nThe measurement itself is straightforward and the paper is honest about its own limits — it flags the mechanism as unclear and the ion-capture explanation as a hypothesis. That is good practice.\n\nThe soft spots are statistical and one control issue. The scatter plot has no error bars, no significance test backs the \"no increase\" claim, and the selection of the removed PMTs is not described. Ten tubes is a small sample. More importantly, the Spare sample is not a clean helium-aging control: the spares were stored roughly nine years, presumably near sea level, while the used tubes spent five years at 2200 m. The helium dose is not the same, and the paper does not correct for exposure or report storage conditions. The stress-test note overreaches, though, when it says the exposure difference alone could explain the absence of an increase: the used tubes actually decreased slightly, and the dose difference is only a factor of about two, not enough to erase a 4.4x growth. But the comparison is still confounded enough that the equilibrium interpretation is underdetermined. A careful referee should ask for the storage history, a helium-dose argument, and a significance test.\n\nWho gets value? Anyone working on vacuum photodetector aging, especially for Cherenkov telescopes. It is a single-telescope study, so within-field impact is modest, but it is the kind of real-operations check that only comes along rarely.\n\nI would send it to review. The measurement deserves to be in the literature with the caveats made explicit. If I were editing, I'd ask for a revision that tightens the statistics and the control discussion, then accept.","headline":"Useful first hard look at PMT afterpulsing aging in a working Cherenkov camera, with a plausible but statistically underpowered self-cleaning story.","tokens_in":6523,"tokens_out":4210,"would_cite":true,"duration_ms":40095,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Operated photomultiplier tubes did not accumulate afterpulsing over five years of LST-1 observing; the paper attributes this to a self-cleaning effect that balances helium infiltration.","keywords":["afterpulsing","photomultiplier tube","imaging atmospheric Cherenkov telescope","helium infiltration","ion feedback","self-cleaning","LST-1","night-sky background"],"falsifier":"Track afterpulsing rates of a larger, randomly selected set of PMTs measured before installation and again after another several years of operation; if the operated distribution shifts upward or approaches the 2e-4 limit, the equilibrium claim is wrong.","tokens_in":5412,"feed_emoji":"🔭","tokens_out":5560,"duration_ms":51322,"temperature":0.7,"pith_summary":"The paper tests whether the afterpulsing rate of PMTs in the first Large-Sized Telescope grew over five years of operation, as helium infiltration into the vacuum would predict. It removed ten operated tubes and twelve stored spares, measured them in the lab, and compared the results with pre-installation quality-control data. The operated tubes did not increase; they slightly decreased and remained below the 2e-4 specification, while the stored spares increased, with a helium-associated afterpulse peak growing 4.4-fold. The paper's central claim is that normal night-sky observing, which supplies high voltage and constant light exposure, removes residual gas from the tube and offsets the helium-driven increase, keeping long-term performance stable.","feed_headline":"After five years, telescope PMTs show no afterpulse rise","feed_subtitle":"Stored spare tubes worsened while operated tubes improved, pointing to a self-cleaning effect under night-sky light.","key_machinery":"The central comparison is between Used PMTs removed from the telescope after five years of operation and Spare PMTs stored since 2013. The measurement setup is a dark box illuminated by fast light pulses of 800–920 ps width corresponding to roughly 50 photoelectrons, with high voltage set for a gain of 40,000 and afterpulses counted in a 0–3 microsecond window using a charge threshold of 4 photoelectron equivalents. The proposed mechanism is ion-feedback self-cleaning: accelerated electrons ionize residual gas inside the tube, and some ions are captured by the photocathode, thereby reducing the gas content during operation. The afterpulse arrival-time distributions, with peaks near 167 ns and 387 ns, identify the ionic species and let the authors attribute the spare-tube growth to helium.","core_discovery":"The paper reports that after five years of regular operation in LST-1, the removed PMTs showed afterpulsing rates slightly lower than their pre-installation values and still below the 2e-4 quality criterion. In contrast, spare PMTs from the same production era that were merely stored showed increased afterpulsing rates, most visibly in a late arrival-time peak at about 387 ns that the authors attribute to He+ and which grew by a factor of 4.4. The discovery is that the operating environment itself cleans the tube: the ion-feedback process that creates afterpulses also ionizes residual gas molecules, and some of those ions are captured by the photocathode and removed from the vacuum. Over years of observation, this self-cleaning keeps the net afterpulsing rate roughly constant.","pith_inferences":["Editorial inference: If the self-cleaning effect can be reproduced on the bench, stored spare tubes could be rejuvenated by operating them under high voltage and illumination before installation, potentially extending the usable lifetime of PMT stockpiles.","Editorial inference: The equilibrium interpretation depends on regular operation; extended downtime, bright-moon shutdowns, or periods of low night-sky light could let helium infiltration dominate and push afterpulsing upward.","Editorial inference: The arrival-time peak positions offer a per-tube helium monitor; comparing the 167 ns and 387 ns peak heights in future removals would test whether the helium equilibrium shifts as tubes age beyond five years."],"forward_implications":["After five years of regular observing, the ten removed PMTs had afterpulsing rates below the 2e-4 specification, so the telescope's false-trigger background from afterpulsing has not worsened.","Stored spare tubes showed rising afterpulsing, with a helium-associated peak growing 4.4-fold, confirming that helium infiltration is a real long-term aging process when tubes are not operated.","The operational environment of an IACT, high voltage plus night-sky light, appears to remove residual gas from the tube and offset helium entry.","The equilibrium between helium infiltration and self-cleaning implies that LST-1's PMT performance can be maintained over the telescope's intended operating lifetime under similar observing conditions."],"supporting_citations":[{"why":"Supplies the quality-control data, the afterpulsing specification, and the measurement setup reused for the 2023 re-measurements.","marker":"[9]"},{"why":"Establishes the ion-feedback mechanism by which gas molecules inside the tube produce afterpulses.","marker":"[10]"},{"why":"Demonstrates that helium contamination raises afterpulsing rates in photomultiplier tubes, the aging mechanism the Used tubes are expected to show.","marker":"[11]"},{"why":"Provides the ionization cross-section ratio used to identify the 387 ns afterpulse peak as He+.","marker":"[13]"},{"why":"Documents the development of the low-afterpulsing PMT candidates that define the baseline performance class.","marker":"[5]"},{"why":"Reports the evaluation of the novel PMTs selected for the Cherenkov telescopes, supporting the claimed initial afterpulsing performance.","marker":"[6]"}],"fun_headline_variants":["PMTs self-clean under night-sky light, halting afterpulse drift","Used PMTs improve, stored degrade: self-cleaning effect","Self-cleaning PMTs offset helium afterpulse growth","Operating PMTs self-clean, preserving afterpulse rate","Stored spare PMTs degrade; used ones self-clean"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The result depends on the ten removed PMTs being a fair sample of the full camera and on the only meaningful difference between Used and Spare tubes being operating history rather than storage history, selection, or manufacturing batch.","fun_headline_variants_meta":{"raw":{"variants":["PMTs self-clean under night-sky light, halting afterpulse drift","Used PMTs improve, stored degrade: self-cleaning effect","Self-cleaning PMTs offset helium afterpulse growth","Operating PMTs self-clean, preserving afterpulse rate","Stored spare PMTs degrade; used ones self-clean"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001153,"raw_usage":{"total_tokens":4773,"prompt_tokens":933,"completion_tokens":3840,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":549,"completion_tokens_details":{"reasoning_tokens":3750}},"tokens_in":549,"tokens_out":3840,"duration_ms":27522,"temperature":1.0,"reasoning_tokens":3750,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T12:11:27.756510+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Track afterpulsing rates of a larger, randomly selected set of PMTs measured before installation and again after another several years of operation; if the operated distribution shifts upward or approaches the 2e-4 limit, the equilibrium claim is wrong.","supporting_citations":[{"cited_title":"Saito, M","cited_arxiv_id":null,"evidence_quote":"Supplies the quality-control data, the afterpulsing specification, and the measurement setup reused for the 2023 re-measurements."},{"cited_title":"Coates,The origins of afterpulses in photomultipliers,Journal of Physics D: Applied Physics6 (1973) 1159","cited_arxiv_id":null,"evidence_quote":"Establishes the ion-feedback mechanism by which gas molecules inside the tube produce afterpulses."},{"cited_title":"Bartlett, A.L","cited_arxiv_id":null,"evidence_quote":"Demonstrates that helium contamination raises afterpulsing rates in photomultiplier tubes, the aging mechanism the Used tubes are expected to show."},{"cited_title":"Shah, D.S","cited_arxiv_id":null,"evidence_quote":"Provides the ionization cross-section ratio used to identify the 387 ns afterpulse peak as He+."},{"cited_title":"Mirzoyan, D","cited_arxiv_id":null,"evidence_quote":"Documents the development of the low-afterpulsing PMT candidates that define the baseline performance class."},{"cited_title":"Mirzoyan, D","cited_arxiv_id":null,"evidence_quote":"Reports the evaluation of the novel PMTs selected for the Cherenkov telescopes, supporting the claimed initial afterpulsing performance."}],"review_version":1}