{"id":"cb6c8ef9-9dbb-4fc6-bdd4-539769bad1d7","arxiv_id":"2511.08898","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Aluminum KIDs with a-Si:H parallel-plate capacitors are dominated by GR/photon noise down to ~0.1 Hz, with TLS noise upper limits consistent with prior high-frequency data.","lead":"This paper measures low-frequency noise in superconducting light detectors that use a hydrogenated amorphous silicon capacitor, finding they stay quiet down to about 0.1 Hz. The result validates a compact detector design for the new NEW-MUSIC submillimeter camera, which needs slow, stable readout.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Central photon-noise-limited claim depends on unmeasured low-frequency optical-load noise; the off-resonance monitor tone that could remove correlated noise was not commissioned.","rationale":"The reader's weakest_assumption correctly identifies the missing direct low-frequency optical-load measurement and the extrapolation of >100 Hz data. My stress-test confirms that this is the most load-bearing concern: the paper's central claim—that the design will be photon-noise-limited at ~0.1 Hz under observing conditions—is a prediction based on scaling and extrapolation, not a demonstrated result. The authors themselves acknowledge the absence of the off-resonance monitor tone and the possibility that the low-frequency dark rise is correlated electronics, not TLS. Therefore, the verdict of CONDITIONAL is appropriate: the claim is plausible and the analysis is conservative, but the condition of a direct low-frequency optical-load measurement is required to 'establish' it. No additional concern rises to the level of changing the verdict. The reader and I agree that the load-bearing weakness is the same.","tokens_in":5118,"tokens_out":4925,"duration_ms":47600,"concrete_test":"Measure the optical-load noise PSD under T_load = 180 K using the two-tone setup with the off-resonance monitor tone enabled, from 0.01 Hz to 1 kHz. Compare the measured PSD at 0.1 Hz to the extrapolated GR+photon noise level shown in Fig. 4. If the measured 0.1-Hz PSD is within ~2x of the extrapolation, the photon-noise-limited claim is supported; if it is significantly higher, the claim must be revised.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The abstract's central claim — that the MS-PPC-LEKID design 'will be photon-noise-limited' down to ~0.1 Hz under observing conditions — rests on an extrapolation, not a measurement. In Sec. IV, the optical-load noise is 'only been reliably measured above 100 Hz' (reported in [1], without the two-tone setup), and is extrapolated down to 0.01 Hz in Fig. 4. The two-tone system's off-resonance monitor tone, which was designed to remove correlated multiplicative low-frequency electronics noise, was not commissioned (footnote 1). The dark data's low-frequency rise is attributed by the authors to 'correlated electronics noise' rather than TLS, so the TLS upper limits derived from dark data are approximate and may overestimate TLS. Under optical load, the actual low-frequency PSD could be higher than the extrapolated GR+photon level if any additional noise source (e.g., correlated readout noise, optical or thermal drift, or an unexpected load-dependent component) appears. The body text uses 'likely' and 'conservatively concluded,' but the abstract says 'establish,' overstating the evidence. The central claim is therefore not yet empirically anchored at the frequencies relevant to the stated application.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports low-frequency noise measurements of Al/a-Si:H microstrip-coupled parallel-plate-capacitor lumped-element kinetic inductance detectors (MS-PPC-LEKIDs) developed for the NEW-MUSIC camera. Under dark conditions, the measured PSDs are claimed to be generation-recombination (GR) noise dominated down to 0.1 Hz. Using scaling relations based on prior TLS measurements of a-Si:H PPCs [7], the authors derive approximate upper limits on TLS noise (3.5x the broken power-law model and 56x the f^-0.5 extrapolation) from the dark PSDs. Under optical load (T=180 K), the GR+photon noise measured above 100 Hz in earlier work is extrapolated to low frequency and compared with scaled TLS-noise curves; the authors conservatively conclude that photon noise will dominate down to tenths of a Hz and possibly lower. The abstract states that these results 'establish' that the design will be photon-noise-limited under observing conditions.","tokens_in":1453,"tokens_out":1533,"duration_ms":76484,"significance":"If the conclusions hold, this work provides an important validation of a-Si:H parallel-plate capacitors as a viable low-TLS-noise technology for KIDs in astronomy, where low-frequency stability is required. The paper's strengths include the use of an independent TLS prediction from prior published measurements (no free parameter is tuned to force agreement), conservative upper limits derived from dark data, and a two-tone readout system designed to address correlated electronics noise even though the monitor tone was not fully commissioned. The results are useful for the NEW-MUSIC detector development and for the broader KID community considering a-Si:H PPCs. However, the central quantitative claim rests on an extrapolation rather than a direct low-frequency optical-load measurement, and the manuscript itself contains a contradiction in the TLS scaling description that needs resolution.","major_comments":[{"comment":"The abstract states that the results 'establish that our MS-PPC-LEKID design for NEW-MUSIC will be photon-noise-limited,' but the body text uses 'likely' and 'conservatively concluded.' Under optical load, noise is only reliably measured above 100 Hz (from [1], without the two-tone setup) and is extrapolated to 0.01 Hz in Fig. 4. This is a projection, not a direct measurement, and any new low-frequency noise source under optical load would invalidate it. The abstract and conclusion should be reworded to match the evidence, e.g., 'consistent with being photon-noise-limited' or 'likely photon-noise-limited.'","section":"Abstract; Sec. IV, Fig. 4"},{"comment":"The TLS upper limits are described as 'approximate fits' and 'approximate upper limits' of 3.5x and 56x, but no fitting procedure, confidence level, or uncertainty on the PSD estimates is provided. These factors are load-bearing for the conclusion that TLS will not dominate at low frequency. Please give a reproducible fitting method and state statistical uncertainties or explicitly label these as order-of-magnitude estimates.","section":"Sec. IV, Fig. 3 caption"},{"comment":"There is a contradiction in the TLS scaling. Eq. (1) gives S_TLS ∝ E/E_r, so if the stored power (and hence the electric field E) decreases under optical load, S_TLS should decrease. However, the Fig. 4 caption states that the scaled TLS model curves 'increase because the stored power decreases.' This inconsistency affects the quantitative comparison and the conclusion. Please correct either the equation or the text and justify the scaling direction.","section":"Sec. III, Eq. (1); Sec. IV, Fig. 4 caption"},{"comment":"The off-resonance monitor tone intended to remove correlated multiplicative electronics noise was not commissioned. The authors note that the low-frequency rise in dark data may be correlated electronics rather than TLS. This makes the TLS upper limits conservative, but it also means there is no direct low-frequency measurement under optical load. The manuscript should present the photon-noise-limited conclusion as a testable prediction, not an established result, and should emphasize this limitation in the abstract or conclusion.","section":"Sec. II, footnote 1; Sec. IV"}],"minor_comments":[{"comment":"Typo: 'provide access to a a wide range' should be 'to a wide range.'","section":"Sec. I, Introduction"},{"comment":"Typo: 'dependence of responsibility and photon noise' should likely be 'dependence of responsivity and photon noise.'","section":"Sec. IV, Optical load paragraph"},{"comment":"Duplicate word: 'If the prediction from from [7]' should be 'from [7].'","section":"Sec. IV, last paragraph"},{"comment":"The acronym expansion 'MUltiband' is unconventional; please use 'Multiband' or provide the official expansion.","section":"Abstract"},{"comment":"The color names 'mustard' and 'yellow-green' are difficult to identify; use standard color names or line styles in the caption and text.","section":"Fig. 3 caption"},{"comment":"The parameters a, b, c, d of the broken power law in Eq. (2) are not defined here; please provide the values or refer explicitly to [7] with a short summary, since the scaling and extrapolation rely on them.","section":"Sec. III, Eq. (1)"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is within the scope of IEEE TAS and presents useful device characterization. The main concern is that the abstract overstates what is a conservative extrapolation; the authors themselves are appropriately cautious in the body. The contradiction between Eq. (1) and the Fig. 4 caption regarding TLS scaling with stored power must be resolved, and the TLS upper limits need a firmer statistical basis. These are fixable but require more than a light edit."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe short version: this is the first low-frequency noise characterization of aluminum microstrip-coupled KIDs with a-Si:H parallel-plate capacitors, and it mostly does what it says. It deserves a serious referee, though the abstract oversells the optical-load claim.\n\nWhat's new: they push TLS noise measurements of a-Si:H down to 0.1 Hz, two orders of magnitude below prior work, and show dark GR noise dominates down to that frequency. The two-tone setup is sensible, the upper limits on TLS (3.5x and 56x the scaled models) are honestly approximate, and they explicitly flag that the low-frequency rise in dark PSDs may be correlated electronics, not TLS, and that the off-resonance monitor tone was not yet commissioned. That is the right kind of candor.\n\nThe soft spot is the central application claim. Under optical load, the noise was only reliably measured above 100 Hz (from prior work) and is extrapolated down to 0.01 Hz. The body text says \"likely\" and \"conservatively concluded,\" but the abstract says \"establish.\" Those are different epistemic claims. The TLS scaling itself is reasonable but rests on a broken power-law fit from prior work with no parameter uncertainties carried through. The conclusion would be firmer with a direct low-frequency optical-load PSD, or at least a sentence in the abstract acknowledging that low-frequency performance under load is extrapolated.\n\nFairness check: the extrapolation is not reckless. The scaling is conservative, the crossing-point analysis includes the pessimistic broken-power-law case, and if the low-frequency rise is indeed electronics, the TLS limit improves. The paper is transparent about what is measured and what is scaled.\n\nBottom line: for the KID detector community this is a useful, incremental validation of a-Si:H PPCs for slow-scan instruments. It is not a new physics result, but it is a clean instrumentation characterization with honest error accounting. I'd send it to review, and ask the authors to align the abstract with the body's hedged language or, better, add the missing low-frequency optical-load measurement.","headline":"Useful first low-frequency noise data on a-Si:H PPC KIDs, but the abstract's 'establish' oversells an extrapolated optical-load claim.","tokens_in":5940,"tokens_out":1561,"would_cite":true,"duration_ms":15458,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Kinetic inductance detectors with hydrogenated amorphous silicon capacitors stay photon-noise-limited down to about 0.1 Hz, despite the dielectric's potential two-level-system noise.","keywords":["kinetic inductance detectors","two-level-system noise","hydrogenated amorphous silicon","parallel-plate capacitors","low-frequency noise","generation-recombination noise","submillimeter astronomy","NEW-MUSIC"],"falsifier":"Measure the noise PSD of the same detectors under an 180 K optical load down to 0.01 Hz using the two-tone setup with the off-resonance monitor tone enabled; if the PSD below 0.1 Hz rises above the extrapolated GR+photon level (the blue dashed line in Fig. 4) by more than a factor consistent with the dark low-frequency excess, the photon-noise-limited claim fails.","tokens_in":5030,"feed_emoji":"🔭","tokens_out":10336,"duration_ms":87379,"temperature":0.7,"pith_summary":"This paper establishes that a detector architecture—kinetic inductance detectors (KIDs) that use hydrogenated amorphous silicon (a-Si:H) parallel-plate capacitors—does not suffer from the low-frequency two-level-system (TLS) noise that such dielectric layers can introduce. Under dark conditions, the detectors are generation-recombination noise dominated down to 0.1 Hz; under optical load, they are likely dominated by generation-recombination plus photon noise down to tenths of a hertz or lower. This matters because the compact, microstrip-coupled design is needed for a planned six-band submillimeter camera, and the result means the detectors can be read out at the slow modulation rates used in astronomical scanning. The paper also sets upper limits on TLS noise in the a-Si:H material at sub-hertz frequencies that are consistent with extrapolations from higher-frequency measurements.","feed_headline":"Down to 0.1 Hz, a-Si:H KIDs stay photon-noise-limited","feed_subtitle":"Compact detector design for a new submillimeter camera avoids low-frequency dielectric noise, enabling slow sky scans.","key_machinery":"Two elements carry the argument. First, a two-tone IQ readout that places an on-resonance probe tone and an off-resonance monitor tone at baseband, which removes correlated multiplicative electronics noise (when the off-resonance tone is commissioned) and avoids additive amplifier 1/f noise, allowing noise PSDs to be measured down to 0.1 Hz. Second, a TLS noise scaling relation, S_TLS = S_r^TLS (T/246mK)^{-0.8} (A/A_r)(E/E_r), which extrapolates prior higher-frequency TLS measurements of a-Si:H PPCs to the test detector's area, temperature, and stored electric field, enabling direct comparison of expected TLS noise with the observed dark and optical-load spectra.","core_discovery":"The central claim is that aluminum microstrip-coupled, parallel-plate-capacitor, lumped-element kinetic inductance detectors (Al/a-Si:H MS-PPC-LEKIDs) are generation-recombination noise dominated down to 0.1 Hz under dark conditions and likely dominated by generation-recombination plus photon noise down to tenths of a hertz or lower under optical load. This holds despite the a-Si:H parallel-plate capacitors, a potential source of two-level-system noise. By comparing measured noise power spectral densities to TLS noise scaled from prior 0.1–10 kHz measurements using a power-law extrapolation, the authors obtain upper limits on low-frequency TLS noise (3.5 times the broken power-law model and","pith_inferences":["If the low-frequency rise in the dark PSD is, as the authors suspect, correlated electronics noise rather than TLS, the actual TLS noise could lie far below their upper limits; the photon-noise-limited regime could then extend to 0.01 Hz or lower, a testable prediction once the off-resonance tone is fully commissioned.","The two-tone IQ readout's noise-removal scheme is not KID-specific; applying it to other superconducting detector readouts (e.g., transition-edge sensors or microwave SQUID multiplexers) could push their low-frequency noise characterization to millihertz levels.","The success with a-Si:H suggests other low-TLS-loss deposited dielectrics (such as silicon nitride or silicon dioxide) might be substituted into PPC-KIDs, provided they pass the same low-frequency noise test.","A direct extension would be to measure the same detectors with the off-resonance tone active under dark and 180 K loads down to 0.01 Hz; a flat PSD at the extrapolated GR+photon level would confirm the photon-noise-limited claim without relying on extrapolation."],"forward_implications":["NEW-MUSIC's detectors will be photon-noise-limited under on-sky loads of 40–150 K, supporting slow-scan (0.5 deg/s) observations at angular scales of 1–5 degrees.","a-Si:H PPCs are validated as a viable dielectric for KIDs in low-modulation-rate applications, not only in the NEW-MUSIC design.","The upper limits on TLS noise (3.5x broken power law, 56x f^−0.5) quantify the worst-case low-frequency noise contribution of the dielectric, bounding detector performance.","The two-tone IQ measurement approach can be reused for sub-0.1 Hz noise characterization of other KID or resonator architectures.","The crossing point where TLS noise would overtake photon noise is conservatively 0.1–0.2 Hz, leaving a comfortable margin for the 1–5 degree scan requirement."],"fun_headline_variants":["a-Si:H KIDs stay quiet to 0.1 Hz, enabling slow sky scans","Detector design dodges TLS noise down to 0.1 Hz for new camera","Al KIDs with a-Si:H caps: noise dominated by GR to 0.1 Hz","Low-frequency noise no barrier for a-Si:H KID detectors","New KIDs keep noise low to 0.1 Hz despite a-Si:H dielectrics"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The photon-noise-limited conclusion rests on extrapolating the noise level measured above 100 Hz under a 180 K optical load unchanged down to 0.01 Hz, assuming no new low-frequency noise source—TLS or electronics—appears under optical load.","fun_headline_variants_meta":{"raw":{"variants":["a-Si:H KIDs stay quiet to 0.1 Hz, enabling slow sky scans","Detector design dodges TLS noise down to 0.1 Hz for new camera","Al KIDs with a-Si:H caps: noise dominated by GR to 0.1 Hz","Low-frequency noise no barrier for a-Si:H KID detectors","New KIDs keep noise low to 0.1 Hz despite a-Si:H dielectrics"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00049,"raw_usage":{"total_tokens":2257,"prompt_tokens":761,"completion_tokens":1496,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":505,"completion_tokens_details":{"reasoning_tokens":1385}},"tokens_in":505,"tokens_out":1496,"duration_ms":12412,"temperature":1.0,"reasoning_tokens":1385,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T22:42:37.885472+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the noise PSD of the same detectors under an 180 K optical load down to 0.01 Hz using the two-tone setup with the off-resonance monitor tone enabled; if the PSD below 0.1 Hz rises above the extrapolated GR+photon level (the blue dashed line in Fig. 4) by more than a factor consistent with the dark low-frequency excess, the photon-noise-limited claim fails.","supporting_citations":[],"review_version":1}