{"id":"864db722-d58d-4cda-b00e-5403515a7a9e","arxiv_id":"1908.06205","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Multimode-fiber-coupled superconducting nanowire detectors can reach about 80% system efficiency and 19.5 ps timing resolution, the best reported timing for such detectors.","lead":"This paper reports superconducting nanowire detectors coupled to multimode optical fibers that combine high system detection efficiency with fast timing. The best figures are about 80 percent efficiency at visible wavelengths and under 20 picoseconds timing resolution with graded-index multimode fibers.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'simultaneously achieved' >80% efficiency and <20 ps jitter are never demonstrated on the same detector, fiber, and wavelength; the 80% SDE is for a 25 µm visible detector and the 19.5 ps jitter is for a 20 µm detector at 1064 nm.","rationale":"The paper contains a real experimental advance: multimode-fiber-coupled SNSPDs with 19.5 ps jitter and useful system efficiencies, with polarization dependence supported by FDTD simulations and a polarization-insensitive fractal detector demonstrated. The central weakness, however, is the abstract's 'simultaneously achieved' claim, which requires both metrics on the same system. The body separates the two record values across different detectors and wavelengths, and the 80% figure is qualified as 'about 80%' without error bars, so the strict >80% inequality is not firmly established for any configuration. The reader's weakest assumption (electronic jitter not deconvolved) is related but distinct: if anything, including electronic jitter makes 19.5 ps an upper bound on detector-fiber jitter, so deconvolution would not weaken the <20 ps claim; the missing link is that the high-efficiency and low-jitter results are never measured together. This does not invalidate the individual results, but it means the headline claim should be conditional on a same-system demonstration or should be softened. Since the reader's verdict is already CONDITIONAL and the recommended condition is compatible with this concern, no change to the verdict is needed.","tokens_in":7506,"tokens_out":3706,"duration_ms":36839,"concrete_test":"Construct a single-system verification for the headline claim: take the 20 µm detector of Fig. 4(c) with the same graded-index multimode fiber and cryogenic amplifier, and measure system detection efficiency at 1064 nm; alternately, take the 25 µm detector of Fig. 3(a) with the 20 µm fiber and measure jitter at 516 nm using the same readout chain. If neither configuration yields SDE >80% and FWHM <20 ps, the abstract must be revised to report the two records separately (e.g., 'SDE >80% for a 25 µm visible detector and jitter <20 ps for a 20 µm detector at 1064 nm') rather than claiming simultaneous achievement.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The abstract's central claim is that the authors 'simultaneously achieved system efficiency >80% and time resolution <20 ps' for multimode-fiber-coupled SNSPDs. For this claim to hold, a single detector-fiber-wavelength configuration would need to show both SDE >80% and jitter <20 ps. The paper does not provide such a configuration. The SDE value supporting the >80% claim is 'about 80%' for a 25 µm diameter visible-wavelength detector coupled to a 20 µm fiber (Fig. 3(a), yellow curve at 516 nm), while the jitter value supporting the <20 ps claim is 19.5±0.2 ps for a 20 µm diameter detector measured at 1064 nm with graded-index multimode fiber and cryogenic amplifier (Fig. 4(c)). No jitter measurement is reported for the 25 µm detector at 516 nm, and no system efficiency measurement is reported for the 20 µm detector at 1064 nm. The 50 µm detector at 516 nm reaches only 70% SDE, and the 20 µm detector at 878 nm reaches 80% only under optimized TE polarization with single-mode fiber, dropping to 60% under multimode randomized illumination. Thus no datapoint in the paper demonstrates both high efficiency and low jitter in the same system. The reader's concern about electronic jitter contributions is valid, but even if the electronics contributed zero jitter, the simultaneous-achievement claim would still lack direct support. The conclusion sentence 'For visible wavelength, polarization dependence was negligible thus both single-mode and multi-mode coupled SNSPDs showed >80% system detection efficiency' likewise refers only to efficiency, not to jitter. This is a load-bearing gap because the paper's headline distinguishes it from earlier multimode-fiber SNSPD work that reported either high efficiency or moderate timing resolution, but not both together.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript reports the fabrication and characterization of NbTiN superconducting nanowire single-photon detectors (SNSPDs) with 20, 25, and 50 µm diameters coupled to multimode fibers, targeting visible, near-infrared, and telecom wavelengths. The authors measure system detection efficiency (SDE) under single-mode and deliberately randomized multimode illumination, observe wavelength-dependent TE/TM polarization sensitivity that increases toward telecom wavelengths, and compare the results with FDTD simulations. They also measure timing jitter at 1064 nm for a 20 µm detector, showing that graded-index fiber avoids the multi-peak instrument response obtained with step-index fiber and that a cryogenic amplifier reduces the FWHM to 19.5±0.2 ps. The abstract and conclusions claim simultaneous system efficiency >80% and time resolution <20 ps.","tokens_in":7818,"tokens_out":6213,"duration_ms":61807,"significance":"If fully supported, the results would be a useful engineering advance for applications such as quantum-dot photon collection, bio-optics, and lidar, where multimode fiber coupling relaxes alignment constraints. The paper's strengths are the detailed efficiency measurements under randomized multimode illumination, the FDTD comparison with no fitted parameters, and the explicit demonstration that graded-index fiber prevents modal-dispersion broadening of the instrument response. However, the headline 'simultaneously achieved' claim is not demonstrated by the reported data, and the jitter numbers are system-level values without a decomposed jitter budget. These issues are correctable with additional data or a qualified claim.","major_comments":[{"comment":"The claim 'We simultaneously achieved system efficiency >80% and time resolution <20 ps' is not supported by any single experimental configuration in the paper. The efficiency value supporting the claim is 'about 80%' for a 25 µm detector at 516 nm coupled to a 20 µm fiber (Fig. 3(a), yellow curve, Section 3), while the sub-20 ps jitter is 19.5±0.2 ps for a 20 µm detector at 1064 nm with graded-index multimode fiber and cryogenic amplifier (Fig. 4(c), Section 4). No jitter measurement is reported at 516 nm and no efficiency measurement is reported at 1064 nm for the same device. Section 5 repeats the 'at the same time' claim; the paper should either add a configuration that shows both properties or explicitly qualify the claim as a combination of results from different devices and wavelengths.","section":"Abstract and Section 5"},{"comment":"The quantitative basis for '>80%' is not established. The text states 'An SDE about 80% was achieved' for the 25 µm visible detector, yet the abstract states '>80%'; no uncertainty or error bars are given for the SDE curves. In addition, the conclusion 'both single-mode and multi-mode coupled SNSPDs showed >80% system detection efficiency' is contradicted by Fig. 3(a), where the 50 µm detector reaches 70% with both SM and MM fibers, and no SM-fiber data are shown for the 25 µm detector. The authors should present the actual saturation value with uncertainty and restrict the claim to the configurations measured.","section":"Section 3, Fig. 3(a)"},{"comment":"The jitter measurement does not separate the detector/fiber timing response from the measurement electronics. The setup includes a 4.2 ps pulsed laser, a 4 GHz oscilloscope, a fast photodiode reference, and rising-edge triggering (Section 4), and the reported 19.5±0.2 ps FWHM is a system-level instrument response function without deconvolution or an independent estimate of the electronic jitter. Because the claim 'best reported time resolution for multimode fiber coupled SNSPDs' is a comparative claim, the authors need to provide a jitter budget or explicitly state that all comparisons are system-level values obtained with the same electronics.","section":"Section 4"}],"minor_comments":[{"comment":"There are several typographical errors, including 'croystat' (Fig. 1 caption), 'sysytem' (Section 1), and 'nanowiwe' (Section 3); these should be corrected.","section":"Throughout"},{"comment":"The phrase 'Similar to [9], we fabricated SNSPDs' cites reference [9], which concerns quantum-dot microcavity sources and appears unrelated to SNSPD fabrication; a relevant NbTiN SNSPD fabrication reference, such as [14], would be more appropriate.","section":"Section 2, fabrication"},{"comment":"The randomization of modes and polarization in the U-bench is described qualitatively; a measurement of the output mode distribution or residual degree of polarization would strengthen the claim that the multimode illumination is randomized.","section":"Section 3, U-bench setup"},{"comment":"The fitting procedure for the jitter histograms could be described more fully; Fig. 4(c) and (d) show asymmetric responses and two-peak fits, but it is not stated whether the reported 19.5±0.2 ps FWHM comes from a single Gaussian fit or from the main peak of a two-peak fit.","section":"Section 4, fitting"}],"recommendation":"major_revision","confidential_remarks":"The engineering content is solid and the paper is likely publishable after revision, but the abstract and conclusions overclaim simultaneous high efficiency and low jitter on a single system. Please ask the authors to either provide a data point with both properties on the same detector and wavelength or substantially qualify the claim. The jitter budget should also be addressed carefully."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Let me cut to the point. The 19.5 ps jitter measured with a graded-index multimode fiber and cryogenic amplifier is a real step forward; earlier multimode-fiber SNSPDs were stuck in the 76–105 ps range. That alone makes the paper worth a careful read. The systematic polarization study under randomized multimode illumination is also genuinely new, and the TE/TM absorption ratios match FDTD simulation without fitting. The fractal detector’s near-zero polarization dependence is a useful addition.\n\nWhere it gets soft: the abstract says “simultaneously achieved system efficiency >80% and time resolution <20 ps.” The supporting data are an ~80% SDE at 516 nm on a 25 µm detector with a 20 µm fiber, and a 19.5 ps jitter at 1064 nm on a 20 µm detector. No single detector/wavelength shows both. The 20 µm detector’s SDE at 1064 nm isn’t reported, and the 25 µm detector’s jitter isn’t reported. So the “simultaneously” claim is an inference, not a measurement. I suspect the components could be combined to meet the claim, but the paper doesn’t show it. That’s load-bearing because the headline is what distinguishes this work from previous multimode results.\n\nThe jitter analysis also leaves one stone unturned. The 19.5 ps is the system IRF, but the paper never deconvolves the 4.2 ps laser pulse, the 4 GHz scope, or the amplifier chain. The SM-fiber/cryo-amp result of 18.5 ps shows the fiber isn’t the bottleneck, but without a separate electronic-jitter measurement you can’t know how much of the 19.5 ps is detector. That’s a moderate concern, not fatal; the comparison against step-index fiber (multi-peak, ~24 ps best) still makes the graded-index improvement credible.\n\nMinor: SDE values have no error bars, and the telecom multimode dark count from blackbody is flagged but not deeply discussed.\n\nThe citation pattern is fine. This is an experimental paper with real measurements and honest comparison to prior work. The main fix is to rewrite the abstract and conclusions so the claim matches the data. A serious referee should see it; I’d recommend acceptance after the simultaneous-claim is either demonstrated or reworded.","headline":"The 19.5 ps multimode-fiber jitter is a genuine advance, but the 'simultaneously achieved >80% and <20 ps' claim is not directly supported by any single measurement in the paper.","tokens_in":8479,"tokens_out":2356,"would_cite":true,"duration_ms":23287,"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":"Superconducting nanowire detectors can couple to multimode fibers and still beat 20 ps timing while exceeding 80% efficiency.","keywords":["superconducting nanowire single-photon detectors","multimode fiber coupling","system detection efficiency","timing jitter","polarization dependence","graded-index fiber","NbTiN","cryogenic amplifier"],"falsifier":"Measure the instrument response with a much shorter laser pulse (below 1 ps) and a higher-bandwidth oscilloscope, or compare jitter across two different amplifier chains; if the FWHM drops noticeably below 19.5 ps, the reported jitter is partly electronic rather than intrinsic to the detector-fiber system.","tokens_in":7295,"feed_emoji":"⚡","tokens_out":7600,"duration_ms":64709,"temperature":0.7,"pith_summary":"This paper reports superconducting nanowire single-photon detectors that retain high system detection efficiency when light arrives through a large-core multimode fiber, and shows how the combination can also keep sub-20 ps timing. The central claim is that efficiency above 80% and timing resolution below 20 ps can be achieved together, which earlier multimode-fiber-coupled detectors had not demonstrated under randomized mode illumination. The authors show that polarization is the main obstacle at longer wavelengths, with the TE/TM absorption ratio growing from near 1 at 516 nm to 3.75 at 1550 nm, and demonstrate a fractal nanowire geometry that cuts polarization sensitivity to about 4%. They also identify the timing bottleneck as modal dispersion in step-index multimode fiber, which splits the instrument response into multiple peaks, and solve it with graded-index fiber and a cryogenic amplifier, reaching 19.5±0.2 ps. If correct, the work makes multimode fiber practical for collecting light from quantum dots, tissue, and lidar returns without sacrificing timing.","feed_headline":"Multimode fiber detectors hit 80% efficiency and 19 ps timing","feed_subtitle":"Graded-index fiber and a cryogenic readout remove the timing penalty of wide-core light collection.","key_machinery":"The central objects are NbTiN superconducting nanowire meanders (20, 25, and 50 µm diameter) inside cavity stacks that push absorption toward saturation: an aluminum/SiO2 cavity for visible wavelengths and a distributed Bragg reflector for 900 and 1550 nm. The argument about polarization is carried by the TE/TM absorption ratio of the nanowire, which grows with wavelength and is also the mechanism behind the efficiency measured with randomized modes. Two design elements do the practical work: the fractal nanowire geometry suppresses polarization sensitivity, and the graded-index multimode fiber suppresses modal dispersion, so the cryogenic amplifier's low electrical noise can reveal a 19.5 ps IRF.","core_discovery":"The paper's core discovery is that the multimode fiber itself is not the fundamental limit: polarization of the detected light is. At 516 nm, the TE and TM absorptions of the NbTiN meander are nearly equal, so a 25 µm detector coupled to a 20 µm fiber saturates at about 80% system efficiency with negligible polarization dependence. At 878 nm the TE/TM efficiency ratio is about 2, giving 60% efficiency under randomized modes, and at 1550 nm the ratio is about 3.75, giving 50%. A fractal nanowire layout reduces the polarization sensitivity to about 4% between maximum and minimum, which points to a broadband route for multimode detection. On the timing side, step-index multimode fiber spreads the instrument response into several peaks through modal dispersion, whereas graded-index fiber keeps a single Gaussian peak; with a cryogenic amplifier, the jitter is 19.5±0.2 ps for a 20 µm detector, close to the 18.5±0.1 ps measured through single-mode fiber with the same readout.","pith_inferences":["The reported 19.5 ps jitter still contains the 4.2 ps laser pulse width, the 4 GHz oscilloscope bandwidth, and the amplifier chain; deconvolving those would test whether the detector-fiber system itself is closer to 15 ps or better.","The same fractal geometry that removes polarization dependence at 1550 nm could be combined with larger-area meanders or arrays, potentially extending polarization-insensitive multimode efficiency to the 100 µm core fibers used in lidar and bio-optics.","Since the TE/TM ratio grows with wavelength, cavity designs that equalize absorption at telecom—for example, thicker index-matching layers—could lift random-mode efficiency from 50% toward the 75% TE ceiling, at the cost of bandwidth.","A direct check of the two-peak IRF interpretation would be to vary the bias current: if the side peak grows at low bias, it matches the probabilistic bend-detection regime; if it stays constant, another electronic source is likely."],"forward_implications":["With graded-index multimode fiber and a cryogenic readout, a 20 µm detector shows 19.5±0.2 ps jitter, so multimode coupling no longer forces a timing penalty below the single-mode value.","Random-mode efficiency is predictable from the TE/TM absorption average: 60% at 878 nm and 50% at 1550 nm match the mean of the two polarization-efficiency curves.","The 25 µm detector saturates at about 80% system efficiency through a 20 µm fiber at 516 nm, giving alignment tolerance for visible-light experiments that need large collection areas.","Fractal nanowire layouts keep polarization sensitivity near 4%, meaning efficiency stays stable even if the fiber scrambles the input polarization.","At telecom wavelengths, multimode fiber coupling raises the dark count rate roughly 1000 times over single-mode coupling due to fiber-coupled blackbody radiation, so system design must trade collection area against dark counts."],"supporting_citations":[{"why":"Supplies the high-efficiency SNSPD platform, the resistive-bridge readout, and the end-facet reflection correction used in the efficiency measurements.","marker":"[14]"},{"why":"Earlier multimode-fiber-coupled SNSPD with 70% system efficiency at visible wavelength; the baseline this work improves on in timing.","marker":"[15]"},{"why":"Large-sensitive-area SNSPD at 850 nm; comparison point for multimode-coupled large-area detectors.","marker":"[16]"},{"why":"100 µm diameter single-nanowire detector; earlier large-active-area multimode result whose timing this paper contrasts.","marker":"[17]"},{"why":"16-pixel NbN detector coupled to 300 µm fiber; the largest-core multimode baseline for collected area.","marker":"[18]"},{"why":"Establishes polarization-dependent absorption in superconducting nanowires, the mechanism behind the TE/TM efficiency analysis.","marker":"[19]"},{"why":"Fractal SNSPD design with reduced polarization sensitivity; the structure used to remove polarization dependence in this work.","marker":"[22]"},{"why":"Identifies deterministic and probabilistic jitter regimes in nanowire detection; used to interpret the asymmetric two-peak instrument response.","marker":"[24]"}],"fun_headline_variants":["Multimode SNSPDs hit 80% efficiency, 19 ps jitter","Polarization solved: multimode detectors reach 80% and 19 ps","Fractal layout tames polarization in multimode SNSPDs","Graded-index fiber keeps timing in wide-core SNSPDs","Multimode photons: 80% efficiency and 19 ps in one detector"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The reported 19.5 ps timing width is treated as the detector-fiber system's own jitter, but the measurement also includes the 4.2 ps laser pulse, the oscilloscope bandwidth, the trigger electronics, and the amplifier; if those add substantially, the detector's true timing improvement is overstated.","fun_headline_variants_meta":{"raw":{"variants":["Multimode SNSPDs hit 80% efficiency, 19 ps jitter","Polarization solved: multimode detectors reach 80% and 19 ps","Fractal layout tames polarization in multimode SNSPDs","Graded-index fiber keeps timing in wide-core SNSPDs","Multimode photons: 80% efficiency and 19 ps in one detector"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000731,"raw_usage":{"total_tokens":3268,"prompt_tokens":941,"completion_tokens":2327,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":557,"completion_tokens_details":{"reasoning_tokens":2227}},"tokens_in":557,"tokens_out":2327,"duration_ms":15676,"temperature":1.0,"reasoning_tokens":2227,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:52:59.839600+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the instrument response with a much shorter laser pulse (below 1 ps) and a higher-bandwidth oscilloscope, or compare jitter across two different amplifier chains; if the FWHM drops noticeably below 19.5 ps, the reported jitter is partly electronic rather than intrinsic to the detector-fiber system.","supporting_citations":[{"cited_title":"Single-photondetectorscombininghighefficiency,highdetectionrates,andultra-hightimingresolution,","cited_arxiv_id":null,"evidence_quote":"Supplies the high-efficiency SNSPD platform, the resistive-bridge readout, and the end-facet reflection correction used in the efficiency measurements."},{"cited_title":"Multimode fiber-coupled superconducting nanowire single-photon detector with 70% system efficiency at visible wavelength,","cited_arxiv_id":null,"evidence_quote":"Earlier multimode-fiber-coupled SNSPD with 70% system efficiency at visible wavelength; the baseline this work improves on in timing."},{"cited_title":"Large-sensitive-area superconducting nanowire single-photon detector at 850 nm with high detection efficiency,","cited_arxiv_id":null,"evidence_quote":"Large-sensitive-area SNSPD at 850 nm; comparison point for multimode-coupled large-area detectors."},{"cited_title":"Large active area superconducting single-nanowire photon detector with a 100µm diameter,","cited_arxiv_id":null,"evidence_quote":"100 µm diameter single-nanowire detector; earlier large-active-area multimode result whose timing this paper contrasts."},{"cited_title":"A 16-pixel NbN nanowire single photon detector coupled with 300 micrometer fiber","cited_arxiv_id":"1811.09779","evidence_quote":"16-pixel NbN detector coupled to 300 µm fiber; the largest-core multimode baseline for collected area."},{"cited_title":"Superconducting single photon detectors with minimized polarization dependence,","cited_arxiv_id":null,"evidence_quote":"Establishes polarization-dependent absorption in superconducting nanowires, the mechanism behind the TE/TM efficiency analysis."},{"cited_title":"Fractal superconducting nanowire single-photon detectors with reduced polarization sensitivity,","cited_arxiv_id":null,"evidence_quote":"Fractal SNSPD design with reduced polarization sensitivity; the structure used to remove polarization dependence in this work."},{"cited_title":"Physical mechanisms of timing jitter in photon detection by current-carrying superconducting nanowires,","cited_arxiv_id":null,"evidence_quote":"Identifies deterministic and probabilistic jitter regimes in nanowire detection; used to interpret the asymmetric two-peak instrument response."}],"review_version":1}