{"id":"89328504-93d3-4d71-a8f7-e3e10a55e7ea","arxiv_id":"2510.02598","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Microstructured Si-PIN diodes with honeycomb walls demonstrate angular-selective electron detection, reproducing simulated angular-response curves.","lead":"A silicon detector etched into a honeycomb of active walls and dead floors counts electrons differently depending on how steeply they arrive. This angular selectivity could filter background electrons in the KATRIN neutrino-mass experiment.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Demonstration lacks a flat-diode control or beam-intensity monitor, so tilt-dependent setup artifacts are not excluded.","rationale":"The reader's weakest assumption was the unverified 12 keV detection threshold in the Geant4/Penelope simulation, which affects the quantitative agreement. My concern is complementary: even the qualitative central claim would be strengthened by a flat-diode control or a beam-intensity monitor, because the angular scan could in principle be biased by setup-related changes in electron flux or beam position. This does not overturn the CONDITIONAL verdict—the reported effect is still plausible and the paper is transparent about limitations—but it identifies an additional experimental control that would make the demonstration more robust. I therefore agree partially with the reader: the threshold is a real issue, but the missing control is at least as load-bearing for the central claim.","tokens_in":11385,"tokens_out":12517,"duration_ms":127384,"concrete_test":"Repeat the tilt scan with a flat (unstructured) Hamamatsu S3590 diode of the same size in the same holder and beam conditions, measuring net count rate vs tilt angle. If the flat diode shows a flat or decreasing rate, the microstructured sample's increase is confirmed as due to the sidewall geometry. Additionally, monitor the source intensity during the scan with a Faraday cup or a second detector to correct for drift and verify that the beam position on the detector is stable. If the flat diode also shows an increasing rate, the claim of angular-selective efficiency for the aTEF is not supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that the measured count-rate increase with incidence angle demonstrates angular-selective detection efficiency of the microstructured Si-PIN diodes. This inference requires that the tilt sequence does not itself change the number or position of electrons hitting the detector in a way that biases the rate. The paper presents no control measurement with an unstructured reference diode under identical conditions, and no monitor of the incident beam intensity or position during the angular scan. The fixed 5 mm aperture in the detector holder has a projected area that changes with tilt (cos(theta)), which would reduce transmitted flux; the observed increase means this effect is dominated, but other effects (magnetic steering, beam-spot movement across a nonuniform microstructure, source drift) are unquantified. The simulation comparison does not close this gap because the simulated curves are scaled by a free factor and use an unverified 12 keV detection threshold (footnote o, Sec. 4), so the shape match could be achieved for reasons unrelated to the microstructure. Without a flat reference or a beam monitor, the qualitative demonstration rests on a single uncalibrated measurement series.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes an 'active Transverse Energy Filter' (aTEF) based on a microstructured Si-PIN diode: deep hexagonal holes are etched from the front side so that electrons with high incidence angles hit the active side walls and produce a signal, while low-angle electrons pass into insensitive channel grounds. The authors fabricate prototypes by ICP-RIE from commercial Hamamatsu S3590 diodes, model the depletion-zone electrostatics with a 1D Poisson treatment and COMSOL, and report proof-of-principle measurements in a magnetically guided photoelectron test stand. The central experimental result is that the measured net count rate increases with detector tilt angle, and this angular profile is compared with Geant4/Penelope simulations including backscattering, with simulated counts scaled by a common factor. They also test resist removal and SiO2 passivation, observing reduced dark current but degraded detector performance.","tokens_in":11583,"tokens_out":3587,"duration_ms":31646,"significance":"If the angular-selective detection efficiency is real, the silicon aTEF could provide a novel, active background-suppression method for MAC-E-filter spectrometers such as KATRIN, where signal and background electrons differ mainly in incidence angle. The paper contains a direct demonstration of the qualitative effect—count rate increasing with incidence angle—and a simulation study that supports the interpretation via backscattering. The electrostatic modeling is a useful contribution, and the authors are candid about many limitations, including unverified threshold, simplified geometry, and fabrication variability. However, the absence of a flat-diode control or beam monitor leaves a plausible setup-artifact explanation for the central measurement, and the quantitative simulation comparison relies on a free scale factor and an unverified 12 keV detection threshold. These issues are load-bearing for the claim of a demonstrated angular-selective detection efficiency.","major_comments":[{"comment":"The central inference is that the measured count-rate increase with tilt angle demonstrates angular-selective detection. This requires excluding tilt-dependent setup artifacts. No flat-diode control measurement under the identical tilt sequence is shown, and no beam-intensity or beam-position monitor is described. The fixed 5-mm aperture in the detector holder has a projected area that decreases as cos(theta), so the observed increase is not a trivial consequence of that geometric effect; however, other artifacts—magnetic steering, beam-spot movement across a nonuniform microstructure, source drift during the angular scan—are unquantified. The simulation comparison does not close this gap because the simulated curves are scaled by a free factor and use an unverified detection threshold. Please provide a flat-diode reference or an independent beam monitor, or explicitly restrict the claim","section":"Sec. 4, Fig. 10"},{"comment":"The quantitative agreement between measurement and simulation is weakened by two unconstrained inputs: the simulated count rate is multiplied by a free common scale factor to match the measured counts, and the detection threshold of 12 keV is stated in footnote o to be 'an estimate based on experience with the samples, but was not verified quantitatively.' Because the scale factor absorbs the absolute efficiency, the comparison only tests the shape of the angular profile. The shape match with backscattering could be fortuitous if the effective detection threshold varies with angle or if omitted effects such as surface roughness and geometry changes compensate. Please either calibrate the threshold by an independent measurement (e.g., energy-loss spectra from a flat diode) or show that the simulated angular profile is robust over a plausible range of thresholds and scale factors.","section":"Sec. 4, footnote o, and Fig. 10"},{"comment":"The simulation assumes a simplified geometry with uniform hexagonal channels, constant wall thickness, a 25-um photoresist layer, and no surface roughness or channel-geometry variation along the etch depth. Footnote o acknowledges these omissions, and Sec. 6 describes the actual prototypes as having 'inhomogeneities and an overall rough surface after microstructuring.' This mismatch between simulated and real geometry limits the strength of any quantitative conclusion about the angular-response mechanism. Please quantify the sensitivity of the simulated profile to plausible variations in wall thickness, channel depth, photoresist thickness, and surface roughness, or clearly present the simulation as illustrative rather than as a validated model.","section":"Sec. 4, simulation geometry"}],"minor_comments":[{"comment":"The phrase 'angular-selective photoelectron source' is not defined in the abstract. The setup actually tilts the detector relative to a magnetically guided beam; please clarify this in the abstract or introduction.","section":"Abstract / Sec. 1"},{"comment":"The condition distinguishing the two cases in Eq. (5) is written as 'U < U_aTEF' and 'U > U_aTEF', but the potential U is a function of x, while the boundary between regimes should be on the depletion width w relative to x_aTEF (as used in the text 'for w < x_aTEF'). Please correct the notation to avoid confusion.","section":"Eq. (5) and Appendix A"},{"comment":"The photoresist density is given as '1.2 g cm^-2'; presumably this should be 'g cm^-3'.","section":"Sec. 4, simulation parameters"},{"comment":"The top panel shows spectra with a color scale labeled 'alpha_D (°)' but the caption does not explain which curve corresponds to which angle in the bottom panel. The overlap of the two panels could be clarified in the caption.","section":"Sec. 4, Fig. 10"},{"comment":"The statement that 'sensitive channels could only be demonstrated with wall thicknesses >25 um so far' is important for the applicability of the concept, but it is buried in the fabrication section. It deserves to be highlighted in the summary or outlook, because the optimal KATRIN geometry requires d = 10 um.","section":"Sec. 3"},{"comment":"The sentence 'Especially, all of them exhibited a reduced signal rate after the removal of the photoresist' could be more specific: is the reduction relative to the same sample before resist removal, or relative to unprocessed diodes? Please clarify the comparison.","section":"Sec. 5"}],"recommendation":"major_revision","confidential_remarks":"The paper's core concept is interesting and the qualitative rate increase with angle is visible in the data. However, the proof-of-principle claim is currently underdetermined by the measurements: the lack of a flat-diode control or beam monitor leaves a plausible artifact channel, and the simulation comparison is flexible enough that the shape agreement is not decisive. The authors are transparent about many limitations, which is commendable, but the manuscript would need either a control measurement or a careful rewriting of the claim to be suitable for publication. I recommend major revision with the expectation that the authors add a flat-diode reference run or a beam-intensity/position monitor, and that they quantitatively address the sensitivity of the simulation to the threshold and geometry assumptions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The thing to know: this is the first time anyone has shown a microstructured Si-PIN diode detecting more electrons at higher incidence angles, and the effect is directly visible in their count-rate-vs-angle data. That is a real step for the aTEF concept. The paper is also honest about its own limitations—the 12 keV threshold is admitted to be an estimate, the geometry is simplified, surface roughness is ignored, and the prototypes degrade after resist removal and passivation. Credit where due: the fabrication itself is nontrivial, and the simple Poisson-equation model for depletion in a microstructured diode is a useful piece of theory. The Geant4/Penelope comparison is a good-faith effort, and the shape agreement with backscattering included is not nothing.\n\nNow the soft spots, in proportion. The stress-test note is right that there is no flat-diode control and no beam-intensity or beam-position monitor during the tilt sequence. That means the measured angular dependence could in principle be contaminated by beam spot movement across a nonuniform structure, magnetic steering, or source drift. The observed increase is opposite to the cos(theta) aperture effect, which argues against a trivial geometric explanation, but it does not close the loophole. The simulation comparison also has a free scale factor and an unverified threshold, so the shape match is suggestive rather than quantitative proof. None of this kills the paper—the central qualitative effect is directly measured—but it does mean the demonstration is a proof-of-principle, not a validated detector technology. The authors themselves frame it that way, and that framing is accurate.\n\nThe paper deserves a serious referee. It is a first experimental result in a line of work that could matter for KATRIN, and the authors have made a genuine attempt to test their own device. What needs attention in review: the missing control, the unquantified threshold, and the absence of error bars on the angle-dependent rates. Those are addressable, and the authors should be asked to supply a flat-diode tilt scan or a beam monitor before publication.\n\nWho gets value from this: anyone working on MAC-E-filter background suppression, low-energy electron detectors, or silicon detectors with engineered angular response. It is not a paradigm shift, but it is a legitimate incremental advance with clear next steps.","headline":"First experimental demonstration of an angular-selective silicon detector, but the proof-of-principle is softer than the abstract suggests—no flat-diode control, and the simulation match leans on a scaled curve and an unverified 12 keV threshold.","tokens_in":12146,"tokens_out":1609,"would_cite":true,"duration_ms":22240,"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":"A microstructured silicon PIN diode with hexagonal holes detects electrons preferentially by their angle of incidence, an effect the paper demonstrates and matches with simulations.","keywords":["angular-selective detector","Si-PIN diode","transverse energy filter","electron spectroscopy","MAC-E filter","microstructured silicon","backscattering","neutrino-mass experiment"],"falsifier":"Measure the angular response of a flat (unstructured) reference diode under identical illumination conditions: if a comparable rise in count rate with tilt is observed, the effect is not due to the microstructure. Alternatively, calibrate the absolute detection threshold of the microstructured diode by recording pulse-height spectra from monoenergetic electrons and check whether the simulated angular profile is reproduced when the known threshold is used.","tokens_in":11273,"feed_emoji":"🔬","tokens_out":3816,"duration_ms":30857,"temperature":0.7,"pith_summary":"This paper claims that a Si-PIN diode whose surface is etched into a honeycomb of deep hexagonal holes—with the side walls remaining active and the channel floors made insensitive—acts as an angular filter for low-energy electrons. Using a rotatable photoelectron source, the authors measured the detector's count rate as a function of incidence angle and found it rises steeply with angle, in line with Monte Carlo simulations that include electron backscattering. The underlying physics is that electrons hitting a side wall deposit energy and are counted, while near-normal electrons land on the insensitive channel floors and are lost. If the effect holds, such active transverse energy filters could suppress low-angle background electrons in MAC-E-filter spectrometers, where signal and background have similar energy but different angular spreads. The paper also shows the microstructuring degrades energy resolution and that a SiO2 passivation layer reduces dark current but did not restore full performance.","feed_headline":"Honeycomb silicon diode sorts electrons by angle","feed_subtitle":"Microstructured PIN detector could slash background in tritium neutrino-mass spectrometers.","key_machinery":"The load-bearing element is the microstructured Si-PIN diode, or aTEF: a honeycomb of hexagonally arranged deep holes with etched silicon side walls that remain part of the charge-collecting depletion zone, while the channel floors and bulk are depleted only at much higher bias and therefore stay insensitive. A one-dimensional Poisson-equation model, validated by field simulation, shows that the microstructure can be fully depleted at modest reverse bias (around 30 V for 150 µm deep channels) and explains how the charge-neutrality condition shifts the potential. The measured angular response is reproduced by Monte Carlo simulation only when electron backscattering from the walls and floor is","core_discovery":"The central claim is that deep etching of a commercial Si-PIN diode into a hexagonal channel structure with active side walls and insensitive channel bottoms yields an angular-dependent detection efficiency for keV electrons. In a test stand with an angular-selective photoelectron source, the measured net signal rate increased monotonically with tilt angle from normal incidence, and the angular profile agreed with simulated profiles when backscattering of electrons in the sensitive volume was included. The authors infer that low-incidence-angle electrons are absorbed in the insensitive channel grounds, while steeper electrons hit the active side walls and produce a detectable charge signal.","pith_inferences":["The angular response depends sensitively on the absolute detection threshold; if the true threshold differs from the 12 keV estimate, the match to simulation might weaken, so a calibrated charge-collection measurement is the obvious next test.","An array of such microstructured pixels with individual readout could turn the angular filter into an imaging pitch-angle spectrometer, effectively measuring the angular distribution of the incoming electron flux in one shot.","Because the detector is a single diode, the same angular-selective readout could be implemented on existing silicon detector geometries without changing the electronics chain, lowering the barrier for adoption."],"forward_implications":["An aTEF-based focal-plane detector could suppress low-angle background electrons in MAC-E-filter spectrometers, reducing background by a factor of 3–5 for the optimized geometry while leaving the steep-angle signal largely intact.","The same angular selectivity could be used in any magnetically guided low-energy electron beam where signal and background differ only in pitch angle, without needing tracking detectors or ΔE-E telescopes.","Fabrication improvements (smoother channels, thin walls, passivation) are identified as the path to practical devices; the current prototypes demonstrate the principle but degrade energy resolution.","The Poisson-model treatment of the microstructured depletion zone is transferable to other microstructured silicon detectors, predicting at what bias the walls become sensitive."],"fun_headline_variants":["Honeycomb Si detector filters electrons by angle","Angular-selective silicon detector for cleaner neutrino spectra","Microstructured diode weeds out background electrons","Tilt-sensitive honeycomb detector boosts neutrino-mass sensitivity","Angle-sorting silicon chips sharpen electron spectrometry"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The convincing part of the claim—that the count rate rises with incidence angle—rests on a detection threshold of 12 keV that the authors estimated from past experience but never verified, so the quantitative match to simulation could be accidental.","fun_headline_variants_meta":{"raw":{"variants":["Honeycomb Si detector filters electrons by angle","Angular-selective silicon detector for cleaner neutrino spectra","Microstructured diode weeds out background electrons","Tilt-sensitive honeycomb detector boosts neutrino-mass sensitivity","Angle-sorting silicon chips sharpen electron spectrometry"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000298,"raw_usage":{"total_tokens":1493,"prompt_tokens":607,"completion_tokens":886,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":351,"completion_tokens_details":{"reasoning_tokens":830}},"tokens_in":351,"tokens_out":886,"duration_ms":6165,"temperature":1.0,"reasoning_tokens":830,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T12:38:01.391694+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the angular response of a flat (unstructured) reference diode under identical illumination conditions: if a comparable rise in count rate with tilt is observed, the effect is not due to the microstructure. Alternatively, calibrate the absolute detection threshold of the microstructured diode by recording pulse-height spectra from monoenergetic electrons and check whether the simulated angular profile is reproduced when the known threshold is used.","supporting_citations":[],"review_version":1}