{"id":"71f35460-3ed7-42a4-8d4f-e06c796c6229","arxiv_id":"2501.05208","paper_version":1,"verdict":"CONDITIONAL","confidence":"LOW","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Scanning noise microscopy detects a near-field signal from hot electrons at a buried InSb/CdTe interface, but the claimed origin in Coulomb scattering by charged ions is inferred rather than directly proven.","lead":"A scanning probe technique called a scanning noise microscope maps hot electron scattering at a buried InSb/CdTe interface, with strong signals only when the interface is close to the surface. The result suggests the microscope could serve as a nondestructive probe of heterojunction quality for spintronic devices.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Depth-dependent signal (60 vs 400 nm InSb) does not establish Coulomb scattering by interfacial ions; surface-channel phonon scattering is excluded only by assertion, so the causal mechanism in the abstract is not supported.","rationale":"The reader's conditional verdict is appropriate. The paper's strongest contribution is the SNoiM visualization of a signal originating from a buried interface, evidenced by the thin-vs-thick comparison and the near-field decay measurement. However, the abstract's mechanism statement is the central claim, and it overreaches relative to the data. The authors themselves hedge on page 6 with 'can be speculated' and 'may come from,' indicating that the charged-ion Coulomb scattering attribution is not definitively demonstrated. The depth control alone does not rule out phonon scattering in the surface InSb channel, nor does it isolate the scattering mechanism from other thickness-dependent sample properties. An independent control with an interdiffusion-free interface would directly test the attribution. Until such a control is provided, the claim should be softened to 'consistent with' rather than 'originates from,' and the verdict should remain conditional. I therefore agree with the reader's assessment and propose no change to the verdict.","tokens_in":5297,"tokens_out":5583,"duration_ms":59330,"concrete_test":"Grow a reference heterostructure with the same 60 nm InSb top layer on a lattice-matched, interdiffusion-free barrier (e.g., InSb/AlInSb) that hosts a 2DEG but lacks interfacial charged ions, process it into identical nano-constrictions, and compare SNoiM signals under the same bias conditions. If a comparable near-field signal is observed, the charged-ion Coulomb scattering attribution is not unique; if the signal is absent, the mechanism is supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that the near-field signal 'originates from the Coulomb scattering of charged ions on electrons at the interface of the embedded layer heterojunction.' On page 6 the authors state the strong signal 'can be speculated... may come from the electronic random motion formed by the intense scattering of charged ions'—explicitly speculative. The only evidence adduced is the comparison of 60 nm and 400 nm InSb thicknesses (Fig. 3(d)): the signal is strong for thin InSb and negligible for thick InSb. This shows the source must be near the interface, because the near-field decays within about 100 nm, but it does not discriminate among scattering mechanisms. The surface InSb conducting channel, which the authors say exists, could produce hot-electron noise via phonon or neutral-impurity scattering. The authors dismiss phonon scattering by asserting it is 'much weaker' than impurity scattering, but provide no quantitative estimate or control experiment. Thickness-dependent changes in strain relaxation, dislocation density, surface Fermi level, or carrier confinement could also account for the observed difference. Thus the causal mechanism attribution is not established by the data presented.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports scanning noise microscope (SNoiM) measurements on InSb/CdTe heterojunction nanodevices with a buried 2DEG interface. The near-field signal is found to scale roughly linearly with bias current, to decay within ~40 nm of the surface, and to be strong in a 60 nm-thick InSb device while negligible in a 400 nm-thick InSb device. From these observations the authors conclude that the detected evanescent field originates from hot-electron noise generated by Coulomb scattering of interfacial charged ions on the 2DEG at the buried heterojunction, and they propose SNoiM as a nondestructive tool for probing such buried interfaces.","tokens_in":5477,"tokens_out":2009,"duration_ms":20271,"significance":"If the mechanism attribution were established, the result would be of appreciable interest: it would demonstrate a real-space, nondestructive probe of scattering processes at a buried heterointerface, with direct relevance to InSb/CdTe spintronic devices. The paper has clear strengths: the central observation does not rely on fitted parameters, the evanescent decay check is a useful control, and the comparison with the standard shot-noise scaling ⟨S∝2e|I|⟩ is a reasonable first test. The depth-dependent comparison between 60 nm and 400 nm InSb layers does support a buried-interface origin of the signal. However, the paper's central causal claim—that the noise is specifically due to Coulomb scattering by charged ions—is not established by the presented data, and the text itself hedges this attribution as speculation. Because the abstract states this mechanism as a demonstrated result, the significance of the work as currently written is limited by the gap between claim and evidence.","major_comments":[{"comment":"The abstract states that 'the near-field signal originates from the Coulomb scattering of charged ions on electrons at the interface of the embedded layer heterojunction,' but the body of the paper, on page 6, explicitly says that the strong signal 'can be speculated' and 'may come from' the electronic random motion formed by intense scattering of charged ions. The thickness comparison in Fig. 3(d) shows that the source is near the buried interface, but it does not discriminate among Coulomb scattering by charged ions, phonon scattering, neutral impurity scattering, or thickness-dependent changes in strain, dislocation density, or confinement. The dismissal of phonon scattering is made by the assertion that it is 'much weaker' than impurity scattering, with no quantitative estimate or control experiment. This is a load-bearing gap: the abstract's causal mechanism is not supported by the data presented.","section":"Abstract and page 6 (mechanism attribution)"},{"comment":"The only evidence for the interfacial origin of the signal is the comparison between a 60 nm and a 400 nm InSb device, with no error bars, no number of measured devices, and no information on whether the 400 nm device had identical growth and processing except for the InSb thickness. Differences in strain relaxation, surface Fermi level pinning, interface interdiffusion, or carrier confinement between the two growths could also produce the observed difference. The authors should provide statistics, repeated devices, and ideally a series of intermediate thicknesses or a control with modified interface conditions to support the claim that the signal disappears solely due to the distance from the surface.","section":"Fig. 3(d) and page 6"},{"comment":"No quantitative model connects the density of interfacial charged ions (or their scattering strength) to the expected SNoiM noise amplitude. Without such a model, the statement that the observed signal is dominated by Coulomb scattering rather than by other mechanisms remains one of several plausible interpretations. A quantitative estimate of the relative contributions of phonon and charged-ion scattering to the evanescent noise, or a prediction of how the signal should vary with temperature or ion density, would make the mechanism claim testable.","section":"Section 2 (model of signal generation)"},{"comment":"The linear dependence of the near-field signal on current (Fig. 2(c)) is cited as consistent with hot-electron shot noise, but shot-noise-like scaling is not mechanism-specific; it does not by itself distinguish Coulomb scattering from phonon or other elastic scattering. The authors should clarify what additional information, if any, the linear scaling provides for the mechanism attribution.","section":"Section 2 (linear dependence)"}],"minor_comments":[{"comment":"There are typographical errors that should be corrected, including 'siginificant', 'heterojucntion', 'makes its suitable', 'interfaced', and 'th at'.","section":"Throughout"},{"comment":"The text refers to 'Fig. 2(h)' when discussing the evanescent decay of the near-field signal, but the figure and its caption only show panels (a)–(f). The reference should be corrected.","section":"Page 6, Fig. 2(h)"},{"comment":"Several reference entries have formatting issues, such as reference 8 ('L. B Abdalla' missing a period), reference 11 with inconsistent punctuation, and reference 13 with 'Sci. Rep' missing the period after 'Rep'. These should be cleaned up.","section":"References"},{"comment":"The caption for Fig. 3(d) states both that the weak signal 'can be considered as system background noise and neglected' and that 'Undetectable near-field signal is tested in the 400 nm thick InSb/CdTe device'; these statements are redundant and should be unified.","section":"Fig. 3(d)"}],"recommendation":"major_revision","confidential_remarks":"The paper is likely within the scope of the journal, and the SNoiM technique itself is of interest. The main concern is that the abstract overstates the mechanism attribution, which the body hedges as speculation. The authors should either (a) temper the abstract and conclusions to match the evidence, or (b) add the missing quantitative or control data needed to support the Coulomb-scattering claim. I would not recommend rejection because the depth-dependent observation is a useful experimental result, but the central claim as currently stated needs substantial revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: the paper has a real observation—SNoiM sees a near-field signal only when the InSb layer is thin, which puts the source near the buried InSb/CdTe interface—but the abstract overclaims the mechanism. The body itself says the Coulomb-scattering attribution is speculative, and it is.\n\nWhat's genuinely new here is applying the group's SNoiM tool to InSb/CdTe and showing that a buried 2DEG generates detectable evanescent noise. That is a useful nondestructive probe for interface quality in a system where interdiffusion matters. The depth comparison (60 vs 400 nm) is a reasonable control for surface-versus-buried origin, and the near-field signal scaling roughly with current is consistent with hot-electron shot noise. The authors also deserve credit for clearly describing the two conducting channels and acknowledging that SNoiM should, naively, be more sensitive to the surface channel.\n\nThe soft spots are real, though. The mechanism attribution to charged-interface-ion Coulomb scattering is not established by the data. The 400 nm sample only rules out signal from the surface InSb channel if you accept the near-field decay argument; it does not discriminate between Coulomb scattering, phonon scattering, or other interface disorder. The paper discounts phonon scattering with a single sentence and no quantitative estimate. There are no error bars or statistics on the signal comparison, and the figures themselves are not accessible from the text, so the experimental claims cannot be independently checked. The passage on page 6 ('can be speculated... may come from') is more honest than the abstract, which states the origin as a demonstrated fact. Thickness-dependent changes in strain, dislocation density, or surface Fermi level could also explain the difference.\n\nMinor: the self-citation to the prior SNoiM papers is fine—they built the tool—and the novelty is appropriately incremental rather than paradigm-shifting.\n\nWho is this for? Researchers working on InSb/CdTe spintronics or on near-field noise microscopy of buried heterojunctions. They would get a plausible new data point and a technique demonstration. I would not cite it in my own work, but I would bring it to a reading group as an example of a promising method with a mechanism claim that outruns the evidence.\n\nRecommendation: yes, send this to peer review. The observation is likely real and the technique is useful. The referee should push for a softened abstract, error bars, and a serious discussion of alternative scattering mechanisms. It deserves referee time, not desk rejection.","headline":"Plausible buried-interface detection with SNoiM in InSb/CdTe, but the Coulomb-scattering mechanism is asserted in the abstract and only speculated in the body.","tokens_in":6053,"tokens_out":1620,"would_cite":false,"duration_ms":16662,"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":"The near-field signal in InSb/CdTe nanodevices comes from Coulomb scattering of hot electrons at the buried heterojunction interface, not from the surface channel.","keywords":["InSb/CdTe heterojunction","scanning noise microscope","near-field imaging","hot electron scattering","two-dimensional electron gas","Coulomb scattering","interfacial interdiffusion","evanescent field"],"falsifier":"Grow InSb/CdTe samples in which the interfacial interdiffusion is suppressed or separated from the 2DEG by a thin spacer layer, and compare the SNoiM signal; if the signal stays strong when charged ions are absent or remote, the charged-ion mechanism is wrong. Alternatively, measure the near-field signal versus temperature: Coulomb scattering by fixed ions is nearly temperature-independent, while phonon scattering rises strongly with temperature.","tokens_in":5075,"feed_emoji":"🔬","tokens_out":7605,"duration_ms":64759,"temperature":0.7,"pith_summary":"This paper claims that a scanning noise microscope can detect hot-electron scattering at the buried InSb/CdTe heterojunction interface, and that the detected near-field signal is dominated by Coulomb scattering of the interfacial two-dimensional electron gas by charged ions produced by interdiffusion. The authors support the claim by showing that the evanescent signal appears only when the InSb top layer is thinner than about 100 nm, vanishes when the interface is buried 400 nm deep, follows the expected shot-noise scaling with current, and localizes at the nano-constriction where heating is strongest. If correct, the result turns SNoiM into a nondestructive, real-space probe of buried heterojunction interface quality, which matters for improving InSb/CdTe growth and spintronic device performance.","feed_headline":"Noise microscope sees hot electrons at a buried interface","feed_subtitle":"Thin InSb devices show strong near-field signal from interfacial charged-ion scattering; thick ones show none.","key_machinery":"The load-bearing object is the scanning noise microscope (SNoiM), a near-field technique in which a sharp metal tip scatters evanescent electromagnetic fields generated by local current fluctuations, converting them into a detectable far-field signal. The argument's geometric lever is the measured decay length: the evanescent field vanishes within tens of nanometers above the surface, so a signal that survives in 60 nm-thick InSb but disappears at 400 nm can only originate from the buried interface. That interface hosts a type-I ladder-band 2DEG, and interdiffusion supplies charged ions that strongly scatter the 2DEG, which is the mechanism the paper identifies as generating the detected noise.","core_discovery":"The central discovery is the local origin of the SNoiM near-field signal in InSb/CdTe nanodevices: it comes from the two-dimensional electron gas at the buried heterojunction interface, not from the conductive InSb surface layer. In devices with a thin InSb layer (60 nm), a strong near-field hot spot appears at the constriction and scales roughly linearly with bias, consistent with hot-electron shot noise; in devices with 400 nm InSb the signal falls to background. Since the evanescent field decays within tens of nanometers of the tip, the survival of the signal at 60 nm depth and its disappearance at 400 nm place the source at the buried interface. The authors attribute the strong scattering there to Coulomb interaction between the 2DEG and positively and negatively charged ions formed by interdiffusion across the InSb/CdTe interface.","pith_inferences":["A natural testable extension is temperature-dependent SNoiM: if Coulomb scattering dominates, the signal should be only weakly temperature dependent, whereas phonon scattering should produce a strong increase at higher temperatures.","If the charged-ion interpretation holds, the SNoiM signal could be calibrated against interfacial ion density, turning the microscope into a quantitative probe of interdiffusion for growth optimization.","The same approach should transfer to other heterojunction or oxide-interface systems where buried 2DEGs coexist with interfacial charged defects, as long as the active interface lies within the evanescent decay length."],"forward_implications":["For InSb/CdTe devices with InSb thickness below about 100 nm, SNoiM can image hot-electron scattering at the buried heterojunction interface in real space and nondestructively.","The near-field signal follows the shot-noise form ⟨S_shot⟩ ∝ 2e|I|, so its amplitude can be used as a local measure of the scattering rate at the interface.","The absence of signal for 400 nm InSb means SNoiM's reach is limited to near-surface interfaces, but it also provides a check that the signal truly comes from the buried layer.","Interfacial charged-ion scattering, previously inferred from band-structure and transport studies, is shown to be the dominant noise source probed by SNoiM, linking interfacial interdiffusion to local hot-electron dynamics.","Because the measurement is nondestructive, it can be used to evaluate interface quality of as-grown InSb/CdTe heterostructures before device processing."],"supporting_citations":[{"why":"Introduces SNoiM and demonstrates its ability to map nonlocal hot-electron transport in GaAs/AlGaAs devices; this paper applies that method to InSb/CdTe.","marker":"19"},{"why":"Sets out the SNoiM detection principle of sensing evanescent fields from local current fluctuations.","marker":"20"},{"why":"Gives the band-structure result that interfacial band reversal and charged InSb/CdTe surfaces arise from interdiffusion, the source of the charged ions.","marker":"8"},{"why":"Documents negatively charged InSb and positively charged CdTe surface states at the interface, supporting the charged-ion scattering picture.","marker":"18"},{"why":"Describes the MBE growth of the InSb/CdTe heterostructures used in this study.","marker":"22"}],"fun_headline_variants":["Scanning noise microscope finds hot electrons at buried interface","Hot electron scattering localized to InSb/CdTe interface","Interface scattering visualized by noise microscope in InSb/CdTe","Buried interface source of hot electron signal in nanodevice","Local hotspots from charged ions seen at heterojunction"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The claim depends on the assumption that the dominant source of the near-field noise is Coulomb scattering of the interface electrons by charged ions, rather than phonon scattering, surface traps, or other disorder; the paper asserts this attribution but does not measure ion density or fit a quantitative scattering model.","fun_headline_variants_meta":{"raw":{"variants":["Scanning noise microscope finds hot electrons at buried interface","Hot electron scattering localized to InSb/CdTe interface","Interface scattering visualized by noise microscope in InSb/CdTe","Buried interface source of hot electron signal in nanodevice","Local hotspots from charged ions seen at heterojunction"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000528,"raw_usage":{"total_tokens":2534,"prompt_tokens":922,"completion_tokens":1612,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":538,"completion_tokens_details":{"reasoning_tokens":1530}},"tokens_in":538,"tokens_out":1612,"duration_ms":12988,"temperature":1.0,"reasoning_tokens":1530,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T21:13:07.873199+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Grow InSb/CdTe samples in which the interfacial interdiffusion is suppressed or separated from the 2DEG by a thin spacer layer, and compare the SNoiM signal; if the signal stays strong when charged ions are absent or remote, the charged-ion mechanism is wrong. Alternatively, measure the near-field signal versus temperature: Coulomb scattering by fixed ions is nearly temperature-independent, while phonon scattering rises strongly with temperature.","supporting_citations":[],"review_version":1}