{"id":"fad4df6d-8296-4b5a-8674-cd9a5830297c","arxiv_id":"1908.10472","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A Pt/YIG spin Seebeck detector exhibits flat photovoltage response from 390 to 2200 nm, with an estimated spin Seebeck coefficient of about 60 nV/K.","lead":"This paper shows that a platinum-on-magnetic-insulator device can detect light from 390 to 2200 nanometers by converting heat into spin currents. The response is flat and follows how strongly platinum absorbs light, pointing to a thermal spin-based detector rather than a conventional photodiode.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Reported SLSSE calibration pairs the 3.457 µV signal with 1190 W/cm², but Appendix B assigns that same voltage to 18 mW (~360 W/cm²); the coefficient may be off by ~3.3×.","rationale":"The reader's CONDITIONAL verdict is appropriate. I read the paper as claiming (i) a broadband photovoltage in Pt/YIG/GGG caused by optically generated LSSE and ISHE, and (ii) a new amplitude-modulation method that yields a precise thermal-gradient calibration and SLSSE = 60 ± 7.8 nV/K. The qualitative evidence for (i) is strong: field and angle dependence, linear power scaling, independence of pulse repetition rate, Pt thickness scaling, front/back/edge illumination contrasts, and spatial maps all point to a thermally driven spin-current ISHE signal. The PSV contribution is plausibly small, though the estimate is parameter-dependent. The load-bearing weak point is in (ii), but it is not primarily the absolute TGGG ≈ TCu assumption: because the AM method compares zero crossings with and without light, a constant thermal offset between GGG and the heat sink largely cancels. What does not cancel is a mismatch in the conditions used for VISHE and ΔT. As written, Sec. II F uses VISHE = 3.457 µV at 1190 W/cm², while Appendix B assigns the same voltage to 18 mW on an 80 µm spot, i.e., ~360 W/cm², a factor of 3.3 lower intensity. Linear power scaling (Fig. 3b) then implies ~11.5 µV at the higher intensity and SLSSE ≈ 200 nV/K. This is an internal inconsistency, not an outside-consensus disagreement; it can be settled by reporting the raw voltage at the AM intensity. The correct fix likely leaves the qualitative mechanism unchanged but changes the numerical coefficient and the claimed precision. Therefore the verdict remains CONDITIONAL, with the conditions expanded to include reconciling the VISHE/intensity pairing.","tokens_in":16322,"tokens_out":15775,"duration_ms":163770,"concrete_test":"Check the raw 405 nm power-series data for the device used in Sec. II F: identify the measured VISHE at 1190 W/cm² on an 80 µm spot. If it is ≈11.5 µV rather than 3.457 µV, recompute SLSSE = (VISHE/80 µm)/(11 K/15 µm); the expected value is ≈200 nV/K, which would make the stated 60 ± 7.8 nV/K unsupported. If it is truly 3.457 µV, verify which power and device produced the Appendix B value and correct the PSV comparison accordingly.","verdict_should_be":"UNCHANGED","load_bearing_attack":"In Sec. II F, the paper states that 1190 W/cm² illumination at 405 nm produces an 11 K optically induced temperature difference across YIG, and that “this intensity generates a VISHE of 3.457 µV across the ~80 µm beam spot size,” leading to SLSSE = 60 ± 7.8 nV/K. Appendix B, however, uses a 405 nm beam with 18 mW and 80 µm diameter, computes I = 3.6 MW/m² (~360 W/cm²), and cites the same “observed signal, 3.457 µV” for a 2 nm Pt device. Since 1190 W/cm² is 3.3× larger than 360 W/cm², a linear detector (Fig. 3b) would produce ~11.5 µV at the AM intensity. If the correct voltage is 11.5 µV, then EISHE ≈ 143 mV/m and SLSSE ≈ 200 nV/K, not 60 nV/K. If 3.457 µV really was measured at 1190 W/cm², then the Appendix B comparison is mislabeled. Either way, the quantitative calibration, the uncertainty estimate, and the claim of a “precise” ∇T measurement are not self-consistent. The qualitative SSE-based detection claim is not destroyed, but the headline SLSSE value rests on this pairing.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports that Pt/YIG/GGG bilayers produce a photovoltage across a wide wavelength range (390–2200 nm) under illumination, and argues that the signal arises from the longitudinal spin Seebeck effect (LSSE). The authors show that the spectral responsivity is roughly flat and correlates with the Pt absorption coefficient (data shown to 1000 nm), that the photovoltage is independent of light polarization, scales linearly with average power and repetition rate, and matches the thermal dynamics of the Pt film. They introduce a field-amplitude-modulation technique to infer the optically induced temperature difference across the YIG (about 11 K at 1190 W/cm², 405 nm) and combine this with the measured electric field (3.457 µV across 80 µm) to estimate an LSSE coefficient of 60 ± 7.8 nV/K. They also estimate the competing photo-spin-voltaic effect to be negligible. The paper claims a new broadband optical detection mechanism based solely on spin-current generation and detection.","tokens_in":16570,"tokens_out":8985,"duration_ms":85417,"significance":"If the conclusions hold, this work demonstrates a conceptually new type of broadband photodetector that operates via thermally generated spin currents rather than photocarriers, with potential advantages for infrared detection and spin-caloritronic devices. The field-modulation null method is a useful addition to the spin-Seebeck measurement toolkit, and the paper includes a commendable set of cross-checks: polarization independence, power linearity, repetition-rate insensitivity, front/back illumination symmetry, edge illumination, Pt-removal spatial mapping, and the null-point shift itself. However, the quantitative calibration of the reported LSSE coefficient contains an internal inconsistency with Appendix B, and the thermal-resistance assumption underlying the temperature-gradient measurement is not quantified. These issues put the headline numerical value on shaky ground, although the qualitative spin-Seebeck-based detection claim is supported by the ensemble of measurements.","major_comments":[{"comment":"The reported calibration of the LSSE coefficient is internally inconsistent. In §II F, the authors state that an intensity of 1190 W/cm² (405 nm) generates a VISHE of 3.457 µV and use this to obtain EISHE = 43 mV/m and SLSSE = 60 ± 7.8 nV/K. Appendix B, however, uses a 405 nm beam with 18 mW and 80 µm diameter (intensity ≈ 3.6 MW/m² = 360 W/cm²) and cites the same observed signal, 3.457 µV, for the 2 nm Pt device. Since 1190 W/cm² is a factor of 3.3 larger than 360 W/cm², a linear detector (Fig. 3b) would produce roughly 11.5 µV at the higher intensity, leading to EISHE ≈ 143 mV/m and SLSSE ≈ 200 nV/K. If the authors instead measured 3.457 µV at 360 W/cm², then the statement in §II F that 'this intensity' (1190 W/cm²) generates 3.457 µV is incorrect. The manuscript must clarify the exact power/intensity used in each measurement, reconcile the two pairs, and recompute the LSSE coefficient and its uncertainty accordingly.","section":"§II F and Appendix B"},{"comment":"The field-modulation null method rests on the assumption that TGGG ≈ TCu, both with and without illumination. In the light-off state, there is no heat flow, so TGGG = TCu is reasonable. In the light-on state, however, heat flows from the Pt through the YIG and GGG to the copper sink, and any thermal resistance between GGG and the sink makes TGGG exceed TCu. The zero-crossing shift then equals ΔT_optical minus QR_th, not ΔT_optical itself, so the inferred 11 K and the derived SLSSE would be systematically biased. The claim that 'the only thing changing between the curves is the presence of an additional heating source' is therefore incomplete. The sensitivity of the result to the GGG–sink thermal resistance should be quantified with a thermal model or a direct temperature measurement at the GGG surface, and this contribution should be included in the uncertainty budget.","section":"§II F, Eq. (for ∇T)"},{"comment":"The abstract states that the device responsivity 'closely following the Pt absorption coefficient' across 390–2200 nm, but the Pt absorption coefficient is only measured and displayed from 390 to 1000 nm. The agreement between the spectral response and α_Pt is therefore demonstrated only over part of the claimed range. To support the broadband-correlation claim, the authors should either extend the absorption-coefficient measurement to 2200 nm or clearly state that the correlation is established only in the 390–1000 nm range and that the flat responsivity at longer wavelengths is consistent with an expected flat α_Pt.","section":"Abstract and Fig. 3(e)"}],"minor_comments":[{"comment":"The statement that the responsivity drop is 'less than a factor of 100' over four orders of magnitude of incident power appears inconsistent with the claim of linear power scaling in Fig. 3(b). If the voltage were strictly linear, the responsivity would be constant. Please clarify the power range over which linearity holds and reconcile the apparent sublinearity in Fig. 3(c).","section":"Fig. 3(c) and Sec. II B"},{"comment":"In Sec. II D the authors state that 'we still cannot conclusively determine whether the SSE and/or PSV effect is the dominant spin-current generation mechanism,' but later in Sec. II F and the Conclusion they conclude that the signal comes predominantly from the bulk SSE. This wording is confusing; please revise to reflect the evidence presented in Appendix B.","section":"Sec. II D vs. Sec. II F/Conclusion"},{"comment":"The symbol 'SSSE' appears in the sentence 'theoretical models of how SSSE varies with Pt thickness'; this appears to be a typo for the spin Seebeck coefficient (SLSSE or SSE).","section":"Sec. II C"},{"comment":"The spin Hall angle is stated to range 'from 10^-4 to 100'; this likely should be 10^-4 to 1.0 (or another intended bound). Please correct the typographical representation of the exponent.","section":"Appendix B"}],"recommendation":"major_revision","confidential_remarks":"This is a potentially interesting contribution, but the quantitative inconsistency between §II F and Appendix B is a serious calibration error that must be resolved before the numerical claim of SLSSE = 60 nV/K can be taken seriously. I recommend a major revision focusing on: (1) reconciling the voltage–intensity pairing, (2) quantifying the thermal-resistance bias in the null-point method, and (3) either extending the Pt absorption measurement or softening the 'closely following' claim. The qualitative demonstration of broadband spin-Seebeck detection appears sound and would survive these corrections."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Kolya—quick take on arXiv:1908.10472. The paper reports a Pt/YIG device that detects light from 390 to 2200 nm via the spin Seebeck effect, with a nearly flat spectral responsivity. The core evidence is strong: signal scales linearly with power, is independent of polarization and pulse repetition rate, follows Pt heating dynamics, responds symmetrically to front/back illumination, and vanishes when Pt is removed or disconnected. The field-modulation technique for locking onto the SSE derivative is genuinely new and useful. This is a real experimental contribution, not a hype-driven claim.\n\nThe soft spot is the quantitative calibration of SLSSE. Section II F states that 1190 W/cm^2 at 405 nm produces an 11 K temperature difference and 'this intensity generates a VISHE of 3.457 µV.' Appendix B, however, uses 18 mW with an 80 µm beam, computes 360 W/cm^2, and cites the same 3.457 µV as the observed signal at that intensity. 1190 W/cm^2 is 3.3× larger than 360 W/cm^2. Since the detector is linear in power (Fig. 3b), the 3.457 µV cannot correspond to both intensities unless something else changed (device thickness, spot size, etc.). If the 3.457 µV was actually measured at 1190 W/cm^2, then the Appendix B comparison is mislabeled and the observed voltage at 360 W/cm^2 would be ~1 µV. If it was measured at 360 W/cm^2, then the 11 K calibration was obtained at a different intensity and the quoted SLSSE = 60 nV/K is off by roughly 3.3×, yielding ~200 nV/K. Either way, the paper needs to untangle this before the coefficient is taken seriously. This is not a fatal flaw for the qualitative claim—the SSE mechanism is supported by multiple independent measurements—but it does undermine the headline number and the 'precise' ∇T language.\n\nAlso worth noting: no error bars are shown on the central spectral responsivity data, and the match to Pt absorption is only plotted to 1000 nm. The PSV estimate in Appendix B is an order-of-magnitude calculation with assumed parameters, but that's a secondary support, not the main evidence.\n\nWho this is for: anyone working on spin caloritronics, spintronic photodetection, or thermal measurement techniques. The field-modulation method alone is worth a look. The paper deserves peer review—a good referee will catch the calibration inconsistency and the authors can fix it with clearer reporting of which intensity corresponds to which voltage. I'd engage with it, but I wouldn't quote the SLSSE value until it's rederived.","headline":"Solid broadband SSE photodetection demonstration with an internal inconsistency in the SLSSE calibration that likely inflates or misassigns the reported coefficient.","tokens_in":17199,"tokens_out":3165,"would_cite":true,"duration_ms":28222,"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":"Spin currents convert absorbed light into a flat photovoltage from 390 to 2200 nm.","keywords":["spin Seebeck effect","inverse spin Hall effect","broadband photodetection","yttrium iron garnet","platinum thin films","spin caloritronics","thermal gradient calibration","photo-spin-voltaic effect"],"falsifier":"Measure the backside GGG temperature directly with a calibrated sensor while sweeping the heat-sink temperature under the same 1190 W/cm² illumination, and compare the zero-crossing temperature of the field-modulated signal with the measured bottom-surface temperature; if they differ, the 11 K optical heating and ∇T = 0.73 K/µm are over- or underestimated.","tokens_in":16069,"feed_emoji":"💡","tokens_out":4519,"duration_ms":43019,"temperature":0.7,"pith_summary":"This paper claims that a platinum-on-yttrium-iron-garnet bilayer detects light across 390–2200 nm by converting optical heating into a spin current rather than by creating mobile photocarriers. The photovoltage is generated in the longitudinal spin Seebeck geometry and read out through the inverse spin Hall effect in the platinum layer, and the measured responsivity is nearly flat across the whole range because it follows the absorption coefficient of platinum. The authors introduce a field-modulation calibration that lets them infer the optically induced temperature difference across the YIG (about 11 K at 1190 W/cm²) and estimate the longitudinal spin Seebeck coefficient as 60 ± 7.8 nV/K. If correct, this establishes a thermally driven, spin-based broadband photodetector that does not depend on semiconductor band gaps.","feed_headline":"Spin currents convert light to voltage from 390 to 2200 nm","feed_subtitle":"A Pt/YIG detector converts absorbed light into a thermal spin current, with responsivity set by platinum absorption.","key_machinery":"The load-bearing mechanism is the longitudinal spin Seebeck effect: a temperature gradient across the magnetic insulator YIG drives a pure spin current into an adjacent platinum film, and the inverse spin Hall effect in the platinum converts that spin current into a transverse electric field. The named detection relation is E_ISHE = (1/σ_c) D_ISHE (J_s × σ), with the spin polarization σ set by the in-plane magnetic field. The paper's new experimental tool is a field-modulation (amplitude-modulated) measurement in which a small ≈1 G modulation at 13.1 Hz on a swept field gives a derivative signal whose zero marks ∇T = 0, allowing the optical heating contribution to be read as the lateral shift between light-on and light-off curves. This machinery connects the measured photovoltage dynamics directly to platinum thermal response and supports the inference that the bulk SSE, not photocarrier creation, is the detection channel.","core_discovery":"The central claim is that the voltage observed when light strikes Pt/YIG/GGG devices is the inverse spin Hall voltage from a thermally generated spin current: absorbed light heats the platinum, a temperature gradient across the YIG launches a spin current, and the platinum converts that spin current into a measurable voltage. The evidence is spectral, dynamic, and geometric: responsivity tracks the Pt absorption coefficient and is nearly featureless from 390 to 2200 nm; the signal is independent of polarization and pulse repetition rate; the temporal response matches platinum heating; the signal disappears where the Pt layer is discontinuous; and edge illumination of the GGG produces almost no response. Using a 13.1 Hz field modulation and a swept heat-sink temperature, the authors locate zero thermal gradient with and without illumination and infer an optical temperature rise of about 11 K, giving ∇T ≈ 0.73 K/µm and SLSSE ≈ 60 ± 7.8 nV/K. They further argue that the photo-spin-voltaic mechanism is negligible because its estimated contribution (~4.9 nV) is three orders of magnitude below the measured signal.","pith_inferences":["If the detection is purely thermal, the spectral range is bounded only by Pt absorption and not by any electronic resonance, so the same architecture could be tested at longer infrared wavelengths or with different high-absorption metals.","The calibration's dependence on the assumption T_GGG ≈ T_Cu could be tested by inserting a known thermal resistance between the GGG and heat sink; a shift of the zero crossing would quantify the systematic error.","A follow-on direct comparison of measured Pt film temperature rise (via resistance) with the 11 K inferred gradient would separate bulk thermal diffusion effects from interfacial Kapitza resistance and sharpen the bulk-versus-interfacial SSE assignment."],"forward_implications":["A single Pt/YIG detector can respond uniformly from 390 to 2200 nm, spanning the silicon and InGaAs ranges without a semiconductor band-gap cutoff.","Because the response follows platinum absorption, extending the operating range further into the infrared is plausible whenever the Pt film still absorbs.","The field-modulation method gives a direct, sample-specific calibration of optical heating, which can be used to extract SSE coefficients in other metal/insulator bilayers.","Thinner platinum films yield higher responsivity (more than an order of magnitude from 6 to 2 nm), so interface and thickness engineering are the levers for improving sensitivity.","The estimated SSE coefficient is consistent with prior Pt/YIG measurements, and published routes for enhancing SSE efficiency imply larger photovoltages should be attainable."],"supporting_citations":[{"why":"Supplies the photo-spin-voltaic baseline whose estimated contribution must be ruled out and whose device preparation is reused.","marker":"[18]"},{"why":"Establishes the longitudinal spin Seebeck effect in magnetic insulators, the geometry used throughout.","marker":"[15]"},{"why":"Provides the inverse spin Hall effect detection relation that converts the spin current into voltage.","marker":"[8]"},{"why":"Supports the field-dependent signal analysis and prior time-resolved LSSE imaging for Pt/YIG.","marker":"[21]"},{"why":"Provides picosecond interfacial SSE timescales used to argue the steady-state signal is not interfacial.","marker":"[23]"},{"why":"Supplies bulk magnon transport dynamics that set the expected timescale for the bulk SSE.","marker":"[29]"},{"why":"Shows how laser-induced temperature gradients in ultrathin magnetic layers should be modeled, motivating the AM calibration.","marker":"[42]"},{"why":"Documents methods for evaluating thermal gradients in longitudinal spin Seebeck measurements, supporting the S_LSSE estimate.","marker":"[48]"}],"fun_headline_variants":["Spin Seebeck effect detects light from 390 to 2200 nm","Pt/YIG spin Seebeck detector covers 390-2200 nm","Broadband light sensing via spin current heat","Spin current converts absorbed light to voltage","Thermal spin Seebeck effect yields broadband photodetector"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The field-modulation calibration assumes the GGG backside sits at the copper heat-sink temperature and that the unilluminated platinum surface is at 20 °C, so the zero of the integrated amplitude-modulated signal marks zero temperature gradient across the YIG; a thermal resistance between GGG and the heat sink, or a different light-off Pt temperature, would shift the inferred optical heating and the 60 nV/K coefficient.","fun_headline_variants_meta":{"raw":{"variants":["Spin Seebeck effect detects light from 390 to 2200 nm","Pt/YIG spin Seebeck detector covers 390-2200 nm","Broadband light sensing via spin current heat","Spin current converts absorbed light to voltage","Thermal spin Seebeck effect yields broadband photodetector"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000653,"raw_usage":{"total_tokens":2988,"prompt_tokens":934,"completion_tokens":2054,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":550,"completion_tokens_details":{"reasoning_tokens":1971}},"tokens_in":550,"tokens_out":2054,"duration_ms":16310,"temperature":1.0,"reasoning_tokens":1971,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T10:42:23.920549+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the backside GGG temperature directly with a calibrated sensor while sweeping the heat-sink temperature under the same 1190 W/cm² illumination, and compare the zero-crossing temperature of the field-modulated signal with the measured bottom-surface temperature; if they differ, the 11 K optical heating and ∇T = 0.73 K/µm are over- or underestimated.","supporting_citations":[{"cited_title":"Imaging magnetization structure and dynamics in ul- trathin Y3Fe5O12/Pt bilayers with high sensitivity using the time-resolved longitudinal spin Seebeck effect,","cited_arxiv_id":null,"evidence_quote":"Supports the field-dependent signal analysis and prior time-resolved LSSE imaging for Pt/YIG."},{"cited_title":"Photo-spin-voltaic effect,","cited_arxiv_id":null,"evidence_quote":"Supplies the photo-spin-voltaic baseline whose estimated contribution must be ruled out and whose device preparation is reused."},{"cited_title":"Observation of longitu- dinal spin-Seebeck effect in magnetic insulators,","cited_arxiv_id":null,"evidence_quote":"Establishes the longitudinal spin Seebeck effect in magnetic insulators, the geometry used throughout."},{"cited_title":"Con- version of spin current into charge current at room tem- perature: Inverse spin-Hall effect,","cited_arxiv_id":null,"evidence_quote":"Provides the inverse spin Hall effect detection relation that converts the spin current into voltage."},{"cited_title":"Picosecond spin Seebeck effect,","cited_arxiv_id":null,"evidence_quote":"Provides picosecond interfacial SSE timescales used to argue the steady-state signal is not interfacial."},{"cited_title":"Role of bulk- magnon transport in the temporal evolution of the lon- gitudinal spin-Seebeck effect,","cited_arxiv_id":null,"evidence_quote":"Supplies bulk magnon transport dynamics that set the expected timescale for the bulk SSE."},{"cited_title":"Precise determination of the temperature gradients in laser-irradiated ultrathin magnetic layers for the analysis of thermal spin current,","cited_arxiv_id":null,"evidence_quote":"Shows how laser-induced temperature gradients in ultrathin magnetic layers should be modeled, motivating the AM calibration."},{"cited_title":"Evaluation of thermal gradients in longitudinal spin Seebeck effect mea- surements,","cited_arxiv_id":null,"evidence_quote":"Documents methods for evaluating thermal gradients in longitudinal spin Seebeck measurements, supporting the S_LSSE estimate."}],"review_version":1}