{"id":"103dc867-1717-4910-aab6-69aef5882a98","arxiv_id":"2412.01449","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":4,"one_line_summary":"A gold/LiNbO3 quasi-BIC metasurface modulates reflected telecom light with 95% depth at ±30 V and 125 MHz bandwidth, tunable by angle of incidence.","lead":"A thin lithium niobate metasurface with gold electrodes reflects telecom-wavelength light whose intensity is switched by an applied voltage, reaching 95% modulation depth at ±30 V with a 125 MHz electrical bandwidth. The resonance is angle-tunable over more than 30 nm, and the authors demonstrate switchable phase contrast imaging and project GHz-class speeds for smaller pixels.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Measured core claims are credible; the weak link is the 39 GHz projection, which assumes pure area-scaling RC and a critical-coupling pixel size that the finite fabricated grating does not satisfy.","rationale":"I agree with the reader's weakest-assumption analysis: the measured core claims are credible and directly supported, while the 39 GHz projection is the least secure load-bearing element. My stress-test sharpens the concern with a concrete order-of-magnitude check based on the stated 90 µm × 60 µm grating area: pure area scaling from the 350 MHz compact electrode to a 22 µm × 22 µm pixel gives about 3.9 GHz, not 39 GHz, unless additional unstated assumptions about electrode geometry and parasitics are invoked. This does not undermine the measured 125/350 MHz results or the demonstrated modulation performance; it only means the 'ultrafast' framing and the spatiotemporal/nonreciprocal promises should be treated as conditional extrapolations. The reader's CONDITIONAL verdict is therefore appropriate, and I would not change it. No raw data or code is provided, but the internal consistency of the static and dynamic measurements and the agreement with finite-grating COMSOL simulations make the central experimental claims reasonably secure.","tokens_in":12440,"tokens_out":7968,"duration_ms":78186,"concrete_test":"Report the exact top/bottom electrode overlap areas and measured parasitic R and C for the 125 MHz and 350 MHz samples, then recompute the projected cutoff for a 22 µm pixel from the same RC model. In parallel, fabricate or simulate a 22 µm pixel electrode with the same 120-period grating and measure the -3 dB electrical cutoff with the same probe and detector. If the pixel cutoff does not approach 39 GHz within about a factor of two, the ultrafast projection fails; if the RC model reproduces the existing 125/350 MHz data and the pixel measurement follows it, the projection lands.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The fabricated-device results (95% relative and 35% absolute modulation at ±30 V, 125/350 MHz cutoffs, angle tuning, phase-contrast imaging) are direct measurements, internally consistent, and I have no sound objection to them. The load-bearing weakness is the 39 GHz 'ultrafast' projection in the Abstract and Discussion. It requires two unverified assumptions. (i) Bandwidth is RC-limited by the electrode-LN-backreflector capacitor and scales inversely with electrode area; no equivalent circuit, electrode dimensions, series/contact resistance, pad capacitance, or probe parasitics are given, so the 125→350→39 GHz chain cannot be checked. A rough area check is worrying: the fabricated grating is 90 µm × 60 µm; shrinking to a 22 µm × 22 µm pixel reduces area by only about 11×, giving roughly 3.9 GHz from 350 MHz, an order of magnitude below 39 GHz unless the compact electrode was much larger than the grating or additional area-reduction of the bottom electrode is assumed. Neither is specified. (ii) The 22 µm pixel itself comes from Eq. (2) with Q_U = 2 Q_L and n_eff = 2, i.e., critical coupling, while the authors state the finite fabricated grating is shifted away from critical coupling (Supplementary Fig. S4). The 39 GHz number is therefore an optimistic extrapolation, not a demonstrated property, and the abstract's 'ultrafast' framing leans heavily on it. The measured 125 and 350 MHz bandwidths are unaffected by this concern.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports reflection-mode electro-optic metasurfaces based on quasi-bound states in the continuum (qBIC) in a gold/lithium-niobate (LN) platform. The authors design, fabricate, and characterize symmetric and asymmetric gold gratings on an 880 nm z-cut LN film with a gold back-reflector. The fabricated device achieves 95% modulation depth (35% absolute modulation of incident power) at ±30 V bias, an electrical -3 dB bandwidth of 125 MHz (350 MHz with a more compact electrode), wavelength tunability over more than 30 nm by varying the angle of incidence, and a proof-of-concept electrically tunable phase-contrast imaging demonstration. The paper also projects a potential 39 GHz bandwidth for a 22 µm pixel size based on an RC-area-scaling argument and a critical-coupling assumption.","tokens_in":12673,"tokens_out":3762,"duration_ms":31097,"significance":"If the measured claims hold, this is a notable advance in free-space electro-optic intensity modulation, combining near-unity relative modulation depth with MHz-scale speed, telecom-wavelength operation, and angle tunability. The central experimental results are direct measurements, are internally consistent with finite-grating simulations, and are well documented in the Methods. The 35% absolute modulation and 125/350 MHz bandwidths are credible and represent a significant step beyond prior ITO-based metasurface modulators. The 39 GHz projection, however, is an extrapolation based on assumptions that are not fully verified, and it currently overstates the demonstrated performance.","major_comments":[{"comment":"The 39 GHz bandwidth projection is not supported by the data presented. The argument assumes that the -3 dB cutoff is set purely by the electrode capacitance and scales inversely with electrode area, but no equivalent circuit, electrode dimensions, series/contact resistances, pad capacitances, or probe parasitics are given. The only experimental points are 125 MHz for the macroscopic electrode and 350 MHz for the compact electrode. A naive area scaling from the stated grating size (90 µm × 60 µm) to a 22 µm × 22 µm pixel would give a cutoff near 3.9 GHz, an order of magnitude below 39 GHz, unless additional area reductions of the bottom electrode are assumed. Since neither the electrode geometry nor the scaling model is specified, the 39 GHz number in the Abstract and Discussion cannot be checked and should be either substantiated with a concrete RC model and electrode layout or removed.","section":"Abstract and Results (Dynamic characterization)"},{"comment":"The 22 µm pixel size used for the 39 GHz projection is derived from Eq. (2) under the assumptions of critical coupling (QU = 2 QL) and an effective mode index of 2. However, the authors state in 'Design and optimization of metasurface' that the fabricated finite grating is shifted away from critical coupling (Supplementary Fig. S4). Because the realized device is not critically coupled, the unloaded Q used in Eq. (2) does not correspond to the measured device, so the resulting pixel size and hence the bandwidth projection are not grounded in the experimental realization. The measured 125 MHz and 350 MHz bandwidths are unaffected by this issue, but the 39 GHz claim relies on an assumption that is known to be violated.","section":"Results (Eq. (2) and pixel-size estimate)"}],"minor_comments":[{"comment":"The rendering of the title and abstract contains missing spaces ('Highlyefficient', 'electro-opticmetasurfaces'); these should be corrected in the final version.","section":"Title and Abstract"},{"comment":"The FOM comparison is made against the authors' own previous value of 0.046. While this is legitimate, placing the new FOM (0.47 and 0.30) in the context of other free-space EO metasurface modulators would strengthen the claim of order-of-magnitude improvement.","section":"Results (Design and optimization of metasurface)"},{"comment":"The measured absolute modulation (35%) is lower than the simulated 50%. The qualitative attribution to finite-grating effects is plausible, but the paper does not quantify how the finite grating reduces the modulation. Extracting an effective Q or modulation from the measured spectrum and comparing it with the finite-grating simulation would make the discussion more rigorous.","section":"Results (Dynamic characterization)"},{"comment":"The statement that the electrical bandwidth is limited by the electrode size is not backed by any dimensions for the macroscopic or compact electrodes. Providing approximate electrode areas and capacitances would allow readers to evaluate the RC scaling more directly.","section":"Results (Dynamic characterization)"}],"recommendation":"major_revision","confidential_remarks":"The measured device performance is credible and the work is likely of interest to the journal. The main weakness is the 39 GHz projection, which is presented as a headline result in the Abstract but rests on an unverified RC-scaling model and a critical-coupling assumption that the fabricated device does not satisfy. The authors should either provide the missing electrical details (electrode dimensions, equivalent circuit) or clearly reframe the 39 GHz figure as a speculative upper bound, revising the Abstract and Discussion accordingly."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe paper delivers: measured 95% modulation depth and 35% absolute modulation at ±30 V in a lithium niobate qBIC metasurface, with a 125 MHz bandwidth, angle tuning over 30 nm, and a phase-contrast imaging demo. The simulated and measured spectra agree well, and the FOM of 0.47 is an order of magnitude over their earlier 0.046. This is a solid experimental advance for free-space electro-optic modulators, not a revolution, but a real step.\n\nThe soft spot is the 39 GHz projection. The abstract and discussion lean on it for the 'ultrafast' framing, but it is an extrapolation the measurements do not support. No equivalent circuit is given, no electrode dimensions, no series resistance or pad capacitance. The 22 µm pixel comes from Eq. 2 with QU = 2 QL and n_eff = 2, i.e. critical coupling, yet the authors themselves say the finite fabricated grating is shifted away from critical coupling (Supplementary Fig. S4). And the scaling check is worrying: the fabricated grating is 90 µm × 60 µm; shrinking to 22 × 22 µm gives about an 11× area reduction, so from the 350 MHz compact-electrode result you would expect roughly 4 GHz, not 39 GHz. Unless the bottom electrode is shrunk as well and the top electrode is much smaller than the grating, which is not stated. The 39 GHz number is optimistic and not demonstrated. It should be qualified or removed.\n\nThe other weaknesses are minor: the gap between simulated 50% and measured 35% absolute modulation is only qualitatively blamed on finite-grating effects, and no raw data or code is deposited. Neither affects the central measured claims. The measurements are direct, internally consistent, and the modulation depth and bandwidth are respectable.\n\nWho is this for? Anyone working on dynamic metasurfaces, electro-optic modulators, or qBIC devices. It deserves a serious peer review. My recommendation: send it out, but ask for a revision that either derives the RC scaling properly or tones down the 39 GHz claim. The paper is worthy of publication with that fix.","headline":"Measured results are credible and useful; the 39 GHz ultrafast projection is unsupported and needs to be toned down or fully derived.","tokens_in":13355,"tokens_out":2927,"would_cite":true,"duration_ms":25581,"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":"A thin lithium niobate metasurface reaches 95% modulation depth at 125 MHz, driven by a ±30 V bias.","keywords":["electro-optic metasurface","quasi-bound states in the continuum","lithium niobate","Pockels effect","free-space intensity modulation","telecom wavelengths","angle-tunable resonance","phase contrast imaging"],"falsifier":"Fabricate a 22 µm-pixel version of the gold-electrode configuration and measure its -3 dB electrical cutoff; if it stays well below 39 GHz, or if the small-pixel reflection dip is not at critical coupling, the projection is refuted. A complementary check is to measure the series resistance of the contact and grating lines to see whether resistance, not geometric capacitance, sets the 125 MHz cutoff of the current device.","tokens_in":12145,"feed_emoji":"⚡","tokens_out":8291,"duration_ms":65937,"temperature":0.7,"pith_summary":"The paper reports a free-space intensity modulator built from a thin lithium niobate film on a gold back-reflector, topped by a grating of gold nanoridges that also serves as the control electrode. The design uses a quasi-bound state in the continuum (qBIC), a very narrow guided-mode resonance, to make reflected light strongly sensitive to the refractive-index change produced by the Pockels effect. The authors claim a measured modulation depth of 95% (35% of the total incident power) at a ±30 V bias, with a -3 dB electrical bandwidth of 125 MHz, at telecom wavelengths around 1550 nm. They further show that the resonance is angle-tunable over more than 30 nm and use the device for electrically switchable phase contrast imaging. If these results hold, the configuration is a practical route to thin, fast, free-space spatial light modulators.","feed_headline":"Thin metasurface hits 95% modulation at 125 MHz","feed_subtitle":"A gold-nanoridge grating on lithium niobate shifts a sharp resonance with ±30 V for fast free-space light control.","key_machinery":"The central object is the quasi-bound state in the continuum (qBIC), a very narrow guided-mode resonance formed in a symmetric gold grating on a thin lithium niobate film when first-order grating coupling ($\\lambda_{\\text{eff}} = \\Lambda$) and second-order Bragg reflection ($2\\Lambda = 2\\lambda_{\\text{eff}}$) hold together. This creates a distributed Bragg resonator whose mode field sits mainly under the gold stripes, producing a sharp reflection dip at about 1550 nm. Applying a voltage between the grating electrode and the gold back-reflector changes the lithium niobate refractive index through the Pockels effect ($\\Delta n \\simeq \\tfrac12 n^3 r_{hk} V/d$), shifting the dip; tuning the angle of incidence moves the whole resonance. The design targets critical coupling, equal scattering and absorption losses, so that the resonance can fully absorb the incoming light and the relative modulation approaches 100%.","core_discovery":"The central claim is that combining a qBIC resonance with the Pockels effect in a lithium niobate-on-gold configuration produces efficient free-space intensity modulation at telecom wavelengths. A symmetric grating of gold ridges excites two counter-propagating waveguide modes via first-order diffraction, and second-order Bragg reflection forms a distributed Bragg resonator; the resulting qBIC appears as a narrow reflection dip at about 1554 nm. A ±30 V bias shifts this dip by roughly 2.5 nm, giving an absolute modulation of 35% and a modulation depth up to 95%. The measured cutoff frequency is 125 MHz for the large top electrode and 350 MHz for a compact electrode, and the authors estimate that a 22 µm pixel would reach 39 GHz if the RC limit scales with electrode area. The same resonance enables angle-tunable operation and a proof-of-concept switchable phase contrast image.","pith_inferences":["Beyond the paper: the 39 GHz number is a projection that assumes the lumped electrode capacitance alone sets the cutoff and that a tiny pixel inherits the critical-coupling condition; neither assumption is experimentally verified here.","Beyond the paper: because the simulated phase swing exceeds 180 degrees at the resonance, a similar structure operated away from the reflection minimum could be used as a phase modulator, not only an intensity modulator.","Beyond the paper: a direct test of the scaling model would be to fabricate a single 22 µm pixel and measure its -3 dB electrical cutoff; any contact-resistance or pad-capacitance limit would show up as a far lower frequency.","Beyond the paper: the grating period and lithium niobate thickness set the qBIC wavelength, so the same design can be scaled to other telecom bands by adjusting the period."],"forward_implications":["Intensity modulation of a free-space beam at telecom wavelengths can be done with an 880 nm lithium niobate layer plus a gold grating, with no external cavity.","The same device acts as an electrically switchable phase-contrast element: at resonance it suppresses paraxial rays and leaves oblique rays nearly untouched.","One fabricated grating covers a 30 nm wavelength range by changing the angle of incidence by about 1 degree, with modulation depth of 30–40%.","The estimated 22 µm pixel size and 39 GHz bandwidth projection, if the RC scaling holds, would put this design in the regime needed for harmonic beam steering and spatiotemporal shaping."],"supporting_citations":[{"why":"It supplies the comb-like electrode configuration and the free-space modulator platform that this design extends.","marker":"[31]"},{"why":"It establishes the nonlocal electro-optic metasurface design procedure and the figure of merit used for performance comparison.","marker":"[38]"},{"why":"It provides the BIC and qBIC physics in waveguide gratings, including the distributed Bragg resonator picture used here.","marker":"[39]"},{"why":"It supplies the lithium niobate Pockels coefficients used to model the voltage-induced index change.","marker":"[32]"},{"why":"It represents the state-of-the-art ITO-based tunable metasurface used as the bandwidth and pixel-size comparison.","marker":"[24]"},{"why":"It states the critical-coupling condition of equal scattering and absorption that the design aims for.","marker":"[43]"},{"why":"It provides the nonlocal angle-selective metasurface principle behind the phase contrast imaging demonstration.","marker":"[46]"}],"fun_headline_variants":["Quasi-BIC metasurface hits 95% modulation at 125 MHz","±30 V tunes metasurface to 95% reflection modulation","Angle-tunable electro-optic metasurface for fast light control","Thin metasurface achieves 95% modulation with 125 MHz bandwidth","Electro-optic metasurface estimated for 39 GHz spatiotemporal control"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The speed projection rests on the assumption that the electrical bandwidth is set only by the electrode's lumped capacitance, so shrinking the electrode to a 22 µm pixel raises the -3 dB cutoff to 39 GHz, with that pixel also assumed to sit at critical coupling with an effective mode index of 2.","fun_headline_variants_meta":{"raw":{"variants":["Quasi-BIC metasurface hits 95% modulation at 125 MHz","±30 V tunes metasurface to 95% reflection modulation","Angle-tunable electro-optic metasurface for fast light control","Thin metasurface achieves 95% modulation with 125 MHz bandwidth","Electro-optic metasurface estimated for 39 GHz spatiotemporal control"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000901,"raw_usage":{"total_tokens":3913,"prompt_tokens":1011,"completion_tokens":2902,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":627,"completion_tokens_details":{"reasoning_tokens":2808}},"tokens_in":627,"tokens_out":2902,"duration_ms":19410,"temperature":1.0,"reasoning_tokens":2808,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T04:22:16.736714+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Fabricate a 22 µm-pixel version of the gold-electrode configuration and measure its -3 dB electrical cutoff; if it stays well below 39 GHz, or if the small-pixel reflection dip is not at critical coupling, the projection is refuted. A complementary check is to measure the series resistance of the contact and grating lines to see whether resistance, not geometric capacitance, sets the 125 MHz cutoff of the current device.","supporting_citations":[{"cited_title":"Electro-optic metasurface- based free-space modulators,","cited_arxiv_id":null,"evidence_quote":"It supplies the comb-like electrode configuration and the free-space modulator platform that this design extends."},{"cited_title":"Nonlocal electro-optic metasurfaces for free-space light modulation,","cited_arxiv_id":null,"evidence_quote":"It establishes the nonlocal electro-optic metasurface design procedure and the figure of merit used for performance comparison."},{"cited_title":"Engineering quasi-bound states in the continuum in asymmetric waveguide gratings,","cited_arxiv_id":null,"evidence_quote":"It provides the BIC and qBIC physics in waveguide gratings, including the distributed Bragg resonator picture used here."},{"cited_title":"Material tensor parameters of LiNbO3 relevant for electro- and elasto-optics,","cited_arxiv_id":null,"evidence_quote":"It supplies the lithium niobate Pockels coefficients used to model the voltage-induced index change."},{"cited_title":"Electrically tunable space–time metasurfaces at optical frequencies,","cited_arxiv_id":null,"evidence_quote":"It represents the state-of-the-art ITO-based tunable metasurface used as the bandwidth and pixel-size comparison."},{"cited_title":"Large-area wide-angle spectrally selective plasmonic absorber,","cited_arxiv_id":null,"evidence_quote":"It states the critical-coupling condition of equal scattering and absorption that the design aims for."},{"cited_title":"Quantitative phase contrast imaging with a nonlocal angle-selective metasurface,","cited_arxiv_id":null,"evidence_quote":"It provides the nonlocal angle-selective metasurface principle behind the phase contrast imaging demonstration."}],"review_version":1}