{"id":"cf1d8b74-7684-4990-ab70-db227aa069dc","arxiv_id":"2411.13751","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A 16 GHz ScAlN-on-SiC solidly mounted resonator reports Qm=380, kt2=4.5%, FOM=17, claimed as the best Ku-band SMR to date.","lead":"Researchers built a tiny acoustic resonator from scandium-doped aluminum nitride on a silicon carbide substrate that vibrates at about 16 GHz. It achieves a high mechanical quality factor of 380 and strong electromechanical coupling of 4.5%, which the authors say is the best reported for solidly mounted resonators in the Ku band, potentially enabling compact high-power RF filters.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No independent validation of the MBVD-fitted Qm and kt2; the record-KPI claim rests on a single un-deembedded fit with no residual or repeatability data.","rationale":"The reader's weakest assumption and my concern are the same core issue: the MBVD fit is not independently validated and could bias Qm, kt2, and FOM. I agree with the reader that this is the most load-bearing assumption, and the paper provides no calibration, de-embedding, residual, or repeatability information. I mark agreement as 'partial' because I place additional weight on the fact that the headline KPI numbers are aggregated from different devices and extraction methods (Y12 MBVD fit versus matched Smith-chart Bode extraction), and on the absence of any comparison table supporting 'highest KPIs.' These are not fatal flaws; device letters commonly report best devices. However, they make the central claim sensitive to one unverified fit. The proposed test would settle the issue: independent de-embedding and a model-free Q extraction would either confirm the reported values or reveal a systematic bias. Secondary concerns include the claim of power handling 'greater than 20 dBm' when the sweep stops at the 20 dBm setup limit, and the lack of error bars; these are addressable but do not change the conditional verdict.","tokens_in":8357,"tokens_out":3145,"duration_ms":33621,"concrete_test":"Request the calibrated, de-embedded raw S-parameters for the devices in Fig. 4 and re-run the MBVD fit using an independently measured pad/open/short dummy to set the parasitic branches. Then compare the fitted Qm, kt2, and FOM with the paper's values, and also compute Qm from the same de-embedded data by a model-independent method, for example 3-dB bandwidth on Y12 and a circle fit on the Smith chart. If the de-embedded values differ by more than 10% from the reported values, or if the MBVD residual shows systematic deviation near resonance, the headline record claim is not supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that S2MRs show the highest KPIs among Ku-band solidly mounted resonators requires the reported Qm=380, QBode~500, kt2=4.5%, and FOM~17 to be unbiased intrinsic metrics. Section IV says only that 'equivalent circuit parameters are extracted by fitting the transmission admittance (Y12) response to a Modified Butterworth-Van Dyke (MBVD) equivalent model.' No VNA calibration, probe-tip reference plane, pad de-embedding, fit residuals, or uncertainty are reported. MBVD models include parasitic branches; if pad/probe capacitance or resistance is absorbed into the motional branch, Qm and kt2 can be inflated. The concern is amplified because the headline numbers come from different devices: Fig. 4a shows Q3dB=277, kt2=4%, FOM~14 for one device; Section IV reports Qm=380, Q3dB=216, kt2=4.5%, FOM=17 for a separate device 'most closely matched to 50 Ω'; and QBode=500 is extracted from a matched/smoothed Smith-chart loop rather than the raw Y12 fit. Without cross-device statistics or a comparison table, the record claim rests on a single best-case fit that cannot be checked from the paper alone.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This letter reports solidly-mounted bidimensional mode resonators (S2MRs) fabricated in 30% ScAlN on 6H-SiC and operating near 16 GHz. The design is based on COMSOL FEA optimization of a Sezawa mode with h/λ = 0.65 and tm/λ = 0.125. The fabrication uses a 230 nm ScAlN layer on a 20 nm AlN seed, e-beam patterned AlSiCu IDTs with 100 nm fingers, and Au pads. RF characterization of two-port devices uses a Keysight P5008A VNA; MBVD fitting to Y12 gives Qm = 380, kt2 = 4.5%, and FOM = 17, while Q3dB = 277 and 216 are reported on different devices and QBode ≈ 500 is extracted from a matched Smith-chart loop. Power sweeps to 20 dBm and TCF of -95 ppm/K are also reported. The central claim is that these are the highest KPIs among solidly mounted resonators in the Ku band.","tokens_in":8607,"tokens_out":5605,"duration_ms":48120,"significance":"If the reported numbers are reliable, this is a meaningful advance for Ku-band acoustic filtering: a FOM exceeding 17 at 16 GHz on SiC, with good film crystallinity, a measured TCF, and a plausible path to integration with SiC power electronics. Strengths include the independent FEA design using a priori ScAlN parameters, the standard MBVD extraction procedure, quantitative film-quality characterization (AOG 0.15%, XRD FWHM), and explicit power-sweep and temperature measurements. I find no circularity: the FEA design does not assume the measured outcome, and the MBVD fit is a standard extraction. The significance, however, hinges on the unvalidated assumption that the extracted Qm and kt2 are unbiased intrinsic values, and on a fair quantitative comparison with prior SMR work.","major_comments":[{"comment":"The headline metrics are not from a single device: Fig. 4a reports Q3dB = 277, kt2 = 4%, and FOM ≈ 14 for the device with the largest Q3dB, while Qm = 380, Q3dB = 216, kt2 = 4.5%, and FOM = 17 are reported for a separate '50 Ω-matched' device in Fig. 4c, and QBode = 500 is extracted from the Fig. 4a device. The abstract combines these into a single set of KPIs. Please report the full metric set for each device, the number of devices measured, and the device-to-device spread (e.g., mean ± standard deviation). Without this, the 'highest KPIs' claim rests on a single best-case fit that cannot be assessed.","section":"IV, Fig. 4a-c"},{"comment":"The paper states that equivalent circuit parameters are extracted by fitting the Y12 response to an MBVD model, but it does not describe the VNA calibration (e.g., SOLT), the reference plane, pad/probe de-embedding, or the fit residuals/uncertainty. Because Y12 includes pad and interconnect parasitics, an un-deembedded fit can absorb parasitic capacitance or resistance into the motional branch and bias Qm and kt2. Please provide the calibration/de-embedding procedure, the fitted MBVD element values, and residual or confidence information for the reported Qm = 380 and kt2 = 4.5% values.","section":"IV, MBVD fit"},{"comment":"The abstract claims 'power handling greater than 20 dBm,' but the experiment sweeps only to the maximum setup power of 20 dBm and observes no deterioration. This supports 'tested to 20 dBm' or '≥20 dBm,' not 'greater than 20 dBm.' Please either add data beyond 20 dBm or revise the wording to match the measurement.","section":"Abstract and IV, power handling"},{"comment":"The claim of 'highest KPIs among solidly mounted resonators in the Ku band' is not backed by a quantitative comparison. The text cites prior SMR work [25]-[28] but gives no table or list of their Q, kt2, and FOM at comparable frequencies. Please add a comparison table with the same metric definitions and cite the specific numbers from each prior work, so the record claim can be verified.","section":"V, Conclusion and I, Introduction"}],"minor_comments":[{"comment":"The text says 'a motional quality factor of Qm' without giving its value for the Fig. 4a device; please state Qm for that device so the comparison with the Fig. 4c device is meaningful.","section":"IV, Fig. 4a"},{"comment":"The zoomed-in view of the peak region (blue box) is not clearly visible in the figure as printed; please enlarge the inset or annotate it more distinctly.","section":"IV, Fig. 4c"},{"comment":"'Fig. 4c show the admittance response' should be 'Fig. 4c shows the admittance response.'","section":"IV"},{"comment":"'K u-band' appears to be broken by a markup artifact; it should read 'Ku-band.'","section":"Abstract"},{"comment":"The phrase 'possessing similar quality factor (Qm = 380 and Q3dB = 216)' is ambiguous because Q3dB = 216 is not similar to the Q3dB = 277 of the Fig. 4a device; please specify which quality factor is being compared.","section":"IV"}],"recommendation":"major_revision","confidential_remarks":"I recommend major revision rather than rejection because the underlying approach is sound and the missing items (calibration/de-embedding details, device statistics, a comparison table, and corrected power-handling wording) are within the scope of a revision. I see no indication of novelty or attribution problems."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe paper reports a genuine new device result: a 16 GHz Sezawa-mode S2MR on ScAlN-on-SiC with Qm around 380, kt2 around 4.5%, and FOM around 17. That is a new frequency/substrate combination and a real step beyond the cited 6 GHz AlScN-on-SiC SAW. The fabrication characterization (AOG, XRD) is solid, and the power sweep to 20 dBm with return-to-baseline is useful. The FEA co-optimization of h/λ and tm/λ is standard but competently done.\n\nThe soft spots are mostly in the reporting. The \"highest KPIs among solidly mounted resonators in Ku band\" claim is asserted without a comparison table. The headline numbers come from different devices: one gives Q3dB=277, kt2=4%, FOM~14; another gives Qm=380, kt2=4.5%, FOM=17; and QBode=500 comes from a matched/smoothed Smith-chart loop. No calibration, de-embedding, or fit-residual information is given for the MBVD extraction, so we cannot check whether pad parasitics are absorbed into the motional branch. And the abstract claims power handling \"greater than 20 dBm\" while the setup maximum is exactly 20 dBm—that overstatement should be fixed.\n\nNone of this is fatal. The central device result is plausible and the gaps are readily addressable. I would send it to peer review rather than desk reject, asking for device-to-device statistics, calibration/de-embedding details, fit residuals, a normalized comparison table, and corrected power-handling wording. The reader's conditional verdict is about right.\n\nRecommendation: worth a serious referee, with expectation of major/minor revision.","headline":"A plausible new 16 GHz ScAlN-on-SiC S2MR data point that overstates its record claim and power-handling number, but the core device results are worth refereeing.","tokens_in":9191,"tokens_out":2215,"would_cite":true,"duration_ms":22246,"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":"Solidly mounted ScAlN-on-SiC resonators reach a 16 GHz Sezawa mode with Qm of 380, coupling of 4.5%, and power handling above 20 dBm, setting a Ku-band record among solidly mounted resonators.","keywords":["ScAlN","Sezawa mode","solidly mounted resonator","Ku band","quality factor","electromechanical coupling","radio-frequency MEMS","modified Butterworth-Van Dyke"],"falsifier":"A calibrated two-port measurement of the same devices with on-wafer de-embedding of pad and interconnect parasitics, followed by an MBVD fit with residuals reported, would settle whether $Q_m = 380$ and $k_t^2 = 4.5\\%$ survive; a direct comparison of the extracted motional parameters against a finite-element model of the full electrode layout would also expose any discrepancy.","tokens_in":8173,"feed_emoji":"📡","tokens_out":10370,"duration_ms":78793,"temperature":0.7,"pith_summary":"This paper reports solidly mounted bidimensional-mode resonators (S2MRs) that operate near 16 GHz, built from a 30% scandium-doped aluminum nitride film on a silicon carbide substrate. The authors claim these devices achieve the highest key performance indicators of any solidly mounted resonator in the Ku band: a mechanical quality factor of $Q_m = 380$, a Bode quality factor near $Q_{\\mathrm{Bode}} \\approx 500$, an electromechanical coupling coefficient of $k_t^2 = 4.5\\%$, and a figure of merit exceeding 17. If the claim is correct, these resonators offer a compact, high-power platform for RF filtering in 5G/6G, satellite communications, and military systems, and they demonstrate that nanoacoustic devices can be integrated on fast substrates alongside high-power electronics.","feed_headline":"16 GHz resonators hit record Q of 380, coupling 4.5%","feed_subtitle":"ScAlN-on-SiC Sezawa-mode devices beat all solidly mounted resonators in the Ku band, enabling high-power filters.","key_machinery":"The S2MR is a solidly mounted resonator in which a thin ScAlN film on a high-velocity SiC substrate sustains a slow-on-fast Sezawa mode: a dispersive surface acoustic wave with displacement in both the lateral and thickness directions, excited by interdigitated transducers. The design is carried by a co-optimized ratio of film thickness to acoustic wavelength ($h/\\lambda = 0.65$) and electrode thickness to wavelength ($t_m/\\lambda = 0.125$), which finite-element simulations show maximizes electromechanical coupling. The mechanical quality factor, coupling, and figure of merit are extracted by fitting the measured $Y_{12}$ admittance to a Modified Butterworth-Van Dyke equivalent circuit, whose fit is the load-bearing step connecting raw measurements to the headline numbers.","core_discovery":"The paper's central claim is that a ScAlN-on-SiC S2MR operating in a Sezawa mode reaches a previously unreported combination of quality factor, coupling, and power handling at roughly 16 GHz, with the best device showing $Q_m = 380$, $Q_{\\mathrm{Bode}}$ near 500, $k_t^2 = 4.5\\%$, FOM = 17, and power handling above 20 dBm. The authors argue that this record performance follows from co-optimizing the piezoelectric film thickness and the electrode thickness relative to the acoustic wavelength, guided by finite-element simulations, together with a high-quality, low-defect ScAlN film. They position the result as the strongest reported key performance indicators for solidly mounted resonators in the Ku band.","pith_inferences":["If the MBVD fit were reported with residuals or a de-embedding of pad parasitics, the record claim would be easier to verify; without it, some uncertainty about the true Qm and coupling remains.","The same slow-on-fast design could be transferred to even faster substrates such as diamond, likely pushing operating frequencies further into the millimeter-wave range while retaining high Q.","The finite-element trend of superlinear kt2 growth with scandium doping suggests that close-to-40% doping could push the figure of merit above 20, provided the mechanical losses do not rise proportionally.","The methodology of co-optimizing h/λ and tm/λ for a targeted mode could be applied to other bidirectional modes, enabling design trade-offs between frequency, coupling, and power handling."],"forward_implications":["If the reported values hold, S2MRs become a leading candidate for Ku-band RF filters with low insertion loss and high selectivity in a compact footprint.","The demonstrated power handling beyond 20 dBm makes the platform suitable for base-station and satellite front ends where large signals are common.","The lithographic frequency scaling, noted by the authors, could produce multi-frequency filter banks on a single chip for 5G/6G radios.","The ScAlN-on-SiC stack offers a path toward monolithic co-integration of acoustic filters with SiC high-power electronics, removing interconnects and module losses."],"supporting_citations":[{"why":"Supplies the material parameters for scandium-doped aluminum nitride used in the finite-element simulations of coupling and velocity.","marker":"[29]"},{"why":"Provides the Modified Butterworth-Van Dyke equivalent circuit model used to extract quality factor and coupling from the measured admittance.","marker":"[35]"},{"why":"Reports an 18 GHz solidly mounted resonator in scandium aluminum nitride on a Bragg reflector, serving as the closest prior comparison for the record KPI claim.","marker":"[25]"},{"why":"Demonstrates solidly mounted two-dimensional guided modes in scandium aluminum nitride on sapphire, establishing the S2MR concept that this paper extends to SiC.","marker":"[26]"},{"why":"Shows near-6 GHz Sezawa-mode resonators using AlScN on SiC, the direct platform precursor for the 16 GHz devices reported here.","marker":"[21]"},{"why":"Provides the Sezawa-mode dispersion and low-loss behavior in a GaN/SiC platform that motivates the mode choice at super-high frequencies.","marker":"[22]"}],"fun_headline_variants":["ScAlN-on-SiC S2MRs set Ku-band record: Qm=380, kt2=4.5%","16 GHz Sezawa-mode S2MRs on SiC hit Qm 380, kt2 4.5%","Best solidly mounted resonators in Ku band: ScAlN S2MRs on SiC","S2MRs on SiC reach Qm 380 and kt2 4.5% at 16 GHz"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The record numbers depend on the Modified Butterworth-Van Dyke fit to the measured admittance cleanly separating the motional resonance from pad parasitics and spurious modes; if that separation is imperfect, the extracted quality factor and coupling could be overstated.","fun_headline_variants_meta":{"raw":{"variants":["ScAlN-on-SiC S2MRs set Ku-band record: Qm=380, kt2=4.5%","16 GHz Sezawa-mode S2MRs on SiC hit Qm 380, kt2 4.5%","Best solidly mounted resonators in Ku band: ScAlN S2MRs on SiC","S2MRs on SiC reach Qm 380 and kt2 4.5% at 16 GHz"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000905,"raw_usage":{"total_tokens":3866,"prompt_tokens":891,"completion_tokens":2975,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":507,"completion_tokens_details":{"reasoning_tokens":2861}},"tokens_in":507,"tokens_out":2975,"duration_ms":18326,"temperature":1.0,"reasoning_tokens":2861,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T15:56:32.061319+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A calibrated two-port measurement of the same devices with on-wafer de-embedding of pad and interconnect parasitics, followed by an MBVD fit with residuals reported, would settle whether $Q_m = 380$ and $k_t^2 = 4.5\\%$ survive; a direct comparison of the extracted motional parameters against a finite-element model of the full electrode layout would also expose any discrepancy.","supporting_citations":[{"cited_title":"Piezoelectric coef- ficients and spontaneous polarization of ScAlN,","cited_arxiv_id":null,"evidence_quote":"Supplies the material parameters for scandium-doped aluminum nitride used in the finite-element simulations of coupling and velocity."},{"cited_title":"Modified Butterworth-Van Dyke circuit for FBAR resonators and automated mea- surement system,","cited_arxiv_id":null,"evidence_quote":"Provides the Modified Butterworth-Van Dyke equivalent circuit model used to extract quality factor and coupling from the measured admittance."},{"cited_title":"Solidly Mounted Two-Dimensional Guided Modes in 30% Scandium Aluminum Nitride on Sapphire,","cited_arxiv_id":null,"evidence_quote":"Demonstrates solidly mounted two-dimensional guided modes in scandium aluminum nitride on sapphire, establishing the S2MR concept that this paper extends to SiC."},{"cited_title":"Super-High- Frequency Low-Loss Sezawa Mode SAW Devices in a GaN/SiC Plat- form,","cited_arxiv_id":null,"evidence_quote":"Provides the Sezawa-mode dispersion and low-loss behavior in a GaN/SiC platform that motivates the mode choice at super-high frequencies."}],"review_version":1}