{"id":"7876cab4-5ed0-47f9-9a42-95ca90a95e8d","arxiv_id":"2509.06199","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"ZST dielectric resonator antennas show stable, high-Q behavior from 296 K to 10 K and can form a 1 mW through-window wireless sensing link, while MCT resonators drift, lose Q, and exhibit hysteresis.","lead":"This paper compares two ceramic microwave antennas at cryogenic temperatures and finds that one material, ZST, stays far more stable than the other, MCT, then uses the stable one to send a wireless signal from a 10 K chamber through a window to detect a room-temperature object. It suggests a path to wireless cryogenic links that could reduce heat load from cables in quantum computers.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Apparent ZST Q-enhancement may be an artifact of temperature-dependent cable loss: S11 is measured with the reference plane outside the cryostat and no de-embedding is reported.","rationale":"The paper's headline comparison is ZST stability versus MCT degradation. The 30 MHz frequency drift and the MCT Q collapse are visible in raw spectra and are less sensitive to cable-loss artifacts. However, the specific quantitative claim that ZST's Q improves by ~20–25% is prominent in the abstract and conclusion, and it is used to argue that ZST is a foundation for high-coherence quantum interfaces. This claim rests on extracting Q_L from the S11 linewidth and β from |S11_min|. Because the reference plane is outside the cryostat, the measured S11 includes the cryogenic cable, whose attenuation falls as the cable cools, causing the resonance to appear deeper and narrower even if the resonator is unchanged. No de-embedding or cryogenic calibration is described. This is therefore the most load-bearing weak point: if it holds, the Q-enhancement claim is an artifact, and the 'foundational material' argument loses a quantitative pillar, though the frequency stability and wireless detection would remain. The reader flagged both Q extraction and thermal hysteresis; I focus on the Q extraction because it affects the positive claim about ZST rather than only the negative MCT characterization. A cryogenic calibration or weakly coupled two-port measurement would settle this directly. If the test shows Q_u still improves with cooling, the paper's claims stand; if not, the manuscript should be revised to remove or reframe the Q-enhancement claim. The current CONDITIONAL verdict remains appropriate, so I recommend no change.","tokens_in":11085,"tokens_out":8921,"duration_ms":110897,"concrete_test":"Perform a temperature-dependent calibration at the cryogenic reference plane: measure the reflection (or the S21 of a thru) of a short/open/load standard mounted at the same location as the resonator from 296 K to 10 K using the same cables. If the round-trip cable loss changes by more than ~0.5 dB over this range, correct the measured S11 spectra accordingly and re-extract Q_L. Alternatively, measure the ZST disk with a weakly coupled two-port transmission probe to obtain Q_u directly; if Q_u at 10 K is not more than ~30% higher than at 296 K, the claimed 20–25% loaded-Q enhancement is an artifact of cable loss rather than a material improvement.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim that ZST's loaded Q-factor improves by 22.6% upon cooling—and the derived unloaded Q increase from ~48 to ~64—is extracted from one-port S11 linewidths and |S11_min|, with the VNA reference plane outside the cryostat. The text states 'Identical VNA settings and reference plane were used across all temperatures' but does not correct for the temperature dependence of the cable between the reference plane and the resonator. Coaxial cable loss decreases as the cable cools, which changes the round-trip attenuation in S11, making a resonance appear deeper and narrower even if the resonator itself is unchanged. The observed dip deepening (from −16.5 dB to −25.3 dB) and linewidth narrowing (from 93.5 MHz to 77.0 MHz) are of precisely the form expected from a few dB of cable-loss reduction. Without a cryogenic calibration or a measured cable S21 over temperature, the Q enhancement and the inferred intrinsic loss improvement are not established. This is load-bearing because the abstract and conclusion use the Q enhancement to argue that ZST is suitable for 'high-coherence quantum interfaces.' A direct test can resolve the ambiguity.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"Torres and Krasnok report a side-by-side cryogenic comparison of MCT and ZST dielectric resonator antennas from 296 K down to 7–10 K. They find that MCT drifts by ~230 MHz, shows ~150 MHz hysteresis at 150 K, and its loaded Q collapses to ~7 at 7 K, whereas ZST shifts by only ~30 MHz, its loaded Q improves by 22.6% (27.6 to 33.9), and no hysteresis is observed. The ZST device is then used as a 10 K through-window antenna at 1 mW input power to detect a water-sphere target at distances up to 4λ, attributed to near-field frequency shifts and far-field S11 magnitude changes. The paper concludes that ZST is a promising cryogenic dielectric for quantum interfaces and wireless links.","tokens_in":11365,"tokens_out":5351,"duration_ms":63099,"significance":"If correct, the work provides a useful material benchmark and a proof-of-concept for low-power cryogenic wireless links. Its strengths are the side-by-side identical-fixture comparison, direct f0(T) and S11(T) measurements, CST design predictions, and the use of window-open/closed baselines to isolate target responses. The τf values are derived from measured frequencies without free parameters. However, the quantitative claims—especially the Q-factor enhancement and inferred intrinsic-loss reduction—rest on one-port S11 measurements without cryogenic calibration or de-embedding, and the near/far-field classification is not self-consistent. The results therefore constitute a promising but not yet fully established comparative study.","major_comments":[{"comment":"The central claim that ZST's loaded Q improves by 22.6% and that its unloaded Q rises from ~48 to ~64 is not established. The one-port S11 data are measured with the VNA reference plane outside the cryostat, and no cryogenic calibration or de-embedding is reported. The statement that 'identical VNA settings and reference plane were used across all temperatures' does not correct for temperature-dependent cable loss, which decreases upon cooling and would deepen and narrow the observed resonance dips exactly in the pattern reported (−16.5 dB to −25.3 dB; linewidth 93.5 MHz to 77.0 MHz). The limitations paragraph at the end of Section 3 acknowledges only simulation limitations, not this calibration gap. Please provide a cold calibration (e.g., reflection calibration at the resonator plane or measured cable S21 versus temperature) or otherwise quantify the cable contribution, and report the","section":"Section 2C, Fig. 3b"},{"comment":"The MCT hysteresis value (Δf0^hyst ≈ 150 MHz at 150 K) and the Q_L collapse to ~7 at 7 K may reflect thermal lag or noise-floor artifacts rather than intrinsic material behavior. No soak times, independent sample-temperature readings, or repeated thermal cycles are reported, and each material is represented by a single sample. Without error bars or a thermalization protocol, the path dependence and the low-temperature Q value are not quantitatively supported. Please add soak-time details, repeated-cycle data, and uncertainty estimates for f0 and Q_L.","section":"Section 2B, Fig. 3a"},{"comment":"The near-field/far-field classification is internally inconsistent. The text defines the far-field onset as rFF ≳ 2D²/λ and states that d < rFF is near-field, but the disk diameter D is not reported anywhere. The subsequent classification uses d ≤ λ as near-field and d > λ as far-field without showing that λ is comparable to rFF. If rFF exceeds 4λ, the 'far-field' magnitude modulations at 4λ would actually be reactive near-field effects, undermining the dual-modality sensing claim. Please report D, compute rFF, and relabel the sensing regimes consistently with the stated criterion.","section":"Section 2D, Fig. 4"},{"comment":"None of the quantitative metrics—f0 shifts, Q_L values, τf, or hysteresis—are reported with error bars or repeat measurements. With a single sample per material and no indication of measurement variability, the comparison between MCT and ZST is not statistically grounded. At minimum, provide repeated measurements, fitting residuals, or a conservative uncertainty estimate for each extracted quantity, especially the shallow MCT resonance at 7 K where Q_L=7 is extracted from a dip nearly indistinguishable from the noise floor.","section":"Section 2B, Section 2C"}],"minor_comments":[{"comment":"The τf definition in the text uses f_i, f_f, and f_avg without defining them; please add explicit definitions and units.","section":"Section 2A, Eq. (1)"},{"comment":"The cooldown and warmup traces in Fig. 3a are difficult to distinguish in grayscale; add markers or a clearer legend. Also label the temperature of each trace or provide a colorbar.","section":"Figure 3"},{"comment":"The inset reports approximate dip depths without uncertainties or repeated measurements; add error bars or state the measurement repeatability.","section":"Figure 4 inset"},{"comment":"Balanis (ref. 36) is listed but not cited in the text around the near/far-field definition; please add the citation where the criterion is introduced.","section":"References"},{"comment":"There are several typographical issues, including missing spaces ('antenna s', 'en ding') and inconsistent equation formatting. A careful proofread is needed.","section":"General"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses a timely topic and the qualitative trend (ZST stable, MCT unstable) is visible in the spectra. The main gap is measurement calibration: the Q-enhancement claim is load-bearing and currently vulnerable to cable-loss artifacts. The authors should be asked for cold-calibrated data or a direct cable-loss measurement before the quantitative conclusions can be accepted. I would also recommend tempering the 'high-coherence quantum interfaces' language given the modest absolute Q values (~64)."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a genuine side-by-side cryogenic characterization of MCT and ZST as dielectric resonator antennas, plus a through-window wireless sensing demo at 10 K. That specific combination is not in the cited literature, as far as I can tell. The qualitative contrast—ZST frequency-stable, MCT drifting badly—is visible in the raw spectra, and the paper is honest about its own limitations (room-temperature material models in simulation, window omitted from pattern calculations). The 1 mW wireless link is a credible proof of concept.\n\nWhere I'd push back: the quantitative Q and hysteresis claims are not yet supported. The loaded-Q extraction relies on one-port S11 linewidths and dip depths measured with the VNA reference plane outside the cryostat, and no de-embedding or cable-loss correction is reported. Coaxial cable loss is temperature-dependent; as the cable cools, loss drops, which can make the same resonator look deeper and narrower. The observed ZST change (dip from −16.5 to −25.3 dB, linewidth 93.5 to 77 MHz) is exactly the fingerprint of a few dB of cable-loss variation. So the 22.6% Q enhancement and the inferred unloaded Q increase from 48 to 64 are not established. This matters because the abstract and conclusions lean on it for the 'high-coherence quantum interface' framing. A simple fix: measure the cable S21 over temperature or calibrate at the cryostat plane, or at least report a raw through response. Without that, the qualitative claim of 'stable and no worse' survives, but the enhancement claim doesn't.\n\nAlso: the 150 MHz MCT hysteresis at 150 K could be thermal lag between sensor and sample; no soak times are reported. Single samples per material means no sample-to-sample variability. The near/far-field classification is internally inconsistent: they state the standard near-field limit rFF ~ 2D²/λ, then label d ≤ λ as near-field, and λ/2 is likely well beyond rFF. That doesn't kill the sensing demo but muddles the interpretation.\n\nThe central qualitative message—ZST holds frequency and MCT doesn't—is plausible and probably right. But the paper oversells the Q enhancement without a calibration correction. Send it to peer review, but require the de-embedding/through-calibration check, error bars, and an honest rewrite of the Q claims. If the cable effect explains the Q change, the paper still has value as a thermal-stability benchmark and a 1 mW through-window sensing demonstration.","headline":"Useful side-by-side cryogenic DRA data, but the ZST Q-enhancement claim needs a cable-calibration check before it can carry the 'foundational material' conclusion.","tokens_in":11816,"tokens_out":2315,"would_cite":true,"duration_ms":26802,"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":"In a head-to-head cryogenic test, the ceramic ZST holds its frequency, improves its quality factor, and senses a room-temperature target through a cryostat window at only 1 mW of power.","keywords":["cryogenic microwave materials","dielectric resonator antenna (DRA)","(Zr,Sn)TiO₄ (ZST)","MgTiO₃–CaTiO₃ (MCT)","quality factor","thermal hysteresis","wireless cryogenic link","TE₀₁δ mode"],"falsifier":"Repeat cooldown–warmup on a second MCT sample with a soak of 30 minutes at every setpoint and a full two-port calibration at each temperature; if the 150 K cooldown/warmup frequency gap shrinks below roughly 50 MHz, or if the ZST Q gain at 10 K disappears after reference-plane correction, the reported hysteresis and Q enhancement are measurement artifacts rather than material behavior.","tokens_in":10998,"feed_emoji":"📡","tokens_out":11717,"duration_ms":119024,"temperature":0.7,"pith_summary":"Quantum processors must sit at millikelvin temperatures, but every cable that carries microwave signals into the cryostat also carries heat—a bottleneck the paper attacks by asking whether a dielectric resonator antenna can send signals through a window instead. The paper compares two ceramic candidates side by side from 296 K down to 7–10 K: MgTiO₃–CaTiO₃ (MCT), engineered to be frequency-stable at room temperature, and (Zr,Sn)TiO₄ (ZST). MCT fails in the cold: its resonance drifts roughly 230 MHz, its loaded quality factor collapses from about 34 to 7, and it shows roughly 150 MHz of thermal hysteresis, behavior the authors trace to incipient-ferroelectric and relaxor-like losses in its CaTiO₃ component. ZST, by contrast, drifts only about 30 MHz, improves its loaded quality factor by about 22%, and shows no discernible hysteresis. The authors then run the ZST disk as a 1 mW antenna at 10 K and demonstrate a through-window wireless link that detects a room-temperature water sphere out to about four wavelengths—evidence that cable-free sensing and, eventually, wireless interconnects from the cold stage are physically plausible.","feed_headline":"At 1 mW, a 10 K ceramic antenna senses through a cryostat window","feed_subtitle":"ZST stays stable in the cold, making cable-free readout from cryogenic stages a practical direction.","key_machinery":"The argument runs on two coupled objects. First is the TE₀₁δ dielectric-resonator mode of high-permittivity ceramic disks, whose azimuthal electric field keeps energy confined inside the dielectric—the same property that gives a high unloaded Q and, through the magnetic field enhancement (about 26 to 33 times the feed field at resonance), the strong coupling a qubit would need. Second is the material relation τf = −(½ τ_εr + α_L), which ties resonant-frequency stability to the temperature coefficient of permittivity and thermal expansion; the paper's entire contrast is that MCT relies on room-temperature cancellation between components (MgTiO₃ and CaTiO₃, an incipient ferroelectric whose Bar","core_discovery":"On the paper's own terms, the discovery is comparative: operated as dielectric resonator antennas under identical fixtures and protocols from 296 K to 7–10 K, MCT and ZST diverge completely. The MCT resonator drifts roughly 230 MHz by 10 K, its loaded Q collapses from about 34 to 7 below 150 K, and it shows roughly 150 MHz of thermal hysteresis at the 150 K setpoint—behavior the authors attribute to incipient-ferroelectric and relaxor-like losses in its CaTiO₃ component. The ZST resonator drifts only about 30 MHz over the same range (average τf ≈ −40 ppm/K), improves its loaded Q by 22.6% (estimated unloaded Q rising from about 48 to 64), and shows negligible hysteresis. The authors then ope","pith_inferences":["The paper stops at sensing; a plausible next step it does not claim is using the same ZST disk both as a qubit-coupling element (its H-field enhancement reaches about 26 times the feed field) and as the radiator for a wireless control or readout link in a quantum chip package.","Because each material was represented by a single sample with no error bars and no stated soak times, the general claim that ZST outperforms MCT at 10 K would be strengthened by a multi-sample repeat with calibrated reference planes—a test the reader could carry out rather than a conclusion the paper already proves.","The dual near/far-field readout scheme should generalize to any dielectric target, which suggests a fast screening test for candidate cryogenic dielectrics: watch for a loaded-Q collapse below about 150 K as an early-warning signature of relaxor-type loss."],"forward_implications":["ZST becomes a candidate baseline dielectric for deep-cryogenic resonators: a 30 MHz total frequency drift (average τf ≈ −40 ppm/K), a 22.6% increase in loaded Q at 10 K, and no measurable thermal hysteresis.","The measured Q improvement implies roughly an 18% narrower resonance linewidth at 10 K, which supports denser frequency-division multiplexing with fewer frequency collisions.","A through-window wireless link works at 1 mW input power, detecting a dielectric target from λ/2 to 4λ via near-field frequency shifts (up to −2.5 MHz) and far-field amplitude modulation—evidence that cable-free readout from a 10 K stage is physically feasible.","The low-temperature Q of ZST is attributed to temperature-independent extrinsic losses (oxygen vacancies, impurities, grain boundaries), so cryogenic Q measurements become a practical screening tool for material purity and processing quality.","Room-temperature τf ≈ 0 engineering does not survive deep cooling: the MCT result implies that materials relying on cancellation between components with opposed permittivity-temperature signs need re-evaluation or suppression of the incipient-ferroelectric phase before use below about 150 K."],"supporting_citations":[{"why":"Supplies the room-temperature microwave dielectric properties of the MgTiO₃–CaTiO₃ composite whose cryogenic behavior is being measured.","marker":"(21)"},{"why":"Identifies CaTiO₃ as an incipient ferroelectric, the mechanism invoked for MCT's steep low-temperature frequency drift.","marker":"(22)"},{"why":"Establishes the processing and microwave characteristics of (Zr,Sn)TiO₄, the material whose cryogenic stability the wireless link relies on.","marker":"(24)"},{"why":"Reports cryogenic microwave characteristics of (Zr0.8Sn0.2)TiO₄ ceramics, the earlier low-temperature data this work extends to radiative antenna operation.","marker":"(25)"},{"why":"Supplies design relations for the TE₀₁δ dielectric resonator antenna mode used for both disks.","marker":"(26)"},{"why":"Provides the one-port coupling equations (β and Qu = (1+β)QL) and the near-field criterion used to extract Q and classify sensing regimes.","marker":"(29)"},{"why":"Gives the Barrett formula for perovskite-type dielectric constant versus temperature, used to model MCT's permittivity rise near 0 K.","marker":"(30)"},{"why":"Documents relaxor-ferroelectric loss processes, the proposed physical origin of MCT's Q collapse and hysteresis.","marker":"(31)"},{"why":"Provides the cavity perturbation theorem (Slater) used to predict the sign and size of near-field frequency shifts from a dielectric target.","marker":"(34)"}],"fun_headline_variants":["Cryo antenna at 10 K senses through window with 1 mW","ZST ceramic antenna stays stable at 10 K, enables wireless sensing","At 1 mW, a 10 K antenna reads through cryostat window","Stable ZST antenna at 10 K: cable-free cryogenic readout"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The headline numbers come from one-port S11 curves of a single sample of each material, with no error bars or stated soak times, no correction for cable or reference-plane losses, and temperature-independent simulation models used for design—so the reported hysteresis and Q values assume the S11 trace itself, not cable drift, thermal lag, or fixture effects, is what changed.","fun_headline_variants_meta":{"raw":{"variants":["Cryo antenna at 10 K senses through window with 1 mW","ZST ceramic antenna stays stable at 10 K, enables wireless sensing","At 1 mW, a 10 K antenna reads through cryostat window","Stable ZST antenna at 10 K: cable-free cryogenic readout"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000606,"raw_usage":{"total_tokens":2717,"prompt_tokens":854,"completion_tokens":1863,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":598,"completion_tokens_details":{"reasoning_tokens":1778}},"tokens_in":598,"tokens_out":1863,"duration_ms":15493,"temperature":1.0,"reasoning_tokens":1778,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T23:56:45.646136+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Repeat cooldown–warmup on a second MCT sample with a soak of 30 minutes at every setpoint and a full two-port calibration at each temperature; if the 150 K cooldown/warmup frequency gap shrinks below roughly 50 MHz, or if the ZST Q gain at 10 K disappears after reference-plane correction, the reported hysteresis and Q enhancement are measurement artifacts rather than material behavior.","supporting_citations":[],"review_version":1}