{"id":"c68dc9d8-1a8f-49e9-861f-36849cc11a87","arxiv_id":"2411.17872","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A dielectric metasurface of disk resonators is designed and demonstrated to support localized guided waves with nearly identical TE and TM dispersion for all in-plane directions.","lead":"This paper shows that a flat metasurface made of small dielectric disks can guide light that is both tightly confined and polarization-insensitive, with the two polarization states (TE and TM) traveling at nearly the same speed. Such a platform could lead to compact, planar optical devices that control polarization without sacrificing miniaturization.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The cross-material design rule is an unpublished fit with no sensitivity analysis; the single microwave validation also appears unable to resolve the claimed Δk, so the universality claim is not yet established.","rationale":"The reader's conditional verdict is reasonable. The weakest point is indeed the universality of the design rule: the paper's headline generality ('different constitutive materials', 'required operating wavelength') depends almost entirely on the fitted curves in Fig. 3, and no functional form, residuals, uncertainty, or sensitivity is reported. The microwave experiment is a single point and, as quantified above, the stated scan parameters may not resolve the claimed Δk, further weakening the empirical anchor. The numerical demonstration for Si3N4 (cases #1-#3) is solid as an existence proof for those specific parameters: MPB is a standard, reproducible code, the design maps in Fig. 4 show a clear optimum ridge, and the reported Δk values are concrete. Therefore I would not reject or accept unconditionally; the paper needs to make the design rule reproducible (fit function/table), demonstrate it on at least one additional non-fitted material, and quantify experimental extraction uncertainty. This leaves the reader's CONDITIONAL verdict unchanged.","tokens_in":8,"tokens_out":16155,"duration_ms":228695,"concrete_test":"Use the Fig. 3 fit to predict d/h, h/a, and d/a for a material not among the fitted points, e.g., ε=9 (n=3) at κ=1.2, then run the same MPB band-structure calculation for that geometry; if the achieved Δkmax exceeds the claimed ~4.1e-3π/a, or if the optimized geometry differs from the fitted prediction by more than 0.01, the claimed universality of the design rule fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that the design rule is universal rests on the fitted curves in Fig. 3. The paper gives no fitting function, no residuals, no measure of how the optimum ridge broadens, and no study of fabrication-tolerance sensitivity; the statement 'optimized with a tolerance of 0.001a' does not quantify how Δk grows when d/a or h/a deviates by 0.001a. The one experimental point (ε=25, κ=1.25) is not enough to validate a global curve. Moreover, the experimental estimate of Δk is at risk: with a 340 mm scan, 1 mm step, and a=12.5 mm, the nominal Fourier resolution is 2π/340 mm ≈ 0.074π/a, which is more than six times larger than the claimed Δk<12e-3π/a; unless a sub-bin extraction procedure with quantified uncertainty is provided, the experimental DBW values in Fig. 7(b) cannot be resolved by the stated method. If the fitted curve is an artifact of the specific parameter sweep, the paper's extension to 'different constitutive materials' does not follow.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript proposes a design principle for dielectric metasurfaces consisting of disk resonators in a square lattice, aiming to support guided TE and TM modes with nearly identical dispersion (small Δk) over a broad frequency range and all in-plane directions, while maintaining strong vertical localization (large κ). The authors use MPB simulations to optimize d/a and h/a for given disk permittivity and localization, present silicon nitride designs with Δk_max = 4.1×10^-3 π/a at κ_max = 1.2, analyze the effect of a quartz substrate, and report a microwave experiment on a ceramic (ε = 25) metasurface that allegedly confirms Δk < 12×10^-3 π/a between 8 and 9.6 GHz. The paper claims that the design rule is universal across constitutive materials and spectral ranges.","tokens_in":14733,"tokens_out":4309,"duration_ms":34954,"significance":"If the claims hold, the work would provide a practical route to planar polarization-controlling devices based on localized guided modes, a useful complement to existing single-direction or narrowband degeneracy schemes. The paper's strengths include concrete, reproducible numerical designs (MPB, geometric parameters listed), a clear operational principle based on overlapping Mie-type magnetic dipole modes, and an independent microwave experiment. However, the two load-bearing pillars — the universal design curves of Fig. 3 and the experimental Δk resolution — are not quantitatively secured, so the central claims currently outrun the evidence.","major_comments":[{"comment":"The experimental DBW values are not resolvable with the stated acquisition parameters. With a scan length L = 340 mm and lattice constant a = 12.5 mm, the nominal Fourier resolution in the in-plane wave vector is 2π/L ≈ 0.0185 rad/mm, which corresponds to 2a/L × π/a ≈ 0.074π/a. This is more than six times larger than the claimed Δk < 12×10^-3 π/a. Unless a sub-bin extraction procedure with quantified uncertainty is described, the red curve in Fig. 7(b) cannot support the stated DBW values. Please provide the dispersion extraction method (e.g., peak fitting, interpolation, windowing) and error bars.","section":"§III.B, Fig. 7(b)"},{"comment":"The universal design dependencies are presented as fits (\"fitting distributions\") without the fitting function, residuals, number of fitted points, or a sensitivity analysis. Since these curves are the basis for the claim that the approach extends to arbitrary constitutive materials and spectral ranges, the manuscript needs to provide the fit equation, its residuals, and an analysis of how Δk grows when d/a or h/a deviate from the optimal values, including the stated tolerance of 0.001a. Without this, the universality claim is a descriptive summary of the numerical search rather than a validated design rule.","section":"§II.C, Fig. 3"},{"comment":"The microwave experiment tests only one point (ε = 25, κ_max = 1.25, d/a = 0.426, h/a = 0.4) on the proposed design surface. The paper does not show where this geometry falls relative to the fitted curves of Fig. 3 at ε = 25, nor does it quantify the agreement between the measured DBW and the simulation for that geometry. A single point cannot validate a global design rule; at minimum, the experimental geometry should be overlaid on Fig. 3 and the measured Δk(f) compared with the simulation for that specific geometry with uncertainty bounds.","section":"§III.A, Fig. 3"}],"minor_comments":[{"comment":"The colored regions in Fig. 3(a) are not clearly defined in the main text; specify which materials correspond to which ranges and cite the data sources for the material dispersion.","section":"Fig. 3(a)"},{"comment":"The axis labels \"F(Ez)(c)\" and \"F(Hz)\" are confusing; rename them to indicate the Fourier magnitudes, e.g., |F(Ez)| and |F(Hz)|.","section":"Fig. 6(c)"},{"comment":"The text states \"case #1 (d/a = 0.655, h/a = 0.843)\" while Table I lists h/a = 0.823 for case #1; please reconcile the discrepancy.","section":"§II.D and Table I"},{"comment":"Reference [58] appears incomplete (\"The rise of Mie-tronics (2022)\"); provide full citation details.","section":"Reference [58]"},{"comment":"The phrase \"discover the near-field polarization degree of freedom\" is repeated in the abstract and conclusions; consider toning down \"discover\" to \"demonstrate\" for objectivity.","section":"Abstract and Conclusions"}],"recommendation":"major_revision","confidential_remarks":"The experimental resolution issue in §III.B is serious: if the authors cannot provide a validated sub-bin extraction with quantified uncertainties, the experimental verification should be downgraded to a qualitative demonstration, and the universality claim should be restricted to the simulated designs. The missing sensitivity analysis for the Fig. 3 curves also needs to be addressed before the design rule can be considered established."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Worth a careful read if you work on all-dielectric metasurfaces or integrated polarization control. The genuinely new step is moving from single-direction TE-TM degeneracy (their earlier work) to a two-dimensional square-lattice platform claimed to be broadband and multi-directional, with numerical designs for Si3N4 (free-standing and on quartz) plus a microwave ceramic prototype. The prior work is cited clearly, and the extension is real. The physical idea is sensible: overlapping orthogonal Mie-type magnetic dipole modes can be tuned to bring the TE and TM band edges together at localization κ ≈ 1.2–1.5. The numerics are standard MPB band-structure calculations, and the experimental section is substantial — measured field maps and isofrequency contours, not just a sketch.\n\nThe soft spots are real but not fatal. First, the 'universal design dependencies' in Fig. 3 are fitted curves with no fitting form, residuals, or sensitivity analysis. Since each point is chosen to minimize DBW, small Δk is partly restating the optimization objective, so the fitted curve is a description of the search, not a first-principles rule. The single microwave point (ε=25) checks one material but can't validate the global curve. Second, the experimental DBW claim in Fig. 7(b) is under-resolved as presented: with a 340 mm scan and a=12.5 mm, the nominal Fourier bin width is about 0.074π/a, about six times the claimed Δk < 12×10^-3 π/a. The authors may be using sub-bin peak extraction, but they don't say so or quantify the uncertainty. That needs to be fixed.\n\nFor whom: this is useful for people designing planar polarization devices and for reviewers who care whether design rules extrapolate across materials. My own verdict is that the existence claim — a metasurface with near-degenerate TE and TM guided modes across in-plane directions — is well supported for the specific designs, while the universality claim is not yet established. Send it to peer review, but ask referees to demand the fitting function, a tolerance/robustness study, and experimental error analysis.","headline":"Extends prior single-direction TE-TM degeneracy to a 2D multi-directional dielectric metasurface platform, with a credible microwave proof-of-principle but an over-sold universal design rule and under-resolved experimental DBW.","tokens_in":15205,"tokens_out":3263,"would_cite":true,"duration_ms":29962,"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":"A square-lattice metasurface of dielectric disks can guide waves that are both tightly confined and polarization-degenerate across all in-plane directions.","keywords":["TE-TM polarization degeneracy","dielectric metasurface","guided waves","Mie resonances","magnetic dipole modes","near-field polarization","localized light","microwave verification"],"falsifier":"Fabricate a silicon-nitride metasurface with $d/a=0.655$ and a lattice period of 300 nm, designed for $\\kappa_{\\max}=1.2$, and measure the TE and TM guided dispersion across 500 to 1700 nm; if the wavenumber difference exceeds the predicted $4.1\\times10^{-3}\\,\\pi/a$ by more than the stated fabrication tolerance, the degeneracy claim is falsified. Alternatively, choose a disk material with permittivity between 4 and 49, read $d/a$ and $d/h$ from the fitted curves, build the array at the target $\\kappa$, and check whether the predicted $\\Delta k$ actually holds, since the experiment tests only the $\\varepsilon=25$ point.","tokens_in":1788,"feed_emoji":"🔬","tokens_out":4496,"duration_ms":89387,"temperature":0.7,"pith_summary":"The paper claims that a square array of dielectric disk resonators can guide waves that are strongly localized to the surface while preserving the polarization degeneracy normally lost in miniaturized waveguides: transverse-electric (TE) and transverse-magnetic (TM) modes travel with almost the same wavenumber at the same frequency and along every in-plane direction. This matters because compact planar optics typically trades away polarization control, forcing polarization-handling components to rely on anisotropic crystals or one-dimensional waveguides. The authors derive universal design curves for disk diameter, height, and lattice period as functions of the disk permittivity at several localization levels, then support those curves with silicon-nitride designs and a microwave ceramic prototype whose TE and TM dispersion stays close over a broad band.","feed_headline":"Disk metasurface restores polarization freedom to tightly bound light","feed_subtitle":"Silicon-nitride and ceramic disk arrays keep TE and TM modes in lockstep across all in-plane directions.","key_machinery":"The load-bearing object is the pair of orthogonal Mie magnetic-dipole modes of a cylindrical dielectric resonator: HE11ℓ (a horizontal magnetic dipole whose resonance is set by the disk height) and TE01ℓ (a vertical magnetic dipole whose resonance is set by the disk diameter). When the two resonances overlap and the disks are arranged in a square lattice, the TM and TE guided modes of the metasurface follow nearly identical dispersion. The paper's quantitative yardsticks are the wavenumber difference $\\Delta k = |k_p - k_s|$, the localization degree $\\kappa = k/(n_{\\text{sub}} k_0)$, and the normalized decay length $Z_\\lambda$; the fitted design curves in Fig. 3 give $d/h$ and $d/a$ versus disk permittivity at fixed $\\kappa$, and the designs are optimized with a tolerance of $0.001a$ using a frequency-domain eigenmode solver.","core_discovery":"The central claim is that near-field TE-TM degeneracy of guided waves is achieved by spectrally overlapping two mutually orthogonal Mie-type magnetic dipole modes of each disk resonator: the horizontal-dipole HE11ℓ mode, tuned by the disk height $h$, and the vertical-dipole TE01ℓ mode, tuned by the disk diameter $d$. Because these orthogonal modes separately control the TM and TE guided polarizations, choosing the geometry so the two resonances coincide makes the array's TE and TM branches nearly identical. The paper states that array interactions (spatial dispersion) prevent the single-resonator Kerker condition from being sufficient, so a parametric optimization over $d/a$, $h/a$, and the lattice period is required. Quantitatively, for silicon nitride at localization $\\kappa_{\\max}=1.2$ the wavenumber difference stays below $\\Delta k_{\\max}=4.1\\times10^{-3}\\,\\pi/a$, corresponding to a polarization-preserving length up to $L_{\\max}/\\lambda=898$; a microwave ceramic ($\\varepsilon=25$) disk array confirms $\\Delta k<12\\times10^{-3}\\,\\pi/a$ from 8 to 9.6 GHz for all in-plane directions.","pith_inferences":["The universality claim could be stress-tested by porting the fitted $d/a$ and $d/h$ curves to a material not used here, such as titanium dioxide in the visible, and measuring the resulting $\\Delta k$; the paper's evidence covers only silicon-nitride simulations and one ceramic experimental point.","Because the degeneracy is purely geometric, combining it with a tunable disk material (e.g., phase-change or thermo-optic) could switch between degenerate and non-degenerate operation, turning the passive platform into an active polarization controller.","The mechanism relies on overlapping two orthogonal magnetic-dipole resonances, so similar degeneracies might be realized with other resonator shapes or lattice symmetries such as hexagonal, but those extensions are not demonstrated.","The paper leaves fabrication sensitivity open: designs are optimized to $0.001a$, but the measured impact of random errors in $d/a$ and $h/a$ on $\\Delta k$ is not quantified, so a tolerance study would be a direct next experiment."],"forward_implications":["For a silicon-nitride metasurface with $\\kappa_{\\max}=1.2$, the TE and TM guided branches stay within $\\Delta k_{\\max}=4.1\\times10^{-3}\\,\\pi/a$, so a guided wave preserves its polarization for up to $L_{\\max}/\\lambda=898$ wavelengths.","The same design curves apply to materials from visible/NIR silicon nitride to microwave ceramics, so one geometry recipe serves different spectral ranges by scaling the lattice period.","A low-index substrate such as quartz does not destroy the degeneracy; enlarging the disks compensates for its presence.","The microwave prototype demonstrates that the degeneracy holds for all in-plane directions over 8 to 9.6 GHz, not just along a single propagation direction.","This offers a planar, all-dielectric route to polarization converters, filters, demultiplexers, and sensors that work with tightly localized light."],"supporting_citations":[{"why":"Supplies the block-iterative frequency-domain eigenmode solver used to compute all dispersion curves and optimize the designs.","marker":"[32]"},{"why":"Gives the HE11ℓ and TE01ℓ mode notation and the dependence of resonant frequency on disk height and diameter.","marker":"[39]"},{"why":"Establishes the prior single-directional all-dielectric meta-waveguide TE-TM degeneracy that this work generalizes to two dimensions.","marker":"[24]"},{"why":"Provides a previous example of TE-TM polarization degeneracy for surface waves, used as a reference for multidirectional degeneracy.","marker":"[23]"},{"why":"Introduces the lattice Kerker effect; the paper argues the single-resonator Kerker condition is insufficient once array interactions are included.","marker":"[40]"},{"why":"Supplies the refractive index and material properties used for the silicon-nitride metasurface designs.","marker":"[28]"},{"why":"Gives the ceramic permittivity ε=25 and loss tangent used for the microwave prototype.","marker":"[53]"},{"why":"Provides the near-field Fourier-transform procedure for extracting isofrequency contours and dispersion from measured field maps.","marker":"[56]"}],"fun_headline_variants":["TE-TM degeneracy restored for tightly bound guided waves","Metasurface restores polarization control for tightly confined light","Dielectric disk arrays lock TE and TM modes for localized light","Polarization-degenerate guided waves on dielectric metasurfaces"],"cache_read_input_tokens":17408,"weakest_assumption_plain":"The design recipe derived from simulations is assumed to work for any transparent material and wavelength, even though only one experimental confirmation point is provided.","fun_headline_variants_meta":{"raw":{"variants":["TE-TM degeneracy restored for tightly bound guided waves","Metasurface restores polarization control for tightly confined light","Dielectric disk arrays lock TE and TM modes for localized light","Polarization-degenerate guided waves on dielectric metasurfaces"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000991,"raw_usage":{"total_tokens":4202,"prompt_tokens":948,"completion_tokens":3254,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":564,"completion_tokens_details":{"reasoning_tokens":3185}},"tokens_in":564,"tokens_out":3254,"duration_ms":23910,"temperature":1.0,"reasoning_tokens":3185,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T11:45:12.040908+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Fabricate a silicon-nitride metasurface with $d/a=0.655$ and a lattice period of 300 nm, designed for $\\kappa_{\\max}=1.2$, and measure the TE and TM guided dispersion across 500 to 1700 nm; if the wavenumber difference exceeds the predicted $4.1\\times10^{-3}\\,\\pi/a$ by more than the stated fabrication tolerance, the degeneracy claim is falsified. Alternatively, choose a disk material with permittivity between 4 and 49, read $d/a$ and $d/h$ from the fitted curves, build the array at the target $\\kappa$, and check whether the predicted $\\Delta k$ actually holds, since the experiment tests only the $\\varepsilon=25$ point.","supporting_citations":[{"cited_title":"Shkondin, O","cited_arxiv_id":null,"evidence_quote":"Supplies the block-iterative frequency-domain eigenmode solver used to compute all dispersion curves and optimize the designs."},{"cited_title":"Allayarov, A","cited_arxiv_id":null,"evidence_quote":"Gives the HE11ℓ and TE01ℓ mode notation and the dependence of resonant frequency on disk height and diameter."},{"cited_title":"Yermakov, V","cited_arxiv_id":null,"evidence_quote":"Establishes the prior single-directional all-dielectric meta-waveguide TE-TM degeneracy that this work generalizes to two dimensions."},{"cited_title":"Kruk and Y","cited_arxiv_id":null,"evidence_quote":"Provides a previous example of TE-TM polarization degeneracy for surface waves, used as a reference for multidirectional degeneracy."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Introduces the lattice Kerker effect; the paper argues the single-resonator Kerker condition is insufficient once array interactions are included."},{"cited_title":"Aguilar, S","cited_arxiv_id":null,"evidence_quote":"Supplies the refractive index and material properties used for the silicon-nitride metasurface designs."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the ceramic permittivity ε=25 and loss tangent used for the microwave prototype."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the near-field Fourier-transform procedure for extracting isofrequency contours and dispersion from measured field maps."}],"review_version":1}