{"id":"0a2607d5-831a-49db-8d10-7961d10c4242","arxiv_id":"2411.15928","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"A metasurface made of rapidly spinning dielectric cylinders is shown numerically to provide strong one-way transmission and reflection, including nonreciprocal wavefront deflection.","lead":"This paper uses computer simulations to show that a grid of tiny silicon cylinders, spinning very fast above a silicon sheet, lets light pass in one direction while blocking it in the opposite direction. It is a proposal for a new kind of nonreciprocal optical metasurface that could shrink optical isolators and enable one-way light control without magnets or nonlinear materials.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Static effective-medium model is applied to non-axisymmetric rotating pillar-cylinder meta-atoms in Sec. III B without justification; the reported wavefront-manipulation isolation ratios are not established.","rationale":"The reader's weakest assumption identifies the Minkowski constitutive relations as unvalidated for a rapidly spinning solid. My concern is more specific and, I argue, more load-bearing for a central part of the paper: the wavefront-manipulation section applies a time-independent effective-medium model to a composite meta-atom (cylinder plus rectangular pillar) that is not invariant under rotation. For a homogeneous axisymmetric cylinder, a frequency-domain effective-medium simulation is internally consistent within the Minkowski model; the pillars, however, break rotational symmetry, making the lab-frame structure time-periodic. A static COMSOL model of a stationary bianisotropic domain cannot capture the resulting sidebands, and the paper supplies no estimate of their magnitude. Since nonreciprocal beam deflection is explicitly advertised as a key achievement (abstract, Sec. I, Sec. III B), this invalidates a substantial portion of the claimed demonstrations. Section III A, dealing with the purely axisymmetric spinning cylinder, is not affected by this specific flaw and may remain valid subject to the broader constitutive-model uncertainties. The verdict should therefore stay CONDITIONAL rather than move to REJECT: the basic mechanism might be sound, but the wavefront-manipulation results require either a rigorous justification of the static approximation or a full time-modulated simulation. This is a concrete, checkable condition, consistent with the reader's conditional acceptance.","tokens_in":12697,"tokens_out":18567,"duration_ms":177609,"concrete_test":"Recompute the supercell transmission of the designs in Figs. 7-9 using a time-domain or Floquet solver that explicitly includes the rotating geometry: model the pillar-cylinder meta-atom as a time-periodic material distribution (e.g., via a moving mesh or time-varying permittivity in COMSOL) and extract the transmitted power in the 249.56 THz carrier channel for both forward and backward incidence. Compare the resulting isolation ratio |S21|^2 - |S12|^2 with the static effective-medium values; if the carrier-channel isolation ratio differs by more than 10% relative, the static approximation used in Sec. III B is invalid and the wavefront-manipulation claims need revision.","verdict_should_be":"UNCHANGED","load_bearing_attack":"In Sec. II, the moving-medium constitutive relations (Eqs. (1) and (2)) are recast as a time-independent bianisotropic medium. This treatment is valid for a homogeneous cylinder spinning about its own axis, because the lab-frame material distribution remains static. In Sec. III B, rectangular dielectric pillars (width w, height h) are added to the meta-atoms (Fig. 6(a)) and the same stationary effective-medium model is used in the COMSOL frequency-domain simulations. However, a cylinder plus a rectangular pillar is not rotationally symmetric: as the assembly rotates about the cylinder axis, the lab-frame geometry and permittivity distribution are time-periodic with period 2π/Ω. The scattered field then contains sidebands at frequencies ω ± nΩ, and the carrier-frequency transmission cannot be obtained from a static simulation of a stationary bianisotropic object unless the time modulation is demonstrated to be negligible. The paper provides no such justification. At the simulated normalized speed ΩR/c = 0.01, the rotation frequency is about 2.4 THz, and the sideband coupling for a non-axisymmetric body of width up to 200 nm is not obviously small. Consequently, the reported isolation ratios for the beam-deflection designs (44.8%, 57.1%, and 37.6% in Figs. 7-9) are not reliable evidence for nonreciprocal wavefront manipulation, even if the underlying Minkowski constitutive model for a homogeneous rotating cylinder is accepted.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript proposes a nonreciprocal optical metasurface composed of a periodic array of spinning silicon cylinders placed above a silicon substrate. The cylinders are modeled using the Minkowski constitutive relations for moving media, which are recast as an equivalent stationary bianisotropic (Tellegen-type) medium, giving an effective azimuthal gauge field that splits the degeneracy of clockwise and counterclockwise chiral multipole modes. Using COMSOL full-wave simulations, the authors report strongly asymmetric transmission and reflection for opposite incident angles, an isolation ratio of about 65% at f = 249.56 THz for opposite incidences, and nonreciprocal beam deflection with isolation ratios of 44.8%, 57.1%, and 37.6% obtained by adding dielectric pillars to the meta-atoms to control the transmission phase. The paper claims that this constitutes a new mechanism for nonreciprocal light manipulation in free space, distinct from magneto-optic, nonlinear, and temporal-modulation approaches.","tokens_in":12994,"tokens_out":4301,"duration_ms":40400,"significance":"If the moving-medium constitutive model and the static-equivalent simulation approach are valid, the paper offers a conceptually new route to nonreciprocity based on the Sagnac effect in spinning dielectric meta-atoms, with no need for magnets or nonlinear materials. The manuscript includes useful internal controls: the Ω=0 simulations show symmetric transmission and reflection spectra, and the multipole expansion provides a physical picture of the chiral-mode excitation asymmetry. The proposed mechanism is falsifiable and could be extended to topological or non-Hermitian photonic platforms. However, the entire evidence is numerical, and the Sec. III B designs involve non-axisymmetric rotating bodies for which the static-equivalent model is not automatically valid; without addressing this, the wavefront-manipulation results are not yet credible. The paper also omits simulation parameters and independent validation of the constitutive assumption, which currently limits its significance to a conditional proof-of-principle.","major_comments":[{"comment":"The meta-atoms in Sec. III B are not rotationally symmetric: a rectangular pillar attached to the cylinder breaks the cylindrical symmetry, so the lab-frame permittivity distribution is time-periodic with period 2π/Ω. The effective stationary bianisotropic model of Eq. (2) is not applicable to such a body; a rotating non-axisymmetric scatterer produces frequency sidebands at ω ± nΩ. The frequency-domain COMSOL simulation of a static equivalent object does not include these sidebands, and the paper provides no estimate of their amplitude at ΩR/c=0.01 (Ω≈2.4 THz). The isolation ratios reported in Figs. 7(c), 8(c), and 9(c) (44.8%, 57.1%, 37.6%) therefore do not yet establish nonreciprocal wavefront manipulation. A Floquet or time-domain treatment of the actual rotating geometry, or a rigorous bound on sideband coupling, is required.","section":"Sec. III B, Figs. 6-9"},{"comment":"The entire simulation rests on the Minkowski constitutive relations applied pointwise to a rigidly rotating dielectric cylinder. The paper neither validates this model for a submicron solid silicon object at ΩR/c=0.01 nor discusses the effect of material dispersion, mechanical stress, and the assumption of rigid-body rotation. Since this constitutive step is the origin of the Tellegen response that drives all the reported phenomena, the authors should state clearly that this is a modeling assumption and delineate its range of validity, ideally with an independent check such as a comparison with an alternative formulation of moving-medium electrodynamics or with published experimental data on spinning microresonators.","section":"Sec. II, Eqs. (1)-(2)"},{"comment":"The paper reports full-wave simulations but provides no simulation parameters (boundary conditions, mesh size, solver type, port definitions, or convergence checks). The text also does not state how periodic boundary conditions are imposed, how the incidence angles are mapped to Floquet ports, or how the reflection/transmission coefficients are extracted. Because the central results are entirely computational, the absence of these details prevents reproducibility and verification. Please add a Simulation Methods paragraph or provide the COMSOL models and data.","section":"Sec. III A, all simulations"}],"minor_comments":[{"comment":"There is a typo in the first paragraph of the introduction: 'The airticle is organized as follows' should be 'The article is organized as follows'.","section":"Sec. I"},{"comment":"In the sentence defining the spinning speed, the notation is inconsistent: 'spin at angular velocity Ω = Ωˆ z' uses the same symbol for the vector and its magnitude, and the clause 'where c is the speed of light' appears abruptly; please clarify the definitions of Ω, R, and the normalized speed ΩR/c.","section":"Sec. II"},{"comment":"Equation (5) as printed is ambiguous: the denominator 'Aθ 2 − ε′rµ′z' lacks parentheses, and the dependence of Aθ on Ω and r is not repeated; a cleaner expression with clear notation for the mode quantities would help the reader.","section":"Sec. II, Eq. (5)"},{"comment":"The labels in Fig. 6(b) and 6(c) appear to contain typographical artifacts such as '1n 2n 3n 4n'; these should be corrected to '1-4' or 'No. 1-4' consistently with the text.","section":"Fig. 6"},{"comment":"The Fourier analysis of transmission channels does not specify the number of harmonics retained, the spatial window used, or how the channel amplitudes are normalized; please add this information for reproducibility.","section":"Sec. III B, Figs. 7-9"},{"comment":"Reference [8] is cited in the context of metasurface applications but is about a self-biased non-reciprocal magnetic metasurface; the citation placement in the introduction may confuse readers and should be reconsidered.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The manuscript addresses an interesting and timely topic, and the Sec. III A results for axisymmetric spinning cylinders are internally consistent, with a good Ω=0 control. The decisive issue is Sec. III B: the use of a static effective-medium model for assemblies that are not rotationally symmetric is not justified, and the reported beam-deflection isolation ratios are therefore not established. I would encourage the editor to seek a revision in which the authors either restrict the wavefront-manipulation claims to cases where the static model holds, or provide a rigorous Floquet/time-domain simulation of the actual rotating geometry. I would also request that the constitutive-model validity and the simulation parameters be explicitly documented before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my read. The paper's core idea is applying the Minkowski moving-medium model to a periodic array of spinning silicon cylinders above a substrate. For the bare cylinders, which are rotationally symmetric, the model is a standard and well-defined approximation, and the numerical work is solid: Ω=0 gives symmetric spectra, the spinning case shows strong asymmetric transmission/reflection, and the ~65% isolation at 249.56 THz is a clean, nonreciprocal result. The multipole explanation is qualitative but reasonable, and the frequency-splitting formula comes from the constitutive relations, not from a fit. This part of the paper is worth engaging seriously.\n\nThe soft spots are real, and one is serious. The Minkowski constitutive model is unvalidated for a solid dielectric spinning at ΩR/c=0.01, with no experimental or independent numerical check, and the paper provides no simulation parameters or code for faithful reproduction. Calling simulations 'demonstrations' is an overstatement, though a minor one.\n\nThe load-bearing flaw is in Sec. III B. Adding rectangular pillars to the cylinders breaks the rotational symmetry of each meta-atom. A spinning cylinder-plus-pillar is a time-varying scatterer; its lab-frame permittivity is periodic with period 2π/Ω, and scattering contains sidebands at ω ± nΩ. The paper applies the same stationary effective-medium model in a frequency-domain COMSOL simulation, effectively treating the structure as time-independent. That is unjustified. At ΩR/c=0.01 the rotation frequency is about 2.4 THz, and the pillar geometry is large enough that sideband coupling is not obviously negligible. Without a time-domain simulation or at least a quantitative bound on how much power leaks into sidebands, the reported isolation ratios for the beam-deflection designs (44.8%, 57.1%, 37.6%) are not established. The stress-test note is right on this.\n\nMy recommendation: send it to peer review, because the axisymmetric case is a credible and interesting numerical result and the wavefront issue is clearly fixable. But the referee should concentrate on the time-modulation problem and ask for a proper justification or for that section to be revised. I'd bring this to a reading group as a useful case of how easy it is to accidentally use a static effective-medium model on a non-axisymmetric moving structure, but I wouldn't cite the wavefront numbers.","headline":"The spinning-cylinder metasurface simulation is credible for the axisymmetric case, but the wavefront-shaping section's static effective medium for non-axisymmetric spinning pillars has a real time-modulation gap, so those isolation ratios are unestablished.","tokens_in":62,"tokens_out":6047,"would_cite":false,"duration_ms":91144,"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 metasurface of spinning silicon cylinders makes light pass one way and block the other, without magnets or nonlinear materials.","keywords":["nonreciprocal metasurface","spinning cylinders","Sagnac effect","Tellegen response","time-reversal symmetry breaking","nonreciprocal transmission","wavefront manipulation","moving medium"],"falsifier":"Build a test cell where a dielectric cylinder or disk of radius $R$ spins at $\\Omega R/c = 0.01$ next to a dielectric half-space and measure the transmission contrast for opposite incidence angles at the predicted resonance; if the measured isolation ratio falls far below the simulated 65 percent, the Minkowski constitutive assumption is the first suspect. A less demanding check is to compare the full-wave simulation against a rigorous moving-boundary or Doppler-shifted solver on a single cylinder and verify the predicted first-order splitting $\\Delta\\omega = 2m\\Omega(\\varepsilon\\mu-1)/(\\varepsilon\\mu)$.","tokens_in":1826,"feed_emoji":"🌀","tokens_out":2550,"duration_ms":74318,"temperature":0.7,"pith_summary":"The paper proposes a nonreciprocal optical metasurface whose nonreciprocity comes from mechanically spinning its meta-atoms, not from magnets or nonlinear materials. It claims that an array of silicon cylinders rotating just above a silicon substrate breaks time-reversal symmetry through the relativistic Sagnac effect, turning each cylinder into a so-called Tellegen-type bianisotropic element. In full-wave simulations, the structure yields strongly asymmetric transmission and reflection for opposite incidence angles, with an isolation ratio around 65 percent at 249.56 THz. Adding silicon pillars lets the same structure deflect a forward beam while blocking its time-reversed counterpart, with isolation as high as 57.1 percent. If true, this gives a free-space platform for one-way light routing that could be switched by starting or stopping the rotation.","feed_headline":"Spinning silicon cylinders make light flow one way","feed_subtitle":"A spinning-cylinder metasurface claims 65 percent isolation without magnets, nonlinear materials, or moving gratings.","key_machinery":"The load-bearing object is the spinning silicon cylinder treated as a moving medium with the Minkowski constitutive relations, which the paper rewrites as a stationary bianisotropic medium with Tellegen-type tensors. The crucial identity is the resulting eigenmode frequency splitting $\\Delta\\omega = 2m\\Omega(\\varepsilon\\mu-1)/(\\varepsilon\\mu)$ between the clockwise and counterclockwise chiral multipole modes, a photonic analogue of the Zeeman effect that the paper traces to the Sagnac effect. This splitting is what creates the effective azimuthal gauge field $A_\\theta$ in the Helmholtz equation and, together with spin-momentum locking of the substrate guided mode, produces direction-dependent coupling between the cylinders. The phase-gradient extensions rely on propagation phase from silicon pillars and the generalized Snell law to steer the transmitted beam.","core_discovery":"On the paper's own terms, the central discovery is that a periodic lattice of spinning dielectric cylinders coupled through a substrate behaves as a nonreciprocal metasurface. The rotation enters Maxwell's equations through the Minkowski constitutive relations and can be re-expressed as an effective azimuthal gauge field; this gauge field lifts the degeneracy of clockwise and counterclockwise chiral multipole modes of the cylinders, producing a Tellegen-type bianisotropic response that breaks reciprocity. Near the resonance of one chiral dipole mode, the cylinders couple asymmetrically to the forward and backward guided modes of the substrate, so an incident plane wave is largely transmitted from one side and largely reflected from the other. The paper demonstrates this with full-wave simulations, showing a maximum transmission contrast of 65.4 percent at $f = 249.56$ THz for opposite incident angles $\\pm 8.18^\\circ$, and extends the effect to nonreciprocal beam deflection by adding pillars that impose a phase gradient. The mechanism is presented as distinct from magneto-optic, nonlinear, and temporal-modulation schemes because it relies on the relativistic response of a moving medium.","pith_inferences":["One extension the paper leaves implicit is that the same gauge-field mechanism should work at other frequencies and with other dielectric materials, as long as the mode number and rotation speed keep the splitting first-order; a sweep of $\\varepsilon$ and $\\mu$ would be a direct test.","A practical device would likely need a different way to spin submicron silicon cylinders at $\\Omega R/c = 0.01$; the paper does not discuss mechanical drive, friction, or thermal load, so the simulated configuration may be an idealisation rather than a construction blueprint.","The Minkowski-relation step is the main unvalidated modelling assumption, so a comparison against a rigorous moving-boundary simulation or a carefully scaled rotating-dielectric experiment would either confirm the scheme or reveal where it breaks.","Because the effect appears at resonance of a chiral mode, losses in the silicon could reduce the isolation ratio; designing around that trade-off is a natural next step."],"forward_implications":["A compact optical isolator could be built from an array of spinning cylinders over a substrate, with no external magnetic field and no high-intensity nonlinear threshold.","The same metasurface acts as an angle-selective filter: light incident at $+8.18^\\circ$ is transmitted while light incident at the mirror angle is reflected, so beam direction becomes a control knob for one-way routing.","Adding subwavelength pillars turns the device into a one-way beam deflector, so forward light can be sent to a chosen diffraction channel while the time-reversed channel is suppressed.","The isolation ratio of roughly 65 percent is achieved in the subwavelength regime, so the device volume can be near the wavelength scale rather than requiring bulky magnets or long modulators.","Starting and stopping the rotation provides a dynamic, reconfigurable on-off switch for nonreciprocity at fixed frequency and geometry."],"supporting_citations":[{"why":"Supplies the Minkowski constitutive relations used to model the spinning cylinders as a moving medium.","marker":"[39]"},{"why":"Shows that a subwavelength spinning particle induces optical isolation via spin-orbit interaction, establishing the single-particle nonreciprocity the metasurface generalises.","marker":"[15]"},{"why":"Demonstrates nonreciprocal light propagation induced by a subwavelength spinning cylinder, giving the guided-mode coupling picture the metasurface builds on.","marker":"[16]"},{"why":"Introduces the gauge-field description of the Sagnac frequency shift in a rotating cavity, the basis for the effective azimuthal gauge field and chiral-mode splitting.","marker":"[36]"},{"why":"Analyses coupled spinning cylinders, providing the coupled-resonator framework used for the lattice.","marker":"[37]"},{"why":"Defines Tellegen-type bianisotropic media, the response class the spinning cylinders acquire.","marker":"[40]"},{"why":"Supplies the spin-momentum locking property of evanescent guided waves that produces asymmetric directional coupling between the cylinders and the substrate.","marker":"[42]"},{"why":"Provides the multipole expansion used to compute chiral dipole coefficients $a_{+1}$ and $a_{-1}$ that explain the nonreciprocity.","marker":"[44]"},{"why":"Gives the generalized Snell law used to predict and verify the nonreciprocal beam-deflection angles.","marker":"[54]"}],"fun_headline_variants":["Spinning cylinders bend light one way, no magnets","Spun cylinders make light nonreciprocal without magnets","Relativistic spinning creates one-way light metasurface","No magnets: spinning cylinders isolate light 65%","Spinning silicon metasurface gives light a one-way ticket"],"cache_read_input_tokens":15616,"weakest_assumption_plain":"The whole prediction rests on treating a rigidly spinning submicron silicon cylinder as a moving medium governed by the Minkowski constitutive relations, with the same permittivity and no extra mechanical, thermal, or dispersive effects; if that effective description is wrong, the simulated one-way effects may not appear in a real device.","fun_headline_variants_meta":{"raw":{"variants":["Spinning cylinders bend light one way, no magnets","Spun cylinders make light nonreciprocal without magnets","Relativistic spinning creates one-way light metasurface","No magnets: spinning cylinders isolate light 65%","Spinning silicon metasurface gives light a one-way ticket"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000484,"raw_usage":{"total_tokens":2408,"prompt_tokens":983,"completion_tokens":1425,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":599,"completion_tokens_details":{"reasoning_tokens":1348}},"tokens_in":599,"tokens_out":1425,"duration_ms":11146,"temperature":1.0,"reasoning_tokens":1348,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T13:44:25.807681+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Build a test cell where a dielectric cylinder or disk of radius $R$ spins at $\\Omega R/c = 0.01$ next to a dielectric half-space and measure the transmission contrast for opposite incidence angles at the predicted resonance; if the measured isolation ratio falls far below the simulated 65 percent, the Minkowski constitutive assumption is the first suspect. A less demanding check is to compare the full-wave simulation against a rigorous moving-boundary or Doppler-shifted solver on a single cylinder and verify the predicted first-order splitting $\\Delta\\omega = 2m\\Omega(\\varepsilon\\mu-1)/(\\varepsilon\\mu)$.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Demonstrates nonreciprocal light propagation induced by a subwavelength spinning cylinder, giving the guided-mode coupling picture the metasurface builds on."},{"cited_title":"Mazor and A","cited_arxiv_id":null,"evidence_quote":"Introduces the gauge-field description of the Sagnac frequency shift in a rotating cavity, the basis for the effective azimuthal gauge field and chiral-mode splitting."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Analyses coupled spinning cylinders, providing the coupled-resonator framework used for the lattice."},{"cited_title":"Minkowski, Die grundgleichungen f¨ ur die elektromag- netischen vorg¨ ange in bewegten k¨ orpern, Nachrichten von der Gesellschaft der Wissenschaften zu G¨ ottingen, Math","cited_arxiv_id":null,"evidence_quote":"Defines Tellegen-type bianisotropic media, the response class the spinning cylinders acquire."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the spin-momentum locking property of evanescent guided waves that produces asymmetric directional coupling between the cylinders and the substrate."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the generalized Snell law used to predict and verify the nonreciprocal beam-deflection angles."}],"review_version":1}