betaFeedback

WorksPhysics and Astronomy

Geometry-Derived Capacitance Dyadics with Floquet Sideband Correction for Anisotropic Periodic Patch Metasurfaces

Abstract

Homogenised sheet models describe periodic metallic patch layers using a scalar sheet capacitance, which is only valid when the effective response is isotropic. An isotropic response exists for unit cells with patch geometries that have rotational symmetry of order three or higher. Lower-symmetry unit cells with irregular or random geometries generically require a full 2 x 2 surface-capacitance dyadic. We present a framework that computes this dyadic directly from the arbitrary gap-connected patch geometry on a unit cell of a hexagonal lattice. The two-dimensional Bravais periodicity fixes the two independent macroscopic loading directions, the connectivity of the gap ensures a full-rank response for the geometries considered here, and the point-group symmetry of the patch constrains the form of the dyadic. A quasi-static boundary value problem on a single periodic cell then determines its independent entries. We use the surface-capacitance dyadic in the anisotropic surface- wave dispersion relation, which includes the leading spatially-dispersive first shell Floquet-sideband correction. For each new patch geometry, the dyadic and the first-shell correction are obtained from a single quasi-static cell problem and from the reconstructed boundary. So anisotropic surface- wave dispersion follows without a per-geometry full-wave eigenmode band sweep or sheet-tensor retrieval. We validate this geometry-to-dispersion chain against full-wave eigenmode simulations across six representative symmetry classes, showing that the correction captures band flattening near the Brillouin-zone boundary and improves agreement relative to the local dyadic model.

Introduction

Metasurfaces (MTSs) are planar arrays of subwavelength scatterers that control the amplitude, phase, and polarization of electromagnetic waves through the engineered response of their unit cells [1, 2, 3, 4, 5]. Beyond shaping free-space wavefronts, suitably designed metasurfaces can also support and guide bound surface waves [6]. In a homogenised description, the MTS is represented by an effective sheet boundary condition rather than treated element by element [7, 8], in which the entire unit-cell geometry is reduced to a small set of effective parameters. For capacitive metallic patch arrays, the isotropic response is described by a scalar sheet capacitance, for which simple and accurate closed forms are available for canonical square grids [9, 10]. Such a scalar description underlies many analysis and synthesis approaches for surface-wave MTSs, from high-impedance surfaces [11] to modulated-MTS antennas [12]. The more general homogenisation challenge, however, is not merely to reproduce a particular electromagnetic observable, such as a dispersion curve, but to establish a direct mapping from the unit-cell geometry to its effective sheet response. Such a constitutive description can then be employed independently of the particular electromagnetic problem in which the sheet is embedded.

A scalar sheet capacitance, however, presumes that the induced current is parallel to the applied tangential field for every in-plane direction. This holds when the unit cell is sufficiently symmetric [13, 14]. A patch with three-fold or higher-rotational symmetry responds isotropically, whereas any lower-symmetry cell couples the two tangential field components and requires a full 2×22 \times2 surface-capacitance dyadic. Anisotropic sheet responses are an established design resource across the spectrum. At microwave frequencies, tensor holographic impedance surfaces shape aperture fields through modulated anisotropy [15, 16], and printed-circuit tensor impedance surfaces have been characterised by retrieving the effective sheet tensor from full-wave simulations [17], and spatially varying sheet response has been proposed as a platform for surface-wave transformation optics [18]. At optical frequencies, extreme sheet anisotropy produces hyperbolic surface-wave regimes with open isofrequency contours, topological transitions [19, 20, 21], and strongly enhanced light–matter interaction [22, 23, 24, 25], a regime realized not only in structured sheets but natively in anisotropy van der Waals materials [26, 27, 28]. Self-complementary metasurfaces can also exhibit dual-mode hyperbolicity and supercanalization [29]. Closed-form dispersion models exist for canonical anisotropic elements such as elliptical patches [30, 31].

Despite their different formulations, these approaches generally rely on canonical geometries, per-geometry full-wave characterization, or reduced models whose parameters must be determined for the specific structure under consideration. For general patch geometries, including irregular, multiply connected, or random shapes, a direct geometry-to-sheet relation is generally unavailable. A second limitation, common to both scalar and dyadic local sheet models, arises as the Bloch wavevector approaches the Brillouin-zone boundary. In this regime, the higher Floquet orders of the lattice are no longer deeply evanescent and the sheet response becomes non-local, i.e., spatially dispersive [32, 33, 34], with related non-local homogenization effects also arising in structured metallic media such as space–time modulated wire media [35]. Such non-locality is increasingly exploited as a design degree of freedom in its own right [36]. The local dispersion branch then departs from the actual Bloch band [37]. Multimodal equivalent-network models explicitly account for higher-order Floquet contributions to the response of periodic patch and aperture arrays [38, 39]. Accurate semianalytical Bloch approaches for square and hexagonal unit cells [40, 41] rely on full-wave unit-cell characterization, which the present geometry-driven framework avoids. Such approaches provide an accurate description of the Bloch spectrum, but they do not directly establish a reduced geometry-to-sheet relation from which both the local anisotropic response and its leading spatial-dispersion correction can be obtained.

The practical consequence is that obtaining the surface-wave band diagram for a new irregular patch still typically requires a full-wave eigenmode computation, or an equivalent multi-modal Floquet analysis re-built for that geometry. The purpose of this work is to replace that step. Once the patch polygon and its point-group class are given, the local capacitance dyadic and the leading Floquet correction are computed from geometry alone, and the dispersion relation is solved without full-wave simulation. Full-wave eigenmode data are used only for validation.

In this paper, we present a geometry-driven framework that constructs the local anisotropic surface-capacitance response and its leading nonlocal correction directly from a general gap-connected patch geometry on a hexagonal lattice. The framework distinguishes three aspects of the homogenization problem. First, the two-dimensional Bravais periodicity defines two independent macroscopic tangential loading directions, associated with the two primitive translation directions of the lattice. In the periodically identified unit cell, these correspond to the two independent winding directions of the flat-torus representation [42, 43]. For the patch geometries considered here, the gap remains connected across both periodic directions, so that both macroscopic loadings produce non-degenerate electrostatic responses and the resulting surface-capacitance dyadic is full rank. Second, the admissible form of the dyadic is dictated by the point-group symmetry of the complete unit-cell geometry [13, 44], spanning the symmetry classes from the fully symmetric hexagonal cell down to a patch with no point symmetry. Third, the numerical values of the allowed entries follow from a quasi-static boundary-value problem posed on a single periodic cell, based on homogenization theory for composite media [45, 46]. These steps establish the local effective sheet response from geometry before any particular propagation problem is introduced.

Surface-wave propagation is then used to develop and validate the electromagnetic consequences of this effective description. The resulting dyadic drives the anisotropic surface-wave dispersion through its projection onto the transverse-electric (TE) polarization of the guided mode. Near the Brillouin-zone boundary, the six Floquet orders of the first reciprocal shell are retained explicitly [38, 39]. Their coupling to the fundamental guided mode is computed from the reconstructed patch boundary. A boundary Fourier overlap, motivated by coupled-mode theory [47, 48, 49], determines the coupling strength of each retained order, while the discrete turning of the boundary, where the induced edge charge concentrates [50], provides its directional weight. The detuning of each order is determined by its own sheet-loaded decay constant. Mirror symmetries of the patch enter through the coherent pairing of the sidebands, as in the symmetry analysis of photonic crystal bands [51, 52]. Thus, both the local anisotropic response and its leading spatial-dispersion correction are derived from geometric quantities rather than fitted to the Bloch spectrum. The resulting chain from geometry to dispersion is validated against full-wave eigenmode simulations for representative structures spanning the symmetry classes considered in this work.

The paper is organized as follows. Section II develops the geometry-driven surface-capacitance dyadic. The two-dimensional periodicity defines the macroscopic loading space. The connectivity of the gap ensures a non-degenerate response. The tensor form is constrained by symmetry. A quasi-static cell problem determines the independent entries and their calibration to an equivalent sheet capacitance. Section III uses the effective sheet-capacitance dyadic to derive the local anisotropic surface-wave dispersion relation and its quasi-TE reduction. Section IV develops the geometry-derived first-shell Floquet correction, incorporating the coupling to higher-order harmonics through boundary Fourier overlaps and directional turning measures. Section V validates the resulting model against full-wave eigenmode simulations across the symmetry classes considered, and Section VI concludes.

Geometry-driven capacitance dyadic for arbitrary gap-connected patch-type metasurfaces

We represent the patch-type metasurface (MTS) in the quasi-static regime by an infinitesimally thin, purely electric sheet at z=0z = 0. The macroscopic fields satisfy the jump conditions [7, 8]

z^×[H]=Js=Y‾‾s⋅Et,\hat{\mathbf{z}} \times[\mathbf{H}] = \mathbf{J}_{s} = \overline{\overline{Y}}_{s} \cdot\mathbf{E}_{t},
z^×[E]=0.\hat{\mathbf{z}} \times[\mathbf{E}] = 0.

Here [F]=F+−F−[\mathbf{F}] = \mathbf{F}^{+} - \mathbf{F}^{-} denotes the field jump, with F+\mathbf{F}^{+} and F−\mathbf{F}^{-} evaluated on the upper and lower sides of the sheet, respectively. The quantities Js\mathbf{J}_{s} and Y‾‾s\overline{\overline{Y}}_{s} are the equivalent electric surface-current density and the sheet-admittance dyadic. Equation (1b) implies continuity of the tangential electric field, Et+=Et−≡Et\mathbf{E}_{t}^{+} = \mathbf{E}_{t}^{-} \equiv\mathbf{E}_{t}.

For a lossless reciprocal patch array in the quasi-static capacitive regime, we write the induced tangential surface polarization as

Ps,t=C‾‾eqv⋅Et,(1)\mathbf{P}_{s,t} = \overline{\overline{C}}_{\mathrm{eqv}} \cdot\mathbf{E}_t, \tag*{(1)}

where C‾‾eqv\overline{\overline{C}}_{\mathrm{eqv}} is the equivalent surface-capacitance dyadic, in units of F. With the ejωte^{j\omega t} time convention, the associated polarization current is Js=jωPs,t\mathbf{J}_s = j\omega\mathbf{P}_{s,t}. Substitution into Eq. (1a) gives

Y‾‾s=jωC‾‾eqv.(2)\overline{\overline{Y}}_s = j\omega\overline{\overline{C}}_{\mathrm{eqv}}. \tag*{(2)}

Thus, C‾‾eqv\overline{\overline{C}}_{\mathrm{eqv}} is the equivalent sheet quantity entering the macroscopic electromagnetic boundary conditions. In the following, we determine this dyadic directly from the unit-cell geometry for gap-connected patch arrays on a hexagonal lattice.

An example of the unit-cell geometry considered in this work is shown in Figure 1. The metallic (i.e., patch) region Σ\Sigma repeats on a hexagonal lattice generated by the primitive vectors a1\mathbf{a}_1 and a2\mathbf{a}_2, and the gap is Ω=U∖Σ\Omega= U \setminus\Sigma, with UU being the entire unit cell. For a patch having three-fold rotation symmetry (C3C_3) or higher, the sheet response is a scalar surface capacitance [9, 10]. For a patch with lower symmetry, the induced charge depends on the direction of the applied tangential field, and the response is generally described by a symmetric 2×22 \times2 surface-capacitance dyadic [15, 16].

Geometry of the periodic patch layer

Figure 1. An example of the geometry of the periodic patch layer. The unit cell UU (black outline) repeats on a hexagonal lattice generated by the primitive vectors a1\mathbf{a}_1 and a2\mathbf{a}_2. Inside the unit cell, the metallic patch Σ\Sigma (dark blue) is a random geometry, and Ω=U∖Σ\Omega= U \setminus\Sigma is the gap.

To determine C‾‾eqv\overline{\overline{C}}_{\mathrm{eqv}} from this geometry, the quasi-static cell problem first produces the auxiliary two-dimensional dyadic

C‾‾=[CxxCxyCxyCyy].(3)\overline{\overline{C}} = \begin{bmatrix} C_{xx} & C_{xy} \\ C_{xy} & C_{yy} \end{bmatrix}. \tag*{(3)}

in units of (F/m). Its entries characterize the electric energy stored in Ω\Omega under tangential excitation along the two in-plane directions (Sec. II C). Because the quasi-static cell problem is two-dimensional, C‾‾\overline{\overline{C}} is a per-unit-length quantity and is not yet the equivalent sheet capacitance appearing in Eq. (2). The calibration introduced in Sec. II D converts it into equivalent sheet-capacitance dyadic C‾‾eqv\overline{\overline{C}}_{\mathrm{eqv}}, in units of (F), which describes the electromagnetic response of the patch layer. The auxiliary dyadic C‾‾\overline{\overline{C}} and its calibrated counterpart C‾‾eqv\overline{\overline{C}}_{\mathrm{eqv}} therefore characterize the effective quasi-static response of the unit-cell geometry, with C‾‾eqv\overline{\overline{C}}_{\mathrm{eqv}} providing the geometry-dependent input to the local surface-wave dispersion model of Sec. III.

The derivation relies on three complementary elements. The two-dimensional Bravais periodicity defines two independent macroscopic tangential loading directions, while connectivity of the gap across both periodic directions ensures that the corresponding quasi-static response is non-degenerate (Sec. II A). The point-group symmetry of the patch constrains the admissible form of the dyadic (Sec. II B), and a quasi-static boundary-value problem posed on a single periodic unit cell determines its independent entries (Sec. II C). Once C‾‾\overline{\overline{C}} has been obtained and calibrated, C‾‾eqv\overline{\overline{C}}_{\mathrm{eqv}} completely determines the local homogenised response. The nonlocal Floquet correction of Sec. IV additionally requires the reconstructed boundary ∂Ω\partial\Omega, which supplies the first-shell overlaps and directional turning measures.

Periodic loading space and non-degenerate capacitance response

We consider a periodic patch-type MTS invariant under the two-dimensional Bravais lattice generated by the primitive translation vectors a1\mathbf{a}_1 and a2\mathbf{a}_2,

Rn1n2=n1a1+n2a2,n1,n2∈Z,(4)\mathbf{R}_{n_1 n_2} = n_1 \mathbf{a}_1 + n_2 \mathbf{a}_2,\qquad n_1,n_2 \in\mathbb{Z}, \tag*{(4)}

as depicted in Figure 1. The infinite periodic structure can be represented exactly by a single primitive unit cell equipped with periodic boundary conditions. Identifying opposite edges of the cell maps the periodically repeated plane onto a flat torus [42], providing a compact representation of the infinite periodic problem [Figure 2(a)].

Periodic loading directions and connectivity of the surface-capacitance problem

Figure 2. Periodic loading directions and connectivity of the surface-capacitance problem. (a) Identifying opposite edges of the unit cell maps the periodic plane onto a flat torus. The two highlighted winding cycles, γ1\gamma_1 (red) and γ2\gamma_2 (blue), correspond to the two primitive translation directions of the Bravais lattice and therefore to the two independent macroscopic tangential loadings. (b) Representative hexagonal patch geometry for which the gap Ω=U∖Σ\Omega= U \setminus\Sigma remains connected across both periodic directions. The existence of winding gap paths in both directions allows independent macroscopic potential variations to be sustained and gives a non-degenerate 2×22 \times2 capacitance response for the class of geometries considered here.

The significance of this construction for the present homogenisation problem is that a two-dimensional periodic sheet admits two independent macroscopic tangential loading directions. These may be represented by the two primitive lattice directions, or equivalently by any two independent Cartesian directions in the tangential plane. A prescribed macroscopic electrostatic loading therefore has two independent components, and the corresponding linear effective response is naturally represented by a 2×22 \times2 surface-capacitance dyadic C‾‾\overline{\overline{C}}.

In the torus representation shown in Figure 2(a), the primitive periodic directions are associated with two independent non-contractible winding cycles, γ1\gamma_1 and γ2\gamma_2, corresponding respectively to translations along a1\mathbf{a}_1 and a2\mathbf{a}_2 [42]. These cycles identify the two independent macroscopic directions in which a potential variation may be imposed across the periodically identified cell.

The topology of the gap region enters separately through its connectivity. Let

Ω=U∖Σ(5)\Omega= U \setminus\Sigma \tag*{(5)}

denote the unmetallized area of the unit-cell gap, where UU is the complete unit cell and Σ\Sigma is the metallic patch. A capacitive response to a macroscopic loading in a given periodic direction requires the gap region to remain connected across the cell in that direction, so that a macroscopic potential variation can be sustained. For every geometry considered in this work, Ω\Omega contains connected winding paths across both primitive directions, as illustrated schematically in Figure 2(b).

Consequently, neither of the two independent macroscopic loadings is suppressed by the geometry, and the corresponding two column responses of the capacitance dyadic remain linearly independent. Therefore for this class of patch geometries considered here,

rank⁡(C‾‾)=2.(6)\operatorname{rank}\left(\overline{\overline{C}}\right) = 2. \tag*{(6)}

Thus, the two-dimensional Bravais periodicity fixes the dimension of the macroscopic tangential loading space, whereas connectivity of the gap across both periodic directions ensures that the resulting 2×22 \times2 capacitive response is non-degenerate.

It is useful to distinguish these macroscopic winding directions from the complete topology of the gap domain. Additional holes, indentations, or disconnected metallic components may introduce further closed paths within Ω\Omega, but they do not increase the dimension of the imposed macroscopic tangential loading space, which remains two-dimensional. Such geometric features instead modify the solution of the quasi-static cell problem and therefore the numerical entries of C‾‾\overline{\overline{C}}.

Equation (7) fixes only the number of independent macroscopic responses; it does not prescribe the individual entries of C‾‾\overline{\overline{C}}. The admissible tensor form is instead constrained by the point-group symmetry of the complete unit-cell geometry, as discussed in Sec. II B and summarised in Table I and Figure 3. The numerical values of the independent entries are subsequently determined by the quasi-static cell problem of Sec. II C.

Directional-capacitance indicators for four symmetry classes

Figure 3. Directional-capacitance indicator C(u^)=u^⋅C‾⋅u^/C‾C(\hat{\mathbf{u}})=\hat{\mathbf{u}}\cdot\overline{C}\cdot\hat{\mathbf{u}}/\overline{C}, normalised by the mean eigenvalue C‾=12tr⁡(C‾)\overline{C}=\frac{1}{2}\operatorname{tr}(\overline{C}), for the four symmetry classes of Table I. The light-blue filled region bounded by the solid blue curve is the directional indicator; purple and orange arrows mark the two principal eigenvectors scaled by their normalised eigenvalues; the dashed grey circle is the isotropic reference (λ1/λ2=1)(\lambda_1/\lambda_2=1). The inset in each panel shows a representative patch geometry Σ\Sigma belonging to that symmetry class. (a) C6v/C3vC_{6v}/C_{3v}: full rotational symmetry forces C‾=CI\overline{C}=C I, giving a perfect circle. (b) C2vC_{2v}: two distinct eigenvalues with eigenvectors aligned to the lattice axes, giving an axis-aligned directional-capacitance indicator. (c) CsC_s: eigenvectors rotated to the mirror-axis orientation φ\varphi, giving a rotated directional-capacitance indicator. (d) C1C_1: a nonzero off-diagonal entry Cxy=CyxC_{xy}=C_{yx} rotates the principal axes freely; at the strong anisotropy of this example the indicator develops a pronounced waist, reflecting the fully general three-parameter dyadic.

Capacitance dyadic and symmetry constraints: group-theoretic tensor form

Let GG denote the in-plane point group of the complete unit-cell geometry, including both the patch and the underlying lattice. The shape of the patch Σ\Sigma determines the symmetry of the structure. Since the hexagonal lattice can have at most six-fold symmetry, GG is always a subgroup of C6vC_{6v}.

The point groups relevant here are the subgroups of the regular hexagonal cell. The symmetry classes relevant to the present analysis are C1C_1, CsC_s, C2C_2, C2vC_{2v}, C3C_3, C3vC_{3v}, C6C_6, and C6vC_{6v}. In Schönflies notation [44], CnC_n denotes a group generated by an nn-fold rotation about the axis normal to the MTS, whereas CnvC_{nv} additionally contains vertical mirror symmetries. The groups CsC_s and C1C_1 denote, respectively, a single-mirror group and the trivial group containing only the identity. Since CnC_n denotes the cyclic group generated by a rotation of 2π/n2\pi/n about the axis normal to the sheet, Cn={E,R2π/n,…,R2π(n−1)/n}C_n = \{E, R_{2\pi/n}, \ldots, R_{2\pi(n-1)/n}\} has order nn, where EE is the identity. Adding a mirror plane containing that axis (a vertical mirror, σv\sigma_v) generates a group CnvC_{nv}, of the order 2n2n, which contains the nn rotations of CnC_n together with nn vertical mirror reflections. The group Cs={E,σ}C_s = \{E, \sigma\} contains the identity and a single mirror, and C1={E}C_1 = \{E\} is the trivial group with no symmetry beyond the identity. The subgroup chain relevant to a hexagonal lattice is therefore

C1⊂C2, Cs⊂C2v, C3⊂C3v, C6⊂C6v(7)C_1 \subset C_2,\ C_s \subset C_{2v},\ C_3 \subset C_{3v},\ C_6 \subset C_{6v} \tag*{(7)}

References

  1. [1]N. Yu, P. Genevet, M. A. Kats, F. Aieta, J.-P. Tetienne, F. Capasso, and Z. Gaburro, Science 334, 333 (2011).
  2. [2]A. V. Kildishev, A. Boltasseva, and V. M. Shalaev, Science 339, 1232009 (2013).
  3. [3]N. Yu and F. Capasso, Nature Materials 13, 139 (2014).
  4. [4]A. I. Kuznetsov, A. E. Miroshnichenko, M. L. Brongersma, Y. S. Kivshar, and B. Luk’yanchuk, Science 354, aag2472 (2016).
  5. [5]W. T. Chen, A. Y. Zhu, and F. Capasso, Nature Reviews Materials 5, 604 (2020).
  6. [6]S. Sun, Q. He, S. Xiao, Q. Xu, X. Li, and L. Zhou, Nature Materials 11, 426 (2012).
  7. [7]E. Kuester, M. Mohamed, M. Piket-May, and C. Holloway, IEEE Transactions on Antennas and Propagation 51, 2641 (2003).
  8. [8]C. L. Holloway, E. F. Kuester, J. A. Gordon, J. O’Hara, J. Booth, and D. R. Smith, IEEE Antennas and Propagation Magazine 54, 10 (2012).
  9. [9]S. Tretyakov, Analytical Modeling in Applied Electromagnetics, Artech House electromagnetic analysis series (Artech House, 2003).
  10. [10]O. Luukkonen, C. Simovski, G. Granet, G. Gousssetis, D. Lioubtchenko, A. V. Raisanen, and S. A. Tretyakov, IEEE Transactions on Antennas and Propagation 56, 1624 (2008).
  11. [11]D. Sievenpiper, L. Zhang, R. Broas, N. Alexopolous, and E. Yablonovitch, IEEE Transactions on Microwave Theory and Techniques 47, 2059 (1999).
  12. [12]T. Smajic, G. Minatti, M. Casaletti, and M. Bosiljevac, IEEE Antennas and Wireless Propagation Letters 10, 1499 (2011).
  13. [13]J. Nye, Physical Properties of Crystals: Their Representation by Tensors and Matrices, Oxford science publications (Clarendon Press, 1985).
  14. [14]W. J. Padilla, Opt. Express 15, 1639 (2007).
  15. [15]B. H. Fong, J. S. Colburn, J. J. Ottusch, J. L. Visher, and D. F. Sievenpiper, IEEE Transactions on Antennas and Propagation 58, 3212 (2010).
  16. [16]R. Quarfoth and D. Sievenpiper, IEEE Transactions on Antennas and Propagation 61, 3597 (2013).
  17. [17]A. M. Patel and A. Grbic, IEEE Transactions on Antennas and Propagation 61, 211 (2013).
  18. [18]A. Vakil and N. Engheta, Science 332, 1291 (2011).
  19. [19]G. Hu, Q. Ou, G. Si, Y. Wu, J. Wu, Z. Dai, A. Krasnok, Y. Mazor, Q. Zhang, Q. Bao, C.-W. Qiu, and A. Alù, Nature 582, 209 (2020).
  20. [20]M. Chen, X. Lin, T. H. Dinh, Z. Zheng, J. Shen, Q. Ma, H. Chen, P. Jarillo-Herrero, and S. Dai, Nature Materials 19, 1307 (2020).
  21. [21]H. Hu, N. Chen, H. Teng, R. Yu, Y. Qu, J. Sun, M. Xue, D. Hu, B. Wu, C. Li, J. Chen, M. Liu, Z. Sun, Y. Liu, P. Li, S. Fan, F. J. García de Abajo, and Q. Dai, Nature Nanotechnology 17, 940 (2022).
  22. [22]A. Poddubny, I. Iorsh, P. Belov, and Y. Kivshar, Nature Photonics 7, 948 (2013).
  23. [23]J. S. Gomez-Diaz, M. Tymchenko, and A. Alù, Phys. Rev. Lett. 114, 233901 (2015).
  24. [24]A. A. High, R. C. Devlin, A. Dibos, M. Polking, D. S. Wild, J. Perczel, N. P. de Leon, M. D. Lukin, and H. Park, Nature 522, 192 (2015).
  25. [25]O. Y. Yermakov, A. I. Ovcharenko, M. Song, A. A. Bogdanov, I. V. Iorsh, and Y. S. Kivshar, Phys. Rev. B 91, 235423 (2015).
  26. [26]D. N. Basov, M. M. Fogler, and F. J. G. de Abajo, Science 354, aag1992 (2016).
  27. [27]W. Ma, P. Alonso-González, S. Li, A. Y. Nikitin, J. Yuan, J. Martín-Sánchez, J. Taboada-Gutiérrez, I. Amenabar, P. Li, S. Vélez, C. Tollan, Z. Dai, Y. Zhang, S. Sriram, K. Kalantar-Zadeh, S.-T. Lee, R. Hillenbrand, and Q. Bao, Nature 562, 557 (2018).
  28. [28]A. Nemilentsau, T. Low, and G. Hanson, Phys. Rev. Lett. 116, 066804 (2016).
  29. [29]E. Martini, F. Giusti, A. Benini, and S. Maci, Nanophotonics 12, 2973 (2023).
  30. [30]H. Bilow, IEEE Transactions on Antennas and Propagation 51, 2788 (2003).
  31. [31]M. Mencagli, E. Martini, and S. Maci, IEEE Transactions on Antennas and Propagation 63, 2992 (2015).
  32. [32]P. A. Belov, R. Marqués, S. I. Maslovski, I. S. Nefedov, M. Silveirinha, C. R. Simovski, and S. A. Tretyakov, Phys. Rev. B 67, 113103 (2003).
  33. [33]D. Correas-Serrano, J. S. Gomez-Diaz, M. Tymchenko, and A. Alù, Opt. Express 23, 29434 (2015).
  34. [34]K. Shastri and F. Monticone, Nature Photonics 17, 36 (2023).
  35. [35]A. B. Yakovlev, M. G. Silveirinha, A. Pal, and M. H. Zadeh, arXiv preprint arXiv:2607.27362 10.48550/arXiv.2607.27362 (2026), arXiv:2607.27362 [physics.optics].
  36. [36]A. C. Overvig, S. C. Malek, and N. Yu, Phys. Rev. Lett. 125, 017402 (2020).
  37. [37]M. Qiu, M. Jia, S. Ma, S. Sun, Q. He, and L. Zhou, Phys. Rev. Appl. 9, 054050 (2018).
  38. [38]R. Rodríguez-Berral, F. Mesa, and F. Medina, IEEE Transactions on Antennas and Propagation 63, 1969 (2015).
  39. [39]F. Mesa, R. Rodríguez-Berral, M. García-Vigueras, F. Medina, and J. R. Mosig, IEEE Transactions on Antennas and Propagation 64, 1106 (2016).
  40. [40]F. Giusti, Q. Chen, F. Mesa, M. Albani, and O. Quevedo-Teruel, IEEE Transactions on Antennas and Propagation 70, 5555 (2022).
  41. [41]M. Petek, J. Rico-Fernández, J. A. T. Vásquez, G. Valerio, F. Mesa, O. Quevedo-Teruel, and F. Vipiana, IEEE Transactions on Antennas and Propagation 73, 1650 (2025).
  42. [42]M. Nakahara, Geometry, Topology and Physics (CRC Press, 2018).
  43. [43]T. Ozawa, H. M. Price, A. Amo, N. Goldman, M. Hafezi, L. Lu, M. C. Rechtsman, D. Schuster, J. Simon, O. Zilberberg, and I. Carusotto, Rev. Mod. Phys. 91, 015006 (2019).
  44. [44]M. Dresselhaus, G. Dresselhaus, and A. Jorio, Group Theory: Application to the Physics of Condensed Matter (Springer Berlin Heidelberg, 2007).
  45. [45]D. J. Bergman, Physics Reports 43, 377 (1978).
  46. [46]G. W. Milton, The Theory of Composites, Cambridge Monographs on Applied and Computational Mathematics (Cambridge University Press, 2002).
  47. [47]H. Kogelnik and C. V. Shank, Journal of Applied Physics 43, 2327 (1972).
  48. [48]A. Yariv, IEEE Journal of Quantum Electronics 9, 919 (1973).
  49. [49]H. Haus, Waves and Fields in Optoelectronics, Prentice-Hall series in solid state physical electronics (Prentice-Hall, 1984).
  50. [50]J. Meixner, IEEE Transactions on Antennas and Propagation 20, 442 (1972).
  51. [51]K. Sakoda, Optical Properties of Photonic Crystals, Springer Series in Optical Sciences (Springer Berlin Heidelberg, 2004).
  52. [52]T. Ochiai and K. Sakoda, Phys. Rev. B 63, 125107 (2001).
  53. [53]S. Haussühl, Physical Properties of Crystals: An Introduction (Wiley, 2008).
  54. [54]See Supplemental Material at [URL will be inserted by publisher] for additional derivations, numerical details, normalization procedure, and Floquet-sideband correction.
  55. [55]F. Giusti, S. Maci, and E. Martini, IEEE Transactions on Antennas and Propagation (Under Review) , 1 (2026).
  56. [56]G. Allaire, SIAM Journal on Mathematical Analysis 23, 1482 (1992).
  57. [57]C. L. Holloway and E. F. Kuester, IEEE Transactions on Antennas and Propagation 64, 4671 (2016).
  58. [58]J. Jackson, Classical Electrodynamics (Wiley, 1998).
  59. [59]N. J. Higham, Functions of Matrices (Society for Industrial and Applied Mathematics, 2008).
  60. [60]J. S. Gomez-Diaz, M. Tymchenko, and A. Alù, Opt. Mater. Express 5, 2313 (2015).
  61. [61]L. Brillouin, Wave Propagation in Periodic Structures: Electric Filters and Crystal Lattices, Chemische Reihe. Lehrbucher und Monographien aus dem Gebiete der exakten Wissenschaften (Dover Publications, 1953).
  62. [62]B. Munk, Frequency Selective Surfaces: Theory and Design (Wiley, 2005).
  63. [63]R. Collin, Field Theory of Guided Waves, IEEE Press Series on Electromagnetic Wave Theory and Applications (Wiley, 1990).
  64. [64]U. Hohenester and J. Krenn, Phys. Rev. B 72, 195429 (2005).
  65. [65]I. Silveiro, J. M. Plaza Ortega, and F. J. García de Abajo, New Journal of Physics 17, 083013 (2015).
  66. [66]E. Baladi and S. V. Hum, IEEE Transactions on Antennas and Propagation 69, 5691 (2021).
  67. [67]J. G. Van Bladel, Singular Electromagnetic Fields and Sources, IEEE Series on Electromagnetic Wave Theory (IEEE Press / John Wiley & Sons, New York, NY, 1996).DOI
  68. [68]J. W. Milnor, Annals of Mathematics 52, 248 (1950).
  69. [69]F. Zhang, The Schur Complement and Its Applications, Numerical Methods and Algorithms No. 4 (Springer, 2005).DOI
  70. [70]L. P. Bouckaert, R. Smoluchowski, and E. Wigner, Phys. Rev. 50, 58 (1936).
  71. [71]C. J. Bradley and A. P. Cracknell, The Mathematical Theory Of Symmetry In Solids: Representation theory for point groups and space groups (Oxford University Press, 2009).

Paper details

vlnc:W_5akeecrd
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC0-1.0
Usage

views · 3

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents