Table of Contents
Fetching ...

Giant bias-free nonreciprocity for unpolarized light via synthetic motion

L. M. Máñez-Espina, B. Amrahi, I. Faniayeu, R. Cichelero, A. Dmitriev, A. Díaz-Rubio, V. S. Asadchy

Abstract

Reciprocity breaking at optical frequencies typically relies on bulky magnets, dynamic modulation, or nonlinearities, all of which hinder chip-scale integration and the handling of unpolarised light. We introduce a fully passive, subwavelength metasurface that achieves polarisation-insensitive one-way transparency by combining self-magnetised ferrite nanodisks in a vortex state with symmetry-protected quasi-bound states in the continuum. The metasurface exhibits a pure synthetic moving-medium response at optical frequencies, yielding giant nonreciprocal directional dichroism. We report near-unity values for both the transmittance contrast and the emissivity-to-absorptivity ratio with experimentally widely available ferrite materials, all under unpolarised illumination and without external bias. Using temporal coupled-mode theory, we identify the design conditions necessary to maximise directional dichroism: critical coupling, Huygens-type resonance overlap, and strong inter-mode coupling. Furthermore, we propose a deterministic, stamp-assisted protocol for imprinting arbitrary, uniform, or patterned vortex configurations across large arrays of nanodisk meta-atoms, enabling scalable fabrication. This work establishes a practical route toward compact nonreciprocal photonics with applications in photonic gyrators, nonreciprocal wavefront engineering, and nonreciprocal solar cell technologies.

Giant bias-free nonreciprocity for unpolarized light via synthetic motion

Abstract

Reciprocity breaking at optical frequencies typically relies on bulky magnets, dynamic modulation, or nonlinearities, all of which hinder chip-scale integration and the handling of unpolarised light. We introduce a fully passive, subwavelength metasurface that achieves polarisation-insensitive one-way transparency by combining self-magnetised ferrite nanodisks in a vortex state with symmetry-protected quasi-bound states in the continuum. The metasurface exhibits a pure synthetic moving-medium response at optical frequencies, yielding giant nonreciprocal directional dichroism. We report near-unity values for both the transmittance contrast and the emissivity-to-absorptivity ratio with experimentally widely available ferrite materials, all under unpolarised illumination and without external bias. Using temporal coupled-mode theory, we identify the design conditions necessary to maximise directional dichroism: critical coupling, Huygens-type resonance overlap, and strong inter-mode coupling. Furthermore, we propose a deterministic, stamp-assisted protocol for imprinting arbitrary, uniform, or patterned vortex configurations across large arrays of nanodisk meta-atoms, enabling scalable fabrication. This work establishes a practical route toward compact nonreciprocal photonics with applications in photonic gyrators, nonreciprocal wavefront engineering, and nonreciprocal solar cell technologies.
Paper Structure (10 sections, 7 equations, 5 figures, 2 tables)

This paper contains 10 sections, 7 equations, 5 figures, 2 tables.

Figures (5)

  • Figure 1: Different dichroism effects and their physical origins.a, polarisation-insensitive nonreciprocal directional dichroism, characterised by differing transmissivity for light incident from opposite directions. This effect arises from the synthetic moving medium mechanism and requires the breaking of both parity and time-reversal symmetries. In this work, the synthetic moving medium effect is realised using vortex magnetisation ${\bf M\mit}$, as indicated by the circular arrows within the material slab. b, Natural circular dichroism, which occurs in reciprocal materials with broken parity symmetry, resulting in direction-independent preferential absorption of light with a specific polarisation handedness. The illustration shows a chiral material slab embedded with helical inclusions. c, Magnetic circular dichroism, observed in nonreciprocal materials with broken time-reversal symmetry. As in b, light of a particular handedness predominantly transmits through the slab; however, the handedness that is transmitted depends on the direction of incidence. An example is illustrated with a Faraday material slab composed of an array of ferromagnetic cylinders magnetised by an external magnetic field ${\bf B\mit}$ oriented out-of-plane.
  • Figure 2: Comparison of dichroism observables and the different contributing electromagnetic effects. Green ticks and red crosses indicate whether a specific electromagnetic effect contributes to a particular observable. Here, 'NDD' stands for nonreciprocal directional dichroism, 'MCD' for magnetic circular dichroism, and 'NCD' for natural circular dichroism.
  • Figure 3: Electromagnetic response of a bias-free metasurface supporting vortex-type static magnetisation and quasi-BICs. a, Metasurface geometry (left) and static magnetisation within each nanodisk (right). The dashed contour depicts the square unit cell. The neighboring nanodisks have slightly different diameters $l_{\rm a}$ and $l_{\rm b}$ and are arranged in the checkerboard order, enabling the excitation of quasi-BICs in the metasurface. The colorbar depicts the amplitude of the normalised magnetisation ${\bf M\mit}/M_{\rm s}$ inside the nanodisk, while the arrows show the orientation of the local magnetisation. b, Field profiles of the quasi-BICs orthogonal eigenmodes excited in the metasurface unit cell in the absence of magnetisation. The left and right plots depict an electric (TE) and magnetic (TM) dipole modes, respectively. c, Scattering parameters of the metasurface consisting of idealistic ferrite material when illuminated normally with unpolarised light. Both simulated data and the theoretical data from the TCMT model are shown. The chosen ferrite parameters are dispersionless with $\varepsilon_r=8$, $\varepsilon_i=0.01$, and $\varepsilon_a=0.1$. Geometric parameters of the metasurface are, in terms of the period, $D=0.6446p$, $h=0.3261p$, and $\Delta=l_a-l_b=0.0557p$, with diameters defined as $l_a=D+\Delta/2$ and $l_b=D-\Delta/2$. The wavelength in the plot is normalised by the metasurface period $p$. d, Absolute values of the normalised collective polarizabilities of the metasurface unit cell extracted from the simulated scattering coefficients. e, Simulated directional and circular dichroisms of the metasurface. Indices "i" and "a" correspond to dichroism quantities defined in the text in terms of intensities and field amplitudes, respectively. f, Emissivities and absorptivities of the metasurface for two illumination directions. g, Transmittance and transmission phase difference between opposite directions of propagation for the same metasurface as in c but in the absence of material losses ($\varepsilon_{\rm i}=0$).
  • Figure 4: Micromagnetic simulations of BIG nanodisks. a, Phase diagram of magnetic states in a single BIG nanodisk for different values of its diameter $D$ and height $h$. The color denotes the ratio between the $z$- and $x$-components of the vorticity on a logarithmic scale. A larger ratio corresponds to a greater in-plane curl and therefore a more vortex-like state. The white dashed lines depict the approximate boundaries between different magnetic states in the diagram. The insets depict principal states, where the arrows depict the directions of normalised local magnetisation ${\bf m\mit}={\bf M\mit}/M_{\rm s}$ and the arrow color denotes the $m_x$ component. The star refers to the point with optimised dimensions of the nanodisks for the metasurface design. b, Spatial distribution of the normalised magnetisation component $m_x$ in the single nanodisk with optimised parameters, $D=269$ nm and $h=135$ nm, corresponding to the point marked by the star in a. The nanodisk requires no external magnetisation after the vortex state is established. c, Hysteresis plot for the nanodisk in b for varying applied external magnetic field oriented along the $z$-direction. d-h, An original methodology for generating identical vortex states in nanodisk arrays. Here, the nanodisks have the same dimensions as the nanodisk in b. The method is applied to an initial $3 \times 3$ array using a micromagnetic simulator, with magnetisation configurations extracted at each step. The colormap shows the normalised magnetisation component $m_x$ in the nanodisks and nanobars at different stages of the proposed magnetisation procedure. The array of rectangular bars above the nanodisk array represents an auxiliary stamp metasurface.
  • Figure 5: Self-magnetised metasurface with BIG nanodisks.a, Magnetisation field for a $5 \times 5$ array of BIG nanodisks with the same unit cell configuration as in Fig. \ref{['fig:Panel2']}a. The nanodisks require no external magnetisation. The arrows depict the directions of normalised local magnetisation ${\bf m\mit}={\bf M\mit}/M_{\rm s}$ and the arrow color denotes the $m_x$ component. The green dashed contour depicts the unit cell. Geometric parameters of the metasurface are $p=435$ nm, $D=269$ nm, $h=135$ nm, $\Delta=25$ nm. Material parameters in the considered wavelength range: $\varepsilon_{\rm r} = 8.077-0.016j$, $\varepsilon_{\rm a} = 0.0733-0.007j$Meta-BiYIG-material. b, Scattering parameters of the metasurface consisting of BIG nanodisks when illuminated normally with unpolarised light. Both simulated data and the theoretical data from the TCMT model are shown. Subscripts "co" and "cr" denote co- and cross-polarised coefficients in the linear polarisation basis. c, Simulated directional and circular dichroisms of the metasurface. d, Emissivities and absorptivities of the metasurface for two illumination directions. e, Absolute values of the normalised collective polarizabilities of the metasurface unit cell extracted from the simulated scattering coefficients.