Visible-Frequency Metasurfaces for Broadband Anomalous Reflection and High-Efficiency Spectrum Splitting

Abstract

Ultrathin metasurfaces have recently emerged as promising materials that have huge potential to enable novel, flat optical components, and surface-confined, miniature photonic devices. Metasurfaces offer new degrees of freedom in molding the optical wavefronts by introducing abrupt and drastic changes in the amplitude, phase, and/or polarization of electromagnetic radiation at the wavelength scale. By carefully arranging multiple subwavelength anisotropic or gradient optical resonators, metasurfaces have been shown to enable anomalous transmission, anomalous reflection, optical holograms, and spin-orbit interaction. However, experimental realization of high-performance metasurfaces that can operate at visible frequency range has been a significant challenge due to high optical losses of plasmonic materials and difficulties in fabricating several plasmonic resonators of subwavelength size with high uniformity. Here, we propose a highly efficient yet a simple metasurface design comprising of a single, anisotropic silver antenna in its unit cell. We demonstrate broadband (450-850 nm) anomalous reflection and spectrum splitting at visible and near-IR frequencies with high conversion efficiency. Average power ratio of anomalous reflection to the strongest diffraction mode was calculated to be on the order of 10(3) and measured to be on the order of 10. The anomalous reflected photons have been visualized using a charge-coupled device camera, and broadband spectrum splitting performance has been confirmed experimentally using a free space, angle-resolved reflection measurement setup. Metasurface design proposed in this study is a clear departure from conventional metasurfaces utilizing multiple, anisotropic and/or gradient optical resonators and could enable high-efficiency, broadband metasurfaces for achieving flat high signal-to-noise ratio optical spectrometers, polarization beam splitters, directional emitters, and spectrum splitting surfaces for photovoltaics.

Tailor the Functionalities of Metasurfaces Based on a Complete Phase Diagram

“…Recently, metasurfaces in metal/insulator/metal (MIM) configuration have been widely used in photonic research [2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18]. Such structures typically consist of a layer of planar metallic resonators and a continuous metallic film separated by a dielectric spacer (see Fig.…”

“…1(a)). Based on MIM metasurfaces, a variety of wave-manipulation effects have been realized in a wide frequency range, including perfect absorption [2][3][4][5], polarization control [6][7][8], phase modulation [9][10][11], anomalous light reflection [12][13][14], flat-lens focusing [15,16], and holograms [17,18].…”

“…However, whereas full-wave simulations can well reproduce the discovered phenomena on such systems [2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20], the inherent physics underlying these distinct effects remains obscure. In particular, why and with what parameters can these MIM metasurfaces exhibit certain functionalities (e.g., perfect absorption, phase modulation, etc.)…”

Metasurfaces-Based Absorption and Reflection Control: Perfect Absorbers and Reflectors

“…In [82] Li et al suggested a simple design to realise a broadband perfect reflector across the visible frequency range by using a single gradient meta-atom design. Rather than several discrete sized elements in the unit cell [83,84], this design is basically an array of trapezoid shaped Ag nanorods coupled with an optically thick Ag film separated by a SiO 2 spacer [5,85].…”

“…Because the thickness of the antenna is of deep-subwavelength scale the light propagation path can be neglected and the phase discontinuity is only induced by the ultrathin metasurface. This produces a reflection of the beam to an arbitrary direction determined by the gradient of the phase change, and since in the configuration described in [82] there is no spatial-variant phase shift in the -direction, generalised Snell's law for reflection [83] is reduced to…”