Optical Metasurfaces: Evolving from Passive to Adaptive

Abstract

of ultrathin, artificial surfaces, so-called metasurfaces, for manipulating the propagation of light at will. [1,2] Metasurfaces control the propagation of light by abruptly changing its phase, amplitude, polarization, or spectrum at a surface and within a thin (quasi-2D) layer using large arrays of optical scatterers with subwavelength separation. Ideally, each scatterer can be independently engineered in order to locally influence the wavelets of the impinging light, and therefore, arbitrarily reform the wavefront. This concept has been exploited to create a variety of flat optical elements such as beam deflectors, [3][4][5] lenses, [6][7][8] and holograms. [9,10] As compared to bulky refraction-based optical elements, these surfaces are of subwavelength or wavelength-scale thickness, which enables their integration into compact devices. A main feature of the majority of such flat optical elements is that they are passive, and once fabricated, their optical functionalities are invariable. A beam deflector of this kind will direct light of a certain wavelength and direction always into the same direction, or a lens will focus light always to the same focal point. A broad range of new applications could be enabled if it were possible to actively and reversibly change the functionality of metasurfaces after their fabrication. With such "adaptive" metasurfaces beam deflectors with adjustable angle of deflection, metalenses with variable focal length or dynamic holograms may be realized. Advances toward such active devices are highly relevant for applications such as light detection and ranging [11] (LIDAR) sensing for driverless cars, dynamic zoom lenses for miniaturized camera systems [12] (e.g., in smartphones), or holographic displays for augmented reality devices. [13] An adaptive metasurface allows for the dynamic adjustment of an optical functionality of the surface. As shown in Figure 1, for the specific case of a metalens, these surfaces can be categorized based on their degree of dynamic adaptivity. While a passive surface has a fixed functionality, a switchable metasurface allows for switching back and forth between two (or more) predefined states (e.g., switching between different focal lengths, deflection angles, or hologram images). A continuously tunable metasurface enables continuous adjustment of a property within a range (e.g., tunable focal length or deflection angle). Finally, a freely tunable metasurface allows for continuous, arbitrary adjustment of a property, for example, 3D adjustment of the focal point of the metalens shown in Figure 1. Adaptive functionalities can be added to metasurfaces with different methods, for example, by mechanically rearranging the scatterers on the surface with respect to each other, or modifying

All‐Dielectric Programmable Huygens' Metasurfaces

“…[ 1–8 ] While originally only enabling static functionality, [ 9 ] these concepts have recently been extended towards active optical devices by employing tunable metasurfaces for sensing and light focusing applications, where the optical response can be controlled using light inclination, [ 10–12 ] stretchable substrates, [ 13 ] electrostatic biasing of two‐dimensional (2D) materials such as graphene, [ 14,15 ] and phase change materials. [ 16–18 ]…”

Biomimetic ultra-broadband perfect absorbers optimised with reinforcement learning

“…Metamaterials that exhibit a wide range of functionalities have been designed over the past few decades [1][2][3][4][5][6]. Of these metamaterials, perfect absorbers offer numerous benefits over conventional absorbers such as increased effectiveness, miniaturisation and the ability to be designed for a specific narrow or broadband range [7][8][9][10].…”

Diffractive metalens: from fundamentals, practical applications to current trends

“…How to design and to optimize such metalens systems and how to postprocess the vast captured images remain questions. Recently, AI techniques have drawn much attention in both commercial and the scientific community [155,161,[212][213][214], and may enable the realization of super-resolution imaging, photonic quantum computing, and a plenty of intriguing branches. AI techniques even provide possibilities for loop-locked design from materials choosing, geometric design, array arrangement, to fabrication processes, which may give birth to future high-performance metalenses and corresponding systems.…”