The advent of metasurfaces in recent years has ushered in a revolutionary means to manipulate the behavior of light on the nanoscale. The design of such structures, to date, has relied on the expertise of an optical scientist to guide a progression of electromagnetic simulations that iteratively solve Maxwell's equations until a locally optimized solution can be attained. In this work, we identify a solution to circumvent this conventional design procedure by means of a deep learning architecture. When fed an input set of customer-defined optical spectra, the constructed generative network generates candidate patterns that match the on-demand spectra with high fidelity. This approach reveals an opportunity to expedite the discovery and design of metasurfaces for tailored optical responses in a systematic, inverse-design manner.

No abstract

Recent progress in metamaterial research has successfully exceeded the limitations imposed by conventional materials and optical devices, enabling the manipulation of electromagnetic waves as desired. The distinct characteristics and controlling abilities of metamaterials make them ideal candidates for novel photonics devices not only in traditional optics but also for biological detection, medical science, and metrology. However, the controllability and functionality of both single-layer metasurfaces and bulk metamaterials are not sufficient to meet the requirements of emerging technologies; hence, new solutions must be found. As such technologies advance, new functionalities will emerge as different or identical single-layer metasurfaces are combined. Thus, innovation in few-layer metasurfaces will become an increasingly important line of research. Here, these metasurfaces are classified according to their functionalities and the few-layer metasurfaces that have been proposed up to now are presented in a clear sequence. It is expected that, with further development in this area, few-layer metasurfaces will play an important role in the family of optical materials.

controllability of light due to the limited flexibility rendered in periodic metastructures of simple unit cells. To overcome these deficiencies, metasurfaces comprised of multiple meta-atoms, such as gradient and multilayered metasurfaces, have been proposed and developed. [7][8][9] Relying on the collective effects of multiple meta-atoms, these metasurfaces present intriguing properties such as anomalous deflection, [7,10] arbitrary phase control, asymmetric polarization conversion, [8,11] wave-front shaping, [12][13][14] etc., which brings about extensive applications for imaging, optical signal processing, emission control, and much more. Here in our following discussion, we refer to unit cells composed of various meta-atoms as metamolecules, analogous to the hierarchical relationship between atoms and molecules in nature. In our definition of a metamolecule, we assume every two adjacent meta-atoms are not strongly coupled, in which case the overall properties of the metamolecule can be analytically predicted by the properties of its constituent meta-atoms. Such an assumption is valid in most metasurfaces that consist discrete, spatially variant building blocks.Despite the extraordinary properties of metasurfaces made up of metamolecules, designing multiple meta-atoms that collectively function as a device is a time-consuming task that requires labor-intensive trial-and-error simulations. The difficulty of the inverse design of such metamolecules arises from the intricate mechanisms of multistructured systems, the vast number of possible combinations of distinct meta-atoms, as well as the expensive 3D full wave simulations required. Traditionally, a practical solution to such a design follows three steps: 1) specifying a class of geometry with a few parameters as candidate meta-atoms, 2) carrying out parametric sweeps on these parameters, and 3) enumerating possible combinations of meta-atoms to meet the design objective. However, the limitation of the geometry in the strategy largely restricts the variety of the shapes of meta-atoms, which usually does not lead to an optimal solution, even after extensive and expensive simulations.Alongside the evolution of nanophotonics, various methods for expediting the design of photonic structures have been developed. Gradient-based adjoint methods, such as topology optimization, are a class of widely applied approaches for Molecules composed of atoms exhibit properties not inherent to their constituent atoms. Similarly, metamolecules consisting of multiple meta-atoms possess emerging features that the meta-atoms themselves do not possess. Metasurfaces composed of metamolecules with spatially variant building blocks, such as gradient metasurfaces, are drawing substantial attention due to their unconventional controllability of the amplitude, phase, and frequency of light. However, the intricate mechanisms and the large degrees of freedom of the multielement systems impede an effective strategy for the design and optimization of metamolecules. Here, a hybrid artificial-i...

Machine learning, as a study of algorithms that automate prediction and decision‐making based on complex data, has become one of the most effective tools in the study of artificial intelligence. In recent years, scientific communities have been gradually merging data‐driven approaches with research, enabling dramatic progress in revealing underlying mechanisms, predicting essential properties, and discovering unconventional phenomena. It is becoming an indispensable tool in the fields of, for instance, quantum physics, organic chemistry, and medical imaging. Very recently, machine learning has been adopted in the research of photonics and optics as an alternative approach to address the inverse design problem. In this report, the fast advances of machine‐learning‐enabled photonic design strategies in the past few years are summarized. In particular, deep learning methods, a subset of machine learning algorithms, dealing with intractable high degrees‐of‐freedom structure design are focused upon.

Designing complex physical systems, including photonic structures, is typically a tedious trial-and-error process that requires extensive simulations with iterative sweeps in multidimensional parameter space. To circumvent this conventional approach and substantially expedite the discovery and development of photonic nanostructures, here we develop a framework leveraging both a deep generative model and a modified evolution strategy to automate the inverse design of engineered nanophotonic materials. The capacity of the proposed methodology is tested through the application to a case study, where metasurfaces in either continuous or discrete topologies are generated in response to customer-defined spectra at the input. Through a variational autoencoder, all potential patterns of unit nanostructures are encoded into a continuous latent space. An evolution strategy is applied to vectors in the latent space to identify an optimized vector whose nanostructure pattern fulfills the design objective. The evaluation shows that over 95% accuracy can be achieved for all the unit patterns of the nanostructure tested. Our scheme requires no prior knowledge of the geometry of the nanostructure, and, in principle, allows joint optimization of the dimensional parameters. As such, our work represents an efficient, on-demand, and automated approach for the inverse design of photonic structures with subwavelength features.

wileyonlinelibrary.comantennas and metallic patches were also proved to be able to manipulate wavefront of scattered waves for refracted or refl ected light. [4][5][6][7][8][9][10][11][12] Among these new-type anomalous-scattering devices, the control of circularly polarized (CP) light is testifi ed feasible in both theory and experiment-with spatially various antennas arranged properly. The phase of CP light is modulated without paying attention on scattered intensity generated by each antenna. [13][14][15][16][17] Harnessing light for modern photonic applications often involves the control and manipulation of light intensity or effi ciency. The most important issue for generating anomalous scattered lights is also the effi ciency by reasonable arrangement of the various shaped or layered plasmonic metasurfaces. Compared with normal light conforming to ordinary Snell's law, the effi ciency of the anomalous scattered lights is far too low in the previous works. Some methods have been proposed to improve the intensity of abnormal light but at the same time suppress that of normal one. For example, plasmonic metasurface can be sandwiched by gratings to increase the effi ciency of cross-polarized light with the help of Fabry-Pérot effect; [ 18,19 ] or few-layer composite metascreen made of dielectric and metal is used to enhance the performance of control. [ 20,21 ] These approaches dramatically enhance the effi ciency of anomalous light, but at the same time increase the processes and diffi culty in fabrication because of their extra structures (gratings or board) and precise distance between antennas and gratings. However, so far it is still challenging to further improve the effi ciency of anomalous light with circular polarization by a single-layer plasmonic metasurface.Here, we introduce a new approach to describe the function of a polarizer and show an intuitive method to describe the extra phase of CP light due to its orientation. We present the design, fabrication, and characteristics of the multinanorod plasmonic metasurfaces, in which each unit can be regarded as an imperfect polarizer with extraordinary broadband. The high effi ciency and broadband effects are theoretically and experimentally demonstrated. The correctness of GSL is also demonstrated by experimental results. At last, arbitrary polarized incidence is considered, and an easy way to determine the polarization degree of the incident light based on Poincaré sphere is feasibly proved by the proposed plamonic metasurfaces. High-Performance Broadband Circularly Polarized Beam Defl ector by Mirror Effect of Multinanorod MetasurfacesZhaocheng Liu , Zhancheng Li , Zhe Liu , Jianxiong Li , Hua Cheng , Ping Yu , Wenwei Liu , Chengchun Tang , Changzhi Gu , Junjie Li , * Shuqi Chen , * and Jianguo Tian * Manipulation of light phase and amplitude by plasmonic metasurfaces has immensely promising applications in optical imaging, information processing, communications, and quantum optics. However, the controllability of effi ciency and bandwidth is relati...

Optical activity is the rotation of the plane of linearly polarized light along the propagation direction as the light travels through optically active materials. In existing methods, the strength of the optical activity is determined by the chirality of the materials, which is difficult to control quantitatively. Here we numerically and experimentally investigated an alternative approach to realize and control the optical activity with non-chiral plasmonic metasurfaces. Through judicious design of the structural units of the metasurfaces, the right and left circular polarization components of the linearly polarized light have different phase retardations after transmitting through the metasurfaces, leading to large optical activity. Moreover, the strength of the optical activity can be easily and accurately tuned by directly adjusting the phase difference. The proposed approach based on non-chiral plasmonic metasurfaces exhibits large optical activity with a high controllable degree of freedom, which may provide more possibilities for applications in photonics.