Evolutionary Optimization of All‐Dielectric Magnetic Nanoantennas

Abstract

Magnetic light and matter interactions are generally too weak to be detected, studied, and applied technologically. However, if one can increase the magnetic power density of light by several orders of magnitude, the coupling between magnetic light and matter could become of the same order of magnitude as the coupling with its electric counterpart. For that purpose, photonic nanoantennas, in particular dielectric, are proposed to engineer strong local magnetic field and therefore increase the probability of magnetic interactions. Unfortunately, dielectric designs suffer from physical limitations that confine the magnetic hot spot in the core of the material itself, preventing experimental and technological implementations. Here, it is demonstrated that evolutionary algorithms can overcome such limitations by designing new dielectric photonic nanoantennas, able to increase and extract the optical magnetic field from high refractive index materials. It is also demonstrated that the magnetic power density in an evolutionary optimized dielectric nanostructure can be increased by a factor 5 compared to state‐of‐the‐art dielectric nanoantennas and that the fine details of the nanostructure are not critical in reaching these aforementioned features, as long as the general shape of the motif is maintained. This advocates for the feasibility of nanofabricating the optimized antennas experimentally and their subsequent application.

Numerical Optimization Methods for Metasurfaces

“…In Ref. [60], the authors identified the optimal configuration of a binary configuration, made of Si and air, which maximizes the magnetic field intensity. The geometry and its binary representation are shown in Figure 5a, in which each "1" represents Si rectangular meta-atoms and each "0" represents an air void.…”

“…So far, our discussion has focused on the optimization of only a single target in the objective function. A more challenging [60] Copyright 2019, Wiley-VCH. d-f): another example to illustrate the usefulness of utilizing evolutionary approach based technique in enhancing near field intensity.…”

“…(a–c) Reproduced with permission. [ 60 ] Copyright 2019, Wiley–VCH. d–f): another example to illustrate the usefulness of utilizing evolutionary approach based technique in enhancing near field intensity.…”

Diffractive metalens: from fundamentals, practical applications to current trends

“…However, with the striking increment of the multi-dimensional parameter spaces in the metasurfaces designs, the realization of some highperformance metasurfaces becomes time-consuming, such as the high-NA, high-efficiency metalens, and the ultra-wideband multifunctional meta-devices. Recently, the development of AI techniques has become a research hot spot, and enables novel microwave and photonic devices, which may relieve the researchers from tedious trials and errors-testing works [155][156][157][158][159][160][161][162][163]. For example, a topology-optimized method was applied to optimize the phase design of the metalens [164].…”

Inverse design of higher-order photonic topological insulators

“…Topology optimization employs a robust numerical algorithm to seek the layout of materials within the design domain, so as to minimize or maximize the defined objective function [36]. Topology optimization has been proven to be an efficient and effective tool for the design of PCs [37][38][39][40][41][42][43][44][45][46]. Here, we will extend topology optimization to design the unit cell of nontrivial and trivial PCs for SPTIs.…”