Enhanced light–matter interactions in dielectric nanostructures via machine-learning approach

Abstract

A key concept underlying the specific functionalities of metasurfaces, i.e. arrays of subwavelength nanoparticles, is the use of constituent components to shape the wavefront of the light, on-demand. Metasurfaces are versatile and novel platforms to manipulate the scattering, colour, phase or the intensity of the light. Currently, one of the typical approaches for designing a metasurface is to optimize one or two variables, among a vast number of fixed parameters, such as various materials' properties and coupling effects, as well as the geometrical parameters. Ideally, it would require a multi-dimensional space optimization through direct numerical simulations. Recently, an alternative approach became quite popular allowing to reduce the computational cost significantly based on a deeplearning-assisted method. In this paper, we utilize a deep-learning approach for obtaining high-quality factor (high-Q) resonances with desired characteristics, such as linewidth, amplitude and spectral position. We exploit such high-Q resonances for the enhanced light-matter interaction in nonlinear optical metasurfaces and optomechanical vibrations, simultaneously. We demonstrate that optimized metasurfaces lead up to 400+ folds enhancement of the third harmonic generation (THG); at the same time, they also contribute to 100+ folds enhancement in optomechanical vibrations. This approach can be further used to realize structures with unconventional scattering responses.

Infrared upconversion imaging in nonlinear metasurfaces

“…54 Further optimization could be achieved by employing machine learning approaches to simultaneously enhance light-matter interactions. 55 We believe that by enhancing the SFG conversion efficiency, continuous-wave nonlinear upconversion is achievable. Furthermore, the employment of sensitive CMOS cameras and additional devices such as intensifiers can ease the conversion efficiency requirements.…”

Determination of steep sidewall angle using polarization-sensitive asymmetric scattering

“…There should be also special attention to the detector, as this should have low noise amplifiers and probably with heterodyne techniques given that the expected voltage differences are small. We envision in the future using machine learning [33] to solve the inverse problem in scatterometry by building a comprehensive database based on calculations as presented in this paper to analyze the experimental data.…”

Plasmonic tweezers: for nanoscale optical trapping and beyond

“…The machine-learning approach is a thriving frontier field that provides an approach to improve the efficiency of research. It has been demonstrated that machine learning could enhance light–matter interactions 343 , providing good insight into plasmonic tweezers design and application research 316 . In addition, nanostructures composed of dielectric resonators can exhibit many of the same features as plasmonics, such as high surface-enhanced spectroscopies with low heat conversion 344 and nonlinear optical response processes 345 , giving rise to superior performance in comparison to their lossy plasmonic counterparts.…”