Deep Learning-Assisted Enhanced Fano Resonances in Symmetry-Breaking SOI Metasurface

Zan Hui Chen,Weicheng Chen,Jiaqi Wang,Guo-Wei Lu,Zhenzhou Cheng

doi:10.1109/jphot.2021.3127220

Abstract

Metasurfaces analogues of Fano resonances provide a powerful platform for high sensitivity sensing, nonlinear optics, and light manipulation. However, previous Fano-resonant metasurfaces usually are not compatible with silicon complementary metal-oxide semiconductor circuits due to their hybrid material structures and large non-radiative loss. Herein, we theoretically demonstrate a silicon-on-insulator metasurface (SOIM) enhancing Fano resonances by using a tandem neural network design. Multiple Fano resonances with high Q-factor have been observed in the symmetry-breaking SOIM. The Fano-resonant mechanism of the SOIM is analyzed. Additionally, the spectral features of the Fano-resonant SOIM as a function of the symmetry tuning factor of the double silicon nanobars and the environment refractive index are also investigated. The result shows that the Fano-resonant SOIM as a methanol sensor with a sensitivity of 310 nm/RIU can achieve an overall figure of merit of 195 in the near-infrared spectral regime. The designed Fano-resonant SOIM shows enormous potential applications in highly sensitive sensors and light-matter interaction enhancement.

Highlights

FANO resonance was firstly proposed by Ugo Fano to explain the phenomena of asymmetric line shape of the absorption spectra in the experiment of noble gasses [1]
While our design is based on the SOI platform, which has the advantage of the scalability and standardization, for silicon complementary metal-oxide semiconductor (CMOS) (Si-COMS) circuits [18]
The predicted geometry parameters (Px, Py, W, L, GT, and Gm, as shown in Fig. 4(b)) of silicon-on-insulator metasurface (SOIM) were obtained via deep neural networks (DNNs), which correspond to our input target spectra with the high Q-factor of 1000

Summary

INTRODUCTION

FANO resonance was firstly proposed by Ugo Fano to explain the phenomena of asymmetric line shape of the absorption spectra in the experiment of noble gasses [1]. The inverse design based on neural networks shows enormous potential applications in the design and engineering of materials [22], [23], structures [24], [25], and devices [26], [27]. Deep neural networks (DNNs) can save the time and cost of the device design by accelerating the optimization process. We theoretically propose and demonstrate a silicon-on-insulator metasurface (SOIM) by using a deep neural. Based on the deep learning-assisted optimization, multiple Fano resonances with high Q-factor have been observed in symmetry-breaking SOIM. The results suggest that the designed SOIM shows enormous potential applications in developing highly sensitive sensors and nonlinear optics devices

OPTIMIZATION PROCESS OF SOIM BY USING DNNS

DESIGN OF FANO-RESONANT SOIM

FANO RESONANCE ENHANCEMENT IN SOIM

METHANOL SENSING APPLICATION OF THE FANO-RESONANT SOIM

CONCLUSION