Generalized Brewster-Kerker effect in dielectric metasurfaces

TL;DR

This paper predicts and demonstrates a generalized Brewster effect in dielectric metasurfaces, enabling polarization control at any angle, wavelength, or polarization, with potential for advanced photonic applications.

Contribution

It introduces a theoretical prediction and experimental demonstration of a generalized Brewster effect in dielectric metasurfaces, expanding polarization control beyond traditional limits.

Findings

Theoretical prediction of a generalized Brewster effect for any angle, wavelength, and polarization.

Experimental demonstration using silicon nanodisks at visible wavelengths.

Suppressed scattering due to interference of electric and magnetic dipoles in nanoparticles.

Abstract

Polarization is one of the key properties defining the state of light. It was discovered in the early 19th century by Brewster, among others, while studying light reflected from materials at different angles. These studies led to the first polarizers, based on Brewster's effect. One of the most active trends in photonics now is devoted to the study of miniaturized, sub-wavelength devices exhibiting similar, or even improved, functionalities compared to those achieved with bulk optical elements. In the present work, it is theoretically predicted that a properly designed all-dielectric metasurface exhibits a generalized Brewster effect potentially for any angle, wavelength and polarization of choice. The effect is experimentally demonstrated for an array of silicon nanodisks at visible wavelengths. The underlying physics of this effect can be understood in terms of the suppressed scattering at certain angles that results from the interference between the electric and magnetic dipole resonances excited in the nanoparticles, predicted by Kerker in early 80s. This reveals deep connection of Kerker's and Brewster's legacies and opens doors for Brewster phenomenon to new applications in photonics, which are not bonded to a specific polarization or angle of incidence.