Herein, a new approach for using the intelligence aspects of artificial intelligence for knowledge discovery rather than device optimization in electromagnetic (EM) nanostructures is presented. This approach uses training data obtained through full‐wave EM simulations of a series of nanostructures to train geometric deep learning algorithms to assess the range of feasible responses as well as the feasibility of a desired response from a class of EM nanostructures. To facilitate the knowledge discovery, this approach combines the dimensionality reduction technique with convex‐hull and one‐class support‐vector‐machine (SVM) algorithms to find the range of the feasible responses in the latent response space of the EM nanostructure. More importantly, the one‐class SVM algorithm can be trained to provide the degree of feasibility of a response from a given nanostructure. This important information can be used to modify the initial structure to an alternative one that can enable an initially unfeasible response. To show the applicability of this approach, it is applied to two important classes of binary metasurfaces (MSs), formed by an array of plasmonic nanostructures, and periodic MSs formed by an array of dielectric nanopillars. These theoretical and experimental results confirm the unique features of this approach for knowledge discovery in EM nanostructures.

In this paper, the interplay of Bragg scattering and local resonance is theoretically studied in a phononic crystal (PnC) structure composed of a silicon membrane with periodic tungsten pillars. The comparison of phononic band gaps (PnBGs) in three different lattice types (i.e., square, triangular, and honeycomb) with different pillar geometries shows that different PnBGs have varying degrees of dependency on the lattice symmetry based on the interplay of the local resonances and the Bragg effect. The details of this interplay is discussed. The significance of locally resonating pillars, specially in the case of tall pillars, on PnBGs is discussed and verified by examining the PnBG position and width in perturbed lattices via Monte Carlo simulations. It is shown that the PnBGs caused by the local resonance of the pillars are more resilient to the lattice perturbations than those caused by Bragg scattering.

We have designed interlayer grating couplers with single/double metallic reflectors for Si/SiO(2)/SiN multilayer material platform. Out-of-plane diffractive grating couplers separated by 1.6 μm thick buffer SiO(2) layer are vertically stacked against each other in Si and SiN layers. Geometrical optimization using genetic algorithm coupled with electromagnetic simulations using two-dimensional (2D) finite element method (FEM) results in coupler designs with high peak coupling efficiency of up to 89% for double- mirror and 64% for single-mirror structures at telecom wavelength. Also, 3-dB bandwidths of 40 nm and 50 nm are theoretically predicted for the two designs, respectively. We have fabricated the grating coupler structure with single mirror. Measured values for insertion loss and 3-dB bandwidth in the fabricated single-mirror coupler confirms the theoretical results. This opens up the possibility of low-loss 3D dense integration of optical functionalities in hybrid material platforms.

We present strong experimental evidence for the existence of a complete phononic bandgap, for Lamb waves, in the high frequency regime (i.e., 800 MHz) for a pillar-based phononic crystal (PnC) membrane with a triangular lattice of gold pillars on top. The membrane is composed of an aluminum nitride film stacked on thin molybdenum and silicon layers. Experimental characterization shows a large attenuation of at least 20 dB in the three major crystallographic directions of the PnC lattice in the frequency range of 760 MHz–820 MHz, which is in agreement with our finite element simulations of the PnC bandgap. The results of experiments are analyzed and the physics behind the attenuation in different spectral windows is explained methodically by assessing the type of Bloch modes and the in-plane symmetry of the displacement profile.

No abstract