Gradient-Based Optimization of Core-Shell Particles with Discrete Materials for Directional Scattering

TL;DR

This paper presents a gradient-based optimization framework using deep learning to design core-shell nanoparticles with desired directional scattering, overcoming discrete parameter challenges efficiently.

Contribution

It introduces a novel method combining generative deep learning and surrogate modeling for efficient inverse design of nanophotonic structures with discrete materials.

Findings

Achieved improved directional scattering in core-shell nanoparticles.

Demonstrated increased computational efficiency over global optimization.

Provided design guidelines for strong forward and minimized backscattering.

Abstract

Designing nanophotonic structures traditionally grapples with the complexities of discrete parameters, such as real materials, often resorting to costly global optimization methods. This paper introduces an approach that leverages generative deep learning to map discrete parameter sets into a continuous latent space, enabling direct gradient-based optimization. For scenarios with non-differentiable physics evaluation functions, a neural network is employed as a differentiable surrogate model. The efficacy of this methodology is demonstrated by optimizing the directional scattering properties of core-shell nanoparticles composed of a selection of realistic materials. We derive suggestions for core-shell geometries with strong forward scattering and minimized backscattering. Our findings reveal significant improvements in computational efficiency and performance when compared to global optimization techniques. Beyond nanophotonics design problems, this framework holds promise for broad applications across all types of inverse problems constrained by discrete variables.