Advanced multi-objective and surrogate-assisted optimization of topologically diverse metasurface architectures

Abstract

Advancements in micro-and nano-fabrication techniques are enabling the realization of high-performance metasurfaces which exploit the generalized Snell’s law of refraction to achieve disruptive optical functionalities. Moreover, metasurfaces can be used in conjunction with conventional optical elements to achieve massive size, weight, and power (SWaP) reduction. However, no commercial tools exist which can efficiently model optical systems whose geometrical features span many orders-of-magnitude in spatial scale. Therefore, new forward solvers must be developed in order to make such multiscale problems tractable. Furthermore, optimization of multiscale optical systems is crucially-important in order to maximize system performance and minimize SWaP. While how one achieves a specific set of desired performances with conventional optical elements is generally well understood and thoroughly presented in design textbooks, it is not always clear how to design a nanoscale optical device to best achieve a desired set of performances. Therefore, a small subset of well-understood and/or canonical structures such as split-ring resonators are typically employed to achieve the targeted functionality. However, relaxing the device’s topological constraints may lead to improved performance albeit at the expense of a larger solution space to explore. To mitigate this issue, we employ custom state-of-the art multi-objective and surrogate-assisted optimization algorithms to explore the solution space afforded by emerging manufacturing techniques in order to design metasurface topologies that achieve an arbitrary number of user-specified performance criteria.