Accelerating the discovery of multilayer nanostructures with analytic differentiation of the transfer matrix equations

Abstract

Multilayer nanostructures represent an important class of materials with tunable optical and thermal radiative properties that can be leveraged for a wide range of energy applications. We present a theoretical framework for optimizing the geometry of such structures that utilizes gradients of various objective functions that are enabled through analytic differentiation of the transfer-matrix equations. We demonstrate the usefulness of this method by applying it to the local optimization of many-degree-of-freedom structures for incandescent light sources, and the global optimization of few-degree-of-freedom structures that serve as solar cell coatings and optical cavities for enhancing the absorption of organic chromophores embedded in thin films.