Microcavities optimization under uncertainty by evolutionary algorithms

In this paper we present the first quantitative study of parameters optimization for semiconductor microcavities synthesis under uncertainty using genetic algorithm. A microcavity is a system in which a gain medium can interact with a single cavity mode. These structures have been used in important studies of several areas for technological or purely scientific purposes. However, the definition of the optimal parameters for the fabrication of microcavities is a difficult task. Moreover, the difficulty further increases during the growth process due to experimental uncertainties that may occur, mainly related to the thickness of the layers. Thus, the device can present different properties from those desired. Based on the reflectance spectra of a Al_xGa_1-xAs semiconductor microcavity, our goal is to find the optimal parameter set (aluminum concentrations x, thickness of layers and the number of layers). This set of parameters may offer increased robustness in the growth process, while providing a considerable Quality Factor and the desired position of the cavity resonance. The results indicate that the proposed algorithm is able to find satisfactory solutions, minimizing the problems caused by inaccuracy in the growth of these devices.