Plasmonic metasurfaces have recently attracted much attention because of their novel characteristics with respect to light polarization and wave front control on deep-subwavelength scales. The development of metasurfaces with reconfigurable optical responses is opening new opportunities in high-capacity communications, real-time holograms and adaptive optics. Such tunable devices have been developed in the mid-infrared spectral range and operated in light intensity modulation schemes. Here we present a novel optically reconfigurable hybrid metasurface that enables polarization tuning at optical frequencies. The functionality of tuning is realized by switching the coupling conditions between the plasmonic modes and the binary isomeric states of an ethyl red switching layer upon light stimulation. We achieved more than 20° nonlinear changes in the transmitted polarization azimuth using just 4 mW of switching light power. Such design schemes and principles could be easily applied to dynamically adjust the functionalities of other metasurfaces.

Ellipsometry is a powerful method for determining both the optical constants and thickness of thin films. For decades, solutions to ill-posed inverse ellipsometric problems require substantial human–expert intervention and have become essentially human-in-the-loop trial-and-error processes that are not only tedious and time-consuming but also limit the applicability of ellipsometry. Here, we demonstrate a machine learning based approach for solving ellipsometric problems in an unambiguous and fully automatic manner while showing superior performance. The proposed approach is experimentally validated by using a broad range of films covering categories of metals, semiconductors, and dielectrics. This method is compatible with existing ellipsometers and paves the way for realizing the automatic, rapid, high-throughput optical characterization of films.