The design and creation of two-dimensional nanostructured materials have created a new paradigm in material science: so-called metasurfaces have emerged as a new class of integrated photonic elements featuring only a single or few surface layers composed of subwavelength plasmonic nanostructures. Recently, significant breakthroughs in the wavefront manipulation, such as abnormal and out-of-plane refraction and reflection, have been achieved by laterally inhomogeneous metasurfaces. Another example is the spectrally selective image formation demonstrated by computationally encoded metasurfaces using the principles of digital holography. In the view of the growing structural complexity of contemporary metasurfaces, the lack of comprehensive experimental methods to assess and characterize their building blocks performance becomes a critical issue, hampering development of this field towards real-world applications. Due to restrictions of the state of the art experimental techniques for the phase measurements on metasurfaces, the preference is commonly given to indirect characterization methods relying heavily on rigorous numerical simulations.
In this talk I will demonstrate, based on the Kramers-Kronig transformation analysis, that the use of indirect methods is inadequate for the accurate characterization of complex metasurfaces. In order to provide experimental access to the complex transmission and reflection coefficients of optical metasurfaces and as a prerequisite to assess their broadband performance, we developed an original experimental technique, which I will present in the talk. Additionally, I will demonstrate an experimentally realized compute generated hologram based on metasurfaces, which generates different holographic images at two distinct wavelengths.