Microlens arrays placed not far from the image plane of infrared focal-plane arrays (IR FPAs) can, by focusing light onto each pixel, reduce the size of the active area of each pixel, which can increase the sensitivity of the FPA and allow it to run at a higher temperature. Now, researchers at Harvard University (Cambridge, MA) and the Jet Propulsion Laboratory (JPL) at the California Institute of Technology (Pasadena, CA) are designing solid-immersion metasurface lenses to accomplish the same purpose. In these solid-immersion metalenses, the transmissive metasurface is atop a solid substrate that totally fills the gap between the metasurface lens and the FPA itself. As a result, the assembly is straightforward to put together and forms a rugged unit. The configuration is compatible with the architecture of back-illuminated FPAs.
The experimental 10 × 10 metalens array focuses light at a 3–5 µm wavelength through a gallium antimonide (GaSb) substrate. However, the technique is applicable to other mid-IR optical materials as well. Each metasurface-lens array element is 30 µm in diameter and is composed of many round posts created by etching the GaSb, with the diameter of the posts decreasing as a function of lens radius. Two types of metalenses were created: amonochromatic and a broadband version. For testing, a microscope situated to focus on the other side of the substrate (where the FPA would be placed) imaged the focused spots and compared them to unfocused light, showing that the monochromatic metalenses produced foci with a beam waist of 13.9 µm at 1/e2 of the peak intensity at a 3.9 µm wavelength, while the broadband version produced foci with a 16.4 µm size over a 3–5 µm wavelength. Strehl ratios for the monochromatic and broadband lenses were 52 and 0.38, respectively. (The researchers note that for this type of use, diffraction-limited performance is not required.) Both types of metalens showed an intensity enhancement of about 3X at the focus. Reference: S. Zhang et al., Appl. Phys. Lett. (2018); https://doi.org/10.1063/1.5040395.
