Synthesization of multiple functionalities over a flat metasurface platform offers a promising approach to achieving integrated photonic devices with minimized footprint. Metasurfaces capable of diverse wavefront shaping according to wavelengths and polarizations have been demonstrated. Here we propose a class of angle-selective metasurfaces, over which beams are reflected following different and independent phase gradients in the light of the beam direction. Such powerful feature is achieved by leveraging the local phase modulation and the non-local lattice diffraction via inverse scattered field and geometry optimization in a monolayer dielectric grating, whereas most of the previous designs utilize the local phase modulation only and operate optimally for a specific angle. Beam combiner/splitter and independent multibeam deflections with up to 4 incident angles are numerically demonstrated respectively at the wavelength of 700 nm. The deflection efficiency is around 45% due to the material loss and the compromise of multi-angle responses. Flexibility of the approach is further validated by additional designs of angle-switchable metagratings as splitter/reflector and transparent/opaque mirror. The proposed designs hold great potential for increasing information density of compact optical components from the degree of freedom of angle.

The combination of graphene and a metasurface holds great promise for dynamic manipulation of the electromagnetic wave from low terahertz to mid-infrared. The optical response of graphene is significantly enhanced by the highly-localized fields in the meta-atoms, and the characteristics of meta-atoms can in turn be modulated in a large dynamic range through electrical doping of graphene. Graphene metasurfaces are initially focused on intensity modulation as modulators and tunable absorbers. In this paper, we review the recent progress of graphene metasurfaces for active control of the phase and the polarization. The related applications involve, but are not limited to lenses with tunable intensity or focal length, dynamic beam scanning, wave plates with tunable frequency, switchable polarizers, and real-time generation of an arbitrary polarization state, all by tuning the gate voltage of graphene. The review is concluded with a discussion of the existing challenges and the personal perspective of future directions.

Mid-infrared perfect absorbers (PAs) based on metamaterials have many applications in material analysis and spectral detection thanks to the associated strong light–matter interaction. Most of the PAs are built as ‘metal nanostructure’-insulator-metals (MIM). In this paper, we propose an ultra-narrow band absorber based on dielectric metasurface with a metal film substrate. The absorptance comes from the plasmonic absorption in the metal film, where the absorption is enhanced (while the band of that is compressed) by the super cavity effect of the dielectric metasurface. Based on our numerical calculation, the full-width at half-maximum (FWHM) can reach 67 nm at 8 μm (8‰), which is more than two orders of magnitude smaller than the resonance wavelength and much narrower than the theoretical FWHMs of MIM absorbers. Moreover, we studied their application in infrared thermal imaging, which also has more benefits than MIM absorbers. This kind of hybrid dielectric metasurface provides a new route to achieve ultra-narrow band perfect absorbers in the mid-infrared regime and can be broadly applied in detectors, thermal emitters and bio-spectroscopy.

In the remote sensing field, synthetic aperture radar (SAR) is a type of active microwave imaging sensor working in all-weather and all-day conditions, providing high-resolution SAR images of objects such as marine ships. Detection and instance segmentation of marine ships in SAR images has become an important question in remote sensing, but current deep learning models cannot accurately quantify marine ships because of the multi-scale property of marine ships in SAR images. In this paper, we propose a multi-scale feature pyramid network (MS-FPN) to achieve the simultaneous detection and instance segmentation of marine ships in SAR images. The proposed MS-FPN model uses a pyramid structure, and it is mainly composed of two proposed modules, namely the atrous convolutional pyramid (ACP) module and the multi-scale attention mechanism (MSAM) module. The ACP module is designed to extract both the shallow and deep feature maps, and these multi-scale feature maps are crucial for the description of multi-scale marine ships, especially the small ones. The MSAM module is designed to adaptively learn and select important feature maps obtained from different scales, leading to improved detection and segmentation accuracy. Quantitative comparison of the proposed MS-FPN model with several classical and recently developed deep learning models, using the high-resolution SAR images dataset (HRSID) that contains multi-scale marine ship SAR images, demonstrated the superior performance of MS-FPN over other models.

In this Letter, we propose a hybrid graphene-dielectric metasurface as a bi-functional polarization converter. It can switch between a reflective half-wave plate and a quarter-wave plate around 1 THz by merely applying external biasing voltage, without reoptimizing the dielectric structure. Switching of the two wave plates originates from distinct dispersion of the orthogonal eigenmodes with the chemical potential, which is further explained by the overlapping of graphene and the dielectric resonance modes. Compared with graphene-metallic metasurfaces, a combination of graphene with dielectric microstructures offers an alternative solution for active terahertz devices with high efficiency and large flexibility.

Terahertz (THz) technology offers unparalleled opportunities in a wide variety of applications, ranging from imaging and spectroscopy to communications and quality control, where lack of efficient modulation devices poses a major bottleneck. Spatial modulation allows for dynamically encoding various spatial information into the THz wavefront by electrical or optical control. It plays a key role in single-pixel imaging, beam scanning and wavefront shaping. Although mature techniques from the microwave and optical band are not readily applicable when scaled to the THz band, the rise of metasurfaces and the advance of new materials do inspire new possibilities. In this review, we summarize the recent progress of THz spatial light modulators from the perspective of functional materials and analyze their modulation principles, specifications, applications and possible challenges. We envision new advances of this technique in the near future to promote THz applications in different fields.