Fig. 1. Schematic of the fabrication process for the proposed (a) perovskite-based and (b) graphene-based controllable THz asymmetric metasurface devices. (c) Unit cell of the bare metasurface with parameters: py=190  μm, px=255  μm, L=75  μm, x=110  μm, y=20  μm, w=15  μm, and t=3  μm.

Fig. 2. Optical microscopy images of the bare metasurface device (a) without MAPbI3 film and (b) with spin-coated MAPbI3. (c) Representative top-view scanning electron microscopy image of the proposed perovskite-based device. (d) Graphical representation of the proposed perovskite-based THz metasurface device illuminated by a THz beam and optical pump.

Fig. 3. Simulated transmission spectra, surface currents, and electric field distributions for (a) the perovskite-based symmetrical metasurface when the perovskite photoconductivity value is 0 S/m and the proposed perovskite-based THz metasurface device when the perovskite photoconductivity values are (b) 0 S/m and (c) 6000 S/m.

Fig. 4. (a) Simulated and (b) measured transmission spectra of the perovskite-based THz metasurface device for different values of perovskite photoconductivity and photoexcitation intensity. (c), (d) Variation of the transmission amplitude at frequencies f1, f2, and f3 when changing (c) the photoconductivity of perovskite in the simulation or (d) the laser power density in the experiment. (e), (f) Amplitude modulation depth calculated by (e) simulation and (f) experiment.

Fig. 5. (a) Transmission spectra of analytical fits to the two-oscillator model for different values of perovskite photoconductivity. (b) Fit parameters as a function of the perovskite photoconductivity (κ is in units of THz2).

Fig. 6. (a) Three-dimensional schematic of the controllable graphene-based THz metasurface device with optical microscopy images on both sides. (b), (c) Diagrams of (b) optical modulation via pumping and (c) electrical modulation using bias voltages. (d) Raman spectra of the graphene layer measured with a 514 nm excitation laser at three different locations.

Fig. 7. (a) Transmission spectra of the simulation result from electromagnetic simulation software and analytical fitting curves by two-oscillator model under the different Fermi energy of graphene. Simulated electric field distributions at the frequency of transmission peak f2 when the Fermi energy are (b) 0.01 eV and (c) 0.1 eV. (d) Variation of the transmission amplitude at the frequency of f1, f2, and f3 when changing the Fermi energy of graphene in simulation. (e) Fitting parameters as a function of the Fermi energy (the unit of κ is THz2).

Fig. 8. (a) Measured transmission spectra of the graphene-based THz metasurface device with varying Fermi energy and photoexcitation intensity. (b) Variation of the transmission amplitude at frequencies f1, f2, and f3 when changing the laser power density in the experiment. (c) Modulation depth amplitude calculated by experiment.

Fig. 9. (a) Measured transmission spectra of the graphene-based THz metasurface device with and without ion-gel film coating. (b) Electrical modulation of the transmission spectrum at different gate voltages.