Fig. 1. Schematics of multiplexing metadevices based on coherent wave interferences. Illuminated by a light beam with polarization continuously tuned characterized by two continuously varying circular-polarization expansion coefficients, wave reflected by the metadevice can exhibit different wave-fronts and polarization distributions dictated by two expansion coefficients.

Fig. 2. Schematic, phase diagrams and experimental characterizations of metaatoms. (a) Schematic of the designed metal-insulator-metal meta-atom with geometric parameters d_Str = 30 nm, d_Sub = 120 nm, d_Int = 120 nm, w = 80 nm, and P_x = P_y = 625 nm fixed. FDTD-simulated (b) phase difference ΔΦ and (c) resonant phase Φ_Res of metaatoms with different L_u and L_v at the wavelength of 1550 nm. Orange dashed lines denote the cases ΔΦ = 0.5π and − 1.5π and black dashed lines denote the cases ΔΦ = ±π. Orange and black stars represent metaatoms labelled as Nos. 1–4 (functioning as QWPs) and Nos. 5–12 (functioning as HWPs), correspondingly. Experimentally measured LPR spectra of metaatoms (d) No.1, (e) No.2, (f) No.3, (g) No.4 under the RCP incidence and metaatoms (h) No.5, (i) No.6, (j) No.7, (k) No.8, (l) No.9, (m) No.10, (n) No.11, (o) No.12 under the incidence of LP light polarized along the angle 135°. Dashed lines denote the positions of 1550 nm.

Fig. 3. Experimental characterizations on metadevice I. (a) Schematic illustration of working principle of metadevice I: The device can generate two spatially separating vortex beams carrying different topological charges with strengths tuned by varying the LCP/RCP components inside the incident beam. (b) SEM image of the fabricated sample and a zoom-in picture (inset). (c) Amplitudes of two beams (|A₊| (red) and |A₋| (blue)) versus Θ₀ obtained from theoretical calculations (line) and experimental measurements (stars); Five polarization states labelled by 1, 2, 3, 4 and 5 on the Poincare’s sphere, adopted as incident polarizations for further experimental characterizations. (d–h) Experimentally measured angular distributions of intensity of light reflected by our metasurface under illuminations of normally incident light with different wavelengths exhibiting polarizations corresponding to those 5 points on the Poincare’s sphere as shown in (c). (i–m) Measured interference patterns between a spherical wave and the waves reflected by the metasurface, as shined by normally incident light at 1550 nm with polarizations corresponding to those 5 points on the Poincare’ sphere as shown in (c). Only a single interference pattern is shown in (i) and (m) as only one reflected beam exists in these two cases.

Fig. 4. Experimental characterizations on metadevice II with incident polarization changing along path-I on the Poincare’s sphere. (a) SEM image of the fabricated sample and a zoom-in picture (inset). (b) Path-I on the Poincare’s sphere on which the incident polarization varies, with 1, 2, 3, 4 and 5 denoting five incident polarization states adopted for experimental characterizations. Theoretically calculated (line) and experimentally measured (stars) values of (c) L_z (d) φ₀ and (e) γ₀ versus the parameter δ, which is the relative angle between a linear polarizer and a QWP employed in experiments to generate the desired incident polarization state. Measured intensity patterns of light reflected by our metasurface, under the illuminations of light at 1550 nm with polarizations with number (f) 1, (g) 2, (h) 3, (i) 4, and (j) 5 on the Poincare’s sphere as shown in (b), with a rotatable polarizer placed in front of the detecting CCD tiled at the angles indicated by the white double-head arrows shown in right-up corners of each panels. Green symbols (arrows/circles/ellipse) illustrate the theoretically predicted polarization distributions in different cases. (k-o) Measured interference patterns between a spherical wave and the waves reflected by the metasurface, as shined by normally incident light at 1550 nm with polarizations corresponding to those 5 points on the Poincare’ sphere as shown in (b).

Fig. 5. Experimental characterizations on metadevice II with incident polarization changing along path-II on the Poincare’s sphere. (a) Path-II on the Poincare’s sphere on which the incident polarization varies, with 1, 2, 3, and 4 denoting four representative incident polarization states adopted for experimental characterizations. Theoretically calculated (line) and experimentally measured (stars) values of (b) L_z (c) φ₀ and (d) γ₀ versus the parameter Ψ₀, which is the azimuthal angle on the Poincare’s sphere. Measured intensity patterns of light reflected by our metasurface, under the illuminations of light at 1550 nm with polarizations denoted by number (e) 1, (f) 2, (g) 3 and (h) 4 on the Poincare’s sphere as shown in (a), with a rotatable polarizer placed in front of the detecting CCD tiled at the angles indicated by the white double-head arrows shown in right-up corners of each panels. Green symbols (arrows/circles/ellipse) illustrate the theoretically predicted polarization distributions in different cases. (i-l) Measured interference patterns between a spherical wave and the waves reflected by the metasurface, as shined by normally incident light at 1550 nm with polarizations corresponding to those four points on the Poincare’ sphere as shown in (a).