Full-Color Plasmonic Metasurface Holograms
Supporting Information
Full-Color Plasmonic Metasurface Holograms Weiwei Wan, Jie Gao* and Xiaodong Yang* Department of Mechanical and Aerospace Engineering, Missouri University of Science and Technology, Rolla, MO 65409, USA *E-mail: [email protected], [email protected]
1
Figure S1. Control the refraction angles by constant gradient phase distribution. Assuming that two light beams with a distance ∆x normally incident from the bottom medium, there is a constant gradient phase delay ∆Φ/∆x along x direction at the interface between the two media. Compared with the left-side transmitted beam, the phase of the right-side transmitted beam is delayed by phase shift ∆Φ = 2πntd/λ. Based on the geometry relationship d = ∆xsinθ, we can get sinθ = ∆Φ/(k∆x), where k = 2π/λ is the wavevector. The refraction angle θ depends on the gradient phase distribution. If ∆Φ/∆x = 0, we can get a special case that θ = 0 with two normally transmitted beams.
2
Figure S2. Simulated amplitude deviation caused by the near field coupling of adjacent nanoslits with different orientation angles. (a) Simulated cross-polarization transmission spectra of periodic nanoslit unit cells with various orientation angles. Since the circularly polarized light is symmetrical in both x and y directions, the transmission spectrum will not change when the nanoslit is rotated by π/2 so that only four groups of spectra are displayed. (b) The amplitude deviations for three working wavelengths. The amplitude deviations from the mean value for eight orientation angles are calculated based on (a). The deviations are very small (about ±0.2%), which is much less than the transmission values (1%-4%). Thus, the effect of orientation angle on the amplitude can be ignored, providing the potential to control phase and amplitude independently.
3
Figure S3. Optical properties of uniform periodic nanoslit array, with period Px = Py = 230 nm, length L = 160 nm and width W = 80 nm. (a) Simulated cross-polarization and co-polarization transmission spectra with normally incident light. (b) Experimentally measured cross-polarization and co-polarization transmission spectra. (c) The normalized electric field |E| distributions for different incident circular polarizations RCP and LCP. (Top view, the light is incident from the bottom).
4
Figure S4. Experimental optical setup for the observation of 2D and 3D full-color holographic images. The incident angles for red, green, and blue laser light are 35°, 0°, and -35°, respectively.
Figure S5. The simulated color holographic image for an eight-level phase modulated CGH without the amplitude modulation. To calculate the phase-only CGH, a randomdistributed phase (from 0 to 2π) is added to each pixel of the virtual object to achieve a more uniform amplitude distribution [A(x,y)≈1], only the phase distribution is recorded. Compared with the simulated holographic images with both two-level amplitude and eight-level phase modulations (as shown in Figure 4b), the color of each letter suffers granular noise, i.e., speckle noise for the phase-only CGH.
5
Figure S6. The reconstruction of 3D full-color holographic images of six letters located on three different planes with red ‘R’ and cyan ‘C’ on z1 = 180 µm plane, green ‘G’ and magenta ‘M’ on z2 = 210 µm plane, and blue ‘B’ and yellow ‘Y’ on z3 = 240 µm plane. (a) The original letters placed at different z planes. (b, c) Simulated and measured 3D holographic images focused at different z planes of 180 µm, 210 µm, and 240 µm.
6
Figure S7. The reconstruction of 3D full-color holographic images of a color pyramid object with RGBW side lines. (a) The side and top views of the original color pyramid. (b, c) Simulated and measured 3D holographic images focused at different z planes of 160 µm, 185 µm, and 200 µm.
7
Full-Color Plasmonic Metasurface Holograms Weiwei Wan, Jie Gao* and Xiaodong Yang* Department of Mechanical and Aerospace Engineering, Missouri University of Science and Technology, Rolla, MO 65409, USA *E-mail: [email protected], [email protected]
1
Figure S1. Control the refraction angles by constant gradient phase distribution. Assuming that two light beams with a distance ∆x normally incident from the bottom medium, there is a constant gradient phase delay ∆Φ/∆x along x direction at the interface between the two media. Compared with the left-side transmitted beam, the phase of the right-side transmitted beam is delayed by phase shift ∆Φ = 2πntd/λ. Based on the geometry relationship d = ∆xsinθ, we can get sinθ = ∆Φ/(k∆x), where k = 2π/λ is the wavevector. The refraction angle θ depends on the gradient phase distribution. If ∆Φ/∆x = 0, we can get a special case that θ = 0 with two normally transmitted beams.
2
Figure S2. Simulated amplitude deviation caused by the near field coupling of adjacent nanoslits with different orientation angles. (a) Simulated cross-polarization transmission spectra of periodic nanoslit unit cells with various orientation angles. Since the circularly polarized light is symmetrical in both x and y directions, the transmission spectrum will not change when the nanoslit is rotated by π/2 so that only four groups of spectra are displayed. (b) The amplitude deviations for three working wavelengths. The amplitude deviations from the mean value for eight orientation angles are calculated based on (a). The deviations are very small (about ±0.2%), which is much less than the transmission values (1%-4%). Thus, the effect of orientation angle on the amplitude can be ignored, providing the potential to control phase and amplitude independently.
3
Figure S3. Optical properties of uniform periodic nanoslit array, with period Px = Py = 230 nm, length L = 160 nm and width W = 80 nm. (a) Simulated cross-polarization and co-polarization transmission spectra with normally incident light. (b) Experimentally measured cross-polarization and co-polarization transmission spectra. (c) The normalized electric field |E| distributions for different incident circular polarizations RCP and LCP. (Top view, the light is incident from the bottom).
4
Figure S4. Experimental optical setup for the observation of 2D and 3D full-color holographic images. The incident angles for red, green, and blue laser light are 35°, 0°, and -35°, respectively.
Figure S5. The simulated color holographic image for an eight-level phase modulated CGH without the amplitude modulation. To calculate the phase-only CGH, a randomdistributed phase (from 0 to 2π) is added to each pixel of the virtual object to achieve a more uniform amplitude distribution [A(x,y)≈1], only the phase distribution is recorded. Compared with the simulated holographic images with both two-level amplitude and eight-level phase modulations (as shown in Figure 4b), the color of each letter suffers granular noise, i.e., speckle noise for the phase-only CGH.
5
Figure S6. The reconstruction of 3D full-color holographic images of six letters located on three different planes with red ‘R’ and cyan ‘C’ on z1 = 180 µm plane, green ‘G’ and magenta ‘M’ on z2 = 210 µm plane, and blue ‘B’ and yellow ‘Y’ on z3 = 240 µm plane. (a) The original letters placed at different z planes. (b, c) Simulated and measured 3D holographic images focused at different z planes of 180 µm, 210 µm, and 240 µm.
6
Figure S7. The reconstruction of 3D full-color holographic images of a color pyramid object with RGBW side lines. (a) The side and top views of the original color pyramid. (b, c) Simulated and measured 3D holographic images focused at different z planes of 160 µm, 185 µm, and 200 µm.
7
