Vivid images can be created with metasurface color holograms by configuring their subwavelength planar structures. However, it is a challenge to arbitrarily tune the colors of the images after the metasurfaces have been fabricated. The common method of balancing the incident intensities of individual wavelengths directly or indirectly does not change the relative brightness/color of different parts of the image. Here, we use spin manipulation and wavelength multiplexing to change the color of not only the entire holographic image but also of defined parts of it. This is achieved by designing polarization-dependent dual images for each primary color. This work demonstrates potential for exploiting dielectric metasurfaces to reconstruct colorful images, which is highly desired in the holographic display industry.
© 2017 Optical Society of America
Color holograms play an increasingly important role in display area because of the capability of producing a colorful image of an object through reconstructing its light field by illuminating the recording media with a white light or a beam consisting of three primary colors (red, green, and blue). Recently, metasurfaces have been reported to realize color holograms by tailoring the size, shape, and material of the constituents, known as meta-atoms or metamolecules, to control the local light scattering properties at the subwavelength scale [1–12]. The plasmonic resonance effect of metal nanostructures has been extensively employed to explore color holograms with metal metasurfaces [8,13–18]. Usually, images with dual or triple colors can be reconstructed by a collinear multiwavelength beam at normal incidence by arranging two or three kinds of metal meta-atoms in a supercell metamolecule to encode the binarized phase or the amplitude distributions for two or three wavelengths independently [8,14–16]. Alternatively, the nonindependent encoding for different wavelengths allows one to select the wanted image by obliquely illuminating the metasurface with a specific wavelength beam at a specific angle [17,18]. Recently, dielectric metasurfaces consisting of relatively thick Mie scatters of high refractive indices have also been used to manipulate wavefronts [3–6,11,19–24]. It has been demonstrated that dielectric metasurfaces work efficiently at three or two wavelengths as achromatic lenses using aperiodic arrays of coupled silicon resonators or metamolecules . We have recently presented color holograms using three kinds of silicon nanoblocks that are independently multiplexed in subwavelength metamolecules . The light fields of three primary colors are highly confined within the corresponding nanoblocks, which allows individual phase for each primary color to be modulated independently to reconstruct color images without cross talk.
The dynamic full-color holographic display requires metasurface color holograms to be capable of being actively modulated, which is a challenge for the fabricated meta-atoms according to the predesigned sizes, shapes, and orientations. A feasible way to achieve holographic color modulation is to directly or indirectly balance the incident intensities of individual wavelengths. Directly tuning the intensity of one color wavelength simply alters its brightness of the entire image and thus cannot change the relative brightness among different parts of the image. Similarly, the metasurfaces consisting of polarization- and wavelength-selective meta-atoms are capable of indirectly balancing the intensities of two wavelengths by changing the orthogonally linear polarizations of the reconstruction beam [16,27,28]. Although the colors can also be modulated by singling out definite wavelengths from the incident white light by various color filters with high resolution and stability [29–35], they are not suitable for holographic display.
Here we design and demonstrate photon spin-controlled color-tunable holograms using Si metasurfaces. With this approach, each photon spin (right- or left-handed circular polarization, RCP or LCP, respectively) for each primary color reconstructs purposefully designed spin-controlled dual holographic images based on holographic multiplexing  or the shared-aperture technique , which is different from the previous spin-controlled holograms with a single reconstructed image or multiple uncoupled images. Any other incident polarization, consisting of the RCP and LCP components with a certain intensity ratio, reconstructs the twin images whose intensities vary according to the ratio, resulting in tunable colors by color mixing. We have successfully modulated the color of not only the entire holographic image but also its definite part by using two kinds of silicon nanoblocks to encode the wavefronts of two primary colors by the geometric phase. This approach is further extended to tuning the colors of holographic images using three primary colors by multiplexing three kinds of Si nanoblocks. The combination of the dielectric metasurface color holograms and the holographic multiplexing is expected to play a critical role in the holographic display industry by providing an effective pathway of tuning the color of holographic images and enlarging the information capacity of metasurfaces at multiwavelengths.
The designed reflective metahologram with polarization-controlled colors is illustrated in Fig. 1. The green and red beams are collimated to generate a mixed yellow beam to normally illuminate the metasurface. The coordinates are constructed with the origin defined at the center of the incident beam, the plane at the screen parallel to the metasurface, and the axis along the beam. In the experiment, the holographic images of interest are all reconstructed at , while their conjugate images at are blocked. When the polarization of a beam is changed by a polarizer and a quarter-wave plate, the distinct colorful off-axis images are reconstructed. Particularly, we design the target chameleon from changing its color from red to green and the target tree branch remains green.
The metaholograms are fabricated with the standard silicon-on-insulator technology. The subwavelength supercells consist of two types of silicon nanoblocks ( and ) on a 2 μm thick silicon-oxide layer and bulk handle wafer, as illustrated in Fig. 2(a). The thickness of the silicon nanoblock is . The period of the supercell is set to be , which is shorter than the wavelengths to reduce higher order diffractions. The local phase is determined by the dimensions of the nanoblocks along the and directions ( and ) and the in-plane orientation (). When a circularly polarized light beam is normally incident, the reflected light is generally composed of two parts: one with the handedness opposite to the incident light without a phase delay, and the other with the same handedness with a phase delay of with depending on the handedness of the incident beam, RCP or LCP. The energy ratio of the reflected beam engineered by the geometric phase to the incident beam is defined as diffraction efficiency (DE). The corresponding DEs as functions of and are illustrated in Figs. 2(b) and 2(c). The maximum DE is 95% or 75% at 632.8 nm [the red triangle in Fig. 2(b)] or at 532 nm [the red diamond in Fig. 2(c)], respectively. The and form a metamolecule that provides multichannel full-phase manipulations at both 632.8 nm and 532 nm. To maximize the difference in DE for two wavelengths upon illumination by dual beams simultaneously, we set the DEs of at 632.8 nm and at 532 nm to be 51% and 0.8%, respectively, and the DEs of at 532 nm and at 632.8 nm to be 64% and 1.7%, respectively. For the chosen , , and the supercell, each silicon nanoblock can strongly confine the light to avoid interacting with others, which ensures the independent manipulation of dual-wavelength wavefronts.
The above design enables the entire holographic image to continuously change color without chromatic aberration for the two wavelengths upon varying the incident polarization, as shown in Fig. 3. The reconstructed “chameleon” in Fig. 3 is designed to be identical for both RCP and LCP but with red and green colors (at 632.8 nm and 532 nm), respectively. For other polarizations, a solid mixed color results. The required phase distributions at the two wavelengths are retrieved individually using the Gerchberg–Saxton (GS) algorithm, and thus the nanoblocks are accordingly oriented in plane. Since the images are reconstructed with the opposite circular polarization states at the two wavelengths, rotates clockwise and anticlockwise. When we continually rotate the quarter-wave plate after a polarizer to select the incident polarization state, as shown in Fig. S1 in Supplement 1, the polarization states change accordingly and form a curved trajectory on the chromatic Poincare sphere, as shown in Fig. 3(c). The reconstructed image is red for RCP at the north pole of the sphere, while it is green for LCP at the south. The color can keep on changing from red, orange, yellow to green with the polarization states shown on the sphere, depending on the ratio between RCP and LCP or the Stokes parameter . The colors of the captured images for six incident polarization states are consistent with those shown on the Poincare sphere. The replacement of one by two smaller nanoblocks marked as ( and ) with the flip of the in-plane orientations of all can change the image colors between green and blue. This is demonstrated by the colorful Poincare sphere and experimental results, as revealed in Fig. 3(d).
Figure 4 demonstrates that multiple polarization-dependent colorful images can be multiplexed onto a single metasurface. For the identical polarization operation, the image can be decomposed into two solid color images marked as and (RCP) or and (LCP) in Figs. 4(a) and 4(b) or Figs. 4(e) and 4(f), respectively, where the first and second subscript letters stand for the color and the polarization, respectively. () means the image reconstructed using the RCP (LCP) at the wavelength of 632.8 nm (532 nm). The identical part of the image showing red at RCP changes into green at LCP and vice versa. The reconstructed images in Figs. 4(d) and 4(h) agree well with the target images in Figs. 4(c) and 4(g). Note that this metasurface achieves the phase manipulation of at dual wavelengths, and thereby polarization-dependent color information for one object can be thus distinguished. This makes it possible to detect polarization information at dual wavelengths simultaneously using a single metadevice.
Based on the above results, we aim to design a metasurface capable of realizing the modulation of polarization-dependent and -independent color for different parts of an image by simply varying the polarization. Figure 5 shows the reconstructed holographic image with the color of one part of the image changed (polarization-dependent color) and that of the left part remaining unchanged (polarization-independent color) upon varying the polarization of an incident mixed beam. Rows A and B illustrate the target images (a chameleon on a tree branch) and the experimental results when both lasers at 632.8 nm and 532 nm illuminate simultaneously. By continually changing the incident polarization state from RCP to linear to LCP, the chameleon changes its color from green to yellow to red while the tree branch is always green. This can be understood by the experimental results upon illumination by either the 632.8 nm or 532 nm laser. The RCP beam reconstructs a red image and the LCP beam reconstructs a green one for the chameleon, as explained above. To realize the polarization-independent color of the tree branch, the key is to encode the complex combination of the phase shifts of the identical image for both RCP and LCP into the in-plane orientations of . Thus, any combination of RCP and LCP generates the green tree branch. The reconstructed images with a high signal-to-noise ratio are of good quality, indicating the potential of silicon nanoblocks with PB phase to create polarization and color multiplexing holograms. The measured efficiencies of the holograms are 25% and 48% for green and red wavelengths, respectively.
The combination of spin manipulation and wavelength multiplexing provides a feasible way to actively control the color of the holograms within a large color gamut, as demonstrated in the following. As any colorful images can be divided into three RGB subimages, the phase distribution for each subimage is independently retrieved using the GS algorithm. Then, it is converted to the distribution of or or . Finally, , , and are multiplexed in subwavelength scale to form the desired metasurface. To reduce the possible cross talk between the three wavelengths, the sizes of the (, ), (, ), and (, ) are slightly altered compared with the cases for two primary colors. The period of the unit is also changed to 350 nm so that one , one , and two can be multiplexed into a single unit.
The scanning electron microscope (SEM) image of a part of the fabricated dielectric metasurface is shown in Figs. 6(a) and 6(b), where the three kinds of Si nanoblocks are densely arranged to make sure the multiwavelength phase controlling at the subwavelength scale. Experimental results are shown in Figs. 6(c)–6(g). Benefited from the independent wavefront manipulations at the three wavelengths, every alphabet is reconstructed with the correct size and position, which is vital for realizing the spin-controlled tunable colors. Furthermore, distinct multiple colors are presented for different alphabets while we alter the incident polarization. For instance, the alphabet “A” in the first row changes from red to blue as the incident polarization changes from RCP to LCP. At the same time, the alphabet “A” in the second row follows a completely different color path (changing from yellow to purple). Accordingly, other alphabets follow their own different color paths. Therefore, by using the spin manipulation and the wavelength multiplexing, abundant holographic images with their colors controlled by the incident polarization state can be realized, which will facilitate the display industry with miniaturized tunable metadevices.
In conclusion, we have demonstrated dielectric metasurface color holograms with polarization-controlled tunable colors by integrating space, polarization, and wavelength multiplexings. With the PB phase to manipulate the wavefronts of two wavelengths, high-quality color holograms are successfully reconstructed and their colors can be changed entirely or in specific parts of the image by varying the polarization of the incident beams. Furthermore, this approach is extended to the full-color tuning for three primary colors and multiple images by multiplexing three kinds of Si nanoblocks to form a metamolecule. The combination of the color holograms and the holographic multiplexing for dielectric metasurfaces not only provides an effective way to continually tune the color of holographic images but increases the information capacity at multiwavelengths, which renders dielectric metasurfaces very promising in both holographic display industry and information processing.
National Basic Research Program of China (2013CB921904); National Natural Science Foundation of China (NSFC) (11474010, 11627803, 61590933); Youth Innovation Promotion Association of the Chinese Academy of Sciences (2015030); Chinese Academy of Sciences (CAS) Key Technology Talent Program, Key Research Program of Frontier Sciences, (QYZDB-SSW-SYS031).
See Supplement 1 for supporting content.
1. N. Yu, P. Genevet, M. A. Kats, F. Aieta, J. P. Tetienne, F. Capasso, and Z. Gaburro, Science 334, 333 (2011). [CrossRef]
2. X. Chen, L. Huang, H. Mühlenbernd, G. Li, B. Bai, Q. Tan, G. Jin, C. Qiu, S. Zhang, and T. Zentgraf, Nat. Commun. 3, 1198 (2012). [CrossRef]
3. D. Lin, P. Fan, E. Hasman, and M. L. Brongersma, Science 345, 298 (2014). [CrossRef]
4. S. Vo, D. Fattal, W. V. Sorin, Z. Peng, T. Tran, M. Fiorentino, and R. G. Beausoleil, IEEE Photon. Technol. Lett. 26, 1375 (2014). [CrossRef]
5. Y. Yang, W. Wang, P. Moitra, I. I. Kravchenko, D. P. Briggs, and J. Valentine, Nano Lett. 14, 1394 (2014). [CrossRef]
6. A. Arbabi, Y. Horie, A. J. Ball, M. Bagheri, and A. Faraon, Nat. Commun. 6, 7069 (2015). [CrossRef]
7. X. Ding, F. Monticone, K. Zhang, L. Zhang, D. Gao, S. N. Burokur, A. de Lustrac, Q. Wu, C. Qiu, and A. Alù, Adv. Mater. 27, 1195 (2015). [CrossRef]
8. Y. Huang, W. T. Chen, W. Tsai, P. C. Wu, C. Wang, G. Sun, and D. P. Tsai, Nano Lett. 15, 3122 (2015). [CrossRef]
9. L. Huang, X. Chen, H. Mühlenbernd, H. Zhang, S. Chen, B. Bai, Q. Tan, G. Jin, K. Cheah, C. Qiu, J. Li, T. Zentgraf, and S. Zhang, Nat. Commun. 4, 2808 (2013). [CrossRef]
10. X. Ni, A. V. Kildishev, and V. M. Shalaev, Nat. Commun. 4, 2807 (2013). [CrossRef]
11. A. Arbabi, Y. Horie, M. Bagheri, and A. Faraon, Nat. Nanotechnol. 10, 937 (2015). [CrossRef]
12. E. Hasman, V. Kleiner, G. Biener, and A. Niv, Appl. Phys. Lett. 82, 328 (2003). [CrossRef]
13. M. Ozaki, J. Kato, and S. Kawata, Science 332, 218 (2011). [CrossRef]
14. Y. Montelongo, J. O. Tenorio-Pearl, C. Williams, S. Zhang, W. I. Milne, and T. D. Wilkinson, Proc. Natl. Acad. Sci. USA 111, 12679 (2014). [CrossRef]
15. B. Walther, C. Helgert, C. Rockstuhl, F. Setzpfandt, F. Eilenberger, E. Kley, F. Lederer, A. Tuennermann, and T. Pertsch, Adv. Mater. 24, 6300 (2012). [CrossRef]
16. Y. Montelongo, J. O. Tenorio-Pearl, W. I. Milne, and T. D. Wilkinson, Nano Lett. 14, 294 (2014). [CrossRef]
17. W. Wan, J. Gao, and X. Yang, ACS Nano 10, 10671 (2016). [CrossRef]
18. X. Li, L. Chen, Y. Li, X. Zhang, and M. Pu, Sci. Adv. 2, e1601102 (2016). [CrossRef]
19. Y. F. Yu, A. Y. Zhu, R. Paniagua-Domínguez, Y. H. Fu, B. Luk’Yanchuk, and A. I. Kuznetsov, Laser Photon. Rev. 9, 412 (2015). [CrossRef]
20. F. Aieta, M. A. Kats, P. Genevet, and F. Capasso, Science 347, 1342 (2015). [CrossRef]
21. W. Zhao, B. Liu, H. Jiang, J. Song, Y. Pei, and Y. Jiang, Opt. Lett. 41, 147 (2016). [CrossRef]
22. K. Huang, Z. Dong, S. Mei, L. Zhang, and Y. Liu, Laser Photon. Rev. 10, 500 (2016). [CrossRef]
23. M. Khorasaninejad, A. Ambrosio, P. Kanhaiya, and F. Capasso, Sci. Adv. 2, e1501258 (2016). [CrossRef]
24. Q. Li, F. Dong, B. Wang, F. Gan, J. Chen, Z. Song, L. Xu, W. Chu, Y. Xiao, Q. Gong, and Y. Li, Opt. Express 24, 16309 (2016). [CrossRef]
25. E. Arbabi, A. Arbabi, S. M. Kamali, Y. Horie, and A. Faraon, Optica 3, 628 (2016). [CrossRef]
26. B. Wang, F. Dong, Q. Li, D. Yang, C. Sun, J. Chen, Z. Song, L. Xu, W. Chu, Y. Xiao, Q. Gong, and Y. Li, Nano Lett. 16, 5235 (2016). [CrossRef]
27. O. Eisenbach, O. Avayu, R. Ditcovski, and T. Ellenbogen, Opt. Express 23, 3928 (2015). [CrossRef]
28. E. Arbabi, A. Arbabi, S. M. Kamali, Y. Horie, and A. Faraon, Opt. Express 24, 18468 (2016). [CrossRef]
29. D. Franklin, Y. Chen, A. Vazquez-Guardado, S. Modak, J. Boroumand, D. Xu, S. Wu, and D. Chanda, Nat. Commun. 6, 7337 (2015). [CrossRef]
30. L. Duempelmann, A. Luu-Dinh, B. Gallinet, and L. Novotny, ACS Photon. 3, 190 (2016). [CrossRef]
31. Z. Li, J. Kapraun, J. Ferrara, and C. J. Changhasnain, Optica 2, 255 (2015). [CrossRef]
32. L. Cao, P. Fan, E. S. Barnard, A. M. Brown, and M. L. Brongersma, Nano Lett. 10, 2649 (2010). [CrossRef]
33. Y. Gu, L. Zhang, J. K. W. Yang, S. P. Yeo, and C. Qiu, Nanoscale 7, 6409 (2015). [CrossRef]
34. A. Kristensen, J. K. W. Yang, S. I. Bozhevolnyi, S. Link, P. Nordlander, N. J. Halas, and N. A. Mortensen, Nat. Rev. Mater. 2, 160881 (2017). [CrossRef]
35. V. Vashistha, G. Vaidya, R. S. Hegde, A. E. Serebryannikov, N. Bonod, and M. Krawczyk, ACS Photon. 4, 1076 (2017). [CrossRef]
36. L. Huang, H. Mühlenbernd, X. Li, X. Song, B. Bai, Y. Wang, and T. Zentgraf, Adv. Mater. 27, 6444 (2015). [CrossRef]
37. E. Maguid, I. Yulevich, D. Veksler, V. Kleiner, M. L. Brongersma, and E. Hasman, Science 352, 1202 (2016). [CrossRef]