1. Introduction

Photonic printing based on optical resonators [1–5] has been a focus of intense research due to its potential as an alternative to conventional printing methods using chemical pigments. Compared to conventional printing methods where the color of each pixel is determined by the inherent absorption of pigments, the geometrical dimensions of a pixelated optical resonator determine the color of each pixel in photonic printing technologies. This new type of a color element enables to achieve higher image resolution, flexibility in material choice, excellent mechanical durability, and enhanced stability against color fading. In addition, photonic prints can change colors or display different contents under external stimuli such as mechanical deformations [3,6,7], electrical biases [8–10], and chemical reactions [11,12], or different viewing conditions defined by optical polarizations [13–15] and viewing angles [16], by employing unconventional materials whose mechanical, chemical, and optical properties change in response to such stimuli. This dynamic aspect of a photonic print that is inaccessible with conventional printing methods can open new opportunities for a variety of applications including displays [10,12], anti-counterfeiting labels [17–19], barcodes [5], artistic decorations [2,20], optical data storages [21,22], and holography [23–26].

Toward development of new photonic printing technologies, a number of device platforms such as plasmonic nanostructures [2,20,27], photonic crystals [1,28], dielectric Mie scatterers [4,14], and Fabry-Perot (FP) resonators [29–31] have been demonstrated so far (see Table 1 for summary of photonic printing technologies). Among them, a FP-resonator-based approach is very promising owing to not only high optical efficiency but also simplicity in fabrication and scalability that inherit from their simple architecture. Since the optical resonance of a FP resonator is determined by the thickness of a resonant cavity between two mirrors, however, printing an image using FP resonators still requires a complicated fabrication process to integrate pixelated FP resonators with different thicknesses onto a single substrate, such as multiple vacuum depositions [32], sequential transfer-printings [29], and grayscale photolithography [30]. In addition, the surface of a printed image forms a stepwise landscape, which is undesirable for certain applications such as displays and anti-counterfeiting labels where an uneven surface can result in inhomogeneous electrical responses or leave traces for mechanical replication. From the viewpoint of a functionality, the optical nearfields of a FP resonator are mostly concentrated within the resonant cavity and accordingly, many of conventional strategies for the dynamic tuning of a printed image that rely on the sensitivity of the nearfields to external stimuli become ineffective. Except for a few cases [12,31], therefore, most of previous works based on a FP-resonator-based approach have demonstrated printing of static images. For use in a wider range of applications, aforementioned limitations should be overcome to facilitate scalable fabrication and enable new functionalities.

In this work, we demonstrate the new device platform based on anisotropic FP resonators that allows for the dynamic selection of an image by the polarization of incident light. In contrast to conventional FP resonators having an isotropic film for a resonant cavity, the dynamic selection becomes possible by employing a film of liquid crystal polymers (LCPs) aligned by directional molecular registration (DMR). Note that LCs and LCPs have been conventionally used for the optical storage of black-and-white or grayscale images [24,33–35]. With the help of FP resonances, our system can display colors that are determined by the molecular directions of LCPs for the given incident polarization and thickness of a resonant cavity and accordingly, construct a color image with FP resonators of the same thickness, yielding the continuity and uniformity of colors across the pixels. In other words, the DMR in our case produces no geometrical barrier between two adjacent FP resonators (or from one pixel to the other). Based on this feature, we demonstrate a prototype of an anticounterfeiting label that can hide and disclose different images under specific viewing conditions defined by polarizations of incident light with inherent anti-replicability. In addition, we show that this concept can be extended to the holographic image selection.

Table 1. Selective photonic printing technologies.

View Table

2. Color encoding by directional molecular registration

In our photonic print, each FP resonator incorporating the LCP layer was used as an individual pixel. As shown in Fig. 1(a), the LCP layer (thickness of d₁ and refractive index of n_LCP) was placed on a photoalignment layer (thickness d₂ and refractive index n_AL) prepared on a transreflective layer (thickness: t_T). Basically, the optical medium (with the total thickness d = d₁ + d₂), composed of the photoalignment layer and the LCP layer, provides an anisotropic resonant cavity between two TRLs. Note that the optical path depends on the total thickness and the effective refractive index of the resonant cavity in the FP resonator [38]. For the given thickness of the resonant cavity, the resonant wavelength changes with the alignment direction of the LCP molecules with respect to the polarization state of incident light, i.e., in terms of the polarization angle of θ as shown in Fig.Oka 1(b). In a certain range of optical paths (n_LCP·d₁ + n_AL·d₂), the resonant wavelengths fall into the visible regime so that the direction of the DMR can be recognized with naked eyes.

 

Fig. 1. Device structure and optical properties. (a) Schematic illustration of a Fabry-Perot resonator having a liquid crystal polymer (LCP) layer inside the resonant cavity. (b) The definition of the polarization angle of θ with respect to the LCP director. (c) Optical transmittance in the visible range for θ from 0° to 90° at the interval of 15° upon the normal incidence of broadband light. (d) Microscope images of different colors for θ from 0° to 90° at the interval of 22.5°. Scale bars represent 500 µm.

Download Full Size | PDF

Figure 1(c) shows how the resonant peaks in the visible spectra of the FP device change with the polarization angle of 0° and 90°. For the device, d₁ = 603, d₂ = 84, and t_T = 28 nm for silver. Three resonant peaks at the 3rd, 4th, and 5th orders appeared in the visible range. The wavelengths, the orders, and the total number of the resonant peaks vary with the magnitudes of d₁ and d₂ together with the refractive indices of the materials comprising the resonant cavity (see Fig. 5 for the peak wavelengths versus the resonant cavity thickness and Fig. 6 for the color palette). The optical transmittance at the resonant peak was over 45%. Note that this magnitude is significantly larger than the transmittances of other polarization-dependent photonic prints based on surface plasmon resonances [13,15,39] and comparable to prints based on all-dielectric scatterers [14,40,41]. On the other hand, the linewidths of the resonance peaks (around 25 nm) are much smaller than those of other photonic prints. The optical transmittance can be further improved at the expense of the linewidth by decreasing the thicknesses of TRLs. Two spectra at θ = 0° and 90° in Fig. 1(c) (denoting s₁ and s₂, respectively) were chosen as two orthogonal components of the encoding basis for the information to be stored. The resonance peaks of the spectra s₁ shift to longer wavelengths compared to those of the spectra s₂ since the extraordinary refractive index of n_LCP, corresponding to the fast optic axis of the LCP, is larger than the ordinary refractive index. For the polarization angle θ of the incident light, the spectra are then given as cos²θ·s₁+sin²θ·s₂. As shown in the microscopic images in Fig. 1(d), different colors were indeed realized at different polarization angles of the incident light. Such spectral change with the incident polarization enables to utilize the anisotropic FP resonator as a polarization-dependent color pixel. We define the “reading” direction as the direction along which the incident light is polarized perpendicular to the fast optic axis of the LCP, which renders corresponding pixel to be purple. According to this definition, the polarization for the DMR and the reading beam, θ_DMR and θ_OUT, are identical while the alignment direction of the LCP (θ_AL) should be perpendicular to both the recording and reading polarizations (θ_DMR = θ_OUT $\bot $ θ_AL).

3. Polarization-dependent image selection

Based on the findings above, we demonstrate the dynamic selection of a color image among predefined ones according to the polarization state of the incident light. Such dynamic selection functionality has been a focus of interest in anti-counterfeiting applications based on photonic printing. An example of the photonic print carrying two randomly generated QR codes, A and B, was illustrated in Fig. 2(a). For recording images through the DMR, the two QR codes were first decomposed into three independent patterns of A-C, C (= A∩B), and B-C, as shown in Fig. 2(b). The three patterns were then recorded along three directions of the DMR corresponding to θ_DMR = 0°, 45°, and 90° in Fig. 2(c) for the LCP alignment. Different images were stored through a series of the DMR processes using different photomasks for different images. Note that the directions of the DMR for different images should be separated by 180°/m to reduce the crosstalk between them where m is the number of images. The DMR processes for the patterns of C, A-C, and B-C were carried out in sequence. Note that the DMR process was not performed for the pattern for (AUB)^C in this proof of concept experiment since each image could be still resolved as shown in the following results. For reading one out of two original QR codes separately, the polarization states of θ_OUT = 22.5° and 67.5° (bisectional directions between two adjacent directions of the DMR) of a reading beam were used as shown in Fig. 2(d).

 

Fig. 2. Polarization-dependent quick-response (QR) code. Schematic illustrations of (a) randomly generated QR codes, A and B, and (b) three different sets of the QR images decomposed from QR codes, A and B, for recording. (c) Recording conditions of the polarization states (θ_DMR) for three different sets of the QR images. (d) Reading conditions of the polarization states (θ_OUT) of incident broadband light for reconstructing the original QR codes, A and B, separately. Microscopic images of the recorded polarization-dependent QR images upon the incidence of (e) unpolarized light, (f) polarized light parallel to the directions of the DMR for recording, and (g) polarized light according to the reading conditions (or along the bisectional directions between two adjacent directions of the DMR) for the QR codes, A and B, given in (d). In all cases, scale bars represent 1 cm.

Download Full Size | PDF

Under unpolarized light, no discernable image was observed as shown in Fig. 2(e). This is because the unpolarized light experiences the average value of n_eff and there are no geometrical barriers that can diffract incident light, across individual pixels or elements over the whole sample. In contrast to other polarization-selective photonic prints where geometrical discontinuities are inevitably involved and can produce obscure boundaries between different color pixels, thus, our photonic print under unpolarized light is completely clear of any visual traces for the recorded images. Also, the absence of geometrical barriers and the resultant flat top surface of our photonic print makes almost impossible to counterfeit by mechanical replication. Figure 2(f) shows the photographs of the recorded images of A-C, C, and B-C (in deep blue color in the background of green color) that were observed by the reading beam with the polarization states of θ_OUT = 0°, 45°, and 90°, respectively, i.e., along the directions of the DMR. Note that essentially the same patterns in Fig. 2(b) were seen. Clearly, different images appeared due to the difference of n_LCP in pattern by pattern. Upon the incidence of the polarization states of the bisectional directions, θ_OUT = 22.5° and 67.5°, QR codes of A and B were well reproduced, respectively, as shown in Fig. 2(g). Under these circumstances, n_LCP for A-C is the same as that for C. Similarly, n_LCP for B-C is identical to that for B. In addition to the dynamic selection functionality that is defined at the molecular level and accordingly, inherently impossible to replicate, our photonic print allows for scalable and cost-effective fabrication and makes an excellent platform for anti-counterfeiting applications.

4. Holographic image selection

We describe how anisotropic FP resonators can be used for the dynamic selection among two holographic images. As shown in Fig. 3(a), the original images of two alphabets, ‘A’ and ‘B’, are Fourier-transformed into the binary-type hologram patterns. In a similar fashion to the case of the QR codes, three different patterns of C (=A∩B), A-C, and B-C, derived from two Fourier-transformed patterns of ‘A’ and ‘B’, were recorded through a series of the DMR processes. The directions of the DMR during recording were denoted by green double-headed arrows. Under two different states of the reading polarization of the monochromatic light, the holographic images were reconstructed from the binary-type hologram patterns through the inverse Fourier transform as shown in Fig. 3(a). Depending on the reading polarization lying along the bisectional direction of two adjacent directions of the DMR (denoted by red double-headed arrow), only one of two original images (‘A’ and ‘B’) can be reconstructed. It is desirable that the center wavelength of a monochromatic reading beam should coincide with one of the resonance wavelengths in the FP resonator to achieve the high contrast between the image and the background.

 

Fig. 3. Recording and reading Fourier-transformed patterns. (a) Schematic illustration of recording the Fourier transformed patterns of the original images, ‘A’ and ‘B’, and reconstructing them. (b) Fourier transformed patterns of ‘A’ and (c) the corresponding microscopic image observed at the reading polarization angle of θ_OUT = 60°. (d) Fourier transformed patterns of ‘B’ and (e) the corresponding microscopic image observed at θ_OUT = 0°. (f) The values of the normalized intensity in red channel for A-C, B-C, and C as a function of θ_OUT. Two grey bands in (f) denote the ranges of the reading polarization angle for ‘A’ and ‘B’ at which the images of ‘A’ and ‘B’ were reconstructed and discernable with high contrast. In (c) and (e), scale bars represent 300 µm.

Download Full Size | PDF

We chose d₁ = 603 nm and d₂ = 84 nm for device fabrication. The pixels in the background, (AUB)^C, were covered with a 30 nm-thick aluminum film. Figures 3(b) and 3(c) show the binary-type Fourier-transformed patterns of ‘A’ and the corresponding microscopic image observed at θ_OUT = 60°, respectively. In Fig. 3(c), the pixels for both A-C and C exhibited the same color (purple) since the reading polarization was along the bisectional direction between A-C and C. The pixels for B-C appeared green in color and those for (AUB)^C (background) were completely black. Figures 4(d )and 4(e) are the binary-type Fourier-transformed patterns of ‘B’ and the corresponding microscopic image observed at θ_OUT = 0°, respectively. Similar to ‘A’, in Fig. 3(e), the pixels for both B-C and C showed the identical color (purple) and those for A-C became distinct from others with the black background. Note that the resonance wavelength of the third order in our FP resonator was fairly close to the central wavelength of a red diode-pumped solid-state laser (660 nm) upon the input polarization parallel to the recording polarization (see Fig. 1(b)). This leads to the enhancement of the contrast between the transmitted state and the blocked state of light in an individual pixel.

 

Fig. 4. Holographic image reconstruction. (a) Optical setup for reconstructing holographic images from Fourier-transformed patterns. Here, f represents the focal point of the Fourier lens. In addition, f_x and f_y denote the Cartesian axes in the Fourier space. Photographs showing the selection and reconstruction of one (in the complex or conjugate field) of two holographic images (of the zeroth order) at the reading polarization angle of (b) θ_OUT = 60° and (c) θ_OUT = 0°.

Download Full Size | PDF

Figure 3(f) shows the normalized transmitted intensity measured at the wavelength of 660 nm as a function of θ_OUT. For the patterns of A-C, the intensity was minimum at θ_OUT = 0°, i.e., when the input polarization was parallel to the recording polarization while it was maximum at θ_OUT = ±90°. For the patterns of B-C, the intensity was maximum at θ_OUT = -60° and minimum at θ_OUT= 60°. The requirements for the selection of the reading polarization angles were that (1) the contrast (or the relative intensity difference) between A and B should be the highest and (2) the magnitude of the intensity itself (A or B) should be sufficiently large among all possible combinations of A-C, B-C, and C. In our case, such requirements were θ_OUT = 60° for the patterns of A and θ_OUT = 0° for the patterns of B as described already in Fig. 3(a).

We now describe the reconstruction of the holographic images from the photonic print. Figure 4(a) is the schematic diagram showing the optical setup for the selection and reconstruction of the holographic images. In our case, the presence of optical resonances enables to obtain the intensity contrast between binary-type hologram pixels without the help of crossed-polarizers. Details of the experiments performed for the image reconstruction were given in Appendix 1. Figures 4(b) and 4(c) show the photographs of the holographic image of the zeroth order observed at θ_OUT = 60° and 0°, respectively. Clearly, the original information of alphabet ‘A’ and ‘B’ was reconstructed in the form of an individual holographic image (either in complex or conjugate field) selected according to the reading polarization. The crosstalk between ‘A’ and ‘B’ at θ_OUT = 0° and 60° can be substantially reduced by matching the resonance wavelength of the FP resonator with the central wavelength of a probing laser and also by increasing the thickness of transreflective layer to suppress the background transmission. In contrast to holography based plasmonic [25,42,43] and dielectric [44,45] metasurfaces, our DMR-based FP architecture allows high throughput, the scalability (over 40,000 pixels in 2 cm $\times $ 2 cm area), the cost-effective fabrication, and the flexibility in design. Although it is limited by the spatial resolution of the coarse photomask (70 µm) used in this proof-of-concept experiment, we expect the resolution of our photonic print to scale with the spatial resolution of a photomask up to a micron, which is the minimum resolution for a typical photoalignment process [46–48].

5. Conclusions

We demonstrated the photonic printing using the anisotropic FP resonators that allow the polarization-dependent selection of an image. In contrast to previous approaches based on plasmonic and dielectric nanostructures, our device platform is compatible with a scalable and high-throughput photolithography system by employing the DMR processes of the LCP molecules. Since the optical information is recorded by the DMR from the interface in a collective manner, our photonic print is difficult to replicate by mechanical copying, which makes it promising for anticounterfeiting applications. Furthermore, our photonic print is capable of storing holographic data and selectively reconstructing them at the specific polarization as well as center wavelength of incident light in a simpler setup without crossed polarizers. Our device platform for photonic prints will bring much impact on the next generation of anti-counterfeiting labels and information storage media.

Appendix 1. Methods and Materials

Image recording by the directional molecular registration: For the DMR, a photosensitive polymer solution of AL 1084 (JSR) was spin-coated onto the TRL. In principle, upon the illumination of ultra-violet (UV) light, the photosensitive polymer layer is capable of aligning the LCP molecules along the direction perpendicular to the polarization of the UV light. In other words, the DMR can be achieved on the photo-alignment layer. In order to record different images, the multiple exposure of the UV light (with the intensity of 16 mW/cm²), polarized linearly in different directions, was performed on the photo-alignment layer for 10 minutes each through different photomasks. Note that the alignment marks were needed for precisely defining the patterns of the photo-alignment layer without leaving out optical traces or boundaries across pixels. During spin-coating the solution of photo-curable LCP (RMS-013C; Merck) on the photo-alignment layer, the LCP molecules were spontaneously aligned according to the polarization direction of the UV light. The LCP layer was then cured by the UV exposure for 1 min. The top TRL of a 28 nm-thick silver film was finally prepared by the thermal evaporation in vacuum as described above.

Optical transmittance measurement: The optical transmittance of an anisotropic Fabry-Perot resonator was measured using a commercial visible spectrometer (V530; Jasco). A polarizer was placed between a halogen lamp and the sample. The polarization angle (θ) was adjusted by rotating the sample with respect to the optic axis of the polarizer. The optical transmittance through the sample was normalized relative to that through the polarizer.

Holographic image reconstruction: For restoring the original information through reconstruction in the form of a holographic image, the sample was mounted on an optical setup in Fig. 4(a). A diode-pumped solid-state laser with the center wavelength of 660 nm was used as a monochromatic light source. The polarized direction of the laser light was adjusted by rotating a half-wave plate in front of the light source. For high coherency and uniformity, the laser beam was focused using a microscope objective lens with 20× magnification, and transmitted through a high energy pinhole with an aperture of 5 um in diameter to selectively filtrate the center region of beam. It was then collimated using another lens with the focal length of 200 mm to be incident on the sample along the nearly normal direction. After being converged using a Fourier transforming lens, the light rays from individual pixels interfere and reconstruct the holographic image with its conjugate and the constant residual. By spatially filtering the conjugate term and the residual term, a reconstructed hologram of the complex-valued object can be obtained. The holographic image was finally captured using a digital single-lens reflex (Sony), whose focal point was adjusted to locate in the image plane.

Appendix 2. Thickness-dependent color palette

From the viewpoint of the design versatility, it is important to establish the relationship between the geometrical parameters and the optical properties of the photonic print. Figure 5 shows the peak wavelength of each resonance band as a function of the resonant cavity thickness varying from 415 nm to 1027 nm. Here, filled and open symbols represent the data for the input polarization parallel and perpendicular to the alignment direction of LCP molecules, respectively. In general, they are covariant with each other in a certain range of the spectral distance resulting from the difference in n_eff. The order of resonance increases with the thickness of the resonant cavity. For given order of resonance, as the thickness of the resonant cavity increases, the peak wavelengths shift toward longer wavelengths as expected from the resonant equation. For the given thickness of the resonant cavity, the peak wavelength becomes shorter with increasing the order of resonance. Note that the variations of the peak wavelength with the RC thickness and the order of resonance lead to the color change in the visible range. As shown in Fig. 6, the color design of the photonic print is available by adjusting the thickness of the resonant cavity. The color coordinates calculated from the optical microscopic images in Fig. 6 are summarized in Fig. 7 along with the numerical results for thinner resonant cavities. By using thinner cavities, color saturation can be improved at the expense of a tuning range by optical polarization.

 

Fig. 5. The peak wavelengths at the resonance orders from the third to the ninth in the Fabry-Perot resonator as a function of the thickness of the resonant cavity. Filled and open symbols represent the data for the input polarization parallel and perpendicular to the alignment direction of liquid crystal polymer molecules, respectively.

Download Full Size | PDF

 

Fig. 6. Microscopic images, comprising a color palette, for different values of the thickness of the resonant cavity. Scale bars represent 500 µm.

Download Full Size | PDF

 

Fig. 7. Chromaticity diagram (CIE 1931) showing the color coordinates of different samples presented in Fig. 6 (black dots). Red and blue squares represent the numerical results for Fabry-Perot resonators with thinner resonant cavities.

Download Full Size | PDF

Appendix 3. Generation of binary holograms

For facilitating the experimental demonstration, a binary computer-generated hologram was used in the experiment. According to the Fraunhofer diffraction condition, the optical field of a complex-valued object, U(ξ, η), i.e, the letter “A” and the random phase distribution can be directly calculated from a Fourier transform of the object. Considering that the fabricated sample functions as an amplitude modulator, the real part of U(ξ, η), S(ξ, η), is extracted to obtain the amplitude of the CGH which is given by

(1)$$S({\xi ,\eta } )= \frac{1}{2}[{U({\xi ,\eta } )+ {U^ \ast }({\xi ,\eta } )} ],$$

and the binary hologram, B(ξ, η), is then determined as

(2)$$B({\xi ,\eta } )= \left\{ \begin{array}{ll} 1, & S({\xi ,\eta } )\ge 0\\ 0, & S({\xi ,\eta } )\le 0. \end{array} \right.$$

The pixels for B(ξ, η) = 1 transmit the light while those for B(ξ, η) = 0 block the light.

Funding

Ministry of Education (Brain Korea 21 Plus Project).

References

1. H. Fudouzi and Y. Xia, “Colloidal crystals with tunable colors and their use as photonic papers,” Langmuir 19(23), 9653–9660 (2003). [CrossRef]  

2. K. Kumar, H. Duan, R. S. Hegde, S. C. Koh, J. N. Wei, and J. K. Yang, “Printing colour at the optical diffraction limit,” Nat. Nanotechnol. 7(9), 557–561 (2012). [CrossRef]  

3. S. Ye, Q. Fu, and J. Ge, “Invisible photonic prints shown by deformation,” Adv. Funct. Mater. 24(41), 6430–6438 (2014). [CrossRef]  

4. S. Sun, Z. Zhou, C. Zhang, Y. Gao, Z. Duan, S. Xiao, and Q. Song, “All-dielectric full-color printing with TiO2 metasurfaces,” ACS Nano 11(5), 4445–4452 (2017). [CrossRef]  

5. H. Kim, J. Ge, J. Kim, S.-E. Choi, H. Lee, H. Lee, W. Park, Y. Yin, and S. Kwon, “Structural colour printing using a magnetically tunable and lithographically fixable photonic crystal,” Nat. Photonics 3(9), 534–540 (2009). [CrossRef]  

6. H. Fudouzi and T. Sawada, “Photonic rubber sheets with tunable color by elastic deformation,” Langmuir 22(3), 1365–1368 (2006). [CrossRef]  

7. R. Zhang, Q. Wang, and X. Zheng, “Flexible mechanochromic photonic crystals: routes to visual sensors and their mechanical properties,” J. Mater. Chem. C 6(13), 3182–3199 (2018). [CrossRef]  

8. I. Lee, D. Kim, J. Kal, H. Baek, D. Kwak, D. Go, E. Kim, C. Kang, J. Chung, and Y. Jang, “Quasi-amorphous colloidal structures for electrically tunable full-color photonic pixels with angle-independency,” Adv. Mater. 22(44), 4973–4977 (2010). [CrossRef]  

9. D. Franklin, Y. Chen, A. Vazquez-Guardado, S. Modak, J. Boroumand, D. Xu, S.-T. Wu, and D. Chanda, “Polarization-independent actively tunable colour generation on imprinted plasmonic surfaces,” Nat. Commun. 6(1), 7337 (2015). [CrossRef]  

10. S.-U. Kim, S.-H. Lee, I.-H. Lee, B.-Y. Lee, J.-H. Na, and S.-D. Lee, “Generation of intensity-tunable structural color from helical photonic crystals for full color reflective-type display,” Opt. Express 26(10), 13561–13572 (2018). [CrossRef]  

11. L. Bai, Z. Xie, W. Wang, C. Yuan, Y. Zhao, Z. Mu, Q. Zhong, and Z. Gu, “Bio-inspired vapor-responsive colloidal photonic crystal patterns by inkjet printing,” ACS Nano 8(11), 11094–11100 (2014). [CrossRef]  

12. X. Duan, S. Kamin, and N. Liu, “Dynamic plasmonic colour display,” Nat. Commun. 8(1), 14606 (2017). [CrossRef]  

13. Z. Li, A. W. Clark, and J. M. Cooper, “Dual color plasmonic pixels create a polarization controlled nano color palette,” ACS Nano 10(1), 492–498 (2016). [CrossRef]  

14. V. Vashistha, G. Vaidya, R. S. Hegde, A. E. Serebryannikov, N. Bonod, and M. Krawczyk, “All-dielectric metasurfaces based on cross-shaped resonators for color pixels with extended gamut,” ACS Photonics 4(5), 1076–1082 (2017). [CrossRef]  

15. E. Heydari, J. R. Sperling, S. L. Neale, and A. W. Clark, “Plasmonic color filters as dual-state nanopixels for high-density microimage encoding,” Adv. Funct. Mater. 27(35), 1701866 (2017). [CrossRef]  

16. L. Duempelmann, D. Casari, A. Luu-Dinh, B. Gallinet, and L. Novotny, “Color rendering plasmonic aluminum substrates with angular symmetry breaking,” ACS Nano 9(12), 12383–12391 (2015). [CrossRef]  

17. H. Hu, Q.-W. Chen, J. Tang, X.-Y. Hu, and X.-H. Zhou, “Photonic anti-counterfeiting using structural colors derived from magnetic-responsive photonic crystals with double photonic bandgap heterostructures,” J. Mater. Chem. 22(22), 11048–11053 (2012). [CrossRef]  

18. H. Nam, K. Song, D. Ha, and T. Kim, “Inkjet printing based mono-layered photonic crystal patterning for anti-counterfeiting structural colors,” Sci. Rep. 6(1), 30885 (2016). [CrossRef]  

19. Y. Heo, H. Kang, J. S. Lee, Y. K. Oh, and S. H. Kim, “Lithographically encrypted inverse opals for anti-counterfeiting applications,” Small 12(28), 3819–3826 (2016). [CrossRef]  

20. S. J. Tan, L. Zhang, D. Zhu, X. M. Goh, Y. M. Wang, K. Kumar, C.-W. Qiu, and J. K. Yang, “Plasmonic color palettes for photorealistic printing with aluminum nanostructures,” Nano Lett. 14(7), 4023–4029 (2014). [CrossRef]  

21. P. Zijlstra, J. W. Chon, and M. Gu, “Five-dimensional optical recording mediated by surface plasmons in gold nanorods,” Nature 459(7245), 410–413 (2009). [CrossRef]  

22. X. Zhu, C. Vannahme, E. Højlund-Nielsen, N. A. Mortensen, and A. Kristensen, “Plasmonic colour laser printing,” Nat. Nanotechnol. 11(4), 325–329 (2016). [CrossRef]  

23. Y.-W. Huang, W. T. Chen, W.-Y. Tsai, P. C. Wu, C.-M. Wang, G. Sun, and D. P. Tsai, “Aluminum plasmonic multicolor meta-hologram,” Nano Lett. 15(5), 3122–3127 (2015). [CrossRef]  

24. L. De Sio, D. E. Roberts, Z. Liao, S. Nersisyan, O. Uskova, L. Wickboldt, N. Tabiryan, D. M. Steeves, and B. R. Kimball, “Digital polarization holography advancing geometrical phase optics,” Opt. Express 24(16), 18297–18306 (2016). [CrossRef]  

25. W. Wan, J. Gao, and X. Yang, “Full-color plasmonic metasurface holograms,” ACS Nano 10(12), 10671–10680 (2016). [CrossRef]  

26. K. T. Lim, H. Liu, Y. Liu, and J. K. Yang, “Holographic colour prints for enhanced optical security by combined phase and amplitude control,” Nat. Commun. 10(1), 25 (2019). [CrossRef]  

27. A. Kristensen, J. K. Yang, S. I. Bozhevolnyi, S. Link, P. Nordlander, N. J. Halas, and N. A. Mortensen, “Plasmonic colour generation,” Nat. Rev. Mater. 2(1), 16088 (2017). [CrossRef]  

28. Z. Z. Gu, H. Uetsuka, K. Takahashi, R. Nakajima, H. Onishi, A. Fujishima, and O. Sato, “Structural color and the lotus effect,” Angew. Chem., Int. Ed. 42(8), 894–897 (2003). [CrossRef]  

29. I.-H. Lee, S.-H. Lee, C.-M. Keum, S.-U. Kim, and S.-D. Lee, “Combinatorial color arrays based on optical micro-resonators in monolithic architecture,” Opt. Express 22(12), 15320–15327 (2014). [CrossRef]  

30. Z. Yang, Y. Chen, Y. Zhou, Y. Wang, P. Dai, X. Zhu, and H. Duan, “Microscopic interference full-color printing using grayscale-patterned Fabry–Perot resonance cavities,” Adv. Opt. Mater. 5(10), 1700029 (2017). [CrossRef]  

31. Y. Chen, X. Duan, M. Matuschek, Y. Zhou, F. Neubrech, H. Duan, and N. Liu, “Dynamic color displays using stepwise cavity resonators,” Nano Lett. 17(9), 5555–5560 (2017). [CrossRef]  

32. S.-J. Kim, H.-K. Choi, H. Lee, and S.-H. Hong, “Solution-processable nanocrystal-based broadband Fabry–Perot absorber for reflective vivid color generation,” ACS Appl. Mater. Interfaces 11(7), 7280–7287 (2019). [CrossRef]  

33. M. Shih, A. Shishido, and I. Khoo, “All-optical image processing by means of a photosensitive nonlinear liquid-crystal film: edge enhancement and image addition–subtraction,” Opt. Lett. 26(15), 1140–1142 (2001). [CrossRef]  

34. S. R. Nersisyan and N. V. Tabiryan, “Polarization imaging components based on patterned photoalignment,” Mol. Cryst. Liq. Cryst. 489(1), 156–168 (2008). [CrossRef]  

35. X. Zhao, A. Bermak, F. Boussaid, and V. G. Chigrinov, “Liquid-crystal micropolarimeter array for full Stokes polarization imaging in visible spectrum,” Opt. Express 18(17), 17776–17787 (2010). [CrossRef]  

36. S. Yokogawa, S. P. Burgos, and H. A. Atwater, “Plasmonic color filters for CMOS image sensor applications,” Nano Lett. 12(8), 4349–4354 (2012). [CrossRef]  

37. S. D. Rezaei, R. J. Hong Ng, Z. Dong, J. Ho, E. H. Koay, S. Ramakrishna, and J. K. Yang, “Wide-gamut plasmonic color palettes with constant subwavelength resolution,” ACS Nano 13(3), 3580–3588 (2019). [CrossRef]  

38. M. Vaughan, The Fabry-Perot interferometer: history, theory, practice and applications (Routledge, 2017).

39. T. Ellenbogen, K. Seo, and K. B. Crozier, “Chromatic plasmonic polarizers for active visible color filtering and polarimetry,” Nano Lett. 12(2), 1026–1031 (2012). [CrossRef]  

40. Y. Nagasaki, M. Suzuki, and J. Takahara, “All-dielectric dual-color pixel with subwavelength resolution,” Nano Lett. 17(12), 7500–7506 (2017). [CrossRef]  

41. B. Yang, W. Liu, Z. Li, H. Cheng, S. Chen, and J. Tian, “Polarization-sensitive structural colors with hue-and-saturation tuning based on all-dielectric nanopixels,” Adv. Opt. Mater. 6(4), 1701009 (2018). [CrossRef]  

42. L. Huang, X. Chen, H. Mühlenbernd, H. Zhang, S. Chen, B. Bai, Q. Tan, G. Jin, K.-W. Cheah, and C.-W. Qiu, “Three-dimensional optical holography using a plasmonic metasurface,” Nat. Commun. 4(1), 2808 (2013). [CrossRef]  

43. G. Zheng, H. Mühlenbernd, M. Kenney, G. Li, T. Zentgraf, and S. Zhang, “Metasurface holograms reaching 80% efficiency,” Nat. Nanotechnol. 10(4), 308–312 (2015). [CrossRef]  

44. B. Wang, F. Dong, Q.-T. Li, D. Yang, C. Sun, J. Chen, Z. Song, L. Xu, W. Chu, and Y.-F. Xiao, “Visible-frequency dielectric metasurfaces for multiwavelength achromatic and highly dispersive holograms,” Nano Lett. 16(8), 5235–5240 (2016). [CrossRef]  

45. G. Yoon, D. Lee, K. T. Nam, and J. Rho, “Pragmatic metasurface hologram at visible wavelength: the balance between diffraction efficiency and fabrication compatibility,” ACS Photonics 5(5), 1643–1647 (2018). [CrossRef]  

46. S. Slussarenko, O. Francescangeli, F. Simoni, and Y. Reznikov, “High resolution polarization gratings in liquid crystals,” Appl. Phys. Lett. 71(25), 3613–3615 (1997). [CrossRef]  

47. X. H. Li, V. M. Kozenkov, F. S. Y. Yeung, P. Z. Xu, V. G. Chigrinov, and H. S. Kwok, “Liquid-crystal photoalignment by super thin azo dye layer,” Jpn. J. Appl. Phys. 45(1A), 203–205 (2006). [CrossRef]  

48. V. G. Chigrinov, V. M. Kozenkov, and H.-S. Kwok, Photoalignment of liquid crystalline materials: physics and applications (John Wiley & Sons, 2008), Vol. 17.