“Crypto-Display” in Dual-Mode Metasurfaces by Simultaneous Control

www.acsnano.org

“Crypto-Display” in Dual-Mode Metasurfaces by Simultaneous Control of Phase and Spectral Responses Gwanho Yoon,† Dasol Lee,† Ki Tae Nam,‡ and Junsuk Rho*,†,§,∥

Downloaded via EASTERN KENTUCKY UNIV on August 25, 2018 at 09:49:02 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

†

Department of Mechanical Engineering, Pohang University of Science and Technology (POSTECH), Pohang 37673, Republic of Korea ‡ Department of Materials Science and Engineering, Seoul National University, Seoul 08826, Republic of Korea § Department of Chemical Engineering, Pohang University of Science and Technology (POSTECH), Pohang 37673, Republic of Korea ∥ National Institute of Nanomaterials Technology (NINT), Pohang 37673, Republic of Korea S Supporting Information *

ABSTRACT: Although conventional metasurfaces have demonstrated many promising functionalities in light control by tailoring either phase or spectral responses of subwavelength structures, simultaneous control of both responses has not been explored yet. Here, we propose a concept of dual-mode metasurfaces that enables simultaneous control of phase and spectral responses for two kinds of operation modes of transmission and reflection, respectively. In the transmission mode, the dual-mode metasurface acts as conventional metasurfaces by tailoring phase distribution of incident light. In the reflection mode, a reflected colored image is produced under white light illumination. We also experimentally demonstrate a crypto-display as one application of the dual-mode metasurface. The crypto-display looks a normal reflective display under white light illumination but generates a hologram that reveals the encrypted phase information under single-wavelength coherent light illumination. Because two operation modes do not affect each other, the crypto-display can have applications in security techniques. KEYWORDS: holography, structural color, dielectric nanoantennas, broadband, cryptography

M

such as lenses and holograms exhibit trivial reflective images under white light illumination, which is the general lighting environment of our lives, because spectral responses cannot be controlled in such metasurfaces. There is another type of metasurface that can produce nontrivial reflective images by tailoring spectral responses of each unit structure.31−33 However, these metasurfaces cannot control their phase responses, resulting limited applications. Here, we propose a concept of dual-mode metasurfaces that can control phase and spectral response simultaneously. In this article, we use the term of “coherent” to represent spatial coherence of light sources for convenience because temporal coherence is irrelevant to the operation of metasurfaces. Our dual-mode metasurface uses dielectric nanoantennas based on Pancharatnam-Berry (PB) phase which is also called as geometric phase to control spatial phase distribution of single-wavelength coherent light as well as the reflection spectra under white light illumination. Conventional meta-

etasurfaces consisting of ultrathin subwavelength antennas provide an approach to realize flat optical devices. A primary challenge of metasurfaces is to design such antennas to control optical properties of incident optical waves in a desired way. Most effort has been focused on wavefront manipulation owing to its versatile applications such as beam steering devices,1−3 lenses,4−7 holograms,8−14 skin cloaks,15 and other optical components.16,17 The functionality of the metasurface can be further improved by embedding multifunctionality, and this capacity is a distinct advantage of metasurfaces compared to conventional refractive optical elements. For example, meta-holograms can be designed to produce different images as the incident polarization18−21 or wavelength22−25 changes. Nonlinear metasurfaces can generate different images or different states of orbital angular momentum based on second-harmonic generation by metallic nanoantennas.26−28 Furthermore, metasurface functionalities can be superposed by careful integration of several metasurfaces.29,30 Despite many promising functionalities of metasurfaces by controlling either phase or spectral responses of subwavelength structures, simultaneous control of both responses has not been investigated yet. Metasurfaces based on phase modulation © 2018 American Chemical Society

Received: February 19, 2018 Accepted: June 20, 2018 Published: June 20, 2018 6421

DOI: 10.1021/acsnano.8b01344 ACS Nano 2018, 12, 6421−6428

Article

Cite This: ACS Nano 2018, 12, 6421−6428

Article

ACS Nano

“crypto-display” as one application of the dual-mode metasurface (Figure 1). Under white light illumination, the cryptodisplay shows the colored reflective image (reflection mode), whereas under single-wavelength coherent light illumination, it acts as a typical meta-hologram (transmission mode). The orientation of nanostructures does not affect the reflected image, so two modes of the crypto-display are independently controllable; that is, two images do not affect each other. As the holographic image of the crypto-display cannot be inferred from the reflected image, our device can provide a way to develop security technologies such as steganography, anticounterfeiting measure, and ghost imaging applications.34,35 Figure 1. Operation schematic of our crypto-display. The transmission mode uses single-wavelength coherent light to produce the holographic image of “3.141592...” in the image plane, whereas the reflection mode uses white light to represent a reflected colored image of “π”. The transmission mode requires the spatial coherence of the light source for its operation, but the reflection mode does not require any coherence of the light source; i.e., the reflection mode can be activated using incoherent white light such as sunlight.

RESULTS AND DISCUSSION To realize the crypto-display, each unit cell should satisfy two requirements that their CPTs should be the same at λ = 635 nm, and their reflectance spectra should be different as much as possible to produce distinct reflective colors. The CPT is defined as the ratio of power of transmitted cross-polarized light to incident optical power. Because our device is based on PB phase, circularly polarized light is used as incident light. When we shine right circular polarization (RCP) into the metasurface, the outgoing wave can be decomposed into the RCP and left circular polarization (LCP), which is the crosspolarized state of the RCP. The outgoing LCP contains the desired phase information on the metasurface, whereas the transmitted RCP does not. Therefore, CPT is directly related to the hologram efficiency of the metasurface. In order to improve the CPT, a double-nanorod configuration is used as our unit cell design (Figure 2a). The double-nanorod has a

surfaces based on the PB phase control the orientation of each nanoantenna to manipulate phase distribution of crosspolarized light, but we focus on the fact that the spectral response is also controllable by changing the size of nanoantennas. The nanoantenna size also affects crosspolarization transmittance (CPT), so we retrieve a pair of nanoantenna designs that have equal CPT near our target wavelength of 635 nm as well as different reflection spectra. Based on the pair of designs, we experimentally demonstrate a

Figure 2. Optical characteristics of the parallel double-nanorod made of a-Si:H on a glass substrate. (a) Illustration of the unit cell which has two identical nanoantennas. Design A has length L = 300 nm, width W = 100 nm, and gap g = 100 nm; design B has L = 300 nm, W = 50 nm, and g = 100 nm. Designs A and B have the same height h = 300 nm and pitch p = 500 nm. (b) Calculated CPTs of each nanoantenna design for circularly polarized light illumination. Magenta, design A; green, design B. The CPT is higher in the parallel double nanoantennas (solid lines) than in the single nanoantennas (dotted lines) at target λ = 635 nm in the case of the same pitch, and the double nanorod is designed to have the identical CPT near this λ. (c) Calculated reflection spectra of double-nanorod designs. Magenta, design A; green, design B. Based on the spectra, reflective colors can be calculated; i.e., design A shows a magenta color, and design B shows a green color. (d) Comparison of complex refractive indices of amorphous silicon in hydrogenated (solid lines) and unhydrogenated36 (dotted lines) states; red lines, refractive index; blue lines, extinction coefficient. Hydrogen impurities can reduce defect density of amorphous silicon and thereby reduce absorption at visible wavelengths. 6422

DOI: 10.1021/acsnano.8b01344 ACS Nano 2018, 12, 6421−6428

Article

ACS Nano

Figure 3. Characterization of reflection spectra and colors from hologram-encoded metasurfaces composed of the same kind of nanoantennas. (a,b) Comparison of measured reflection spectra (solid lines) and simulated reflection spectra (dotted lines). Magenta, design A; green, design B. (c,d) Optical microscope images of the fabricated metasurfaces composed of either design A (c) or design B (d) nanoantennas using a white LED source. The white point of CCD in the optical microscope is adjusted to match the white point of standard D65 illuminant because the white LED does not belong to the standard illuminant (Figure S1). (Inset) Calculated colors based on the simulated reflection spectra. The D65 illuminant is used for this calculation. The brightness of the observed color varies depending on the brightness of the optical source in the microscope.

initial random phase mask. The paraxial condition is assumed in our algorithm, but our hologram does not satisfy the paraxial condition due to the large viewing angle. So, the original image is adjusted in advance to compensate distortion of the holographic image. At the end, continuous phase distribution retrieved from the algorithm is converted to discrete phase distribution by dividing the phase range from 0 to 2π into eight uniform phase steps. Off-axis configuration is used for the separation of the generated holographic image from the zeroth-order beam (ZOB). If the pixel number of the metasurface is equal to that of the image, the ZOB appears at the center of the generated image and degrades the image fidelity. The intensity of the ZOB decreases as the CPT increases. Unfortunately, the CPT of a-Si:H is not high enough to ignore the ZOB. To realize the off-axis configuration in the Fourier hologram, we cannot avoid the image size reduction as well as resolution loss because the pixels near the center part of the image cannot be used. Hence, all the images used in this work for the hologram are enlarged to 600 × 600 pixels in size by zero padding to remove such obstacle (Figure S3). Reflection spectra of metasurfaces composed of either design A or B nanoantennas are calculated by simplified conditions to explore how the orientation disarrangement of nanoantennas affects the reflected colors. Conventional metasurfaces for color generation are usually composed of equally oriented nanoantennas. This situation is easily described by using periodic boundary conditions in numerical simulations because the computational cost is acceptable. However, the cryptodisplay has an array of randomly oriented nanoantennas in order to generate holographic images. This random orientation cannot be described by the periodic boundary condition, but

structure density higher than that of a typical single nanorod in the case of a same pitch size, resulting in higher CPT by >10% at λ = 635 nm (Figure 2b). To satisfy the aforementioned conditions, we use FDTD simulation to retrieve two kinds of unit cell designs, which are called “design A” and “design B”. Two designs have almost same CPT at λ = 635 nm. In the calculation of phase distribution for the hologram, we assume an outgoing optical wave from the metasurface has a constant amplitude distribution. Because the CPT determines the amplitude of cross-polarized light, their CPTs should be the same to satisfy our assumption. On the other hand, their reflection spectra are different enough to represent different reflective colors (Figure 2c). We calculate the reflective colors of each design by using a D65 standard illuminant spectrum (Figure S1). Design A shows a magenta color, and design B exhibits a green color. Hydrogenated amorphous silicon (aSi:H) is used as structuring material of our metasurface (Figure 2d). Although normal amorphous silicon has a high refractive index at visible wavelengths, its diffraction efficiency is very low due to severe optical loss in the visible region.36,37 In the case of a-Si:H, hydrogen impurities can reduce defect density of normal amorphous silicon and thereby improve its optical properties for visible metasurfaces, that is, reduce its low optical loss at visible wavelengths.38 Computer-generated phase-only Fourier holography is used for the holographic functionality of the crypto-display. Fourier holography enables observation of the generated image without optical components such as lenses. A 300 × 100 pixelated image of “3.141592...” is encoded into the cryptodisplay using the Gerchberg-Saxton algorithm (Figure S2), which is an iterative method for phase-only holograms.39 Two hundred iterative processes are conducted, combined with an 6423

DOI: 10.1021/acsnano.8b01344 ACS Nano 2018, 12, 6421−6428

Article

ACS Nano

Figure 4. Experimental demonstration of the crypto-display. Each figure set contains (i) an optical microscope image with the white point correction (scale bar = 100 μm), generated holographic images under (ii) 635 nm and (iii) 532 nm laser illumination, and (iv) a SEM image (scale bar = 1 μm, with full-version SEM images in Figure S12). The holographic images of 532 nm wavelength have barrel distortion because our precompensation process of the original image only corresponds to the target λ = 635 nm. (a,b) Metasurfaces composed of either design A or B nanoantennas, respectively. They generate holographic images under single-wavelength coherent light illumination but no reflection image under white light illumination. (c) Demonstrated crypto-display consists of both design A and B nanoantennas. Under coherent light illumination, the desired holographic image is generated as conventional meta-holograms. Furthermore, the crypto-display provides the reflected colored image of “π” for the illumination of white light. (d) Another crypto-display encoded by a different holographic image of an “apple pie”. Modification of the encoded image reorganizes the orientation of nanoantennas, but the reflection image remains the same. Hence, the transmission and reflection modes are independently controllable in the crypto-display.

numerical simulations but at enormous computational cost. Our simulation does not consider this aspect. Despite the difference of the reflection spectra, observed colors reasonably correspond to our simplified simulation (Figure 3c,d). Both of the metasurfaces are composed of randomly oriented nanoantennas due to hologram encoding, but the effect of the orientation disarrangement on the reflective color is negligible. Each image is captured by a charge-coupled device (CCD) of a conventional optical microscope with unpolarized white light from a light-emitting diode (LED). Although our LED source belongs to the white LED, its color looks yellowish due to its strong peak near λ = 550 nm. If we see the metasurfaces directly through the eyepiece of the microscope, all colors become yellowish; nevertheless, observed colors agree well with our simulation (Figure S1). However, white LED has not been standardized yet. It is not appropriate to directly compare the colors captured under our LED illumination. Hence, we adjust a white point of the CCD to match with the D65 illuminant, which is one of the standard light sources to show what our metasurfaces look like under general lighting environments. Rigorously, this method is not perfect, but it is widely used for image conversion from an image captured under one illuminant to the same image captured under another illuminant. As a result, the metasurface composed of design A nanoantennas represents a magenta color, and the metasurface consisting of design B nanoantennas shows a green color. We experimentally demonstrate the crypto-display composed of both design A and B nanoantennas. Conventional metasurfaces are also fabricated as a control group. The same holographic image “3.141592...” is encoded on the three metasurfaces (Figure 4a−c), and they produce the same

we obtain our simulation results from a periodic array of nanoantennas with identical orientations as usual (Figure S4). The simulation condition is much simplified compared to the real situation because the real one needs tremendous computational costs; that is, the real situation is not perfectly described in our simulation. Therefore, we should compare the simulated reflective colors with experimentally measured colors. Reflection spectra of each metasurface are experimentally obtained using conventional Fourier transform infrared spectroscopy (FT-IR) (Figure 3a,b). The difference between simulation and measurement is mainly caused by structure size variation as well as irregular interaction among unit cells. Process conditions change during fabrication steps such as electron beam lithography, resist development, and material etching. Hence, dimensions of nanoantennas are not consistent even in the same metasurface. As a result, the final dimensions of the metasurface differ slightly from the original design rather than being of equal dimensions. Because the reflection spectrum is strongly affected by the structure dimensions, the measured spectra differ from the simulation results. For example, the reflection spectrum mismatch of design B can be explained by additional simulation results in which the peak of the calculated reflection spectrum near λ = 550 nm moves to λ = 540 nm as the width of the structure changes from 50 to 40 nm. The irregular interaction among nanoantennas occurs because the pitch of the unit cell is not long enough to prevent electromagnetic interaction among neighboring unit cells. Moreover, in the crypto-display, distance between each unit cell also differs depending on the position due to the random orientation of nanoantennas. This difference provokes different electric field distributions compared with our simulation. Ideally, this irregular interaction can be fully included in the 6424

DOI: 10.1021/acsnano.8b01344 ACS Nano 2018, 12, 6421−6428

Article

ACS Nano

Hologram efficiency of each metasurface is measured to investigate the integration effect of different nanoantennas (Figure 5). The hologram efficiency is defined by the ratio of

holographic image when we shine single-wavelength coherent light on the metasurfaces. The control group shows typical results: under single-wavelength coherent light, an encoded holographic image is generated on the image plane, but under white light illumination, no image is obtained (Figure 4ai,bi). In the case of the crypto-display, we divide the metasurface into two parts (i.e., the area of the character “π” and the background). Design A nanoantennas are arranged on the character region, and design B nanoantennas are arranged on the background. Therefore, the crypto-display produces not only a holographic image under single-wavelength coherent light (transmission mode) but also a reflected colored image under white light illumination (reflection mode) (Figure 4ci). The reflection mode of the crypto-display is based on the spectrum control, which is irrelevant to the source coherence. Therefore, theoretically, it is possible to observe the same reflected image using incoherent white light such as sunlight. Propagation phase that does not need to be considered in the control group should be considered in the crypto-display because it consists of two kinds of unit cells. When the incident light is RCP, the transmitted light from the dielectric nanorod can be described by the Jones vector as follows.40 Ä É ÄÅ ÉÑ ÅÅ1ÑÑ TL + TS ÅÅÅ 1 ÑÑÑ TL − TS ÅÅ ÑÑ + ÅÅ ÑÑ exp( 2 ) i α Å Ñ ÅÅ i ÑÑ 2 ÅÅÇ−i ÑÑÖ 2 ÅÇ ÑÖ

Figure 5. Measured hologram efficiency of the fabricated metasurfaces. The hologram efficiency is defined by the ratio of the optical power of the holographic image to the incident optical power. The hologram efficiency is strongly related to the CPT of nanoantennas, so design A and B metasurfaces have a similar hologram efficiency at λ = 635 nm. The hologram efficiency of the crypto-display is always intermediate between the values of design A and B metasurfaces because the crypto-display includes both nanoantennas.

TL and TS denote complex transmission coefficients of the nanorod when the incident light is linearly polarized along the long and short axis of the nanorod, and α is the rotation angle of the nanorod. The first term is outgoing RCP, which does not contain the desired phase distribution, whereas the second term represents the outgoing LCP containing the desired phase distribution originated from the geometric phase. Phase delay of the LCP is determined by the propagation phase and geometric phase. The propagation phase is determined by the angle of the complex number (TL − TS)/2, which is frequencydependent, and the geometric phase is determined by exp(i2α), which is frequency-independent. If same-sized nanorods are used for the metasurface, the propagation term is constant over the metasurface. We do not need to consider the propagation term in this case. The phase distribution of outgoing LCP is determined by the geometric phase only. On the other hand, the crypto-display uses different-sized nanorods (i.e., designs A and B). Each design has different propagation phases, so we compensate the difference by adjusting the geometric phase (Figures S5 and S6). As the propagation phase is frequency-dependent, the propagation phase difference at λ = 532 nm is not the same as that of λ = 635 nm. Therefore, the phase error at λ = 532 nm cannot be compensated together, resulting in the noise in the holographic image (Figure 4ciii). Another crypto-display which is encoded by the different holographic image is demonstrated to prove that the transmission and reflection modes are independently controllable (Figure 4d). We encode another image of “apple pie” which is totally different from the previous holographic image of “3.141592...” (Figure S3). The orientations of nanoantennas also differ from the previous crypto-display because the required phase distribution changes; nevertheless, the reflected colored image “π” remains the same. The holographic image cannot be inferred from the reflected image, and it enables us to realize a security device based on our crypto-display by hiding secret information in the form of a hologram.

optical power of the holographic image to the incident optical power. The hologram efficiency and the CPT are positively correlated because the transmitted cross-polarized light forms the holographic image in PB phase metasurfaces. The measured results reasonably agree with the tendency of the calculated CPT. Two metasurfaces that consist of either design A or B nanoantennas have almost the same hologram efficiency at λ = 635 nm, whereas they are different at λ = 532 nm. The hologram efficiency of the crypto-display is always intermediate between those metasurfaces because the crypto-display is composed of both types of nanoantennas. The incident light on the crypto-display is not perfectly transmitted or reflected. Some incident light is reflected or transmitted, so it may disturb the other operation modes of our device, resulting in the crosstalk. However, each operation mode of our crypto-display can be fully separated. Our cryptodisplay does not aim simultaneous operation of transmission and reflection mode (Figure S7). The transmission mode is activated under single-wavelength coherent light illumination, whereas the reflection mode uses incoherent white light. Under the incoherent light illumination, the reflection mode can be activated, but the transmission mode cannot be activated in this case because the desired phase distribution cannot be developed from the random phase distribution of the incoherent light. On the other hand, both operation modes can be activated simultaneously under the coherent light illumination. However, two operation modes can also be fully separated because image locations differ from each operation mode. In transmission mode, the holographic image of “3.141592...” produced by the phase modulation is generated on the image plane. The image of “π” produced by the amplitude modulation is also generated on the image plane, but it is concealed in the zeroth-order beam. It does not spread over the image plane, so two images do not overlap each other (Figure 4cii,iii). In the reflection mode, the image of “π” is produced on the metasurface itself, but the holographic image 6425

DOI: 10.1021/acsnano.8b01344 ACS Nano 2018, 12, 6421−6428

ACS Nano

l o AE0B(E − EG)2 o o o when E > EG o 2 2 2 2 2 ε″(E) = o m E[(E − E0 ) + B E ] o o o o 0 when E ≤ EG o n

of “3.141592...” is not produced on the metasurface. So, when we observe the reflective image of our crypto-display using single-wavelength coherent light, we see the image of “π” only (Figure S8). The holographic image of “3.141592...” does not appear in the reflected image. In short, two operation modes of our crypto-display do not crosstalk with each other. Our crypto-display uses two kinds of nanoantennas that have distinct reflective colors under white light illumination, but further improvement on the reflective color is possible. If we use additional nanoantenna designs that have dimensions between designs A and B, colors that change gradually from magenta (design A) to green (design B) can be expressed. Another possibility is a change of structuring material. We use a-Si:H which has severe intrinsic absorption near blue color, so the range of colors that we can represent is limited. On the other hand, other materials such as crystalline silicon have absorption much lower than that of a-Si:H in short visible wavelengths, enabling more versatile color presentation. Therefore, if we have three kinds of nanoantenna designs showing three primary colors, appropriate combination of them will allow diverse reflective colors to be expressed.

Article

where A is the strength of the absorption peak, B is the broadening term, E0 is the energy position of the absorption peak, and EG is the optical band gap energy. The fitting parameters for our model were A = 104.4084 eV, B = 1.387 eV, E0 = 4.216 eV, and EG = 1.651 eV (Figure S9). The mean square error between the fitting curve and the measured data was 56.000. Numerical Simulation. All numerical simulations were performed using Lumerical FDTD software. Periodic boundary conditions and perfectly matched layers were used as boundary conditions in the calculation of CPT and reflection spectra of the nanoantennas (Figure S4). The incident polarization angle was 45° by the long axis of the nanorod for the reflectance calculation. The refractive index of SiO2 substrate was 1.5. The measured refractive index of our a-Si:H using the ellipsometer was used in the numerical simulations. Optical Measurement. Reflection spectra of metasurfaces were measured by FT-IR, and reflection images were captured by a CCD of a conventional optical microscope with a white LED. Because our LED source spectrum has a strong peak near λ = 550 nm, its actual color looks yellowish compared to a typical white color (Figure S1). The holographic images were generated using a customized optical setup (Figure S11). The holographic images were formed on an image plane ∼10 cm away from the metasurface and captured using the camera of a conventional mobile phone.

CONCLUSIONS In summary, we propose the crypto-display which can work as not only conventional meta-holograms under single-wavelength coherent light illumination but also reflective displays under white light illumination. Our crypto-display is composed of two kinds of unit cells consisting of double dielectric nanoantennas. Each unit cell has the same CPT with different reflection spectra. The double nanorod has the advantage of a CPT higher than that of the single nanorod. The orientations of nanoantennas determine the phase distribution of singlewavelength coherent light, and their sizes determine the reflection spectrum under white light illumination. We experimentally demonstrate the crypto-display and verify that both operation modes can be independently controllable. It looks like a normal reflective display under white light illumination but reveals encrypted information in the form of a hologram under coherent light illumination. The encoded hologram cannot be inferred from the reflected image, so our approach can be applied to develop security devices.

ASSOCIATED CONTENT S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsnano.8b01344. Reflective color calculation, simulated holographic images, off-axis configuration in Fourier holography, boundary conditions in reflection spectrum simulation, propagation phase calculation, operation modes of the crypto-display, coherent imaging of the crypto-display, ellipsometry data of a-Si:H, optical setup for the hologram generation, full version of SEM images (PDF)

AUTHOR INFORMATION Corresponding Author

METHODS

*E-mail: [email protected].

Sample Fabrication. A 300 nm thick a-Si:H thin film was deposited on a fused silica substrate using plasma-enhanced chemical vapor deposition (HiDep-SC, BMR Technology). The deposition rate of a-Si:H was 1.3 nm/s under the temperature of 300 °C with 10 sccm flow rate of SiH4 gas and 75 sccm flow rate of H2 gas. Then electron beam lithography (ELS-7800, Elionix) was used to define a chromium (Cr) mask on the a-Si:H, and it was etched along the Cr mask by using inductively coupled plasma reactive ion etching. An 80 sccm flow rate of Cl2 gas and a 120 sccm flow rate of HBr gas were used in etching process under the pressure of 5 mTorr, resulting in an etching rate of 4 nm/s. The remaining Cr mask was removed using Cr etchant. Thin Film Characterization. The a-Si:H thin film was characterized using an ellipsometer (M-2000, J.A. Woollam). For the measurement, we used a Si substrate with ∼100 nm SiO2 layer on the top to enhance measurement accuracy. The incident beam angle was 65°. The Tauc-Lorentz model was employed because this model generally works well with the absorptive amorphous material. The imaginary part of the dielectric function of the Tauc-Lorentz model is given by

ORCID

Ki Tae Nam: 0000-0001-6353-8877 Junsuk Rho: 0000-0002-2179-2890 Author Contributions

G.Y. and J.R. conceived the idea and initiated the project. G.Y. designed and fabricated the devices. G.Y. and D.L. did optical experiments. K.T.N. provided materials characterization. G.Y. and J.R. wrote the manuscript. All authors confirmed the final manuscript. J.R. guided the entire project. Notes

The authors declare no competing financial interest.

ACKNOWLEDGMENTS This work is financially supported from the LGD-SNU Incubation program funded by LG Display and the National Research Foundation grants (NRF-2017R1E1A1A03070501, NRF-2015R1A5A1037668, and CAMM6426

DOI: 10.1021/acsnano.8b01344 ACS Nano 2018, 12, 6421−6428

Article

ACS Nano

Subwavelength Spatial Resolution and High Transmission. Nat. Nanotechnol. 2015, 10, 937−943. (20) Balthasar Mueller, J. P.; Rubin, N. A.; Devlin, R. C.; Groever, B.; Capasso, F. Metasurface Polarization Optics: Independent Phase Control of Arbitrary Orthogonal States of Polarization. Phys. Rev. Lett. 2017, 118, 113901. (21) Xie, Z.; Lei, T.; Si, G.; Wang, X.; Lin, J.; Min, C.; Yuan, X. Meta-Holograms with Full Parameter Control of Wavefront over a 1000 nm Bandwidth. ACS Photonics 2017, 4, 2158−2164. (22) Huang, Y. W.; Chen, W. T.; Tsai, W. Y.; Wu, P. C.; Wang, C. M.; Sun, G.; Tsai, D. P. Aluminum Plasmonic Multicolor MetaHologram. Nano Lett. 2015, 15, 3122−3127. (23) Wang, B.; Dong, F.; Li, Q. T.; Yang, D.; Sun, C.; Chen, J.; Song, Z.; Xu, L.; Chu, W.; Xiao, Y. F.; Gong, Q.; Li, Y. Visible-Frequency Dielectric Metasurfaces for Multiwavelength Achromatic and Highly Dispersive Holograms. Nano Lett. 2016, 16, 5235−5240. (24) Wan, W.; Gao, J.; Yang, X. Full-Color Plasmonic Metasurface Holograms. ACS Nano 2016, 10, 10671−10680. (25) Li, X.; Chen, L.; Li, Y.; Zhang, X.; Pu, M.; Zhao, Z.; Ma, X.; Wang, Y.; Hong, M.; Luo, X. Multicolor 3D Meta-Holography by Broadband Plasmonic Modulation. Sci. Adv. 2016, 2, e1601102. (26) Ye, W.; Zeuner, F.; Li, X.; Reineke, B.; He, S.; Qiu, C. W.; Liu, J.; Wang, Y.; Zhang, S.; Zentgraf, T. Spin and Wavelength Multiplexed Nonlinear Metasurface Holography. Nat. Commun. 2016, 7, 11930. (27) Walter, F.; Li, G.; Meier, C.; Zhang, S.; Zentgraf, T. Ultrathin Nonlinear Metasurface for Optical Image Encoding. Nano Lett. 2017, 17, 3171−3175. (28) Li, G.; Wu, L.; Li, K. F.; Chen, S.; Schlickriede, C.; Xu, Z.; Huang, S.; Li, W.; Liu, Y.; Pun, E. Y. B.; Zentgraf, T.; Cheah, K. W.; Luo, Y.; Zhang, S. Nonlinear Metasurface for Simultaneous Control of Spin and Orbital Angular Momentum in Second Harmonic Generation. Nano Lett. 2017, 17, 7974−7979. (29) Lin, D.; Holsteen, A. L.; Maguid, E.; Wetzstein, G.; Kik, P. G.; Hasman, E.; Brongersma, M. L. Photonic Multitasking Interleaved Si Nanoantenna Phased Array. Nano Lett. 2016, 16, 7671−7676. (30) Maguid, E.; Yulevich, I.; Yannai, M.; Kleiner, V.; L Brongersma, M.; Hasman, E. Multifunctional Interleaved Geometric-Phase Dielectric Metasurfaces. Light: Sci. Appl. 2017, 6, e17027. (31) Kumar, K.; Duan, H.; Hegde, R. S.; Koh, S. C.; Wei, J. N.; Yang, J. K. Printing Colour at the Optical Diffraction Limit. Nat. Nanotechnol. 2012, 7, 557−561. (32) Olson, J.; Manjavacas, A.; Liu, L.; Chang, W. S.; Foerster, B.; King, N. S.; Knight, M. W.; Nordlander, P.; Halas, N. J.; Link, S. Vivid, Full-Color Aluminum Plasmonic Pixels. Proc. Natl. Acad. Sci. U. S. A. 2014, 111, 14348−14353. (33) Sun, S.; Zhou, Z.; Zhang, C.; Gao, Y.; Duan, Z.; Xiao, S.; Song, Q. All-Dielectric Full-Color Printing with TiO2 Metasurfaces. ACS Nano 2017, 11, 4445−4452. (34) Liu, H.-C.; Yang, B.; Guo, Q.; Shi, J.; Guan, C.; Zheng, G.; Mühlenbernd, H.; Li, G.; Zentgraf, T.; Zhang, S. Single-Pixel Computational Ghost Imaging with Helicity-Dependent Metasurface Hologram. Sci. Adv. 2017, 3, e1701477. (35) Duan, X.; Kamin, S.; Liu, N. Dynamic Plasmonic Colour Display. Nat. Commun. 2017, 8, 14606. (36) Pierce, D. T.; Spicer, W. E. Electronic Structure of Amorphous Si from Photoemission and Optical Studies. Phys. Rev. B 1972, 5, 3017−3029. (37) Huang, K.; Dong, Z.; Mei, S.; Zhang, L.; Liu, Y.; Liu, H.; Zhu, H.; Teng, J.; Luk’yanchuk, B.; Yang, J. K. W.; Qiu, C.-W. Silicon Multi-Meta-Holograms for the Broadband Visible Light. Laser Photonics Rev. 2016, 10, 500−509. (38) Park, C.-S.; Shrestha, V. R.; Yue, W.; Gao, S.; Lee, S.-S.; Kim, E.-S.; Choi, D.-Y. Structural Color Filters Enabled by a Dielectric Metasurface Incorporating Hydrogenated Amorphous Silicon Nanodisks. Sci. Rep. 2017, 7, 2556. (39) Gerchberg, R. W.; Saxton, W. O. A Practical Algorithm for the Determination of Phase from Image and Diffraction Plane Pictures. Optik 1972, 35, 237−246.

2014M3A6B3063708) funded by the Ministry of Science, ICT and Future Planning (MSIP) of the Korean government.

REFERENCES (1) Yu, N.; Genevet, P.; Kats, M. A.; Aieta, F.; Tetienne, J.-P.; Capasso, F.; Gaburro, Z. Light Propagation with Phase Discontinuities: Generalized Laws of Reflection and Refraction. Science 2011, 334, 333−337. (2) Huang, L.; Chen, X.; Muhlenbernd, H.; Li, G.; Bai, B.; Tan, Q.; Jin, G.; Zentgraf, T.; Zhang, S. Dispersionless Phase Discontinuities for Controlling Light Propagation. Nano Lett. 2012, 12, 5750−5755. (3) Aieta, F.; Kats, M. A.; Genevet, P.; Capasso, F. Multiwavelength Achromatic Metasurfaces by Dispersive Phase Compensation. Science 2015, 347, 1342−1345. (4) Aieta, F.; Genevet, P.; Kats, M. A.; Yu, N.; Blanchard, R.; Gaburro, Z.; Capasso, F. Aberration-Free Ultrathin Flat Lenses and Axicons at Telecom Wavelengths Based on Plasmonic Metasurfaces. Nano Lett. 2012, 12, 4932−4936. (5) Lin, D.; Fan, P.; Hasman, E.; Brongersma, M. L. Dielectric Gradient Metasurface Optical Elements. Science 2014, 345, 298−302. (6) Arbabi, A.; Arbabi, E.; Kamali, S. M.; Horie, Y.; Han, S.; Faraon, A. Miniature Optical Planar Camera Based on a Wide-Angle Metasurface Doublet Corrected for Monochromatic Aberrations. Nat. Commun. 2016, 7, 13682. (7) Khorasaninejad, M.; Chen, W. T.; Devlin, R. C.; Oh, J.; Zhu, A. Y.; Capasso, F. Metalenses at Visible Wavelengths: DiffractionLimited Focusing and Subwavelength Resolution Imaging. Science 2016, 352, 1190−1194. (8) Walther, B.; Helgert, C.; Rockstuhl, C.; Setzpfandt, F.; Eilenberger, F.; Kley, E.-B.; Lederer, F.; Tünnermann, A.; Pertsch, T. Spatial and Spectral Light Shaping with Metamaterials. Adv. Mater. 2012, 24, 6300−6304. (9) Huang, L.; Chen, X.; Mühlenbernd, H.; Zhang, H.; Chen, S.; Bai, B.; Tan, Q.; Jin, G.; Cheah, K.-W.; Qiu, C.-W.; Li, J.; Zentgraf, T.; Zhang, S. Three-Dimensional Optical Holography Using a Plasmonic Metasurface. Nat. Commun. 2013, 4, 2808. (10) Ni, X.; Kildishev, A. V.; Shalaev, V. M. Metasurface Holograms for Visible Light. Nat. Commun. 2013, 4, 2807. (11) Huang, L.; Muhlenbernd, H.; Li, X.; Song, X.; Bai, B.; Wang, Y.; Zentgraf, T. Broadband Hybrid Holographic Multiplexing with Geometric Metasurfaces. Adv. Mater. 2015, 27, 6444−6449. (12) Zheng, G.; Muhlenbernd, H.; Kenney, M.; Li, G.; Zentgraf, T.; Zhang, S. Metasurface Holograms Reaching 80% Efficiency. Nat. Nanotechnol. 2015, 10, 308−312. (13) Li, Z.; Kim, I.; Zhang, L.; Mehmood, M. Q.; Anwar, M. S.; Saleem, M.; Lee, D.; Nam, K. T.; Zhang, S.; Luk’yanchuk, B.; Wang, Y.; Zheng, G.; Rho, J.; Qiu, C.-W. Dielectric Meta-Holograms Enabled with Dual Magnetic Resonances in Visible Light. ACS Nano 2017, 11, 9382−9389. (14) Yoon, G.; Lee, D.; Nam, K. T.; Rho, J. Pragmatic Metasurface Hologram at Visible Wavelength: the Balance between Diffraction Efficiency and Fabrication Compatibility. ACS Photonics 2018, 5, 1643−1647. (15) Ni, X.; Wong, Z. J.; Mrejen, M.; Wang, Y.; Zhang, X. An Ultrathin Invisibility Skin Cloak for Visible Light. Science 2015, 349, 1310−1314. (16) Khorasaninejad, M.; Crozier, K. B. Silicon Nanofin Grating as a Miniature Chirality-Distinguishing Beam-Splitter. Nat. Commun. 2014, 5, 5386. (17) Liu, Z.; Li, Z.; Liu, Z.; Cheng, H.; Liu, W.; Tang, C.; Gu, C.; Li, J.; Chen, H.-T.; Chen, S.; Tian, J. Single-Layer Plasmonic Metasurface Half-Wave Plates with Wavelength-Independent Polarization Conversion Angle. ACS Photonics 2017, 4, 2061−2069. (18) Montelongo, Y.; Tenorio-Pearl, J. O.; Williams, C.; Zhang, S.; Milne, W. I.; Wilkinson, T. D. Plasmonic Nanoparticle Scattering for Color Holograms. Proc. Natl. Acad. Sci. U. S. A. 2014, 111, 12679− 12683. (19) Arbabi, A.; Horie, Y.; Bagheri, M.; Faraon, A. Dielectric Metasurfaces for Complete Control of Phase and Polarization with 6427

DOI: 10.1021/acsnano.8b01344 ACS Nano 2018, 12, 6421−6428

Article

ACS Nano (40) Chen, W. T.; Zhu, A. Y.; Sanjeev, V.; Khorasaninejad, M.; Shi, Z.; Lee, E.; Capasso, F. A Broadband Achromatic Metalens for Focusing and Imaging in the Visible. Nat. Nanotechnol. 2018, 13, 220−226.

6428

DOI: 10.1021/acsnano.8b01344 ACS Nano 2018, 12, 6421−6428