Dynamically Reconfigurable Metadevice Employing Nanostructured

Jul 21, 2017 - ... will open new opportunities for signal processing, memory, security, ..... camera images of the final arrays at 30 and 82 °C. The ...
0 downloads 0 Views 2MB Size
Subscriber access provided by UNIV OF NEWCASTLE

Communication

Dynamically reconfigurable metadevice employing nanostructured phase change materials Zhihua Zhu, Philip Gareth Evans, Richard F. Haglund, and Jason Valentine Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.7b01767 • Publication Date (Web): 21 Jul 2017 Downloaded from http://pubs.acs.org on July 23, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Nano Letters is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Dynamically reconfigurable metadevice employing nanostructured phase change materials Zhihua Zhu †, Philip G. Evans ‡, Richard F. Haglund Jr. ||, Jason G. Valentine*⊥ †

Department of Electrical Engineering and Computer Science, Vanderbilt University, Nashville, Tennessee 37235, USA ‡ Computational Science and Engineering Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, USA || Department of Physics and Astronomy, Vanderbilt University, Nashville, Tennessee 37235, USA ⊥Department of Mechanical Engineering, Vanderbilt University, Nashville, Tennessee 37235, USA

Abstract: Mastering dynamic free-space spectral control and modulation in the near-infrared and optical regimes remains a challenging task that is hindered by the available functional materials at high frequencies. Here, we realize an efficient metadevice capable of spectral control by minimizing the thermal mass of a vanadium dioxide phase-change material (PCM) and placing the PCM at the feed gap of a bow-tie field antenna. The device has an experimentally measured tuning range up to 360 nm in the NIR and a modulation depth of 33% at the resonant wavelength. The metadevice is configured for integrated and local heating leading to faster switching and more precise spatial control compared to devices based on phase change thin films. We envisage that the combined advantages of this device will open new opportunities for signal processing, memory, security, and holography at optical frequencies.

Keywords: metadevice, metamaterials, perfect absorber, vanadium dioxide, plasmon, nanoantenna

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 21

Metadevices – artificial electromagnetic media comprising plasmonic or dielectric nanostructures and active functional materials1 – have attracted interest for a wide range of applications including sensing2, memory or data storage3, 4, communications5-7 and imaging8, 9. A number of techniques have been proposed for achieving dynamic control of metadevices at near-infrared (NIR) and visible frequencies, including carrier injection5-7,10-14, mechanical actuation15,16, liquid crystals17,18, chemical activation2, and the deployment of phase-change materials19-27. However, these techniques have roadblocks that must be addressed in order to move towards commercialization of specific metadevice technologies. For instance, while carrier injection can be quite fast it is difficult to achieve optical frequency modulation due to the large carrier concentrations and high voltages required10-14. Even with sufficient voltage, the changes in the effective optical constants are small which can limit modulation depth. Thermal and electromechanical metadevices generally work at slower speeds and while modulation depth can in theory be large, to date reversible depths on the order of 10% have been achevied16 at optical frequencies. Phase-change materials such as chalcogenide glass (GST) (non-volatile) and vanadium dioxide (volatile) have been used for decades in memory storage, and dynamic optical elements such as tunable absorbers and polarizers3-4,

19-27

. Upon undergoing a

phase change, these materials exhibit a large change in their optical properties providing a means for modulation. Vanadium dioxide is particularly interesting as it has a reversible phase transition at low temperatures (~340K) which is critical for realizing low power devices. As a comparison, the transition threshold for GST is higher and repeated cycling is limited by degradation associated with repetitive amorphous-crystalline phase changes.

ACS Paragon Plus Environment

Page 3 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

The majority of previously demonstrated PCM-based metadevices have employed PCM films that can be readily incorporated into structured metal resonators19-21, 23-26. However, when switched thermally the large thermal mass of continuous films necessitates large switching power per bit while also requiring an external heating element. At the same time, this limits switching speed as this thermal energy must then be dissipated to recover the device. Here, we present a metadevice that integrates VO2 nanocrystals with plasmonic metamaterials. By designing the unit cells so that each nanocrystal is in the feed gap of a bow-tie antenna, we achieve strong field concentration within the VO2 nanocrystals. This allows the metadevice to be more sensitive to the optical properties of the VO2 while also utilizing a small thermal mass for reduced switching power and time. In addition, the metallic unit cells serve as the heating element minimizing the volume that must be heated and improving integration. This results in an integrated device with a modulation depth of 33% within the telecommunications band and a 1.27 ms recovery time. The metadevice is based on the perfect absorber (MPA) architecture with each unit cell consisting of a gold bow-tie antenna with a small VO2 patch placed in its feed gap. This layer is separated from a thick gold backplane by a thin Al2O3 dielectric spacer layer (Figure 1a, b). Fabrication began with thermal deposition of 100 nm of gold (Au) on fused silica to form the backplane (see the material selection in S1). This was followed by atomic-layer deposition (ALD) of 37 nm of Al2O3 to form a transparent dielectric spacer layer. The VO2 patterns (120 x 120 x 37 nm3) were then defined on the Al2O3 film by electron beam lithography. After development, the VO2 was sputtered using a vanadium metal source followed by a standard lift-off procedure and then annealing for 5

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

min to crystallize the VO2 (see detail in Methods). Figure 2a shows the atomic-forcemicroscope (AFM) topographical image of the patterned VO2 nanocrystal array. The index and absorption coefficient (figure 2b) were measured using spectroscopic ellipsometry and show that the VO2 film has excellent contrast between the optical properties of the semiconducting and metallic states. Gold bow-tie antennas with 34 nm wide feed-gaps were then overlaid onto the VO2 nanocrystals using electron beam lithography, thermal deposition and lift-off (figure 2c, d). The smallest possible gap size is preferred as it increases the capacitance in the semiconducting state, red shifting the resonance and resulting in a larger wavelength change upon switching. More details can be found in Section 2 of the Supplementary Information. These antennas were connected by bus bars which were connected to large external gold electrodes so that the antennas can serve as integrated heating elements. In order to characterize the modulation depth of the metadevice, it was first mounted on a temperature-controlled stage. A white-light supercontinuum source was incident onto the metadevice using a 50x objective, with the polarization perpendicular to the bus wires. The MPA absorption curve is shown in Figure 3a. The measured resonant position shifts from 1,590 nm to 1,230 nm as the stage temperature increases from 21 ℃ to 87 ℃. This results in a figure of merit of FOM = ∆λ/λFWHM = 82%. The reversible modulation depth (defined as ℎ = |  −  | ) was experimentally measured to be 27% and 33% at 1,232 nm and 1,588 nm, respectively. These are the highest values measured in free space within this wavelength regime, to the best of our knowledge. It should also be noted that compared to VO2-film based metadevices, we observe a change in the switching temperature range resulting from the small size of the

ACS Paragon Plus Environment

Page 4 of 21

Page 5 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

VO2 nanocrystals28 (Figure 3d). The modulation starts earlier at about 40.3 ℃ and finishes at 86 ℃, which provides greater dynamic range for grey-scale control. From a materials perspective, VO2 has a lower switching threshold than PCMs like GST due to the near-room-temperature phase transition. In order to electrically switch the device, a current is applied to the antenna layer through the bus bars, heating the metal. Compared to external heaters, this technique only generates heat where it is required, reducing power consumption. The measured optical properties, as a function of current injection, are shown in Figure 3b. Local Joule heating of the device results in the same modulation depth as using the temperature controlled stage. The threshold current to switch the entire 24 µm x 24 µm sample was found to be 56 mA (2.2 V), yielding a power consumption of 123.2 mW, which agrees well with the calculated switching threshold (see S4 for the theoretical model). The performance of the metadevice agrees well with full-wave finite-difference time-domain (FDTD) simulations shown in Figure 3c. In these simulations the VO2 properties were extracted from the experimentally measured optical properties shown in Figure 2b. Based on the simulated absorption spectrum, when the VO2- based metadevice is in the semiconducting state (25 °C), the device exhibits almost unity absorption (99.7%) at a resonant frequency of 1,590 nm. At an intermediate state of 65 °C the resonance shifts to approximately 1,380 nm with a peak absorption of 73%. The resonance further shifts to 1,190 nm as the VO2 becomes metallic (85 °C) with a peak absorption of 90%. The simulated structure has a slightly larger tuning range (400 nm) and modulation depth (48% and 51% at 1,590 nm and 1,190 nm, respectively). The deviation between the simulated and experimental modulation depth can be attributed to several factors. The

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ellipsometry data used in the FDTD simulations were acquired from a 60 nm VO2 continuous film on a silicon substrate, while for the metadevice, VO2 nanoparticles were patterned and grown on Al2O3 and have a height of 37 nm. These films were deposited at the same time but shadowing by PMMA reduces the height of the nanoparticles. The different substrate and thickness can both affect the grain size and grain boundary29 leading to deviations with the simulations. The VO2 phase transition also depends on the nanocrystal size and it has been shown that nanocrystals result in a broader hysteresis loop and phase transition compared to thin films28. Lastly, inhomogeneities in the fabricated bowtie antennas will broaden the resonances of the metadevice compared to the simulations. The temporal response of the device was measured using a function generator to modulate the applied current while the reflectivity at 1,107 nm was monitored (see figure S6). The results are presented in Figures 3e,f. First, it should be noted that the modulation contrast of the temporal intensity measurements is 37%, matching the steady-state measurements in Figure 3b and indicating that the device is fully switched. Second, the reflected power follows the trigger signal and has a rise time of 2.3 ms, which could be further reduced by increasing the applied current. The relaxation time was measured to be 1.27 ms, which is a function of the thermal conductivity of the substrate and the thermal capacitance of the VO2. When plasmonic structures are fabricated on or under continuous VO2 films, the absorption associated with the continuous VO2 film increases insertion loss, reduces modulation depth, and lengthens response time. Here, the patterned nanocrystals have a thermal mass that is 18% that of a thin film of the same thickness. The substantially reduced thermal capacitance of the system results in a response time

ACS Paragon Plus Environment

Page 6 of 21

Page 7 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

400 times shorter than film-based metadevices25 while also lowering the power consumption of the device. Overall, the modulator compares favorably to past work in this wavelength regime, as shown in Table 1. Furthermore, the switching speed observed here is by no means a lower limit. VO2 switching times on the order of 600 ns have been observed in waveguide-based structures in which current is passed directly through the VO233. In this case, the faster speed is due to the fact that only the region between the electrodes is heated, further reducing the thermal capacitance and collateral heat generation. This technique could in principle be readily translated to the device presented here to realize a several order of magnitude decrease in switching time. We also note that our metadevice was modulated for over 24,000 cycles and showed no degradation in the either the modulation depth or the speed, consistent with tests of current-driven VO2 devices which have been operated up to 260 million cycles without failure, far greater than the cycling reversibility of other phase change materials such as GST34,35. One application of this approach could be in infrared and visible identification tags and coding. To demonstrate how one could employ the device for such applications, VO2 nanocrystal arrays were patterned to form the letters ‘V’ and ‘U’ and then overlaid by a uniform, unpatterned, bow-tie antenna layer (24 µm x 24 µm)(Figure 4a, d). The array was imaged at a wavelength of 1,010 nm, corresponding to the point where the bow-tie with semiconducting VO2 in the feed gap (30 °C) have the same reflectivity as the bare bow-tie antenna array. In this case the array appears completely uniform. However, heating the metadevice to 82 °C switches the VO2 into the metallic state and shifts the absorption maximum to shorter wavelengths, resulting in a reflection contrast

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

of 23% (see measured spectrum in Figure S7). As shown in Figure 4(c-d, e-f), this results in the “V” and “U” becoming visible, revealing the previously hidden image. In summary, by fabricating unit cells with strong field concentration in a region occupied by nanocrystalline phase-change elements we have significantly reduce the required switching time and power compared to thin film-based spatial light modulators. The device exhibits a tuning range as large as 360 nm and a modulation depth up to 33%, with a response time of 1.27 ms. We believe this reconfigurable metasurface is a promising candidate for dynamic coatings and active beam shaping and with further improvements in switching speed it could be employed for signal processing and display applications.

Methods: Simulation: Numerical simulations were carried out with CST Microwave Studio. The measured permittivity was acquired from a 60 nm thick VO2 film that was grown using the same deposition parameters as the nanocrystals. The film was deposited on a silicon substrate and measured at 25 ℃ and 85 ℃. The optical properties of gold were taken from Johnson and Christy36 though the damping was increased three times compared to bulk gold to better match the experimental results.

Sample fabrication and measurement: All of the patterned structures were defined using electron beam lithography (Raith), then developed using MIBK:IPA=1:3 developer followed by a 3 second O2 descum. The VO2 structures were written using 120 nm lift-off

ACS Paragon Plus Environment

Page 8 of 21

Page 9 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

photoresist (50 nm PMMA 495 A2 and 70 nm 950 A2 stack). Thermal evaporation was used to deposit gold and RF magnetron sputtering of VO2 (Angstrom Engineering). VO2 was sputtered using a two-inch diameter vanadium target in a 6 mTorr Ar and O2 environment (20 sccm :1 sccm), with a deposition rate of 1.2 Å/s. After lift-off, the VO1.7 nanoparticles were annealed for 5 min at 450 ℃ in a 250 mtorr O2 environment to form VO2 nanocrystals. The gold bow-tie structure and backplane were deposited thermally under 5e-6 Torr vacuum, with a 3 nm Cr adhesion layer. The optical measurements were carried out using a super-continuum light source (Fianium) and the reflected signal from the sample was analyzed using a grating spectrometer with an InGaAs detector (Horiba iHR320).

Acknowledgement We gratefully acknowledge support from the National Science Foundation (CBET1336455) and the United States Department of Energy, Office of Science (DE-FG0201ER45916). A portion of this research was also sponsored by the Laboratory Directed Research and Development Program of Oak Ridge National Laboratory, managed by UTBattelle, LLC, for the U. S. Department of Energy. A portion of this research was conducted at the Center for Nanophase Materials Sciences, which is a DOE Office of Science User Facility.

Correspondence

and

requests

for

materials

should

[email protected]

ACS Paragon Plus Environment

be

addressed

to

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

References [1] Zheludev, N. I., Kivshar, Y. S. Nat. Mater. 2012, 11, 917–924. [2] Sterl, F., Strohfeldt, N., Walter, R., Griessen, R., Tittl, A., Giessen, H. Nano Lett. 2015, 15, 7949–7955. [3] Driscoll, T., Kim, H. T., Chae, B.G., Kim, B. J., Lee, Y. W., Jokerst, N. M., Palit, S., Smith, D. R., Di Ventra, M., Basov, D. N. Science 2009, 325, 1518-1521. [4] Simpson, R. E., Fons, P., Kolobov, A. V., Fukaya, T., Krbal, M., Yagi, T., Tominaga, J. Nat. nanotechnol. 2011, 6, 501-505. [5] Phare, C. T., Daniel Lee, Y.-H., Cardenas, J., Lipson, M. Nature Photon. 2015, 9, 511–514. [6] Liu, M., Yin, X., Ulin-Avila, E., Geng, B., Zentgraf, T., Ju, L., Wang, F., Zhang, X. Nature 2011, 474, 64-67. [7] Li, W., Chen, B., Meng, C., Fang, W., Xiao, Y., Li, X., Hu, Z., Xu, Y., Tong, L., Wang, H. Liu, W. Nano Lett. 2014, 14, 955-959. [8] Ee, H. S., Agarwal, R. Nano Lett. 2016, 16, 2818–2823. [9] Tittl, A., Michel, A. K. U., Schäferling, M., Yin, X., Gholipour, B., Cui, L., Wuttig, M., Taubner, T., Neubrech, F., Giessen, H. Adv. Mater. 2015, 27, 4597–4603. [10] Huang, Y. W., Lee, H. W. H., Sokhoyan, R., Pala, R., Thyagarajan, K., Han, S., Tsai, D. P., Atwater, H. A. Nano Lett. 2015, 16, 5319-5325. [11] Abb, M., Albella, P., Aizpurua, J., Muskens, O.L. Nano Lett. 2011, 11, 2457-2463. [12] Chen, C. F., Park, C. H., Boudouris, B. W., Horng, J., Geng, B., Girit, C., Zettl, A., Crommie, M. F., Segalman, R. A., Louie, S. G. et al. Nature 2011, 471, 617-620.

ACS Paragon Plus Environment

Page 10 of 21

Page 11 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

[13] Emani, N. K., Chung, T. F., Ni, X., Kildishev, A. V., Chen, Y. P., Boltasseva, A. Nano Lett. 2012, 12, 5202-5206. [14] Emani, N. K., Chung, T. F., Kildishev, A. V., Shalaev, V. M., Chen, Y. P., Boltasseva, A. Nano Lett. 2013, 14, 78-82. [15] Ou, J. Y., Plum, E., Jiang, L., Zheludev, N. I. Nano Lett. 2011. 11, 2142-2144. [16] Ou, J. Y., Plum, E., Zhang, J., Zheludev, N. I. Nat. Nanotechnol. 2013, 8, 252–5. [17]Shrekenhamer, D., Chen, W. C., Padilla, W. J. Phys. Rev. Lett. 2013, 110, 1–5. [18] Zhao, Y., Hao, Q., Ma, Y., Lu, M., Zhang, B., Lapsley, M., Khoo, I. C., Huang, T. J. Appl. Phys. Lett. 2012, 100, 053119. [19] Wang, H., Yang, Y., Wang, L. Appl. Phys. Lett. 2014, 105, 2012–2017. [20] Cueff, S., Li, D., Zhou, Y., Wong, F. J., Kurvits, J. A., Ramanathan, S., Zia, R. Nat. Commun. 2015, 6, 8636. [21] Kats, M. A., Sharma, D., Lin, J., Genevet, P., Blanchard, R., Yang, Z., Qazilbash, M. M., Basov, D. N., Ramanathan, S., Capasso, F. Appl. Phys. Lett. 2012, 101, 221101. [22] Kim, M., Jeong, J., Poon, J. K., Eleftheriades, G. V. JOSA B, 2016, 33, 980-988. [23] Michel, A. K. U., Chigrin, D. N., Maß, T. W., Schönauer, K., Salinga, M., Wuttig, M., Taubner, T. Nano Lett. 2013, 13, 3470-3475. [24] Peng, X. Y., Wang, B., Teng, J., Kana, J. K., Zhang, X., J. Appl. Phys. 2013, 114, 163103. [25] Liu, L., Kang, L., Mayer, T. S., Werner, D. H. Nat. Commun. 2016, 7, 13236. [26] Zhou, Y., Chen, X., Ko, C., Yang, Z., Mouli, C., Ramanathan, S. IEEE Electron Device Lett. 2013, 34, 220-222.

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

[27] Wang, D., Zhang, L., Gu, Y., Mehmood, M. Q., Gong, Y., Srivastava, A., Jian, L., Venkatesan, T., Qiu, C. W., Hong, M. Sci. Rep. 2015, 5, 15020. [28] Lopez, R., Haynes, T. E., Boatner, L. A., Feldman, L. C., Haglund, R. F. Phys. Rev. B 2002, 65, 224113. [29]Marvel, R. E., Harl, R. R., Craciun, V., Rogers, B. R., Haglund, R. F. Acta Materialia, 2015, 91, 217-226. [30] Valente, J., Ou, J. Y., Plum, E., Youngs, I. J., Zheludev, N. I. Nat. commun. 2015, 6, 7021 [31] Samson, Z. L., MacDonald, K. F., De Angelis, F., Gholipour, B., Knight, K., Huang, C. C., Di Fabrizio, E., Hewak, D. W., Zheludev, N.I. Appl. Phys. Lett. 2010, 96, 143105. [32] Gholipour, B., Zhang, J., MacDonald, K. F., Hewak, D. W., Zheludev, N. I. Adv. Mater. 2013, 25, 3050-3054. [33] Markov, P., Marvel, R.E., Conley, H.J., Miller, K.J., Haglund Jr, R.F. and Weiss, S.M., ACS Photon. 2015, 2, 1175-1182. [34] Aurelian C., Julien G., Jonathan L., David M., Corinne C., Jean-Christophe O., Alain C. and Pierre B. Sci. Tech. Adv. Mater., 2010, 11, 065002. [35] Kolobov, A. V., Fons, P., Frenkel, A. I., Ankudinov, A. L., Tominaga, J., and Uruga, T. Nat. mater., 2004, 3, 703-708. [36] Johnson, P. B. and Christy, R. W. Phys. Rev. B, 1972, 6, 4370–4379.

ACS Paragon Plus Environment

Page 12 of 21

Page 13 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figures and Captions

Figure 1 Schematic of the MPA structure. Schematic of (a) the multilayer MPA structure array and integrated local heater; (b) a unit cell with px=264 nm, py=300 nm, wire width w=100 nm, and gap width g=34 nm.

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2 Atomic force microscope and SEM images of the fabricated MPA structure. (a) AFM image of the uniform VO2 nanocrystals. The AFM data indicates the VO2 nanocrystal height is 37 nm. (b) Measured optical properties for both semiconducting and metallic VO2. Solid lines correspond to refractive index (left axis) and dash lines correspond to absorptive coefficient (right axis); (c) false color SEM image of the final device. (d) Magnified SEM image of the final device, the VO2 nanocrystals (purple) are located in the bow-tie antenna (yellow) feed-gaps.

ACS Paragon Plus Environment

Page 14 of 21

Page 15 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 3 Measured and simulated dynamic absorption spectra and temporal response. (a) Experimental absorption spectra as a function of the device temperature. (b) Experimental absorption spectra as a function of input current. (c) Simulated MPA absorption in the semiconducting, intermediate and metallic states. (d) Resonant wavelength (blue and left axis), and absorption amplitude (red and right axis) as a function of temperature; (e) Reflected power (blue circle) within one modulation cycle and the fitted exponential decay curve (red solid). The decay time is 1.27 ms based on the fit. (f) Trigger signal (red right axis) and monitored reflection (blue, left axis) over multiple modulation cycles.

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 21

Figure 4 Dynamic metadevices displaying images. (a, d), VO2 nanocrystal array in the shape of V and U. This array is covered with a uniform bow-tie antenna array. (b-c), (e-f), IR camera images of the final array at both 30 ℃ and 82 ℃. The illumination wavelength is 1010nm. At this wavelength the MPA in the absence of VO2 has the same reflection amplitude as an array with semiconductor VO2 in the feedgaps. Table 1 Comparison of near-infrared free space optical modulators trigger type

∆λ (nm)

Modulation depth

speed

Thermal15

NA

2%

NA

NA

Electrical16

250

8%

NA

NA

Graphene

Electrical14

NA

3%

NA

NA

ITO

Optical11

NA