The marriage of micro/nanoelectromechanical systems with metamaterials offers a viable route to achieving reconfigurable devices, which control the emission of energy. Here we propose and demonstrate the idea of a metamaterial microelectromechanical system (MEMS) capable of tailoring the energy emitted from a surface, without changing the temperature, but, instead, only altering the spectral emissivity. Our metamaterial achieves a range of emissivities equivalent to a nearly 20°C temperature change when viewed with a thermal infrared camera. We tessellate a surface with individually reconfigurable MEMS metamaterial pixels, thus realizing a spatiotemporal emitter capable of displaying thermal infrared patterns up to 110 kHz. Our results may be scaled to nearly any sub-optical range of the electromagnetic spectrum, and validate the potential of MEMS metamaterials to operate as reconfigurable multifunctional devices with unprecedented energy control capabilities.
© 2017 Optical Society of America
The first investigation into the ability of materials to achieve varied emissive properties  took place only 60 years after infrared (IR) radiation was first discovered by William Herschel in 1800 . Not surprisingly, study of the static control of thermal IR emission is still active owing to the many great potential applications ranging from thermal control and energy harvesting, to IR imaging. The capacity to dynamically tune a material’s emissive properties would be an even more exciting prospect—due to the extended range of possible applications—and has garnered significant attention recently [3–8]. For example, emitters based on microelectromechanical system (MEMS) were studied and their potential for combat identification—a technique for differentiating targets as friendly, enemy, or neutral—was highlighted [9,10]. Although some progress has been made [11,12], most systems demonstrated to date are either based on a change in the device’s temperature or due to a specific property of a natural material, and are thus limited due to high-temperature operation, slow modulation speed, or constrained operational frequency.
Electromagnetic metamaterials are promising candidates to overcome these limitations, since their radiative properties are controlled via their geometrical parameters rather than through their kinetic temperature or chemical composition [13–16]. Further, modulation rates may be several orders of magnitude faster with metamaterials—compared with thermal-based methods—making them a realistic solution for dynamic thermal control applications [17,18]. Here we propose and demonstrate the concept of an IR emitter that can be spatially and temporally controlled in real-time—based on electrically actuated MEMS metamaterials.
The MEMS metamaterial emitter (MME) studied here consists of a top (free-standing) moving section and a bottom metallic ground plane covered with a dielectric layer (aluminum oxide). A gold–germanium–gold segment comprises the top section and is connected to a support structure by eight cantilever arms, lying around the perimeter, which provide a restoring force when the layer is displaced from equilibrium [see the left configuration in Fig. 1(a)]. This three-layered design reduces the stress-induced surface curvature of the suspended sections . A bias applied between the bottom ground plane and the top layer provides an electrostatic force, thus actuating the device and bringing the free-standing section down into contact with the bottom layer [depicted in Fig. 1(a), right configuration]. The high absorption/emission state of our metamaterial is achieved due to the confluence of the electric and magnetic resonances, and is determined by the geometry of the unit cell . Our metamaterial geometry has been designed to achieve low emissive properties in the free-standing state, which we term the “off state”. As a bias is applied, the top layer is pulled down due to the electrostatic force and, at a certain level called the “pull-in voltage” , we reach a critical distance and the suspended film snaps down to the ground plane. In this “on-state” configuration [see Fig. 1(a), right configuration], our MME achieves high emissivity at a wavelength near 9 μm (see Supplement 1 for more information).
The MME investigated here is fabricated via a micromachining process (see Supplement 1). SEM images of the fabricated emitter are shown in Fig. 1(b) and 1(c) and, due to an unbalanced stress in the suspended films, a slight curvature across the sample surface can be observed . The fabricated sample was mounted at the focal plane of an IR microscope, which was coupled to a Fourier-transform IR spectrometer. The wavelength-dependent reflectivity () of our MME in the on state () and off state () was characterized and is shown in Supplement 1. We find zero transmissivity due to the 120 nm thick gold ground plane. In our study, we ensure our sample is always in thermal equilibrium by connecting it to a thermal bath at room temperature. We thus determine the spectral emissivity and absorptivity as .
The measured voltage-dependent spectral absorptivity of our MME is shown in Fig. 2(a). In the off state, the top patterned metallic film is suspended from the ground plane [inset in Fig. 2(a)] and realizes relatively low spectral absorptivity with between 6 μm and 8.5 μm. At longer wavelengths, increases and realizes a broad feature with a peak value of 0.5 at 10.2 μm (blue curve), before slowly diminishing to values near 0.25 at 16 μm. Upon application of a 15 V bias, the suspended film snapped down to the ground plane and a sharp resonant absorptivity peak was achieved with a value of at a wavelength of 8.9 μm (red curve). A natural absorption peak of aluminum oxide —coupled to the resonance of the metamaterial—causes the spectral absorptivity to be broad and relatively wavelength-independent for values greater than 10 μm. In order to highlight the ability of our MEMS metamaterial to modulate , we calculate the differential spectral absorptivity, defined as , where is the spectral absorptivity plotted in Fig. 2(a). The differential absorptivity realizes a sharp peak with a value near 70% at a wavelength of 8.9 μm. We perform finite-difference time domain simulations in order to elucidate the nature of the and values obtained. As can be observed from Fig. 2(b), although our simulation consists of only a single unit cell, and thus realizes “perfect” on and off states with no top layer curvature, we find relatively good agreement with experimental measurements.
An important metric often used to quantify dynamical properties is the modulation speed —determined by where the modulated signal falls to 3 dB of its low frequency value. We characterized in a custom IR microscope setup where the sample was driven by a periodic signal supplied by a function generator. The modulated thermal IR radiation was measured by a mercury cadmium telluride detector with the synchronized output used as a reference for lock-in detection. The detected signal obtained from the lock-in amplifier is shown in Fig. 2(c) as a function of modulation frequency. At low modulation frequencies, our response is relatively flat and begins to increase near 50 kHz [see Fig. 2(c)]. We find a broad mechanical resonant response peak—due to the MEMS structure—at 85 kHz before the modulated signal begins to fall off for increased modulation speeds (Supplement 1). The 3 dB point of modulation is determined with respect to the signal level at 30 kHz, i.e., before the mechanical resonance, and we find a value of , denoted by the vertical dashed red line in Fig. 2(c). Remarkably, although our metamaterial achieves its dynamic properties through mechanical actuation, we realize modulation speeds that far surpass those possible with a thermal approach [3,10].
We next turn toward characterization of the IR radiance of the MME device. A thermal IR camera was used to record the time-dependent radiance from the surface of the emitter. According to Kirchhoff’s law of thermal radiation, in thermodynamic equilibrium, the spectral emissivity of any material is equal to its spectral absorptivity . The measured is used to calculate the spectral radiant exitance by multiplying with the blackbody spectrum at room temperature and the spectral sensitivity of the camera ,2(d). We integrate the spectral radiant exitance over the range of our IR camera, 6 μm to 16 μm, and calculate that our MEMS metamaterial is able to change its radiosity (emitted power density) from in the off state to a value of approximately in the on state. Most natural materials realize temperature-independent spectral emissivities, and, thus, the change in radiosity exhibited by our MME would be equivalent to a temperature change from 23°C to 42°C—a difference of 19°C (see Supplement 1).
In order to verify the ability of our MME to achieve high dynamic radiant exitance, we perform direct imaging measurements of the surface of the metamaterial with a thermal IR camera. The sample was mounted with the surface normal oriented parallel to the optical axis of the IR camera, and located at the focal plane of the lens. The pixel size of our IR camera is and the magnification of the lens is 1.5. Since IR imaging was performed at room temperature, and our emitter was partially reflective in the detecting wavelength range of the camera, a normalization procedure was implemented (see Supplement 1) . The total power density detected by the camera () consisted of two terms, the emitted power density from the metamaterial () and the ambient power density reflected by the metamaterial (), which we write as,Supplement 1) with and . Then of the sample in both on and off states is determined by Eq. (4), and is derived from Eq. (3).
In the measurement, a ¼ Hz periodic voltage with magnitude from 0 V to 20 V was applied to the MME and the surface was imaged with our thermal IR camera. The time dependence of the emitted power density (averaged over the emitter area) of our device is shown in Fig. 3(a) (left axis) (Supplement 1). We observe in Fig. 3(a) that the power density is modified from to ; thus, the modulation index, defined as , of our device with respect to thermal radiation can be as large as 23.7%. To verify the emitted power density shown in Fig. 3(a), we calculated the emitted power density in both on and off states from Eq. (5), and the results are consistent with that derived above within 4% error (see Supplement 1).
The thermal radiation from a surface is dependent on both the emissivity and kinetic temperature of the surface. As mentioned before, most natural materials realize temperature-independent spectral emissivity, and, thus, changes in their thermal radiation must be from a change in their kinetic temperature. Our MME, on the other hand, is able to dynamically tailor its spectral emissivity, thus resulting in changes in thermal radiated power without a change in temperature. For example, a thermal IR camera integrates the received energy over its operational wavelength range and assumes a temperature-independent emissivity. Applying these approximations to our MME, we find an equivalent temperature change of nearly 20°C from room temperature [Fig. 3(a), right axis] (see Supplement 1). We study the impact of the slight curvature of the suspended portion of our MME and plot the spatial distribution of the temperature over the surface of the emitter in both on and off states in Fig. 3(b) and 3(c), respectively. We find that in the off state, the spatially averaged temperature is 25.4°C, whereas when a bias is applied, the temperature increases to 44.3°C—a change of 18.9°C.
We next highlight the potential application of our MME for combat identification. A device composed of 8 by 8 emitter pixels was designed and fabricated (Supplement 1). A ½ Hz sine wave signal from 0 V to 13 V was applied to the device and we write a “D” pattern in our pixelated MME. The thermal radiation of the surface was then imaged with an IR camera. Although all measurements were performed at room temperature, the letter “D” clearly appears [on state, Fig. 4(a)] and disappears [off state, Fig. 4(b)] periodically at a rate of ½ Hz (see Visualization 1).
We have designed, fabricated, and demonstrated a room temperature dynamic metamaterial IR emitter, whose modulation speed can be as high as 110 kHz. A measured modulation index of 23.7% at thermal IR wavelengths was obtained and corresponds to a nearly 20°C difference when viewed with an IR camera. In comparison, a traditional IR emitter for friend or foe identification systems operates at several hundred degrees centigrade and, since they are thermal-based devices [3,10], our MME achieves modulation speeds that are approximately ten thousand times faster . Our results verify the capability of metamaterials to enable real-time control of the emissivity of surfaces and demonstrate a new path forward to related applications in IR camouflage [4,25], friend or foe identification [9,10,26], and IR scene projectors [27,28]. We believe that the control of IR radiation enabled by reconfigurable MEMS metamaterials has the potential to significantly impact nanoscience and nanotechnology due to their novel design approach and multifunctional capabilities.
U.S. Department of Energy (DOE) (DE-244 SC0014372); National Science Foundation (NSF) (ECCS-1542015).
We thank Dr. Kebin Fan for useful discussions and helpful fabrication suggestions. We acknowledge support from the U.S. Department of Energy (DOE) (DE-244 SC0014372). This work was performed in part at the Duke University Shared Materials Instrumentation Facility (SMIF), a member of the North Carolina Research Triangle Nanotechnology Network (RTNN), which is supported by the National Science Foundation (NSF) (Grant ECCS-1542015) as part of the National Nanotechnology Coordinated Infrastructure (NNCI).
See Supplement 1 for supporting content.
1. B. Stewart and T. Roy, Soc. Edinb. 22, 1 (1861). [CrossRef]
2. W. Herschel, Philos. Trans. R. Soc. London 90, 284 (1800). [CrossRef]
3. J. Hildenbrand, J. Korvink, J. Wollenstein, C. Peter, A. Kurzinger, F. Naumann, M. Ebert, and F. Lamprecht, IEEE Sens. J. 10, 353 (2010). [CrossRef]
4. M. A. Kats, R. Blanchard, S. Zhang, P. Genevet, C. Ko, S. Ramanathan, and F. Capasso, Phys. Rev. X 3, 041004 (2013). [CrossRef]
5. T. Inoue, M. De Zoysa, T. Asano, and S. Noda, Nat. Mater. 13, 928 (2014). [CrossRef]
6. V. W. Brar, M. C. Sherrott, L. A. Sweatlock, M. S. Jang, L. Kim, M. Choi, and H. A. Atwater, Nat. Commun. 6, 7032 (2015). [CrossRef]
7. X. Liu and W. J. Padilla, Adv. Mater. 28, 871 (2016). [CrossRef]
8. A. Tittl, A. U. Michel, M. Schäferling, X. Yin, B. Gholipour, L. Cui, M. Wuttig, T. Taubner, F. Neubrech, and H. Giessen, Adv. Mater. 27, 4597 (2015).
9. M. U. Pralle, I. Puscasu, J. Daly, K. Fallon, P. Loges, A. Greenwald, and E. Johnson, Proc. SPIE 6542, 65422Z (2007). [CrossRef]
10. H. San, X. Chen, P. Xu, F. Li, and M. Cheng, Proc. SPIE 6940, 69401T (2008). [CrossRef]
11. J. S. Hale and J. A. Woollam, Thin Solid Films 339, 174 (1999). [CrossRef]
12. H. Demiryont and D. Moorehead, Solar Energy Mater. Sol. Cells 93, 2075 (2009). [CrossRef]
13. I. Puscasu and W. L. Schaich, Appl. Phys. Lett. 92, 233102 (2008). [CrossRef]
14. L. Luo, I. Chatzakis, J. Wang, F. B. P. Niesler, M. Wegener, T. Koschny, and C. M. Soukoulis, Nat. Commun. 5, 3055 (2014). [CrossRef]
15. X. Liu, T. Tyler, T. Starr, A. F. Starr, N. M. Jokerst, and W. J. Padilla, Phys. Rev. Lett. 107, 045901 (2011). [CrossRef]
16. S. Molesky, C. J. Dewalt, and Z. Jacob, Opt. Express 21, A96 (2013). [CrossRef]
17. X. Liu and W. J. Padilla, Adv. Opt. Mater. 1, 559 (2013). [CrossRef]
18. J. Y. Ou, E. Plum, J. Zhang, and N. I. Zheludev, Nat. Nanotechnol. 8, 252 (2013). [CrossRef]
19. W. S. Yang, S. M. Cho, H. Ryu, S. H. Cheon, B. G. Yu, and C. A. Choi, Proceedings of the 2nd International Conference on Sensor Technologies and Applications (2008), p. 391.
20. C. M. Watts, X. Liu, and W. J. Padilla, Adv. Mater. 24, OP98 (2012). [CrossRef]
21. E. Saucedo-Flores, R. Ruelas, M. Flores, and J.-C. Chiao, Proc. MRS 782, A5.86 (2003). [CrossRef]
22. W. D. Cowan, V. M. Bright, A. A. Elvin, and D. A. Koester, Proc. SPIE 3225, 56 (1997). [CrossRef]
23. J. Kischkat, S. Peters, B. Gruska, M. Semtsiv, M. Chashnikova, M. Klinkmüller, O. Fedosenko, S. Machulik, A. Aleksandrova, G. Monastyrskyi, Y. Flores, and W. Ted Masselink, Appl. Opt. 51, 6789 (2012). [CrossRef]
24. M. A. Noginov, A. Mozafari, T. U. Tumkur, J. K. Kitur, and E. E. Narimanov, Opt. Mater. Express 5, 1511 (2015). [CrossRef]
25. C.-W. Cheng, M. N. Abbas, C.-W. Chiu, K.-T. Lai, M.-H. Shih, and Y.-C. Chang, Opt. Express 20, 10376 (2012). [CrossRef]
26. I. Puscasu, E. Johnson, A. Taylor, and W. Schaich, Proc. SPIE 8031, 80312Y (2011). [CrossRef]
27. O. M. Williams, Infrared Phys. Technol. 39, 473 (1998). [CrossRef]
28. B. E. Cole, C. J. Han, R. E. Higashi, and J. Ridley, Sens. Actuators A 48, 193 (1995). [CrossRef]