We report a multifunctional metamaterial composite structure that not only provides the broadband radar and thermal infrared bi-stealth function but also possesses an in-band microwave transmission window and high optical transparency. It is composed of four metasurface layers made of indium tin oxide (ITO) films with different surface resistances, which are specifically designed to sequentially control the infrared emission, microwave absorption and transmission. The fabricated sample exhibits a low reflectivity less than 10% in 1.5-9 GHz and a transmission peak of 50% around 3.8 GHz up to the incident angle of 30 degrees. In the infrared atmosphere window, a low thermal emissivity of about 0.52 is achieved. Meanwhile, it keeps good optical transparency by the use of the ITO films. The optically transparent, low-infrared-emissivity, radar-reflectionless and frequency-selective-transmission properties will enable the promising application of communication-compatible multispectral stealth technology.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Functional integration of multi-spectrum stealth is vital to counteract the comprehensive multiband development of modern military detection technologies [1–3]. For radar and infrared (IR), the two mostly employed detection measures, it is essential to develop materials with variable loss dispersions that can satisfy the conflicting parametric requirement for radar and IR stealth. High absorption and low reflection are required in radar stealth, while low absorption or thermal emissivity is demanded in infrared stealth [4–6]. They have different requirements on the material impedance value. Traditionally, metallic particles embedded composites have been explored for this specific purpose [7–12]. For examples, radar absorption materials with controlled IR emissivity around 0.5, such as using ordered mesoporous C-MO (M = Al, Si, or Ti, etc) nanocomposites [7, 8] or carbon nanotube loaded perovskites , have been investigated. Lv et al. fabricated FeAl flake mixtures with a narrow microwave absorption bandwidth of 1.7 GHz and IR emissivity of 0.15 . However, the pure composition optimization scheme is physically of low efficiency in controlling the bi-band loss dispersions and the microwave and infrared properties can be hardly simultaneously well manipulated. On the other hand, metamaterials and their two-dimensional structures, metasurfaces , have the advantages in engineering the dispersions of electromagnetic (EM) parameters through the selection and optimization of subwavelength resonance “atoms” and have enabled the designs of various EM functional devices, for instance, perfect absorbers [14–20], frequency selective transmission and absorption [21–23], polarimeters [24, 25], anomalous beam bending [26, 27], etc. Compared with the natural composite, the artificial structures with engineered “atoms” could have more freedoms to control the band response across different spectrum regions. As an example, recently, the authors took this designing advantages and experimentally demonstrated a metasurfaces-based radar-IR bi-stealth structure that has a strong absorptivity over 90% in 3-8 GHz and a low IR emissivity of 0.2 . It is believed that more complex EM properties or requirements could be realized using the artificial structures.
In some real applications, besides the multiband response, the stealth coats should also have the transparent windows for the purposes of signal communication, such as the antenna radome or the cockpit window. In microwaves, it needs materials with selective transmission, while in optics, transparent materials have to be employed. The implementation becomes much harder due to the complex parametric requirements, i.e., combination of multispectral absorption and transmission. Such a possibility for simultaneous multifunctional engineering has been explored in the literature using metamaterials but only for the integration of partial functions. For example, by using optically transparent aluminum grid  or ITO-coated-PET , broadband microwave absorbers with optical transparency have been designed. But neither their infrared properties nor microwave frequency-selective transmission functions are involved. In this work, we show a composite metasurface structure that simultaneously possesses the features of transmission-selective broadband microwave absorption, low infrared emission and high optical transparency, thus enabling the application of multispectral stealth and communication. These properties are realized based on the usage of transparent conductive ITO-coated-PET thin sheets which are subtly patterned to realize the different sub-functions. The measurements show our sample has a low reflectivity less than 10% from 1.5 to 9 GHz and a narrow in-band transmission window around 3.8 GHz at illumination angles up to 30 degrees. The emissivity in the infrared atmosphere window of 8-14 μm is about 0.52 and the optical transmittance of the overall structure is about 33%. These results indicate that our proposed structure has potential applications in multispectral stealth-communication compatibility technology.
2. Design and methods
Our aim in this work is to design a multifunctional EM coat that not only provides the basic radar-IR bi-stealth function but also has the in-band microwave transmission channel and high optical transparency. For this complex purpose, we need to overcome three major difficulties related to mediate the conflictions between: (1) optical transparency and microwave controllability, (2) broadband radar wave absorption and in-band transmission and (3) microwave absorption and IR-shielding. In our approach, highly conductively ITO film coated PET sheets have been primarily adopted to solve the first problem. While for the second, we pursue a combined broadband metamaterial absorber design which is grounded by a frequency selective surface. Controlling of the insertion loss in the transmission window is realized through the optimization of the quality factor of the local resonance cavity. For the third one, the size-dependent resonance character of metasurface has been fully utilized to configure a microwave-transparent IR screen. The details are discussed in the following.
As shown in Fig. 1(a), the proposed structure from the top to bottom consists of infrared shielding layer (IRSL), radar absorption layer 1 (RAL1), radar absorption layer 2 (RAL2) and frequency selective transmission layer (FSTL). All the layers are designed by use of transparent ITO-coated PET films with different surface resistances, 5 Ω/sq for IRSL and FSTL, 15 Ω/sq for RAL1, and 35 Ω/sq for RAL2, respectively. The thickness of the ITO-coated PET film is 125 μm for 15 Ω/sq and 35 Ω/sq types and 175 μm for 5 Ω/sq type. The PET has dielectric constant of 3 and loss tangent of 0.06. The layers are separated by air with each other at distances t1 = 3 mm, t2 = 18 mm and t3 = 12 mm, respectively, as shown in Fig. 1(a). The size of the unit cell shown in Fig. 1(a) is a = 100 mm, and there are 5 × 5 sub-cells in the RAL2 and 2 × 2 sub-cells in FSTL. The IRSL is composed of ITO patch array metasurface with periodicity 1.724 mm and gap width 0.5 mm. The RAL1 is made of square loop with length l1 = 60 mm and 2 mm line-width. The square loop of the RAL2 has length of l2 = 14.2 mm and 1 mm line-width. The FSTL is designed by etching square rings with outer and inner lengths l3 = 35 mm and l4 = 15 mm away from ITO film, and then two copper square rings with strip width of 2.5 mm are connected at the sides of the complementary square ring, which generates complementary square ring with 5-mm width. Copper is introduced here to partially replace ITO in order to decrease the transmission loss at the selective frequency. Figures 1(b) and 1(c) present a fabricated 100 mm × 100 mm unit cell, which demonstrates good optical transparency of the 4-layer structure. All the layers are fabricated by using the standard photolithography technique. The copper loops in the FSTL is fabricated by copper foils with 2.5-mm width, and adhered along the sides of ITO. The layers are supported by plastic posts of different heights at the corners. In the measurements, we assemble 9 unit cells [Fig. 1(b)] into the total sample with a size of 300 mm × 300 mm. Three spacing foams with suitable thickness are added between adjacent layers for convenience of measurement.
2.1 Microwave absorption and frequency selective transmission
Resistive close ring resonators were widely used to design broadband perfect absorbers from microwave to terahertz regimes [18, 20]. In this work, transparent ITO films with intrinsic resistive loss are employed to realize wideband absorption in the microwave absorption layers. By replacing the former full ground with a square-slot band-pass frequency selective surface (FSS), a transmission window within the absorption band can be opened, preserving a low reflection in the whole band. Full-wave simulations are carried out using a frequency domain solver. Excitations propagating along the z direction from port 1 with the E field along the x direction and the H field along the y direction (the inset in Fig. 2) are used to calculate the S parameters. Periodic boundary conditions are set in both the x and y directions. The reflectivity (R), transmissivity (T) and absorptivity (A) are calculated by R = |S11|2, T = |S21|2, and A = 1-R-T, respectively. Figure 2 shows the simulated reflectivity, transmissivity and absorptivity (RTA) spectra of the structure at normal incidence. It is seen that the reflectivity is less than 10% from 1.1 GHz to 9.9 GHz, and there is a transmission peak of 48% at 3.5GHz between two absorption bands. The RTA spectra are polarization-independent for both TE and TM waves at normal incidence due to the structure symmetry.
To reveal the influence of the IRSL on the microwave performance, the simulated reflectivity and transmissivity of the structure with the IRSL under normal incidence are compared with the case without the IRSL in Fig. 3(a). When the IRSL is added, the transmission peak has a slight change, and a little rise of the reflectivity below and above the transmission band occurs. It can be explained by the mutual coupling between the IRSL and the other layers. The IRSL is effectively a thin dielectric layer with over-95% transmissivity and less-than-12° phase shift in this band, as shown in Fig. 3(b). The experimental transmissivity is over-90%, which agrees with the simulated result well, despite a little discrepancy due to background noise. The IRSL could be simply regarded as a low-pass FSS, passing waves below its resonant frequency. By tailoring the element size and the gap, it is designed to be transparent to the radar wave in the working band, ensuring the absorption and frequency selective transmission not disturbed. A larger metallic filling ratio of the IRSL is required to obtain a lower emissivity for the IR stealth function.
In order to improve the peak transmissivity of FSTL, conventional slot type band-pass FSS  is modified by loading copper loops near the slot sides. Figure 4 shows the R and T spectra of the ground FSTL without and with introduced copper. As can be seen, in both cases, the FSTLs have transmission peaks around 3.5 GHz, and high reflectivity below and above the passband (except a dip at 8.8 GHz which is caused by high order resonance). However, for the case of ITO pattern without copper, the peak transmissivity is 68%, much lower than the case with copper introduced (97%). It can be explained from the current distributions depicted in the right panel of Fig. 4, showing that at the transmission frequency 3.5GHz, the current concentrate near the sides of the slot. As a consequence, the ITO case has a lower transmission due to high resistive loss. By employing low resistive copper near the slot sides, corresponding to stronger resonance with a higher quality factor, the transmission of FSS can be increased, although, at the cost of optical transparence.
The mechanism of the proposed absorption-transmission-integrated structure can be explained by equivalent transmission line model. As shown in Fig. 5(a), the spacers between adjacent layers are equivalent to transmission lines with different lengths, and all the layers can be regarded as impedance sheets for the incident radar wave, with impedance ZF1, ZF2, ZF3 and ZF4, respectively, since they are thin enough compared to the radar wavelength. The impedance of each employed sheet is frequency-dependent due to resonances, which is computed by matching the normal incidence full-wave response of each layer in the freestanding configuration. According to the equivalent circuit described in Fig. 5(a), the ABCD matrix of the network is written as,29], we can get the expressions for S11 and S21, and then calculate the RTA spectra. Based on the above equation, we adopt the analytical approximate analysis on the structural parametrical dependence of the RTA characteristics, avoiding time-consuming simulations for the total structure. For instance, to optimize the separation t1 between IRSL and RAL1, the RTA spectra of the structure for different t1 are calculated according to the transmission-line equation, as shown in Fig. 5(b). It is seen that the separation t1 has little influence on the RTA spectra in the lower band and in the higher band, the reflectivity decreases slightly as the separation t1 increases, primarily caused by the mutual coupling between the IRSL and RAL1 which degrades the impedance match. In our fabricated sample, the separation t1 is designed to be 3 mm, which offers satisfactory absorptivity and sample thickness.
The RAL1 and RAL2 play major roles in the absorption of lower and higher bands, respectively. As depicted in Fig. 6, in the absence of RAL2, the structure with RAL1 has an absorption peak in 1-2 GHz. On the contrary, for the case without RAL1 (only with RAL2), the absorption drops down in the lower band and rises in the higher band (5-9 GHz). This is because these two layers with different sizes of ITO loops resonate at the lower and higher frequencies, respectively. The simulated surface currents and power loss density of the unit cell at different frequencies are illustrated in Fig. 7 in order to get more insight into the absorption mechanism. It is shown that at 1.7 GH and 7.6 GHz, most of the currents are excited at the loops in RAL1 and RAL2, respectively. Then the power will be dramatically dissipated at the ITO loops where the currents accumulate, as shown in Figs. 7(d) and 7(f), since the currents will lead to significant ohmic loss with Ploss = I2Rs, in which Rs is the surface resistance. At the transmission frequency of 3.5 GHz, electric currents are excited in the loops of RAL1 and RAL2 [see Fig. 7(b)], which give rise to power loss [see Fig. 7(e)], resulting in only 50% transmissivity at this transmission window.
The radar reflection and transmission of the fabricated sample are measured in an anechoic chamber by the free space method  using Agilent E8361C vector network analyzer (VNA). Two horn antennas working in 1-12 GHz are connected to the VNA as the transmitter and receiver. The sample is placed in the hole of an absorbing screen and its center is carefully aligned with the center of the antenna apertures, as shown in the Figs. 8(a) and 8(b). In the reflection measurement, orientations of the antennas are varying for different incidence angles, namely 10° and 30°. The results are normalized with a control experiment using a metal plate that has a same size with the sample. Moreover, in the transmission measurement, the absorbing screen with the sample is rotated for oblique incidence, and the transmissions of the absorbing screen without the sample are tested for normalization. The absorptivity is calculated by A = 1-R-T = 1-|S11|2-|S21|2. As illustrated in Fig. 8(c), at the 10° incidence angle, both polarization waves show low reflectivity with R < 10% in a wide range of 1-9 GHz. When the incident angle increases to 30°, the R value increases in the lower band (< 3 GHz) and decreases in the higher band (> 9 GHz). The sample has a transmission window around 3.8 GHz, with a peak value of about 50% for both TE and TM polarizations at different incidence angles. Two absorption bands with over 90% absorptivity appear below and above the transmission band. The absorptivity drops to 80% in the lower band for the larger incidence angle. The experimental results agree well with the simulation results except for a little discrepancy of the transmission frequency from 3.5 GHz to 3.8 GHz as observed in Fig. 2, which is attributed to the fabrication flaws of the copper loops in the sample.
2.2 Low emissivity property
For infrared stealth application, metals are ideal materials due to their high reflectivity and low emissivity for infrared. ITO, like metals, also shows high IR reflectivity and low IR emissivity, which is suitable to control the IR emission by itself and block the IR light from inner objects of the structure. We examine both transmission and emission characteristics of the IRSL in the infrared range. A 15 × 15 mm2 square is cut from the original IRSL for the tests. The transmissions and reflections of PET, ITO-coated PET and IRSL are measured by a FTIR Spectrometer (Bruker Vertex 70). The reflectance and transmittance are normalized by a high reflective mirror made of 200 nm thick gold film and the empty case without sample, respectively. The spot size of IR beam is about 5 mm, covering several unit cells to get average reflectance and transmittance of the IRSL sample. Then the absorptions are calculated by A = 1-T-R. According to Kirchhoff's law, at the same temperature, emissivity is equal to absorptivity. Consequently, the IR emission property can be measured indirectly according to the absorption characteristic. As shown in Fig. 9, for the PET substrate, two transmission peaks of about 16% appear at 10.8 and 13.2 μm, and the absorption is about 90% within the infrared atmosphere window of 8-14 μm. The ITO has zero transmission and only 5% absorption in the whole band, which manifests a superior IR stealth property. From the blue solid line, we can find that the transmittance peaks of the IRSL are about 8% at the same wavelengths, which is about half of the PET. It is because the ITO filling 50% of the IRSL area blocks the transmission wave. The measured absorption of the IRSL is about 52% in the 8-14 μm range, which means that the surface emissivity is about 0.52. The surface infrared emissivity of IRSL is 40% lower than the PET substrate, originating from half of the surface occupied by ITO with low emissivity (0.05). The infrared emissivity of the IRSL can also be evaluated by an empirical formula [11, 12]12], when the ITO is patterned into square patch arrays [see Fig. 3(b)], this layer becomes nearly transparent to microwaves below 10 GHz with transmissivity larger than 95%, which has little influence for the radar absorption and frequency selective transmission functions.
2.3 Optical transparency
In our design, the use of transparent ITOs to achieve different functions is the premise for transparent characteristic of the total device. The optical transmittances of different layers of the structure are measured by Ocean-Optics-2000 Spectrometer. As depicted in Fig. 10, the light transmittance of the ITO-coated PET (5 Ω/sq) has a peak of about 60% at 580 nm. The transmittance of FSTL in the structure can be calculated according to the filling ratios of the ITO and PET (ITO area/total area and PET area/total area). The FSTL has a lower transmittance than the ITO-coated PET due to the introduced opaque copper loops, as shown in Fig. 10. By reducing the cooper parts, the light transparence can be enhanced further. The total structure transmittance can be obtained by multiplying the transmittance of each layer, which reaches 33% at 610 nm. The use of ITO layer with higher transparence in the FSTL and IRSL will be beneficial to improve the light transmittance despite microwave transmission at the window will attenuate due to the increased resistive loss of ITO. Further optimization of the structure is required to get a better optical transparence without affecting the microwave transmission property in the future design.
In summary, an optically transparent broadband radar stealth structure with a microwave transmission window and low infrared emissivity made of four specifically designed ITO-based metasurface stacks was proposed. Two different radar absorbing layers and a frequency selective transmission layer are employed to achieve ultra-broadband reflectivity less than 10% from 1.5 GHz to 9 GHz, with a window at 3.8 GHz with transmissivity 50%. Moreover, a radar transparent metasurface with low surface IR emissivity of 0.52 in the atmosphere window is designed to realize radar-infrared bi-stealth function. The overall structure has good optical transparence with measured light transmittance of 33%. Both the simulated and experimental results indicate that our multifunctional structure will find promising applications for multispectral stealth and communication integration technology.
National Natural Science Foundations of China (NSFC) (61701268, 61775195 and 61631012); Natural Science Foundations of Ningbo City (2016A610066) and Zhejiang Province (LQ17F010003 and LZ17A040001).
This work is partially sponsored by K. C. Wong Magna Fund in Ningbo University. We acknowledge Kequn Chi for his great help in the sample test.
References and links
1. E. Knott, J. F. Shaeffer, and M. T. Tuley, Radar Cross Section, 2nd ed. (Scitech Publishing Inc., 2004).
2. D. Qi, X. Wang, Y. Z. Cheng, R. Z. Gong, and B. W. Li, “Design and characterization of one-dimensional photonic crystals based on ZnS/Ge for infrared-visible compatible stealth applications,” Opt. Mater. 62, 52–56 (2016). [CrossRef]
3. Z. P. Mao, W. Wang, Y. Liu, L. P. Zhang, H. Xu, and Y. Zhong, “Infrared stealth property based on semiconductor (M)-to-metallic (R) phase transition characteristics of W-doped VO2 thin films coated on cotton fabrics,” Thin Solid Films 558, 208–214 (2014). [CrossRef]
4. Q. Fu, W. W. Wang, D. L. Li, and J. J. Pan, “Research on surface modification and infrared emissivity of In2O3:W thin films,” Thin Solid Films 570, 68–74 (2014). [CrossRef]
5. L. Phan, W. G. Walkup 4th, D. D. Ordinario, E. Karshalev, J. M. Jocson, A. M. Burke, and A. A. Gorodetsky, “Reconfigurable infrared camouflage coatings from a cephalopod protein,” Adv. Mater. 25(39), 5621–5625 (2013). [CrossRef] [PubMed]
6. L. Xiao, H. Ma, J. Liu, W. Zhao, Y. Jia, Q. Zhao, K. Liu, Y. Wu, Y. Wei, S. Fan, and K. Jiang, “Fast adaptive thermal camouflage based on flexible VO2/graphene/CNT thin films,” Nano Lett. 15(12), 8365–8370 (2015). [CrossRef] [PubMed]
7. T. Wang, J. P. He, J. H. Zhou, X. C. Ding, J. Q. Zhao, S. C. Wu, and Y. X. Guo, “Electromagnetic wave absorption and infrared camouflage of ordered mesoporous carbon–alumina nanocomposites,” Microporous Mesoporous Mater. 134(1-3), 58–64 (2010). [CrossRef]
8. J. H. Zhou, J. P. He, G. X. Li, T. Wang, D. Sun, X. C. Ding, J. Q. Zhao, and S. C. Wu, “Direct incorporation of magnetic constituents within ordered mesoporous carbon-silica nanocomposites for highly efficient electromagnetic wave absorbers,” J. Phys. Chem. C 114(17), 7611–7617 (2010). [CrossRef]
9. L. Chen, C. H. Lu, Z. G. Fang, Y. Lu, Y. R. Ni, and Z. Z. Xu, “Infrared emissivity and microwave absorption property of Sm0.5Sr0.5CoO3 perovskites decorated with carbon nanotubes,” Mater. Lett. 93, 308–311 (2013). [CrossRef]
10. H. Lv, G. Ji, X. Li, X. Chang, M. Wang, H. Zhang, and D. Youwei, “Microwave absorbing properties and enhanced infrared reflectance of FeAl mixture synthesized by two-step ball-milling method,” J. Magn. Magn. Mater. 374, 225–229 (2015). [CrossRef]
11. H. Tian, H. T. Liu, and H. F. Cheng, “A thin radar-infrared stealth-compatible structure: design, fabrication, and characterization,” Chin. Phys. B 23(2), 025201 (2014). [CrossRef]
12. S. M. Zhong, W. Jiang, P. P. Xu, T. J. Liu, J. F. Huang, and Y. G. Ma, “A radar-infrared bi-stealth structure based on metasurfaces,” Appl. Phys. Lett. 110(6), 063502 (2017). [CrossRef]
14. S. M. Zhong, Y. G. Ma, and S. L. He, “Perfect absorption in ultrathin anisotropic ε-near-zero metamaterials,” Appl. Phys. Lett. 105(2), 023504 (2014). [CrossRef]
15. C. M. Watts, X. Liu, and W. J. Padilla, “Metamaterial electromagnetic wave absorbers,” Adv. Mater. 24(23), OP98–OP120 (2012). [PubMed]
17. J. F. Zhu, Z. F. Ma, W. Sun, F. Ding, Q. He, L. Zhou, and Y. G. Ma, “Ultra-broadband terahertz metamaterial absorber,” Appl. Phys. Lett. 105(2), 021102 (2014). [CrossRef]
19. T. Jang, H. Youn, Y. J. Shin, and L. J. Guo, “Transparent and flexible polarization-independent microwave broadband absorber,” ACS Photonics 1(3), 279–284 (2014). [CrossRef]
20. C. Zhang, Q. Cheng, J. Yang, J. Zhao, and T. J. Cui, “Broadband metamaterial for optical transparency and microwave absorption,” Appl. Phys. Lett. 110(14), 143511 (2017). [CrossRef]
21. Y. P. Shang, Z. X. Shen, and S. Q. Xiao, “Frequency-selective rasorber based on square-loop and cross-dipole arrays,” IEEE Trans. Antenn. Propag. 62(11), 5581–5589 (2014). [CrossRef]
22. V. Asadchy, I. Faniayeu, Y. Ra’di, S. Khakhomov, I. Semchenko, and S. Tretyakov, “Broadband reflectionless metasheets: frequency-selective transmission and perfect absorption,” Phys. Rev. X 5(3), 031005 (2015). [CrossRef]
23. Y. Shen, J. Q. Zhang, Y. Q. Pang, Y. F. Li, Q. Q. Zheng, J. F. Wang, H. Ma, and S. B. Qu, “Broadband reflectionless metamaterials with customizable absorption-transmission-integrated performance,” Appl. Phys., A Mater. Sci. Process. 123(8), 530 (2017). [CrossRef]
24. E. Maguid, I. Yulevich, D. Veksler, V. Kleiner, M. L. Brongersma, and E. Hasman, “Photonic spin-controlled multifunctional shared-aperture antenna array,” Science 352(6290), 1202–1206 (2016). [CrossRef] [PubMed]
25. M. Juhl, C. Mendoza, J. P. B. Mueller, F. Capasso, and K. Leosson, “Performance characteristics of 4-port in-plane and out-of-plane in-line metasurface polarimeters,” Opt. Express 25(23), 28697–28709 (2017). [CrossRef]
26. B. Ratni, A. de Lustrac, G.-P. Piau, and S. N. Burokur, “Reconfigurable meta-mirror for wavefronts control: applications to microwave antennas,” Opt. Express 26(3), 2613–2624 (2018). [CrossRef] [PubMed]
28. B. A. Munk, Frequency selective surfaces: theory and design (Wiley, 2000).
29. D. M. Pozar, Microwave Engineering, 4th ed. (Wiley, 2012).
30. S. Kitagawa, R. Suga, K. Araki, and O. Hashimoto, “Active absorption/transmission FSS using diodes,” in IEEE International Symposium on Electromagnetic Compatibility (2015). [CrossRef]