Optically Transparent Single‐Layer Frequency‐Selective Surface Absorber for Dual‐Band Millimeter‐Wave Absorption and Low‐Infrared Emissivity

Current research on multispectral stealth material technology focuses on practical functionalities such as optical transparency, flexibility, and lightweight. Herein, an optically transparent and single‐layer frequency‐selective surface (OTSF) absorber for dual‐band millimeter‐wave (MMW) absorption and low‐infrared (IR) emissivity is proposed. By adopting indium tin oxide and polyethylene terephthalate, the proposed OTSF absorber exhibits good optical transparency (76% transmittance in the 400–800 nm range on an average) and flexibility. The OTSF absorber exhibits high absorption at the dual‐band frequency in the MMW range (99.7% at 35 GHz and 89.7% at 103 GHz). In the IR band, the average band emissivity of the OTSF absorber for the midwave IR (MWIR) band (3–8 μm) and the long‐wave IR (LWIR) band (8–15 μm) is measured to be 0.26 and 0.23, respectively. Thermal images of the OTSF absorber clearly show its low‐emission characteristics that are similar to those of metal. The proposed OTSF absorber structure has significant potential for practical applications of MMW–IR multispectral stealth materials.

DOI: 10.1002/adpr.202200009 Current research on multispectral stealth material technology focuses on practical functionalities such as optical transparency, flexibility, and lightweight. Herein, an optically transparent and single-layer frequency-selective surface (OTSF) absorber for dual-band millimeter-wave (MMW) absorption and low-infrared (IR) emissivity is proposed. By adopting indium tin oxide and polyethylene terephthalate, the proposed OTSF absorber exhibits good optical transparency (76% transmittance in the 400-800 nm range on an average) and flexibility. The OTSF absorber exhibits high absorption at the dual-band frequency in the MMW range (99.7% at 35 GHz and 89.7% at 103 GHz). In the IR band, the average band emissivity of the OTSF absorber for the midwave IR (MWIR) band (3-8 μm) and the long-wave IR (LWIR) band (8-15 μm) is measured to be 0.26 and 0.23, respectively. Thermal images of the OTSF absorber clearly show its low-emission characteristics that are similar to those of metal. The proposed OTSF absorber structure has significant potential for practical applications of MMW-IR multispectral stealth materials.
In the past few years, an optically transparent FSS absorber, which exhibits high transmission in the visible range, has attracted increasing attention, with the increase in the practical application in stealth technology. [23][24][25][26][27][28][29] Optical transparency is required for applications to the windows of vehicles or airplanes and the EM-shielding windows of a building. [23][24][25][26] To achieve optical transparency, indium tin oxide (ITO) and transparent dielectric materials, such as polyethylene terephthalate (PET), polyvinyl chloride (PVC), and colorless polyimide (CPI), have been utilized. However, ITO exhibits a relatively lower conductivity than metals (%1/100), leading to the poor absorption performance of ITO-based FSS absorber in resonance modes compared with a metal-based FSS absorber. [24,28] Furthermore, additional layers of the conventional multilayered FSS absorber reduce its optical transmittance in the visible wavelength range. Accordingly, the design of the ITO/PET-based bistealth FSS is more challenging in terms of the absorber thickness or FSS pattern design. Zhang et al. first suggested an absorber for broadband MMW absorption using a mesh-type ITO FSS pattern, PET, and PVC with a low filling ratio and IR reflectivity. [23] Multilayered FSS absorbers comprising ITO, PET, and other transparent substrates have been proposed to achieve low-IR emissivity. [21,[29][30][31][32][33][34] An optically transparent single-layer FSS (OTSF) absorber and a metasurface consisting of a high-filling-ratio ITO pattern and PET were also reported. [19,20] Although numerous studies have reported various OTSF absorbers with single-band microwave-IR bistealth performance, no study has reported the development of a single-layer multiband MMW-IR bistealth FSS absorber. A single-layered bistealth FSS absorber suggested by Shim et al. showed dual-band MMW absorption and selective IR stealth; however, still it has no optical transparency. [22] In this study, we present an OTSF absorber, which can achieve dual-band absorption in the MMW range and low emissivity in the IR range. ITO and PET were used to ensure the optical transparency of the fabricated FSS absorber. In addition, a high-fillingratio ITO pattern on the PET layer, which can transmit dual-band MMW (K a band and W band) and exhibit low-IR emissivity simultaneously, was designed. To enhance the MMW absorption performance of the OTSF absorber with low ITO conductivity, we proposed three types of unit cell designs of the OTSF absorber, calculated their MMW absorptivity, and determined the optimal design. The MMW measurement results revealed that the absorptivity of the optimal OTSF absorber was higher than 90% at 35 and 103 GHz, which was consistent with the calculation results and higher than the FSS pattern design proposed by a previous work. [22] Furthermore, in terms of the IR stealth performance, the Fourier-transform IR (FTIR) spectroscopic measurement result revealed that the average band emissivity of the absorber was 0.26 in the midwave IR (MWIR) band (3-8 μm) and 0.23 in the long-wave IR (LWIR) band (8-15 μm). In addition, thermal images of the designed OTSF absorber, PET, and ITO confirmed that our OTSF absorber effectively suppresses IR radiation. Finally, the measurement of the optical transmittance of the fabricated OTSF absorber revealed that the average transmittance of the absorber was 76% in the range of 400-800 nm. The dielectric constant of PET was obtained from a previous study. [35] First, the upper layer of the unit cell consists of square and rectangular conductor patches with two different side lengths, a and b, for transmitting the EM waves at dual-band resonant frequencies. The shorter side length, a, is mainly responsible for transmitting the EM waves at a higher resonant frequency, whereas the longer side length, b, is responsible for transmitting the EM waves at a lower resonant frequency.
To investigate the absorption mechanism of the proposed OTSF absorber structure, the magnetic field distributions of the absorber along the z-direction (side view) at the dual-band resonant frequencies are analyzed, and the results are shown in Figure 2. These dual resonant modes in the structure were reported and can be theoretically analyzed using effective resonant cavities and equivalent circuit analysis. [22] At a lower resonant frequency, a strong magnetic field is induced under the larger square patch, indicating that a resonant current is induced in the cavity constructed with the inductor and capacitor www.advancedsciencenews.com www.adpr-journal.com components of the larger square patch and back reflector ( Figure 2a). From the equivalent circuit of the effective cavities, the lower resonant frequency (ω r1 ) is given as follows.
where μ 0 and ε 0 are the permeability and permittivity of free space, respectively. Moreover, d and ε r are the thickness and relative permittivity of the dielectric layer, respectively. The typical value of the geometrical factor β is 0.23 for a plate. [36] The constant K is defined as C g,eq /βε 0 ε r W, where C g,eq is an equivalent gap capacitance among the square patches. The equivalent gap capacitance includes the gap capacitance induced between adjacent larger and smaller patches, adjacent smaller patches, and the larger and smaller patches separated by a þ 2g. W is the width of the patches. Definitions of the gap capacitance, detail analysis, and derivation are provided in Chapter 1 in the Supporting Information. At higher resonant frequencies, two kinds of strong magnetic fields are induced in divided cavities ( Figure 2b). These magnetic fields are induced by two kinds of resonant current with opposite directions: one is induced by the cavity constructed with the LC components of the larger square patch and back reflector and the other is induced by the coupled cavity constructed with the LC components of the smaller square patch, part of the larger square patch, and back reflector. Using the same analysis for the lower resonant frequency, the higher resonant frequency is given as where r is the ratio of the effective side length (l eff ) to the period without the gap distances (P 0 ) (i.e., r = l eff /P 0 ). The effective side length is defined as a part of the side length b, under which a strong magnetic field is induced with the LC components of the larger patch and back reflector and l eff is determined by balancing the impedances of the divided cavities (see Chapter 1 in the Supporting Information). The derived lower and higher resonant frequencies in Equations (1) and (2a) imply that the lower-resonant-frequency mode is predominantly affected by the value of b, and the higher-resonant-frequency mode is predominantly affected by P. Based on this, the resonant frequency of the proposed FSS absorber was easily predicted and compared with the simulated MMW absorption spectra optimized with different values of the design parameters b and P (Figure 3). A finite-difference timedomain (FDTD) software was utilized for the simulation (Lumerical, ANSYS). Boundary conditions for the xand y-axes were set as antisymmetric and symmetric, respectively, as the incident light is an s-polarized plane wave with a normal incident angle. On the z-axis, the bottom and top boundary were set as metal and the perfectly matched layer, respectively. Figure 3a shows the variation in the resonant frequency modes of the OTSF absorber with a change in the side length of the larger square, b, at a constant gap distance, g (0.05 mm), and unit cell period, P (2.25 mm). The value of b significantly affects the lowerresonant-frequency mode of the absorber. With an increase in the value of b, the absorption peak of the absorber at lower frequencies shifts toward lower frequency. In contrast, the value of b has no significant effect on the absorption peak at higher frequencies. Figure 3b shows the variation in the resonant frequency modes of the absorber with a change in P at a constant g (0.05 mm) and b (1.7 mm). With an increase in P, the absorption peak at higher frequencies shifts toward lower frequency; however, it has no significant effect on the absorption peak at lower frequencies. As shown in Figure 3, the analytic and simulated results are consistent with the FDTD simulation results. In addition to the squares with side length a or b in the unit cell, the rectangles which have both side length a and b may contribute to the dual resonant modes to a certain degree for each polarization of incident wave, though they are designed to fill the area among the squares.
A similar upper layer design was proposed by Shim et al., [22] which comprises a pair of metal (gold) square patches and a pair of rectangular patches. Compared with ITO, which was employed in this study, it exhibits a lower conductivity; thus, the absorption performance of the FSS absorber is expected to be significantly lower than that of the FSS absorber composed www.advancedsciencenews.com www.adpr-journal.com of a metallic upper layer. In contrast, the unit cell design proposed in this study includes more smaller patches with the side length a from which we can expect an enhanced higher-frequency resonance. To investigate the contribution of smaller patches on the higher-frequency resonance, we obtained simulated absorption spectra of the proposed OTSF absorber with three types of unit cell designs, as shown in Figure 4a.  Figure 4b. Despite optimization, %88% absorption is achieved at a higher resonant frequency for the unit cell design 1 with one smaller square patch. For the unit cell design 2 with four smaller square patches, the higher frequency absorption increases by 11.6% (to 99.7%)  www.advancedsciencenews.com www.adpr-journal.com and slightly increases to 99.9% for the unit cell design 3 with nine smaller square patches. There are slight increases in the lower resonant frequency (98.1%-99.9%) for the unit cell design variation. We can clearly notice the contribution of the smaller patches to the higher frequency resonance because the unit cell period P increases as more patches of side length a are included in the unit cell according to Equation (2a), with the fixed dimension of the larger square patch b.

Low-IR Emission of the Optically Transparent FSS Absorber
The ITO square patches on the top layer are arranged to exhibit a higher filling ratio than the dielectric layer at the top view of the structure, as shown in Figure 1b, to ensure the high reflectivity and low emissivity of the absorber in the IR range. The IR emissivity of the FSS absorber ε can be estimated as where ε ITO and f ITO are the emissivity and filling ratio of the ITO layer area at the top view of the FSS absorber, respectively, and ε d is the emissivity of the dielectric layer. ITO behaves like a metal in the IR band because of the negative value of the real part of its permittivity, and its emissivity in the IR band is %0.1. [29] In contrast, transparent dielectric materials, such as PET, exhibit a higher emissivity of %0.9. [37] According to Equation (1), the filling ratio and the estimated IR emissivity of the unit cell design 1 are 0.90 and 0.18, respectively. The filling ratio and the estimated IR emissivity of the unit cell designs 2 and 3 are 0.87 and 0.20 and 0.80 and 0.26, respectively, which can be attributed to the increased number of gaps between the square patches. At a constant P, because the filling ratio of the ITO pattern is constant, there is no change in the IR emissivity. However, the IR emissivity of the absorber decreases with an increase in P because the increase in dielectric area is larger than that in the ITO area at a constant b.
Considering both the MMW absorption performance and the IR emissivity of each unit cell design in terms of the filling ratio, the unit cell design 2 can be considered to be an optimal design.

Fabrication of the Optically Transparent FSS Absorber
To experimentally verify the performance of the proposed OTSF absorber, a transparent and flexible sample of dimensions 240 Â 240 mm 2 was fabricated. Figure 5a indicates that the fabricated sample exhibits excellent optical transparency. In addition, the thickness of the sample was significantly thin (%0.35 mm), and the sample exhibited good flexibility (Figure 5b).

MMW Absorption Measurement
To evaluate the multispectral stealth performance of the proposed OTSF absorber, the MMW absorption spectrum of the absorber was measured. The measured absorption spectra for each band are shown in Figure 6, and the results are consistent   with the calculated results. The measured absorptivity at lower and higher resonant frequencies (35 and 103 GHz) was %99.7% and 89.7%, respectively. The standard deviations of the measured absorptivity among the experiment sets at each frequency were 1.3% and 2.3%, respectively. Due to the considerable noise and deviation of the high-frequency signal for the measurement setup, the measured absorptivity of the higher frequency resonance was lower than the calculated one. Nevertheless, it was still higher than the calculated absorptivity for the unit cell design 1 (88.1%). The extra minor absorption peak is observed near 90 GHz, which is attributed to unwanted resonant modes that originated from the FSS patterns fabricated with different sizes.

Evaluation of the IR Stealth Performance
For the evaluation of the IR emission performance, a target OTSF absorber sample and reference metal plate of dimensions 35 Â 55 mm 2 were prepared, and the spectral reflectivity was measured first using an FTIR. The measured spectral emissivity of the OTSF sample in the entire IR band was 0.26 on an average in the MWIR band (3-8 μm) and 0.23 on an average in the LWIR band (8-15 μm), which are in good agreement with the values estimated using Equation (3) (Figure 7a). The slightly fluctuating absorption in the range of 5-7 μm can be attributed to the ITO material property. To further evaluate the IR stealth performance of the absorber, thermal IR images of the sample were obtained using an IR camera for the LWIR (8-14 μm) band. Figure 7b and c shows the measured thermal IR images of the samples on the hot plate at heating temperatures of 40 and 80°C, respectively. The stage surface of the hot plate was composed of a ceramic material (emissivity of 0.8-0.9); thus, it emitted similar energy to the blackbody radiation. In addition, the PET film exhibited a high radiant temperature due to its high emissivity (%0.9). In contrast, the measured radiant temperatures of the aluminum plate and ITO film were significantly low due to their low emissivity in the IR band (0.1-0.2). The radiant temperature of the OTSF absorber sample was similar to those of the bare ITO film and aluminum plate, indicating the low emissivity of the FSS absorber. This performance was observed under higher-temperature conditions, as shown in Figure 7c, where the radiant temperature of the OTSF absorber sample was 32.6°C and that of the hot plate was 72.1°C.

Optical Transmittance Measurement
Finally, the optical transparency of the proposed OTSF absorber sample was quantitatively characterized using an ultravioletvisible spectrophotometer (V-650, JASCO Corp.). Figure 8 shows the transmittance spectrum of the proposed OTSF absorber sample and bare ITO-PET film with the same thickness (%0.35 mm). The average optical transmittance of the FSS absorber sample and ITO-PET film in the range of 400-800 nm was %76% and reached 82% at a wavelength of %540 nm, indicating good optical transparency of the samples in the visible band. The samples were measured three times repetitively, and the standard deviations of the measurement of the ITO-PET film and FSS absorber sample were 0.5% and 1.1%, respectively.

Conclusion
In this study, we proposed an OTSF absorber for dual-band MMW absorption and low-IR emissivity. An ITO thin film was used as the top layer and back reflector of the OTSF absorber,  and a PET film was utilized as a dielectric spacer to achieve optical transparency in the proposed OTSF absorber. The top FSS layer was designed to exhibit dual-band frequency resonance by combining square patches and a high filling ratio to achieve high reflectivity (low emissivity) in the IR band. We investigated the contribution of the small patches to the higher-frequency resonance and suggested a design rule to enhance the absorption performance for ITO with low conductivity. An OTSF absorber sample of dimensions 240 Â 240 mm 2 was fabricated using the laser etching method. The fabricated sample exhibited a very thin structure (%0.35 mm) with excellent optical transparency and flexibility. Furthermore, the measured MMW absorptivity of the as-fabricated OTSF absorber was consistent with the calculation results, with an absorption of 99.7% at a lower resonant frequency and 89.7% at a higher resonant frequency. In addition, the measured average spectral emissivity of the as-fabricated OTSF absorber in the MWIR and LWIR bands using FTIR spectroscopy was 0.26 and 0.23, respectively. Furthermore, the thermal IR image of the OTSF absorber revealed where the IR stealth performance and radiant temperature of the absorber were similar to those of the metal and bare ITO film. In addition, the average optical transmittance of the OTSF absorber measured in the 400-800 nm wavelength range was 76%. We believe that the findings of this study based on the proposed OTSF absorber will provide useful insights into the practical development of multispectral stealth and shielding materials applied to the windows of vehicles and buildings.

Experimental Section
Fabrication: For the fabrication of the proposed OTSF absorber sample, an ITO-deposited PET film (%0.175 mm thickness, 240 Â 240 mm 2 ) was prepared, and the FSS pattern on the top ITO layer was fabricated via laser etching. The fabricated ITO-PET film was combined with another ITO-PET film with the same thickness using an adhesive tape.
MMW Measurement: For the MMW absorption measurement of the proposed OTSF absorber, a horn antenna (Rx/Tx) and a vector network analyzer (Agilent E8363B) were used in an anechoic chamber. For the analysis, an OTSF absorber sample of dimensions 100 Â 100 mm 2 was prepared. Two types of antennas were used for the analysis: one was used for the K a -band (26-40 GHz) measurement (SGH-28, MTG) and the other was used for the W-band (75-110 GHz) measurement (HO10R, Custom Microwave Inc.). The monostatic setup was used in this study and the S 11 parameter (reflectivity in dB unit) was measured at normal incidence. The absorptivity of the OTSF absorber was calculated by subtracting the reflectivity from unity under the assumption that there is no transmission. An aluminum plate with the same dimension as the target OTSF absorber sample was used for reference measurement, and the time-gating method was used to accurately measure the absorptivity of the reflected MMW. Because of the frequency coverage of each horn antenna, the measurement was not performed in the frequency band range from 40 to 75 GHz.
IR Stealth Performance Evaluation: For the spectral measurement of the proposed OTSF absorber in IR band, a target sample and reference metal plate of dimensions 35 Â 55 mm 2 were prepared, and the spectral reflectivity was measured using an FTIR spectrometer (Nicolet iN10MX, Thermofisher Scientific).
For the IR image acquisition of the proposed OTSF absorber, IR camera for the LWIR (8-14 μm) band was used (TE-SQ1, i3 System). For this, the target FSS absorber, bare ITO film, and PET film of dimensions 35 Â 55 mm 2 and an aluminum plate of dimensions 100 Â 100 mm 2 were prepared. The aluminum plate was placed on a hot plate system (HP120D, Misung Scientific) to control the surface temperature, after which the target samples were placed on the aluminum plate. Thermal IR images were taken at the heating temperatures of 40 and 80°C. The emissivity was set at unity during the measurements.

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.