Open Access
Add to BibSonomy
Abstract
An optical transparent metasurface for dual-band Wi-Fi shielding is presented in this paper. The unit cell of the proposed metasurface is composed of a hexagonal ring and a three-petal oval flower which resonate at 2.4 and 5.5 GHz, respectively. The corresponding equivalent circuit is modelled to better understand the physical phenomena of electromagnetic shielding. Based on transmission line theory and curve fitting technique, a convenient and efficient method for extracting permittivity of substrate is presented. Simulation results show that the proposed metasurface is insensitive to the polarization of incoming wave under normal incidence and offers excellent angular stability. For verifying the design, two prototypes are fabricated using different manufacturing technologies, flexible printed circuit and ink-jet printing of silver nano-particles. The measured results are in good agreement with the simulated ones. The proposed metasurface has potential applications of electromagnetic wave suppression and information security in indoor environments.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Wireless communication plays a significant role in day-to-day life and wireless technology has become an indispensable part of our daily activities. Compare with wired communication, wireless communication has numerous advantages including mobility, reliability, ease of installation, etc. Various wireless devices, such as mobile phones, GPS units, wireless computer parts, cordless telephones, and satellite televisions, are used for wireless communication. However, the large number of wireless devices working in proximity to each other cause electromagnetic (EM) interference (EMI), which easily affects the function of electronic devices. Therefore, different standards for the maximum EM radiation and spurious emission for each device have been released [1]. In some specific environments, such as special research centers, hospitals, and military installations, EM shielding is necessary. In the past few decades, many metasurfaces and frequency-selective surfaces (FSSs) have been proposed for different applications including radar absorber [2,3], EM energy harvesting and wireless power transfer [4], and EM shielding [5–13].
In recent years, Wi-Fi signal has become one of main contributions to EM pollution after numerous wireless access points (WAPs) have been installed to broadcast Wi-Fi signals for clients. For concrete buildings, the main considerations of Wi-Fi shielding are windows and glass doors, because concrete wall has high attenuation characteristic to EM waves. Driven by demand, researchers are striving to develop new optical-transparent materials for electromagnetic shielding. Several novel fabrication techniques, such as indium tin oxide (ITO) sheet with conductive patterns [8,9], screen or 3D printing with conductive ink [10,11], graphene-based FSS [12,13], and active FSS [14], are applied. To realize better shielding effect, metasurfaces which have characteristics of polarization-insensitive, stable performance under wide angle of incidence, and operating in multiband or wideband, are desirable. Besides the shielding effect, two other important considerations in designing metasurfaces for shielding windows are optical transparency and aesthetics.
A metasurface, which has characteristics of EM shielding in dual-band Wi-Fi frequencies but transparent for optical and microwave signals outside the shielding bands, is presented in this paper. The unit cell is designed with a three-petal oval flower surrounded with a hexagonal ring. Two manufacturing technologies, one of which is flexible printed circuit (FPC) [15], and another is ink-jet printing of silver nano-particles (Ag NPs) [16], are applied to fabricate two prototypes. Measured results show that both prototypes have characteristics of high transparency and good angular stability. The contributions of this work include, (i) Taking into account the requirements of aesthetics, optical transparency, high angular stability, etc., a metasurface for dual-band Wi-Fi shielding is designed. (ii) Using ink-jet printing technology, conductive ink of silver nano-particles is directly printed on the glass surface. (iii) A convenient and efficient method for extracting the permittivity of substrate is presented based on the transmission line theory and curve fitting technique.
The rest of this paper is arranged as follows. In Section 2, the structure and the equivalent circuit model (ECM) are presented firstly, and performances of metasurface for single and double-paned windows are analyzed. Additionally, method for extracting the electromagnetic characteristics of materials is briefly discussed. In Section 3, the measurement system and method, measured results are provided and analyzed. At last, conclusions are drawn.
2. Design and analysis of the metasurface
2.1 Structure
For the transparent window and door applications, transparency and aesthetics are two important considerations in design. To realize dual stopbands at 2.4 and 5.5 GHz, two separated metallic resonant structures, one of which is a hexagonal ring and the other is a three-petal oval flower, are configured, as shown in Figs. 1(a-d). The periodic outer ring resonates at 2.4 GHz, while the array of inner flower shape resonates at 5.5 GHz, respectively. The three-petal oval flower consists of three elliptic rings connected by a small ring in the center. The rotationally symmetrical structure provides not only aesthetic effect but also the characteristic of polarization insensitive. Substrate for supporting the array can be chosen from different types of glass and polymethyl methacrylate (PMMA) because they are common materials for transparent windows or doors.
Fig. 1. The structure of metasurface, (a) planar periodic array, (b) top view of the array, (c) geometry of unit cell, (d) top view of unit cell.
When designing of metasurface which has two stopbands with a large frequency interval, one intractable problem is the onset of grating lobes, especially under oblique incidence. The grating lobes are often excited at high frequency, and mostly determined by the incident angle and the space of two adjacent cells. Consequently, the grating lobes might appear at the second stopband and significantly affect the angle stability, if the unit cell and the space between cells are inappropriately designed and arranged.
In Figs. 2(a-c), we compare the performance of three conventional unit cells, circular ring, square ring and hexagonal ring, which are often used for bandstop applications. For fairness, same minimum space between adjacent rings and same width of ring, namely, S1 = 0.4 mm and W1 = 0.4 mm, are chosen. Substrates for all structures are PMMA, whose thickness, the permittivity and the loss tangent are h1 = 3 mm, ɛr = 2.7 and tanδ = 0.005, respectively.
Fig. 2. Results of different structures under oblique incidence, (a) ring, (b) square, (c) hexagonal.
Figure 2(a) shows the simulated results of circular rings arranged in a square array, under the normal incidence and the oblique incidence of 45°. The cell size is Dx = 21.1 mm and the space between two cells is S1 = 0.4 mm. The frequency of fundamental mode is 2.4 GHz under the normal incidence, and grating lobes appear at 5.44 GHz for the parallel polarization (TM) and 6 GHz for the perpendicular polarization (TE), both of which locate around the second stopband. Figure 2(b) shows the results of square ring array. The cell size Dx = 14.8 mm is selected to generate the fundamental mode of same frequency at 2.4 GHz. In this case, grating lobes similarly appear at 5.74 GHz for the parallel polarization and 6.6 GHz for the perpendicular polarization. Figure 2(c) shows the results of the hexagonal ring structure with honeycomb arrangement. Within the band from 1 to 7 GHz, only one grating lobe shows at 6.58 GHz for the perpendicular polarization under the oblique incidence of 45°. The side length of the hexagon is Dx = 12.2 mm.
The results shown in Figs. 2(a-c) illustrate that both circular ring and square ring for generating the first stopband excite grating lobes at the second stopband under the oblique incidence. No matter which structure is designed to produce the second stopband, the grating lobes will significantly affect the angular stability. Among these three conventional structures, the hexagonal ring has only one grating lobe appears at the frequency much higher than the second stop band under the perpendicular polarization. According to the results, the hexagonal ring is chosen for the first stopband and the array is arranged in a honeycomb structure.
2.2 Equivalent circuit model
The equivalent circuit model is shown in Fig. 3(a), where Z0 = 120π Ω is the characteristic impedance of free space, and ${Z_\textrm{B}}\textrm{ = }{{{Z_0}} / {\sqrt {{\varepsilon _{1r}}} }}$ is the equivalent impedance of the substrate under the normal incidence, here ɛ1r is the effective permittivity of the substrate. The periodic hexagonal ring can be equivalent to a series resonant circuit of L1 and C1, where L1 is the equivalent inductance of the hexagonal ring itself, and C1 represents the coupling capacitance between two adjacent rings. Similarly, L2 denotes the equivalent inductance of the flower shape resonator, and C2 represents the coupling capacitance between two flower resonators. The equivalent admittances of two series circuits are Y1 and Y2, respectively. The presence of two series resonant circuits means that there are two stopband frequencies, one of which is mainly determined by L1 and C1, and the other is depended on L2 and C2, namely, ${f_1}\textrm{ = }{1 / {\left( {2\pi \sqrt {{L_1}{C_1}} } \right)}}$ and ${f_2}\textrm{ = }{1 / {\left( {2\pi \sqrt {{L_2}{C_2}} } \right)}}$. There are many combinations of L1 and C1 (L2 and C2) that can resonate at f1 (f2). Nevertheless, the combination is deterministic and unique if both the resonant frequency and the bandwidth are considered.
According to microwave network analysis [17], the transmission coefficient represented by S21 can be derived and written as,
Figure 3(b) compares the simulation results obtained from the ECM and HFSS. The two curves almost overlap in the band of 1 - 7 GHz, which indicates that an exact ECM has been modelled. The values of the components are L1 = 13.62 nH, C1 = 0.317 pF, L2 = 5.34 nH, and C2 = 0.157 pF, correspond to the geometric dimensions are, Dx = 12.2 mm, Lma = 8.6 mm, Lmi = 4 mm, Din = 2 mm, W1 = 0.4 mm, and S1 = 0.4 mm, respectively.
2.3 Electromagnetic characteristics of PMMA
Polymethyl methacrylate is often used as a lightweight and shatter-resistant replacement for regular glass because of its high transparency up to 92%. Unfortunately, for most polymers, the published data of the permittivity and the dielectric loss at microwave frequencies in literature are not available. Therefore, the electromagnetic parameters of PMMA should be tested firstly if we choose it as the substrate. Different methods, such as resonant cavity approach [18], capacitive method [19], transmission line [20], free-space method [21], and coaxial probe [22], have been presented to measure the complex permittivity and complex permeability of materials. Each technique has its pros and cons, and no single technique can accurately characterize all materials over wide frequency band.
In this article, we established a convenient and efficient method to extract the permittivity of PMMA, based on the transmission line theory and curve fitting technique. The procedure of the method is as follows. Firstly, scattering parameters of a practical microstrip line fabricated on the PMMA substrate are measured using a two-port network analyzer. Secondly, the microstrip and two SMA connectors are modelled and simulated using HFSS. The permittivity ${\varepsilon _r}(\omega )$ of the substrate is set to be frequency-dependent parameter which can be swept during simulation. Then, simulate the model by tuning ${\varepsilon _r}(\omega )$ to obtain the simulated result which agrees with the measured result. Figure 4(a) compares the simulated |S11| of both constant and frequency-dependent permittivity with the measured result. Figure 4(b) shows the extracted ${\varepsilon _r}(\omega )$ of PMMA and borosilicate glass which are used in this paper. The relative permittivity of the two materials reduce slightly as the frequency increasing.
Fig. 4. Extraction of the permittivity of PMMA, (a) |S11| of different ɛr, (b) frequency-dependent permittivity for PMMA and borosilicate glass.
2.4 Performance analysis and angular stability
As the above discussion, the outer ring resonates at 2.4 GHz and the inner flower shape resonates at 5.5 GHz, which imply that two stopbands can be independently altered by changing the dimension of the related structures. Figure 5(a) shows the simulated |S21| of different S1 and same Dx, which means different sizes of the hexagonal ring are used. The first resonant frequency increases from 2.1 to 2.6 GHz as S1 increases from 0.2 to 0.6 mm. If Lma is increased from 8.3 to 9.0 mm, the second resonant frequency is decreased from 5.9 to 5.25 GHz, as shown in Fig. 5(b). Slightly variations appear at the higher stopband in Fig. 5(a) and at the lower stopband in Fig. 5(b) mean that there is small mutual coupling between two resonant structures. Other sizes will also affect the resonant frequencies, yet the effects are not as significant as S1 and Lma.
Due to the rotate-symmetric structure, the transmission coefficients for the parallel and the perpendicular polarizations are identical under the normal incidence. However, they are different under the oblique incidence. According to EM field theory [17,23], the expressions of the S21 in terms of the electrical parameters, thickness of the substrate (h) and angle of incidence (θi), can be expressed as,
The angular frequency and wavenumber of air and substrate are defined as ω, k0 and k1, respectively. For perpendicular polarization,
The expresses from Eq. (2) to Eq. (9) imply that the transmission coefficients are different for two polarizations under the oblique incidence.
To measure the shielding performance, shielding effectiveness (SE) is often used and it is defined as the ratio of the magnitude of incident field (Ei) to the transmitted field (Et) [1], namely,
Figure 6 shows the simulated SE of two polarizations under the oblique incidence. The results for the angle of 0°, 45°, 60° and 75° are plotted. It is evident that the bandwidth of stopbands is expanded for the perpendicular polarization and narrowed for the parallel polarization, and the resonant frequencies shift slightly as the incident angle is increased up to 75°. For the perpendicular polarization, grating lobes do not greatly affect the second stopband because they appear at around 7 GHz. Figure 6 also exhibits that the stopbands of the oblique incidences have small deviations from that of the normal incidence. The reason can be explained from Eq. (1) and (2), in which different denominators lead to different transmission zeros.
Fig. 6. Simulated SE under the oblique incidence, (a) perpendicular polarization, (b) parallel polarization.
Figure 7 exhibits the field distributions for different frequencies and different incidences. For the normal incidence at 2.4 and 5.5 GHz, the incident wave from the top is almost fully reflected backward, and nearly no wave transmits through the metasurface. While at the passband around 3.5 GHz, most of the incoming field passes through it and transmits to the area underneath. Under the oblique incidence, specular reflections are observed for 2.4 and 5.5 GHz, and refraction is seen for 3.5 GHz.
Fig. 7. Field distribution at different frequencies, (a) - (c) normal incidence, (d) - (f) oblique incidence of 60° (φ = 0° incident plane, perpendicular polarization, ki is the incident direction, kr is the reflection direction, and kt is the transmission direction).
The incidence angle dependence of the transmission spectrums on both xoz (φ = 0°) and yoz (φ = 90°) incident planes are simulated and depicted in Figs. 8(a-d). It can be observed that at the frequencies close to the two stopbands, the transmission is always small for both polarizations, as the incident angle increases. The bandwidths of both stopbands are expanded with the increasing of the incident angle for the perpendicular polarization, while they are narrowed for the parallel polarization. At the frequencies lower than 6.4 GHz, almost same spectrums are obtained for two incident planes of φ = 0° and 90°.
Fig. 8. Incidence angle dependence of the simulated transmission spectrums, (a) perpendicular polarization (φ = 0°), (b) parallel polarization (φ = 0°), (c) perpendicular polarization (φ = 90°), (d) parallel polarization (φ = 90°).
2.5 Metasurface for double-paned windows
Double-paned windows are often used in cold areas because they have a small space between two glass panes that creates an air pocket to keep the outside elements from affecting the temperature inside of rooms. Figures 9(a-b) show the configuration and its equivalent circuit which is actually a second-order filter. The capacitors, C11 and C22, represent the small coupling between two metasurfaces for the lower and the higher bands, respectively. They are regulated by the space between MS-1 and MS-2, namely, 2h1 + h2. Figure 9(c) shows the SE results of L1 = L'1 = 13.62 nH, C1 = C'1 = 0.317 pF, L2 = L'2 = 5.34 nH, C2 = C'2 = 0.157 pF, and different C11 and C22. The results imply that a tiny value of C22 (0.1 or 0.3 fF) will split the main lobe of the higher stopband into two peaks and do not significantly affect the lower stopband. Figure 9(d) illustrates the SE results of C11 = C22 = 0.1 fF, C1 = 0.317 pF and different C'1. The results exhibit that two peaks at the lower stopband can be obtained by choosing different C1 and C'1, which are affected by the gap between unit cells.
Fig. 9. Metasurface for double-paned windows, (a) structure, (b) equivalent circuit, (c) SE results of different coupling capacitance C11&C22, (d) SE results of different C'1.
Figures 10(a-d) show the simulated SE results with different thicknesses of air gap and different dimensions of metasurface. Figure 10(a) is the SE results under the normal incidence. Same dimensions (Lma = 8.6 mm, Lmi = 3 mm, Din = 2 mm, W1 = 0.4 mm, S1 = 0.4 mm and h1 = 3 mm) are used for MS-1 and MS-2 and different thicknesses of air gap (h2) are considered. The maximum SE values are higher than 40 and 50 dB at the lower and the higher stopbands, respectively. The results verify that the coupling between two metasurfaces results in the splitting of the main lobe at the higher stopband, which is consistent with the results shown in Fig. 9(c). Figure 10(b) shows the SE results of different configurations, S1 = 0.3 mm for MS-1 and 0.5 mm for MS-2, and identical Lma = 8.8 mm for both metasurfaces. The main lobe of the lower stopband splits into two peaks, comparing with the results shown in Fig. 10(a). The SE results in Figs. 10(a-b) demonstrate that the space between unit cells (S1) affects the main lobe of the lower band. Figures 10(c-d) are the SE results of h2 = 9 mm, S1 = 0.3 mm for MS-1 and 0.5 mm for MS-2 under the oblique incidence. The results exhibit that the SEs increase for the perpendicular polarization but decrease for the parallel polarization as the incident angle increasing.
Fig. 10. Simulated SE results of double-paned windows, (a) same dimensions for metasurfaces with different air gaps, (b) different configurations for metasurfaces under the normal incidence, (c) perpendicular polarization under the oblique incidences, (d) parallel polarization under the oblique incidences.
3. Experimental results and discussion
3.1 Measurement system and method
Figure 11 shows the measurement setup in an anechoic chamber. It consists of two wideband antennas that can be mechanically rotated and a fixture with an aperture to stabilize the fabricated prototype. The two antennas were connected to a vector network analyzer (VNA, Keysight N5243A). The prototype was placed between two antennas and was surrounded with the commercial absorbent foam for reducing the effect of the edge diffraction.
Measurements were carried out as follows [24,25].
- (1) Measure the frequency domain transmission coefficient $S_{21}^B(f )$ of the fixture without prototype in the anechoic chamber.
- (2) Apply the inverse fast Fourier transform (IFFT) to convert $S_{21}^B(f )$ to time domain transmission coefficient $S_{21}^B(t )$, and the time gating was activated to cancel the interference.
- (3) Measure $S_{21}^{MS}(f )$ of the prototype installed in the aperture.
- (4) Convert $S_{21}^{MS}(f )$ to time domain $S_{21}^{MS}(t )$ and apply the time gating.
- (5) Apply the fast Fourier transform (FFT) for the background-subtracted data, $S_{21}^{MS}(t )- S_{21}^B(t )$, to return back to the frequency domain ${S_{21}}(f )$.
3.2 Prototypes
Two prototypes were fabricated using two different technologies. Figure 12(a) shows the first prototype which was realized using FPC technology. In view of the metallic periodic structure cannot be directly printed on the surface of PMMA by using conventional printed circuit technology due to the mechanical and thermal properties, the resonant unit cells were firstly etched on a flexible polyethylene terephthalate (PET) sheet using FPC technology. After that, the PET sheet was adhered to the surface of PMMA with transparent double-sided tape. The dimensions need to be optimized because the PET affects the resonant frequency. Finally, Dx = 11.2 mm, Lma = 8 mm and Lmi = 2.6 mm are chosen and the thickness and permittivity of PET are 0.15 mm and 3.5, respectively.
Fig. 12. Prototypes, (a) conductive pattern etched on PET sheet, (b) conductive pattern printed on high borosilicate glass.
Another prototype was realized by ink-jet printing Ag NPs ink on the surface of high borosilicate glass. Because of their high electric conductivity and good oxidation resistance, Ag NPs hold a unique position when making high performance conductive ink among the conductive materials [26]. The Ag NP-based conductive inks were formulated by adding the concentrated Ag NP into a certain amount of ethylene glycol, isopropanol, and 2-butoxy ethanol mixture with the load of 35 to 40 wt.%. The radii of Ag NPs range from 30 to 50 nm. The printed sample was sintered at 150 °C for 30 minutes to obtain a sheet resistance of about 1 ∼ 2 mΩ/sq. Figure 12(b) shows the printed sample. The frequency-dependent permittivity of the high borosilicate glass shown in Fig. 4(b) is used to optimize the unit cell. The dimensions for acquiring two stopbands are Dx = 10.2 mm, Lma = 7.2 mm, Lmi = 2.6 mm and W1 = 0.2 mm, respectively.
3.3 Results and discussions
In Fig. 13 and Fig. 14, the measured SE results for the normal and the oblique incidence up to 75° are presented. Figure 13 is the SE results of the first prototype. It is evident that the signal will be reduced by about 20 to 25 dB in both stopbands. Figure 14 shows that the SE results of the second prototype. The maximum SE is larger than 30 dB at the lower stopband under the normal incidence. Both SE results exhibit that the bandwidth expands for the perpendicular polarization and narrows for the parallel polarization, and the stopbands are stable under the oblique incidence.
Fig. 13. Measurement SE results of metasurface on PMMA, (a) perpendicular polarization, (b) parallel polarization.
Fig. 14. Measurement results of metasurface on high borosilicate glass, (a) perpendicular polarization, (b) parallel polarization.
Figures 15(a-b) compare the measured and the simulated SE results. In Fig. 15(a), the magenta line is the result of PMMA with a constant relative permittivity of 2.7, and the blue line is the result of frequency-dependent permittivity shown in Fig. 4(b). It can be seen by comparison that the result obtained from frequency-dependent permittivity is much closer to the measured result. Some deviations between them are mainly attributed to the used transparent double-sided tape, whose thickness and permittivity change with the pressure of the lamination. Figure 15(b) exhibits good agreement between two results for the second prototype. It should be mentioned that frequency-dependent permittivity is used for obtaining the simulation result in Fig. 15(b).
Fig. 15. Measurement and simulation SE results under the normal incidence, (a) PMMA, (b) high borosilicate glass.
The comparison of the proposed metasurface with various closely related transparent designs in literature is presented in Table 1. Our works realize the highest angular stability and high transparency. Due to the presence of copper wires and nano-silver wires, the light transparency is varied at different positions on the surface. In order to obtain a relatively accurate results, we randomly select different positions to test the transmittance, including the positions with the largest and smallest transparency, then take their average value. The values of VLT shown in Fig. 12 are the measured light transparency. The average light transparency of 82.4% has been realized by ink-jet printing Ag NPs ink on the surface of glass. Comparing our design with [5], both of which realize high angular stability and high transparency, the maximum SE of this work is much higher.
Table 1. Comparison of the Proposed Metasurface with Closely Related Designs in Literature
4. Conclusion
Both transparency and aesthetics are considered in designing the presented metasurface for Wi-Fi shielding at the windows or doors. The rotationally symmetrical structure of periodic hexagonal rings and three-petal flowers are arranged to provide not only the aesthetic effects but also the characteristics of polarization insensitive and angular stability under the oblique incidence. For optical transparency, the FPC technology is used to etch the periodic pattern on a PET sheet, which is adhered to the surface of PMMA. Ink-jet printing Ag NPs ink on the surface of high borosilicate glass is also applied to fabricate the metasurface. Simulated and measured results show that the presented metasurface realizes two stopbands for Wi-Fi shielding at 2.4 and 5.5 GHz with high SE feature.
Funding
National Natural Science Foundation of China (62071435, 62071436).
Disclosures
The authors declare that there are no conflicts of interest related to this article.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. S. Celozzi, R. Araneo, and G. Lovat, Electromagnetic Shielding (John Wiley & Sons, 2008).
2. M. Rahmanzadeh, H. Rajabalipanah, and A. Abdolali, “Analytical investigation of ultrabroadband plasma-graphene radar absorbing structures,” IEEE Trans. Plasma Sci. 45(6), 945–954 (2017). [CrossRef]
3. A. G. D’Aloia, M. D’Amore, and M. S. Sarto, “Adaptive broadband radar absorber based on tunable graphene,” IEEE Trans. Antennas Propag. 64(6), 2527–2531 (2016). [CrossRef]
4. E. Faruk and O. M. Ramahi, “A scalable, dual-band absorber surface for electromagnetic energy harvesting and wireless power transfer,” IEEE Trans. Antennas Propag. 69(10), 6982–6987 (2021). [CrossRef]
5. Y. Yang, W. Li, K. N. Salama, and A. Shamim, “Polarization insensitive and transparent frequency selective surface for dual band GSM shielding,” IEEE Trans. Antennas Propag. 69(5), 2779–2789 (2021). [CrossRef]
6. R. B. Green, M. Guzman, N. Izyumskaya, B. Ullah, S. Hia, J. Pitchford, R. Timsina, V. Avrutin, Ü. Özgür, H. Morkoç, N. Dhar, and E. Topsaka, “Optically transparent antennas and filters: A smart city concept to alleviate infrastructure and network capacity challenges,” IEEE Antennas Propag. Mag. 61(3), 37–47 (2019). [CrossRef]
7. S. K. Sharma, D. Zhou, A. Luttgen, and C. D. Sarris, “A micro copper mesh-based optically transparent triple-band frequency selective surface,” Antennas Wirel. Propag. Lett. 18(1), 202–206 (2019). [CrossRef]
8. H. Jiang, W. Yang, S. W. Lei, H. Q. Hu, B. Chen, Y. F. Bao, and Z. Y. He, “Transparent and ultra-wideband metamaterial absorber using coupled hexagonal combined elements,” Opt. Express 29(18), 29439–29448 (2021). [CrossRef]
9. T. S. Li, K. Chen, G. W. Ding, J. M. Zhao, T. Jiang, and Y. J. Feng, “Optically transparent FSS Salisbury screen with wideband microwave absorption,” Opt. Express 26(26), 34384–34395 (2018). [CrossRef]
10. A. A. Dewani, S. G. O’Keefe, D. V. Thiel, and A. Galehdar, “Window RF shielding film using printed FSS,” IEEE Trans. Antennas Propag. 66(2), 790–796 (2018). [CrossRef]
11. Y. Tayde, M. Saikia, K. V. Srivastava, and S. A. Ramakrishna, “Polarization-insensitive broadband multilayered absorber using screen printed patterns of resistive ink,” Antennas Wirel. Propag. Lett. 17(12), 2489–2493 (2018). [CrossRef]
12. W. B. Lu, J. W. Wang, J. Zhang, Z. G. Liu, H. Chen, W. J. Song, and Z. H. Jiang, “Flexible and optically transparent microwave absorber with wide bandwidth based on graphene,” Carbon 152, 70–76 (2019). [CrossRef]
13. J. Zhang, Z. F. Li, L. D. Shao, and W. R. Zhu, “Dynamical absorption manipulation in a graphene-based optically transparent and flexible metasurface,” Carbon 176, 374–382 (2021). [CrossRef]
14. J. F. Yang, J. Chen, L. Quan, Z. Z. Zhao, H. Y. Shi, and Y. J. Liu, “Metamaterial-inspired optically transparent active dual-band frequency selective surface with independent wideband tunability,” Opt. Express 29(17), 27542–27553 (2021). [CrossRef]
15. T. Tian, X. J. Huang, K. Cheng, Y. L. Liang, S. R. Hu, L. X. Yao, D. F. Guan, Y. L. Xu, and P. G. Liu, “Flexible and reconfigurable frequency selective surface with wide angular stability fabricated with additive manufacturing procedure,” Antennas Wirel. Propag. Lett. 19(12), 2428–2432 (2020). [CrossRef]
16. X. F. Zhang, Z. G. Liu, W. Shen, and S. Gurunathan, “Silver nanoparticles: Synthesis, characterization, properties, applications, and therapeutic approaches,” Int. J. Mol. Sci. 17(9), 1534–1568 (2016). [CrossRef]
17. D. M. Pozar, Microwave Engineering (John Wiley & Sons, 2009).
18. M. S. Ramachandraiah and M. C. Decreton, “A resonant cavity approach for the nondestructive determination of complex permittivity at microwave frequencies,” IEEE Trans. Instrum. Meas. 24(4), 287–291 (1975). [CrossRef]
19. C. J. Mathai, S. Saravanan, M. R. Anantharaman, S. Venkitachalam, and S. Jayalekshmi, “Characterization of low dielectric constant polyaniline thin film synthesized by ac plasma polymerization technique,” J. Phys. D: Appl. Phys. 35(3), 240–245 (2002). [CrossRef]
20. P. M. Narayanan, “Microstrip transmission line method for broadband permittivity measurement of dielectric substrates,” IEEE Trans. Microwave Theory Tech. 62(11), 2784–2790 (2014). [CrossRef]
21. D. K. Ghodgaonkar, V. V. Varadan, and V. K. Varadan, “Free-space measurement of complex permittivity and complex permeability of magnetic materials at microwave frequencies,” IEEE Trans. Instrum. Meas. 39(2), 387–394 (1990). [CrossRef]
22. D. V. Blackham and R. D. Pollard, “An improved technique for permittivity measurements using a coaxial probe,” IEEE Trans. Instrum. Meas. 46(5), 1093–1099 (1997). [CrossRef]
23. C. A. Balanis, Advanced Engineering Electromagnetics (John Wiley & Sons, 2012).
24. C. A. Grosvenor, R. T. Johnk, J. Baker-Jarvis, M. D. Janezic, and B. Riddle, “Time-domain free-field measurements of the relative permittivity of building materials,” IEEE Trans. Instrum. Meas. 58(7), 2275–2282 (2009). [CrossRef]
25. Z. Akhter and M. J. Akhtar, “Free-space time domain position insensitive technique for simultaneous measurement of complex permittivity and thickness of lossy dielectric samples,” IEEE Trans. Instrum. Meas. 65(10), 2394–2405 (2016). [CrossRef]
26. L. X. Mo, Z. X. Guo, L. Yang, Q. Q. Zhang, Y. Fang, Z. Q. Xin, Z. Chen, K. Hu, L. Han, and L. H. Li, “Silver nanoparticles based ink with moderate sintering in flexible and printed electronics,” Int. J. Mol. Sci. 20(9), 2124–2152 (2019). [CrossRef]
27. U. Farooq, M. F. Shafique, and M. J. Mughal, “Polarization insensitive dual band frequency selective surface for RF shielding through glass windows,” IEEE Trans. Electromagn. Compat. 62(1), 93–100 (2020). [CrossRef]
28. F. Bagci, C. Mulazimoglu, S. Can, E. Karakaya, A. E. Yilmaz, and B. Akaoglu, “A glass based dual band frequency selective surface for protecting systems against WLAN signals,” Int. J. Electron. Commun. (AEU) 82, 426–434 (2017). [CrossRef]
