Electromagnetic characteristics of carbon nanotube

Chinese Journal of Aeronautics, (2015), 28(4): 1245–1254

Chinese Society of Aeronautics and Astronautics & Beihang University

Chinese Journal of Aeronautics [email protected] www.sciencedirect.com

Electromagnetic characteristics of carbon nanotube film materials Zhang Wei a, Xiong Huagang a, Wang Shaokai a b

b,*

, Li Min b, Gu Yizhuo

b

School of Electronic and Information Engineering, Beihang University, Beijing 100191, China School of Materials Science and Engineering, Beihang University, Beijing 100191, China

Received 20 January 2015; revised 11 February 2015; accepted 1 April 2015 Available online 13 May 2015

KEYWORDS Carbon nanotube; Permeability; Permittivity; Shielding effectiveness; Waveguide

Abstract Carbon nanotube (CNT) possesses remarkable electrical conductivity, which shows great potential for the application as electromagnetic shielding material. This paper aims to characterize the electromagnetic parameters of a high CNT loading film by using waveguide method. The effects of layer number of CNT laminate, CNT alignment and resin impregnation on the electromagnetic characteristics were analyzed. It is shown that CNT film exhibits anisotropic electromagnetic characteristic. Pristine CNT film shows higher real part of complex permittivity, conductivity and shielding effectiveness when the polarized direction of incident wave is perpendicular to the winding direction of CNT film. For the CNT film laminates, complex permittivity increases with increasing layer number, and correspondingly, shielding effectiveness decreases. The five-layer CNT film shows extraordinary shielding performance with shielding effectiveness ranging from 67 dB to 78 dB in X-band. Stretching process induces the alignment of CNTs. When aligned direction of CNTs is parallel to the electric field, CNT film shows negative permittivity and higher conductivity. Moreover, resin impregnation into CNT film leads to the decrease of conductivity and shielding effectiveness. This research will contribute to the structural design for the application of CNT film as electromagnetic shielding materials. ª 2015 The Authors. Production and hosting by Elsevier Ltd. on behalf of CSAA & BUAA. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction Carbon nanotube (CNT) possesses remarkable mechanical and physical performances, which has a tensile strength of * Corresponding author. Tel.: +86 10 82339575. E-mail address: [email protected] (S. Wang). Peer review under responsibility of Editorial Committee of CJA.

Production and hosting by Elsevier

200 GPa, Elastic modulus of 1.3 TPa, and electrical conductivity of 2 · 107 S/m.1–4 In order to transfer the multifunctional properties of CNTs to macroscale material, CNTs could be assemblied to form film materials. The resultant film inherits the excellent properties of individual carbon nanotubes and shows great potential as structural and multi-functional material. Carbon nanotube film is a freestanding mat of intertwined carbon nanotubes (CNTs), which has ultrahigh strength, very low density and high electrical and thermal conductivity. By incorporating CNT film into fiber reinforced polymer composites, it has shown great potential in electromagnetic shielding application. These materials have been extensively evaluated

http://dx.doi.org/10.1016/j.cja.2015.05.002 1000-9361 ª 2015 The Authors. Production and hosting by Elsevier Ltd. on behalf of CSAA & BUAA. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1246 to advance its application in the field of aviation and aerospace. Due to the excellent electrical functionality, such as electrostatic dissipation and shielding effectiveness, CNT film has been applied as outermost ply on fiber reinforced composites and successfully used on the Juno spacecraft.5 In order to optimize the electrical functionality of hybrid structure, it is essential to understand the electromagnetic characteristics of CNT film. Shielding effectiveness has attracted more and more attention not only in military applications but also in civil communication. The field of astronautics and aerospace presents higher demand for the weight and performance of shielding materials. Due to the extremely high electrical conductivity and light weight, carbon nanotube materials are very promising to be the next-generation shielding material. Huang et al. investigated the influence factors of shielding efficiency of single-walled carbon nanotube (SWCNT)/epoxy composites, including the type of SWCNT, aspect ratio and wall integrity.6 A 2 mm thick composite with 15% SWCNTs shows a 20–30 dB EMI shielding efficiency in X-band. Much effort has been made to further improve shielding efficiency by introducing highly conductive metal particles. For instance, a 4 lm CNT film with approximately 35wt% iron showed an electromagnetic interference shielding effectiveness (EMI SE) as high as 61–67 dB.7 Permittivity and permeability are the most important parameters representing electromagnetic properties of material. Permittivity stands for the resistance to the formation of electric field within a substance. Permeability represents the ability to sustain a magnetic field. Their real parts represent the energy storage ability, while imaginary parts are related to the dissipation of energy. The research on the permittivity and permeability of CNTs mainly focused on its polymer matrix composites with low CNT content.8–10 The research about macroscopic CNT film materials was rarely reported. It is important to reveal the relationship between structure and electromagnetic characteristics for further structural design and optimization of shielding materials. CNT film has a thickness of tens of micrometer, and its wavetransparent rate is extremely low. These features bring the difficulty to accurately measure its complex permittivity and permeability. The most widely used methods to measure complex permittivity and permeability of thin film materials include microstrip line method and resonant cavity method.11–17 Although microstrip line and resonant cavity methods have relatively high test precision, narrow frequency range and the difficulty in sample preparation limit their application for the characterization of CNT film. Some attempts have been made to characterize the electromagnetic characteristic of coating or film materials by using waveguide method. Kamarei and Daoud built a test sample with 1 mm thick ferrite placed on plexiglass substrate and obtained accurate results by waveguide method below 2 GHz.18 These attempts developed a new direction for electromagnetic characterization of CNT film. In this paper, CNT film sample supported by Teflon substrate is prepared to meet the needs of sample holder in waveguide method. The theoretical analysis of complex permittivity and permeability was presented based on two independent scattering parameters (S11 and S21) measured by vector network analyzer. The complex permittivity and shielding effectiveness of different CNT film sample were compared, including pristine CNT film, film laminates, resin impregnated

W. Zhang et al. film, and aligned CNT film to reveal the structure–properties relationship. 2. Experimental 2.1. Materials CNT film was synthesized using floating catalyst chemical vapor deposition (CVD) growth method. The film is made of multiwalled carbon nanotubes (MWCNTs). Fig. 1(a) shows the as-prepared multiwalled carbon nanotube film. The film is 10–20 lm thick and CNTs randomly oriented in the plane. CNTs are bonded together through numerous junctions forming a random CNT network. The CNTs in the film tend to assemble into bundles. These synthesized CNTs had an average diameter of 7.8 nm and 3–8 tube walls. Thermogravimetry analytical results show that the CNT film consists of 5.0wt% amorphous carbon, 15.3wt% Fe catalyst, and 79.7wt% carbon nanotube. Also, Raman spectroscopy analysis shows that IG/ID of CNT is about 6.3, indicating its high-quality graphite structure.

Fig. 1

SEM morphology of different CNT film materials.

Electromagnetic characteristics of carbon nanotube film materials

1247

E-51 epoxy with an imidazole hardener was used to prepare resin impregnated film, which were produced by Lanxing Chemical New Materials Co., Ltd. and Beijing Xiangshan United Assistant Factory, respectively. Acetone was used to prepare epoxy solution, which was purchased from Beijing Chemical Works. The substrate Teflon sheets with constant electromagnetic parameters were purchased from Shanghai Daoguan Rubber Hardware Co., Ltd. The pristine CNT film has a big elongation under tension. The aligned CNT film was prepared by a simple mechanical stretching process. The sample was stretched along its winding direction. In the stretching process, CNTs oriented along the applied tensile force, and big CNT bundles were formed. The degree of alignment increased with increasing strain. Correspondingly, the physical properties of CNT film such as electrical conductivity become anisotropic. The morphology of aligned CNT film was shown in Fig. 1(b). In order to strengthen the CNT film, polymer can be introduced to increase the interaction between carbon nanotubes. Epoxy and acetone solution with a resin content of 15 wt% was prepared to facilitate resin permeation. Considering the nanoscale pores in CNT film, it was soaked in the solution for 60 min to ensure full impregnation. After the vaporization of acetone, the impregnated CNT film was cured at 55 C for 0.5 h, 80 C for 1 h and 125 C for 2 h. The morphology of resin impregnated film is shown in Fig. 1(c). 2.2. Characterization Due to the small thickness and flexibility of CNT film, it remains challenging to prepare suitable sample for waveguide measurement. In this paper, we designed a compliant test sample by placing thin CNT film on the surface of Teflon cuboid with cross sectional dimensions of 22.76 mm · 10.06 mm as shown in Fig. 2. The Teflon substrate has a thickness of approximately 5.1 mm. All Teflon substrates in the experiment were machined from the same Teflon sheet and CNT films were cut from the same sheet. The surface of Teflon was polished to avoid the trapped air in the rough surface. Different CNT samples were prepared to explore structure and property relationship, including pristine CNT film, CNT laminate, resin impregnated CNT film, and aligned CNT film. Table 1 lists the structural parameters of these samples. Both winding and stretching directions for pristine and aligned samples were defined as 0 direction. The electromagnetic characteristics were measured by HP 85072E vector network analyzer. Scattering parameters of 11 points at regular interval in X-band were recorded. In order to reduce the error caused by Teflon substrate, CNT film was selected to be incident plane. The sample was placed in the fixture with CNT film at the plane of its open edge. Complex permittivity and permeability can be calculated by reflection and transmission coefficients (S11 and S21) of sample. 3. Theoretical analysis For the waveguide measurement, CNT film sample with Teflon substrate is placed into fixture which is a rectangular waveguide. Port 1 and 2 of vector network analyzer (VNA) were calibrated by through-reflect-line (TRL) method. When calibrating Port 2, empty sample fixture was connected with

Fig. 2 Measurement of electromagnetic characteristics with waveguide measurement.

Table 1

Dimensions of CNT film samples.

No.

Type

Film thickness (mm)

Substrate thickness (mm)

1 2 3 4 5 6 7 8

Monolayer sample (0) Trilayer sample (0) Five-layer sample (0) Monolayer sample (90) Stretched five-layer sample [0]5 Stretched five-layer sample [90]5 Monolayer sample with resin (0) Trilayer sample with resin (0)

0.017 0.039 0.060 0.020 0.059 0.059 0.019 0.036

5.11 5.12 5.17 5.16 5.12 5.10 5.06 5.26

the adapter at Port 2, as shown in Fig. 2(b). Thereby, the interface between fixture and adapter at Port 1 was chosen as the reference plane of scattering parameter. Samples were carefully loaded in the fixture to keep lower surface of fixture and CNT film in the same plane, as shown in Fig. 2(c). CNT film faced Port 1. The thickness of test sample has no influence on reference plane. The test system could be seen as a two-port network; S11 and S21 are its scattering parameters. A column vector X related to scattering parameters can be defined when no reflection occurs at Port 2.

1248

W. Zhang et al.

X¼ "

x1 x2

¼

1 ec2 d2 2Z0 0 #

Z0 þ Z2 0 Z0 Z2 ec2 d2

S21 ec0 ðd1 þd2 Þ 0

Z0 Z2 Z0 þ Z2

ð1Þ

r ¼ xe0 e001r

where Z0, c0 and Z2, c2 are dielectric impendence and propagation constants of free space and Teflon substrate, respectively. d1 and d2 are the thicknesses of CNT film and Teflon substrate, respectively. The dielectric impendence of CNT film Z1 can be calculated as sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð1 þ S11 Þ2 ðx1 þ x2 Þ2 ð2Þ Z1 ¼ Z0 Z2 2 Z2 ð1 S11 Þ2 Z20 ðx1 x2 Þ2 Z1 in Eq. (2) is chosen based on the fact of positive real part. The transmission coefficient of CNT film P can be expressed as: P ¼ jPjej/P ¼

Z0 ½Z2 ðx1 þ x2 Þ þ Z1 ðx1 x2 Þ Z2 ½Z0 þ Z1 þ ðZ0 Z1 ÞS11

ð3Þ

where /P is the phase of P. According to the relationship between transmission coefficient P and propagation constant c1 in CNT film, P ¼ ec1 d1 , c1 can be further calculated as c1 ¼

1 1 2np /P þj ln d1 jPj d1

ð4Þ

where n = 0, ±1, ±2, . . .. . . The CNT film is tens of micrometer thick, which is smaller than a dielectric wavelength. In this case, n value is chosen to be 0 in Eq. (4). Based on the waveguide theory, complex permittivity and permeability are calculated according to Eq. (5). 8 c1 Z1 > > > l1r ¼ j xl < "0 # ð5Þ 2 2 c 2 > k 1 1 0 > > e1r ¼ : h 4l1r p where l0 and k0 are permeability and wavelength of free space; h is waveguide width. The analysis method has been proved by the measurement of epoxy film with plexiglas substrate. The calculated permittivity and permeability of epoxy film are very close to the results of pure epoxy casting. When electromagnetic wave reaches the surface of CNT film with an incident angle h, wave constant k1z and impendence g1z of CNT film could be calculated from e1r and l1r as follows: qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2p l1r e1r sin2 h g1z ¼ xl0 l1r =k1z k1z ¼ ð6Þ k0 Thus, the transmittance T of CNT film can be further written as: 1 j g0z g1z T ¼ cos k1z d1 þ þ ð7Þ sin k1z d1 2 g1z g0z where g0z is wave impendence of free space in propagation direction and g0z = cl0/cosh; c is light velocity of free space. Shielding effectiveness (SE) of CNT films can be written as SE ¼ 20 lg jTj

permeability and thickness. In the following discussion, farfield shielding effectiveness was analyzed. Complex permittivity of CNT film can be expressed as e1r ¼ e01r je001r . The conductivity r was further calculated according to Eq. (9).

ð8Þ

For far-field shielding effectiveness, incident angle h is considered to be zero. According to Eqs. (7) and (8), far-field shielding effectiveness of CNT film is determined by permittivity,

ð9Þ

4. Results and discussion 4.1. Electromagnetic characteristics of pristine CNT film During the fabrication process of CNT film, it was winded on a roller. Even though very small force was applied, the resultant film showed slight anisotropy. In order to investigate the influence of anisotropic morphology on electromagnetic characteristics, two samples were prepared with their winding direction parallel or normal to the lengthwise direction of the rectangular sample, i.e. the polarized direction of electrical field was parallel or perpendicular to the winding direction of CNT film, which was denoted as 0 or 90 sample, respectively. Fig. 3 shows the measured real part of complex permittivity, electrical conductivity and shielding effectiveness under different polarized direction. The 0 sample of CNT film has a real part of complex permittivity in the range of 3000–3500. The 90 sample shows higher real part of complex permittivity ranging from 3500 to 6800 as shown in Fig. 3(a). The conductivity calculated from Eq. (10) also exhibits similar anisotropic feature. CNT film of 90 direction has an electrical conductivity of 1700 S/m in X-band, while the conductivity in 0 direction is much lower. The shielding effectiveness was further calculated according to Eq. (9) as shown in Fig. 3(c). The shielding effectiveness also shows obvious anisotropic feature. The 0 sample has a shielding effectiveness in the range of 15 dB–18 dB, while 90 one has a larger shielding effectiveness ranging from 20 dB to 25 dB. Table 2 lists the reported shielding effectiveness of other CNT materials and the present results. 17 lm thick CNT film shows the same electromagnetic shielding ability as 2 mm thick SWCNT/epoxy composites with 15 wt% CNT. This is caused by the superior electrical conductivity of CNT film with long carbon nanotubes and high CNT content. Fig. 3(d) shows the reflectance (S11) of sample 1. Its absolute value of magnitude is very small and its phase is in the range of 175.4 to 174.2. The high reflectance of CNT film plays the most important role in its excellent shielding effectiveness. 4.2. Electromagnetic characteristics of CNT film laminates In order to enhance the shielding effectiveness, several layers of CNT film can be laminated. Fig. 4 shows the permittivity, permeability and shielding effectiveness of CNT film laminates with different layer numbers. The samples with monolayer, trilayer and five-layer CNT film were compared. For all these samples, the polarized direction of electrical field was parallel to the winding direction. For monolayer CNT film the real part of complex permittivity was calculated to be 3000–3500. It was obviously observed that the real part of complex permittivity increased with increasing layer number of CNT film. The trilayer sample had a larger real part of 5000 and the real part for five-layer sample reached as high as 30000–40000.


Fig. 3

1249

Complex permittivity, conductivity, shielding effectiveness and reflectance of pristine CNT film.

Table 2 Comparison of EMI SE results of CNT film and other materials. Material

EMI SE (dB)

Frequency Thickness range (mm) (GHz)

SWCNT/epoxy composites6 PU/SWCNT composites10 CNT/crystalline Fe dispersed into epoxy resin19 CNT film with iron particles7 Present monolayer CNT film 90 Present five-layer CNT film 0 Present stretched five-layer CNT film 90 Present monolayer CNT film 0 with epoxy resin

20–25 16–19 12–25

8–12.4 8.2–12.4 2–18

2 2 1.2

61–67 20–25 67–78 54–57

8–12.6 8.2–12.4 8.2–12.4 8.2–12.4

0.004 0.02 0.06 0.039

12–14

8.2–12.4

0.019

Fig. 4(c) shows the shielding effectiveness of CNT laminates with different layer numbers. The shielding effectiveness of monolayer CNT film was in the range of 15–18 dB. The addition of CNT layer greatly increased the shielding effectiveness. The shielding effectiveness of trilayer sample increased to 37 dB. The five-layer sample showed a shielding effectiveness in the range of 67–78 dB, which is much higher than reported results of other CNT materials as listed in Table 2. The superior shielding effectiveness of five-layer sample meets the requirement of avionics system and sophisticated electronics

where 70 dB shielding effectiveness is demanded. The shielding effectiveness of five-layer CNT film laminate is higher than that of copper and aluminum sheet with the same thickness (0.06 mm), which is only about 58 dB in X-band calculated from dc conductivity. The real part of permeability shows interesting tendency with the increase of layer number. Trilayer and five-layer CNT film laminates exhibit a negative real part of permeability (Fig. 4(b)). Random CNT film consists of numerous CNT bundles, which are intertangled and have disordered arrangement. These CNT bundles construct a lot of closed paths. Fig. 5(a) shows the closed path consisting of CNT bundles. Although the path is closed in xy plane, splits exist between bundles in xz plane. The split closed path made of CNT bundles is similar to split-ring resonator (SRR), which is the basic structural unit of artificial negative permeability material. Fig. 5(c) shows a circular SRR. If magnetic field H has a component perpendicular to the plane of SRR, induced current i may appear when magnetic flux through SRR changes, as shown in Fig. 5(d). SRR can be seen as a magnetic moment. Meanwhile, effective inductance may occur in closed path and the split can yield effective capacitance. SRR could also be seen as a resonator circuit. Base on magnetization and circuit, effective permeability of SRR will have a negative real part when frequency of incident wave is in a special range above resonance frequency of SRR.20 SRR shape is not an essential factor to negative permeability. And hexagonal and square SRR were also reported to have

1250

W. Zhang et al.

Fig. 5 Schematic diagram of split closed path made of CNT bundles, and split-ring resonator model.

Fig. 4 Comparison of electromagnetic characteristics of CNT film laminates with different layer numbers.

negative permeability behavior.21,22 Two essential requirements to induce negative permeability of CNT film can be concluded as: (1) numerous split closed paths consisting of CNT bundles; (2) incident magnetic field must have component perpendicular to these closed paths. In the waveguide measurement of CNT film, magnetic field H and wave vector k has component perpendicular to the plane of these split closed paths, while electrical field E is parallel to split closed paths as shown in Fig. 5(e). The amount of split closed paths through thickness direction is limited for a single layer. Multilayer CNT film is easier to behave negative permeability than monolayer CNT film, which is proved by experimental results.

The negative permeability of multilayer CNT film laminate is decided by the arrangement of CNT bundles and the direction of incident wave. The influence of wave incidence angle on the permeability indicates the anisotropic feature of CNT film laminates. The negative permeability contributes to a significant shielding effectiveness. The addition of layer number of CNT film not only increased the thickness but also caused strong coupling between layers. The materials with negative permeability are difficult to be constructed. Further study should be carried out to investigate the unique negative permeability of CNT film laminates and reveal the relationship between microstructure and properties of CNT film laminate.

4.3. Effect of CNT alignment on its electromagnetic characteristics Stretching process could greatly induce the alignment of CNTs in the stretching direction. Five-layer samples were used in the test. The samples with the stretching direction parallel (0) or perpendicular (90) to polarization direction were compared. Fig. 6(a) shows the real part of complex permittivity of stretched CNT film samples. The permittivity of 0 sample has a negative real part. From the morphology of stretched CNT film in Fig. 1(b), CNT bundles become straighter and


1251

Fig. 7 Conductivity of stretched CNT film and schematic illustration of mechanical stretching to induce CNT alignment.

Fig. 6 Complex permittivity of stretched CNT film and structural schematic of periodic wire array.

their configuration also becomes more ordered when CNT film is stretched in a certain direction. Meanwhile, the distance between adjacent bundles becomes smaller. When stretching ratio reaches a certain level, CNT bundles become very straight and the distance of adjacent bundles is very tiny. The CNT bundles were almost equally spaced and CNT bundles were very similar to periodic wire array as shown in Fig. 6. Periodic wire array can be equalized as plasma and the plasma frequency xP can be written as follows.23–25 sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi c 2p ð10Þ xP ¼ a lnða=rÞ where a is lattice constant of array, r the radius of CNT bundles, and c the light velocity in free space. When the polarized direction of electrical field is parallel to the stretching direction, the effective permittivity eeff can be written as 1 x2P e0 a2 x2P xþj eeff ¼ 1 ð11Þ x pr2 r where e0 is vacuum permittivity, r is electrical conductivity. According to the SEM and TEM morphology, the dimensions of CNT bundles are as follows: a = 105 nm, r = 50 nm. The simulated real part of complex permittivity from Eqs. (10) and (11) keeps stable at approximately 3.76 · 104 in X-band, which is consistent with the experimental results. Fig. 7(a) shows the electrical conductivity of 0 and 90 samples. The conductivity of 0 sample is fivefold bigger than 90 sample. CNT bundles were straightened in the stretching process as shown in Fig. 1(b). Fig. 7(b) illustrates the

Fig. 8

Shielding effectiveness of stretched CNT film.

morphology evolution during stretching process. S1 and S2 are two parallel planes perpendicular to the stretching direction. With the increase of stretching ratio the intertangled CNT bundles were put up straight and CNTs in bundle were tightly packed. As a result, the effective conductive path between S1 and S2 became shorter and shorter. Both the short conductive path and tightly packed structure are beneficial to the electrical conductivity. Meanwhile shielding effectiveness of 0 sample is better than 90 sample (shown in Fig. 8) because of higher conductivity and negative permittivity. 4.4. Effect of resin impregnation on its electromagnetic characteristics Fig. 9 shows the electromagnetic parameters of resin impregnated CNT film in X-band. Real part of complex permittivity of resin impregnated CNT film is approximately 1000, which is much lower than pristine CNT film. For the resin impregnated

1252

W. Zhang et al.

Fig. 9

Electromagnetic characteristics of CNT films with and without resin.

CNT film, CNTs were coated by resin on its surface, and resin enriched in the junction, as shown in Fig. 1(c). Due to the poor electrical conductivity, polymer matrix acts as insulation for conductive path between CNTs. The electronic field conduction and electron motion are blocked because resin molecules enter the space between nanotubes. Trilayer sample shows the same tendency. Real part of complex permittivity reduces from 6000–9000 to 1000–2000, and conductivity decreases by more than 70%. A simple model of CNT film and polymer sandwich structure was established for resin impregnated CNT film as shown

Fig. 10

Simplified model of resin impregnated CNT film.


Fig. 11

1253

Comparisons of electromagnetic characteristics of resin impregnated CNT film with different thicknesses of resin layer.

in Fig. 10. The thicknesses of both polymer layers are the same. In order to simulate the resin impregnated CNT film, the experimental permittivity and permeability of pristine CNT film were used as input parameters. The permittivity and permeability of epoxy resin casting were characterized using waveguide method. Fig. 11(a–c) shows the simulated electromagnetic characteristic of resin impregnated CNT film with different resin content. Both the real part of complex permittivity and electrical conductivity of resin impregnated CNT film decrease with increasing resin content (i.e. the thickness of polymer layer in the model). Compared to the pristine CNT film without resin, impregnated CNT film has smaller shielding effectiveness. However, after resin impregnation, resin content has no obvious influence on the shielding effectiveness. Introducing resin into CNT film shows greater influence on the trilayer sample than the monolayer one. Moreover, that real part of permeability becomes positive for resin impregnated CNT film (Fig. 11(d)). This may be caused by the weaker resonance in the resin impregnated sample. 5. Conclusion In this paper, CNT film supported by a Teflon substrate was designed to meet the requirement of waveguide measurement for the characterization of electromagnetic parameters of CNT film. The theoretical analysis of permittivity,

permeability and shielding effectiveness was developed based on measured S11 and S21. The CNT film shows anisotropic features in electromagnetic parameters. When the polarized direction of incident wave is perpendicular to the winding direction of CNT film, it shows higher real part of complex permittivity, electrical conductivity and shielding effectiveness. For the CNT film laminates, complex permittivity increases with increasing layer number and shielding effectiveness decreases. The five-layer CNT film shows extraordinary shielding properties with shielding effectiveness in the range of 67–78 dB in Xband. Stretching process induces the alignment of CNTs, and when aligned direction is parallel to the electronic field, CNT film shows negative permittivity and higher conductivity. Moreover, the resin impregnation into the CNT film leads to the decrease of conductivity and shielding effectiveness. Acknowledgements The research was supported by the National Science Foundation of China (No. 51403009), the ‘‘Fundamental Research Funds for the Central Universities’’ (No. YWF-14RSC-002) of China, and the 54th Research Institute of China Electronics Technology Group Corporation. The authors thank Suzhou Institute of Nano-tech and Nanobionics, Chinese Academy of Sciences for supplying carbon nanotube film.

1254 References 1. Zhang R, Wen Q, Qian W, Su DS, Zhang Q, Wei F. Superstrong ultralong carbon nanotubes for mechanical energy storage. Adv Mater 2011;23(30):3387–91. 2. Mann D, Javey A, Kong J, Wang Q, Dai H. Ballistic transport in metallic nanotubes with reliable Pd ohmic contacts. Nano Lett 2003;3(11):1541–4. 3. Berber S, Kwon YK, Tomanek D. Unusually high thermal conductivity of carbon nanotubes. Phys Rev Lett 2000;84(20): 4613–6. 4. Ebbesen TW, Lezec HJ, Hiura H, Bennett JW, Ghaemi HF, Thio T. Electrical conductivity of individual carbon nanotubes. Nature 1996;382(6586):54–6. 5. Queffelec P, Gelin P, Gieraltowski J, Loaec J. A microstrip device for the broad band simultaneous measurement of complex permeability and permittivity. IEEE Trans Magn 1994;30(2): 224–31. 6. Huang Y, Li Y, Ma Y, Du F, Li F, He X, et al. The influence of single-walled carbon nanotube structure on the electromagnetic interference shielding efficiency of its epoxy composites. Carbon 2007;45(8):1614–21. 7. Wu ZP, Li MM, Hu YY, Li YS, Wang ZX, Yin YH, et al. Electromagnetic interference shielding of carbon nanotube macrofilms. Scr Mater 2011;64(9):809–12. 8. Watts PCP, Hsu WK, Barnes A, Chambers B. High permittivity from effective multiwalled carbon nanotubes in the X-band. Adv Mater 2003;15(7–8):600–3. 9. Lu S, Xu W, Xiong X, Ma K, Wang X. Preparation magnetism and microwave absorption performance of ultra-thin Fe3O4/carbon nanotube sandwich buckypaper. J Alloy Compd 2014;606:171–6. 10. Chen R, Peng LM, Duan XF, Chen Q, Liang X. Microwave absorption enhancement and complex permittivity and permeability of Fe encapsulated within carbon nanotubes. Adv Mater 2004;16(5):401–5. 11. Rawal S, Brantley J, Karabudak N. Development of carbon nanotube-based composite for spacecraft components. Proceedings of 6th international conference on recent advances in space technologies, 2013 Jun 12–14. Istanbul, Turkey. Piscataway, NJ: IEEE; 2013. p. 12–4. 12. Queffelec P, Gelin P. Influence of higher order modes on the measurements of complex permittivity and permeability of materials using a microstrip discontinuity. IEEE Trans Microw Theory 1996;44(6):816–24. 13. Queffelec P, Floc’h ML, Gelin P. Broad-band characterization of magnetic and dielectric thin films using a microstrip line. IEEE Trans Instrum Meas 1998;47(4):956–63. 14. Hinojosa J. S-parameter broad-band measurements on-microstrip and fast extraction of the substrate intrinsic properties. IEEE Microw Wirel Co 2001;11(7):305–7.

W. Zhang et al. 15. Wu Y, Tang Z, Xu Y, Gou Y, Zhang B. Support vector regression for measuring electromagnetic parameters of magnetic thin-film materials. IEEE Trans Magn 2007;43(12):4071–5. 16. Kozlowski G, Kleismit G, Boeckl G, Campbell A, Munbodh K, Hopkins S, et al. Electromagnetic characterization of carbon nanotube films by a two-point evanescent microwave method. Physica E 2009;41(8):1539–44. 17. Huang H, Free CE, Pitt KEG, Berzins AR, Shorthouse GP. Relative permittivity measurements of thick-film dielectrics at microwave frequencies. Electron Lett 1995;31(21):1812–4. 18. Kamarei M, Daoud N. Measurement of complex permittivity and permeability of dielectric materials placed on a substrate. Electron Lett 1991;27(1):68–70. 19. Lozano K, Espinoza L, Hernandez K, Adhikari AR, Radharrishnan G, Adams PM. Investigation of the electromagnetic interference shielding of titanium carbide coated nanoreinforced liquid crystalline polymer. J Appl Phys 2009;105(10): 103511. 20. Smith DR, Padilla WJ, Vier DC, Nemat-Nasser SC, Schultz S. Composite medium with simultaneously negative permeability and permittivity. Phys Rev Lett 2000;84(18):4184–7. 21. Chandrasekhar K, Mishra D, Poddar DR, Mishra RK. Design of left-handed metamaterials using hexagonal split ring resonator at S-band frequencies. Proceedings of applied electromagnetics conference, 2007 Dec 19–20. Kolkata, India. Piscataway, NJ: IEEE; 2007. p. 1–4. 22. Shelby RA, Smith DR, Schultz S. Experimental verification of a negative index of refraction. Science 2001;292(5514):77–9. 23. Rotmant W. Plasma simulation of artificial dielectrics and parallel-plate media. IRE Trans Antennas Propag 1962;10(1): 82–95. 24. Larsen T. A survey of the theory of wire grids. IRE Trans Microw Theory Tech 1962;10(3):191–201. 25. Pendry JB, Holden AJ, Robbins DJ, Stewart WJ. Low frequency plasmons in thin-wire structures. J Phys Condens Matter 1998;10(22): 4785–809. Zhang Wei received Bachelor degree of Engineering in electronic and information engineering and Master’s degree in electromagnetic and microwave technology from Beihang University in 2003 and 2006, respectively. He is currently pursuing Ph.D. degree of information and communication engineering at Beihang University. His current research interests include electromagnetic theory, measurement of electromagnetic parameters, and material shielding design. Wang Shaokai is currently working at Beihang University as a lecturer. He received his Ph.D. degree of materials science at Beihang University in 2011. His research interests include polymer composite and processing, processing-structure–property relationship of polymer nanocomposites.