Electromagnetic interference shielding properties of polymer-grafted carbon nanotube composites with high electrical resistance

Electromagnetic interference shielding properties of polymer-grafted carbon nanotube composites with high electrical resistance

CARBON 8 5 ( 2 0 1 5 ) 3 6 3 –3 7 1 Available at www.sciencedirect.com ScienceDirect journal homepage: www.elsevier.com/locate/carbon Electromagne...

1MB Sizes 0 Downloads 31 Views

CARBON

8 5 ( 2 0 1 5 ) 3 6 3 –3 7 1

Available at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevier.com/locate/carbon

Electromagnetic interference shielding properties of polymer-grafted carbon nanotube composites with high electrical resistance Kenichi Hayashida a b

a,* ,

Yoriko Matsuoka

b

Materials and Processing Dept. II, Toyota Central R&D Labs., Inc., Nagakute, Aichi 480-1192, Japan Materials Analysis and Evaluation Dept., Toyota Central R&D Labs., Inc., Nagakute, Aichi 480-1192, Japan

A R T I C L E I N F O

A B S T R A C T

Article history:

Poly(methyl methacrylate) (PMMA)-grafted multiwalled CNTs were prepared, and then dis-

Received 26 September 2014

persed into additional PMMA matrix, yielding highly insulated PMMA–CNT composites.

Accepted 4 January 2015

The volume resistivity of PMMA–CNT was as high as 1.3 · 1015 X cm even at 7.3 wt% of

Available online 9 January 2015

the CNT. The individual CNTs electrically-isolated by the grafted PMMA chains in PMMA– CNT transmitted electromagnetic (EM) waves in the frequency range of 0.001–1 GHz, whereas the percolated CNTs in a conventional composite prepared by blending PMMA with the pristine CNTs strongly shielded the EM waves. This result suggests that the intrinsic conductivity of the CNT itself in PMMA–CNT does not contribute to the EM interference (EMI) shielding in the frequency range of 0.001–1 GHz. On the other hand, PMMA–CNT exhibited EMI shielding at the higher frequency range than 1 GHz because the dielectric loss of the CNT itself was rapidly increased over 1 GHz. At 110 GHz, PMMA–CNT with 7.3 wt% of the CNT had EMI SE of as high as 29 dB (0.57 mm thickness), though is slightly lower than that of the percolated conventional composite (35 dB). Thus, it is demonstrated that the highly insulated PMMA–CNT has the good EMI shielding at extremely high frequency range (30–300 GHz).  2015 Elsevier Ltd. All rights reserved.

1.

Introduction

Polymer/carbon nanotube (CNT) composites have been paid much attention as electromagnetic interference (EMI) shielding materials because the CNTs impart high electrical conductivity and good mechanical properties to the polymer materials at low loading of the CNT [1–5]. It is well-known that the electrical conductivity enhances the EMI shielding, and the highly conductive polymer/CNT composites exhibit superior EMI shielding effectiveness (SE) [6–29]. The polymer/CNT composite becomes conductive owing to the

* Corresponding author. E-mail address: [email protected] (K. Hayashida). http://dx.doi.org/10.1016/j.carbon.2015.01.006 0008-6223/ 2015 Elsevier Ltd. All rights reserved.

formation of a three-dimensional conductive network of the CNTs within the polymer matrix when the CNT content exceeds a critical value, known as a percolation threshold [1–5]. At over the percolation threshold, the CNTs approach each other within several nanometers, allowing efficient electron tunneling between the CNTs (tunneling conduction) [30,31]. This conductive CNT network can strongly interact with the EM waves. In general, as mentioned above, good EMI shielding materials are always electrically conductive; this fact is not limited to the polymer/CNT composites [32–38]. On the other hand, electronic materials including

364

CARBON

8 5 ( 2 0 1 5 ) 3 6 3 –3 7 1

the EMI shielding materials often require electrical insulation (e.g. EMI shielding for high frequency module and signal transmission cable). For such applications, therefore, the EMI shielding materials have to be covered with other materials with high electrical resistance. Is it impossible to produce a polymer/CNT material with both of high EMI SE and electrical insulation? Also, it is suggested by some research groups that the electrical conductivity of the polymer/CNT composites is not the scientific criterion for the EMI shielding because the shielding does not require the conductive network of the CNTs; the mobile electrons in the CNTs interact with the EM waves [15,18,24,39]. Therefore, the conductivity of the CNT itself is essential for the EMI shielding, and the conductive network of the CNTs only improves the EMI shielding. However, the issue of the relationship between the conductivity of the CNT itself and the EMI shielding has been still unclear because the formation of the conductive networks is inevitable for the polymer/CNT composites owing to the extremely low percolation threshold of the CNTs. If the percolation of the CNTs can be forbidden, the clear understanding of the issue would be obtained, and a polymer/CNT material with both of high EMI SE and electrical insulation might be produced. A solution to prevent the percolation is polymer-grafting on the surface of the CNTs. Recently, we have reported the ultrahigh electrical resistance of poly(cyclohexyl methacrylate) (PCHMA)/multiwalled CNT composite system where all the PCHMA chains were tethered to the CNT surface [40]. The grafted polymer chains isolated the individual CNTs at a sufficient distance (>10 nm) to prevent the tunneling conduction between the CNTs [41]. Here, we report the interaction of EM waves with individual CNTs electrically-isolated by grafted polymer chains. In this study, poly(methyl methacrylate)-grafted multiwalled CNTs ([email protected]) were prepared using our surface-initiated atom transfer radical polymerization (SI-ATRP) technique, and then dispersed into additional (ungrafted) PMMA matrix, yielding highly insulated polymer/CNT composites (PMMA– CNT). The EMI properties of PMMA–CNT were characterized in the broad frequency f range of 0.001–110 GHz, and compared to those of conventional composites prepared by blending PMMA with the pristine CNTs (PMMA/CNT). To our knowledge, this is the first report that the EMI properties are investigated for a polymer/CNT composite system with high electrical resistance.

2.

Experimental

2.1.

Sample preparation

Multiwalled CNT with an average diameter and length of 10 nm and 1.5 lm, respectively, was obtained from Nanocyl SA. The density of the CNT was determined to be 1.8 g cm3 by pycnometry. ATRP-initiator modified CNT (CNT-Br) was synthesized as previously reported in detail [40]. An example of the procedure of the SI-ATRP of MMA using CNT-Br was as follows: N,N-dimethylacetamide (160 ml) was added to CNTBr (1.51 g) and CuBr (85.8 mg, 600 lmol) in a N2 atmosphere. After sonication of the mixture, 2,2 0 -bipyridyl (187 mg, 1.2 mmol) in MMA (80 ml) was added, and the solution was

kept at 60 C for 9 h. The resultant [email protected] was precipitated in methanol and freeze-dried by benzene. PMMA–CNTwas prepared by blending [email protected] with an additional PMMA matrix (Mn = 15 k). First, [email protected] and the PMMA matrix were added in benzene and homogeneously dispersed by sonication. The dispersion was quickly frozen by liquid N2, and then freeze-dried under vacuum. Similarly, PMMA/CNT was also prepared by blending PMMA with the pristine CNTs. PMMA/CNT and PMMA–CNT were molded by hot pressing at 130 C into 15 mm square specimens with a thickness of 0.85 mm for direct current (DC) electrical resistance and impedance measurements, or 150 mm square specimens with a thickness of 0.57 mm for two types of EMI SE measurements.

2.2.

Measurements

Morphologies of [email protected] and PMMA–CNT were observed using scanning electron microscopes (S-4300 and S-5500, respectively, Hitachi) operated at an accelerating voltage of 2 kV. For the SEM observation of PMMA–CNT, the cross-sections of the molded composites that had been planed using an ultramicrotome, were slightly etched with oxygen plasma in order that the PMMA matrix on the surface was removed out. The DC resistance of the composites was measured at a voltage of 500 V using a megohmmeter (SM-8220, Hioki) for PMMA–CNT with high resistance and a multimeter for PMMA/CNTwith low resistance in combination with a test fixture with 10 mm square copper electrodes. In order to minimize the contact resistance between the electrodes and a specimen, two gold electrodes (11 mm square) were deposited by sputter coating on the top and bottom of the specimen at a distance of 2 mm from the edge of the specimen. The measurement system was calibrated with a copper specimen (1 mm thickness) on which the same gold electrodes were deposited in order to remove the resistance of the gold electrodes and the total contact resistances. Dielectric spectra of the composites were obtained in the f range of 0.001–1 GHz using an impedance analyzer (E4991A, Agilent) equipped with a commercially available test fixture (16453A, Agilent). The test fixture was calibrated with a poly(tetrafluoroethylene) film to be minimized contact impedance between the electrode and a specimen. EMI SE of the composites was obtained by two types of methods depending on the f range. In the f range of 0.001– 1 GHz, the KEC method was used [42]. The KEC method was developed by Kansai Electronic Industry Development Center in Japan. The set-up consisted of a test fixture (MA8602B, Anritsu), an RF amplifier (8247A, Agilent), an RF preamplifier (MH-648A, Anritsu), and a spectrum analyzer (R3361C, Advantest). The test fixture is based on a transverse electromagnetic (TEM) cell that is divided into two units, and a specimen is sandwiched between the two units. In the higher f range than 1 GHz, the free space method was used [10]. For the free space method, 5 bands, 5.6–8.2 GHz (C band), 12.4–18.0 GHz (Ku band), 26.5–40 GHz (Ka band), 50.0–68.0 GHz (V band), and 75.0–110.0 GHz (W band), were scanned. S parameters were obtained by combination of a network analyzer (N5227A, Agilent) and a millimeter-wave controller (N5361A, Agilent). This system including spot-focusing horn lens antennas was cali-

CARBON

brated by through-reflect-line (TRL) technique that is the most robust calibration technique for the free space method [43].

3.

Results and discussion

3.1.

Preparation and characterization of PMMA–CNT

365

8 5 (2 0 1 5) 3 6 3–37 1

PMMA/CNT, conventional composites, that has the same UCNT as PMMA–CNT was also prepared by blending PMMA with the pristine CNTs. Thus-obtained two types of composites, PMMA/CNT and PMMA–CNT, were molded by hot pressing. The morphology of PMMA–CNT with UCNT = 0.030 is shown in Fig. 2c. For the SEM observation, a cross section of the molded PMMA–CNT composite was slightly etched with oxygen plasma in order that the grafted PMMA and the PMMA matrix on the surface were selectively removed out [40,44]. Owing to the oxygen plasma etching, the individual CNTs can be clearly observed by the SEM image. The uniform distribution of the CNTs in PMMA–CNT with UCNT = 0.030 is also confirmed by the SEM image with low magnification in Fig. 2d. Fig. 3a shows DC volume resistivity qDC of the two types of composites, and the coefficient of variation for the qDC values is shown in Fig. 3b. The qDC of PMMA/CNT is rapidly decreased to below 103 X cm with the increase in UCNT owing to the percolation of the CNTs. This is typical of polymer/CNT composites prepared by blending. In contrast, the qDC of PMMA–CNT is almost constant (to be accurate, slightly decreased) and as high as 1.3 · 1015 X cm even at UCNT = 0.073, which is 14 orders of magnitude higher than that of PMMA/CNT. The thick PMMA shell over 10 nm isolates the individual CNTs at a sufficient distance (>10 nm) to prevent the tunneling conduction between the CNTs. This ultrahigh electrical resistance of PMMA–CNT is never attributed to degradation of the CNTs in [email protected] because the synthesis condition for [email protected] was very moderate. It was confirmed that the inner walls in the CNTs were intact using high resolution transmission electron microscopy in our previous study where PCHMA was grafted on the same CNTs in the same manner [40].

Fig. 1 shows the synthesis method of [email protected], where multiwalled CNTs with an average diameter of 10 nm were used. First, ATRP-initiator modified CNTs (CNT-Br) were prepared as shown in Fig. 1a and b. The detailed synthesis procedure of CNT-Br was previously reported [40]. Using CNT-Br, SIATRP of methyl methacrylate (MMA) was conducted. Two [email protected] samples, [email protected] and [email protected], were prepared as shown in Table 1. The weight fraction of the CNT UCNT in [email protected] was determined from the weight increment after the SI-ATRP of MMA. Furthermore, the theoretical thickness of the PMMA shell around the CNT core was estimated using a density of 1.8 g cm3 for the CNT and of 1.2 g cm3 for the rest organic components (Table 1). Fig. 2a and b show SEM images of the pristine CNT and [email protected], respectively. Because of the grafted PMMA chains, [email protected] has a much larger diameter than the pristine CNTs. The diameter of [email protected] is estimated to be about 30–50 nm from the SEM image in Fig. 2b, which is consistent with the calculated value of 44 nm from the PMMA-shell thickness of 17 nm and the CNT diameter of 10 nm. [email protected] was dispersed in a PMMA matrix, yielding a PMMA–CNT composite. [email protected] was used for PMMA– CNT with UCNT = 0.015 and UCNT = 0.030, and [email protected] was used for PMMA–CNTwith UCNT = 0.073 (Table 1). Similarly,

(a) H 2N

O

NH 2

NaNO 2, HClaq

(ODA)

5 °C

Cl N N

O

NH 3Cl

(b) ClN 2

CNT

O

NH 3Cl

CNT

5 ºC → 50 °C, 11 h

O

NH 2

(CNT-NH2) BrCH2

CH3 CH2O C C Br O CH3

40 °C, 24 h

CH3

CNT-NH

CH 2

CH 2O

(CNT-Br)

(c)

C C Br O CH3

CH 3 H 2C C

CNT-Br

O

C

OCH 3

60 °C, 9 h

CNT

CH 3 CH 2 O

C Br n C OCH3

=

([email protected]) Fig. 1 – Synthesis procedure of poly(methyl methacrylate)-grafted CNT ([email protected]). (a) Preparation of a diazonium salt from 4,4 0 -oxydianiline (ODA). (b) Preparation of ATRP initiator-modified CNT (CNT-Br). (c) Surface-initiated ATRP of methyl methacrylate using CNT-Br.

366

CARBON

8 5 ( 2 0 1 5 ) 3 6 3 –3 7 1

Table 1 – Characteristics of PMMA–CNT composites. Sample

PMMA shell a Grafted Ungrafted thickness (nm)

UCNT UPMMA

[email protected] 0.075 0.925 [email protected] 0.128 0.872 PMMA–CNTb 0.015 0.185 0.030 0.368 0.073 0.503

– – 0.800 0.602 0.424

17.1 11.8 17.1 17.1 11.8

a Calculated from the average diameter of the CNT (10 nm) and the volume fraction of the CNT in [email protected] The volume fraction of the CNT was estimated using a density of 1.8 g cm3 for the CNT and of 1.2 g cm3 for the organic components. b [email protected] was used for PMMA–CNT with UCNT = 0.015 and 0.030, and [email protected] was used for PMMA–CNT with UCNT = 0.073.

3.2. EMI shielding properties of PMMA–CNT in the f range of 0.001–1 GHz In this study, the EMI SE of the highly insulated PMMA–CNT was investigated over the broad f range of 0.001–110 GHz using two types of methods. For the EMI SE measurements, PMMA/CNT and PMMA–CNTwere molded into 150 mm square specimens with a thickness of 0.57 mm. In the lower f range of 0.001–1 GHz, the KEC method was used [42]. For the KEC method, transmittance T is obtained and EMI SE is calculated as SE = 10 logT. Fig. 4 shows EMI SE of the two types of

(a)

composites in the f range of 0.001–1 GHz. For PMMA/CNT, the EMI SEs of all composites are more than 40 dB at the low f range but quickly decreased with the increase in f as previously reported [9]. At 1 GHz, the SE of PMMA/CNT with UCNT = 0.073 has decreased to 22 dB. On the other hand, the EMI SEs of PMMA–CNT are extremely lower than those of PMMA/CNT, which take constant values of only a few dB over the f range of 0.001–1 GHz. The EMI SE of a material is closely related to the imaginary part of complex relative permittivity e00r . Fig. 5 shows dielectric spectra of the two types of composites in the f range of 0.001–1 GHz obtained using an impedance analyzer. As shown in Fig. 5b, the e00r of PMMA/CNT is much greater than that of PMMA–CNT, resulting in the much higher EMI SE for PMMA/CNT. e00r is the sum of a loss from conductivity e00r (r) and losses from dielectric polarizations e00r (DP) [13,45]: e00r ¼ e00r ðrÞ þ e00r ðDPÞ :

ð1Þ

Furthermore, e00r (r) is expressed using DC conductivity rDC (= 1/qDC) as follows: e00r ðrÞ ¼ rDC =ð2pf e0 Þ;

ð2Þ

where e0 is the permittivity of vacuum. According to Eq. (2), e00r (r) is inversely proportion to f. This characteristic behavior is observed in the spectra for PMMA/CNT. These spectra are roughly consistent with the theoretical lines calculated from Eq. (2) using the qDC values in Fig. 3. This means that the e00r of PMMA/CNT is dominated by e00r (r). Consequently, we infer

(b)

200 nm

(c)

200 nm

(d)

500 nm

2 μm

Fig. 2 – SEM images of (a) pristine CNT, (b) poly(methyl methacrylate)-grafted CNT ([email protected]), and (c and d) PMMA–CNT composites with UCNT = 0.030.

ρDC [Ω cm]

(a)

(b)

10

18

10

15

10

12

10

9

10

6

10

3

10

0

0

0.02

0.04

Φ CNT

0.06

367

8 5 (2 0 1 5) 3 6 3–37 1

(a)

10

2

ε r’

CARBON

10

1

0.08

60

40

0

10 0.001

30

0.1

1

0.1

1

0.1

1

f [GHz]

20

(b)

10 0

0.01

0

0.02

0.04

Φ CNT

0.06

0.08

Fig. 3 – (a) DC volume resistivity qDC of PMMA (diamond) and two types of composites (circles) and (b) coefficient of variation (CV) for qDC as a function of UCNT. Open circles, PMMA/CNT; filled circles, PMMA–CNT.

10

4

10

3

10

2

10

1

10

0

εr”

CV [%]

50

10

−1

10

−2

50

40

10−3 0.001

0.01

SE [dB]

f [GHz] 30

(c)

10

2

10

1

10

0

20

tanδ

10

0 0.001

0.01

0.01

1

f [GHz] Fig. 4 – Shielding effectiveness (SE) of PMMA/CNT (open symbols) and PMMA–CNT (filled symbols) with UCNT = 0.015 (circles), UCNT = 0.030 (triangles), and UCNT = 0.073 (squares) in the frequency f range of 0.001–1 GHz. The thickness of the specimens is about 0.57 mm.

that the high and f dependent EMI SE for PMMA/CNT is mainly attributed to e00r (r). In addition, the high e0r values of PMMA/CNT would improve the EMI SE because as the e0r of a material becomes higher, the reflection of EM waves is enhanced [11,21]. Meanwhile, it is considered that the e00r of

10

−1

10

−2

10

−3

10

−4

0.001

0.01

f [GHz] Fig. 5 – Frequency f dependence of (a) e0r and (b) e00r of relative permittivity and (c) tand of PMMA (diamonds) and two types of composites with UCNT = 0.015 (circles), UCNT = 0.030 (triangles), and UCNT = 0.073 (squares). Open symbols, PMMA/CNT; filled symbols, PMMA–CNT. The e00r (r) values calculated from Eq. (2) are drawn by solid lines in Fig. 5b.

368

CARBON

σAC [S cm−1]

10

8 5 ( 2 0 1 5 ) 3 6 3 –3 7 1

The e00r of a material can be converted the AC conductivity rAC by the following equation:

0

10

−1

10

−2

10

−3

10

−4

rAC ¼ 2pf e0 e00r ¼ rDC þ 2pf e0 e00r ðDPÞ :

10−5 10

−6

10

−7

10

−8

10

−9

0.001

0.01

0.01

1

f [GHz] Fig. 6 – AC conductivity rAC of PMMA (diamonds) and two types of composites with UCNT = 0.015 (circles), UCNT = 0.030 (triangles), and UCNT = 0.073 (squares). Open symbols, PMMA/CNT; filled symbols, PMMA–CNT.

PMMA–CNT originates from the e00r (DP) for both the CNTs and the PMMA matrix. Therefore, a little increase in EMI SE for PMMA–CNT with the increase in UCNT might be mainly attributed to the e00r (DP) of the CNTs.

(a)

ð3Þ

Fig. 6 shows rAC of the two types of composites. For PMMA/CNT, no f-dependence of rAC is observed at the low frequencies. According to Eq. (3), the contribution of e00r (DP) for rAC is negligible, and the constant rAC value corresponds to the rDC of the composite. On the other hand, rAC is increased with the increase in f for PMMA–CNT because the rDC is much smaller and the contribution of e00r (DP) is dominant. From the obtained results mentioned above, the conductivity of a polymer/CNT composite, that is, the conductive network of the CNTs is much more important for the EMI shielding than the intrinsic conductivity of the CNT in the f range of 0.001–1 GHz. It is well-known that the conductivity enhances the EMI shielding, and the highly conductive polymer/CNT composites exhibit superior EMI SE [6–29]. This is because the conductive network spreading overall the composite can strongly reflect and absorb the EM wave as if the network is a conductive bulk plate [7]. EM waves can penetrate only the short distance from the surface of a conductive plate, known as a skin depth [12,15,18,39]. On the other hand, it is found that the electrically-isolated and uniformly dispersed CNT transmits the EM waves because the CNT itself has very little e00r in the f range of 0.001–1 GHz. In addition, the size of the CNTs is much smaller than the wavelength; the f range of 0.001–1 GHz is equivalent to the wavelength k

35

SER, SEA [dB]

30 25 20 15 10 5 0 0

(b)

10

20

30

40

50

60

70

80

90

100

110

60

70

80

90

100

110

f [GHz] 35

SER, SEA [dB]

30 25 20 15 10 5 0 0

10

20

30

40

50

f [GHz] Fig. 7 – Frequency f dependence of shielding effectiveness by reflection (SER) and absorption (SEA) of (a) PMMA/CNT and (b) PMMA–CNT with UCNT = 0.015 (circles), UCNT = 0.030 (triangles), and UCNT = 0.073 (squares). Gray symbols, SER; black symbols, SEA. The thickness of the specimens is around 0.57 mm.

CARBON

range of 300–0.3 m. The much smaller CNTs does not even scatter the EM waves [14].

3.3. EMI shielding properties of PMMA–CNT in the f range of 5.6–110 GHz. Although PMMA–CNT was highly transparent to EM waves at less than 1 GHz, the EM property was drastically changed at more than 1 GHz. The EMI SE of PMMA–CNT was investigated in the f range of 5.6–110 GHz by the free space method [10]. For the free space method, reflectance R is obtained besides T, and hence we can divide EMI SE into two contributions as the following equation [6,21,35]: SE ¼ 10logT ¼ 10logfT=ð1  RÞg  10log ð1  RÞ ¼ SEA þ SER ð4Þ where SEA and SER are SEs by absorption and reflection, respectively. Fig. 7 shows SEA and SER of the two types of composites in the f range of 5.6–110 GHz. The SEA’s of PMMA–CNT are significantly increased with the increase in f, especially at UCNT = 0.073 (Fig. 7b); although the SEA of PMMA–CNT with UCNT = 0.073 is only 1.6 dB at 5.6 GHz, has reached 26 dB at 110 GHz. In other words, PMMA–CNT becomes a good EM wave absorbing material at the higher f range. Similarly, PMMA/CNT also has an increase in SEA (Fig. 7a). This is due to the e00r (DP) of the CNT itself because the contribution of e00r (r) for e00r in Eq. (1) is very small in the f range. It is suggested that the e00r (DP) of the CNT is attributed to interfacial polarization in the CNT [14]. Because the movement of electrons is not much faster than the electric field in the gigahertz frequencies, the e00r (DP) of the CNT itself is rapidly increased over 1 GHz. On the other hand, the SER’s of PMMA–CNT are similar to decay curves owing to the wave interference effect [46]. In this f range, because the k of an EM wave is comparable to the thickness d of a specimen, the wave interference between the reflected waves must be considered in order that the interaction of the EM wave is correctly understood. In short, R takes the maximum values under the condition for constructive interference expressed by the following equation: 2nd ¼ ðm  1=2Þk

369

8 5 (2 0 1 5) 3 6 3–37 1

ð5Þ

where n is the refractive index of the specimen and m is a positive integer. According to Eq. (5), as n is higher, peak-top f is lower. As an example, we estimated the f at which R takes the first maximum (m = 1) for PMMA–CNT with UCNT = 0.073 using the er value at 1 GHz in Fig. 5. The calculated value of 18 GHz was well consistent with the first peak-top at around 15 GHz for the SER of PMMA–CNT with UCNT = 0.073 in Fig. 7b. Also, the total EMI SE (SEA + SER) is shown in Fig. 8. At UCNT = 0.073, the SE of PMMA–CNT with 0.57 mm thickness is as high as 29 dB at 110 GHz, though is slightly lower than that of PMMA/CNT (35 dB). As mentioned above, at more than 1 GHz, the intrinsic conductivity of the CNT becomes important for the EMI shielding in addition to the conductive network of the CNTs as reported by some researchers [15,18,24,39]. Furthermore, in this study, it is demonstrated that the intrinsic conductivity of the CNT predominantly contributes to the EMI shielding at around 100 GHz. However, the reason that the intrinsic conductivity of the CNT is more important for the EMI shielding seems somewhat different from that previously proposed. Although Chung suggested that the mobile electrons in CNTs strongly reflect EM waves [39], the SER of PMMA–CNT is much smaller than SEA except for wave interference effect in our results. Similar results have been reported in X band (8.2– 12.4 GHz) by Sundararaj et al. [15,22]. Therefore, it is newly suggested that the e00r (DP) caused by the conductivity of the CNT is the scientific criterion for the EMI shielding in the higher f range rather than the reflection by the intrinsic conductivity of the CNT. When the mobility of the electrons in the CNT is slower than the electric field of EM wave, the e00r (DP) of the CNT itself becomes large. If the CNTs with the lower conductivity than present study were used, PMMA–CNT might show the comparable EMI SE at the lower f range. A further study in which different CNTs are used is currently in progress. Lastly, it is noted that the effect of multiple-reflections for PMMA–CNT remains unclear because the contribution of the multiple-reflection for the EMI SE could not be distinguished from our results. It is suggested that the multiplereflections also contribute to the EMI SE for polymer/CNT composites in addition to the absorption and the reflection

35 30

SE [dB]

25 20 15 10 5 0 0

10

20

30

40

50

60

70

80

90

100

110

f [GHz] Fig. 8 – Shielding effectiveness (SE) of PMMA (diamond) and two types of composites with UCNT = 0.015 (circles), UCNT = 0.030 (triangles), and UCNT = 0.073 (squares) in the frequency f range of 5.6–110 GHz. Open circles, PMMA/CNT; filled circles, PMMA– CNT. The thickness of the specimens is about 0.57 mm.

370

CARBON

8 5 ( 2 0 1 5 ) 3 6 3 –3 7 1

[15,39]. More characterization is required to clarify the multiple-reflections for PMMA–CNT. Our PMMA–CNT system provides a new concept for EMI materials: the compatibility of high EMI SE and electrical resistance. Therefore, PMMA–CNT can be used for some applications demanding not only good EMI SE but also electrical insulation. Also, PMMA–CNT exhibits EMI shielding only at the higher f range than 1 GHz because the e00r of the CNT itself is rapidly increased over 1 GHz. For example, this finding enables to produce a shielding material for millimeter-wave radars (e.g. automotive radars: around 76 GHz) with high transparency to EM waves for mobile phone (around 1 GHz). Thus, PMMA–CNT shows promise for selective EMI shielding at extremely high frequency range (30–300 GHz).

4.

Conclusion

In summary, an electromagnetic wave absorbing material with high electrical resistance was produced using polymergrafted carbon nanotubes. In this study, poly(methyl methacrylate) (PMMA)-grafted multiwalled CNTs were prepared, and then dispersed into additional PMMA matrix, yielding highly insulated PMMA–CNT composites. The volume resistivity of PMMA–CNT was as high as 1.3 · 1015 X cm even at the weight fraction of the CNT UCNT = 0.073, which is 14 orders of magnitude higher than that of a conventional PMMA/CNT composites prepared by blending PMMA with the pristine CNTs. The individual CNTs electrically-isolated by the grafted chains in PMMA–CNT transmitted electromagnetic (EM) waves in the frequency range of 0.001–1 GHz, whereas the percolated CNTs in PMMA/CNT with high conductivity hardly did. This result suggests that the intrinsic conductivity of the CNT itself in PMMA–CNT does not contribute to the EM interference (EMI) shielding in the f range of 0.001–1 GHz. On the other hand, PMMA–CNT exhibited good EMI shielding at the higher f range than 1 GHz. At UCNT = 0.073, the EMI SE of PMMA–CNT with 0.57 mm thickness is as high as 29 dB at 110 GHz though is slightly lower than that of PMMA/CNT (35 dB). The SE by absorption of PMMA–CNT is much higher than the SE by reflection. Therefore, it is suggested that the dielectric loss caused by the intrinsic conductivity of the CNT is the scientific criterion for the EMI shielding in the higher f range than 1 GHz. Thus, it is demonstrated that PMMA–CNT has both of the good EMI shielding at extremely high frequency range (30–300 GHz) and the high electrical resistance.

R E F E R E N C E S

[1] Coleman JN, Khan U, Blau WJ, Gun’ko YK. Small but strong: a review of the mechanical properties of carbon nanotube– polymer composites. Carbon 2006;44(9):1624–52. [2] Moniruzzaman M, Winey KI. Polymer nanocomposites containing carbon nanotubes. Macromolecules 2006;39(16):5194–205. [3] Li J, Ma PC, Chow WS, To CK, Tang BZ, Kim JK. Correlations between percolation threshold, dispersion state, and aspect ratio of carbon nanotubes. Adv Funct Mater 2007;17(16):3207–15.

[4] Byrne MT, Gun’ko YK. Recent advances in research on carbon nanotube–polymer composites. Adv Mater 2010;22(15):1672–88. [5] Spitalsky Z, Tasis D, Paragelis K, Costas G. Carbon nanotube– polymer composites: chemistry, processing, mechanical and electrical properties. Prog Polym Sci 2010;5(3):357–401. [6] Kim HM, Lee CY, Joo J, Cho SJ, Yoon HS, Pejakovic´ DA, Yoo JW, Epstein AJ. Electrical conductivity and electromagnetic interference shielding of multiwalled carbon nanotube composites containing Fe catalyst. Appl Phys Lett 2004;84(4):589–91. [7] Yang Y, Gupta MC, Dudley KL, Lawrence RW. Novel carbon nanotube–polystyrene form composites for electromagnetic interference shielding. Nano Lett 2005;5(11):2131–4. [8] Yang Y, Gupta MC, Dudley KL, Lawrence RW. A comparative study of EMI shielding properties of carbon nanofiber and multi-walled carbon nanotube filled polymer composites. J Nanosci Nanotechnol 2005;5(6):927–31. [9] Li N, Huang Y, Du F, He X, Lin X, Gao H, Ma Y, Li F, Chen Y, Eklund PC. Nano Lett 2006;6(6):1141–5. [10] Park K-Y, Lee S-E, Kim C-G, Han J-H. Fabrication and electromagnetic characteristics of electromagnetic wave absorbing sandwich structures. Compos Sci Technol 2006;66(3–4):576–84. [11] Liu Z, Bai G, Huang Y, Ma Y, Du F, Li F, Guo T, Chen Y. Reflection and absorption contributions to the electromagnetic interference shielding of single-walled carbon nanotube/polyurethane composites. Carbon 2007;45(4):821–7. [12] Zhang C-S, Ni Q-Q, Fu S-Y, Kurashiki K. Electromagnetic interference shielding effect of nanocomposites with carbon nanotube and shape memory polymer. Comp Sci Technol 2007;67(14):2973–80. [13] Paton KR, Windle AH. Efficient microwave energy absorption by carbon nanotubes. Carbon 2008;46(14):1935–41. [14] Mdarhri A, Carmona F, Brosseau C, Delhaes P. Direct current electrical and microwave properties of polymer-multiwalled carbon nanotubes composites. J Appl Phys 2008;103(5):054303. [15] Al-Saleh MH, Sundararaj U. Electromagnetic interference shielding mechanisms of CNT/polymer composites. Carbon 2009;47(7):1738–46. [16] Wang L-L, Tay B-K, See K-Y, Sun Z, Tan L-K, Lua D. Electromagnetic interference shielding effectiveness of carbon-based materials prepared by screen printing. Carbon 2009;47(8):1905–10. [17] Thomassin J-M, Huynen I, Jerome R, Detrembleur C. Functionalized polypropylenes as efficient dispersing agents for carbon nanotubes in a polypropylene matrix; application to electromagnetic interference (EMI) absorber materials. Polymer 2010;51(1):115–21. [18] Mahmoodi M, Arjmand M, Sundararaj U, Park S. The electrical conductivity and electromagnetic interference shielding of injection molded multi-walled carbon nanotube/ polystyrene composites. Carbon 2012;50(4):1455–64. [19] Wang H, Wang G, Li W, Wang Q, Wei W, Jiang Z, Zhang S. A material with high electromagnetic radiation shielding effectiveness fabricated using multi-walled carbon nanotubes wrapped with poly(ether sulfone) in a poly(ether ether ketone) matrix. J Mater Chem 2012;22(39):21232–7. [20] Qin F, Brosseau C. A review and analysis of microwave absorption in polymer composites filled with carbonaceous particles. J Appl Phys 2012;111(6):061301. [21] Thomassin J-M, Je´roˆme C, Pardoen T, Bailly C, Huynen I, Detrembleur C. Polymer/carbon based composites as electromagnetic interference (EMI) shielding materials. Mater Sci Eng, R 2013;74(7):211–32.

CARBON

8 5 (2 0 1 5) 3 6 3–37 1

[22] Al-Saleh MH, Saadeh WH, Sundararaj U. EMI shielding effectiveness of carbon based nanostructured polymeric materials: a comparative study. Carbon 2013;60:146–56. [23] Villacorta BS, Hubing TH, Ogale AA. Influence of composite electrical properties on the VHF–UHF electromagnetic shielding characteristics of polyethylene–carbon nanoparticle composites. Comp Sci Technol 2013;89:158–66. [24] Theilmann P, Yun D-J, Asbeck P, Park S-H. Superior electromagnetic interference shielding and dielectric properties of carbon nanotube composites through the use of high aspect ratio CNTs and three-roll milling. Org Electron 2013;14(6):1531–7. [25] Kwon S, Ma R, Kim U, Choi HR, Baik S. Flexible electromagnetic interference shields made of silver flakes, carbon nanotubes and nitrile butadiene rubber. Carbon 2014;68:118–24. [26] Liu X, Yin X, Kong L, Li Q, Liu Y, Duan W, Zhang L, Cheng L. Fabrication and electromagnetic interference shielding effectiveness of carbon nanotube reinforced carbon fiber/ pyrolytic carbon composites. Carbon 2014;68:501–10. [27] Zhao T, Hou C, Zhang H, Zhu R, She S, Wang J. Electromagnetic wave absorbing properties of amorphous carbon nanotubes. Sci Rep 2014;4:5619. [28] Maiti S, Suin S, Shrivastava NK, Khatua BB. A strategy to achieve high electromagnetic interference shielding and ultra low percolation in multiwall carbon nanotube– polycarbonate composites through selective localization of carbon nanotubes. RSC Adv 2014;4(16):7979–90. [29] Ren F, Yu H, Wang L, Saleem M, Tian Z, Ren P. Current progress on the modification of carbon nanotubes and their application in electromagnetic wave absorption. RSC Adv 2014;4(28):14419–31. [30] Connor MT, Roy S, Ezquerra TA, Balta´ Calleja FJ. Broadband AC conductivity of conductor-polymer composites. Phys Rev B 1998;57(4):2286–94. [31] Kilbride BE, Coleman JN, Fraysse J, Fournet P, Cadek M, Drury A, Hutzler S, Roth S, Blau WJ. Experimental observation of scaling laws for alternating current and direct current conductivity in polymer–carbon nanotube composite thin films. J Appl Phys 2002;92(7):4024–30. [32] Chung DDL. Carbon materials for structural self-sensing, electromagnetic shielding and thermal interfacing. Carbon 2012;50(9):3342–53. [33] Yan D-X, Ren P-G, Pang H, Fu Q, Yang M-B, Li Z-M. Efficient electromagnetic interference shielding of lightweight graphene/polystyrene composite. J Mater Chem 2012;22(9):18772–4.

371

[34] Zhou Z, Wang J, Zhuang J, Liu Q. A covalent route for efficient surface modification of ordered mesoporous carbon as high performance microwave absorbers. Nanoscale 2013;5(24):12502–11. [35] Song W-L, Cao M-S, Lu M-M, Bi S, Wang C-Y, Liu J, Yuan J, Fan L-Z. Flexible graphene/polymer composite films in sandwich structures for effective electromagnetic interference shielding. Carbon 2014;66:67–76. [36] Wen B, Cao M, Lu M, Cao W, Shi H, Liu J, Wang X, Jin H, Fang X, Wang W, Yuan J. Reduced graphene oxides: light-weight and high-efficiency electromagnetic interference shielding at elevated temperatures. Adv Mater 2014;26(21):3484–9. [37] Shen B, Zhai W, Zheng W. Ultrathin flexible graphene film: an excellent thermal conducting material with efficient EMI Shielding. Adv Funct Mater 2014;24(28):4542–8. [38] Ameli A, Nofar M, Wang S, Park CB. Lightweight polypropylene/stainless-steel fiber composite foams with low percolation for efficient electromagnetic interference shielding. ACS Appl Mater Interfaces 2014;6(14):11091–100. [39] Chung DDL. Electromagnetic interference shielding effectiveness of carbon materials. Carbon 2001;39(2):279–85. [40] Hayashida K, Tanaka H. Ultrahigh electrical resistance of poly(cyclohexyl methacrylate)/carbon Nanotube composites prepared using surface-initiated polymerization. Adv Funct Mater 2012;22(11):2338–44. [41] Hayashida K. Dielectric properties of polymethacrylategrafted carbon nanotube composites. RSC Adv 2013;3(1):221–7. [42] Catrysse JA, de Goeije M, Steenbakkers W, Anaf L. Correlation between shielding effectiveness measurements and alternative methods for the characterization of shielding materials. IEEE Trans Electromagn Compat 1993;35(4):440–4. [43] Engen GF, Hoer CA. Thru-reflect-line: an improved technique for calibrating the dual six port automatic network analyzer. IEEE Trans Microw Theory Tech 1979;27(12):987–93. [44] Hayashida K, Matsuoka Y. Highly enhanced dielectric constants of barium titanate-filled polymer composites using polymer-grafted carbon nanotube matrix. Carbon 2013;60:506–13. [45] Lestriez B, Maazouz A, Gerard JF, Sautereau H, Boiteux G, Seytre G, Kranbuehl DE. Is the Maxwell–Sillars–Wagner model reliable for describing the dielectric properties of a core–shell particle–epoxy system. Polymer 1998;39(26):6733–42. [46] Born M, Wolf E. Principles of optics: electromagnetic theory of propagation, interference and diffraction of light. 4th ed. Oxford: Pergamon Press; 1970.