carbon fiber composite foams

carbon fiber composite foams

CARBON 6 0 ( 2 0 1 3 ) 3 7 9 –3 9 1 Available at www.sciencedirect.com journal homepage: www.elsevier.com/locate/carbon Electrical properties and ...

3MB Sizes 1 Downloads 28 Views

CARBON

6 0 ( 2 0 1 3 ) 3 7 9 –3 9 1

Available at www.sciencedirect.com

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

Electrical properties and electromagnetic interference shielding effectiveness of polypropylene/carbon fiber composite foams A. Ameli, P.U. Jung, C.B. Park

*

Microcellular Plastics Manufacturing Laboratory, Department of Mechanical and Industrial Engineering, University of Toronto, 5 King’s College Road, Toronto, Ontario, Canada M5S 3G8

A R T I C L E I N F O

A B S T R A C T

Article history:

Foamed and solid polypropylene/carbon fiber (PP–CF) composites containing various CF

Received 1 February 2013

contents (0–10 vol.%) were injection-molded. Foamed composites were achieved using dis-

Accepted 13 April 2013

solved pressurized nitrogen gas. The effects of foaming on the fibers inter-connectivity and

Available online 26 April 2013

orientation, electrical percolation threshold, through-plane electrical conductivity, longitudinal and transversal in-plane conductivities, dielectric permittivity, and electromagnetic interference (EMI) shielding effectiveness (SE) were investigated. Cell growth increased the fibers inter-connectivity by biaxial stretching of the matrix and also changed the fiber orientation. The introduction of foaming reduced the density of the injection-molded samples by 25%, lowered the volume fraction of the percolation threshold from 8.5 to 7 vol.% CF, enhanced the through-plane conductivity up to a maximum of six orders of magnitude, increased the dielectric permittivity and resulted in the increase of the specific EMI SE up to 65%. Moreover, the uniformity of in-plane and through-plane conductivities as well as EMI SE along the injection-molded samples was greatly improved by foaming. The relationships between the microstructure and electrical properties were also established. The results reveal that lightweight conductive products with lower fiber content and enhanced electrical and EMI shielding properties can be fabricated with the aid of injection foam molding for applications in electronics, aerospace and automotive industries.  2013 Elsevier Ltd. All rights reserved.

1.

Introduction

Electronic devices radiate and are affected by electromagnetic interference (EMI). Protecting such devices from incoming EMI is essential to maintain their functionality and integrity, as well as controlling their EMI emission level is required to achieve product acceptance in complying with electromagnetic compatibility standards imposed by governmental agencies [1]. Conductive coatings, metal cabinets, foil laminates and conductive polymer composites (CPCs) are the means of EMI shielding. Metal-based shields, as the most

widely used EMI protectors [2] suffer from drawbacks of being heavy, prone to corrosion, and expensive processing. CPCs containing carbon fiber, carbon black and stainless steel fibers have been extensively investigated in an attempt to overcome the shortcomings of metal-based shields [3–14], and more recently, the EMI shielding capabilities of CPCs with carbon nanotubes (CNT) and carbon nanofibers (CNF) have been studied [15–26]. Compared to the conventional CPCs, composites with carbon nanofillers present lower percolation threshold and superior electrical properties. The applications of such conductive additives are however limited due to their

* Corresponding author: Fax: +1 416 978 0947. E-mail address: [email protected] (C.B. Park). 0008-6223/$ - see front matter  2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.carbon.2013.04.050

380

CARBON

6 0 ( 2 0 1 3 ) 3 7 9 –3 9 1

high processing costs [27]. Therefore, micro-sized additives such as CFs are still the most widely used conductive additives. CPCs are not yet very successful in replacing the metal-based shields due to the high filler loading that is required to achieve the percolation threshold and adequate level of shielding, which adversely affects the processability, weight and economic feasibility of the product. CPCs are usually processed using injection molding as one of the costeffective manufacturing methods. However, in-plane orientation of fibers within the injection-molded products yields preferential orientation of fibers and thus anisotropic electrical conductivity where through-plane conductivity is several orders of magnitude lower than in-plane conductivity [28–33]. Physical foaming can provide a route to address the aforementioned issues related to the processing of CPCs in injection molding. First of all, the blowing agent dissolved in the matrix decreases the composite viscosity by the plasticizing effect of gas and eases the processing [30]. Foam injection molding can also effectively disturb the fiber orientation and decrease the size of the skin layer that has highly oriented fibers and thus enhance the through-plane conductivity [28–30,34]. Moreover, the gas also prevents fiber aggregation and improves the dispersion [34] and distribution [28–30] of the fibers, which helps in lowering the electrical percolation threshold [35–38]. Processability will be further increased when a low fiber content is used to facilitate mixing and shaping while foaming is induced at the last stage to make the fibers percolated. The foaming process can thus improve the shortcomings of the injection-molded solid CPCs. Some works have demonstrated that the percolation and conductivity of CPCs can be improved using foaming [39–41]. Recently, some workers have also tried to develop conductive foams for EMI shielding applications [19,41–43]. However, these efforts have been focused on samples made with batch foaming systems and very little attention has been paid to foamed CPCs made with injection molding process. Motlagh et al. [29] and Thompson et al. [33] found that the throughplane conductivity of injection-molded carbon-filled composites may increase when they are foamed with a chemical blowing agent. They observed that the through-plane conductivity was improved via foaming for composites with carbon black or a mix of carbon black and CF, whereas no improvement was observed from the composites filled with CF only. In our earlier work, it was shown that the through-plane electrical conductivity of injection-molded PP–CF composites, containing 20 wt.% fibers can be increased by physical foaming. It was found that the foam injection molding parameters, specifically void fraction, injection flow rate, melt temperature, and dissolved gas content play important roles on the foams’ electrical conductivity through their effects on the foam morphology of the core and thickness of the skin [30]. Also, the optimum processing and foaming conditions to achieve the highest through-plane conductivity were presented [30]. In this work, PP–CF composites containing various fiber contents (0–10 vol.%) were made using injection molding process and the composites were foamed using dissolved pressurized nitrogen gas as the physical blowing agent. The effects of foaming on the electrical percolation threshold, through-plane conductivity, longitudinal and transversal in-

plane conductivities, dielectric permittivity, and EMI SE were characterized. The fiber orientation, skin-core structure, and cellular morphology were also characterized to derive the relationships between the microstructure and electrical properties of the solid and foamed PP–CF composites were derived.

2.

Experimental procedure

2.1.

Materials

The commercial injection molding grade polypropylene (PP), ZS-761 with a specific gravity of 0.9 g cm3 and a melt flow index around 21 dg min1 (230 C/2.16 kg), supplied by SUNOCO Inc., Philadelphia, USA was used as the base resin. PAN-based chopped CF, with a density of 1.8 g cm3 was used as the additive. Proprietary PP–CF composite, grade Carbo-Rite F261 with 10 vol.% CF was provided by Lubrizol Corp., Wickliffe, Ohio, USA and the lower fiber contents were achieved by diluting the PP–10 vol.% CF through mixing with as-received PP before injection molding. Commercial nitrogen (N2), supplied by Linde Gas, Canada, was used as the environmentally benign physical blowing agent.

2.2.

Experimental setup

A 50-ton Arburg Allrounder 270/320 C injection molding machine (Lossburg, Germany) with a 30-mm diameter screw, and equipped with MuCell technology (Trexel, Inc., Woburn, Massachusetts) was used to fabricate solid and foamed PP– CF composites. The mold contained a rectangular cavity with a fan gate after the sprue. The cavity dimensions were 132 · 108 · 3.2 mm. More details can be found in [44]. Table 1 summarizes the processing parameters used in injection molding of solid and foamed composites.

2.3.

Microstructure and fiber orientation

The microstructures were examined using a JEOL JSM-6060 scanning electron microscope (SEM) following the procedure of [44]. The SEM micrographs were used to measure the skin layer thickness of the samples. The length of CFs was measured by optical microscopy of the composite residue after burning in thermal gravimetric analyzer. To characterize the CF orientation, both the core and skin regions were metallographically polished with a series of Struer polishers. Digital images of the polished surface taken by optical microscopy were quantitatively analyzed to calculate the fiber orientation factors. 3-D orientation factors were calculated using approximately 1000 fibers for each case. In the images taken from the length-thickness (LT) plane, the orientation factors were calculated according to the following equations [28–30]: 1X cos2 a  cos2 b n 1X 2 sin a  cos2 b fT ¼ n 1X 2 sin b fW ¼ n fL ¼

ð1Þ ð2Þ ð3Þ

where L, T and W are axes in the length (machine), thickness, and width directions of the injection-molded sample,

CARBON

6 0 (2 0 1 3) 3 7 9–39 1

381

Table 1 – Processing parameters used in injection molding of solid and foamed composites.

a

Parameter

Solid

Foam

Melt temperature (C) Barrel pressure (MPa) Screw speed (rpm) Metering time (s) Injection flow rate (cm3 s1) Mold temperature (C) Pack/hold pressure (MPa) Pack/hold time (s) Gas injection pressure (MPa) N2 content (wt.%) Void fraction (%)

210 17 400 10 100 30 80 4 N/A N/A N/A

210 17 400 10 100 30 N/Aa N/A 24 0.3 25

N/A = not applicable.

respectively. n is the total number of measured fibers. a and b are the fiber in-plane angle measured from L axis and out-ofplane angle, respectively. a was directly determined by the image analyzer while b was calculated from the dimensions of the major and minor axes (a and b, respectively) of the fiber’s elliptical projection on the working plane using: 1

b ¼ sin ðb=aÞ

ð4Þ

Each fiber orientation factor can vary from 0 (meaning that the fiber is normal to the respective direction) to 1 (meaning that the fiber is parallel to the respective direction). In the case of images taken from the LW plane, the equations were reset accordingly. More details can be found in [28–30,45,46].

2.4.

Electrical conductivity and dielectric permittivity

Disk-shape samples of 20 mm in diameter and 3.2 mm in thickness were cut from the middle location of the injection-molded parts using a die cutter under compression molding. Alpha-A high performance conductivity analyzer by Novocontrol Technologies GmbH & Co. KG was used to measure the through-plane (T-P) electrical conductivity and dielectric permittivity of the disks at a frequency range of 1 · 101 to 3 · 10+5 Hz. For comparative purposes, the direct current T-P conductivity, rDC was taken at the frequency of 1 · 101 Hz. At least four injection-molded sample replications were carried out for each case and the average values were reported. To measure the longitudinal and transversal in-plane (I-P) conductivity, rectangular samples of 40 · 10 mm were cut in the length and width directions of the injection-molded sample at similar locations as those for T-P samples. HP 4339B resistance meter was used for I-P conductivity measurements. In the uniformity studies, the conductivity was measured at three different locations along the sample length, i.e., near the gate, middle and end locations. To measure the T-P conductivity of core and skin regions separately, core samples were made by removing a skin layer of 0.5 mm from both surfaces of the samples and the skin samples were prepared by maintaining only 0.5 mm of sample thickness at one side. In all cases, a voltage of 1.0 V was applied.

Fig. 1 – A schematic of electromagnetic interference (EMI) shielding effectiveness (SE) measurement set-up.

2.5.

EMI SE

EMI SE was measured in the X-band frequency range (8.0– 12.4 GHz) using the set-up schematically shown in Fig. 1. The set-up consists of a programmable network analyzer (PNA) HPE8361A network analyzer, two standard WR-90 coaxial launchers to guide EM wave, and a copper sample holder in between the launchers. The sample was cut from the injection-molded part and sized to 25 · 12.5 mm and fitted in the opening of the sample holder, which had bigger dimensions than the opening of the waveguide launchers (22.5 · 10 mm). The sample holder was then bolted in between the launchers. The incident EM wave had a power of 0 dBm, which corresponds to 1 mW. After calibration of the set-up, the wave transmittance and reflectance were directly measured by the PNA and used to calculate the EMI SE.

3.

Results and discussion

3.1.

Microstructure–conductivity relationships

3.1.1.

Effects of physical foaming on microstructure

Fig. 2a schematically illustrates the overall change of the microstructure with foaming. The fiber inter-connectivity,

382

CARBON

6 0 ( 2 0 1 3 ) 3 7 9 –3 9 1

(a)

Solid

Foam

(b)

Solid

Foam

Core

Skin

Fig. 2 – (a) 2-D schematic illustrating the effects of foaming on the structure of injection-molded composites. (b) Representative SEM micrographs of PP–10 vol.% CF composites showing solid’s core, foam’s core, solid’s skin and foam’s skin. LD and TD represent length and thickness directions, respectively.

fiber orientation, fiber length and skin layer thickness of the composites were changed when foaming was introduced. Fig. 2b shows the microstructure of the core and skin regions of the injection-molded samples for both solid and foamed PP–10 vol.% CF composites. In the core of the solid samples, the fibers were highly oriented in the width direction (normal to the micrograph, Fig. 2a) due to filling and packing stages at the end of injection cycle [28–30]. In the skin region of solid

composites, due to rapid cooling and high shear stresses, the fibers were highly oriented in the flow direction (Fig. 2b). However, as clearly seen in Fig. 2b, the state of fiber distribution changed in both regions when the composites were foamed. Fig. 3 schematically illustrates how the growth of cell during foaming contributed to the increased contact points between fibers. The growth of cells resulted in biaxial

Fig. 3 – Schematic illustration of the effect of cell growth on the fibers inter-connectivity. (a) and (b) represent the fiber alignment before and after cell growth, respectively.

CARBON

383

6 0 (2 0 1 3) 3 7 9–39 1

8

Solid

7

Foam

Frequency (%)

6

4 3 2 1 300

250

200

150

100

50

0 0

Fig. 4 – Fiber orientation factors of solid and foamed PP– 10 vol.% CF composites in the length (L), width (W) and thickness (T) directions, measured at the core and skin of the injection-molded samples. The values are the average of measurements at three different locations of the injectionmolded samples (i.e., near gate, middle and end).

5

CF length (μm) Fig. 5 – CF length frequency of the injection-molded solid and foamed PP–10 vol.% CF composites.

stretching of polymer matrix in which the fibers had preferentially oriented. The spherical growth of cell caused different degrees of displacements and rotations of the fibers depending on the fibers relative location with respect to the cell. This action caused the preferentially oriented fibers to redistribute and thus increased the chance of inter-connectivity. Such effect of foaming action has been also demonstrated in [29]. Quantitative analysis of the mechanics of fibers influenced by the biaxial stretching effect of foaming action is the subject of our future study. The presence of dissolved gas in the composite melt also affected the microstructure. By the addition of gas, the viscosity of the melt decreased and thus the shear stresses applied to the fibers were decreased. This resulted in the reduction of the fibers’ preferential orientation in the flow direction in both skin and core regions. It also contributed to the reduction of fiber breakage during the injection. Moreover, as has been demonstrated in [30], the thickness of skin layer that had highly preferential fiber orientation was also decreased in the presence of dissolved gas. To characterize these effects of foaming on the microstructure, the fiber orientations and length as well as skin layer thickness were measured in both solid and foamed composites. Fig. 4 shows the orientation factors in the length, width and thickness directions, measured in the core and skin of the solid and foamed composites. The highly preferential in-plane fiber orientations in the skin and core regions of solid composites resulted in very low orientation factors in the thickness direction (fT), and consequently should have had an adverse effect on the through-plane conductivity. In the core of foamed composites, a relatively more random distribution of fibers with a significant increase of through-plane orientation was achieved. This change in the state of orientation was a consequence of cell growth (Fig. 3). In the skin region, the severity of one-dimensional orientation in the machine direction was also reduced in foamed composites. As seen in Fig. 4, by foaming, fL decreased while fW and fT increased. As explained earlier, this was attributed to the plasticizing effect of the dissolved gas, which reduced the applied shear stresses on the fibers and thus decreased the fiber orientation in the flow direction. Another factor that affects the conductivity of filled composites is the aspect ratio of the fillers [47,48]. Fig. 5 depicts the final CF length distribution measured for both solid and

foamed PP–10 vol.% CF composites. Overall, relatively similar CF length distribution was achieved in both solid and foamed composites. Approximately 50% of CFs was measured to have lengths greater than 90 and 110 lm in the solid and foamed composites, respectively. These corresponded to the aspect ratios of 12.0 and 14.7, respectively, similar to the reported values in the literature for the same processing history of CF composites [28,29,48]. Slightly higher CF length in the foamed composites was originated from the presence of the dissolved gas during injection that decreased the melt viscosity and slightly decreased the fiber breakage.

3.1.2.

T-P conductivity of skin and core regions

The overall resistance of the full sample (Req) should be governed by the resistance of a series of resistors as the skin (Rs)–core (Rc)–skin (Rs) combination following: Req ¼ Rs þ Rc þ Rs ¼ 2Rs þ Rc

ð5Þ

For a resistor, specific resistance (q) is: q ¼ R  A=t

ð6Þ

where A and t are the area and thickness of the disk shape sample. A remained the same for the skin, core and full samples and noting that conductivity is the reciprocal of resistivity, Eq. (5) can be rewritten as: 1 1 r1 eq ¼ 2ars þ ð1  aÞrc

ð7Þ

where req, rs and rc are the conductivity of the full sample and its skin and core, respectively. a is the ratio between the skin layer thickness and the overall thickness of the sample, which remained unchanged. Eq. (7) describes that the overall conductivity is affected by the conductivities of the skin and core regions as well as their relative size. Therefore, to further demonstrate the effects of structure on the overall conductivity, the T-P conductivity of skin and core of the composites were also measured separately. As seen in Fig. 6, the conductivity of the core was in general higher than that of the corresponding skin in both solid and foamed composites. At 5 and 6.25 vol.% CF cases, this difference was not significant since the fibers were too far apart and not inter-connected even in the core. However, near the threshold content and beyond, the core conductivity was up

384

CARBON

6 0 ( 2 0 1 3 ) 3 7 9 –3 9 1

Fig. 6 – T-P DC electrical conductivity (rDC) measured in the skin and core of the solid and foamed PP–CF composites (the conductivity of the corresponding overall samples, i.e., the skin–core–skin combinations are given in Fig. 8).

to three orders of magnitude higher than the skin conductivity. This indicated that the percolation network at the core had more fiber inter-connectivity, compared to the skin. This should have been originated from a higher through-plane orientation factor (fT) in the core, as shown in Fig. 4. Moreover, the T-P conductivity of both core and skin regions of foamed composites was higher than their corresponding solid ones. This is also consistent with the data representing the through-plane fiber orientation changes with foaming in these two regions (Fig. 4). It is also seen that the conductivity improvement in the core region was greater than that in the skin region upon the introduction of foaming, which is again consistent with changes in fT of these regions which should have originated from the cell growth biaxial stretching effect (Fig. 4). The increase of T-P conductivity in both core and skin regions with foaming should contribute to the improvement of overall conductivity (Eq. (7). It is also noted that with 7.5 and 8.75 CF vol.%, the conductivity of the skin region of the foamed composites were higher than those of the corresponding solid cores. This was originated from the reduced in-plane orientation of fibers in the skin layer caused by the plasticizing effect of dissolved gas (Fig. 4). It is also partially attributed to the decreased skin layer thickness via foaming, which affected the conductivity of both skin samples (500 lm thickness) and full sample. The skin layer thickness is further discussed in Section 3.3.1.

3.2.

Fig. 7 – Variation of T-P AC electrical conductivity (rAC) of solid and foamed composites with frequency. Solid and hollow symbols denote solid and foamed samples, respectively.

of foamed composite was about four orders of magnitude higher than that of the solid one and spanned a much wider range of frequency. In the case of the highest CF content (10 vol.%), the conductivity of both solid and foamed samples became frequency-independent in the tested range of frequency, again with a higher conductivity value for the foamed one. Fig. 8 depicts the T-P rDC vs. CF content for both solid and foamed composites. The CF content of foamed composites is given with respect to both polymer volume and total volume of the foams. To assess the sole effect of foaming action on the conductivity, CF content was considered with respect to the polymer volume and to include the density reduction effect of foaming as well, the CF content was calculated with respect to the total volume of the foams. The conductivity clearly exhibited an insulator–conduction transition. The transition started earlier (at 6.25 vol.% CF) in the foamed composites compared to that (7.5 vol.% CF) in the solid ones, indicating that the introduction of foaming lowered the electrical percolation threshold of the PP–CF composites.

Overall T-P conductivity

Fig. 7 depicts the broadband T-P electrical conductivity as a function of frequency for the solid and foamed PP–CF composites with various CF contents. At 5 vol.% CF content, the conductivity of both solid and foamed samples decreased proportionally with frequency in the whole range. This is the characteristic of insulating materials such as PP, indicating that the electrical properties of the composites were controlled by the matrix (rPP = 1 · 1015 S cm1 at 1 · 101 Hz) and the fibers were not in contact. When the CF content increased to 7.5 vol.%, the conductivity showed a conduction behavior that is represented by r = rDC + rAC where rAC is the frequency-dependent part of the total electrical conductivity. This law considers a critical frequency below which conductivity becomes frequency-independent, known as rDC [39]. With this CF content, the frequency-independent conductivity

Fig. 8 – T-P DC electrical conductivity (rDC) of solid and foamed PP–CF composites as a function of CF content (the conductivity of the corresponding skin and core regions are given in Fig. 6).

CARBON

Consequently, at 7.5 and 8.75 vol.% CF, foamed composites exhibited conductivities up to 6 orders of magnitude higher, compared to that of the corresponding solid samples. According to the percolation statistical model [49], the percolation threshold of the solid and foamed composites should have been around 8.5 and 7 vol.%, respectively. It is noted that an accurate percolation analysis was not possible due to the limited number of data points beyond the threshold. It is also seen in Fig. 8 that to achieve the same level of electrical conductivity within a sample with a certain volume, the required final CF vol.% for the foamed composites is significantly lower than that for the solid ones. For example, the foamed sample with a final CF content of 6.5 vol.% resulted in a higher conductivity than a solid sample containing 10 vol.% CF.

3.3.

Uniformity and isotropy of conductivity

3.3.1.

Uniformity of the T-P electrical conductivity

385

6 0 (2 0 1 3) 3 7 9–39 1

To assess its uniformity, the T-P conductivity was measured at three different locations along the injection-molded sample; i.e., near the gate, middle, and near the end of sample, located 50 mm apart. The conductivity measurements of the solid composites showed that the degree of nonuniformity was more severe in lower melt temperatures. Therefore, a lower melt temperature (190 C) was selected for this part to better capture the foaming effects. Fig. 9 depicts the T-P electrical conductivity of the skin and core for the solid and foamed PP–10 vol.% CF composites. In the solid composites, the T-P conductivity of both skin and core of near the gate location was about 5–6 orders of magnitude lower than the corresponding values in the middle and end locations. This resulted in a very nonuniform T-P conductivity along the

sample. However, as foaming introduced to the composites, the T-P conductivity increased in all the locations with the greatest improvement near the gate location. In the middle and end locations, the conductivity increase by foaming was about one order of magnitude whereas this increase was five orders of magnitude near the gate location. This change by foaming resulted in significantly improved uniformity of conductivity along the injection-molded samples in both the core and skin regions. As Table 2 shows, in the skin region of near the gate location, the fiber orientation factor in the flow direction was the highest due to the longest exposure of the shear flow and the highest speed of the flow at that location and thus resulted in the lowest through-plane fiber orientation and lowest T-P conductivity. Similarly, in the core region of near the gate location, the effects of shear flow and packing pressure were greater compared to the middle and end locations, and consequently, a more severe in-plane orientation of fibers formed in this location with a resultant lower fT and T-P conductivity. Overall, the conductivity trends of Fig. 9 correlated well with the states of fiber orientation; i.e., fT increased from near the gate location toward the end of the samples in both solid and foamed composites and also increased when the foaming was introduced to the composites. Compared to the other locations, the cell density was the highest and the average cell size was the smallest in the core of near the gate location and it is believed that this finer cellular morphology also helped achieving the greatest improvement in the T-P conductivity of the core at this location. Knowing the conductivity of the skin and core regions and the thickness of the skin layer (Table 2), Eq. (7) can be used to calculate the conductivity of the full sample. Fig. 10 shows the

Fig. 9 – T-P DC electrical conductivity (rDC) of (a) skin and (b) core of the solid and foamed PP–10 vol.% CF composites.

Table 2 – Skin layer thickness and fiber orientation factors in the longitudinal (fL), transversal (fW) and thickness (fT) directions, measured at different locations. Skin

Core

Skin thickness (mm)

fL

fW

fT

fL

fW

fT

Solid Near gate Middle End

0.61 0.43 0.33

0.81 0.76 0.66

0.18 0.20 0.32

0.01 0.03 0.04

0.33 0.39 0.44

0.63 0.56 0.49

0.04 0.05 0.07

Foama Near gate Middle End

0.49 0.38 0.28

0.74 0.71 0.58

0.21 0.33 0.35

0.05 0.06 0.07

0.27 0.35 0.39

0.54 0.44 0.41

0.19 0.21 0.21

a

a

Standard deviation was less than 24% of the average values in all the cases.

386

CARBON

6 0 ( 2 0 1 3 ) 3 7 9 –3 9 1

inter-connectivity caused by cell growth. Moreover, the fiber orientation state showed a more random distribution of fibers in the foamed cases (Table 2), which is known to increase the I-P conductivity [49]. One possible reason for the decrease of the I-P conductivity at the end location might be the larger and more elongated cells with lower cell density at this location. It has been shown that such morphology can cause a decrease in the conductivity [30]. Overall, foaming improved the uniformity of the I-P conductivity along the injection-molded samples as well. The IP conductivity of foamed sample in both longitudinal and transverse directions varied only within the same order of magnitude when the composites were foamed. Moreover, as Table 3 shows, foaming also significantly decreased the anisotropy of the conductivity defined as the ratio of the I-P and T-P conductivities (rL/rT and rW/rT). The range of anisotropy decreased from 8.7 · 107 to 7.3 · 102 in the solid composite to only 9.7 · 103 to 2 · 101 in the foamed composites.

Fig. 10 – Measured and calculated T-P DC electrical conductivity (rDC) of full samples of solid and foamed PP– 10 vol.% CF composites.

measured and calculated T-P conductivity of the full solid and foamed PP–10 vol.% CF composites. It is seen that the greatest increase of conductivity in the skin and core regions and also the maximum decrease in the skin layer thickness of this location via foaming resulted in significant (five orders of magnitude) increase in the overall conductivity near the gate location and thus significantly improved the uniformity of the overall T-P conductivity. The difference between the conductivity of near the gate and end locations was decreased from seven orders of magnitude in the solid composite to only three orders of magnitude when the sample was foamed. Moreover, the measured and calculated T-P conductivities of the full samples agreed reasonably well and the predicted values fell within the scatter range of the measurement replications, indicating the validity of the used equation and the accuracy of the skin layer thickness measurements.

3.3.2.

I-P electrical conductivity

Fig. 11 shows the I-P conductivity of the solid and foamed PP– 10 vol.% CF composites in the longitudinal and transversal directions. As expected, the I-P conductivities were higher than their corresponding T-P conductivities. Foaming increased both the longitudinal and transversal conductivities of near the gate and middle locations while slightly decreased the conductivity of the end location. The I-P conductivity increase via foaming was in agreement with the increased fiber

3.4.

Dielectric permittivity analysis

Fig. 12 depicts the variation of the dielectric permittivity (e 0 ) in the solid and foamed composites at various CF contents. Fig. 13 also gives the dielectric permittivity of the solid and foamed composites, measured at 0.1 MHz, as a function of CF content. It is seen that below the percolation threshold, the addition of CFs did not change the frequency-independent behavior of the permittivity. However, when the CF content was near the percolation threshold, the permittivity started to exhibit a frequency-dependent behavior. At higher frequencies, the permittivity slowly increased by the increase of the CF content and ranged only between 2.3 and 33.8 at 0.1 MHz. At lower frequencies, however, the permittivity increased by several orders of magnitude, when percolation was achieved. For example, the permittivity of the solid PP– 10 vol.% CF exceeded 5 · 104 at 0.1 Hz. For the CF contents near the threshold or above, the permittivity initially decreased with the increase of frequency until it reached a crossover frequency beyond which remained approximately

Fig. 11 – (a) Longitudinal and (b) transversal I-P DC electrical conductivity (rDC) of solid and foamed PP–10 vol.% CF composites.

Table 3 – Anisotropy of conductivity in the solid and foamed PP–10 vol.% CF composites. rL, rW and rT are longitudinal, transversal and through-plane DC electrical conductivities, respectively. Solid rL/rT Near gate Middle End

Foam rW/rT

7

8.7 · 10 8.4 · 103 7.3 · 102

rL/rT 7

4.1 · 10 2.8 · 103 6.4 · 102

rW/rT 3

9.7 · 10 1.2 · 103 2.9 · 101

2.1 · 103 5.2 · 102 2.0 · 101

CARBON

387

6 0 (2 0 1 3) 3 7 9–39 1

Fig. 12 – Dielectric permittivity spectra of (a) solid and (b) foamed PP–CF composites at several CF contents.

Fig. 13 – Dielectric permittivity of solid and foamed PP–CF composites as a function of CF content, measured at a frequency of 0.1 MHz.

unchanged. As seen in Fig. 12b, this crossover frequency had greater values for the composites having a higher CF content. Similar trends have been reported for other conductive composites [50,51]. The real part of the dielectric permittivity, which is discussed here, is related to the displacement current and mainly affected by the polarization (localized charges) inside the material. In CPCs, polarization of matrix, conductive filler and interfacial polarization all can play role depending on the frequency range [52–55]. The polymer matrix polarization can be effective only at very high frequencies, i.e., optical frequency range. Therefore, in the frequency range studied here, the variations of the permittivity should be originated from the polarization of the matrix/fiber interface and/or polarization of the fibers [52–54]. The fact that the dielectric permittivity at low frequencies increased significantly only when the CF content reached the percolation threshold indicates that the permittivity enhancement is largely due to the interfacial polarization. According to Mawxell Wagner Sillars (MWS) effect [56,57], when a current flows across the two-materials interfaces, charges can be accumulated at the interface between the two materials that have different relaxation times and the giant increase in the permittivity caused by this effect can be predicted by the percolation theory [52]. In the PP–CF composites, the fibers were separated by thin barriers of dielectric regions (i.e., PP matrix), and the adjacent fibers formed a micro-capacity whose effective surface increased drastically near the percolation threshold and caused the giant increase in the permittivity, as has also been shown in [52,56,57].

Comparing the permittivity behaviors of the solid and foamed composites containing 7.5 vol.% CF (Fig. 12) clearly shows that the percolative network was still absent in the solid composite and no significant increase was observed in its permittivity, as was the case for its electrical conductivity (Fig. 8), while an effective interfacial polarization could occur in the foamed PP–7.5 vol.% CF composite, resulting in a significantly increased permittivity at lower frequencies. Overall, the permittivity of the foamed composites was higher than that of the corresponding solid counterparts. The higher permittivity of the foamed composites can be attributed to the enhanced interfacial polarization through (a) localization of the fibers by creating a gaseous phase, (b) decreased in-plane orientation by foaming (Fig. 5), which increased the fiber orientation in the thickness direction, and (c) decreased fiber-tofiber distance caused by the biaxial stretching of the matrix during cell growth. All these actions might have contributed to the formation of more effective capacity between adjacent fibers with a thin layer of matrix in between. Higher permittivity with less oriented fibers has been previously reported for injection-molded carbon nanotube/polystyrene composites [54]. Fig. 13 also shows the permittivity of the foamed composites as a function CF volume percent with respect to the total volume of the foams, which includes the density reduction effect of foaming as well. Similar to the T-P conductivity (Fig. 8), to achieve a certain level of permittivity in a given volume, the CF content required for the foamed composites is significantly lower than that of the corresponding solid ones. For example, to achieve a permittivity of about 25, the CF final vol.% is 6.5 and 10 for the foam and solid cases, respectively, which accounts for 35% less usage of CF, when foaming is used.

3.5. Electromagnetic effectiveness (SE)

interference

(EMI)

shielding

EMI SE is a measure of the material’s ability to attenuate EM waves intensity. For an EM radiation, EMI SE is the logarithm of the ratio of incident power (Pi) to transmitted power (Pt) in decibel, i.e., SE = 10 log (Pi/Pt). For example, a SE of 20 corresponds to the blocking of 99% of EM incident wave. Fig. 14 shows the EMI SE of solid and foamed PP–CF composites in the X-band frequency range for various CF contents. EMI SE was relatively frequency-independent in the X-band range and had greater values at higher CF contents in both solid and foamed composites. As Fig. 15 shows, the foamed com-

388

CARBON

6 0 ( 2 0 1 3 ) 3 7 9 –3 9 1

Fig. 14 – X-band frequency range electromagnetic interference (EMI) shielding effectiveness (SE) of (a) solid and (b) foamed PP– CF composites at several CF contents.

Foam (CF content with respect to total volume) Foam (CF content with respect to polymer volume) Solid 25 EMI SE (dB)

20 15 10 5 0 0

2.5

5 7.5 CF Content (vol. %)

10

Fig. 15 – Electromagnetic interference (EMI) shielding effectiveness (SE) of solid and foamed PP–CF composites as a function of CF content.

posites presented higher SE values than their solid counterparts with the same CF content. The SE values reported in Fig. 15 are the grand average values in the X-band frequency range and for four sample replications. At 10 vol.% CF, SE of the foamed composites reached to about 24.9 dB corresponding to 99.7% blocking of EM waves in the X-band frequency range. At the same CF content, the solid composites presented SE of about 19.8 dB. At even 7.5 vol.% CF, the foamed composites reached a SE of 16.3 dB, which is in the range required for computer devices, i.e., 15–20 dB [58]. Compared to the similar level of EM blocking reported for 7 wt.% multiwalled carbon nanotube [59], using 7.5–10 vol.% carbon fiber in the foamed composites suggests an economical alternative. Fig. 15 also shows the EMI SE of the foamed composites as a function of CF volume percent with respect to the total volume of the foams. It is seen that to achieve a certain level of SE in a given volume, the final CF content required for the foamed composites was significantly lower than that of the corresponding solid ones. For example, to achieve a SE of about 20, the CF final vol.% is 6.5 and 10 for the foam and solid cases, respectively, which accounts for 35% less usage of CF, when foaming is used. In the foamed composites, a final CF vol.% of 5.6 was sufficient to achieve a SE greater than 15 dB. Moreover, the specific EMI SE, defined as SE value divided by the sample density [60], increased up to 65% when the composites were foamed.

Fig. 16 – Absorption and reflection shielding effectiveness (SE) of the solid and foamed CF–PP composites with different CF contents.

Wave absorption and reflection are the major electromagnetic attenuation mechanisms. In order to better understand the effect of foaming on the shielding mechanisms, the contribution of both absorption and reflection mechanisms to the total SE (SET) should be quantified. The utilized EMI SE characterization setup directly measures the transmitted power (T) and reflected power (R) and given that the incident power (I) is known (1 mW), the absorbed power (A) can be calculated using the power balance equation: I¼TþAþR

ð8Þ

Shielding by absorption and reflection can thus be calculated using the following equations [58]: SET ¼ 10 logðI=TÞ ¼ SEA þ SER

ð9Þ

SEA ¼ 10 logððI  RÞ=TÞ

ð10Þ

SER ¼ 10 logðI=ðI  RÞÞ

ð11Þ

Fig. 16 shows the contribution of each shielding mechanism to the total SE of the composites. The shielding by reflection was similar for both solid and foamed composites, reaching to a maximum of about 4–5 dB in the percolation threshold region. The shielding by absorption was the dominant attenuation mechanism and continuously increased with the increase of the CF content in both solid and foamed composites. For instance, in solid and foamed PP–10 vol.% CF composites, 77% and 81% of the shielding was achieved by only absorption. As seen Fig. 16, the shielding by absorption was higher for the foamed composites compared to that of the solid ones with the same CF content and thus resulted in the enhanced total EMI SE for the foamed composites.

CARBON

6 0 (2 0 1 3) 3 7 9–39 1

Fig. 17 – Electromagnetic interference (EMI) shielding effectiveness (SE) of solid and foamed PP–10 vol.% CF composites of Fig. 10.

The improved SE via absorption in the foamed composites is attributed to two factors: (a) change of fiber orientation state and increased inter-connectivity of the fibers as they enhanced the conductivity and permittivity. Increase of EMI SE by more random distribution of conductive filler has already been reported [49,54]. It has also been shown that the higher conductivity and real dielectric permittivity results in higher SE [58]. Another mechanism that contributed to the EM wave absorption was the wave scattering in the sample’s core due to the cellular structure of the foamed samples, as explained in [41]. Fig. 17 shows the EMI SE of solid and foamed PP–10 vol.% CF composites of Fig. 10. The introduction of foaming increased the EMI SE of the composites by approximately 30% near the gate and 15% in the middle location. The EMI SE remained approximately unchanged in the end location. The greatest improvement of SE was achieved near the gate location, which should have been originated from the maximum increase of the T-P and I-P conductivities via foaming in this location (Figs. 10 and 11). The SE improvement was of low extent in the middle location as was the case for the conductivities. In the end location, upon foaming, the T-P conductivity increased while the I-P conductivity decreased and the combination led to relatively unchanged SE. Moreover, all these changes caused by foaming resulted in a great improvement of the SE uniformity along the injection foam molded composites. The SE of near the gate location was 26% lower than that of the end location in the solid composites, while it decreased to only 6% (in the range of replication scatter) in the foamed samples.

4.

Summary

The microstructure, fiber orientation, fiber breakage and electrical properties of injection-molded solid and foamed PP–CF composites with various CF contents were characterized. The effects of foaming and the skin-core structure on the percolation threshold, through-plane and in-plane electrical conductivity, dielectric permittivity, EMI SE, and the uniformity of these parameters were investigated. Introduction of physical foaming changed the microstructure configuration of the PP–CF composites through the biaxial stretching effect of cell growth on the fibers and the plasticizing effect of the dissolved gas on the viscosity of the composites. The biaxial stretching increased the fibers inter-connectivity and fiber orientation in the thickness direc-

389

tion, and the dissolved gas reduced the fiber breakage as well as the severity of the fibers’ preferential orientation and also decreased the size of the skin region. Foaming reduced the density of the injection-molded PP– CF composites by about 25% and improved their electrical properties. These improvements correlated well with the changes that foaming made in the microstructure of the PP– CF composites. Introduction of foaming (i) lowered the volume fraction of the electrical percolation threshold from 8.75 to 7 vol.% CF, (ii) enhanced the through-plane conductivity up to a maximum of six orders of magnitude, (iii) increased the dielectric permittivity, and (iv) resulted in the increase of the specific EMI SE up to 65%. Moreover, the uniformity of microstructure, the T-P and I-P conductivities, and EMI SE along the injection-molded samples was greatly improved upon the introduction of foaming. The results obtained in this investigation reveal that lightweight conductive products with an low filler content and enhanced electrical and EMI shielding properties can be developed with the aid of foaming in the injection molding process for applications in the electronics, aerospace and automotive industries.

Acknowledgements The work was financially supported by MITACS, Ontario, Canada, and the masterbatch was kindly supplied by Dr. Feina Cao of Lubrizol Corp.

R E F E R E N C E S

[1] Paul CR. Introduction to electromagnetic compatibility. 2nd ed. Hoboken, NJ: John Wiley and Sons Inc.; 2006. [2] Schlechter M. EMI/RFI: materials and technology. [accessed 20.08.07]. [3] Sun X, Cheng Y, Wu H. Research on microstructure and electrical resistivity of injection molded metallic fiber-filled polymer composites. Adv Mater Res 2012;472:748–52. [4] Al-Saleh MH, Sundararaj UT. Electromagnetic interference (EMI) shielding effectiveness of PP/PS polymer blends containing high structure carbon black. Macromol Mater Eng 2008;293(7):621–30. [5] Jana PB, Mallick AK, De SK. Effects of sample thickness and fiber aspect ratio on EMI shielding effectiveness of carbon fiber filled polychloroprene composites in the X-band frequency range. IEEE Trans Electromagn Compat 1992;34(4):478–81. [6] Bagwell RM, McManaman JM, Wetherhold RC. Short shaped copper fibers in an epoxy matrix: their role in a multifunctional composite. Comp Sci Technol 2006;66(3– 4):522–30. [7] Wetherhold RC, McManaman JM. EMI shielding effectiveness of copper/epoxy composites. Polym Polym Comp 2005;13(7):657–68. [8] Tan ST, Zhang MQ, Rong MZ, Zeng HM, Zhao FM. Properties of metal fibre filled thermoplastics as candidates for electromagnetic interference shielding. Polym Polym Comp 2001;9(4):257–62. [9] Chen D, Yang J, Chen G. The physical properties of polyurethane/graphite nanosheets/carbon black foaming conducting nanocomposites. Comp Part A 2010;41(11):1636–8.

390

CARBON

6 0 ( 2 0 1 3 ) 3 7 9 –3 9 1

[10] Krueger QJ, King JA. Synergistic effects of carbon fillers on shielding effectiveness in conductive nylon 6,6- and polycarbonate-based resins. Adv Polym Technol 2003;22(2):96–111. [11] Chang H, Kao M-J, Huang K-D, Kuo C-G, Huang S-Y. Electromagnetic shielding effectiveness of thin film with composite carbon nanotubes and stainless steel fibers. J Nanosci Nanotechnol 2011;11(2):1754–7. [12] Das NC, Khastgir D, Chaki TK, Chakraborty A. Electromagnetic interference shielding effectiveness of carbon black and carbon fibre filled EVA and NR based composites. Comp Part A 2000;31(10):1069–81. [13] Chen C-S, Chen W-R, Chen S-C, Chien R-D. Optimum injection molding processing condition on EMI shielding effectiveness of stainless steel fiber filled polycarbonate composite. Int Commun Heat Mass Transfer 2008;35(6): 744–9. [14] Wong KH, Pickering SJ, Rudd CD. Recycled carbon fibre reinforced polymer composite for electromagnetic interference shielding. Comp Part A 2010;41(6): 693–702. [15] Al-Saleh MH, Sundararaj U-T. A review of vapor grown carbon nanofiber/polymer conductive composites. Carbon 2009;47(1):2–22. [16] Das NC, Maiti S. Electromagnetic interference shielding of carbon nanotube/ethylene vinyl acetate composites. J Mater Sci 2008;43(6):1920–5. [17] Arjmand M, Mahmoodi M, Gelves GA, Park S, Sundararaj U-T. Electrical and electromagnetic interference shielding properties of flow-induced oriented carbon nanotubes in polycarbonate. Carbon 2011;49(11):3430–40. [18] Kim HM, Kim K, Lee CY, Joo J, Cho SJ, Yoon HS, et al. Electrical conductivity and electromagnetic interference shielding of multiwalled carbon nanotube composites containing Fe catalyst. Appl Phys Lett 2004;84(4):589–91. [19] Yang YL, Gupta MC, Dudley KL, Lawrence RW. Novel carbon nanotube–polystyrene foam composites for electromagnetic interference shielding. Nano Lett 2005;5(11):2131–4. [20] Arjmand M, Apperley T, Okoniewski M, Sundararaj U-T. Comparative study of electromagnetic interference shielding properties of injection molded versus compression molded multi-walled carbon nanotube/polystyrene composites. Carbon 2012;50(14):5126–34. [21] Lee BO, Woo WJ, Kim MS. EMI shielding effectiveness of carbon nanofiber filled poly(vinyl alcohol) coating materials. Macromol Mater Eng 2001;286(2):114–8. [22] Zhang CS, Ni QQ, Fu SY, Kurashiki K. Electromagnetic interference shielding effect of nanocomposites with carbon nanotube and shape memory polymer. Comp Sci Technol 2007;67(14):2973–80. [23] Huang Y, Li N, 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. [24] Li N, Huang Y, Du F, He XB, Lin X, Gao HJ, et al. Electromagnetic interference (EMI) shielding of single-walled carbon nanotube epoxy composites. Nano Lett 2006;6(6):1141–5. [25] Liu ZF, Bai G, Huang Y, Ma YF, Du F, Li FF, et al. Reflection and absorption contributions to the electromagnetic interference shielding of single-walled carbon nanotube/polyurethane composites. Carbon 2007;45(4):821–7. [26] Yuen SM, Ma CCM, Chuang CY, Yu KC, Wu SY, Yang CC, et al. Effect of processing method on the shielding effectiveness of electromagnetic interference of MWCNT/PMMA composites. Comp Sci Technol 2008;68(3–4):963–8. [27] O’Connell M, editor. Carbon nanotubes: properties and applications. Boca Raton: Taylor and Francis; 2006.

[28] Motlagh GH, Hrymak AN, Thompson MR. Properties of a carbon filled cyclic olefin copolymer. J Polym Sci Part B 2007;45(14):1808–20. [29] Motlagh GH, Hrymak AN, Thompson MR. Improved throughplane electrical conductivity in a carbon-filled thermoplastic via foaming. Polym Eng Sci 2008;48(4):687–96. [30] Ameli A, Jung PU, Park CB. Through-plane electrical conductivity of injection-molded polypropylene/carbon-fiber composite foams. Comp Sci Technol 2013;76:37–44. [31] Weber M, Kamal MR. Microstructure and volume resistivity of composites of isotactic polypropylene reinforced with electrically conductive fibers. Polym Comp 1997;18(6):726–40. [32] Dani A, Ogale AA. Electrical percolation behavior of shortfiber composites: experimental characterization and modeling. Comp Sci Technol 1996;56(8):911–20. [33] Thompson MR, Motlagh GH, Oxby KJ, Hrymak AN. Multiple percolation in a carbon-filled polymer composites via foaming. J Appl Polym Sci 2010;115(2):646–54. [34] Okamoto M, Nam PH, Maiti M, Kotaka T, Nakayama T, Takada M, et al. Biaxial flow-induced alignment of silicate layers in polypropylene/clay nanocomposite foam. Nano Lett 2001;1(9):503–5. [35] Po¨tschke P, Bhattacharyya AR, Janke A. Carbon nanotubefilled polycarbonate composites produced by melt mixing and their use in blends with polyethylene. Carbon 2004;42(5– 6):965–9. [36] Villmow T, Po¨tschke P, Pegel S, Ha¨ussler L, Kretzschmar B. Influence of twin-screw extrusion conditions on the dispersion of multi-walled carbon nanotubes in a poly(lactic acid) matrix. Polymer 2008;49(16):3500–9. [37] Pegel S, Po¨tschke P, Petzold G, Alig I, Dudkin SM, Lellinger D. Dispersion, agglomeration, and network formation of multiwalled carbon nanotubes in polycarbonate melts. Polymer 2008;49(4):974–84. [38] Mu¨ller MT, Krause B, Po¨tschke P. A successful approach to disperse MWCNTs in polyethylene by melt mixing using polyethylene glycol as additive. Polymer 2012;53(15):3079–83. [39] Antunes M, Mudarra M, Velasco JI. Broad-band electrical conductivity of carbon nanofibre-reinforced polypropylene foams. Carbon 2011;49(2):708–17. [40] Hermant MC, Verhulst M, Kyrylyuk AV, Klumperman B, Koning CE. The incorporation of single-walled carbon nanotubes into polymerized high internal phase emulsions to create conductive foams with a low percolation threshold. Comp Sci Technol 2009;69(5):656–62. [41] Zhang HB, Yan Q, Zheng WG, He Z, Yu ZZ. Tough graphene– polymer microcellular foams for electromagnetic interference shielding. ACS Appl Matter Interfaces 2011;3(3):918–24. [42] Thomassin J-M, Pagnoulle C, Bednarz L, Huynen I, Jerome R, Detrembleur C. Foams of polycaprolactone/MWNT nanocomposites for efficient EMI reduction. J Mater Chem 2008;18:792–6. [43] Eswaraiah V, Sankaranarayanan V, Ramaprabhu S. Functionalized graphene–PVDF foam composites for EMI shielding. Macromol Mater Eng 2011;296(10):894–8. [44] Lee JWS, Wang J, Yoon JD, Park CB. Strategies to achieve a uniform cell structure with a high void fraction in advanced structural foam molding. Ind Eng Chem Res 2008;47(23):9457–64. [45] Bay RS, Tucker CL. Fiber orientation in simple injection moldings. Part I: Theory and numerical methods. Polym Comp 1992;13(4):317–31. [46] Bay RS, Tucker CL. Fiber orientation in simple injection moldings. Part II: Experimental results. Polym Comp 1992;13(4):332–41. [47] Li J, Ma PC, Chow WS, To CK, Tang BZ, Kim J-K. Correlations between percolation threshold, dispersion state, and aspect

CARBON

[48]

[49]

[50]

[51]

[52]

[53]

[54]

6 0 (2 0 1 3) 3 7 9–39 1

ratio of carbon nanotubes. Adv Funct Mater 2007;17(16):3207–15. Yi JY, Choi GM. Percolation behavior of conductor-insulator composites with varying aspect ratio of conductive fiber. J Electroceramics 1999;3(4):361–9. Kortschot MT, Woodhams TR. Computer simulation of the electrical conductivity of polymer composites containing metallic fillers. Polym Comp 1988;9(1):60–71. Panwar V, Park J-O, Park S-H, Kumar S, Mehra RM. Electrical, dielectric, and electromagnetic shielding properties of polypropylene–graphite composites. J Appl Polym Sci 2010;115(3):1306–14. Po¨tschke P, Dudkinb SM, Alig I. Dielectric spectroscopy on melt processed polycarbonate–multiwalled carbon nanotube composites. Polymer 2003;44(17):5023–30. Yuan JK, Yao SH, Dang ZM, Sylvestre A, Genestoux M, Bai J. Giant dielectric permittivity nanocomposites: realizing true potential of pristine carbon nanotubes in polyvinylidene fluoride matrix through an enhanced interfacial interaction. J Phys Chem 2011;115(13):5515–21. Jiang M-J, Dang Z-M, Bozlar M, Miomandre F, Bai J. Broadfrequency dielectric behaviors in multiwalled carbon nanotube/rubber nanocomposites. Appl Phys 2009;106(8). 084902-1–6. Mahmoodi M, Arjmand M, Sundararaj U-T, Park S. The electrical conductivity and electromagnetic interference

[55]

[56]

[57]

[58]

[59]

[60]

391

shielding of injection molded multi-walled carbon nanotube/ polystyrene composites. Carbon 2012;50(4):1455–64. Watts PCP, Hsu WK, Randall DP, Kroto HW, Walton DRM. Nonlinear current–voltage characteristics of electrically conducting carbon nanotube–polystyrene composites. Phys Chem Chem Phys 2002;4:5655–62. Effros AL, Shklovskii BI. Critical behavior of conductivity and dielectric constant near the metal–non-metal transition threshold. Phys Status Solidi B 1976;76(2):475–85. Tamura R, Lim E, Manaka T, Iwamotoa MJ. Analysis of pentacene field effect transistor as a Maxwell–Wagner effect element. Appl Phys 2006;100(11). 114515-1–7. Al-Saleh MH, Gelves GA, Sundararaj U-T. Copper nanowire/ polystyrene nanocomposites: lower percolation threshold and higher EMI shielding. Comp Part A 2011;42(1):92–7. Yang YL, 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. 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;22:18772–4.