Electrical and dielectric properties of polypropylene nanocomposites based on carbon nanotubes and barium titanate nanoparticles

Composites Science and Technology 71 (2011) 1706–1712

Contents lists available at SciVerse ScienceDirect

Composites Science and Technology journal homepage: www.elsevier.com/locate/compscitech

Electrical and dielectric properties of polypropylene nanocomposites based on carbon nanotubes and barium titanate nanoparticles Chang-Rong Yu a, Da-Ming Wu a, Ying Liu a, Hui Qiao a, Zhong-Zhen Yu a,⇑, Aravind Dasari b,⇑, Xu-Sheng Du c, Yiu-Wing Mai c a b c

State Key Laboratory of Organic–Inorganic Composites, Beijing University of Chemical Technology, Beijing 100029, China School of Materials Science & Engineering (Blk N4.1), Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore Centre for Advanced Materials Technology (CAMT), School of Aerospace, Mechanical and Mechatronic Engineering J07, The University of Sydney, Sydney, NSW 2006, Australia

a r t i c l e

i n f o

Article history: Received 29 November 2010 Received in revised form 11 April 2011 Accepted 30 July 2011 Available online 10 August 2011 Keywords: A. Nanocomposites A. Carbon nanotubes B. Electrical properties B. Surface treatments

a b s t r a c t Functional polypropylene (PP) nanocomposites were prepared by melt compounding with multiwalled carbon nanotubes (MWNT) as the electrically conductive component and barium titanate (BT) spherical nanoparticles as the ferroelectric component. To make PP electrically conductive, more than 3 wt.% MWNT is required. Surface modification of either MWNT or BT with titanate coupling agent further improves the electrical conductivity of the PP/MWNT/BT ternary nanocomposites. Interestingly, by modifying both MWNT and BT, 2 wt.% MWNT are sufficient to make the ternary nanocomposite electrically conductive. In addition, the incorporation of MWNT greatly increases the dielectric permittivity of PP/ BT nanocomposites. However, to retain a low dielectric loss, the MWNT loading should be slightly less than the percolation threshold of the nanocomposites. The improved electrical conductivity and dielectric properties make the ternary nanocomposites attractive in practical applications. Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction Since the discovery of carbon nanotubes (CNT), their superior electrical [1,2], thermal [3,4] and mechanical properties [5,6] made them very attractive for incorporation in polymer materials for a range of applications including electrostatic dissipation, electromagnetic interference shielding [7,8], electro-thermal nanoprobe [9], dielectric materials with high dielectric permittivity [10–12] and conductive adhesives. Although solution blending [13] and in situ polymerization techniques are widely used to prepare polymer/CNT nanocomposites, considering the economics and industrial needs, melt compounding is still the preferred route [14]. Generally, with the increase of CNT loading, a three-dimensional conducting network is expected to form in a polymer matrix whereby the resulting nanocomposites show a sharp transition from electrically insulating to conducting behavior. But to form an effective conducting network, uniform dispersion of CNT in the polymer matrix is required. Thus, for this purpose, surface modification/functionalization of the CNT is often conducted [15]. Kodgire et al. [16] used a sodium salt of 6-aminohexanoic acid to modify CNT in order to obtain a homogeneous dispersion in polyamide 6 (PA 6) matrix during melt compounding, hence leading to a large improvement in electrical conductivity of the nanocomposites. ⇑ Corresponding authors. Fax: +86 10 6442 8582 (Z.-Z. Yu), +65 6790 9081 (A. Dasari). E-mail addresses: [email protected] (Z.-Z. Yu), [email protected] (A. Dasari). 0266-3538/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.compscitech.2011.07.022

Another method to improve the electrical conductivity of a polymer/conducting filler nanocomposite is to take advantage of the volume-exclusion concept [17–21]. Meincke et al. [18] blended acrylonitrile–butadiene–styrene with PA 6/CNT nanocomposites by melt compounding. Due to the selective localization of the CNT in PA 6 phase, the ternary nanocomposites showed a reduced percolation threshold. Recently, Dasari et al. [21] added maleated polyethylene–octene copolymer (POE-g-MA) to a PA 6/MWNT blend by melt compounding and obtained improvements in not only electrical conductivity but also toughness of the ternary nanocomposite. The presence of MWNT in PA 6 matrix and their absence in the POE-g-MA particles increased the electrical conductivity caused by the volume exclusion effect of POE-g-MA; meanwhile, the dispersed POE-g-MA particles participated in toughening processes similar to PA 6/POE-g-MA binary blend, yielding greatly improved toughness. Instead of an organic third component, inorganic fillers were also utilized [22–25]. For example, Ma et al. [24] used a combination of CNT/carbon black (CB) nanoparticles in epoxy and found a synergistic effect on electrical conductivity resulting in a lower percolation threshold. Liu and Grunlan. [25] introduced electrically insulating clay layers into epoxy/CNT nanocomposites and showed that even with 2 wt.% clay, the percolation threshold could be reduced from 0.05 wt.% to 0.01 wt.% of CNT. The combination of CNT with other inorganic fillers [26] was also used to improve the dielectric properties. Barium titanate (BT) is a ferroelectric crystal with high dielectric permittivity and

1707

C.-R. Yu et al. / Composites Science and Technology 71 (2011) 1706–1712

3 mm-thick plates of PP nanocomposites were made by compression molding at 200 °C with a pressure of 10 MPa. 2.4. Characterization The surface chemistry of BT and MWNT before and after modification with the titanate coupling agent was characterized by Nicolet 6700 Fourier transform infrared spectroscopy (FT-IR). The melt viscosities of PP nanocomposites were measured on a TA ARES rheometer with 25 mm diameter parallel plates at 190 °C with frequencies varying from 0.1 to 40 rad/s. To evaluate the dispersion quality of BT and MWNT in PP matrix, freeze-fractured surfaces of

(a) Transmittance (%)

has been widely used to enhance the dielectric permittivity of polymers [27,28]. However, due to the low dielectric permittivity of most polymers, high loading of BT is required, which often impairs the mechanical properties and processability of the polymers. It is currently a challenge to decrease BT loading in a polymer matrix but retain a satisfactory dielectric permittivity. Surface modification of BT was found to be an effective way of improving its dispersion along with dielectric permittivity of the polymer composites [29–34]. Kim et al. [31] reported that phosphonic acids formed organic shells on the BT surface, leading to well dispersed BT in polymeric films with high dielectric permittivity and high dielectric strength. In addition, an abrupt increase in the dielectric permittivity was observed around the percolation threshold of polymer nanocomposites with conducting fillers [35–37]. Nanocomposites in the neighborhood of percolation threshold could become capacitors with good characteristics of charge storage. Dang et al. [38] reported PVDF/MWNT/BT nanocomposite with high dielectric permittivity based on hybrid fillers and percolation theory. In the present work, functional PP nanocomposites were prepared by melt compounding using MWNT as the electrically conductive component and BT nanoparticles as the ferroelectric component. BT is also expected to have an effect on the electrical conductivity of PP/MWNT nanocomposites via the volume-exclusion mechanism. Similarly, the presence of conducting MWNT is believed to improve the dielectric permittivity of the PP/BT nanocomposites. Further, the effects of surface modification of MWNT and BT on microstructures, electrical and dielectric properties of the PP nanocomposites are studied.

m-MWNT 2848 2917 1631

1151

3433

3500

2. Experimental work

MWNT

3000

2500

2000

1500

1000

500

-1

Wavenumber (cm )

2.1. Materials Polypropylene (S1003) with a melt flow index of 3 g/10 min was purchased from Yanshan Petroleum Chemical Co. Ltd. (China). Vapor-grown MWNTs with a purity of 95% were supplied by Chengdu Organic Chemicals (China). The MWNT has an outer diameter of 20–30 nm and a length of 10–30 lm and it has 1.23 wt.% of the carboxyl group. BT ceramic powders with an average diameter of 100 nm were bought from Shandong Guoci Functional Materials (China). Titanate coupling agent (TTS) with a chemical formula of (CH3)2CHO-Ti-(COO-C17H35R)3 was purchased from Anhui Hongsheng Fine Chemical Co. Ltd. (China). Silver paste (CD-2004) with a silver content of 55 wt.% was provided by the Great Wall Gold– Silver Refinery (China). 2.2. Surface modification of BT and MWNT Prior to modification, BT and MWNT were dried in an air-circulating oven at 80 °C for 16 h. The dried BT or MWNT powders were suspended in ethanol under mechanical stirring. Subsequently, 4 wt.% titanate coupling agent (based on BT or MWNT) dissolved in ethanol was added. The resulting suspension was heated at 90 °C until the solvent was evaporated while stirring continuously. The modified powders were dried in the oven at 100 °C for 16 h and are designated hereafter as m-BT and m-MWNT. 2.3. Preparation of PP nanocomposites PP nanocomposites were prepared by melt compounding using a two-roll mill at 170 °C for 20 min. To minimize the thermal degradation of PP, 0.1 wt.% of antioxidant (1010) based on PP component was added. For the ternary nanocomposites, PP was first melt compounded with BT at 170 °C for 10 min, followed by blending with MWNT at the same temperature for another 10 min. Finally,

Fig. 1. (a) FT-IR spectra of MWNT before and after surface modification with the titanate coupling agent. (b) and (c) SEM micrographs of the PP nanocomposites with 2 wt.% of unmodified and modified MWNT, respectively. White arrows indicate pulled out nanotubes.

1708

C.-R. Yu et al. / Composites Science and Technology 71 (2011) 1706–1712

Transmittance (%)

(a)

BT m-BT 2854 2925

1631

1426

3433

568

4000

3500

3000

2500

2000

1500

1000

500

Wavenumber (cm-1)

Fig. 2. (a) FT-IR spectra of BT nanoparticles before and after surface modification with the titanate coupling agent. SEM micrographs of PP nanocomposites: (b) PP/40 wt.% BT; (c) PP/40 wt.% m-BT; (d) PP/40 wt.% BT/2 wt.% MWNT; and (e) PP/40 wt.% m-BT/2 wt.% m-MWNT.

the nanocomposites were observed with a Hitachi S-4700 fieldemission scanning electron microscope (FESEM). Alternate current (AC) electrical conductivities and dielectric properties were measured with an Agilent 4294A impedance analyzer at ambient temperature within the frequency range of 102–106 Hz. The silver paste was used to ensure good contact between the specimen and the electrodes. 3. Results and discussion 3.1. Influence of surface modification on dispersion of MWNT and BT nanofillers MWNT naturally form bundles owing to their strong inter-tubular van der Waals attractions. To disentangle them and improve their dispersion in PP matrix, a titanate coupling agent with long

alkyl chains was used to modify the MWNT surface. The chemical interaction between them is expected due to the presence of the carboxyl groups of MWNT and the reactive groups of the titanate coupling agent. Fig. 1a shows the FTIR spectra of MWNT before and after the surface modification. MWNT exhibits broad peaks at 3433, 1631 and 1151 cm1, corresponding to –OH stretch vibration, asymmetric –COO stretch vibration and C–O stretch vibration, respectively. After modification with the titanate coupling agent, new peaks appear at 2917 cm1 and 2848 cm1, which are, respectively, asymmetric and symmetric –CH2 stretch vibration from the long alkyl chain (–C17H35) of the coupling agent [39]. Also, the positive charge group +Ti(OCOR)3 induces physical adsorption with negative charges (such as –OH and –COOH) on the surface of MWNT. These results are reflected in the SEM micrographs of freeze-fractured surfaces of the PP nanocomposites with 2 wt.% MWNT before and after the modification (Fig. 1b and c). Besides

1709

C.-R. Yu et al. / Composites Science and Technology 71 (2011) 1706–1712 2

10

-2

10

-4

10

-6

10

-8

10

Neat PP

-10

10

2% MWNT

-12

3% MWNT

10

5% MWNT -14

10

2

3

10

10

4

10

5

10

6

10

Frequency (Hz) 2

10

3.2. Electrical conductivity

(b)

0

10

AC Conductivity (S/m)

Fig. 3a shows the variation of electrical conductivity (r) as a function of frequency for the PP/MWNT nanocomposites from which it is evident that r increases with % MWNT and the percolation threshold is between 3 and 5 wt.%. At 5 wt.% MWNT, there is a dramatic increase in the electrical conductivity by 5–6 orders of magnitude compared to neat PP over the entire frequency range of 102–106 Hz. Note that at 5 wt.% MWNT loading, the nanocomposite exhibits a frequency-independent behavior with a conductivity of 102 S/m, indicating the formation of a threedimensional interconnecting network by the dispersed MWNT. At the lower loading of 2 wt.% and 3 wt.% of MWNT, however, the nanocomposites are still electrically insulating and their conductivity is frequency-dependent being controlled by inter-filler tunneling rather than the filler itself. After the surface modification of MWNT, the percolation threshold is reduced from 3–5 wt.% to 2–3 wt.% (see Fig. 3b). It is clear that r of PP with 3 wt.% MWNT is low and depends on frequency, but for PP with the same amount of m-MWNTs, r is greatly increased (>103 S/m) and almost frequency-independent. Such a high electrical conductivity is close to that of PP nanocomposites with 5 wt.% MWNT (Fig. 3a). The effect of BT nanoparticles on r of PP/MWNT nanocomposites is shown in Fig. 4. At 1 wt.% MWNT, addition of 40 wt.% BT has little effect on r and the ternary nanocomposite is still electrically insulating. This is because 1 wt.% MWNT is still too far to form a conducting network. With 2 wt.% MWNT, although the addition of BT decreases r by more than one order of magnitude, r of the ternary nanocomposite is still higher than the binary nanocomposite with 1 wt.% MWNT. It should be noted that the presence of BT increases the viscosity of the PP matrix (see below). It is wellknown that the higher viscosity of a polymer matrix usually results in a higher percolation threshold of conducting fillers [41,42]. Hence, Ma et al. [42] used acetyl tributyl citrate (ATBC), a plasticizer, to decrease the melt viscosity of polylactic acid (PLA) which in turn decreased the percolation threshold of the PLA/CB nanocomposites. The PLA/ATBC/CB ternary nanocomposites exhibited low electrical percolation thresholds of 0.52, 1.2, 2.5 and 2.7 vol.% CB with 30, 20, 10 and 0 wt.% ATBC, respectively. In the present case with 2 wt.% MWNT (Fig. 4a), the viscosity-dispersion effect is predominant compared to the volume-exclusion effect of the BT nanoparticles, which plausibly explains the decreased conductivity of PP filled with 2 wt.% MWNT and 40 wt.% BT. When the MWNT is increased to 3 wt.%, the PP/MWNT/BT ternary nanocom-

(a)

0

10

AC Conductivity (S/m)

the even distribution of MWNT in Fig. 1c compared to Fig. 1b, many fractured MWNT are still embedded within the matrix (see Fig. 1c). Similar to MWNT, BT nanoparticles were also modified with the titanate coupling agent. Fig. 2a shows the FTIR spectra of BT and mBT. The absorption peak at 3433 cm1 is related to the stretch vibration of –OH, and the peak at 568 cm1 belongs to Ti–O vibration of BT. The peak of 1426 cm1 is probably due to the stretch vibration of CO2 arising from the residual barium carbonate 3 (BaCO3) in BT [40]. After modification, there are new peaks at 2925 and 2854 cm1, confirming the presence of the long alkyl chains on m-BT surface. The dispersion of the unmodified BT nanoparticles is relatively homogeneous in the PP matrix though some aggregates and pullout of BT nanoparticles are observed (Fig. 2b). After surface modification, there is little pullout of the m-BT nanoparticles, demonstrating an enhanced compatibility (Fig. 2c). Similar observations (i.e., improved interaction between nanoparticles and matrix leading to reduced amount of pullout/debonding) are also evident from the PP/BT/MWNT and PP/m-BT/m-MWNT ternary nanocomposites (Fig. 2d and 2e).

-2

10

-4

10

-6

10

-8

2% MWNT 3% MWNT 2% m-MWNT 3% m-MWNT

10

-10

10

-12

10

2

10

3

10

4

10

5

10

6

10

Frequency (Hz) Fig. 3. AC conductivity versus frequency for PP nanocomposites with different loading of (a) MWNT and (b) m-MWNT.

posite exhibits again a significant improvement in electrical conductivity independent of frequency (see Fig. 4b). Obviously, 3 wt.% MWNT is considered in the percolating regime where a conductive network of the MWNT will form. The presence of 40 wt.% BT significantly reduces the volume available for MWNT to occupy and the volume-exclusion effect is predominant, which results in an electrically conductive ternary nanocomposite. Similar to the positive effect of m-MWNT on electrical conductivity, m-BT also increases r of the ternary nanocomposites by over one order of magnitude (Fig. 5). It should, however, be noted that r of ternary nanocomposites with 2 wt.% MWNT and 40 wt.% BT (or m-BT) still depend on frequency. When both m-MWNT and m-BT are used, 2 wt.% m-MWNT suffices to make the ternary composite electrically conductive and almost frequency-independent with a conductivity of 105 S/m at 102 Hz. With 3 wt.% m-MWNT, the conductivity of the ternary nanocomposite can reach 102 S/m at 102 Hz. The positive contribution of the surface modification on electrical conductivity is closely related to the influence of melt viscosity on the dispersion of MWNT. Fig. 6 shows the effect of surface modifications of 2 wt.% MWNT and/or 40 wt.% BT on the melt viscosity of the PP nanocomposites at 190 °C. Similar to neat PP, all nanocomposites exhibit shear-thinning behavior. Under an applied shear force, the entangled molecular chains of PP become oriented and lead to the reduction of the number of entanglements and

1710

C.-R. Yu et al. / Composites Science and Technology 71 (2011) 1706–1712 -2

(a)

10

(a) Viscosity (Pa*s)

AC Conductivity (S/m)

-4

10

-6

10

-8

10

-10

10

-12

4

PP PP + 2% MWNT PP + 2% m-MWNT

1% MWNT 1% MWNT + 40% BT 2% MWNT 2% MWNT + 40% BT

10

10

10

3

10-1

-14

10

2

3

10

10

4

5

10

10

100

101

102

Frequency (rad/s)

6

10

Frequency (Hz)

(b) (b)

0

-2

Viscosity (Pa*s)

AC Conductivity (S/m)

10 10

-4

10

-6

10

10

4

-8

PP PP + 40% BT PP + 40% m-BT

10

-10

10

3% MWNT 3% MWNT + 40% BT

10

3

-12

10

10

10

2

10

3

10

4

10

5

10

-1

10

6

0

10

1

10

2

Frequency (rad/s)

Frequency (Hz)

6

10

(c)

AC Conductivity (S/m)

10

Viscosity (Pa*s)

Fig. 4. AC conductivity versus frequency for PP/MWNT nanocomposites with and without 40 wt.% unmodified BT nanoparticles: (a) 1 wt.% and 2 wt.% MWNT, and (b) 3 wt.% MWNT.

0

10

-2

10

-4

10

-6

10

-8

5

10

4

10

3

10

10

10

-12

10

-14 2

10

3

10

4

10

5

10

-1

10

0

10

1

10

2

Frequency (rad/s)

40% BT + 2% MWNT 40% BT + 3% MWNT 40% m-BT + 2% MWNT 40% m-BT + 3% MWNT 40% m-BT + 2% m-MWNT 40% m-BT + 3% m-MWNT

-10

10

PP PP + 40% BT PP + 40% BT + 2% MWNT PP + 40% m-BT + 2% m-MWNT

Fig. 6. Melt viscosity versus frequency at 190 °C for PP binary and ternary nanocomposites.

6

10

Frequency (Hz) Fig. 5. AC conductivity versus frequency for PP/BT/MWNT ternary nanocomposites.

of PP nanocomposite compared to the ternary nanocomposite with unmodified nanofillers. (See Fig. 6c). Thus, it is reasonable to conclude that the reduced viscosity and the volume-exclusion effect are two key factors which have an influence on the conductivity and percolation threshold of the nanocomposite. 3.3. Dielectric properties of PP with unmodified nanofillers

hence the melt viscosity. The addition of MWNT or BT increases the melt viscosity of PP, which is further increased by simultaneously adding both MWNT and BT nanofillers. However, the surface modification of both MWNT and BT clearly decreases the melt viscosity

For PP/BT nanocomposites, whilst the dielectric permittivity increases with BT loading, the efficiency is not satisfactory; even with 80 wt.% BT, the dielectric permittivity of PP is only 18

1711

C.-R. Yu et al. / Composites Science and Technology 71 (2011) 1706–1712

where e and em are dielectric permittivity of nanocomposite and polymer matrix, respectively, f is volume fraction of MWNT, fc is percolation threshold, and s is a critical exponent equal to 1.0. Hence, it is expected that at slightly lower loadings of MWNT than the percolation threshold, PP/BT/MWNT ternary nanocomposites would exhibit higher dielectric permittivity. Fig. 7 plots the dielectric permittivity e as a function of frequency for PP/40 wt.% BT/ MWNT ternary nanocomposites. e of the nanocomposites with less than 3 wt.% unmodified MWNT is almost independent of frequency, while e of those with 3 wt.% MWNT decreases rapidly at frequencies greater than 104 Hz. Based on the percolation theory [35], as fMWNT u u1 , where x is equal to 2pf and u is a critical ? fc, r / x , e / x exponent. The data for r of the nanocomposite with fMWNT = 0.03 (Fig. 4c) gives u  1.0 which is a little higher than the normal value of 0.7 from the percolation theory [36]. At a given frequency, the addition of MWNT improves the dielectric permittivity of the PP/ BT nanocomposites, which is related to the formation of nanocapacitors between MWNT and PP/BT. As long as the loading of MWNT is insufficient to form a conductive network, they are isolated and covered by the insulating PP matrix, hence forming a large number of nano-capacitors in which the insulating PP/BT components can be considered as the dielectric layers with higher e while the conducting MWNTs act as the electrodes [45]. Additionally, the incorporation of MWNT in PP/BT nanocomposites changes the frequency-dependence of e. Hence, the dielectric permittivity of the nanocomposites with less than 3 wt.% MWNT shows weak frequency-dependence owing to their insulating nature, while the strong frequency-dependence of the ternary nanocomposite with 3 wt.% MWNT is a direct result of its conducting feature. With 3 wt.% MWNT, the ternary nanocomposite shows a gigantic e. But it cannot be used as a dielectric material due to its rather high dielectric loss (Fig. 8) resulting from electric leakage. Hence, the optimal content of MWNT should be slightly lower than the percolation threshold of the ternary nanocomposites, which is 2 wt.% in the current ternary system with BT nanoparticles.

Dielectric Permittivity

ð1Þ

(a)

2% MWNT 2% m-MWNT 2% MWNT + 40% BT 2% MWNT + 40% m-BT

40

30

20

10

0 2

3

10

10

4

5

10

10

6

10

Frequency (Hz) 5

10

(b)

3% MWNT + 40% BT 3% MWNT + 40% m-BT 3% MWNT

4

Dielectric Permittivity

e=em ¼ ðfc  f Þs

50

10

3

10

2

10

1

10

0

10

2

3

10

10

4

5

10

10

6

10

Frequncy (Hz)

tan δ

(Fig. 7). It is also noted that PP and its BT nanocomposites exhibit stable dielectric constants throughout the whole frequency range due to their insulating and less polar nature [43]. From the percolation power law (Eq. (1)) [44], the dielectric permittivity can be enhanced remarkably near the percolation threshold:

10

5

10

4

10

3

10

2

10

1

10

0

(c)

3% MWNT 3% MWNT + 40% BT 3% MWNT + 40% m-BT 2% MWNT + 40% m-BT

10 -1 10 -2 3

Dielectric Permittivity

10

10 -3 10

40% BT 40% BT + 1% MWNT 40% BT + 2% MWNT 40% BT + 3% MWNT 80% BT PP

2

10

2

10

3

10

4

10

5

10

6

Frequncy (Hz) Fig. 8. Dielectric permittivity (a, b) and dielectric loss (c) versus frequency for PP/ MWNT nanocomposites without and with 40 wt.% of BT or m-BT nanoparticles.

3.4. Influence of surface modification of nanofillers on dielectric properties of PP

1

10

0

10

2

10

3

10

4

10

5

10

6

10

Frequency (Hz) Fig. 7. Dielectric permittivity versus frequency for PP/40 wt.% BT nanocomposites with 0–3 wt.% MWNT.

The influences of BT and m-BT on dielectric permittivity (e) of PP/MWNT and PP/m-MWNT nanocomposites are shown in Fig. 8. For the binary nanocomposite with 2 wt.% m-MWNT, e is greatly enhanced from 5 to 15 at 102 Hz. Similarly, using BT and MWNT simultaneously is also very efficient to improve e of PP nanocomposites. Use of m-BT instead of BT further improves e of the ternary nanocomposites. At 2 wt.% MWNT, PP with 40 wt.% m-BT has a much higher e than its unmodified counterpart. The

1712

C.-R. Yu et al. / Composites Science and Technology 71 (2011) 1706–1712

dielectric permittivity of the former reaches 28 at 102 Hz, nearly 3 times that of the latter and 7 times that of neat PP. More importantly, its dielectric loss is kept as low as 0.07 at 102 Hz. When the frequency is 106 Hz the dielectric loss is 0.2. The titanate coupling agent acts as an effective passivation layer, reduces the concentration of the polar groups (ionizable hydroxyl) on the BT surface minimizing the amount and mobility of the charge carriers usually associated with the surface [31,46]. All these ultimately decrease the dielectric loss. With 3 wt.% MWNT, the addition of either BT or m-BT yields huge dielectric permittivity (Fig. 8b). Such nanocomposites are, however, already electrically conductive and have rather high dielectric losses (Fig. 8c). 4. Conclusions Polypropylene nanocomposites using electrically conductive MWNT and ferroelectric BT nanoparticles as nanofillers were prepared by melt compounding. The nanocomposites show a dramatic increase in conductivity by 6 orders of magnitude when the MWNT loading is increased from 3 to 5 wt.% at 102 Hz. Surface modification of MWNT by the titanate coupling agent greatly improves the electrical conductivity of PP nanocomposites and reduces the percolation threshold. When MWNT and BT are both modified, 2 wt.% m-MWNT is sufficient to make the ternary nanocomposite electrically conductive, which is almost independent of frequency. The reduced viscosity due to the surface modification and volumeexclusion effect of BT nanoparticles are responsible for the greatly increased conductivity and decreased percolation threshold of the PP nanocomposites. Dielectric permittivity of the PP nanocomposites is greatly improved by simultaneously using BT and MWNT. Surface modification of BT further enhances the dielectric permittivity of the ternary nanocomposites. At a fixed loading of 2 wt.% MWNT, the dielectric permittivity of the PP nanocomposite with 40 wt.% m-BT reaches 28 at 102 Hz and, importantly, its dielectric loss is maintained low. Acknowledgements Financial supports from the National Natural Science Foundation of China (50873006, 51073012) and the Program for New Century Excellent Talents in Universities, Ministry of Education of China (NCET-08-0711) are gratefully acknowledged. References [1] Grunlan JC, Mechrabi AR, Bannon MV, Bahr JL. Adv Mater 2004;16:150–4. [2] Xu XB, Li ZM, Shi L, Bian XC, Xiang ZD. Small 2007;3:408–11. [3] Gojny FH, Wichmann MHG, Fiedler B, Kinloch IA, Bauhofer W, Windle AH, et al. Polymer 2006;47:2036–45.

[4] Haggenmueller R, Guthy C, Lukes JR, Fischer JE, Winey KI. Macromolecules 2007;40:2417–21. [5] Coleman JN, Khan U, Blau WJ, Gun’ko YK. Carbon 2006;44:1624–52. [6] Coleman JN, Khan U, Gun’ko YK. Adv Mater 2006;18:689–706. [7] Li N, Huang Y, Du F, He XB, Lin X, Gao HJ, et al. Nano Lett 2006;6:1141–5. [8] Al-Saleh MH, Sundararaj U. Carbon 2009;47:1738–46. [9] Lee J, Liao A, Pop E, King WP. Nano Lett 2009;9:1356–61. [10] Zhang J, Mine M, Zhu D, Matsuo M. Carbon 2009;47:1311–20. [11] Yang C, Lin Y, Nan CW. Carbon 2009;47:1096–101. [12] Logakis E, Pandis Ch, Peoglos V, Pissis P, Pionteck J, Pötschke P, et al. Polymer 2009;50:5103–11. [13] Dang ZM, Wang L, Yin Y, Zhang Q, Lei QQ. Adv Mater 2007;19:852–7. [14] Chen GX, Li YJ, Shimizu H. Carbon 2007;45:2334–40. [15] Xie XL, Mai Y-W, Zhou XP. Mater Sci Eng R 2005;49:89–112. [16] Kodgire PV, Bhattacharyya AR, Bose S, Gupta N, Kulkarni AR, Misra A. Chem Phys Lett 2006;432:480–5. [17] Gubbels F, Blacher S, Vanlathem E, Jerome R, Deltour R, Brouers F, et al. Macromolecules 1995;28:1559–66. [18] Meincke O, Kaempfer D, Weickmann H, Friedrich C, Vathauer M, Warth H. Polymer 2004;45:739–48. [19] Xu XB, Li ZM, Yang MB, Jiang S, Huang R. Carbon 2005;43:1479–87. [20] Li YJ, Shimizu H. Macromolecules 2008;41:5339–44. [21] Dasari A, Yu ZZ, Mai Y-W. Polymer 2009;50:4112–21. [22] Sun Y, Bao HD, Guo ZX, Yu J. Macromolecules 2009;42:459–63. [23] Bao HD, Guo ZX, Yu J. Polymer 2008;49:3826–31. [24] Ma PC, Liu MY, Zhang H, Wang SQ, Wang R, Wang K, et al. ACS Appl Mater Interfaces 2009;1:1090–6. [25] Liu L, Grunlan JC. Adv Funct Mater 2007;17:2343–8. [26] Kong LB, Li S, Zhang TS, Zhai JW, Boey FYC, Ma J. Prog Mater Sci 2010;55:840–93. [27] Choudhury A. Mater Chem Phys 2010;121:280–5. [28] Sonoda K, Juuti J, Moriya Y, Jantunen H. Compos Struct 2010;92:1052–8. [29] Iijima M, Sato N, Lenggoro IW, Kamiya H. Colloid Surf A-Physicochem Eng Asp 2009;352:88–93. [30] Beier CW, Cuevas MA, Brutchey RL. Langmuir 2010;26:5067–71. [31] Kim P, Jones SC, Hotchkiss PJ, Haddock JN, Kippelen B, Marder SR, et al. Adv Mater 2007;19:1001–5. [32] Kim P, Doss NM, Tillotson JP, Hotchkiss PJ, Pan MJ, Marder SR, et al. ACS Nano 2009;3:2581–92. [33] Jung HM, Kang JH, Yang SY, Won JC, Kim YS. Chem Mater 2010;22:450–6. [34] Dang ZM, Lin YQ, Xu HP, Shi CY, Li ST, Bai J. Adv Funct Mater 2008;18:1509–17. [35] Dang ZM, Fan LZ, Shen Y, Nan CW. Chem Phys Lett 2003;369:95–100. [36] Dang ZM, Lin YH, Nan CW. Adv Mater 2003;15:1625–9. [37] Pecharroman C, Esteban-Betegón F, Bartolome JF, López-Esteban S, Moya JS. Adv Mater 2001;13:1541–4. [38] Dang ZM, Fan LZ, Shen Y, Nan CW. Mater Sci Eng B 2003;103:140–4. [39] Palma RD, Peeters S, Bael MJV, den Rul HV, Bonroy K, Laureyn W, et al. Chem Mater 2007;19:1821–31. [40] Chang SJ, Liao WS, Ciou CJ, Lee JT, Li CC. J Colloid Interface Sci 2009;329:300–5. [41] Min KT, Kim GH. J Appl Polym Sci 2011;120:95–100. [42] Yu J, Wang N, Ma X. Biomacromolecules 2008;9:1050–7. [43] Logakis E, Pollatos E, Pandis Ch, Peoglos V, Zuburtikudis I, Delides CG, et al. Compos Sci Technol 2010;70:328–35. [44] Shen Y, Lin YH, Li M, Nan CW. Adv Mater 2007;19:1418–22. [45] Dang ZM, Yao SH, Yuan JK, Bai J. J Phys Chem C 2010;114:13204–9. [46] Yang TI, Kofinas P. Polymer 2007;48:791–8.