Multiwalled carbon nanotubes (MWCNTs), first reported by Iijima in
1991 (1), contain exclusive structural, mechanical, and electrical
properties (2). Because of the exceptional properties of MWCNTs, many
investigations have focused on exploiting these surprising
characteristics for engineering applications. including polymer
nanocomposites, field emitters, nanoelectronic devices, chemical
sensors, and biomaterials [3-5]. For polymer/ MWCNT nanocomposites, the
addition of MWCNT could extensively change the mechanical, thermal, and
electronic properties of polymer matrix, which would remarkably extend
their application regions. However, some reports indicate that the
improvements of the mechanicalproperties of polymer/CNT composites are
limited due to the phase separation between the polymer matrix and CNTs
[6, 7]. To obtain uniform distribution of CNTs in the polymer matrix is
still a very important subject until now.
Functionalization of MWCNTs is the strategy used to enhance
dispersion in organic media. There are different methods used for the
functionalization of MWCNTs through covalent and noncovalent reactions
with organic-molecules including polymers. Recently, some investigations
have used the functionalized MWCNTs as reinforced fillers to prepare the
polymer/MWCNT nanocom-posites by using various methods such as solution
processing, melt processing, and in situ polymerization [8-20]. They
have demonstrated that the functionalized MWCNTs were well separated and
randomly distributed in the polymer matrix. For example, Coleman et al.
has effectively used CNTs as fillers to significantly enhance the
mechanical and optical properties of polymers [8-11]. In our previous
studies, conducting polymer/MWCNT composites were successfully prepared
using the in situchemical oxidative polymerization (17), (18). The
electrical conductivities at room temperature of conducting
poly-mer/MWCNT composites are extensively enhanced compared to that of
conducting polymer without MWCNT. But it is relatively difficult to
homogeneously disperse MWCNT in a polymer matrix by melt processing.
Potschke et al. has used commerically available polycarbonate
(PC)/nanotube master batches as a starting material and diluted by
adding an appropriate amount of pure polymer in the subsequent melt
procesing (21-25). Their results suggested that the CNT is randomly
distributed in the polymer matrix. The physical properties of fabricated
PC/CNT nanocomposites significantly depend on the amount of CNTs and
their distributions. Nevertheless, the preparation of PC/CNT master
batch and their following melt processing of PC/CNT nanocomposites are
rarely mentioned among these investigations.
In this study, the PC/MWCNT master batch has been fabricated by
mixing the PC" and carbolic acids containing MWCNT in
tetrahydrofuran (THF) solution. The PC/ MWCNT nanocomposites were then
prepared by adding various amount of PC into PC/MWCNT master batch
through the subsequent melt processing. The structure, morphology, and
physical properties of PC/MWCNT nanocomposites were characterized by
FESEM, HRTEM. thermogravimetric analyzer (TGA), and dynamic mechanical
Preparation of Polycarbonatel Multiwalled Carbon Nanotube
Composites Using Melt Processing
The MWCNTs fabricated by ethylene chemical vapor deposition using
AUO, supported F[e.sub.2][O.sub.3] catalysts and polycarbonate (PC) with
melt index (MI) of 12 g/10 minwere kindly supplied by Material and
Chemical Research Laboratories (Hsinchu. Taiwan). The diameter of MWCNT
is about 40 nm, and the purity of MWCNTs is higher than 90% after
purification. The reagents. including nitric acid, THF. and
cetyltrimethylammonium bromide (CTAB), were used without further
purification. The as-received MWCNTs were treated using a solution of
nitric acid under refluxing conditions at 160 C oil bath with various
times, producing various ratio of carboxylic acid groups (designated as
c-MWCNTs) at the defect sites. To ensure that the nitric acid was
completely removed from the c-MWCNTs. a large amount of distilled water
was used to neutralize the c-MWCNTs for 48 h. To avoid the strong
interaction of the functional groups on the surface of c-MWCNTs during
filtration, the 10 wt% c-MWCNT was first mixed with the CTAB cationic
surfactant in THF solution and ultrasonicated over 3 h to form CTAB-
coated c-MWCNT in solution. Then the filtered CTAB- coated c-MWCNTs were
dried overnight inside a vacuum oven at 60 C to remove any remaining
water vapor and nitric acid.
The PC/MWCNT master batch was produced using a various amount of
the CTAB-coated c-MWCNTs and PC individually dissolved in THF by
ultrasonication process for 3 h and 1 h, respectively. Then the THF-PC
solution and THF-CTAC-coated c-MWCNTs solution were mixed together and
ultrasonicated for 60 min. Finally, a lot of methanol solution was
slowly poured into the solution to precipitate the mixture of PC/MWCNT.
The precipitated PC/MWCNT master batch was filtered and rinsed several
times with distilled water and methanol. The powder of PC/MWCNT master
batch with 10 wt% MWCNT content thus obtained was vacuum dried at
60[degrees]C for 12 h. the 2. 5, and 7 wt% PC?MWCNT nanocomposites were
then prepared using a melt-direct compounding process with PC/MWCNT
master batch and PC pellet in a Haake mixer operated at 260 C and 60 rpm
for 10 min.
Structural and Morphological Analysis
The Raman spectroscopy was used to characterize the structure of
MWCNT. c-MWCNT, and CTAB-coated c-MWCNT. Raman spectra were performed on
a Jobin Yvon TRIAX 550 system using a He-Nc laser operated at 632.8 nm
with a CCD detector. The presented spectrum is an average of three
spectra measured at different regions over the entire sample range. The
morphology of the fabricated nanocomposite was characterized using
Field-emission scanning electron microscopy (FESEM) and high-resolution
transmission electron microscopy (HRTEM). FESEM was performed at 3KV
using a JEOL JSM-6700f field-emission instrument, and high-resolution
transmission electron microscope (HRTEM) was recorded on a Hitachi
HF-2000 at 200 kV.
Thermal stabilities of these samples were obtained using
PerkinElmer thermogravimetric analysis (TGA). The test was performed
from room temperature until 800.sup.degrees]C at a scanning rate of 10
C/min. Dynamic mechanical analysis (DMA) experiments were obtained on a
PerkinElmer instrument DMA 7e apparatus in three-point bending mode. A
rectangular sample of PC matrix and nanoeomposites with dimensions. 18
mm X 10 mm X 2 mm was prepared. The test was carried out at the
temperature range of 40-200[degrees] C using [2.degrees]C/min heating
rate and 1 Hz constant frequency. Calibrations for force, mass,
position, and temperature were made in accordance with PerkinElmer
procedures. For the determination of conductivity, the samples of pure
PC matrix, PC/MWCNT master batch, and PC/MWCNT nanoeomposites were
pressed into pellet under 20 MPa. The conductivity at room temperature
using the standard Van Der Pauw de four-probemethod was measured by a
programmable DC voltage/ current detector. Each data shown here is the
mean value of the measurement from at least three samples, and the
experimental error using standard deviation of these data is obtained.
RESULTS AND DISCUSSION
Figure 1 shows the Raman spectra of the MWCNTs c-MWCNTs. and
CTAB-coated c-MWCNTs. Both spectra of MWCNTs and c-MWCNTs have the same
pattern, revealing that the surface modification using nitric acid
solution under refluxing conditions at 160 C oil bath does not affect
the graphite structure of the MWCNTs. Such chemical functionalization of
the MWCNT can be used to produce carboxylic acid at local defects in the
curved gra-phene sheets and tube ends. The Raman characteristic peaks at
1355 and 1580 [cm.sup.-1] are D and G hands of MWCNT corresponding to
the s[p.sup.3]- and s[p.sup.2]-hybridized carbons of disordered graphite
and the ordered state on the surface of MWCNT, respectively. The
intensity ratios of G and D bands ([I.sub.G]/[I.sub.D]) are 0.88 for
MWCNTs and 0.65 for c-MWCNTs. These results indicate that the chemical
functionalization of nitric acid increases the degree of disorder. The
Raman spectrum of CTAB-coated c-MWCNT has different characteristic peak
compared to that of c-MWCNT, revealing that the cationic surfactant CTAB
is successfully coated on the surface of c-MWCNTs.
[FIGURE 1 OMITTED]
The thermogravimetric analysis was conducted to identify the
thermal stability of MWCNT, c-MWCNTs, and CTAB-coated c-MWCNTs. Figure 2
shows the TGA thermal curves of the MWCNT, c-MWCNTs, and CTAB-coated
c-MWCNTs. As seen from the figure, the weight loss of MWCNT is mainly
attributed to the loss of amorphous carbon. More weight loss of c-MWCNT
compared to the pure MWCNT is due to the organic decomposition for the
carboxylic acid groups after surface modification as suggested by
earlier reports (26). Then the steepest weight loss of CTAB-coated
c-MWCNT is observed in a temperature interval of 200-250 [degrees] C,
which may be attributed to the thermal degradation of the CTAB cationic
surfactant formed on the surface of the c-MWCNTs (27). Both Raman and
TGA data indicate that the cationic surfactant is successfully coated on
the surface of c-MWCNTs.
[FIGURE 2 OMITTED]
The PC/MWCNT master batch was fabricated using a various amount of
the CTAB coated c-MWCNTs and PC individually dissolved in THF solution.
Figure 3 shows the HRTEM images of the PC/MWCNT master batch. From this
image, it can be clearly seen that the tubelike morphology representing
the c-MWCNT in PC matrix is well separated and randomly distributed.
This result suggests that the MWCNT in PC/MWCNT master batch was
effectively separated via solution mixing method under ultrasonic
[FIGURE 3 OMITTED]
Figure 4 shows the FESEM data of 2, 5, and 7 wt% PC/MWCNT
nanocomposites fabricated by mixing the PC/MWCNT master batch and PC
pellet through the melt-direct-compounding process. It is clear that the
tubelike structure of MWCNT is randomly distributed in the PC matrix.
Closer inspection of HRTEM images of PC/MWCNT nanocomposites shown in
Fig. 5 reveals that the distribution of MWCNT for the resulting
nanocomposites is well uniform. Therefore, the fabrication of PC/MWCNT
nanocomposites is successfully prepared through melt processing.
[FIGURE 4 OMITTED]
To study the effect of MWCNT on the thermal behaviors of PC matrix,
TGA analysis was used to identify the thermal degradation behavior of PC
and PC/MWCNT nanocomposites. Figure 6 shows the TGA curves of PC and
PC/MWCNT nanocomposites at a heating rate of 10 [degrees] C/min. The TGA
profiles of PC/MWCNT nanocomposites show similar tendency, and the onset
temperature of degradation ([T.sub.onset]) shown in Fig. 6b can be
determined from these curves by extrapolating from the curve of maximum
degradation back to the initial weight of the polymer. Detailed
[T.sub.onset] of PC matrix and PC/MWCNT nanocomposites are summarized in
Table 1. The [T.sub.onset] of PC is 488 [degrees] C and extensively
increases to 495, 497, and 498 [degrees] C as the loading of 2, 5, and 7
wt% MWCNT content, respectively. From these experimental results, it can
be seen that the presence of MWCNT in PC induced better thermal
stability and therefore the degradation starting temperatures clearly
shifted to higher temperatures.
[FIGURE 6 OMITTED]
Figure 7 shows the dynamic storage modulus G' of neat PC
matrix and PC/MWCNT nanocomposites at a temperature range of 40-180
[degrees] C. The data of storage modulus at 40 and 160 [degrees] C are
listed in Table 1. It is clear that the storage modulus of PC starts to
decrease at about 150 [degrees]C, which is probably as a result of the
glass transition temperature ([T.sub.g]) of PC. The [T.sub.g] of
PC/MWCNT nanocomposites is slightly slower than that of pure PC matrix,
which may be due to the presence of cationic surfactant CTAB coated on
the surface of MWCNT to induce a decrease of glass transition
temperature of PC (shown in Fig. 8). At 40 [degrees] C, the storage
modulus of PC is 3.48 X [10.sub.8] Pa, which decreases with the
increasing temperature; at 160 [degrees] C it drops to 6.36 X [10.sub.6]
Pa. This is attributed to insufficient thermal energy to overcome the
potential barrier for transitional and rotational motions of segments of
the polymer molecules in the glassy region, whereas above the [T.sub.g],
the thermal energy becomes comparable to the potential energy barriers
for the segmental motions. For PC/MWCNT nanocomposite, significant
enhancement of G' can be seen in the lower temperature range,
indicating that the experimental introduction of MWCNT into PC matrix
have strong influence on the elastic properties of the PC matrix. Below
[T.sub.g], the enhancement of G' of PC/MWCNT nanocomposites when
compared with virgin PC matrix is 20.7%, 58.3%, and 102% for 2, 5, and 7
wt% PC/MWCNT nanocomposites, respectively. These results indicate that
the reinforcement effects of PC/MWCNT nanocomposite are predominated by
the presence of MWCNT and possible interaction zone between the
CTAB-coated c-MWCNT and PC induced an annular interphase region of
immobilized polymer surrounding the embedded MWCNTs (28). It is
necessary to point out that the enhancement of storage modulus above
[T.sub.g] for nanocomposites (rubbery plateau region) is remarkably
increased with introducing the MWCNT into PCmatrix. Above [T.sub.g], the
storage modulus of pure PC matrix decreases about two order in
magnitude, but the storage modulus of nanocomposites remains roughly
25-30% in magnitude compared to those data below [T.sub.g], Therefore,
the improvement of storage modulus for PC/MWCNT nanocomposites is more
than one order in magnitude higher than those of PC without MWCNT. These
results reveal that the presence of MWCNT in PC/MWCNT nanocomposites is
possibly confined and retarded the segmental chain motion of PC at the
interface. Uniform distribution of the MWCNT at the nanoscale level is
possible to induce a phenomenal increase in the storage modulus for the
PC/MWCNT nanocomposites. Figure 8 illustrates the tan [delta] for
PC/MWCNT nanocomposites. The peak positions are allocated at
152[degrees]C for pure PC matrix and peak profile of PC/MWCNT is clearly
shifted into lower temperature region with less intense and broader
compared to pure PC samples. A decrease of tan [delta] for fabricated
nanocomposites is probably due to the presence of cationic surfactant
CTAB coated on the surface of MWCNT. The broader profiles of
nanocomposites indicate that a segmental chain motion of the PC/MWCNT
nanocomposites is retarded due to the confinement of the PC by MWCNT
[FIGURE 7 OMITTED]
[FIGURE 8 OMITTED]
[FIGURE 5 OMITTED]
The electrical conductivities of PC matrix. PC/MWCNT master batch,
and PC/MWCNT nanocomposites were measured using the standard Van Der
Pauw dc four-probe method (30). The conductivity for PC matrix at room
temperature was [10.sup.-13] S/cm. As the addition of 2 wt% MWCNT into
PC, the conductivities at room temperature dramatically increase from
[10.sup.-13] S/cm to 2 X [10.sup.-8] S/cm. With the continuous increase
in the content of MWCNT, the conductivities at room temperature
gradually increase from 1.9 X [10.sup.-8] S/cm for 2 wt% PC/MWCNT
nanocomposites to 1.8 X [10.sup.-6] S/cm for 5 wt% PC/MWCNT
nanocomposites. Further increasing content of MWCNT, the conductivity at
room temperature approached the conductivity of PC/MWCNT master batch
(4.9 X [10.sup.-5] S/cm). Detail electrical conductivities of PC matrix
and PC/MWCNT nanocomposites are also summarized in Table 1.
Nevertheless, the conductivities of fabricated nanocomposites with very
low MWCNT con-tent at room temperature are more than four orders in
magnitude higher than those of PC without MWCNT. This result is perhaps
due to the presence of MWCNTs containing a large aspect ratio and
surface area, so serving as a conducting path between the insulated PC
domains and increasing the conductivity with increasing MWCNT content.
The PC/MWCNT nanocomposites have successfully prepared by mixing
the PC/MWCNT master batch and PC pellet through melt processing. The
HRTEM micrograph of nanocomposites shows that the MWCNT is well
separated and uniformly distributed in PC matrices. The storage modulus
of fabricated PC/MWCNT nanocomposites shows significant enhancement
compared to that of pure PC matrix. These results indicate that the
reinforcement effects of PC/MWCNT nanocomposite are predominated by the
presence of MWCNT and better interaction between the MWCNT and PC. The
conductivities of 2 and 5 wt% PC/MWCNT nanocomposites are more than four
and seven orders in magnitude higher than that of PC without MWCNT,
respectively. That is because conductive MWCNT serves as a conducting
path between insulated PC, increasing the conductivity with increasing
(1.) S. lijima and T. lchihashi. Nature. 363. 603 (1993).
(2.) S. Fan. M.G. Chapline, N.R. Franklin, T.W. Tombler, A.M.
Cassell, and H. Dai, Science, 283, 512 (1999).
(3.) H. Dai, J.H. Hafner, A.G. Rinzler, D.T. Colbert, and R.E,
Smalley, Nature 384, 147 (1996).
(4.) S. Frank. P. Poncharal. Z.L. Wang, and W.A. de Heer, Science,
280, 1744 (1998).
(5.) S.J. Tans. A.R.M. Verschueren. and C. Dekker, Nature, 393, 49
(6.) P.M. Ajayan, Chem. Rev., 99. 1787 (1999).
(7.) Y.J. Kim, T.S. Shin, H.D. Choi, J.H. Known, Y.C Chung, and
H.G. Yoon, Carbon, 43, 23 (2005).
(8.) J.N. coleman. U. Khan, W,.J. Balu, and Y.K. Gun'ko.
Car-bon:. 44, 1624 (2006).
(9.) S.M. O'Flaherty, R. Murphy, S.V. Hold, M. Cadek, J.N.
Coleman, and W.J. Balu, J. Phys. Chem. B. 107. 958 (2003).
(10.) J.N. Coleman. U. Khan, and Y.K. Gum'Ko, Adv. Mater., 18,
(11). J.N. Coleman, M. Cadekl, K.P. Ryan. A. Fonseca. J.B. Nagy,
Balu, and M.S. Ferreira, Polymer. 47. 8556 (2006).
(12.) H.T. Ham, Y.S. Choi, M.G. Chee, and I.J, Chang, J. Polym.
Sci, Polym. Chem., 44, 573 (2006).
(13.) Z. Zhang. J. Zhang. P. Chen. B. Zhang. J. He. and G.H. Hu,
Carbon, 44, 692 (2006).
(14.) S. Santra. R. Tapec, N. Theodoropoulou, J. Dobson, A. Hebard,
and W.H. Tan, Langmuir, 17, 2900 (2001).
(15.) Y. Yu, C. Ouyang, Y. Gao. Z. Si, W. Chen. Z. Wang, and G.
Xue.J. Polym. Sci. Polym. Chem., 43. 6105 (2005).
(16.) K.H. An. S.Y. Jeong. H.R. Hwang, and Y.H. Lee. Adv. Mater.,
16. 1005 (2004).
(17.) T.M. Wu and S.H. Lin, J. Polym. Sci. Polym. Phys., 44, 1413
(18.) T.M. Wu and S.H. Lin, J. Polym. Sci. Polym. Chem., 44, 6449
(19.) H.T. Ham, Y.S. Choi, N. Jeong, and I.J. Chung, Polymer. 46.
(20.) J. Deng, Y. Peng. C. He. X. Long. P. Li. and A.S.C. Chan.
Polym. Int., 52, 1182 (2003).
(21.) P. Potschke, H. Brunig, A. Janke, D. Fischer, and D.
Jehnichen, Polylmer, 46, 10355 (2005).
(22.) M. Abdel-Goad and P. Potschke, J. Non-Newtonian Fluid Mech.,
128, 2 (2005).
(23.) E. Hammel. X. Tang, M. Trampert, T. Schmitt, K. Mauth-ner. A.
Eder. and P. Potschke, Carbon. 42. 1153 (2004).
(24.) P. Potschke, A.R. Bhattacharyya, and A. Janke, Eur. Polym.
J., 40, 137 (2004).
(25.) P. Potschke, M, Abdel-Goad, I, Alig, S. Dudkin, and D.
Lelling, Polymer, 45. 8863 (2004).
(26.) D.B. Mawhinney, V. Naumenko, A. Kuznetsova, J.T, Yates Jr..
J. Liu. and R.E. Smalley, J. Am, Chem. Soc., 122, 2383 (2000).
(27.) A. Kuznetsova. D.B. Mawhinney. V.J. Naumenko. J.T, Yates Jr.,
J. Liu. and R.E. Smalley, Chem. Phys. Lett., 321, 292 (2000).
(28.) A. Eitan. F.T. Fisher, R. Andrews, L.C. Brinson, and L.S.
Schadler. Compos. Sci. Technol., 66, 1159 (2006).
(29.) Y.T. Sung, C.K. Kum. H.S. Lee, N.S. Byon. H.G. Yoon. and W.N.
Kim. Polymer, 46. 5656 (2005).
(30.) L.J. Van Der Pauw, Philips Res. Rep., 13. 1 (1958).
Correspondence to: T.-M. Wu; e-mail: tmwu@dragon. nchu.edu.tw
Contract grant sponsor: Material and Chemical Research
Industrial Technology Research Institute (ITRI).
Published online in Wiley InterScience
[C] 2008 Society of Plastics Engineers
Tzong-Ming Wu, (1) Erh-Chiang Chen, (1) Yen-Wen Lin, (1) Ming-Feng
Chiang, (1) Gwo-Yang Chang (2)
(1) Department of Materials Science and Engineering, National Chung
Hsing University, Taichung, Taiwan 402
(2) Material and Chemical Research Laboratories, Industrial
Technology Research Laboratories, Industrial Technology Research
Institute, Chu,Tung, Hsinchu, Taiwan 310
TABLE 1. Onset temperature of degradation (Tonset), dynamic storage
modulus (G'). and electrical conductivity of PC matrix and PC/MWCNT
Sample Tonsei (C) 40 C 160 C Conductivity (S/cm)
PC matrix 488 348 6.36 1.0 X [ 10.sup.-13]
2 wt% PC/MWCNT 495 420 108 1.9 X [10.sup.-8]
5 wt% PC7MWCNT 497 551 145 1.8 X [10.sup.6]
7 wt% PC/MWCNT 498 702 210 2.5 X [10.syp.-5]