Mechanical and Dilatometric Properties

of Carboxylated SWCNT/Epoxy
Composites: Effects of the Dispersion
in the Resin and in the Hardener

C. E. PIZZUTTO, J. SUAVE, J. BERTHOLDI, S. H. PEZZIN AND L. A. F. COELHO*

Santa Catarina State University, Center of Technological Sciences
89223-100 Joinville/SC, Brazil

S. C. AMICO
Universidade Federal do Rio Grande do Sul, Materials Engineering
Dept, P.O. Box 15010, 91501-970 Porto Alegre/RS, Brazil

ABSTRACT: In this study the effects of the dispersion of carboxylated single-walled carbon
nanotubes (SWCNT-c) in epoxy matrices using two different routes were investigated. In the first
route, SWCNT-c were dispersed in the resin using solvent and tip sonication. In the second route,
SWCNT-c were dispersed in the hardener in two different ways: with and without solvent, but both
with tip sonication. For comparison purposes, neat epoxy was also prepared using the same curing
conditions. The samples were characterized via mechanical and dilatometric testing, raman and
FTIR analyses, SEM images, and dilatometric tests. An increase of 88% was found for Young’s
modulus in the route with dispersion in the resin aided by solvent. Dispersing the SWCNT-c into
the hardener showed a significant increase in some mechanical properties, indicating that this is
a possible route for preparing nanocomposites. In addition it was observed that all nanocomposites
presented smaller volumetric expansion than neat epoxy.

KEY WORDS: epoxy, nanotubes, hardener, dispersion.

INTRODUCTION

FTER IIJIMA [1] discovered carbon nanotubes (CNT) in 1991 their great potential has
A been attracting attention in both science and engineering fields. CNT have a unique
combination of mechanical and transport properties that makes them excellent candidates
to substitute or complement the conventional nanofillers. Some nanotubes are stronger
than steel, lighter than aluminum, and more conductive than copper, although there is
currently a gap between this potential and the number of commercial applications [2].
The potential of CNT has been investigated for a wide range of applications from
pollutent removal [3] to nanofillers for reinforcement in polymers [4–8]. The first
nanocomposite using SWCNTs was reported in 1994 by Ajayan et al. [4]. In recent years,

*Author to whom correspondence should be addressed. E-mail: [email protected]

Journal of REINFORCED PLASTICS AND COMPOSITES, Vol. 29, No. 4/2010 524
0731-6844/10/04 0524–7 $10.00/0 DOI: 10.1177/0731684408099924
ß SAGE Publications 2010
Los Angeles, London, New Delhi and Singapore
Properties of Carboxylated SWCNT/Epoxy Composites 525

many different studies have used CNTs as fillers for reinforcement [4–8]. It should be
noted that epoxy has been one of the most studied thermoset materials [9–12]. As shown in
the literature, considerable effects of the addition of CNTs on thermoplastics were
detected, whereas in thermoset matrices there were challenges to overcome. In epoxy
composites especially, major efforts are needed to improve the dispersion of CNT and also
the adhesion between nanotubes and matrix [13].
In SWCNT the attractive force is of the order of 0.5 eV (0.8  1019J) per nanometer
of nanotube-to-nanotube contact [14], and kT (at T ¼ 300 K) is around 0.04  1019J,
which means that the attractive force between two single-walled carbon nanotubes
of 10 nm of length each is of the order of 200 kT, of a typical covalent bond. Also, it should
be taken into account that there is a considerable mechanical anchorage between these
nanofillers. In order to avoid the two major problems mentioned above, functionalization
has been applied to CNT in many different ways. Functionalization varies from covalent
to non-covalent chemical bonds using different chemical groups or compounds.
Although covalent functionalization damages the physical structure of CNT, i.e.,
decreasing transport properties, it helps avoiding hopes and bundles as well as improving
adhesion [2].
A general route for preparing nanocomposites with epoxy matrices involves the use
of tip sonication and/or a solvent. As pointed out in the literature, most of the studies with
epoxy initiated the dispersion of nanofillers in the resin [4–15]. However, only a few studies
initiated the dispersion in the hardener [16–18]. In this context we decided to investigate
the effect of dispersed SWCNT-c into the resin or into the hardener for preparing
nanocomposites with epoxy. In this study, SWCNT-c (0.25 wt%) were dispersed in epoxy
following two different routes for investigating the effects in mechanical and dilatometric
properties.

EXPERIMENTAL

Materials

SWCNT-c used in this study were supplied by the Nanomaterials Laboratory at the
Federal University of Minas Gerais (UFMG), Brazil, and were produced by electrical
arc discharge with 92% purity (supplier information). The bifunctional diglycidyl ether
of bisphenol A (DGEBA) epoxy resin was obtained from Huntsman as Araldite GY 251Õ ,
together with the hardener Aradur HY 956Õ , based on a polyamine. The solvent used was
analytical grade acetone, from Cinética Quı́mica, Brazil.

Preparation of SWNT-C/Epoxy Nanocomposites

For the preparation of nanocomposites, a certain amount of SWCNT-c (0.25 wt%,

relative to the epoxy) was dispersed in acetone (10.0% of the resin weight) under
simultaneous magnetic stirring and sonication (Sonics Vibration 750) at 165 W, for 20 min
[19,20]. The nanotube suspension was then added to the epoxy or to the hardener,
under the same conditions of simultaneous magnetic stirring and sonication. The mixture
was later degassed under vacuum and magnetic stirring at 708C for 1 h and cooled down
to room temperature. After that, the hardener or the resin was added to the
SWNT-c/epoxy or SWCNT-c/hardener system under mechanical or manual stirring, at
an epoxy:hardener weight ratio of 5 : 1. Finally, the mixture was cast into silicone molds
526 C.E. PIZZUTTO ET AL.

and left to cure at 608C for 24 h. For comparison, samples of neat epoxy were prepared
using the same curing conditions. Also, nanocomposites were prepared using the same
conditions described above, but with dispersion in the hardener without acetone. From
here, the following samples used in this work will be designated: S1, with acetone,
mechanical dispersion for stirring resin/hardener/SWCNT-c, dispersion of SWCNT-c
initiated in the hardener; S2, without acetone, mechanical dispersion for stirring
resin/hardener/SWCNT-c, dispersion of SWCNT-c initiated in the hardener; S3, without
acetone, manual dispersion for stirring resin/hardener/SWCNT-c, dispersion of SWCNT-c
initiated in the hardener; and S4, with acetone, mechanical dispersion for stirring
resin/hardener/SWCNT-c, dispersion of SWCNT-c initiated in the resin. Mechanical
stirring was carried out at 500 rpm for 60 s and the manual dispersion performed for 60 s.
It should be stressed that S4 belongs to route 1 and S1, S2, and S3 to route 2.

Characterization

Mechanical tests were performed according to the ASTM D638M-93 using an EMIC
DL 30000 universal testing machine with a 1000 kgf load cell operating at a cross head
speed of 5 mm min1. The strain was measured using an extensometer.
The surface morphology of the mechanical fractured specimens was observed under
a Zeiss DSM 940A scanning electron microscope (SEM) operating at 10 kV. The samples
were previously coated with gold in a sputtering device.
The specimens were also analyzed on a Netzsch 402C dilatometer, being heated
from room temperature to 1508C at 108C min1, under N2 atmosphere, followed by a fast
cooling down to room temperature and a new heating (second run) under the same
conditions.
Raman spectroscopy was employed to characterize specific bands of the SWCNT-c,
between 800 and 3400 cm1, on a Labram Jobim Yvon Micro Raman Spectrometer.
In addition, Fourier transform infrared (FTIR) transmission spectra were collected with
a Nexus microscope (magnification 10, probed region 150  150 mm2) attached to a
Nicolet spectrometer, in the mid-infrared region (650–3500 cm1). The measurements were
performed under nitrogen purge, with 4 cm1 of spectral resolution, by averaging 64 scans.

RESULTS AND DISCUSSION

Table 1 presents the obtained mechanical properties of the samples. It was possible
to achieve an 88% increase in Young’s modulus in the composite prepared with the aid
of acetone and dispersion of nanotubes in the resin. To the best of our knowledge this is

Table 1. Mechanical properties of the nanocomposites.

Sample r (MPa) e (%) E (GPa)

*
Neat 39.7  4.8 1.16  0.23 3.5  0.4
S1* 45.0  7.6 1.22  0.55 4.2  1.2
S2* 33.0  11.0 1.10  0.47 3.5  0.6
S3* 28.3  4.7 0.73  0.14 4.1  0.3
S4* 32.7  9.2 0.70  0.20 6.0  0.5

 ¼ Maximum tensile strength; " ¼ Elongation at break; E ¼ Young’s modulus.

*
Specified in the text.
Properties of Carboxylated SWCNT/Epoxy Composites 527

the highest increase in Young’s modulus for an epoxy matrix using SWCNT-c as nanofiller
in randomly oriented nanocomposites.
Chen and Tao also achieved an increase of up to 88% in Young’s modulus of
composites reinforced with 0.5 wt% of SWCNTs, with alignment of the nanofillers [21].
Also, Andrews et al. achieved increases of up to 86% in Young’s modulus of fibers
reinforced with SWCNTs [22].
On the other hand, the best performance obtained when the dispersion was performed
in the hardener with the aid of acetone, was in sample 1 (S1). In this case, the increasement
in Young’s modulus is 18%. Relating to the elongation at break, no effects of statistical
significance were noted, but in S3 the maximum strength was decreased by 30% in
comparison with neat epoxy. It should be noted that in this case the sample was obtained
by manual stirring.
Figure 1 summarizes the stress–strain curves of neat epoxy and all nanocomposites
prepared in this study. Comparing S1, S4, and neat epoxy it is possible to analyze the effect
of dispersion in the hardener or in the resin. The only advantage that can be noted is that
the dispersion into the hardener promotes an increase of Young’s modulus up to 88%,
with decreasing strain. Taking into account the standard deviations, the tensile strength
did not increase compared to neat epoxy. Comparing S2, S3, and neat epoxy it is possible
to analyze the effect of manual and mechanical dispersion (without acetone). Young’s
modulus did not change significantly between S3 and S1. In the case of mechanical
stirring, the tensile strength barely changed. The use of manual stirring decreases the
tensile strength drastically, probably due to the presence of agglomerates. The effect of the
use of acetone is analyzed by comparing S1 and S2.
Figure 2 shows the linear dilatometric curves for the second heating cycle of some
SWCNT-c/epoxy composites at a heating rate of 108C min1. All samples presented
volume contraction in the first heating at temperatures around Tg due to the presence
of voids, probably created by high residual stress and solvent trapping during the curing
process. In the second heating some of the samples presented a small plateau in the
surroundings of glass transition, evidencing some reorganization of polymer chains due to
stress relaxation during the second heating process. Nevertheless, all nanocomposites
presented lower thermal dilatations than neat epoxy. Gonnet [23] relates this last fact
to the negative thermal expansion of carbon nanotubes, due to their sp2 network. Indeed,
theoretical data suggest that even composites with low contents of SWNTs should have
a lower thermal dilatation than the pure resin [24].
The fracture surfaces from specimens of neat epoxy (a) and S4 (b) submitted to tensile
tests (not presented here) exhibited a relatively smooth surface indicating a typical
fractography feature of brittle fracture behavior. No significant changes were observed in
composites with SWCNT-c.
In the Raman spectra of SWCNT-c two peaks were monitered, one around 1570 cm1
and the other around 1350 cm1, with the first referring to the G band and the
second to the disorder-induced D band, respectively. The low G/D ratio observed
in this study (around one) indicates a high content of disordered structures and defects
[18], which is expected since covalent functionalization with carboxylic groups
change the physical and chemical structure of SWCNT [19]. In the FTIR spectra
obtained, the SWCNTs before and after the oxidation process were noted to be
very similar to others presented in the literature [25,26]. The presence of
characteristics bonds of C¼O and O–H bonds are due to the formation of COOH
groups on the nanotubes after acid treatment. There is a very broad shoulder peak at
528 C.E. PIZZUTTO ET AL.

80
75
Neat
70
S1
65
S3
60
S4
55

Stress (σ) (MPa)

50
45
40
35
30
25
20
15
10
5
0
0.000 0.002 0.004 0.006 0.008 0.010 0.012 0.014 0.016 0.018
Strain (ε)

Figure 1. Comparison of the stress–strain behavior of the neat epoxy and nanocomposites containing
0.25 wt% SWCNT.

0.016
Neat
0.014
S1
0.012
S2
0.01
S3
ΔL / L0

0.008 S4
0.006

0.004

0.002

0
25 45 65 85 105 125 145 165
T (°C)

Figure 2. Thermal dilatation of the composites and neat epoxy in the second heating using a heating rate
of 108C/min.

3100–3600 cm1 that is assigned to the O–H stretches of terminal carboxyl groups,
the peak at 2918 cm1 can be assigned to the C–H stretch, and the peaks at 1730 cm1
correspond to the carboxylic C¼O stretching vibrations. The peak at 1585 cm1
is attributed to the SWCNTs –C¼C– stretching mode [26].

CONCLUSIONS

Mechanical tests showed an increase in Young’s modulus (up to 88%) with addition
of 0.25 wt% SWCNT-c. Tensile strength and elongation at break, however, revealed that
the dispersion and/or SWCNT-c/matrix adhesion must be enhanced to achieve the expected
outstanding mechanical results in these two properties. A slight increase was detected in the
tensile strength and in elongation at break for the samples prepared with acetone,
dispersion of SWCNT-c initiated in the hardener and mechanical stirring (S1) compared to
Properties of Carboxylated SWCNT/Epoxy Composites 529

neat epoxy. It should be noted that the results presented by S1 may be indicative of a
promising route to disperse nanofillers in epoxy nanocomposites. Some samples of the S1
route had the best dispersion and adhesion between the SWCNT-c and matrix since the
elongation at break and tensile strength presented a slight average increase.
All nanocomposites and the neat epoxy presented a contraction around the Tg in
the first heating of dilatometric measurements, revealing a residual stress that disappears
in the second heating. Also, it can be seen that nanocomposites had a lower thermal
expansion than the neat epoxy. According to the literature this is most probably due to the
negative thermal expansion of carbon nanotubes [16].
The fact that the volume occupied by the nanotubes is smaller (1 : 5 ratio wt%) when the
dispersion was initiated in the hardener may contribute strongly to the thermal expansion
and some of the mechanical property results. This geometric restriction may affect the
dispersion of the nanofillers, since they have to diffuse more in the nanocomposites when
dispersion was initiated in the hardener and in a more viscous media which is the resin with
the viscosity increasing very fast. Some results obtained with the dispersion initiated in the
hardener may suggest an alternative route to prepare nanocomposites.

ACKNOWLEDGMENTS

The authors would like to thank Agência Espacial Brasileira (AEB) CAPES-PROCAD
(Project 0303054) for the financial support and for the scholarship to Ms. Jaqueline Suave
and to UDESC for the scholarship to Jonas Bertholdi. We thank Prof. R. L. Moreira
(UFMG) for his assistance in the infrared measurements, Prof. Rodrigo Gibrel Lacerda
(UFMG) for helpful discussions, suggestions, and assistance, and Marcio R. Loos for
many helpful suggestions.

REFERENCES

1. Iijima, S. (1991). Helical Microtubules of Graphitic Carbon, Nature, 354(6348): 56–58.

2. Moniruzzaman, M. and Winey, K. I. (2006). Polymer Nanocomposites Containing Carbon Nanotubes,
Macromolecules, 39: 5194–5205.
3. Wang, X. L., Lu, J. L. and Xing, B. S. (2008). Sorption of Organic Contaminants by Carbon Nanotubes:
Influence of Adsorbed Organic Matter, Environmental Science and Technology, 42(9): 3207–3212.
4. Ajayan, P. M., Stephan, O., Colliex, C. and Trauth, D. (1994). Aligned Carbon Nanotube Arrays Formed by
Cutting a Polymer Resin-nanotube Composite, Science, 265: 1212–1214.
5. Du, F., Fischer, J. E. and Winey, K. I. (2004). A Coagulation Method to Prepare Single-walled Carbon
Nanotube/PMMA Composites and their Modulus, Electrical Conductivity and Thermal Stability. Journal
of Polymer Science B, 41: 3333–3339.
6. Dai, J., Wang, Q., Li, W., Wei, Z. and Xu, G. (2007). Properties of Well-aligned SWNT Modified Poly
(methyl methacrylate) Nanocomposites, Mater. Lett., 61: 27–29.
7. Valentini, L., Puglia, D., Frulloni, E., Armentano, I., Kenny, J. M. and Santucci, S. (2004). Dielectric
Behavior of Epoxy Matrix/Single-walled Carbon Nanotube Composites, Composites Science and
Technology., 64: 23–33.
8. Ounaies, Z., Park, C., Wise, K. E., Siochi, E. J. and Harrison, J. S. (2003). Electrical Properties of Single Wall
Carbon Nanotube Reinforced Polyimide Composites, Composites Science and Technology., 63: 1637–1646.
9. Hernandez-Perez, A., Aviles, F., May-Pat, A., Valadez-Gonzalez, A., Herrera-Franco, P. J. and
Bartolo-Perez, P. (2008). Effective Properties of Multiwalled Nanotube/Epoxy Composites using Two
Different Tubes, Composites Science and Technology, 68(6): 1422–1431.
10. Kim, M. G., Hong, J. S., Kang, S. G. and Kim, C. (2008). Enhancement of the Crack Growth Resistance
of a Carbon/Epoxy Composite by Adding Multi-walled Carbon Nanotubes at a Criogenic Temeprature,
Composites Part A–Applied Science and Manufacturing, 39(4): 647–654.
530 C.E. PIZZUTTO ET AL.

11. Sun, L., Warren, G. L., O’Reilly, J. Y., Everett, W. N., Lee, S. M., Davis, D., Lagoudas, D. and Sue, H.-J.
(2008). Mechanical Properties of Surface-carboxylated SWCNT/Epoxy Composites, Carbon, 46(2): 320–328.
12. Zhou, Y. X., Pervin, F., Lewis, L. and Jeelani, S. (2008). Fabrication and Characterization of Carbon/Epoxy
Composites Mixed with Multi-walled Carbon Nanotubes, Materials Science and Engineering A – Structural
Materials Properties Microstructure and Processing, 475(1–2): 157–165.
13. Du, J.-H., Bai, J. and Cheng, H.-M. (2007). The Present Status and Key Problems of Carbon Nanotube
Based Polymer Composites, eXPRESS Polymer Letters, 1(5): 253–273
14. Dyke, C. A. and Tour, J. M. (2004). Covalent Functionalization of Single-walled Nanotubes for Material
Applications, Journal of Physcial Chemistry A, 108: 11151–11159.
15. Gojny, F. H., Wichmann, M. H. G., Fiedler, B. and Schulte, K. (2005). Influence of Different Carbon
Nanotubes on the Mechanical Properties of Epoxy Matrix Composites – A Comparative Study, Composites
Science and Technology, 65: 2300–2313.
16. Chena, H., Jacobs, O., Wua, W., Rudigerb, G. and Schadelb, B. (2007). Effect of Dispersion Method
on Tribological Properties of Carbon Nanotube Reinforced Epoxy Resin Composites, Polymer Testing
26: 351–360.
17. Gojny, F. H. and Schulte, K. (2004). Functionalisation Effect on the Thermo-mechanical Behaviour
of Multi-wall Carbon Nanotube/Epoxy Composites, Composites Science and Technology,c 64: 2303–2308.
18. Gauthier, C., Chazeau, L., Prasse, T. and Cavaille, J. Y. (2005). Reinforcement Effects of Vapour Grown
Carbon Nanofibres as Fillers in Rubbery Matrices, Composites Science and Technology, 65: 335–343.
19. Loos, M. R., Coelho, L. A. F., Pezzin, S. H. and Amico, S. C. (2008). The Effect of Acetone Addition on the
Properties of Epoxy, Polı´meros: Cieˆncia e Tecnologia, 18(1): 76–80.
20. Suave, J. (2007). Nanocompósitos de Epóxi com Nanotubos de Carbono de Parede Simples Carboxilados:
Influência da adição de solventes e das condições de sonificação no processamento e nas propriedades.
Master of Science Dissertation, State University of Santa Catarina, Joinville.
21. Chen, W. and Tao, X. (2006). Production and Characterization of Polymer Nanocomposite with Aligned
Single Wall Carbon Nanotubes. Applied Surface Science, 252: 3547–3552.
22. Andrews, R., Jacques, D., Rao, A. M., Rantell, T., Derbyshire, F., Chen, Y., Chen, J. and Haddon, R. C.
(1999). Nanotube Composite Carbon Fibers. Applied Physics Letters, 75(9): 1329–1331.
23. Gonnet, P. (2004). Thermal Conductivity and Coefficients of Thermal Expansion of SWNTS/Epoxy
Nanocomposites, Master of Science Dissertation, Florida State University, Tallahassee.
24. Villoria, R. G., Miravete, A., Cuartero, J., Chiminelli, A. and Tolosana, N. (2006). Mechanical Properties
of SWNT/Epoxy Composites using Two Different Curing Cycles. Composites Part B – Engineering,
37(4–5): 273–277.
25. Chen, J., Hamon, M. A., Hu, H., Chen, Y., Rao, A. M., Eklund, P. C. and Haddon, R. C. (1998). Solution
Properties of Single-walled Carbon Nanotubes, Science, 282: 95.
26. Zhang, J., Zou, H., Qing, Q., Yang, Y., Qingwen, L., Zhongfan, L., Xinyong, G. and Du, Z. (2003).
Effect of Chemical Oxidation on the Structure of Single Walled Carbon Nanotubes, Journal of Physical
Chemistry B, 107: 3712–3718.