Content uploaded by Yeqing Wang
Author content
All content in this area was uploaded by Yeqing Wang on Sep 23, 2023
Content may be subject to copyright.
COVER SHEET
Paper Number: 004544
Title: Tensile Properties of Unidirectional Polymer Composites Reinforced by
Aligned Carbon Nanotube Yarns
Authors: Ali Naderi
Amirreza Tarafdar
Wenhua Lin
Yeqing Wang
ABSTRACT1
Carbon nanotubes (CNTs), as they possess outstanding mechanical properties and
low density, are considered as one of the most promising reinforcements in composite
structures. Due to their capability of transferring loads, CNTs in long continuous
forms such as yarns and tapes can withstand 20 times as much load as steel can do at
the same weight. In this research, carbon nanotube yarns were wound onto an
aluminum plate using a custom-built fixture to fabricate a unidirectional strip. Then,
by brushing epoxy resin on the strip and laminating four layers, the unidirectional
CNT reinforced epoxy resin composite beam specimens were produced. The
mechanical properties of the unidirectional CNT-reinforced composite (CNTRC)
were determined using standard tensile tests. This study presents a method for
manufacturing CNTRC out of CNT yarns, determining the CNTRC’s Young’s
modulus as well as the tensile strength, and obtaining its strain field via digital image
correlation (DIC) method. It is observed that the pressure due to sandwiching of the
aluminum plates during the manufacturing process leads to nonuniformity of the
specimen in the width along midspan of the longitudinal direction which results in
the specimen’s not being perfectly unidirectional. This phenomenon can cause the
matrix cracking in tensile test and reduce the ultimate tensile strength up to 78% in
comparison with perfectly unidirectional specimens. However, the Young’s modulus
of such composites is comparable with those obtained from previously existing
research. Also, Results from DIC showed the possible failure prone areas in the
specimens, as it presents a up to 64% difference between the highest and lowest strain
in the tensile loading direction through the specimens. This study will serve as a
foundation for future research involving CNT composites, particularly the use of their
high anisotropy to produce auxetic composites with large negative Poisson’s ratios.
Keywords: CNT yarns; Unidirectional CNT composites; Filament winding;
Mechanical properties
A. Naderi, A. Tarafdar, W. Lin, Y. Wang, Department of Mechanical and Aerospace Engineering,
Syracuse University, Syracuse, NY 13244, USA.
Corresponding Author: Yeqing Wang, ywang261@syr.edu.
INTRODUCTION
Carbon nanotubes (CNTs), due to their exceptional mechanical, electrical, and
thermal properties, are attracting growing attention in the research community [1–4].
The mechanical properties associated with the CNT composites are influenced by the
type and amount of CNT utilized in the composite. One of the emerging types of
CNTs is aligned CNT yarns—made by spinning and twisting CNT into long
continues yarns [5].
CNTs were first discovered by Iijima in 1991 [6]. They have been considered as
the new generation of reinforcements in composite structures [7–9]. CNT composites
are usually constructed by integrating CNTs with various matrices, such as metals
[10], ceramics [11], and polymers [12]. There are various studies on the
characteristics of composite reinforced by chopped CNTs. For instance, in the paper
by Zare [13], a model was proposed to calculate the tensile strength of such
composites. In another research, Shokrieh et al. [14] investigated the effect of adding
multiwall CNT to polyester composites. They demonstrated that adding 0.05 wt% of
CNT can improve the ultimate tensile strength by 45% compared to specimen without
this reinforcement. Also, Hussein et al. [15] conducted a research on the effect of
adding multi-walled CNT as well as carbon fiber in an epoxy resin composite on the
mechanical and thermal properties of such composites. They reported that adding
multi-walled CNT to this composite can elevate the impact energy absorption of
epoxy composites by 14%. Chen et al. [16] utilized multi-walled CNT along with
glass fiber to reinforce the core layer of a sandwiched structure. They showed that
the interlaminar toughness of the specimen can be increased by 124% by adding the
reinforcements to the core layer in comparison with the unmodified sample.
Unlike chopped CNTs, CNT yarns can substantially enhance the load-bearing
capabilities of the composites, leading to enhancement in the toughness, tensile
strength, as well as fatigue resistance of composites [17–21]. Additionally, the
anisotropy of such CNTs make them an ideal choice for manufacturing auxetic
composites [22]. In this regard, Fan and Wang [23], using a shear deformable beam
theory, investigated the low-velocity impact on the CNT yarn composites. The beam
in this study is designed to have a negative Poison’s ratio, i.e., auxeticity. Also, Kim
et al. [24] measured the mechanical properties of aluminum rings covered via CNT
yarn composite. They applied dynamic as well as static loadings on the structure. In
this study, it has been shown that although adding CNT yarns over the aluminum ring
increases the weight by 11%, it can improve the load bearing by over 200%.
Additionally, the mechanical characteristics associated with CNT yarn composites
were investigated [25]. In this study, they examined the impact of different winding
tension on the behavior of the composite. They presented that the tensile strength of
CNT composites is 69% of that of CNT yarn itself. Also, Barber et al. [26] analyzed
the deformation as well as the failure strength of composites made with CNT yarns
and polymer. They utilized X-ray to characterize the specimens and showed that
sharp gradients and the accumulation of the resin are two major reasons for failure.
Up to now, various studies have been conducted on chopped CNT composites to
investigate their mechanical characteristics. However, integrating CNT yarns into
matrices to make composite beams and plates is yet to be explored. Thus, in this
research, using a custom-built winding device, the CNT yarns were wound over a
plate to make CNT strips. Next, by applying epoxy resin as the matrix, unidirectional
CNT yarn composites with four layers, i.e., [0]4, were manufactured. The CNT
composite strips were analyzed via microscope before and after tensile testing. By
testing the specimens via a tensile tester, the mechanical properties of unidirectional
CNT composites were obtained and presented. This article will be used as a basis for
our future studies associated with auxetic CNT composites, specifically, using the
high anisotropy of such composites to make auxetic composites with large negative
Poisson’s ratios [3, 23, 27, 28].
METHODOLOGY
Specimen Fabrication
The MIRALON CNT yarn made by Huntsman with a diameter of 150 µm (0.6
g/cm3 density, 240 MPa tensile strength, and 4.9 GPa tensile modulus) was used to
fabricate unidirectional CNT reinforced epoxy resin composites. The CNT yarns
were wound over an aluminum plate, for which the rotary motion of the winding
process was controlled by a DC stepper motor. The distance between adjacent CNT
fibers was controlled via a linear motor and a controller. By coordinating the rotary
motion of the aluminum plate and the translational motion of the feeder mounted on
the linear motor, aligned CNT strips with controllable distance between adjacent
CNT fibers can be fabricated. The layout of the motors, controllers, CNT feeder, and
aluminum plate is shown in Fig.1. The West System 105/206 epoxy resin was
manually brushed onto each layer of the aligned CNT strip prior to winding the next
layer. This process was repeated until four layers of laminated CNT/resin composites
were made. It is worth mentioning that the tension in the yarns, which was reported
as a critical factor that affects the resulting mechanical properties [25], was controlled
by the friction between the feeder and roller. After that, the laminate was sandwiched
between two additional aluminum plates and placed in a vacuum bag for 24 hours for
curing. Then, four coupon specimens with a dimension of 40 mm by 9.5 mm were
cut out of the original laminate (100 mm by 11 mm) with waterjet cutting.
Figure1. Photograph of the custom-built aligned CNT yarn winder, including DC motors,
controllers, feeder, and aluminum plate.
Specimen Characterization
Figure 2 shows the cured laminate on the aluminum plate. It can be seen, the
CNT-reinforced composite (CNTRC), due to the pressure from the vacuum bag and
different friction—the friction between the plate and CNTs at both ends are higher
than that in the middle of the plate, has different widths in different locations,
resulting in the nonuniformity of the specimens.
Figure 2. Photograph of the cured CNT composite laminate wound over the aluminum plate.
Also, in order to make sure the specimen was aligned unidirectionally and check
for possible defects, a digital microscope, Hirox KH-8700, was employed to examine
the CNT/epoxy resin composite specimens. The obtained images from the digital
microscope are presented in Fig. 3. As it is observed, the CNTs are aligned together
to form a unidirectional composite structure. Additionally, it is observed that there
are some spots where the resin has not completely infiltrated, causing the specimen
to be dry and porous. These dried regions can be considered as a bult-in crack in the
specimen, as they formed a void in the longitudinal direction of the specimen. Also,
the regions with proper impregnation of resin can be seen as well.
Electric
controllers
Friction controller
Aluminum
base plate
CNT
feeder
DC Motors
40mm
10mm
Figure 3. Digital microscope images of the CNT composite specimens: (a) sample #2 (b) sample #3.
Experimental Procedure
The tensile test specimens were prepared by bonding four aluminum tabs of 10
mm by 10 mm to the two ends of the coupon specimens. These specimens were used
for tensile tests using an MTS testing machine. To capture the strain in the specimens,
both digital image correlation (DIC) and strain gauge were utilized. Figure 4 shows
the prepared speckle pattern for DIC. Then, by taking the video during the tensile
process, the deformations in the specimens can be recorded and used for DIC analysis
via the MATLAB toolbox, i.e., DUODIC [29]. After the tensile tests, each specimen
was characterized to identify the failure modes.
(a)
(b)
Dried region
Resin filled region
CNT
Figure 4. (a) DIC pattern, aluminum tabs, and gauge length on the specimens, and (b) the specimen
before bonding tabs and applying DIC pattern.
RESULTS AND DISCUSSION
The obtained tensile properties, including the Young’s modulus, tensile strength,
and failure strain are shown in Table 1. Our results are compared with those by [25],
in which they reported that the Young’s modulus and tensile strength of CNTRC are
42.02 ± 4.48 GPa and 873.12 ± 32.64 MPa, respectively, for the CNT/EPONTM 828
composite laminate with two unidirectional layers. Although the Young’s modulus
is comparable, there exists a large difference between the obtained tensile strength
and those in [22]. Two possible reasons for such a difference are: (i) the
nonuniformity in the width along the longitudinal direction of the specimen resulted
from the vacuum compression of the caul plates which cause the specimens to be not
perfectly unidirectional and (ii) the voids due to regions that are not fully impregnated
by the epoxy resin during the manufacturing process, as shown in Fig. 2, working as
a bult-in crack. It is possible that the nonuniformity and the high porosity of the
specimens could have led to more imminent failure of the specimens and hence the
lowered tensile strength of the specimens.
Table 1. Calculated Young’s modulus, tensile strength, and failure strain for each specimen
Specimen #
Young’s modulus
(GPa)
Tensile strength
(MPa)
Failure strain (%)
1
66.20
187
0.37
2
41.85
236
0.49
3
57.60
435
0.41
Figures 5, 6, and 7 show the strain field εyy, photograph of the fractured
specimens, digital microscope images of the fractured regions, and the microscopic
images from specimens 1, 2 and 3 before testing, respectively. It should be mentioned
that part (c) of Figures 5, 6, and 7 are the regions in which the failure occurred taken
before applying tabs, painting the specimen, and tensile tests. These figures show
that the major failure mode in all three specimens was matrix cracking, instead of
CNT fiber failure, which also explains the lower tensile strength of the prepared
CNTRC specimens in comparison to those reported in [25], as the major failure mode
reported were the CNT fiber failure. The reason why that the matrix cracking occur
is the fact that the original laminate was not uniform in the width in longitudinal
direction, which leads to the specimens not being perfectly unidirectional. It can also
be concluded that the nonuniformity in the width along the longitudinal direction and
voids causes the specimen to be under nonuniform strain during testing, leading to
the matrix cracking failure mode of the CNTRC.
Additionally, the DIC results demonstrated that the specimen #1, #2, and #3 on
the right side, where the fracture occurred, are under strain. In other words, these
region for specimen #1, #2, and #3 has 30%, 64%, and 34% higher strain compared
to regions which are under lowest strain, respectively. Generally, from Figs. 5(c),
6(c), and 7(c), it can be observed that some of the CNT yarns were cut in half and not
connected to both ends of the specimens. Specifically, Fig. 5(c) demonstrated that
due to the fact that fibers are not aligned perfectly unidirectional, some part of CNTs
at the edges are cut which made the specimen prone to failure in those parts. This
incident is more observable in specimen #3, i.e., Fig.7(c), in which the specimen
failed in the exact same spot that CNT yarn is cut and pulled out. Given this, the load
bearing capability of the specimens are significantly reduced, thereby causing the
tensile strength to be dramatically reduced. Also, it can be seen that the tensile
strength and the Young’s modulus of specimen #2 is much higher than the other
specimens, as only a small portion of the yarns are cut and the rest of the specimen
stayed intact to bear the applied load. In addition to this, specimen #2 failed by matrix
cracking towards the center of the laminate, which can be explained by Fig. 6(c), i.e.,
the microscope image taken before the tensile testing, where the region with a large
void due to the lack of resin impregnation acted as the initiation spot of the matrix
cracking.
Figure 5. (a) the DIC strain (εyy), (b) the fractured specimen, and (c) microscopic image before
tensile test for specimen #1.
(a) (b)
Fractured part
(c)
5mm
Figure 6. (a) the DIC strain (εyy), (b) the fractured specimen, and (c) microscopic image before
tensile test for specimen #2.
Fig.7. (a) the DIC strain (εyy), (b) the fractured specimen, and (c) microscopic image before tensile
test for specimen #3.
(c)
(a) (b)
Fractured part
5mm
(a) (b)
Fractured part
(c)
5mm
CONCLUSION
The unidirectional CNT/epoxy resin composite laminate specimens with four
layers were fabricated using a custom-built fixture for winding the CNT yarns. The
specimen was meticulously examined via microscopic images to identify voids and
the alignment of the adjacent yarns. Then, using tensile tests, the Young’s modulus
as well as the tensile strength of prepared CNTRC specimens were obtained. Also,
through DIC, the strain field associated with each specimen was presented. The
results indicate that the flaws during the manufacturing process, causing the yarns to
be curved along the longitudinal direction, led to the specimen’s not being perfectly
unidirectional. This resulted in the nonuniform strain distribution and matrix cracking
failure of the specimens under tension. This can lead to diminishing of the specimens’
tensile strength by up to 78% in comparison with an existing study, of which the
tensile strength of CNT yarn composites was reported. By looking at the DIC results,
the failure prone area can be identified as the strain distribution is nonuniform
through the specimen and has up to 64% difference in between high and low strain
regions in the tensile loading direction. Due to nonuniformity of the CNT yarns, some
of the yarns are cut or narrowed down at the sides of the specimens. Also, it is
conceivable that, in the specimens that are more uniform, higher ultimate tensile
strength and modulus can be achieved since the yarns bear the applied load and not
the matrix. In order to achieve a better CNT reinforced composite, the voids and
nonuniformity must be controlled by optimizing the manufacturing process, after
which the presented fabrication process can be utilized to manufacture unidirectional
CNT yarn prepreg and can be employed to make laminated CNT composites with the
desired orientation to produce high negative Poisson’s ratios.
ACKNOWLEDGMENTS
The authors would like to acknowledge the financial support from National
Science Foundation under Award No. CMMI-2202737.
REFERENCE
[1] S. Lampkin, W. Lin, M. Rostaghi-Chalaki, K. Yousefpour, Y. Wang, and J. Kluss, “Epoxy
resin with carbon nanotube additives for lightning strike damage mitigation of carbon fiber
composite laminates,” in American Society for Composites (ASC) 34th Technical
Conference. Atlanta, 2019.
[2] K. Yousefpour, W. Lin, Y. Wang, and C. Park, “Protection of Carbon Fiber Reinforced
Polymer Matrix (CFRP) Composite Laminate Against Lightning Strike Using Nano-Fillers,”
in 2020 IEEE Conference on Electrical Insulation and Dielectric Phenomena (CEIDP),
IEEE, 2020, pp. 35–38.
[3] Y. Fan and Y. Wang, “A study on effect of auxeticity on impact behavior of carbon
nanotube reinforced composite laminates,” in 35th Annual American Society for Composites
Technical Conference, ASC 2020, DEStech Publications, 2020, pp. 1634–1648.
[4] W. Lin, Y. Wang, K. Yousefpour, C. Park, and V. Kumar, “Evaluating the lightning strike
damage tolerance for CFRP composite laminates containing conductive nanofillers,” Appl.
Compos. Mater., vol. 29, no. 4, pp. 1537–1554, 2022.
[5] K. Jiang, J. Wang, Q. Li, L. Liu, C. Liu, and S. Fan, “Superaligned Carbon Nanotube
Arrays, Films, and Yarns: A Road to Applications,” Adv. Mater., vol. 23, no. 9, pp. 1154–
1161, Mar. 2011, doi: https://doi.org/10.1002/adma.201003989.
[6] S. Iijima, “Helical microtubules of graphitic carbon,” Nature, vol. 354, no. 6348, pp. 56–58,
1991, doi: 10.1038/354056a0.
[7] S. Roy, R. S. Petrova, and S. Mitra, “Effect of carbon nanotube (CNT) functionalization in
epoxy-CNT composites,” vol. 7, no. 6, pp. 475–485, 2018, doi: doi:10.1515/ntrev-2018-
0068.
[8] S. Bellucci, C. Balasubramanian, F. Micciulla, and G. Rinaldi, “CNT composites for
aerospace applications,” J. Exp. Nanosci., vol. 2, no. 3, pp. 193–206, Sep. 2007, doi:
10.1080/17458080701376348.
[9] X.-L. Xie, Y.-W. Mai, and X.-P. Zhou, “Dispersion and alignment of carbon nanotubes in
polymer matrix: A review,” Mater. Sci. Eng. R Reports, vol. 49, no. 4, pp. 89–112, 2005,
doi: https://doi.org/10.1016/j.mser.2005.04.002.
[10] A. V Radhamani, H. C. Lau, and S. Ramakrishna, “CNT-reinforced metal and steel
nanocomposites: A comprehensive assessment of progress and future directions,” Compos.
Part A Appl. Sci. Manuf., vol. 114, pp. 170–187, 2018, doi:
https://doi.org/10.1016/j.compositesa.2018.08.010.
[11] W. A. Curtin and B. W. Sheldon, “CNT-reinforced ceramics and metals,” Mater. Today, vol.
7, no. 11, pp. 44–49, 2004, doi: https://doi.org/10.1016/S1369-7021(04)00508-5.
[12] S. Bal and S. S. Samal, “Carbon nanotube reinforced polymer composites—A state of the
art,” Bull. Mater. Sci., vol. 30, no. 4, pp. 379–386, 2007, doi: 10.1007/s12034-007-0061-2.
[13] Y. Zare, “Effects of interphase on tensile strength of polymer/CNT nanocomposites by
Kelly–Tyson theory,” Mech. Mater., vol. 85, pp. 1–6, 2015, doi:
https://doi.org/10.1016/j.mechmat.2015.02.002.
[14] M. M. Shokrieh, A. Saeedi, and M. Chitsazzadeh, “Evaluating the effects of multi-walled
carbon nanotubes on the mechanical properties of chopped strand mat/polyester
composites,” Mater. Des., vol. 56, pp. 274–279, 2014, doi:
https://doi.org/10.1016/j.matdes.2013.11.017.
[15] S. I. Hussein, A. M. Abd-Elnaiem, T. B. Asafa, and H. I. Jaafar, “Effect of incorporation of
conductive fillers on mechanical properties and thermal conductivity of epoxy resin
composite,” Appl. Phys. A, vol. 124, no. 7, p. 475, 2018, doi: 10.1007/s00339-018-1890-0.
[16] Q. Chen, S. Du, Z. Jiang, Y. Liu, R. Du, and G. Zhao, “Mechanical properties of foam
sandwich with chopped-glass-fiber/carbon nanotube reinforced hierarchical structure
interlayer,” Polym. Compos., vol. 41, no. 8, pp. 3411–3420, Aug. 2020, doi:
https://doi.org/10.1002/pc.25630.
[17] Y. Inoue et al., “Study on the mechanical and electrical properties of twisted CNT yarns
fabricated from CNTs with various diameters,” Carbon N. Y., vol. 176, pp. 400–410, 2021,
doi: https://doi.org/10.1016/j.carbon.2021.01.139.
[18] K. Sugano, M. Kurata, and H. Kawada, “Evaluation of mechanical properties of untwisted
carbon nanotube yarn for application to composite materials,” Carbon N. Y., vol. 78, pp.
356–365, 2014, doi: https://doi.org/10.1016/j.carbon.2014.07.012.
[19] T.-W. Chou, L. Gao, E. T. Thostenson, Z. Zhang, and J.-H. Byun, “An assessment of the
science and technology of carbon nanotube-based fibers and composites,” Compos. Sci.
Technol., vol. 70, no. 1, pp. 1–19, 2010, doi:
https://doi.org/10.1016/j.compscitech.2009.10.004.
[20] X.-H. Zhong et al., “Continuous Multilayered Carbon Nanotube Yarns,” Adv. Mater., vol.
22, no. 6, pp. 692–696, Feb. 2010, doi: https://doi.org/10.1002/adma.200902943.
[21] J. M. Gardner et al., “3-D printing of multifunctional carbon nanotube yarn reinforced
components,” Addit. Manuf., vol. 12, pp. 38–44, 2016, doi:
https://doi.org/10.1016/j.addma.2016.06.008.
[22] Y. Han and J. Elliott, “Molecular dynamics simulations of the elastic properties of
polymer/carbon nanotube composites,” Comput. Mater. Sci., vol. 39, no. 2, pp. 315–323,
2007, doi: https://doi.org/10.1016/j.commatsci.2006.06.011.
[23] Y. Fan and Y. Wang, “The effect of negative Poisson’s ratio on the low-velocity impact
response of an auxetic nanocomposite laminate beam,” Int. J. Mech. Mater. Des., vol. 17,
no. 1, pp. 153–169, 2021, doi: 10.1007/s10999-020-09521-x.
[24] J.-W. Kim et al., “Assessment of carbon nanotube yarns as reinforcement for composite
overwrapped pressure vessels,” Compos. Part A Appl. Sci. Manuf., vol. 84, pp. 256–265,
2016, doi: https://doi.org/10.1016/j.compositesa.2016.02.003.
[25] J.-W. Kim et al., “Undirectional carbon nanotube yarn/polymer composites,” 2018.
[26] F. Baber et al., “Microstructural exploration of a carbon nanotube yarn reinforced composite
using a peridynamic approach,” J. Compos. Mater., vol. 56, no. 6, pp. 861–876, 2022.
[27] W. Lin and Y. Wang, “Low velocity impact behavior of auxetic CFRP composite laminates
with in-plane negative Poisson’s ratio,” J. Compos. Mater., vol. 57, no. 12, pp. 2029–2042,
Apr. 2023, doi: 10.1177/00219983231168698.
[28] Y. Wang, “Auxetic Composite Laminates with Through-Thickness Negative
Poisson’s Ratio for Mitigating Low Velocity Impact Damage: A Numerical Study,”
Materials, vol. 15, no. 19. 2022. doi: 10.3390/ma15196963.
[29] D. Solav and A. Silverstein, “DuoDIC: 3D Digital Image Correlation in MATLAB,” J. Open
Source Softw., vol. 7, no. 74, p. 4279, 2022.
