10.2417/spepro.

005040

Injection-molded hybrid
polymer composites with
improved thermal isotropy
Pablo Fabian Rios, Samuel Kenig, Rina Cohen, and Amir Shechter

As next-generation materials, hybrid polymer composites synergisti-

cally combine the properties of traditional polymer composites and
more recent polymer nanocomposites.

Plastics have a weight advantage over metals but lack their strength,
stiffness, dimensional stability, and electrical conduction. Fiber-
reinforced plastics—specifically, engineering thermoplastics contain-
ing carbon fibers (CFs)—have been used to bridge this properties gap.
Plastics also have a relatively large coefficient of linear thermal expan-
sion (CTE) compared with metals and ceramics. In this respect, CFs are
exceptional. Their CTE is negative in the fiber direction1 (i.e., CFs con-
tract with increasing temperature), which significantly reduces the CTE
of the polymer composite, though mostly in this direction. In thermo-
Figure 1. Composite sample location and direction. A: In-flow.
plastic injection molding, the polymer flows into a constrained cavity
B: Cross-flow.14
mold, causing fiber orientation that cannot be fully controlled. Conse-
quently, injection-molded parts containing oriented fibers will contract
or expand anisotropically (i.e., in a direction-dependent manner) with
represents the result for neat PEI. It shows a relatively high CTE and no
temperature changes. This behavior limits their use in high-precision
significant anisotropy (difference between the CTE-in and CTE-cross).
components.2
The addition of first 1.75wt% CNTs and then 3.5wt% CNTs (second
Carbon nanotubes (CNTs) have proven to be efficient in improv-
and third column pairs from the left) did not reduce the CTE signif-
ing the mechanical, thermal, and electrical properties of polymers.3–7
icantly and produced no serious anisotropy. In contrast, addition of
Some work has been done with polymer hybrid composites containing
7.5wt% CFs (fourth column pair from the left) drastically reduced the
both CFs and CNTs,8–13 but the effects of their addition on injection-
CTE in the cross-flow direction, producing a highly anisotropic sample.
molded polymers have not been thoroughly investigated.
When 3.5wt% CNTs were added to this sample (fifth column pair from
In the work described here, we produced polymer composite for-
the left), the CTE in the in-flow direction was reduced and the sample
mulations of polyether imide (PEI, Ultem 1000) containing CFs (RTP
became less anisotropic.
2183) and/or multiwalled CNTs (Arkema CM18-15) and injection-
This phenomenon is also evident in the samples containing 15wt%
molded them into a center-gated disk mold 10cm across and 3.2mm
CFs. For the sample containing only 15wt% CFs (sixth column pair
thick. We cut samples from the disks and evaluated them in the direc-
from the left), the CTE was significantly reduced, up to a level of nega-
tion of the polymer flow (in-flow) and perpendicular to the polymer
tive CTE in the cross-flow direction, but the sample is very anisotropic.
flow (cross-flow): see Figure 1. We evaluated thermal isotropy (i.e.,
The addition of 1.75wt% CNTs (seventh column pair from left) reduced
orientational symmetry) by measuring the CTE in both directions.
the sample anisotropy. The addition of 3.5wt% CNT (last column pair
Figure 2 shows the results of the in-flow and cross-flow CTE for
eight composite formulations. The first column pair from the left Continued on next page
 10.2417/spepro.005040 Page 2/3

from the left) also reduced anisotropy, though to a lesser extent than
in the previous case. The fact that we observed the lowest CTE values
in the cross-flow direction for samples containing CFs is a strong sign
that the fibers have been preferably aligned in this direction. Although
this result may seem contradictory, fiber orientation in the cross-flow
direction is possible in injection molding of a center-gated disk as a
consequence of spreading radial flow.16
To validate these results, we combined 2.5 and 5.0wt% multi-
walled CNTs (Nanocyl 7000, Nanocyl) with polycarbonate (PC) (Lupi-
lon S-3000-UR, Mitsubishi) containing 10wt% CFs (PZ3306R6 NT,
Polyram). We made the composites at a different production site using
a different compounder, injection machine, and mold, in this case in a
center-gated cover-type mold (see Figure 3).
Figure 4. CTE for the confirmation test samples, in-flow and cross-flow
( 20 to C50◦C range).15

Figure 2. Coefficient of linear thermal expansion (CTE) results, in-flow

and cross-flow ( 20 to C50◦C range).14, 15 CNT: Carbon nanotube.
CF: Carbon fiber.

Figure 5. Scanning electron microscopy (SEM) image of PEI/CF/CNT

sample (cross-flow fracture).15 PEI: Polyether imide.

Figure 4 shows the results of the CTE for this test. The addition
of 5wt% CNTs resulted in no major reduction of the CTE compared
with neat PC, nor any significant anisotropy. In contrast, the addition
of 10wt% CFs did significantly reduce the CTE, especially in the cross-
flow direction, and also produced high anisotropy. Adding first 2.5wt%
and then 5wt% CNTs to the samples containing 10wt% CFs signifi-
cantly reduced the anisotropy of the sample part. The sample contain-
ing 5wt% CNTs and 10wt% CFs had a very low CTE and was very
isotropic.

Figure 3. Confirmation test cover part.15

Continued on next page

 10.2417/spepro.005040 Page 3/3

Israel Institute of Technology, Haifa, and his PhD in plastics engineer-

ing from the University of Massachusetts Lowell.

Rina Cohen and Amir Shechter

Elbit Systems Electro-optics (Elop)
Rehovot, Israel

References
1. D. D. L. Chung, Carbon Fiber Composites, p. 74, Butterworth-Heinemann, Newton,
MA, 1994.
2. R. A. Malloy, Plastic Part Design for Injection Molding: An Introduction, pp. 20
and 171, Hanser Publishers, Munich, 1994.
3. P. J. F. Harris, Carbon Nanotube Science: Synthesis, Properties, and Applications,
Cambridge University Press, Cambridge, UK, 2009.
4. W. Bauhofer and J. Z. Kovacs, A review and analysis of electrical percolation in carbon
nanotube polymer composites, Compos. Sci. Technol. 69, p. 1486, 2009.
5. J. N. Coleman, U. Khan, W. J. Blau, and Y. K. Gun’ko, Small but strong: a review of the
mechanical properties of carbon nanotube-polymer composites, Carbon 44, p. 1624,
Figure 6. CNTs bundle on a CF in a PC/CF/CNT confirmation test 2006.
6. O. Breuer and U. Sundararaj, Big returns from small fibers: a review of polymer/carbon
sample.15 PC: Polycarbonate.
nanotube composites, Polym. Compos. 25, p. 630, 2004.
7. E. T. Thostenson, Z. Ren, and T.-W. Chou, Advances in the science and technology of
carbon nanotubes and their composites: a review, Compos. Sci. Technol. 61, p. 1899,
Scanning electron microscopy examination of the samples helped
2001.
explain the CTE findings. CFs are stiff and straight and orient prefer- 8. P. K. Jain, B. Padya, P. S. Rao, K. M. K. Chowdary, B. Aswanikumar, and G. Anusha,
entially in the cross-flow direction, resulting in CTE anisotropy (see Thermal and mechanical properties of multiscale carbon nanotubes and carbon fiber
reinforcements in epoxy hybrid nanocomposites, J. Nanostruct. Polym. Nanocompos.
Figure 5). In contrast, CNTs are flexible and do not align or disperse. 7, p. 81, 2011.
Consequently, they form bundles between the CFs (see Figure 6). The 9. S. U. Khan, C. Y. Li, N. A. Siddiqui, and J.-K. Kim, Vibration damping characteris-
tics of carbon fiber-reinforced composites containing multi-walled carbon nanotubes,
hybrid composite consists of a polymer matrix containing aligned stiff Compos. Sci. Technol. 71, p. 1486, 2011.
CFs with flexible CNT bundles between them. The bundles act like 10. E. T. Thostenson, J. J. Gangloff Jr., C. Li, and J.-H. Byun, Electrical anisotropy in
multiscale nanotube/fiber hybrid composites, Appl. Phys. Lett. 95, p. 073111, 2009.
‘dampers’ that moderate the orientation effects and increase the ther- 11. E. T. Thostenson, W. Z. Li, D. Z. Wang, Z. F. Ren, and T. W. Chou, Carbon nano-
mal isotropy of the part. tube/carbon fiber hybrid multiscale composites, J. Appl. Phys. 91, p. 6034, 2002.
12. M. Al-Haik, C. C. Luhrs, M. M. Reda Taha, A. K. Roy, L. Dai, J. Phillips, and S. Doorn,
In summary, we found that adding CNTs to PEI/CF composites re-
Hybrid carbon fibers/carbon nanotubes structures for next generation polymeric com-
duces the anisotropy of thermal expansion. A confirmation test repro- posites, J. Nanotechnol., p. 860178, 2010.
duced our experimental results, even though the test composites were 13. A. Godara, L. Gorbatikh, G. Kalinka, A. Warrier, O. Rochez, L. Mezzo, F. Luizi, A.
W. van Vuure, S. V. Lomov, and I. Verpoest, Interfacial shear strength of a glass
made from a different polymer, using a different mold, and produced fiber/epoxy bonding in composites modified with carbon nanotubes, Compos. Sci.
with different processing equipment. Morphological examination pro- Technol. 70, p. 1346, 2010.
14. P. F. Rios, S. Kenig, R. Cohen, and A. Shechter, The effect of carbon nanotubes on the
vided an explanation of the results. Our findings show that CNTs can be thermal expansion isotropy of injection molded carbon fiber reinforced thermoplastics,
used to fine-tune polymer composites with high-dimensional thermal Polym. Compos. 34, p. 1367, 2013. doi:10.1002/pc.22551
15. P. F. Rios, S. Kenig, R. Cohen, and A. Shechter, Injection molded carbon fiber rein-
stability that are suitable for high-performance devices. Further work forced thermoplastics with improved thermal expansion isotropy, Eurotec 2013, 2013.
will include evaluating the effects of other nanoadditives (apart from Paper 1608842
16. S. Kenig, Fiber orientation development in molding of polymer composites, Polym.
CNTs) on the thermal isotropy of reinforced thermoplastics containing
Compos. 7, p. 50, 1986.
either CFs or glass fibers.

Author Information

Pablo Fabian Rios and Samuel Kenig

Shenkar College of Engineering and Design
Ramat Gan, Israel

Pablo Fabian Rios is a senior lecturer. He received his MSc in materials

engineering and his BSc in electrical engineering both from Technion,

c 2013 Society of Plastics Engineers (SPE)