JOURNAL OF

COMPOSITE
Article M AT E R I A L S
Journal of Composite Materials
2016, Vol. 50(4) 435–446
! The Author(s) 2015
Fabrication and characterization of Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
silicon carbide/epoxy nanocomposite DOI: 10.1177/0021998315576378
jcm.sagepub.com
using silicon carbide nanowhisker
and nanoparticle reinforcements

Mohammadreza Naeimirad1, Ali Zadhoush1 and Rasoul

Esmaeely Neisiany2

Abstract
In this study, the influence of filler shape and filler content on the physical and mechanical properties of silicon carbide/
epoxy nanocomposites was investigated using silicon carbide nanoparticles and nanowhiskers. Samples containing 0.5, 1,
2 and 4 wt% of b-silicon carbide nanowhiskers and nanoparticles in epoxy resin were prepared using a high-intensity
ultrasonic liquid processor and casting technique. Mechanical and physical tests such as tensile, flexural, hardness, and
wear along with a morphological investigation by FT-IR, scanning electron microscopy and transition electron microscopy
were performed. Results of mechanical tests indicating about 20% and 40% improvement were achieved in nanoparticle
(1 wt%) and nanowhisker (2 wt%) reinforced samples, respectively. Also mechanical experiment results have been
evaluated with theoretical models. Tribological results show that wear and frictional properties improved about
50% and 30%, respectively, by adding silicon carbide nanoparticles (4 wt%) and nanowhiskers (4 wt%) to epoxy resin.
TEM and SEM observations showed that nano-fillers are dispersed in low weight ratio but agglomerated above a
critical content.

Keywords
Epoxy, silicon carbide, nanoparticle, nanowhisker, tribology, mechanical properties

made to make nanoparticle-filled polymers for improv-

Introduction
ing tribological, thermal and mechanical perform-
In recent years, there has been a great need for ance.3–7 Extremely high surface area is one of the
composite materials with enhanced performance and most attractive characteristics of the nanoparticles
outstanding properties.1 Nanocomposites are new com- as it facilitates creating a large interface in a
posite materials in which the matrix material is rein- composite.8,9 The importance of aspect ratio at the
forced by one or more separate nanomaterials in microscale is well-known theoretically, but the signifi-
order to improve performance properties.2 Epoxy cance of aspect ratio in nanoscale has not been devel-
resin has been of significant importance to the engineer- oped yet. Bulk SiC is a ceramic material with high
ing community for many years and has been made by hardness, wear resistance, high temperature stability
the reaction of Bis-phenol A and Epichlorohydrin. and chemical inertness that has many applications in
Components made of epoxy-based materials have pro-
vided outstanding mechanical, thermal and electrical
properties, and ease of processing. Important factors 1
Department of Textile Engineering, Isfahan University of Technology,
influencing their performance are the molecular struc- Isfahan, Iran
2
ture, curing conditions and the ratio of the epoxy resin Department of Chemical Engineering, Isfahan University of Technology,
to curing agent(s).3,4 Isfahan, Iran
Using an additional phase (e.g. inorganic fillers) to
Corresponding author:
strengthen the properties of epoxy resins has become a Ali Zadhoush, Department of Textile Engineering, Isfahan University of
common practice. With the rapid development of nano- Technology, Isfahan 84156-83111, Iran.
phased materials in recent years, attempts have been Email: [email protected]
436 Journal of Composite Materials 50(4)

abrasive, metallurgy, composite and electronic similar study, presented the effect of surface-modified
industries.3,10,11 SiC nanoparticles on friction and wear properties of
In addition to SiC nanoparticles that have a and b epoxy composites. Chisholm et al.10,22 evaluated the
isomorphs, SiC nanowhiskers are one of the new effect of SiC nanoparticles on the mechanical pro-
synthesized materials made by crystals grown in perties of carbon/epoxy composites and showed
Miller index [1 1 1] direction. There are new methods improvement of mechanical and thermal properties by
for nanowhisker synthesis such as growing inside nano- the addition of SiC nanoparticles up to a critical volume
tubes, chemical vapor deposition, catalytic pyrolisis content. In another study by Rodgers et al.,4 improve-
etc.12 ment of mechanical and thermal properties of epoxy
The mechanical performance of SiC whiskers has resin by infusion of SiC nanoparticles was reported.
been established for many years (1959). Petrovic To the best of our knowledge, the effect of aspect
et al.,13 using a specially developed microtensile ratio in SiC nanocomposites has not been investigated
tester, reported that SiC whiskers exhibit an average yet. This paper aims at the preparation and, mechanical
tensile strength of 8.40 GPa and an average elastic and tribological characterization of SiC/epoxy compos-
modulus of 581 GPa, and were considerably stronger ites with SiC nanowhiskers and nanoparticles aiming to
and stiffer than continuous, polycrystalline SiC fibers. distinguish properties related to the shape. In the pre-
These results indicate the significant potential of SiC sent study, SiC nanoparticles and nanowhiskers were
whiskers as short-fiber reinforcement elements for cer- infused through sonic cavitation with a room tempera-
amic matrix and high-temperature composites.13 There ture cure epoxy matrix. The prepared resin compound
is presently great interest in advanced reinforcement was then cast, cured and characterized for mechanical
materials with improved strength, oxidation resistance properties. In addition to the morphological and tribo-
and chemical stability, for example in epoxy and cera- logical study, one of the main objectives of this work
mic matrix composites.14,15 was to develop the importance of the aspect ratio in
Another study by Yuan et al.,16 using a solution nanoscale on tensile and flexural properties and com-
method, demonstrated that the thermal conductivity, parison with micromodels.
glass transition temperature, thermal stability and
dynamical mechanical properties of epoxy composites
could all be improved by incorporating hyper-branched Experimental
aromatic polyamide grafted silicon carbide (SiC-HBP)
Materials
whiskers. Zhang et al.17 showed that silicon carbide
and alumina whiskers can significantly improve the The resin used in this study was epoxy (Araldite LY564)
flexural modulus and wear resistance of the epoxy com- and its hardener (Aradur 560) obtained from Huntsman
posite. In another study by Gu et al.,18 it has been Co., Germany. It is a low-viscosity two-phased tough-
shown that both the flexural and impact strength of ened epoxy resin system consisting of part A (resin:
the SiC/epoxy composites initially improved with the diglycidyl ether of bisphenol A, (DGEBA)) and part B
increasing content of SiC microparticles, but decreased (hardener: tetramine). The filler b-Silicon carbide (b-SiC)
with excessive addition of SiC. nanoparticles are spherical in shape and SiC nanowhis-
One of the biggest challenges in the manufacturing kers are needle shaped, and were purchased from Sky
process of particulate (nano or micro) composites is Spring Nano Ltd., USA. Physical properties of the
dispersion of the fillers in the matrix material. Only materials are shown in Table 1, according to the data
well-dispersed nanoparticles can lead to unique new sheet provided by the manufacturers.
properties of the composite, such as a simultaneous
improvement of toughness and stiffness in very low
filler content.9 There exist various techniques for the Table 1. Material properties.
dispersion of particle agglomerates in a liquid, in
Epoxy SiC SiC
which the dissolver and the bead mill are the most
Material/property resin Nanowhisker Nanoparticle
common machines. Another possibility is dispersion
with the assistance of ultrasonic waves.8 Ultrasonic Color Yellow Grey green Grey green
treatment could also influence matrix mechanical prop- Density (g/cm3) 1.1–1.2 3.216 3.22
erties.10 The effect of different micro and nanoparticle Diameter (nm) – 50–100 50
fillers on wear performance of polymers have been Length (nm) – 200–800 –
investigated before.19,20 Luo et al.’s21 research work Aspect ratio – 2–16 1
on epoxy’s tribological properties using SiC nanoparti- Tensile modulus 2.2–2.7 GPa 480 GPa 480 GPa
cles shows a reduction of specific wear rate and friction
Tensile strength 81–89 MPa 20.8 GPa 20.8 GPa
coefficient by adding of SiC fillers. Ji et al.,6 in another
Naeimirad et al. 437

Table 2. Samples abbreviations. plates with 100–200 nm thickness were prepared using
ultramicrotome (Reichert, Om-U3) equipped with
Abbreviation Filler type Filler wt% Sonicated
diamond knife. Also SEM observation was carried
Epoxy – 0 No out to study the composite fractography using a
S-Epoxy – 0 Yes high resolution scanning electron microscope
0.5 NW Nanowhisker 0.5 Yes (Hitachi S-4800) under 2 KV, after conductive coating
1 NW Nanowhisker 1 Yes of samples (5 nm gold) by high-vacuum sputter-coat-
2 NW Nanowhisker 2 Yes
ing (Leica ACE-600).
The tensile strength of the samples was measured by
4 NW Nanowhisker 4 Yes
a universal testing machine (Zwick Instrument 1446)
0.5 NP Nanoparticle 0.5 Yes
with a crosshead speed of 1 mm/min according to
1 NP Nanoparticle 1 Yes ASTM D638.23 Flexural tests under three-point bend
2 NP Nanoparticle 2 Yes configuration were performed according to ASTM
4 NP Nanoparticle 4 Yes D790.24 The tests were conducted with a 100-KN
servo hydraulic testing machine (Instron 5052) under
displacement control mode at a crosshead speed of
5 mm/min, at room temperature. The dumbbell-
Preparation of composites shaped specimen for tensile test was type 5, ASTM
Composites were manufactured using different concen- D638. For the flexural test, samples were type 5,
trations of reinforcement: wt% of 0.5, 1, 2 and 4. The ASTM D790. Five replicate specimens from each cat-
dispersion of fillers in resin was carried out with a high- egory (neat and various wt.-% loadings) were tested for
intensity ultrasonic liquid processor (UP 400S, 40 KHz both tensile and flexural properties.
Sonics Vibra Cell, Hielscher, Inc., Germany). The man- The hardness of samples was identified by micronee-
ufacturing of nanocomposites began with the disper- dle probe tester type TH 210 in shore-D. Tribological
sion of SiC nanomaterials in part A of epoxy resin. tests were performed by the pin on disc method accord-
The reinforcements were dispersed in part A since this ing to ASTM G99.25 Three samples in a disc shape,
part is less reactive to ultrasonic waves compared to rotated at a speed of 0.1 m/sec, were abraded with a
part B. Pre-calculated amounts of SiC and part A steel pin pressing a 15-kg weight over the moving com-
were carefully weighed, and mixed together in a suitable posite disc. Tribological parameters (like friction factor
beaker. Mixing was carried out with high-intensity and wearing coefficient) were measured during rotation
ultrasonic waves for one hour. In order to avoid a tem- of the disc (for a length of 1000 meters).
perature rise during the sonication process, external A two-way analysis of variance (ANOVA) method
cooling was employed by submerging the beaker con- was applied using Microsoft Excel software for statis-
taining the mixture in an ice-water-salt bath. Once the tical analysis.26
sonication was completed, part B was added to the
modified part A. It should be noted that the mix ratio
of part A and part B of epoxy was 100:27 according to Results and discussion
the manufacturer’s instructions. The mixing was carried
out using a high-speed mechanical stirrer for about
Morphology
5 minutes. Then the mixture was transferred into Transmission Electron Microscopy (TEM) observa-
Teflon-coated metal rectangular molds and kept for tions were carried out to investigate the distribution
24 hours at room temperature. The cured material of nano-SiC fillers. TEM micrographs of nanoparticle
was then de-molded and trimmed. Samples were and nanowhisker reinforced composites with various
named according to Table 2. Finally, the test specimens filler contents are shown in Figures 1 and 2.
were machined for mechanical and tribological charac- It can be observed from Figure 1 that nanoparti-
terization using CNC machine. cles have a diameter of about 50-nm, and are dis-
persed uniformly in 0.5 and 1 wt% contents. But
when the filler content increased, as it is observed in
Characterization the TEM micrographs (c and d), the particles were
Epoxy resin (part A) was analyzed by FT-IR (Bomem agglomerated. Dispersion of nanomaterials at low
MB-100) before and after sonication to study the content is due to the ultrasonic wave effect and this
ultrasonic influence on the resin structure avoiding dispersion can influence the mechanical properties
chemical destruction. Morphological investigations significantly.
were performed by transition electron microscopy In Figure 2 it is also observed that an approximate
(TEM; Zeiss, EM-900) under 50 KV voltage. Sample dispersion of nanowhiskers for 0.5, 1 and 2 wt% is
438 Journal of Composite Materials 50(4)

Figure 1. TEM micrographs of (a) 0.5 NP, (b) 1 NP, (c) 2 NP and (d) 4 NP samples.

Figure 2. TEM micrographs of (a) 0.5 NW, (b) 1 NW, (c) 2 NW and (d) 4 NW samples.

Figure 3. SEM micrographs of (a) Epoxy, (b) 1 NP, (c) 2 NP and (d) 4 NP samples fracture surface.

obtained while for 4 wt% content, agglomeration epoxy is different from that containing nanocomposites.
occurred. Agglomeration of nano SiC fillers at high Agglomeration of nanoparticles and nanowhiskers was
content is due to the fillers interaction. again observed in high concentration fillers (4 wt%).
Figures 3 and 4 show the SEM images indicating The FT-IR spectrum of neat and sonicated epoxy
the brittle fracture in samples. The surface of neat resin was compared qualitatively, as shown in Figure 5.
Naeimirad et al. 439

Figure 4. SEM micrographs of (a) Epoxy, (b) 1 NW, (c) 2 NW and (d) 4 NW samples fracture surface.

Figure 5. FT-IR spectrum of epoxy resin (before and after ultrasonic).

Analysis of the spectrum by standard graphs confirmed Figures 6 and 7 show that ultrasonic treatment
the epoxy chemical construction and type (DGEBA).27 caused a limited increase in mechanical properties
The same peak places in both spectrums and the absence for neat epoxy. This improvement could be due
of a new peak after the sonication indicated no change to the effect of ultrasonic waves on viscosity of
occurrence in the resin structure due to sonication. Also resin and decrease of voids content. This is in
the most important peaks of epoxy (oxiran bands) accordance with the results published in the litera-
around 3057 cm1 (stretching of C-H of the oxiran ture.10 Within the experimental conditions carried
ring) and 915 cm1 (oxiran C-O band) are same before out, the best tensile results were obtained using
and after ultrasonic treatment. 2 wt% of nanowhisker and 1 wt% of nanoparticle
reinforcement.
These results show that sonication process obtained
Mechanical tests results
a suitable distribution and there was a good compati-
Tensile and flexure tests were performed to evaluate the bility between SiC and epoxy resin. Further increase of
bulk stiffness and strength of nanocomposites. Tensile nanowhiskers up to 4% lowered tensile properties, an
test results are shown in Figures 6 and 7. indication of fillers agglomeration and worsening of the
440 Journal of Composite Materials 50(4)

Table 3. Tensile test data for SiC/epoxy nanocomposites.

Tensile Tensile
modulus Gain/Loss strength Gain/loss
Sample [GPa] (%) [MPa] (%)

Epoxy 2.360  0.008 d 86.0  1.2 d

S-Epoxy 2.445  0.040 þ3.6 91.9  1.9 þ6.8
0.5 NW 2.670  0.018 þ13.1 104.0  3.8 þ20.9
1 NW 2.870  0.090 þ21.6 111.9  2.9 þ30.1
2 NW 3.225  0.080 þ36.6 115.1  2.0 þ33.8
Figure 6. SiC nanowhisker composites tensile test stress– 4 NW 2.990  0.085 þ26.7 105.1  1.1 þ22.2
strain diagram. 0.5 NP 2.605  0.060 þ10.3 99.7  1.8 þ15.9
1 NP 2.751  0.030 þ16.5 109.2  3.9 þ26.9
2 NP 2.670  0.075 þ13.1 105.4  2.6 þ22.5
4 NP 2.657  0.090 þ12.0 102.2  0.7 þ18.8

Table 4. Experimental and theoretical tensile modulus.

Experimental (MPa) Theoretical (MPa)

Wt% Vp NW NP Elower Halpin-Tsai Hui-Shai

0.5 0.186 2670 2605 2364 2420 2369

1 0.372 2870 2751 2368 2490 2378
2 0.745 3225 2670 2377 2621 2398
Figure 7. SiC nanoparticle composites tensile test stress–strain 4 1.49 2990 2657 2395 3920 2427
diagram.

distribution (observed in TEM and SEM images).

Agglomeration of nanoparticles occurred at 2 wt%.
In order to elaborate more on the results, tensile tests
data are shown in Table 3. It can be seen that a system-
atic increase in tensile strength and modulus is observed
from neat-epoxy up to 1% for nanoparticles and up to
2% for nanowhisker reinforced samples. Improvement
of mechanical performance is due to the nano-reinforce-
ments good dispersion and interaction with the polymer
matrix. The overall improvement in tensile properties for
nanowhiskers is higher than the improvement obtained
by nanoparticles due to the shape effect, i.e. the higher Figure 8. Comparison of experimental tensile modulus with
aspect ratio of nanowhiskers. theoretical models.
In order to find a correlation for predicting tensiles
properties of nanocomposites, theoretical models pre-
sented in appendix (Eq. (2), (7) and (15)) were used to factor in composite tensile modulus even at the nano-
calculate the tensile properties of the fabricated com- scale, and that nanowhiskers improved the matrix mod-
posite samples (aspect ratio has been assumed 1 for ulus more than nanoparticles. Experimental tensile
nanoparticles and an average 9 for nanowhiskers). strengths are also compared with theoretical models
Results are presented in Table 4 and Figure 8. (equation 4 and 5) in Table 5 and Figure 9. The S
According to Figure 8, the lower bound model factor in the Young model was selected as the minimum
underestimates experimental results similar to Hui- (9.8  104), due to the absence of any surface modifi-
Shai (H-S) approach. Halpin-Tsai (H-T) model has cation or silane treatment on SiC fillers.
the most compatibility with experimental results. It According to Figure 9 the Nicolais approach under-
can be concluded that aspect ratio is an important estimates experimental strengths while Young model
Naeimirad et al. 441

Table 5. Experimental and theoretical tensile strength.

Experimental (MPa) Theoretical (MPa)

Wt% Vp NW NP Nicolais Young

0.5 0.186 104 100 84 91.4

1 0.372 112 109 83 96.8
2 0.745 115 106 82 107
4 1.49 105 102 79.5 116

Figure 11. SiC nanoparticle composites flexural test stress–

strain diagram.

Table 6. Flexural test data for SiC/epoxy nanocomposites.

Flexural Flexural
modulus Gain/Loss strength Gain/Loss
Sample [GPa] [%] [MPa] [%]

Epoxy 3,390  0.030 d 137.4  1.6 d

S-Epoxy 3.595  0.025 þ5.9 144.4  1.6 þ5.1
0.5 NW 3.728  0.048 þ10.0 154.8  1.2 þ12.6
Figure 9. Comparison of tensile strength with theoretical
1 NW 4.070  0.100 þ20.0 177.2  2.2 þ28.9
models.
2 NW 4.588  0.082 þ35.3 197.6  1.6 þ43.8
4 NW 3.612  0.038 þ6.5 116.8  3.2 14.9
0.5 NP 3.686  0.046 þ8.7 149.8  1.2 þ9.0
1 NP 3.963  0.063 þ16.9 173.8  1.2 þ26.4
2 NP 3.851  0.041 þ13.5 157.8  2.2 þ14.8
4 NP 3.794  0.066 þ11.9 135.8  7.8 1.1

resulting in higher areas of interface for loading.

Mechanical properties were increased about 20% by
increasing SiC fillers aspect ratio. Also, mechanical
results are in accordance with microscopic observa-
Figure 10. SiC nanowhisker composites flexural test stress– tions. Due to the good dispersion up to 1–2 wt%,
strain diagram. improvement in mechanical properties was observed.
However, for higher content (4 wt%), due to the
agglomeration, this improvement in mechanical prop-
results are closer to the experimental data. Tensile erties was not observed.
strength improvement indicates compatibility and
good adhesion between SiC and epoxy resin. Results
Hardness
of the bending tests are shown in Figures 10 and 11.
Flexural test results are similar to the results Results of hardness tests are shown in Table 7.
obtained from tensile tests. The best results were The hardness of cured epoxy resin increased from 80
obtained using 2 wt% of nanowhisker and 1 wt% of to 86 shore D by the addition of SiC nanoparticles and
the nanoparticle. Further increase of fillers decreased nanowhiskers. This could be due to the higher hardness
flexural properties, an indication of particle agglomer- of SiC. It should be note that the maximum hardness
ation and worsening of the distribution. Flexural was achieved in the maximum loading content of SiC
modulus and strength are shown in Table 6. nanofillers (4 wt%).
Comparison of the mechanical test results of nano- As the fillers are nano-sized and hardness tester’s
composite samples shows that nanowhiskers have a microprobe for sensing, the hardness improvement
better improvement in properties. The reason for this was small (7%). The same amount of hardness for
phenomenon is the higher aspect ratio of nanowhiskers, nanoparticle and nanowhisker filled composite is due
442 Journal of Composite Materials 50(4)

Table 7. Hardness tests data.

Samples Epoxy S-Epoxy 0.5 NW 1 NW 2 NW 4 NW 0.5 NP 1 NP 2 NP 4 NP

Hardness 80  0.8 80  1.0 81.3  0.3 82.9  1.1 84.2  0.4 85.9  0.5 81.2  0.2 83  0.6 84.1  0.8 86  0.1
(Shore D)

Table 8. Tribological results obtained from pin on disc test.

Samples Epoxy S-Epoxy 0.5 NW 1 NW 2 NW 4 NW 0.5 NP 1 NP 2 NP 4 NP

Friction 800  16 800  15 750  21 700  9 600  25 550  32 700  12 610  24 400  28 420  9
coefficient – l
(104)
Wear rate – W 60  1.8 60  2.0 50  2.0 45  1.0 40  2.2 39  3.2 42  1.2 30  1.5 12  2.7 15  1.9
(104 mm3/Nm)

to same amount of hardness for SiC nanowhisker and 0.09

0.08 NP
nanoparticles (9.5 Mohs). It can be concluded that 0.07 NW
hardness properties of samples do not relate to the 0.06
filler shape. 0.05
µ

0.04
0.03
Tribological investigation 0.02
0.01
The results of pin on disc tribological tests on neat 0
epoxy and nanocomposite samples are shown in 0 1 2 3 4 5
Table 8. The frictional coefficient (l) and wear rate Filler Wt%

(W) of the samples are compared with each other in Figure 12. Frictional coefficient (l) of samples as a function of
Figures 12 and 13 schematically. SiC wt%.
As shown in Figure 12 and 13, the friction coefficient
and wear rates of samples were reduced with the addi-
tion of nanoparticle and nanowhisker content. There
may be two reasons for this phenomenon. First, by 70
NP
adding hard fillers the load is shared by the hard fillers, 60
W (×10 -4 mm3/Nm)

NW
therefore the soft polymer is subjected to lower load. 50
This reduces the damage to the polymer sliding surface. 40
In simple terms, the use of a filler material that is stiffer 30
than the polymer matrix may lead to a significant 20
amount of normal load being carried by the nanofillers. 10
Secondly, nanofillers help in the formation of transfer
0
film on the interface which promotes the sliding of the 0 1 2 3 4 5
polymer surface against its own transfer film. In add- Filler Wt%
ition, the reduction in friction coefficient occurred due
to formation of the transfer film. The differences of Figure 13. Specific wear rate (w) of samples as a function of
formed transfer film type and thickness may lead to a SiC wt%.
change in wear rate and friction coefficient and this
depends on the nanofiller type and content.28
The other important observation in Table 8 is the important to note that tribological properties of soni-
better tribological effect of SiC nanoparticles compared cated and neat epoxy are identical, indicating no
to the nanowhiskers. This is due to the spherical shape changes in surface and frictional properties of polymer
of nanoparticles and sliding effect on wearing, similar due to ultrasonic treatment. Additionally, by increasing
to a ball bearing that avoids high friction of sample the filler content up to maximum wt%, wear perfor-
against the pin (better transfer film formation). It is mance of matrix improved.19,20
Naeimirad et al. 443

Table 9. Result obtained from two-way ANOVA for tensile modulus.

Effective parameter Square Sum Freedom degree Square mean F p Value F critical

Filler type 538155.1 2 269077.5 54.40481 1.21E-09 5.613591

Filler content 256254.7 3 85418.23 17.27072 3.43E-06 4.718051
Interaction 364130.9 6 60688.48 12.27061 2.65E-06 3.666717
Internal 118700.2 24 4945.841 54.40481
Total 1277241 35

Statistical analysis
and ball bearing. Statistical analysis confirmed the
Using the two-way method of ANOVA on mechanical experimental results.
and tribological results the influence of parameters in
group mean was investigated. A typical statistical ana-
lysis result for tensile modulus is shown in Table 9. Acknowledgements
Statistical analysis indicated that the effect of The authors would like to thanks Iran’s Ministry of Science,
reinforcement wt% and reinforcement types on the mea- Research and Technology (MSRT) for financially supporting
sured properties (Mechanical and Tibological) were this research and Laboratory for Advanced fibers, Swiss
highly significant (p < 0.001). In addition, according to Federal Laboratories for Materials Science and Technology
Table 9, filler type has more effect on tensile modulus (Empa) for SEM images. Special appreciation is also for the
than filler content (demonstrated by smaller P-value). reviewers for their valuable comments and Dr. Matthew
Morrissey for critical edit.

Conclusions
A systematic study was carried out to investigate the References
effect of SiC nanoparticle and nanowhisker fillers on 1. Nhuapeng W, Thamjaree W, Kumfu S, et al.
epoxy matrix properties. A good dispersion of nano- Fabrication and mechanical properties of silicon carbide
SiC fillers was obtained in epoxy resin using ultrasonic nanowires/epoxy resin composites. Curr Appl Phys 2008; 8:
treatment. FT-IR analysis indicated that ultrasonic 295–299.
waves have no degrading effect on the epoxy resin. SiC 2. Hu H, Onyebueke L and Abatan A. Characterizing and
nanoparticles up to 1 wt% caused about 17% increase in modeling mechanical properties of nanocomposites-review
and evaluation. J Mineral Mater Charact Eng 2010; 9: 275.
modulus and 27% in strength. The maximum improve-
3. Abenojar J, Del Real J, Martinez M, et al. Effect of silane
ment was obtained by 2 wt% nanowhisker samples with
treatment on SiC particles used as reinforcement in epoxy
a 35% increase in tensile and flexural modulus and ten- resins. J Adhes 2009; 85: 287–301.
sile strength and 43% in flexural strength. Increasing 4. Rodgers RM, Mahfuz H, Rangari VK, et al. Infusion of
filler content caused mechanical properties to decrease, SiC nanoparticles into SC-15 epoxy: an investigation of
which may have been caused by agglomeration. Due to thermal and mechanical response. Macromol Mater Eng
the needle shape (high aspect ratio) of nanowhiskers, it 2005; 290: 423–429.
can be concluded that nanowhiskers reinforced compos- 5. Jha AK, Mantry S, Satapathy A, et al. Erosive wear per-
ites have better mechanical properties compared to formance analysis of jute-epoxy-SiC hybrid composites.
nanoparticle composites. Also there is a good correlation J Compos Mater 2010; 44: 1623–41.
between experimental results and the results obtained 6. Ji Q, Zhang M, Rong M, et al. Friction and wear of epoxy
from established theoretical models. Hardness tests indi- composites containing surface modified SiC nanoparticles.
cated a limited improvement (7%) by adding SiC nano- Tribol Lett 2005; 20: 115–123.
7. Zhou T, Wang X, Liu X, et al. Improved thermal conduct-
particles and nanowhiskers. Tribological properties
ivity of epoxy composites using a hybrid multi-walled
improved by 31% decrease in friction coefficient (l)
carbon nanotube/micro-SiC filler. Carbon 2010; 48:
for 4 wt% nanowhisker samples and 50% decrease for
1171–1176.
4 wt% nanoparticle samples. Also, it was shown that 8. Bittmann B, Haupert F and Schlarb AK. Ultrasonic dis-
wear properties improved by the addition of SiC fillers: persion of inorganic nanoparticles in epoxy resin.
wear rate (w) decreased by 33% in nanowhisker samples Ultrasonics sonochemistry 2009; 16: 622–628.
and 60% in nanoparticle samples. SiC nanoparticles 9. Jiang Y, Tohgo K and Shimamura Y. Ultrasonic disper-
have a better properties effect compared to SiC nano- sion of SiO2 particles in glassy epoxy resin. J Compos
whiskers due to the effect of spherical shape on sliding Mater 2012; 46: 1159–1168.
444 Journal of Composite Materials 50(4)

10. Chisholm N, Mahfuz H, Rangari VK, et al. Fabrication Infrared Spectroscopy-Materials Science, Engineering
and mechanical characterization of carbon/SiC-epoxy and Technology. 2012.
nanocomposites. Compos Struct 2005; 67: 115–124. 28. Sinha SK and Briscoe BJ. Polymer tribology. World
11. Wang Y, Zhang L and Cheng L. Effect of yarn sizes on Scientific, London: Imperial College Press, 2009.
the tensile damage evolution of a C/SiC composite fabri- 29. Kavitha N, Balasubramanian M and Kennedy AX.
cated by chemical vapor infiltration. Appl Compos Mater Investigation of impact behavior of epoxy reinforced
2011; 18: 165–174. with nanometer-and micrometer-sized silicon carbide
12. Dhage S, Lee H-C, Hassan MS, et al. Formation of SiC particles. J Compos Mater 2013; 47: 1877–1884.
nanowhiskers by carbothermic reduction of silica with 30. Fu S-Y, Feng X-Q, Lauke B, et al. Effects of particle size,
activated carbon. Mater Lett 2009; 63: 174–176. particle/matrix interface adhesion and particle loading on
13. Petrovic J, Milewski J, Rohr D, et al. Tensile mechanical mechanical properties of particulate–polymer compos-
properties of SiC whiskers. J Mater Sci 1985; 20: ites. Compos Part B Eng 2008; 39: 933–961.
1167–1177. 31. Alamri H and Low IM. Characterization of epoxy hybrid
14. Schwartz MM. Composite materials handbook. New composites filled with cellulose fibers and nano-SiC.
York, NY: McGraw-Hill, 1984. J Appl Polym Sci 2012; 126: E222–E32.
15. Wadsworth I and Stevens R. The influence of whisker 32. Sisakht Mohsen R, Saied NK, Ali Z, et al. Theoretical
dimensions on the mechanical properties of cordierite/SiC and experimental determination of tensile properties of
whisker composites. J Eur Ceramic Soc 1992; 9: 153–163. nanosized and micron-sized CaCO3/PA66 composites.
16. Yuan Z, Yu J, Rao B, et al. Enhanced thermal properties Polym Compos 2009; 30: 274–280.
of epoxy composites by using hyperbranched aromatic 33. Tandon G and Weng G. A theory of particle-reinforced
polyamide grafted silicon carbide whiskers. Macromol plasticity. J Appl Mech 1988; 55: 126–135.
Res 2014; 22: 405–411. 34. Abenojar J, Martı́nez M, Pantoja M, et al. Epoxy com-
17. Zhang Y, Pickles C and Cameron J. The production and posite reinforced with nano and micro SiC particles:
mechanical properties of silicon carbide and alumina curing kinetics and mechanical properties. J Adhes
whisker-reinforced epoxy composites. J Reinf Plastics 2012; 88: 418–434.
Compos 1992; 11: 1176–1186. 35. Murasawa G, Tohgo K and Ishii H. The effect of fiber
18. Gu J, Zhang Q, Dang J, et al. Preparation and mechan- volume fraction and aspect ratio on the creation of internal
ical properties researches of silane coupling reagent mod- stress in the matrix and deformation for short-fiber shape
ified b-silicon carbide filled epoxy composites. Polym Bull memory alloy composite. Smart Mater Struct 2006; 15: 33.
2009; 62: 689–697. 36. Rangari VK, Yousuf M and Jeelani S. Influence of SiC/
19. Lim B-S, Ferracane JL, Condon JR and Adey JD. Effect Si3N4 Hybrid Nanoparticles on Polymer Tensile
of filler fraction and filler surface treatment on wear of Properties. Journal of Composites2013. DOI:10.1155/
microfilled composites. Dental Materials. 2002; 18: 1–11. 2013/462914.
20. Sawyer WG, Freudenberg KD, Bhimaraj P and Schadler 37. Mori T and Tanaka K. Average stress in matrix and
LS. A study on the friction and wear behavior of average elastic energy of materials with misfitting inclu-
PTFE filled with alumina nanoparticles. Wear. 2003; sions. Acta metallurgica 1973; 21: 571–574.
254: 573–80. 38. Eshelby JD. The determination of the elastic field of an
21. Luo Y, Zhi Rong M and Qiu Zhang M. Tribological ellipsoidal inclusion, and related problems. Proc Royal
behavior of epoxy composites containing reactive Soc London Series A Math Phys Sci 1957; 241: 376–396.
SiC nanoparticles. J Appl Polym Sci 2007; 104: 39. Tandon G and Weng G. The effect of aspect ratio of
2608–2619. inclusions on the elastic properties of unidirectionally
22. Chisholm N, Mahfuzi H, Rangari V, et al. Synthesis and aligned composites. Polym Compos 1984; 5: 327–333.
mechanical characterization of carbon/epoxy composites
reinforced with SiC nano particles. NSTI-Nanotech 2004;
3: 302–307.
23. Designation A. D 638-03. Standard test method for tensile Appendix
properties of plastics 2003; 1–15.
Theoretical review
24. Standard A. D790-07. Standard test method for flexural
properties of unreinforced and reinforced plastics and elec- Young’s modulus is markedly improved by adding
trical insulation materials. American Society for Testing micro- and nano-particles in a polymer matrix since
and Materials. 1997. hard particles have much higher stiffness values than
25. Standard A. G99, Standard test method for wear testing the matrix.29 For micro- and nano-particulate compos-
with a pin-on-disk apparatus. ASTM International, West
ites strength relies on the effectiveness of stress transfer
Conshohocken, PA. 2006.
26. Levine DM, Stephan DF, Krehbiel TC, et al. Statistics
between the matrix and fillers.30 The interaction
for managers using Microsoft Excel. Upper Saddle Creek, between the nanofillers and matrix is the key to the
NJ: Pearson Prentice Hall, 2005. enhancement of the nanocomposites properties. There
27. González MG, Cabanelas JC and Baselga J. Applications are many factors affecting the interaction, such as
of FTIR on Epoxy Resins -Identification, Monitoring the the filler volume (weight) fraction, degree of dispersion,
Curing Process, Phase Separation and Water Uptake. the filler geometry and orientation etc.2 Studying the
Naeimirad et al. 445

interfaces of nanocomposites is still at the edge of sci- For aligned fiber-reinforced composite materials,
ence and so far no unique mechanism was Halpin and Tsai developed the equations for prediction
developed.31,32 of elastic constants. This model is a semi-empirical
There are many approaches for particulate compos- model, and the longitudinal and transverse moduli are
ite’s mechanical properties estimation such as Einstein, given by
Mixture low, Halpin-Tsai, Nicolais, etc.30,33 The aim of
the most micromechanics approaches is to determine 1 þ 2L
the elastic modulus or stiffness of a composite material EL ¼ Em ð7Þ
1  L
in terms of the elastic properties of the constituent
materials.
1 þ 2T
In the mechanics of materials approach some sim- ET ¼ Em ð8Þ
plifying assumptions are made, the most significant of 1  T
which is that the strain in the matrix is equal to the
strain in the filler. With this assumption the most sim- where a is the fiber aspect ratio, f the volume fraction
plistic of all methods of predicting the modulus of a of reinforcement (vf), L and T take the following
composite, known as the rule of mixtures, is shown expressions:
by Eq. (1).30–32,34–36
Ef  Em
Eupper ¼ Em vm þ Ep vp ð1Þ L ¼ ð9Þ
Ef þ 2Em

where v is the volume fraction, E is the young’s modu-

lus, subscript m denotes the matrix of continuous phase Ef  Em
T ¼ ð10Þ
and subscript p denotes the particulate or filler phase. Ef þ 2Em
In general, the rule of mixtures is regarded as the upper
limit of the elastic modulus or Voight upper bound.2 Halpin-Tsai (H-T) model takes the consideration of the
Reuss lower model can be obtained assuming equal fiber geometry and has five independent parameters.2
stress in the matrix and particles: For fibrous-shaped reinforcements that have an
aspect ratio (L/D > 1) in addition to the rule of mixture
Em Ep and Halpin-Tsai, there are well-known models like the
Elower ¼   ð2Þ
Em vp þ Ep vm Cox model, Mori-Tanaka method and some other simi-
lar ones. The Cox model is based on shear lag analysis
The modulus of nanoparticle filled composites also and shows the importance of critical length (aspect
can be predicted by Einstein Eq. (3).30 ratio) in determining macro-composite strength.
  The Mori-Tanaka method is based on Eshelby’s
E ¼ Em 1 þ 2:5vp ð3Þ elasticity solution for composites. The key assumption
in this model is that the average strain in the inclusion is
Tensile strength of particulate composites can be related to the average strain in the matrix by a fourth
calculated using Nicolais model, Eq. (4).30 order tensor.37,38
   m 
 ¼ m 1  1:21vp 2=3 ð4Þ  ij þ ~ ij ¼ Cm ~kl
ijkl "kl þ " ð11Þ

where  is composite strength and  m is matrix Where Cm ijkl is the matrix stiffness tensor and 2 is
strength. Also Young and Beaumont provided a rela- strain,  is a uniform stress and ~ is the average
tion between strength and particles distance Ds shown stress. According to Eshelby’s equivalent principle the
by Eq. (5). stress in the inclusion may be written as:

 ¼ m þ S=Ds ð5Þ  
ij þ ~ ij þ ijpt ¼ Ccijkl "okl þ "~kl þ "pt
kl
 m 
where S is constant and Ds is defined by Eq. (6). ¼ Cmijkl "kl þ " ~kl þ "pt m
kl  "kl ð12Þ
 
Ds ¼ 2dp 1  vp =3vp ð6Þ
Ccijkl is the stiffness tensor of composite,  pt is stress
where dp is the particle diameter, S depends on component in the inclusion from the average stress
interface adhesion and varies from 9.8  104 to ð ij þ ~ij Þ in the matrix and 2pt is corresponding per-
1.74  103 MPa m.30 turbed strain. Also 2o is denoted the equivalent
446 Journal of Composite Materials 50(4)

transformation strain of the inclusions. Eshelby’s solu- theoretical predictions agree well with experimental
tion renders the relation: results. The H-S model presents the Young’s modulus
as follows
"pt m
kl ¼ Sklmn "mn ð13Þ EL ¼ Em ½1  =1 ð15Þ

 1
 1 3
where Sijkr is Eshelby’s transformation tensor; its com- ET ¼ Em 1  þ ð16Þ
4  þ
ponents depend on the shape (or in the present case to
aspect ratio) of the inclusions and the elastic moduli of where
the matrix.


Tandon and Weng39 presented another model by the Em ð1  gÞ2  g=2
combination of Eshelby’s solution38 and Mori-  ¼þ þ 3ð1  Þ ð17Þ
Ef  Em 2  1
Tanaka’s average stress37 (Eq. (11), (12), (13)) that
shows the effect of aspect ratio in fiber or disc shape
2 
3ð þ 0:25Þ g  22
reinforced composites  ¼ ð1  Þ ð18Þ
2  1
   8 h pffiffiffiffiffiffiffiffiffiffiffiffiffi i
Ccijkl  Cm m m m m m
ijkl 2kl þð1  ÞSklmn "mn þ  2kl þ Cijkl 2kl ¼ 0 < 
 2  1  cos h1  1
ð2 1Þ
g¼ h pffiffiffiffiffiffiffiffiffiffiffiffiffi i ð19Þ
ð14Þ : 
 1  2 þ cos1  1
2 3=2
ð1 Þ
39
Based on the results of Tandon and Weng, Hui
and Shia (H-S)2 derived simplified formulas for pre- where a is the aspect ratio of the filler, defined as the
dicting the overall moduli of composites with aligned ratio of the filler’s longitudinal (with Young’s modulus
reinforcements with an emphasis on fiber-like and EL) length to its transverse (with Young s modulus ET)
flake-like reinforcements, and found that their length.2