-
Materials and Design 143 (2018) 81–92
Contents lists available at ScienceDirect
Materials and Design
j ourna l homepage: www.e lsev ie r .com/ locate /matdes
Enhancing mode-I and mode-II fracture toughness of epoxy and
carbonfibre reinforced epoxy composites using multi-walled carbon
nanotubes
Dong Quana, Josu Labarga Urdánizb, Alojz Ivankovića,*a School
of Mechanical and Materials Engineering, University College Dublin,
Irelandb ETSI Caminos, Canales y Puertos, Universidad Politécnica
de Madrid, Madrid, Spain
H I G H L I G H T S
• MWCNTs were used to enhanceepoxy and carbon fibre
composites.
• Mode-I and Mode-II fracturebehaviour was studied.
• The addition of MWCNTs moderatelyincreased Mode-I fracture
toughness.
• The addition of MWCNTs significantlyincreased Mode-II fracture
toughness.
G R A P H I C A L A B S T R A C T
A R T I C L E I N F O
Article history:Received 17 December 2017Received in revised
form 25 January 2018Accepted 25 January 2018Available online 31
January 2018
Keywords:Multi-walled carbon nanotubesEpoxyCarbon fibre
reinforced epoxy compositesFracture toughness in mode-I and
mode-II.
A B S T R A C T
Multi-walled carbon nanotubes (MWCNTs) were added to an epoxy
resin in an effort to improve the fracturetoughness of bulk epoxy
and also when used as matrix for carbon fibre reinforced epoxy
composites (CFRPs).The incorporation of MWCNTs to bulk epoxy and
CFRPs moderately increased the mode-I fracture energy,and
significantly increased the mode-II fracture energy, i.e. the
average mode-II fracture energy of CFRPsincreased from 2026 J/m2 to
3406 J/m2 due to the addition of 0.5 wt% MWCNTs, and further to
5491 J/m2 dueto the addition of 1 wt% MWCNTs. The superior
toughening performance of MWCNTs in mode-II fractureis attributed
to two reasons: 1) increased MWCNT breaking and crack deflection
mechanisms under shearload; and 2) large fracture process zone
accompanied with extensive hackle markings and micro-cracksahead of
the mode-II crack tip of CFRPs, which resulted in significant
number of MWCNTs contributing totoughening mechanisms.
© 2018 Elsevier Ltd. All rights reserved.
1. Introduction
The application of carbon fibre reinforced plastics (CFRPs)
hasexpanded extensively in a wide range of industries, including
auto-motive, aerospace and wind energy. Epoxies are used most
widely
* Corresponding author.E-mail address: [email protected]
(A. Ivanković).
as the matrices for CFRPs due to their favourable engineering
prop-erties, such as high modulus, high strength, low creep and
excel-lent thermal stability. However, the highly cross-linked
structureof epoxies results in inherently low fracture toughness
and hencepoor resistance to fracture. As a consequence, CFRPs
possess rel-ative low interlaminar fracture toughness. Blending
second phasemodifiers, such as silica particles [1,2], rubber
particles [3,4], car-bon nanotubes [5–7] and graphene [8,9], to
epoxies was reported tobe a prevalent method of improving the
fracture toughness. Further
https://doi.org/10.1016/j.matdes.2018.01.0510264-1275/© 2018
Elsevier Ltd. All rights reserved.
https://doi.org/10.1016/j.matdes.2018.01.051http://www.ScienceDirect.com/http://www.elsevier.com/locate/matdeshttp://crossmark.crossref.org/dialog/?doi=10.1016/j.matdes.2018.01.051&domain=pdfmailto:[email protected]://doi.org/10.1016/j.matdes.2018.01.051
-
82 D. Quan et al. / Materials and Design 143 (2018) 81–92
developments of incorporating hybrid rubber-silica
nanoparticlesto epoxy have been made in [10,11] with considerable
success inenhancing the fracture toughness.
The superior mechanical properties of carbon nanotubes
(CNTs)make them attractive candidates as toughening agents of
epoxyresins and fibre reinforced epoxy composites. Tang et al.
[12,13]reported that blending 1 wt% MWCNTs into an epoxy increased
themode-I fracture energy by 56%, while a concurrent improvement
inYoung’s modulus was also observed. Cha et al. [14] presented
thatthe addition of 1 wt% poly(4-aminostyrene) functionalized CNTs
toan epoxy increased the Young’s modulus and tensile strength
from2.76 GPa and 61.51 MPa to 3.89 GPa and 82.57 MPa, respectively.
Inanother study, Cha et al. [15] employed noncovalently
functional-ized carbon nanotubes to enhance the epoxy
nanocomposites. Themost significant improvements of Young’s
modulus, ultimate tensilestrength and fracture toughness were
reported to be 64%, 22% and95–100%, respectively. Saboori et al.
[16] measured mode-I, mode-IIIand mixed-mode fracture toughness of
MWCNT/epoxy nanocom-posites. It was reported that the mode-I
fracture energy increasedsteadily by 27 wt% as the MWCNT content
increased to 0.5 wt%,and then decreased slightly for 1 wt% MWCNTs.
More pronouncedimprovement was achieved for mode-III and mix-mode
fracturetoughness, with no down-ward trend observed as the
MWCNTincreased to 1 wt%. It is noteworthy that adding third phase
mod-ifiers, such as nanoclay [17,18], to CNT modified epoxies
demon-strates a promising method to further enhance the mechanical
andfracture properties. In general, the addition of a small amount
ofMWCNTs could moderately increase the mode-I fracture
toughness.However, limited work [5] has been performed to study the
effectof CNTs on mode-II fracture behaviour of epoxies and the
fracturemechanisms are not yet fully understood.
CNTs were normally introduced into fibre reinforced epoxy
com-posites (FRPs) either by adding CNT/fibre interleaves between
thelaminates [19–21], or by grafting/growing CNTs directly on the
car-bon fibres [22–24]. Xu et al. [19] employed continuous carbon
nan-otube film as interleave to enhance CFRPs. The flexural
strength andinterlaminar shear strength were increased by 16% and
25%, respec-tively, for adding 0.22 wt% CNT film. Zheng et al. [20]
fabricated sand-wiched carbon nanotube/polysulfone nanofiber
(CNTs/PSF) papersas interleaves to improve the interlaminar
fracture toughness ofCFRPs. It was reported that adding 10% CNT/PSF
interleaves toCFRPs increased the mode-I and mode-II fracture
toughness by 53%and 34%, respectively. Additionally, the flexural
strength and flex-ural modulus were improved by 27% and 29%,
respectively. Zhouet al. [21] evaluated the use of hierarchical
carbon nanotube-shortcarbon fibre (CNT-SCF) as interleaves on the
interlaminar fracturetoughness of CFRPs. Increases of 125% and 98%
of the fracture tough-ness, compared to the control material, were
achieved by adding1.0 and 2.0 mg/cm2 CNT-SCF interleaves. Davis and
Whelan [22]managed to deposit fluorine functionalized CNTs on the
mid-planeof fibre reinforced epoxy composites. It is found that
depositing0.5 wt% CNTs increased the mode-II fracture propagation
energyfrom 1906 J/m2 of the control to 2419 J/m2. In another work
[23],CNTs were grown in-situ on carbon fibres using a flame
synthesismethod. The interfacial shear strength of the CFRPs was
increased byapprox. 70% after the CNTs were grown for only 3 min.
Based on theliterature review, it is clear that the integration of
a small amountof CNTs in the mid-plane of FRPs could significantly
increase thefracture toughness. However, limited work employed CNT
modifiedepoxies as matrices of CFRPs to study the fracture
behaviour of suchcomposite laminates. Also, there is a lack of
studies on the mode-Iand mode-II fracture mechanisms of CFRPs based
on CNT modifiedepoxy.
In the present work, MWCNTs were blended in an epoxy resinto
enhance the fracture toughness. Such MWCNT modified epoxieswere
then used as matrices to manufacture CFRPs. The effects of
MWCNTs on the mechanical and fracture properties of bulk
epoxyand CFRPs were investigated.
2. Experimental
2.1. Materials
The epoxy (CYCOM 890 RTM) is a commercially available, one-part
liquid resin system supplied by Cytec Solvay Group. This
resinsystem possesses high viscosity at room temperature. However,
itcould achieve a low viscosity of 250 cps and stay below 350 cps
for24 h at 80 ◦C. The MWCNTs were obtained in powder form
fromGraphene Supermarket, USA. They have an average outer
diameterof 50–85 nm and a length of 10–15 lm. They appear in an
entan-gled cotton-like form. These MWCNTs were not functionalized
andwere used as-received. The carbon fibres are biaxial non-crimp
fabric(Toray T700Sc 50C), provided by Saertex GmbH, Germany.
2.2. Preparation of MWCNT modified epoxies
High shear mixing process was used to disperse the MWCNTs inthe
epoxy. This process has been reported to be able to achieve
goodMWCNT dispersion [7,12,25]. The MWCNT/epoxy mixture was
firstlypre-mixed using an IKA RW20 digital mixer operating at 2000
rpmfor 2 h at 80 ◦C. The low viscosity of the mixture at 80 ◦C
allowed aneasier processing. Then, a shear mixer (Silverson L4RT)
was used forhigh-shear mixing at 3000 rpm initially at 80 ◦C with
the tempera-ture slowly reduced to 30 ◦C over 2 h. The mixture was
subsequentlyshear mixed for another 2 h at approx. 30 ◦C. In this
process, a vis-cous system at relative low temperature could
generate sufficientshear force to effectively break up MWCNT
agglomerates. Followingthe mixing, the mixture was degassed in a
vacuum oven at 80 ◦C. Theconcentrations of MWCNTs were 0.5 wt% and
1 wt%.
A portion of the mixture was then cast into aluminium mouldsto
manufacture the specimens for evaluating the mechanical andfracture
properties of the bulk epoxy. The cure cycle consists of a 2-hour
ramp from room temperature to 180 ◦C followed by a 2-hourhold, or
dwell, at 180 ◦C. After the curing schedule, the samples
wereallowed to cool down naturally to room temperature in the oven.
Theremaining mixture was used as matrices to manufacture the
CFRPs.
2.3. Fabrication of CFRPs
Due to the high aspect ratio (up to several thousands) of
theMWCNTs, resin transfer moulding process is not suitable for
manu-facturing the MWCNT enhanced CFRPs, i.e. the MWCNTs in the
epoxyresin are not able to flow through multiple layers of fibres
uniformlyand hence results in poor overall nanotube dispersion
[26]. In thecurrent work, to tackle this problem, the carbon fibre
fabrics wereimpregnated by the MWCNT modified epoxy manually.
Detailed pro-cedure for manufacturing the CFRP specimens is
described in thefollowing and schematically shown in Fig. 1.
(a, b) The epoxy resin and one ply of carbon fibre fabric with
aweight ratio of 3/2 were placed into a preheated oven at80 ◦C for
approx. 30 min. The low viscosity of the epoxyat 80 ◦C and high
temperature of the carbon fibre fabricallowed for easier processing
in the next step.
(c, d) The epoxy resin was poured onto the carbon fibre fabric
andthen spread evenly on the fabric using a plastic scraper.
(e, f) The carbon fibre fabric was subsequently placed into
thepreheated oven at 80 ◦C for additional 30 min. This allowedthe
resin to impregnate the carbon fibre fabric.
(g) Eight plies of carbon fibre fabrics with fibre direction
ofeither 0 ◦/90 ◦ or ±45 ◦ were prepared by repeating the
pro-cedure (a–f). They were then laid-up in an order as shown
-
D. Quan et al. / Materials and Design 143 (2018) 81–92 83
80 C
Epoxy Carbon fibreOven
80 C
Oven
a b c d
ef
0 /90
± 45
± 45
0 /90
PTFE film
0 /90
± 45
± 45
0 /90
g
h
i
jk
Bottom Plate Release PlyApplied VacuumSealant Tape
PrepregBagging FilmBreathing Fabric
Bottom Plate Release PlyApplied VacuumSealant Tape
Prepreg Bagging film
Compressed Air at 80 psi
Applied 30 Tonne of Force (from hydraulic press)
Top Lid
Breathing Fabric
Load blocks
PTFE insert as initial crack
Fig. 1. Schematic of the fabrication of CFRP samples.
in Fig. 1 (g). A strip of PTFE release film with a thickness
of12.5 lm was inserted between the fourth and fifth plies andserved
as the initial crack.
(h) A debulking process (under vacuum for 45 min) was
thenapplied to the prepreg layup to remove air pockets and
toconsolidate the layup. An illustration of the layup proce-dure is
shown in Fig. 1 (h).
(i) The prepreg was then cured inside an in-house pressclave,and
the setup is shown in Fig. 1 (i). In the curing process,an internal
pressure of 80 psi (approx. 5.5 bar) was appliedin the chamber from
a compressed air supply line. A vac-uum was also applied to the
base plate throughout thecure cycle. The cure cycle consists of a
2-hour ramp fromroom temperature to 180 ◦C followed by a 2-hour
hold, ordwell, at 180 ◦C. After the curing schedule, the
pressclavewas allowed to cool down naturally to 80 ◦C for approx. 4
hwhile keeping the layup under full pressure and vacuum.
(j, k) After the pressclave cooling down to room temperature,the
composite panel was taken out and cut into requireddimensions for
subsequent tests. Load blocks were thenadhesively attached to the
end of each specimen wherethe crack initiator was located. The
sides of each specimenwere painted with a thin layer of water-based
correctionfluid (Tipp-Ex). Once the fluid had dried, vertical lines
weredrawn on the side of the specimen for indicating
cracklength.
2.4. Test procedure
Uniaxial tensile test was conducted to measure the Young’s
mod-ulus and tensile strength of the bulk polymer according to BS
ISO
527 Standard [27]. Dumbbell specimens with 25 mm gauge lengthand
5 mm × 4 mm cross-sectional area were machined from a curedplate.
The tests were conducted at a loading rate of 0.5 mm/min atroom
temperature (normally 20±1 ◦C). At least five replicate testswere
conducted for each material.
Single edge notch three-point bend (3PB) test, see Fig. 2 (a),
wasemployed to measure the Mode-I fracture toughness of the
bulkepoxy according to ASTM D5045-99 standard [28]. The sharp
pre-crack was introduced by tapping a liquid nitrogen-chilled razor
bladeinto the bottom of the v-shape notch. The tests were conducted
atroom temperature with a constant displacement rate of 1 mm/min.At
least six replicate tests were conducted for each material.
Asymmetric four-point bend (A4PB) test [5], see Fig. 2 (b),
wasused for determining the mode-II fracture toughness of the
bulkepoxy. The tests were conducted at room temperature with a
con-stant displacement rate of 1 mm/min. At least six replicate
tests wereconducted for each material. A static equilibrium
analysis of the A4PBconfiguration reveals that the shear force Q
and the bending momentM at the crack plane can be written in terms
of the load P as:
Q =P(L1 − L2)
L1 + L2and M = cQ (1)
when the crack tip is directly underneath the load, i.e. c = 0,
thebending moment vanishes and the sample is in pure mode-II
loading.The mode-II fracture toughness, KIIC, can be determined
as:
KIIC =Q
BW1/2f(
aW
)(2)
-
84 D. Quan et al. / Materials and Design 143 (2018) 81–92
48mm
a 6mm
=6mm
P
W=12mm
B=6mm
(a) 3PB test
L1=24mm
W=12mma
B=6mm
=6mm
P
L2=12.5mm
12.5mm 24mm
c
(b) A4PB test
a0=45 mm
b=25 mm
h=6 mm
L=150 mm
P
P
(c) DCB test
a0=65 mm
b=25 mm
h=6 mm
LF=117.5 mm
P
L=170 mm
(d) ELS test
Fig. 2. Illustrations of 3PB, A4PB, DCB and ELS tests. The red
lines in (a) and (b) indicate the sharp precrack. The green lines
in (c) and (d) indicate the crack starter.
where f(a/W) is the geometry function expressed as [5]:
f(
aW
)= 9.763
(a
W
)4− 15.036
(a
W
)3+ 8.667
(a
W
)2
+1.695(
aW
)− 0.037
(a
W
)≤ 0.7 (3)
The fracture energy, GIIC, was calculated using the
relation:
GIIC =K2IIC
E
(1 − m2
)(4)
where E is the Young’s modulus and m is the Poisson’s ratio of
theepoxy.
Three-point bend flexure test was used to determine the
flex-ural modulus and flexure strength of the CFRPs according tothe
standard ISO: 14125:1998 [29]. The sample dimension was80 mm × 20
mm × 5.5 mm. The tests were conducted at room tem-perature with a
constant displacement rate of 2 mm/min. Five sam-ples were tested
for each material.
Mode-I double cantilever beam (DCB) test and mode-II endloaded
split (ELS) test were carried out to measure the crackpropagation
energy of the CFRPs according to the standards ISO:15024:2001 [30]
and ISO: 15114:2014 [31], respectively. The testconfigurations are
shown in Fig. 2 (c) and (d). The tests were con-ducted at room
temperature with a constant displacement rate of2 mm/min. Five
tests were repeated for each material. It should benoted that a 5
mm long sharp precrack was generated by loading thesamples under
opening load.
In order to examine the length of the MWCNTs after the
shearmixing process, a small amount of uncured epoxy/MWCNT
mixturewas added to acetone to dissolve the epoxy matrix. The
solution wasthen placed in a low energy ultrasonic bath for 1 h to
disperse theMWCNTs. After that, a drop of solution containing
remaining MWC-NTs was placed on a piece of tin foil. Scanning
electron microscopeequipped with a field emission gun (SEM, FEI
Quanta 3D) was usedto image the MWCNTs on the tin foil. The length
of the MWCNTs on
the SEM images was measured using a Java-based image process-ing
and analysis program called ImageJ. About 400 measurementswere
performed. The dispersion of MWCNTs in the cured epoxywas studied
using transmission optical microscope (TOM, NikonE80i(Orina)). The
samples were ground and fine polished to thinsections of
approximately 40 lm thickness, according to the tech-nique
described in [32]. The fracture surfaces of the 3PB specimens,A4PB
specimens, DCB specimens and ELS specimens were studiedusing SEM.
The samples were gold sputter coated at a current of30 mA for 15 s
to get a gold layer of approximately 5 nm.
3. Results and discussion
3.1. Length of the MWCNTs
Fig. 3 (a) presents a typical SEM micrograph of the
remainingMWCNTs after removing the epoxy matrix, and the measured
length-distribution of the MWCNTs is shown in Fig. 3 (b). The
length ofthe MWCNTs was measured by dividing the curved MWCNTs into
anumber of segments, as shown in the inset of Fig. 3 (b). It is
found thatthe MWCNTs have been severely damaged from their initial
lengthof 10–15 lm to an average length of 2.2 lm. This was also
observedin literature [33], and is typical for brittle MWCNT fibres
after shearmixing with polymers [33,34].
3.2. Mechanical properties
3.2.1. Bulk epoxyThe Young’s modulus and tensile strength of the
bulk epoxies are
summarised in Table 1. A Young’s modulus of 3.20 GPa was
mea-sured for the control. The addition of 0.5 wt% MWCNTs resulted
in amoderate increase of the Young’s modulus to 3.42 GPa. This
indicateseffective load transfer between the MWCNTs and the epoxy
at theelastic deformation stage of the tensile test [35]. The
Young’s modu-lus did not increase further for blending more MWCNTs
(1 wt%) intothe epoxy. This is in agreement with the literature
[7,35,36], and wasattributed to the increasing amount of MWCNT
agglomerates. An
-
D. Quan et al. / Materials and Design 143 (2018) 81–92 85
5 µm
(a)
0
6
12
18
24
Fre
quency (
%)
Length of MWCNTs (µm)
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 7.00
(b)
Fig. 3. (a) SEM micrograph of MWCNTs collected after removing
epoxy matrix, and (b) the length-distribution of MWCNTs and the
inset is a schematic showing the measurementof a curved MWCNT.
indication of the dispersion characteristics of the MWCNTs is
givenin Fig. 4. It presents TOM images of the MWCNTs dispersed in
thecured epoxy. It is found that relatively good MWCNT dispersion
wasachieved for the epoxy with 0.5 wt% MWCNTs, while
considerableMWCNT agglomeration can be observed for the epoxy with
1 wt%MWCNTs.
The tensile strength of the control was measured to be 72.4
MPa,see Table 1. The incorporation of MWCNTs to the epoxy had
negligi-ble effects on the tensile strength. While this behaviour
is in agree-ment with the observation in the literature [7,36],
others [14,15]presented a different trend, i.e. blending a small
amount of MWC-NTs to epoxies considerably increased the tensile
strength. This isdue to the varying interfacial adhesion between
MWCNTs and epox-ies in different systems, i.e. effective load
transfer between epoxymatrix and MWCNTs is required to active
reinforcement potentialof MWCNTs. In the current work, the
interfacial adhesion is suffi-cient to provide effective load
transfer at the elastic deformationstage under loading, and hence
an increase in the Young’s modu-lus was observed for 0.5%MWCNTs.
However, the retention of thetensile strength when adding MWCNTs to
epoxy indicates that theinterfacial adhesion is not strong enough
to maintain effective loadtransfer at the failure stage of the
tensile test. Surface modification ofMWCNTs demonstrates a
promising way to achieve good CNT/epoxyinterfacial adhesion
[14,37,38], and hence, to achieve effective rein-forcement in both
tensile modulus and strength. It is important tonotify that the
mechanical properties of CNT modified epoxies alsodepend on the
dispersion of the CNTs.
3.2.2. CFRPsThe flexural properties of the CFRPs are given in
Table 2. The
flexural modulus and flexural strength of the control were
mea-sured to be 48.6 GPa and 505.7 MPa, respectively.
Interestingly, theaddition of a small amount of MWCNTs enhanced the
mechanicalproperties of CFRPs more efficiently than that of the
bulk epoxy.The flexural modulus and flexural strength of the CFRPs
increased to54.6 GPa and 520.1 MPa, respectively, due to the
addition of 0.5 wt%MWCNTs, and further to 57.6 GPa and 543.0 MPa,
respectively, dueto the addition of 1 wt% MWCNTs. Similar trend was
also reported
Table 1Mechanical properties of the epoxy.
Sample Young’s modulus (GPa) Tensile strength (MPa)
Control 3.20 ± 0.07 72.4 ± 1.60.5 wt% MWCNTs 3.42 ± 0.11 71.2 ±
1.51 wt% MWCNTs 3.37 ± 0.06 72.1 ± 2.1
by Zheng et al. [20], and the favourable effect on the flexural
prop-erties of CFRPs can be explained in terms of an interaction
betweenMWCNTs and carbon fibres under bending load.
3.3. Fracture properties
3.3.1. Bulk epoxyFig. 5 shows the measured mode-I and mode-II
fracture energy
of the bulk epoxies. The mode-I fracture energy, GIC, of the
con-trol was measured to be 118.5 J/m2. The value of GIC was
increasedto 136.5 J/m2 due to the addition of 0.5 wt% MWCNTs, and
furtherto 161.1 J/m2 due to the addition of 1 wt% MWCNTs. A more
pro-nounced enhancement was achieved for the mode-II fracture
energy,i.e. GIIC increased from 177.2 J/m2 of the control to 283.6
J/m2 for0.5 wt% MWCNTs, and further to 411.8 J/m2 for 1 wt%
MWCNTs.Extensive research, such as [7,35,39,40], has reported
similar trendon the mode-I fracture toughness of MWCNT reinforced
polymers,i.e. the addition of a small amount of MWCNTs moderately
increasedthe mode-I fracture toughness. However, to our best
knowledge, verylimited study has been performed on the mode-II
fracture of MWCNTmodified polymers, to date.
3.3.2. Fractographic studies: bulk epoxyTypical SEM images of
the fracture surfaces of the 3PB specimens
are shown in Fig. 6. The dashed line indicates the front of the
precrackand the arrow indicates the crack growth direction. A very
smoothfracture surface is observed for the control, as shown in
Fig. 6 (a). Thefracture surface of MWCNT modified epoxy appears
slightly rougherthan the control (Fig. 6 (b)). Fig. 6 (c) and (d)
compare the fracturesurfaces of the control and the 1 wt% MWCNT
modified epoxy athigher magnification. While a mirror-like surface
is observed for thecontrol, a large number of MWCNT segments along
with many holes(with approx. the same diameter as the MWCNTs) and
scratchinglines can be seen on the fracture surfaces of the 1 wt%
MWCNT mod-ified epoxy. It is worthy noticing that both long and
short MWCNTsegments are observed on the fracture surface of the 1
wt% MWCNTmodified epoxy, see Fig. 6 (d). Similar structures have
been previ-ously reported in the literature [7,12,13,35] and were
identified tobe formed by two different failure mechanisms:
immediate fractureof the MWCNTs once the crack front passes it
(short segments); andMWCNT pull-out (long segments). Hence, these
MWCNT segmentsare the evidence of nanotube breaking and bridging
(nanotube pull-out) mechanisms of the MWCNTs, and the scratching
lines wereaccompanied with the crack deflection mechanism of the
MWC-NTs [7,35]. All these mechanisms contributed to the increase of
themode-I fracture energy.
-
86 D. Quan et al. / Materials and Design 143 (2018) 81–92
200 µm
(a) 0.5 wt.% MWCNTs
200 µm
(b) 1 wt.% MWCNTs
Fig. 4. Dispersion of the MWCNTs in cured epoxy.
Fig. 7 presents typical SEM images of the fracture surfaces of
theA4PB specimens. A large number of river-lines are observed on
thefracture surfaces of both the control and the 1 wt% MWCNT
modi-fied epoxy (Fig. 7 (a) and (b)). The river-lines are more
dense on thefracture surface of the 1 wt% MWCNT modified epoxy.
This mightbe attributed to the crack deflection of the MWCNTs. Fig.
7 (c) and(d) presents the zoom-in of the positions between the
river-lines inFig. 7 (a) and (b). A smooth surface is observed for
the control, asexpected. The fracture surface of the 1 wt% MWCNT
modified epoxyhas a much rougher appearance, identified with
numerous MWCNTsegments and micro-crack marks. Hence, the toughening
mecha-nisms of MWCNTs in mode-II fracture are determined to be
crackbridging, nanotube breaking and crack deflection. These are
similarto the detected toughening mechanisms in mode-I
fracture.
Although similar toughening mechanisms are observed in mode-I
and mode-II fracture, the toughness enhancement of epoxy foradding
MWCNTs was much more pronounced in mode-II than inmode-I (see Fig.
5). For the MWCNT segments on the fracture sur-faces, the short
ones were generated by the breaking of the MWCNTsas the crack front
passes it, while the long ones were formed by thepull-out of MWCNTs
[7,35]. When comparing Figs. 6 (d) and 7(d),it is clear that only a
small proportion of short MWCNT segmentsare observed on the mode-I
fracture surface (all the others are long),while the majority of
the MWCNT segments on the mode-II fracturesurface are very short.
Moreover, a large number of holes generatedby MWCNT pull-out are
observed on Fig. 6 (d), while only limited ofthem present on Fig. 6
(d). Hence, the majority of the MWCNTs werepulled out (crack
bridging mechanisms) during the mode-I fractureprocess, while most
of the MWCNTs fractured (nanotube breakingmechanism) during the
mode-II fracture process. This might be dueto the presence of
higher stress in shear loading of mode-II frac-ture [41]. Given the
superior mechanical properties of MWCNTs, it isnot surprising that
the nanotube breaking mechanism is more effec-tive in preventing
the crack growth and deflecting the crack paththan the crack
bridging mechanism. Hence, the addition of MWCNTsis more effective
for mode-II toughness enhancement.
3.3.3. CFRPsR-curves of the mode-I DCB tests and the mode-II ELS
tests of all
the CFRPs are shown in Fig. 8. No significant R-curve behaviour
isobserved for the DCB tests of the control and the MWCNT
modified
Table 2Flexural properties of the CFRPs.
Sample Flexural modulus (GPa) Flexural strength (MPa)
Control 48.6 ± 3.4 506 ± 160.5 wt% MWCNTs 54.6 ± 1.3 520 ± 201
wt% MWCNTs 57.6 ± 1.5 543 ± 4
CFRPs, i.e. the mode-I fracture energy remains essentially
constantalong the crack length. The average mode-I fracture energy,
GIC, ofthe CFRPs is given in Table 3. The value of GIC was measured
to be427 J/m2 for the control. It was increased to 462 J/m2 due to
the addi-tion of 0.5 wt% MWCNTs, and further to 537 J/m2 due to the
additionof 1 wt% MWCNTs.
As shown in Fig. 8 (b), ‘rising’ R-curve behaviour is
observedfor the ELS tests of all the CFRPs, i.e. the fracture
energy increasedsteadily with the crack length. This can be caused
by an increasingdamage zone ahead of the crack tip [42]. The
average value, GAvgIIC ,minimum value, GMinIIC , and maximum value,
G
MaxIIC , of the R-curves are
presented in Table 3. It is found that the incorporation of
MWCNTsdramatically enhanced the mode-II fracture toughness of the
CFRPs.GAvgIIC increased from 2026 J/m
2 of the control to 3406 J/m2 of the0.5% MWCNTs modified CFRP,
and further to 5491 J/m2 of the 1 wt%MWCNTs modified CFRP.
3.3.4. Fractographic studies: CFRPsFig. 9 presents typical
micrographs of the fracture surfaces of
the DCB specimens. A number of broken fibres are observed on
thefracture surfaces of both the control and the 1 wt% MWCNT
mod-ified CFRPs, as shown in Fig. 9 (a) and (b). All the other
fibres arewell attached to the epoxy matrix. Fig. 9 (c) and (d)
presents micro-graphs of the broken fibres and the surrounding
matrix under highermagnification. It is clear that some epoxy
matrix is still attachedto the fibres. All these observations
demonstrate good interfacialadhesion between the matrix and the
fibres, and the main toughen-ing mechanisms of the carbon fibres in
CFRPs are fibre peeling andfibre breaking. Another observation from
Fig. 9 (a) and (b) is thatthere exist numerous flake-like fracture
features in the interstitial
0
100
200
300
400
500
Control 0.5wt.% MWCNTs 1wt.% MWCNTs
Fra
ctu
re E
nerg
y (
J/m
2) GIIC
GIC
Fig. 5. Fracture energy of the bulk epoxies.
-
D. Quan et al. / Materials and Design 143 (2018) 81–92 87
Precrack
100 µm
(a) Control
Precrack
100 µm
(b) 1 wt.% MWCNTs
5 µm
(c) Control
5 µm
Long MWCNT segments
Short MWCNT segments
(d) 1 wt.% MWCNT
Fig. 6. Typical SEM micrographs of the fracture surfaces of 3PB
specimens for the control and the 1 wt% MWCNTs modified epoxy. The
dashed line indicates the tip of the precrack.The arrow indicates
the crack growth direction.
matrix between the carbon fibres for both the control and the 1
wt%MWCNT modified CFRPs. However, it is more extensive for the 1
wt%MWCNT modified composites, which resulted in a rougher
fracturesurface. The micrographs of the interstitial matrix between
the fibresare shown in Fig. 9 (e) and (f). Apparently, the fracture
surface ofthe 1 wt% MWCNT modified CFRPs appears rougher than that
of thecontrol, and a large number of MWCNT segments (generated by
nan-otube breaking and pull-out) are observed on the fracture
surface of1 wt% MWCNT modified CFRP composite.
Fig. 10 shows typical micrographs of the fracture surfaces of
themode-II ELS specimens as a representative of the unmodified
andMWCNT modified CFRPs. One can see that all the fibres remained
onthe fracture surface of the bottom part of the ELS specimens,
leav-ing corresponding groves on the top part. In the rest of this
work, thefracture surface of the bottom part of the ELS specimens
is referredto as the male surface, and the fracture surface of the
top part isreferred to as the female surface. On the male surfaces,
there is noevidence of fibre breaking, and all the fibres are well
attached tothe matrix. Hence, the main toughening mechanism of the
carbonfibres in mode-II fracture is fibre peeling off in shearing
mode. It isalso found that the female surface appears much rougher
than themale surface for having extensive hackle markings in the
interstitialmatrix between the carbon fibres. The hackle markings
were createdunder the shearing deformation in the fracture process
and they arealigned in a direction of approx. 45 ◦ to the fracture
surface.
Fig. 11 presents SEM images of the male and female fracture
sur-faces of the ELS specimens. By comparing Fig. 11 (a) and (b) to
(c) and(d), it is found that the incorporation of MWCNTs into CFRPs
signifi-cantly increased the roughness of both the male and female
fracturesurfaces. Images of the hackle markings with higher
magnification
are presented in Fig. 11 (e) and (f). Apparently, the surfaces
of thehackle markings are very smooth for the control. This is
typical forbrittle epoxies. Contrarily, there exists extensive
micro cracks in thehackle markings for the 1 wt% MWCNT modified
CFRPs. These microcracks were created by the crack deflection
mechanism due to thepresence of MWCNTs. Moreover, significant
number of MWCNT seg-ments are observed on the fracture surfaces of
the 1 wt% MWCNTmodified CFRPs, see Fig. 11 (f).
Based on the observations, the toughening mechanisms of
theMWCNTs in CFRPs are suggested to be crack bridging,
nanotubebreaking and crack deflection in both mode-I and mode-II
fracture.This is identical to the toughening mechanisms observed in
the bulkepoxy. However, the toughness enhancement of the CFRPs due
to theaddition of MWCNTs was significantly more pronounced in
mode-IIthan in mode-I, i.e. the incorporation of 1 wt% MWCNTs
increased themode-II fracture energy by 171%, but only increased
the mode-I frac-ture energy by 26%. There are two reasons for the
different toughen-ing performance of MWCNTs in mode-I and mode-II
fracture. Firstly,the MWCNTs are more effective to improve the
fracture toughnessof the epoxy matrix under shear loading than
under opening load-ing condition, as discussed in Section 3.3.2.
Secondly, the length ofthe fracture process zone for mode-II
fracture is much longer thanthat for mode-I fracture. Xie et al.
[43] modelled the length of thefracture process zone of CFRPs in
different fracture mode using acohesive zone model. It was reported
that the length of the mode-IIfracture process zone is about 6
times of the mode-I fracture pro-cess zone. It is worth noticing
that Xie et al. [43] also calculated thelength of the fracture
process zone using various analytical models,and all of them
indicated a much longer fracture process zone inmode-II fracture
than in mode-I fracture. Fig. 12 schematically shows
-
88 D. Quan et al. / Materials and Design 143 (2018) 81–92
100 µm
(a) Control
100 µm
(b) 1 wt.% MWCNTs
5 µm
(c) Control
5 µm
(d) 1 wt.% MWCNTs
Fig. 7. Typical SEM micrographs of the fracture surfaces of A4PB
specimens for the control and the 1 wt% MWCNTs modified epoxy. The
red arrow indicates the crack growthdirection.
0
200
400
600
800
40 60 80 100 120
Fra
ctu
re E
nerg
y (
J/m
2)
Crack Length (mm)
Control
0.5wt.% MWCNT
1wt.% MWCNT
(a) Mode-I DCB test
0
2500
5000
7500
10000
65 70 75 80 85
Fra
ctu
re E
nerg
y (
J/m
2)
Crack Length (mm)
Control
0.5wt.% MWCNT
1wt.% MWCNT
(b) Mode-II ELS test
Fig. 8. R-curves for the fracture tests of CFRPs.
the fracture process zone ahead of the crack tip in the CFRPs
undermode-I and mode-II fracture condition. The longer fracture
processzone accompanied with extensive hackle markings and
micro-crackscould include significantly more MWCNTs to introduce
the tough-ening mechanisms. Hence, the incorporation of MWCNTs into
CFRPsdramatically increased the mode-II fracture toughness, but
onlyresulted in moderate increase in the mode-I fracture
toughness.
Previous attempts to improve the interlaminar fracture
tough-ness of CFRPs using CNTs have shown a variety of success.
Zhenget al. [20] et al. used carbon nanotubes/polysulfone
nanofiber(CNTs/PSF) paper as interlayer to enhance the interlaminar
fracture
toughness of CFRPs and achieved 53% and 34% increases of
themode-I and mode-II fracture toughness, respectively. Zhu et al.
[44]reported 52–95 % improvement in the mode-I fracture toughness
and
Table 3Fracture energy of the CFRP composite laminates.
Sample GIC (J/m2) GAvgIIC (J/m
2) GMinIIC (J/m2) GMaxIIC (J/m
2)
Control 428 ± 27 2026 ± 180 903 ± 312 3106 ± 2930.5 wt% MWCNTs
462 ± 25 3406 ± 334 2231 ± 528 4333 ± 2651 wt% MWCNTs 537 ± 34 5491
± 373 4213 ± 260 6315 ± 446
-
D. Quan et al. / Materials and Design 143 (2018) 81–92 89
100 µm
Broken fibres
(a) Control
100 µm
Broken fibres
(b) 1 wt.%MWCNT
5 µm
Carbon Fibre
(c) Control
5 µm
Carbon Fibre
(d) 1 wt.% MWCNTs
5 µm
(e) Control
5 µm
(f) 1 wt.% MWCNTs
Fig. 9. Typical SEM micrographs of the fracture surfaces of the
mode-I DCB specimens of the control and the 1 wt% MWCNTs modified
CFRPs.
74-109% improvement in the mode-II fracture toughness of
CFRPsdue to the incorporation of interlayer made of
glycidyloxypropyl-trimethoxysilane and carbon nanotubes. Carbon
fibre/epoxy inter-leaves were used to toughen CFRPs in [45]. It was
found that themode-I and mode-II fracture energy had been improved
by 26% and
47%, respectively. In current work, the mode-II fracture energy
ofCFRPs was increased by 170% for adding 1% MWCNTs. This is
morepronounced than the other studies. This might be due to the
differ-ent methods of incorporating MWCNTs into the CFRPs. This
resultedin different composite structures, as shown in Fig. 13. The
structure
P
200 µm200 µm
Bottom-male surface Top-female surface
Fig. 10. Representative fracture surfaces of the mode-II ELS
tests for the CFRPs. The arrows indicate the crack growth
direction.
-
90 D. Quan et al. / Materials and Design 143 (2018) 81–92
30 µm
(a) Control-Male
30 µm
(b) Control-Female
30 µm
(c) 1 wt.% MWCNTs-Male
30 µm
(d) 1 wt.% MWCNTs-Female
5 µm
(e) Control-Female
5 µm
Micro cracks
(f) 1 wt.% MWCNT-Female
Fig. 11. Typical SEM micrographs of the fracture surfaces of
mode-II ELS specimens for the control and the 1 wt% MWCNTs modified
epoxy.
of the CFRPs in this work might result in more efficient
interac-tion/load transfer between the MWCNTs and the carbon
fibres, andsubsequently benefit the improvement of mechanical and
fractureproperties.
4. Conclusions
The present work studies the effect of adding MWCNTs to anepoxy
resin on the mechanical and fracture properties of bulk epoxyand
when used as the matrix of CFRPs, with an emphasis on thefracture
toughness and toughening mechanisms. A number of con-clusions can
be drawn from the current work.
The addition of 0.5 wt% MWCNTs moderately increased theYoung’s
modulus of the bulk epoxy. No further improvement in theYoung’s
modulus is observed when adding 1 wt% MWCNTs due tothe increasing
MWCNT agglomeration. The incorporation of MWC-NTs shows negligible
effects on the tensile strength. The adhesion
between the MWCNTs and the epoxy is sufficient to create
effec-tive load transfer at the elastic deformation stage
(resulting in theincrease of the Young’s modulus), but insufficient
to maintain effec-tive load transfer at the failure stage in the
tensile test (resulting inthe retention of the tensile
strength).
The flexural modulus and flexural strength of the CFRPs
weresteadily increased from 48.6 GPa and 506 MPa of the control
to57.6 GPa and 543 MPa of the 1 wt% MWCNT modified CFRPs,
respec-tively. This demonstrates effective load transfer between
the epoxy,the MWCNTs and the carbon fibres under bending load.
Blending MWCNTs into bulk epoxy slightly increased the mode-I
fracture energy but significantly increased the mode-II
fractureenergy. The toughening mechanisms of MWCNTs appeared to
becrack bridging (nanotube pull-out), nanotube breaking and
crackdeflection in both mode-I and mode-II fracture. However,
moreintensive nanotube breaking mechanism accompanied with
higherdensity of crack deflection were detected in the mode-II
fracture.This resulted in more pronounced toughness
enhancement.
Crack
fracture process zone
(a) Mode-I
Crack
hackle markings micro-cracks fracture process zone
(b) Mode-II
Fig. 12. Schematics of the form of the fracture process zone in
CFRPs.
-
D. Quan et al. / Materials and Design 143 (2018) 81–92 91
Crack
MWCNTs
Carbon
fibres
(a) MWCNTs blended in matrix
CrackMWCNT
interlayer
(b) MWCNTs incorporated as interlayer
Fig. 13. Schematics of the form of MWCNT reinforced CFRPs in (a)
this work and (b) the literature [20,44,45].
Similarly to the bulk epoxy, the fracture energy of
CFRPsincreased slightly in mode-I fracture but dramatically in
mode-IIfracture. The outstanding toughening performance of MWCNTs
inmode-II fracture was attributed to two factors. Firstly,
intensive nan-otube breaking and crack deflection mechanisms took
place undershear load. Secondly, the large mode-II fracture process
zone (accom-panied with extensive hackle markings and micro-cracks)
of CFRPsresulted in considerable number of MWCNTs contributing to
tough-ening mechanisms.
Acknowledgments
The authors gratefully acknowledge the financial support fromthe
Irish Composites Centre. The carbon fibres and epoxy matrixwere
supplied by Bombardier Aerospace (Belfast).
References
[1] D. Carolan, A. Ivankovic, A.J. Kinloch, S. Sprenger, A.C.
Taylor, Toughening ofepoxy-based hybrid nanocomposites, Polymer 97
(2016) 179–190. https://doi.org/10.1016/j.polymer.2016.05.007.
[2] S. Rahmanian, A. Suraya, B. Roshanravan, R. Othman, A.
Nasser, R. Zahari, E.Zainudin, The influence of multiscale fillers
on the rheological and mechanicalproperties of carbon-nanotube
silica-reinforced epoxy composite, Mater. Des.88 (2015) 227–235.
https://doi.org/10.1016/j.matdes.2015.08.149.
[3] D. Quan, A. Ivankovic, Effect of core-shell rubber (CSR)
nano-particles onmechanical properties and fracture toughness of an
epoxy polymer, Polymer66 (2015) 16–28.
https://doi.org/10.1016/j.polymer.2015.04.002.
[4] D. Quan, N. Murphy, A. Ivankovic, Fracture behaviour of
epoxy adhesive jointsmodified with core-shell rubber nanoparticles,
Eng. Fract. Mech. 182 (2017)566–576.
https://doi.org/10.1016/j.engfracmech.2017.05.032.
[5] M.R. Ayatollahi, S. Shadlou, M.M. Shokrieh, Fracture
toughness ofepoxy/multi-walled carbon nanotube nano-composites
under bending andshear loading conditions, Mater. Des. 32 (4)
(2011) 2115–2124. https://doi.org/10.1016/j.matdes.2010.11.034.
[6] T. Subhani, M. Latif, I. Ahmad, S.A. Rakha, N. Ali, A.A.
Khurram, Mechanicalperformance of epoxy matrix hybrid
nanocomposites containing carbon nan-otubes and nanodiamonds,
Mater. Des. 87 (2015) 436–444.
https://doi.org/10.1016/j.matdes.2015.08.059.
[7] D. Quan, D. Carolan, C. Rouge, N. Murphy, A. Ivankovic,
Carbon nanotubes andcore-shell rubber nanoparticles modified
structural epoxy adhesives, J. Mater.Sci. 52 (8) (2017) 4493–4508.
https://doi.org/10.1007/s10853-016-0695-9.
[8] B. Ahmadi-Moghadam, M. Sharafimasooleh, S. Shadlou, F.
Taheri, Effectof functionalization of graphene nanoplatelets on the
mechanical response ofgraphene/epoxy composites, Mater. Des. 66,
Part A (2015) 142–149.
https://doi.org/10.1016/j.matdes.2014.10.047.
[9] P. Li, Y. Zheng, M. Li, T. Shi, D. Li, A. Zhang, Enhanced
toughness andglass transition temperature of epoxy nanocomposites
filled with solvent-freeliquid-like nanocrystal-functionalized
graphene oxide, Mater. Des. 89 (Supple-ment C) (2016) 653–659.
https://doi.org/10.1016/j.matdes.2015.09.155.
[10] D. Quan, R.A. Pearson, A. Ivankovic, Interaction of
toughening mechanismsin ternary nanocomposites, Polym. Compos.
(2017) in press. https://doi.org/10.1002/pc.24368.
[11] D. Carolan, A. Ivankovic, A.J. Kinloch, S. Sprenger, A.C.
Taylor, Tough-ened carbon fibre-reinforced polymer composites with
nanoparticle-modifiedepoxy matrices, J. Mater. Sci. 52 (3) (2017)
1767–1788. https://doi.org/10.1007/s10853-016-0468-5.
[12] L.-C. Tang, Y.-J. Wan, K. Peng, Y.-B. Pei, L.-B. Wu, L.-M.
Chen, L.-J. Shu,J.-X. Jiang, G.-Q. Lai, Fracture toughness and
electrical conductivity of epoxy
composites filled with carbon nanotubes and spherical particles,
Compos. A:Appl. Sci. Manuf. 45 (Supplement C) (2013) 95–101.
https://doi.org/10.1016/j.compositesa.2012.09.012.
[13] L.-C. Tang, H. Zhang, X.-P. Wu, Z. Zhang, A novel failure
analysis of multi-walled carbon nanotubes in epoxy matrix, Polymer
52 (9) (2011)
2070–2074.https://doi.org/10.1016/j.polymer.2011.03.002.
[14] J. Cha, S. Jin, J.H. Shim, C.S. Park, H.J. Ryu, S.H. Hong,
Functionalization ofcarbon nanotubes for fabrication of CNT/epoxy
nanocomposites, Mater. Des. 95(2016) 1–8.
https://doi.org/10.1016/j.matdes.2016.01.077.
[15] J. Cha, G.H. Jun, J.K. Park, J.C. Kim, H.J. Ryu, S.H. Hong,
Improvement of mod-ulus, strength and fracture toughness of
CNT/epoxy nanocomposites throughthe functionalization of carbon
nanotubes, Compos. Part B Eng. 129 (2017)169–179.
https://doi.org/10.1016/j.compositesb.2017.07.070.
[16] B. Saboori, M.R. Ayatollahi, Experimental fracture study of
MWCNT/epoxynanocomposites under the combined out-of-plane shear and
tensile load-ing, Polym. Test. 59 (2017) 193–202.
https://doi.org/10.1016/j.polymertesting.2017.01.028.
[17] M.N. Salimi, M.T. Merajin, M.K.B. Givi, Enhanced mechanical
proper-ties of multifunctional multiscale glass/carbon/epoxy
composite reinforcedwith carbon nanotubes and simultaneous carbon
nanotubes/nanoclays,J. Compos. Mater. 51 (6) (2017) 745–758.
https://doi.org/10.1177/0021998316655201.
[18] M. Hosur, T.H. Mahdi, M.E. Islam, S. Jeelani, Mechanical
and viscoelasticproperties of epoxy nanocomposites reinforced with
carbon nanotubes, nan-oclay, and binary nanoparticles, J. Reinf.
Plast. Compos. 36 (9) (2017)
667–684.https://doi.org/10.1177/0731684417691365.
[19] H. Xu, X. Tong, Y. Zhang, Q. Li, W. Lu, Mechanical and
electrical propertiesof laminated composites containing continuous
carbon nanotube film inter-leaves, Compos. Sci. Technol. 127 (2016)
113–118. https://doi.org/10.1016/j.compscitech.2016.02.032.
[20] N. Zheng, Y. Huang, H.-Y. Liu, J. Gao, Y.-W. Mai,
Improvement of inter-laminar fracture toughness in carbon
fiber/epoxy composites with carbonnanotubes/polysulfone
interleaves, Compos. Sci. Technol. 140 (2017)
8–15.https://doi.org/10.1016/j.compscitech.2016.12.017.
[21] H. Zhou, X. Du, H.-Y. Liu, H. Zhou, Y. Zhang, Y.-W. Mai,
Delamination tough-ening of carbon fiber/epoxy laminates by
hierarchical carbon nanotube-shortcarbon fiber interleaves, Compos.
Sci. Technol. 140 (2017) 46–53.
https://doi.org/10.1016/j.compscitech.2016.12.018.
[22] D.C. Davis, B.D. Whelan, An experimental study of
interlaminar shear frac-ture toughness of a nanotube reinforced
composite, Compos. Part B Eng. 42 (1)(2011) 105–116.
https://doi.org/10.1016/j.compositesb.2010.06.001.
[23] X. Du, F. Xu, H.-Y. Liu, Y. Miao, W.-G. Guo, Y.-W. Mai,
Improving theelectrical conductivity and interface properties of
carbon fiber/epoxy compos-ites by low temperature flame growth of
carbon nanotubes, RSC Adv. 6 (2016)48896–48904.
https://doi.org/10.1039/C6RA09839H.
[24] M.S. Chaudhry, A. Czekanski, Z.H. Zhu, Characterization of
carbon nanotubeenhanced interlaminar fracture toughness of woven
carbon fiber reinforcedpolymer composites, Int. J. Mech. Sci.
131-132 (2017) 480–489.
https://doi.org/10.1016/j.ijmecsci.2017.06.016.
[25] W. Zhao, M. Fang, F. Wu, H. Wu, L. Wang, G. Chen,
Preparation of graphene byexfoliation of graphite using wet ball
milling, J. Mater. Chem. 20 (2010) 5817–5819.
https://doi.org/10.1039/C0JM01354D.
[26] A.T. Seyhan, M. Tanoglu, K. Schulte, Mode I and mode II
fracture toughness ofE-glass non-crimp fabric/carbon nanotube (CNT)
modified polymer based com-posites, Eng. Fract. Mech. 75 (18)
(2008) 5151–5162.
https://doi.org/10.1016/j.engfracmech.2008.08.003.
[27] BS ISO 527-2:1996, Plastics - Determinatin of Tensile
Properties - Part 2: TestConditions for Moulding and Extrusion
Plastics, British Standard Institute, BSI.1996.
[28] ASTM D5045-99, Standard Test Method for Plane-Strain
Fracture Toughnessand Strain Energy Release Rate of Plastic
Materials, 1999.
[29] ISO 14125:1998, Fibre-reinforced Plastic Composites
Determination of FlexuralProperties, 1998.
[30] ISO 15024:2001, Fibre-reinforced Plastic Composites
Determination of Mode IInterlaminar Fracture Toughness, GIC for
Unidirectionally Reinforced Materials,2001.
https://doi.org/10.1016/j.polymer.2016.05.007https://doi.org/10.1016/j.polymer.2016.05.007https://doi.org/10.1016/j.matdes.2015.08.149https://doi.org/10.1016/j.polymer.2015.04.002https://doi.org/10.1016/j.engfracmech.2017.05.032https://doi.org/10.1016/j.matdes.2010.11.034https://doi.org/10.1016/j.matdes.2010.11.034https://doi.org/10.1016/j.matdes.2015.08.059https://doi.org/10.1016/j.matdes.2015.08.059https://doi.org/10.1007/s10853-016-0695-9https://doi.org/10.1016/j.matdes.2014.10.047https://doi.org/10.1016/j.matdes.2014.10.047https://doi.org/10.1016/j.matdes.2015.09.155https://doi.org/10.1002/pc.24368https://doi.org/10.1002/pc.24368https://doi.org/10.1007/s10853-016-0468-5https://doi.org/10.1007/s10853-016-0468-5https://doi.org/10.1016/j.compositesa.2012.09.012https://doi.org/10.1016/j.compositesa.2012.09.012https://doi.org/10.1016/j.polymer.2011.03.002https://doi.org/10.1016/j.matdes.2016.01.077https://doi.org/10.1016/j.compositesb.2017.07.070https://doi.org/10.1016/j.polymertesting.2017.01.028https://doi.org/10.1016/j.polymertesting.2017.01.028https://doi.org/10.1177/0021998316655201https://doi.org/10.1177/0021998316655201https://doi.org/10.1177/0731684417691365https://doi.org/10.1016/j.compscitech.2016.02.032https://doi.org/10.1016/j.compscitech.2016.02.032https://doi.org/10.1016/j.compscitech.2016.12.017https://doi.org/10.1016/j.compscitech.2016.12.018https://doi.org/10.1016/j.compscitech.2016.12.018https://doi.org/10.1016/j.compositesb.2010.06.001https://doi.org/10.1039/C6RA09839Hhttps://doi.org/10.1016/j.ijmecsci.2017.06.016https://doi.org/10.1016/j.ijmecsci.2017.06.016https://doi.org/10.1039/C0JM01354Dhttps://doi.org/10.1016/j.engfracmech.2008.08.003https://doi.org/10.1016/j.engfracmech.2008.08.003http://refhub.elsevier.com/S0264-1275(18)30065-0/rf0135http://refhub.elsevier.com/S0264-1275(18)30065-0/rf0140http://refhub.elsevier.com/S0264-1275(18)30065-0/rf0145http://refhub.elsevier.com/S0264-1275(18)30065-0/rf0150
-
92 D. Quan et al. / Materials and Design 143 (2018) 81–92
[31] ISO 15114:2014, Fibre-reinforced Plastic Composites
Determination of ModeII Fracture Resistance for Unidirectionally
Reinforced Materials Using the Cal-ibrated End-loaded Split (C-ELS)
Test and an Effective Crack Length Approach,2014.
[32] A.S. Holik, R.P. Kambour, D.G. Fink, S.Y. Hobbs, Grinding
and polishingtechniques for thin sectioning of polymeric materials
for transmission lightmicroscopy, Microstruct. Sci. 7 (1979)
357–367.
[33] S.-Y. Fu, Z.-K. Chen, S. Hong, C.C. Han, The reduction of
carbon nanotube (CNT)length during the manufacture of CNT/polymer
composites and a method tosimultaneously determine the resulting
CNT and interfacial strengths, Carbon47 (14) (2009) 3192–3200.
https://doi.org/10.1016/j.carbon.2009.07.028.
[34] G. Sui, W. Yu, Q. Zhang, F. Chen, Q. Fu, Achieving
excellent dispersionand electrical conductivity of olefin block
copolymer/MWCNTs composites effi-ciently via high-shear processing,
Polymer 123 (2017) 65–72.
https://doi.org/10.1016/j.polymer.2017.07.007.
[35] F.H. Gojny, M.H.G. Wichmann, U.K. pke, B. Fiedler, K.
Schulte, Car-bon nanotube-reinforced epoxy-composites: enhanced
stiffness and fracturetoughness at low nanotube content, Compos.
Sci. Technol. 64 (15) (2004) 2363–2371. developments in carbon
nanotube and nanofibre reinforced
polymers.https://doi.org/10.1016/j.compscitech.2004.04.002.
[36] F.H. Gojny, M.H. Wichmann, B. Fiedler, K. Schulte,
Influence of differentcarbon nanotubes on the mechanical properties
of epoxy matrix composites -a comparative study, Compos. Sci.
Technol. 65 (15-16) (2005)
2300–2313.https://doi.org/10.1016/j.compscitech.2005.04.021.
[37] T. Tsafack, J.M. Alred, K.E. Wise, B. Jensen, E. Siochi,
B.I. Yakobson, Exploringthe interface between single-walled carbon
nanotubes and epoxy resin, Carbon105 (2016) 600–606.
https://doi.org/10.1016/j.carbon.2016.04.066.
[38] S.A. Sydlik, J.-H. Lee, J.J. Walish, E.L. Thomas, T.M.
Swager, Epoxy functional-ized multi-walled carbon nanotubes for
improved adhesives, Carbon 59 (2013)109–120.
https://doi.org/10.1016/j.carbon.2013.02.061.
[39] P.J.F. Harris, Carbon nanotube composites, Int. Mater. Rev.
49 (1) (2004) 31–43.https://doi.org/10.1179/095066004225010505.
[40] J.N. Coleman, U. Khan, W.J. Blau, Y.K. Gun’ko, Small but
strong: a review ofthe mechanical properties of carbon nanotube
polymer composites, Carbon 44(9) (2006) 1624–1652.
https://doi.org/10.1016/j.carbon.2006.02.038.
[41] D. Bhattacharjee, J.F. Knott, Effect of mixed mode I and II
loading on the frac-ture surface of polymethyl methacrylate (PMMA),
Int. J. Fract. 72 (4) (1995)359–381.
https://doi.org/10.1007/BF00040373.
[42] B.R.K. Blackman, A.J. Kinloch, M. Paraschi, The
determination of the modeII adhesive fracture resistance, GIIC, of
structural adhesive joints: an effectivecrack length approach, Eng.
Fract. Mech. 72 (6) (2005) 877–897.
https://doi.org/10.1016/j.engfracmech.2004.08.007.
[43] J. Xie, A.M. Waas, M. Rassaian, Estimating the process zone
length of fracturetests used in characterizing composites, Int. J.
Solids Struct. 100 (2016) 111–126.
https://doi.org/10.1016/j.ijsolstr.2016.07.018.
[44] Y. Zhu, C.E. Bakis, J.H. Adair, Effects of carbon
nanofiller functionalizationand distribution on interlaminar
fracture toughness of multi-scale reinforcedpolymer composites,
Carbon 50 (3) (2012) 1316–1331.
https://doi.org/10.1016/j.carbon.2011.11.001.
[45] M. Arai, Y. Noro, K. ichi Sugimoto, M. Endo, Mode I and
mode II interlaminarfracture toughness of CFRP laminates toughened
by carbon nanofiber inter-layer, Compos. Sci. Technol. 68 (2)
(2008) 516–525.
https://doi.org/10.1016/j.compscitech.2007.06.007.
http://refhub.elsevier.com/S0264-1275(18)30065-0/rf0155http://refhub.elsevier.com/S0264-1275(18)30065-0/rf0160https://doi.org/10.1016/j.carbon.2009.07.028https://doi.org/10.1016/j.polymer.2017.07.007https://doi.org/10.1016/j.polymer.2017.07.007https://doi.org/10.1016/j.compscitech.2004.04.002https://doi.org/10.1016/j.compscitech.2005.04.021https://doi.org/10.1016/j.carbon.2016.04.066https://doi.org/10.1016/j.carbon.2013.02.061https://doi.org/10.1179/095066004225010505https://doi.org/10.1016/j.carbon.2006.02.038https://doi.org/10.1007/BF00040373https://doi.org/10.1016/j.engfracmech.2004.08.007https://doi.org/10.1016/j.engfracmech.2004.08.007https://doi.org/10.1016/j.ijsolstr.2016.07.018https://doi.org/10.1016/j.carbon.2011.11.001https://doi.org/10.1016/j.carbon.2011.11.001https://doi.org/10.1016/j.compscitech.2007.06.007https://doi.org/10.1016/j.compscitech.2007.06.007
Enhancing mode-I and mode-II fracture toughness of epoxy and
carbon fibre reinforced epoxy composites using multi-walled carbon
nanotubes1. Introduction2. Experimental2.1. Materials2.2.
Preparation of MWCNT modified epoxies2.3. Fabrication of CFRPs2.4.
Test procedure
3. Results and discussion3.1. Length of the MWCNTs3.2.
Mechanical properties3.2.1. Bulk epoxy3.2.2. CFRPs
3.3. Fracture properties3.3.1. Bulk epoxy3.3.2. Fractographic
studies: bulk epoxy3.3.3. CFRPs3.3.4. Fractographic studies:
CFRPs
4. ConclusionsAcknowledgmentsReferences