Composites: Part B 41 (2010) 388–395

Contents lists available at ScienceDirect

Composites: Part B
journal homepage: www.elsevier.com/locate/compositesb

Impact characterization of glass/epoxy composite plates: An experimental

and numerical study
Ramazan Karakuzu a, Emre Erbil a, Mehmet Aktas b,*
a
Dokuz Eylül University, Department of Mechanical Engineering, 35100 Izmir, Turkey
b
Usak University, Department of Mechanical Engineering, 64200 Usak, Turkey

a r t i c l e i n f o a b s t r a c t

Article history: This study presents the effects of impact energy, impactor mass and impact velocity on the maximum
Received 3 September 2009 contact force, maximum deflection, contact time, absorbed energy, and overall damage area of glass/
Received in revised form 10 February 2010 epoxy laminated composites, experimentally and numerically. The stacking sequence of the composite
Accepted 10 February 2010
plates was chosen as [0°/30°/60°/90°]S. The impact event was simulated and analyzed by using 3DIM-
Available online 12 February 2010
PACT finite element code. The overall delamination area obtained from experimental study and numer-
ical analyses were also examined. It is seen that the numerical results are in good agreement with
Keywords:
experimental results.results.
A. Polymer-matrix composites (PMCs)
B. Impact behavior
Ó 2010 Elsevier Ltd. All rights reserved.
B. Delamination
C. Numerical analysis
D. Mechanical testing

1. Introduction although the permanent indentation is greater for high than speed
impact. Aktas et al. [4] and Icten et al. [5] have investigated the im-
Nowadays, fiber-reinforced composite materials are widely pact response of unidirectional glass/epoxy laminates in room and
used in various engineering applications including automotive, low temperature conditions by considering energy profile dia-
aviation, civil engineering structures, etc. due to their lower grams and associated load–deflection curves. Results show that
weight, high specific stiffness and strength, and damping charac- the penetration threshold for [0°/90°/+45°/45°]S is smaller than
teristics. Air vehicles may be subject to impact loads by foreign ob- that of [0°/90°/0°/90°]S. It is also noted that testing temperature af-
jects such as a dropped tool during maintenance or debris from fects significantly variation of impact characteristics beyond im-
runways. Bird strikes or hailstones (during flight) can also cause pact energy of 20 J.
impact damages. The impact damage in composite materials may Freitas et al. [6] have carried out a numerical study to examine
not be detected sometimes by visual inspection. Such impact- the failure mechanism in composite specimens subjected to impact
induced damages occur inside the material and grows up after loading. Results show that the numerical evaluation of impact with
the onset of small delaminations. In an impact event, several a linear static finite element analysis is not very accurate, but it
damage types occur in composite materials such as matrix crack- gives a meaningful insight on the major mechanisms of failure.
ing, delamination and fiber breakage. Consequently, the impact Zhang et al. [7] have done a series of finite element analyses by
behavior of the laminated composite materials is an important using ABAQUS commercial software to predict damage initiation
phenomenon to be studied [1]. and propagation in laminated carbon/epoxy composite plates sub-
Liu et al. [2] have carried out an experimental study to obtain jected to low-velocity impact. Olsson et al. [8] have predicted the
the perforation threshold of laminated composite plates having delamination threshold load for small mass/high velocity impact
different thicknesses and bending stiffnesses. Results show that on transversely isotropic plates with different thicknesses by using
thickness is more efficient than bending stiffness on the perfora- LS-DYNA finite element software.
tion threshold. Symons [3] has investigated the characterization Aslan et al. [9] have carried out an experimental and numerical
of high and low speed impact damage in carbon fiber reinforced study to investigate the in-plane dimension effect on the impact
plastics. For the same impact energies, results show that the same response of cross-ply glass/epoxy laminated composite plates un-
delamination areas occur under high and low-velocity impacts der low-velocity impact. Results show that the impact behavior
of composite structures is directly dependent on their in-plane
* Corresponding author. Tel.: +90 276 2634195; fax: +90 276 2634196. dimensions. Tiberkak et al. [10] have studied on the damage
E-mail address: [email protected] (M. Aktas). prediction in composite plates subjected to low-velocity impact

1359-8368/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.compositesb.2010.02.003
 R. Karakuzu et al. / Composites: Part B 41 (2010) 388–395 389

by the use of finite element analysis. Results showed that the in-
crease of the 90° plies causes the increase in the contact force
and a reduction in the rigidity of laminate.
Mili and Necib [11] have considered the impact behavior of
glass/epoxy laminated composite plates under low-velocity impact
theoretically and experimentally. Tita et al. [12] have conducted
experimental and numerical studies to examine the stacking se-
quence and impact energy effect on thin carbon/epoxy laminated
composite plates under low-velocity impact.
Li et al. [13] have done an experimental and numerical investi-
gation on low-velocity impact-induced damage of continuous fi-
ber-reinforced composite laminates. Results show that the
maximum contact force decreases, while the maximum deflection
increases with increasing of the plate size. In addition, the influ-
ence of the boundary condition on the maximum contact force,
maximum deflection and delamination size is very small. However,
the impactor mass has a significant effect on the impact behavior
of the composite plates.
In this study, the effects of the equal impact energy (40 J), equal
impactor mass (5 kg) and equal velocity (2 m/s) on the contact
force, deflection, contact time, damage area and absorbed energy
of glass/epoxy laminated composite plates with [0°/30°/60°/90°]S
orientation were examined, experimentally and numerically. The
equal impact energy remained constant by changing the impact
velocity or the impactor mass. Four different impact energies,
10 J, 20 J, 30 J and 40 J, and four impactor masses, 5 kg, 10 kg,
15 kg and 20 kg were chosen. The numerical analysis was done
by using 3DIMPACT finite element code.

2. Production of composite laminates

The glass/epoxy composite material used in this study was pro-

duced in Izoreel Composite Isolate Materials Company. The unidi-
rectional E-glass fabric having a weight of 509 g/m2 was used as
reinforcing material. CY225 epoxy and HY225 hardener were
mixed and applied to the E-glass fabric. Then, it was cured under
0.25 MPa pressure during 2 h at 120 °C. The curing process was
carried out by using a hydraulic press with temperature and time
control. Composite plate consists of eight unidirectional layers Fig. 1. (a) Impact test fixture and (b) specimen clamp position.

with approximately 0.36 mm thickness. At the end of the manufac-

turing process, the final thickness of the composite plate was mea-
22.24 kN was used. An automatic clamping device was also used
sured as about 2.9 mm. The mechanical properties of the
in the environmental chamber of the test machine to clamp the
composite plate are given in Table 1 [14].
specimen before impact. The dimensions of the specimens were
chosen as 100 mm  100 mm. However, the inner diameter of
3. Experimental study the clamping ring is 76.2 mm. The impact specimen and clamping
ring are schematically illustrated in Fig. 1b. Four different masses
The impact tests were performed by using Fractovis Plus test including the impactor mass and the crosshead mass were se-
machine at room temperature. The impact test fixture is given in lected, 5 kg, 10 kg, 15 kg and 20 kg.
Fig. 1a. During the tests, the impactor with a hemispherical nose The stacking sequence of the impact specimen was chosen as [0°/
of 12.7 mm in diameter and a piezoelectric force transducer local- 30°/60°/90°]S. Impact tests were performed for three varied param-
ized in the hemispherical impactor nose with the capacity of eters; equal impact energy, equal impactor mass and equal impact
velocity. Nine major experiments were performed for different
Table 1 impactor masses and impact energies. Each experiment was re-
Mechanical properties of the lamina [15]. peated three times. The values of the impact velocity depending on
Property Value the impactor mass and energy are presented in Table 2.
Density, q (g/cm3) 1.83
Longitudinal modulus, E1 (GPa) 40.51
Table 2
Transverse modulus, E2 (GPa) 13.96
The velocity of the impactor for different impact test cases (m/s).
In-plane shear modulus, G12 (GPa) 3.10
Poison’s ratio, m12 0.22 Impactor mass Impact energy
In-plane shear strength, S12 (GPa) 69.00
10 J 20 J 30 J 40 J
Long. tensile strength, Xt (MPa) 783.30
Trans. tensile strength, Yt (MPa) 64.00 5 kg 2.00 2.83 3.46 4.00
Long. comp. strength, Xc (MPa) 298.00 10 kg – 2.00 – 2.83
Trans. comp. strength, Yc (MPa) 124.00 15 kg – – 2.00 2.31
Interlaminar shear strength, S1 (MPa) 38.00 20 kg – – – 2.00
390 R. Karakuzu et al. / Composites: Part B 41 (2010) 388–395

Software called VisualIMPACT is used to calculate the absorbed

energy, velocity and deflection. The impact force versus time, F(t),
is recorded by data acquisition system (DAS). Data points collected
during a test are up to 16,000 for each channel. Sampling rate is
1 kHz to 2 MHz for each channel. Acceleration of the impactor is
obtained by dividing difference between impact force and total
weight of the impactor (gMtotal) to total weight of the impactor.
Deflection derives from a double integration of acceleration of
the impactor as
Z Z
FðtÞ  gM total 2
di ¼ dt ð1Þ
i Mtotal
where di is deflection of the impactor or the center of the specimen
up to point i, g is the gravity acceleration.
In the impact event, rebounding, penetration, and perforation
Fig. 2. The finite element mesh of the impact specimen with dimensions of
cases can be examined depending on the impact velocity or impac-
76.2 mm  76.2 mm.
tor mass. The absorbed energy is calculated by the total area under
contact force–deflection curves.
The delamination expands depending on the different fiber ori- where rij are the stresses, ekl are the strains, q is the density, ui;tt are
entations between two adjacent plies. At this stage, the visible the accelerations, wi are the arbitrary variational displacements, eij
delamination area (VDA) can be mentioned. VDA is composed of are the strains from the arbitrary variational displacements, X is the
two or three delamination areas, between the last three or four lay- plate volume, C is the surface of the plate, nj is the external unit nor-
ers. The VDA was calculated by using graphics design/editor mal vector on the plate surface, and Eijkl are the material properties
software. of the laminates, which differ from layer to layer due to ply orien-
tation of the laminate.
4. Numerical study For solving the Eq. (2), the contact force distribution between
the impactor and the laminate must be known. The projectile is as-
The 3DIMPACT transient finite element code was used for the sumed as an elastic body with a spherical nose. In a force–deflec-
comparison of the experimental results with the numerical ones. tion history curve, the portion of the curve up to the maximum
Dynamic stress analysis was carried out using an eight-point brick contact force is called loading, and after this point other portion
element. The direct Gauss quadrature integration scheme was also of the curve is named unloading. Therefore, the contact force distri-
used through the element thickness to account for the variation of bution can be determined by the loading and unloading case.
the material properties from layer to layer within the element. The According to the Hertzian contact law in the loading case, the con-
Newmark scheme was adopted to perform time integration step by tact force can be expressed as
step. The Newmark time integration constants such as beta and
gamma are used during step by step analysis. These constants F ¼ ja1:5 ð3Þ
are 0.3025 and 0.6, respectively. The classical Hertzian contact where a and j are the indentation depth and the constant of the
law incorporated with the Newton–Raphson method was applied Hertz contact. In the unloading case, the contact force can be deter-
to calculate the contact force between the impactor and the com- mined as
posite plate. 3DIMPACT code can give the components of stresses  
in each layer, the contact force between impactor and plate, the a  a0
F ¼ Fm ð4Þ
variation of impact velocity during an impact event, the maximum am  a0
deflection of the plate, and also energy and contact time histories.
where Fm is the maximum contact force just before unloading, am is
It can also give the delamination between each of two adjacent
the maximum indentation related to Fm, and a0 is the permanent
plies by means of suitable stress analysis and damage criteria.
indentation during loading/unloading process. The permanent
Therefore, the overall delamination area of the specimen was ob-
indentation can be expressed as
tained from combining projection of the each delamination area.
This code has not the fiber fracture criteria. This is the limitation a0 ¼ 0 when am < acr
of the code. In order to obtain the finite element mesh of the lam- "  2=5 #
a ð5Þ
inated composite plate, the impact specimen is divided into a total a0 ¼ am 1  cr when am P acr
of N  M  Q elements for the length, width, and thickness of the am
plate, respectively (Fig. 2). The dimensions of the impact specimen
where acr is the critical indentation and is approximately 0.1 mm
are 76.2 mm  76.2 mm. Choi and Chang [15] have evaluated the
for glass/epoxy.
effect of mesh size on the stability of the numerical model. They
have proposed that the extensive fine mesh does not require for
failure analysis. Therefore; totally, 576 elements were used in the 4.2. Failure mechanism
numerical model. The clamped boundary conditions are applied
on each side of the composite plate. In the numerical study, two failure criteria as critical matrix
cracking criterion and impact-induced delamination criterion are
4.1. Contact force used.

Based on plane strain condition, the equilibrium equation at an 4.2.1. Matrix cracking criterion
instant time = t in a variational form can be expressed as [16] Based on transversely impact loading, there are only three
Z Z Z stresses components that can contribute to initial matrix cracking
0¼ wi qui;tt dv þ eij Eijkl ekl dv þ wi rij nj dA ð2Þ in the 90° layers in the plane strain condition. These are the
X X C interlaminar shear stress r23 , in-plane tensile stress r22 and
 R. Karakuzu et al. / Composites: Part B 41 (2010) 388–395 391

out-of-plane normal stress r33 . However, the out-of-plane normal

stress is very small in comparison with the other two stresses dur-
ing an entire impact event. Hence, the matrix cracking criterion can
be expressed as (r22 P 0) [17]
n 2 n 2 
r 22 r 23 eM P 0 failure
nY
þ nS
¼ e2M ð6Þ
t i eM < 0 no failure

where the subscripts of 1 and 2 are the local coordinates of the nth
layer parallel and normal to the fiber directions, respectively, and 3
is the out-of-plane direction. n Y t and n Si are the in situ ply trans-
verse tensile strength and interlaminar transverse shear strength
within the nth ply of laminate under consideration, respectively.
r23 and n r 22 are the averaged interlaminar transverse shear stress
n

and the averaged in-plane transverse tensile stress, respectively. Fig. 3. Energy profiling diagram for the experimental results.
For determining any additional matrix failure, the criterion
must be applied to the other layers. If no additional matrix crack-
versus impact energy is plotted for two equal parameters; mass
ing is found during impact, then the impactor’s velocity associated
and velocity. It is clearly seen from Fig. 3 that all the specimens
with the first matrix cracking is referred to as the impact velocity
are of the rebounding case. However, the specimen subjected to
threshold, which is the velocity required to just cause the initial
40 J for equal velocity has reached the penetration threshold. In
impact damage of the laminate.
addition, the energy absorption capability of the specimen sub-
jected to equal mass is lower than the specimen subjected to equal
4.2.2. Impact-induced delamination criterion velocity, for the same impact energy.
Basically two types of critical cracks initiating delamination For a better understanding of the impact behavior of glass/
resulting from impact as shear crack and bending crack. For the epoxy composite plates, only the maximum values of the contact
shear crack, delamination can occur due to the interlaminar longi- force, maximum deflection and contact time are given in Fig. 4. As
tudinal shear stress r13 and the interlaminar transverse shear seen in the figure, the equal mass and the equal velocity have
stress r23 . However, for the bending crack, delamination can occur nearly the same effects on the maximum contact force and max-
due to the interlaminar longitudinal shear stress r13 and the in- imum deflection while the impact energy increases (Fig. 4a and
plane bending stress r22 . By taking both failure mechanisms, b). However, the effect of both on the contact time differs from
delamination could occur only when the following two sequential each other (Fig. 4c). In the range of 10–20 J, the maximum contact
conditions are met: force changes linearly with increasing impact energy for the equal
mass and the equal velocity. Beyond the energy level of 20 J, the
 One of the ply groups intimately above or below the concerned maximum contact force is observed to have a nonlinear behavior
interface has failed due to matrix cracking. for both parameters varied, Fig. 4a. Maximum deflection changes
 The combined stresses governing the delamination growth linearly by increasing impact energy for the plates subjected to
mechanisms through the thicknesses of the upper and lower equal mass and equal velocity (Fig. 4b). It can be seen from
ply groups of the interface reaches a critical value. Fig. 4c that the velocity is the most important parameter for
the contact duration. The equal mass does not affect significantly
Based on the above statement, impact-induced delamination contact time. The effect of equal velocity on the contact time is
criterion for low-velocity impact can be given as: much higher than the equal mass when the impact energy
" 2 nþ1 2 nþ1 2 # increases.
n
r 23 r 13 r 22
Da nS
þ nþ1 Snþ1 Y
þ The contact time for the different impactor mass and impact en-
i i
ergy is given in Table 3. The change of contact time between 5 kg
8
>
> e P 1 failure ð7Þ and 20 kg impactor mass for the same impact energy (40 J) dis-
< D plays the energy transfer speeds. Transfer of the energy depends
eD < 1 no failure
¼ e2D
>
>
nþ1
Y ¼ nþ1 Y t if r
 22 P 0 on the magnitude of the impactor mass. High-mass impacts cause
: nþ1
Y ¼ nþ1 Y c if r22 < 0 larger contact time than low-mass impacts for the same impact
energy.
where Da is a constant which is determined from impact tests. The The variation of the contact force and energy versus the contact
superscripts n and n + 1 correspond to the upper and lower plies of time is illustrated in Fig. 5a and b. The contact force versus deflec-
the nth interface, respectively. n r  23 and nþ1 r
 22 are the averaged
tion is also given in Fig. 5c. It can be seen from the figure that, the
interlaminar and in-plane transverse stresses within the nth and contact force increases by increasing impact energy, while the
(n + 1)th ply, respectively. nþ1 r
 13 is the averaged interlaminar longi-
impactor mass remains constant, 5 kg. The loading portions of
tudinal stress within the (n + 1)th ply. the graphs are similar, but unloading portions differ due to dissim-
ilar damage mechanisms. The loading and unloading portions of
5. Results and discussion the curves are smoother for the samples subjected to 5 kg to 10 J
and 5 kg to 20 J. The damages of these specimens are matrix crack-
In this study, the effects of equal impact energy, equal impactor ing and delamination. The damage consists mostly of matrix dam-
mass and equal impact velocity on the maximum contact force, age and fiber fractures for the samples subjected to higher
maximum deflection, contact time, absorbed energy and overall impactor mass, for instance, 15 kg to 40 J, 20 kg to 40 J and 15 kg
damage area of glass/epoxy laminated composite plates were to 30 J. For the same impact energy levels, the lightweight – the
investigated, experimentally and numerically. fastest impactor (for instance, 5 kg to 40 J) causes earlier damage
The impact energy versus the absorbed energy is plotted for and reaches the peak contact force earlier (Fig. 5a). The higher
equal mass and equal velocity as shown in Fig. 3 by using energy mass causes higher absorbed energy at the same impact energy va-
profiling diagrams [2,4]. That is variation of the absorbed energy lue. Because some of the impact energy is dissipated owing to
392 R. Karakuzu et al. / Composites: Part B 41 (2010) 388–395

40J-5kg 30J-5kg

20J-5kg
40J-15kg

30J-15kg

20J-10kg 40J-20kg
10J-5kg

40J-10kg

a
40J-5kg 40J-10kg 40J-20kg
40J-15kg

30J-5kg 30J-15kg

20J-10kg
20J-5kg

10J-5kg

40J-5kg

40J-15kg

20J-10kg
40J-20kg
40J-10kg
20J-5kg 30J-15kg

10J-5kg
30J-5kg

c
Fig. 5. The diagram of impact history for different energy and impactor mass: (a)
contact force-contact time, (b) energy-contact time and (c) contact force–deflection.

The overall delamination areas of the plate subjected to 5 kg

Fig. 4. The variation of impact energy versus (a) maximum contact force, (b)
and 10 kg impactor masses at the same energy (40 J) are given in
maximum deflection and (c) contact time depending on the equal mass and the
equal velocity. Fig. 6. It can be seen from the figure that the sample subjected to
higher impactor mass has a lower delamination length and width.
On the contrary, the delamination area in the plate subjected to
higher impactor mass is greater than that of the lower impactor
Table 3
Contact time (ms) for different impactor masses and impact energies. mass, for the same impact energy value.
To verify the 3DIMPACT code, the overlapped delamination area
Impactor mass Impact energy
for the experimental and numerical study is illustrated in Fig. 7.
10 J 20 J 30 J 40 J Both of the delamination areas are nearly the same. Therefore, it
5 kg 6.67 6.49 6.81 7.35 seems that the 3DIMPACT code can be used for comparison. Over-
10 kg – 9.40 – 10.35 lapped delamination areas for all experimental and numerical
15 kg – – 11.91 12.76
studies are also given in Table 4. It is seen from the table that
20 kg – – – 17.49
delamination area increases for the equal mass (5 kg), the equal
velocity (2 m/s) and the equal energy (40 J) except for 40 J to
20 kg. High-mass impact causes more delaminations than low-
delamination occurred between different adjacent plies. As a result mass impact at the same impact energy.
of this, the more delamination occurs in the samples subjected to In Fig. 8, the contact force-contact time and deflection-contact
higher impactor mass (Fig. 5b). From the curves in Fig. 5c, it can time diagrams of numerical study are shown for all impact
be observed that rebounding cases occur except for the 20 kg to parameters (except for 40 J). It can be seen from the figure that
40 J case resulting in penetration. the contact force and the deflection are nearly the same for
 R. Karakuzu et al. / Composites: Part B 41 (2010) 388–395 393

Fig. 6. The overall delamination areas of the composite plate subjected to (a) 5 kg and (b) 10 kg impactor mass at 40 J impact energy.

12
5kg-10J
10 5kg-20J

Contact Force [kN]

5kg-30J
8
10kg-20J
6 15kg-30J

0
0 2 4 6 8 10 12
Contact Time [ms]
a
8
5kg-10J
5kg-20J
6
5kg-30J
Deflection [mm]

Fig. 7. Comparison of delamination areas obtained experimentally and numerically

for 2 m/s to 20 J specimen. 10kg-20J
4 15kg-30J

5 kg to 20 J and 10 kg to 20 J whereas the contact time at the

maximum contact force differs from each other. In addition, the 2
contact time of the plate subjected to 10 kg to 20 J is higher than
the plate subjected to 5 kg to 20 J. This means higher impactor
mass at the same energy level causes the formation of a higher 0
0 2 4 6 8 10 12
delamination area.
The delamination areas obtained from numerical and experi- Contact Time [ms]
mental studies are given in Fig. 9 for the equal impactor mass, b
the equal impact velocity and the equal impact energy cases. It
can be seen from the figure that, the delamination area increases Fig. 8. Variation of contact time versus (a) contact force and (b) deflection histories
by increasing energy level (Fig. 9a). In the range of 10–20 J of numerical solutions for various impactor masses.

Table 4
Overlapped delamination areas (mm2) for the experimental and numerical studies.

Impactor mass Impact energy (J)

10 20 30 40
Exp. Num. Exp. Num. Exp. Num. Exp. Num.
5 kg 339.6 347.8 686.5 568.3 1688.9 797.4 2054.5 1034.9
10 kg – – 661.7 771.9 – – 2173.1 1488.7
15 kg – – – – 1351.3 1043.4 2214.7 1959.5
20 kg – – – – – – 1753.1 1391.2
394 R. Karakuzu et al. / Composites: Part B 41 (2010) 388–395

Fig. 9. The delamination areas for experimental and numerical studies for, (a) equal
mass (5 kg) and equal velocity (2 m/s) and (b) equal energy (40 J).

energy, the delamination area obtained from experimental study is

nearly the same for the equal mass and the equal velocity. In this
range, the delamination area obtained from numerical study is also
close to the delamination area obtained from experimental study.
However, after energy level of 20 J, since the 3DIMPACT code does
not include the fiber fracture criterion, the gap between the exper-
imental and numerical results widens rapidly. In addition, the effect
of the equal impactor mass on the delamination area is higher than
the effect of the equal impact velocity (Fig. 9a). Fig. 9b shows
delamination area versus impactor mass at equal impact energy.
By decreasing the impact velocity or increasing the impactor mass
Fig. 10. The contact force–contact time history for equal impact velocity.
up to a critical value (15 kg mass or 2.309 m/s impact velocity) the
delamination area increases. After this point, delamination be-
comes smaller. The main reason for this fact is the fiber fracture
and more matrix failure. Increasing of the impactor mass causes 6. Conclusion
the increase of the indentation at equal impact energy.
To compare the experimental and numerical results, the varia- In this study, the effects of the equal mass, the equal velocity
tion of the contact force-contact time history is given in Figs. 10 and the equal energy on the impact behavior of the glass/epoxy
and 11 for the equal impact velocity and the equal impactor mass, laminated composite plates were investigated, experimentally
respectively. It can be seen from the figures that the numerical and numerically. The following conclusions can be drawn from
curve almost agrees with the experimental one. In addition, the the results obtained:
contact force increases by increasing impact energy for both stud-
ies, while the impactor mass and impact velocity are constant, 5 kg  The energy absorption capability of the specimens subjected to
and 2 m/s, respectively. The maximum contact force is nearly the equal mass is lower than the specimens subjected to equal
same for the same impact energy depending on the impactor mass velocity.
and impact velocity. However, the contact time differs from each  The effects of the equal mass and the equal velocity on the max-
other, for instance, the contact time is 12 ms for 2 m/s to 30 J in imum contact force and maximum deflection are nearly the
the experimental study (Fig. 10c) and is 7 ms for 5 kg to 30 J same while the impact energy increases. However, the effects
(Fig. 11c), approximately. of both on the contact time differ from each other.
 R. Karakuzu et al. / Composites: Part B 41 (2010) 388–395 395

 High-mass impact causes more delaminations and contact time

than low-mass impact at equal impact energy except for the
penetration case. In addition, the higher mass impact causes ear-
lier damage initiation and higher contact forces.
 The delamination area in the sample subjected to a lower impac-
tor mass with higher velocity is lower than the delamination
area in the sample subjected to a higher impactor mass with
lower velocity for the same impact energy.
 The delamination increases up to the critical point of 15 kg
impactor mass for the equal impact energy. After this point,
delamination becomes smaller. The main reason for this fact is
the fiber fracture and more matrix failures.

References

[1] Erbil E. Impact loading in laminated composites. MSc thesis, Dokuz Eylül
University; 2008.
[2] Liu D, Raju BB, Dang X. Impact perforation resistance of laminated and
assembled composite plates. Int J Impact Eng 2000;24(6):733–46.
[3] Symons DD. Characterisation of indention damage in 0/90 lay-up T300/914
CFRP. Compos Sci Technol 2000;60(3):391–401.
_
[4] Aktas M, Atas C, Icten BM, Karakuzu R. An experimental investigation on
impact response of unidirectional composite laminates. Compos Struct
2009;87(4):307–13.
[5] Icten BM, Atas C, Aktas M, Karakuzu R. Low temperature effect on impact
response of quasi-isotropic glass/epoxy laminated plates. Compos Struct
2009;91:318–23.
[6] Freitas M, Silva A, Reis L. Numerical evaluation of failure mechanisms on
composite specimens subjected to impact loading. Composite Part B
2000;31(3):199–207.
[7] Zhang Y, Zhu P, Lai X. Finite element analysis of low-velocity impact damage in
composite laminated plates. Mater Des 2006;27:513–9.
[8] Olsson R, Donadon MV, Falzon BG. Delamination threshold load for dynamic
impact on plates. Int J Solids Struct 2006;43(10):3124–41.
[9] Aslan Z, Karakuzu R, Okutan B. The response of laminated composite
plates under low-velocity impact loading. Compos Struct 2003;59(1):
119–27.
[10] Tiberkak R, Bachene M, Rechak S, Necib B. Damage prediction in
composite plates subjected to low velocity impact. Compos Struct
2008;83(1):73–82.
[11] Mili F, Necib B. Impact behavior of cross-ply laminated composite plates under
low velocities. Compos Struct 2001;51(3):237–44.
[12] Tita V, Carvalho J, Vandepitte D. Failure analysis of low velocity impact on thin
composite laminates: experimental and numerical approaches. Compos Struct
2008;83(4):413–28.
[13] Li CF, Hu N, Cheng JG, Fukunaga H, Sekine H. Low-velocity impact-induced
damage of continuous fiber-reinforced composite laminates. Part II.
Verification and numerical investigation. Composites Part A
2002;33(8):1063–72.
[14] Aktas M, Karakuzu R. Determination of mechanical properties of glass-epoxy
composites in high temperatures. Polym Compos 2009;30:1437–41.
[15] Choi HY, Chang FK. A model for predicting damage in graphite/epoxy
laminated composites from low-velocity point impact. J Compos Mater
1992;26(14):2134–69.
[16] Aktas M. Temperature effect on impact behavior of laminated composite
plates. PhD thesis, Dokuz Eylül University; 2007.
[17] Choi HY. Damage in graphite/epoxy laminated composites due to low velocity
Fig. 11. The contact force–contact time history for equal impactor mass. impact. PhD thesis, Stanford University; 1990.