18TH INTERNATIONAL CONFERENCE ON COMPOSITE MATERIALS

IMPACT ENERGY AND DAMAGE BEHAVIOR OF HYBRID

COMPOSITE STRUCTURES UNDER HIGH VELOCITY IMPACT
Sung-Choong Woo1, Jong-Tak Kim2 , Jin-Young Kim3, Tae-Won Kim4*
1
Survivability Technology Defense Research Center, Hanyang University,
17 Haengdang-Dong, Sungdong-Gu, Seoul 133-791, Republic of Korea
2
Department of Automotive Engineering, Hanyang University, 17 Haengdang-Dong,
Sungdong-Gu, Seoul 133-791, Republic of Korea
3
Agency for Defense Development, 462 Jochiwon-Gil, Yuseong-Gu,
Daejeon 305-152, Republic of Korea
4
School of mechanical Engineering, Hanyang University, 17 Haengdang-Dong,
Sungdong-Gu, Seoul 133-791, Republic of Korea

* Corresponding author([email protected])

Keywords: Hybrid composite structure; High velocity impact; Damage behavior; Finite
element analysis

Impact absorption energy together with the material damage of hybrid composite structure under
high velocity impact was investigated. The hybrid composite structure studied in this work
consists of six-layer, namely S2-glass-1, CMC, EPDM rubber, Al7039, Al-foam and S2-glass-2.
A three-dimensional finite element simulation was conducted based on a progressive damage
model using the commercial code program, LS-DYNA. In order to simulate the sufficient
deformations and fractures, an extremely high velocity (5,000 m/s) was applied as impact
loading to the hybrid composite structure. The damage parameter in continuum damage
mechanics determined by the reduction of stiffness, and also the absorbed energy were
calculated to analysis the local fracture of the hybrid composite structure. Results of finite
element analyses revealed that S2-glass showed a wide range of damage and local delamination;
CMC and aluminum foam revealed a narrow band of damage. It is therefore suggested that the
progressive damage model was appropriate to simulate quantitatively the level of damage of
hybrid composite structure under such high velocity impact.

1 Introduction may be found addressing these issues and problems

In a design of composite structures for impact [1-5]. Most of the works to determine the failure
energy absorption, brittle materials such as ceramics characteristics of hybrid composite structures,
are stacked at the front and ductile materials are however, concerned about experimental
arranged at the rear. Ceramic/metal or ceramic/PMC investigation, and hence required a lot of time and
(polymer matrix composite) composition is preferred effort. A popular trend enabling the increase of cost
in the stacking sequence considering material efficiency is to reduce destructive testing schemes
performances, deformation characteristics, and by predicting the performance of materials through
multi-layer manufacturing technologies etc. analytical modeling or numerical simulations. When
Concerning an impact on composite laminates, multi-layered plates including diverse materials are
determination of ballistic limits with the analysis of subjected to ballistic impact, their response is
the penetration process has been widely studied determined by interactions of multiple stress waves
during the last three decades. Numbers of literature generated at the layer interfaces [6]. Many works
have been done to model the failure mechanisms of parameter n was assumed to be identical (n = 0.57)
hybrid composite structures under relatively lower for the four strain softening damage modes.
transverse impact loading [710]. However, limited The failure criterion for isotropic materials such as
studies have been done on the progressive failure of aluminum and ceramic is given by the following
composites under high strain rate impact loading. It relation:
is generally accepted that composites fail in a
progressive manner. 2 2 2

The objective of this work is therefore to understand =f 2 + 23 + 12 = 1 0 (3)
the impact absorption behavior of hybrid composite S 2 S 23 S12
structures consisting of many different materials by
employing the progressive damage model. The where f is the failure function for isotropic materials,
material local damage together with the impact 2 normal axial stress, S 2 failure strength, 23 and
absorption energy was analyzed. In the numerical 12 shear stresses, S 23 and S 12 shear strengths,
simulations, explicit commercial software LS- respectively. Mark < > denotes Macaulay bracket.
DYNA was used. The fiber failure criteria of Hashin for a
2 Numerical Analysis unidirectional layer are generalized to characterize
the fiber damage in terms of strain components for a
2.1 Damage Model plain weave layer. The tensile/shear failure of fill
LS-DYNA provides material model MAT161 and and warp fibers are given by the quadratic
MAT162 (developed by Yen) which capture the interaction between the associated axial and shear
progressive failure mode of composite laminates stresses:
including both unidirectional and plain weave
2
laminates during transverse impact. The material 2 + 2
model MAT162, based on the Hashins failure =
f fill 1 + 12 31= 1 0 (4)
criteria [11] was assigned to model the plain weave S1 S1FS
composite laminate [12].
2
The continuum damage mechanics approach 2 + 2
proposed by Matzenmiller et al. [13] has been =
f warp 2 + 12 23= 1 0 (5)
incorporated into MAT 162. This model enables S 2 S 2 FS
progressive damage of composite laminates to
simulate by controlling strain softening after failure where S 1 and S 2 are the axial tensile strengths in the
during high velocity impact. The continuum damage fill and warp directions, respectively, and S 1FS and
mechanics formulation takes into consideration the S 2FS are the layer shear strengths due to fiber shear
post failure mechanisms in a composite plate as failure in the fill and warp directions [6].
characterized by a reduction in material stiffness. A MAT 162 provides an insight into the physics of
set of damage variable (w i ) to relate the damage the delamination of the composite plate as given by
growth to stiffness reduction (E red ) in the material Eq. (6):
[12] is given by:
2 2 2
=f del S d 3 + 23 + =
2 31
1 0
1
(1r )
ni

S S S 31
(6)

j
wi = 1 e ni (1) 2 23

Ered= (1 wi ) E0 (2) where S 2 is the thickness tensile strength, and S 23

and S 31 are shear strengths assumed to depend on the
compressive normal stress 3 . Delamination factor,
where w i is the damage variable, n i the strain S d was selected iteratively by the fitting the
softening parameter, r j the damage threshold, E 0 the analytical prediction and found to be 0.3.
elastic modulus and E red the stiffness reduction. The
damage variable w i varies from 0 to 1 as r j varies 2.2 Finite Element Analysis
from 1 to infinite. For simplicity, the softening
 DAMAGE BEHAVIOR OF HYBRID COMPOSITE STRUCTURE
UNDER HIGH VELOCITY IMPACT

a rigid body with no deformation. A total of 16,000

Ux=Uy=Uz=0 elements are used. Initial velocity of a spherical
projectile with a mass of 40g and caliber of 9.5 mm
y
was set to 5000 m/s for complete penetration. The
Ux=0, y=z=0

Ux=Uy=Uz=0
symmetry

z
x
rubber layer has been modeled with a hyperelastic
continuum rubber element developed by Blatz and
Materials Ko [14]. In LS-DYNA, contact between the
projectile and the target plate was defined using a
contact eroding single surface [12]. The authors
projectile Uy=0, x=z=0
symmetry handled the penetration of the projectile using
(a) eroding elements with strain based failure criterion.
The impact absorbed energy of the plate was
projectile
z calculated according to the following equation:
y CMC
(646427 mm)

m ( vI2 vR2 )
impact

1
x EPDM50
(64645 mm) E = EI ER = (7)
2
S2-glass-1 Al7039
(64643 mm) (64643 mm)
where E, E I and E R , are the absorbed energy, the
Al-foam kinetic energy of projectile at impact on the target
(646415 mm)
plate and the residual energy of projectile through
S2-glass-2 the target plate, respectively; v I , v R and m are the
(646411 mm)
velocity of projectile at impact (impact velocity), the
(b)
velocity of projectile through the target plate
(residual velocity) and the mass of the projectile,
Fig. 1. Schematic of (a) boundary conditions and (b) respectively.
a lay-up sequence of hybrid composite structure
adopted in three dimensional finite element analyses.
Table 1. Material properties and model No. of the
Fig. 1 shows the schematic of boundary conditions hybrid composite structure [10, 16]. (S, N and Y
and a lay-up sequence of hybrid composite structure denote shear, normal and yield stresses respectively
adopted in three dimensional finite element analyses. and mark * indicates own experimental data)
Aforementioned damage model was applied to the
hybrid composite structure. For the aluminum-foam Material
Elastic or
and rubber, Fleck and Blatz-Ko models were applied Density Shear Strength (GPa)
(LS-DYNA
(g/cm3) modulus Failure Strain (%)
respectively as in the literature [14-15]. model No.)
(GPa)
HypermeshTM (Version 10) was used for pre-
processing in the model development. LS-DYNA S2-glass Ex=Ey=61 S=0.1, Nx=Ny=1.26,
1.40
(Mat 162) Ez=0.24 Nz=0.05 (20%)
(Version 971) was used to analyze perforation
mechanisms, failure modes, and damage evaluation CMC
3.60 G=113 Y=2.48 (0.9%)
during high velocity projectile impact on the six- (Mat 162)
layer hybrid composite target plates. The material
EPDM50
properties and model No. used in the simulations are 1.14* G=1.24* -
(Mat 7)
listed in Table 1. Only a quarter of the target plate
was modeled considering the symmetry conditions Al7039
2.70 E=70 Y=0.055 (17%)
with respect to the central axis as shown in Fig. 1(a). (Mat 162)
Both the projectile and the composite plates have Al-foam Y=0.00133*
been meshed with eight node brick elements with a 0.25* E=1.3*
(Mat 63) (5.0%)*
single integration point. The spherical projectile was
made using 896 brick elements and was assumed as

3
3 Results and Discussion Eq. (7). It is found that the energy absorption of S2-
glass is the largest whereas EPDM rubber is the
The main objective of the finite element analysis is
smallest. As shown in Fig. 2, penetration velocity
to investigate the deformation-failure response of the
reduction rate in the S2-glass layer is larger than
multi-layer composite plate in the event of a
those in other layers. The reason for this is due to the
projectile striking at a velocity of 5000 m/s.
stiffness discrepancies among the materials. The
Analyses of high velocity impact responses in terms
residual velocity of the projectile was influenced by
of energy absorption and stress contour plot are
the stress wave interactions, particularly by the
presented below.
amount of damage growth. Thus damage presence in
3.1 Penetration velocity variation of the projectile each layer is predictable from the Fig. 2.

6000
S2-glass-1
5000 2000
Impact velocity (m/s)

EPDM rubber

Impact absorption energy (J)

4000
CMC Al-foam 1600

3000
Al 1200
S2-glass-2
2000
800
1000
penetration
0 400
0.000 0.005 0.010 0.015 0.020 0.025 0.030
Time (ms) 0
S2-glass-1 CMC EPDM50 Al7039 Al-foam S2-glass-2
Materials
Fig. 2. Velocity variation of the projectile
penetrating the target plate. (a)
1.0x109
Impact absorption efficiency (J mm2/kg)

Fig. 2 shows the velocity variation of the projectile

penetrating the target plate. The arrows in Fig. 2 8.0x108
indicate the impacting moment that the projectile
reaches each layer. It is observed that penetrating 6.0x108
velocity reduces at a fast pace. It seems that the
kinetic energy also may be down as the projectile 4.0x108
goes through the target plate. At first the projectile
with an initial velocity of 5000 m/s penetrates the 2.0x108
S2-glass layer and the residual velocity decreases to
4632 m/s passing through the S2-glass-1 layer. 0.0
S2-glass-1 CMC EPDM50 Al7039 Al-foam S2-glass-2
Immediately the projectile impacts on the CMC
Materials
layer at a speed of 4632 m/s and goes through the
CMC/EPDM interface layer at a speed of 3556 m/s. (b)
After this, the projectile penetrates aluminum, al-
foam and S2-glass-2 layers in order and the final Fig. 3. Comparison of (a) impact absorption energy
velocity of the projectile after complete penetration and (b) impact absorption efficiency of the hybrid
is 1310 m/s. composite structure.

3.2 Impact Absorption Energy

Fig. 3(a) compares the energy absorption of each Fig. 3(b) compares the impact absorption efficiency
material under impact velocity of 5000 m/s. The of each material. The impact absorption efficiency is
impact absorbed energy can be calculated by using defined as the absorption energy divided by area
density of each material. Comparing Fig. 3(a) with
 DAMAGE BEHAVIOR OF HYBRID COMPOSITE STRUCTURE
UNDER HIGH VELOCITY IMPACT

Fig. 3(b), the impact absorption efficiency of S2- layer and (c) brittle fractures in the CMC layer after
glass is the largest and that of Al-foam has second the impact velocity of 5000 m/s.
largest value. Thus, it is confirmed that the roles of
Fig. 4(a) shows von-Mises stress contour plot and
S2-glass and Al-foam are very significant in a view
damage modes in each layer after complete
of impact absorption performance of multi-layer
penetration of the projectile. It is apparent that
composite for protection structures.
during complete penetration of the projectile
3.3 Damage Modes localized fractures and damages occur in the vicinity
of the impacting point along the target plate
thickness. Particularly wide range of damages is
found in the S2-glass layer as shown in Fig. 4(b). In
addition, large delaminations between the S2-
glass/CMC layers and between the CMC/EPDM
rubber layers are found. It is widely accepted that in
a typical composite system, the energy absorption
mechanism during impact is the local deformation
and fiber fracture. However, delamination has a
major role in dissipating a large amount of energy in
such multi-layer hybrid composite system. Fig. 4(c)
shows the damage in the CMC layer. Unlike the
damage mode in the S2-glass layer, damage zone is
distributed only in the contact region that projectile
penetrates. This may be due to the brittleness of
CMC. From the results, a hybrid composite structure
shows various types of damage modes according to
the constituent materials.
4 Conclusions
In this study, the damage behavior and impact
absorption energy in each layer of hybrid composite
structures have been investigated based on the three
dimensional finite element analyses. The results
obtained are summarized as follows:
(1) By evaluating the impact absorption energy and
impact absorption efficiency, S2-glass and Al-
foam are superior to other materials.
(2) From the simulation, a wide range of damages
and local delamination are occurred in S2-glass
layer, whereas CMC and aluminum reveal a
narrow band of damage only in the vicinity of
the zone that the projectile penetrates.
(3) Delaminations between the CMC/EPDM rubber
interlayer and between the S2-glass/CMC
interlayer are found to be representative damage
modes during high velocity impact.
(4) By applying progressive damage model to the
LS-DYNA, it is possible to evaluate
Fig. 4. Simulated damages showing (a) global quantitatively the local fracture behavior of
fractures, (b) wide range of damages in the S2-glass multi-layer composite structure under high
velocity impact, particularly, by comparing

5
 with the amount of impact absorption energy [13] A. Matzenmillar, J. Lubliner, RL. Taylor A
and efficiency. constitutive model for anisotropic damage in fiber
composites Mechanics of Materials, Vol. 20, pp.
Acknowledgment 125-152, 1995.
This work was supported by the Research fund of [14] Blatz. PJ, Ko. WL "Application of finite element
theory to the deformation of rubbery materials"
Survivability Technology Defense Research Center
Transection of Society of Rheology, Vol. 6, pp 223-
of Agency for Defense Development of Korea (No.
251, 1962.
UD090090GD) [15] Deshpande. VS, Fleck. NA "Isotropic models for
References metallic foams" Journal of the Mechanics and
Physics of Solids, Vol. 48, pp 1253-1283, 2000.
[1] Wu. E, Tsai. C, Chen. Y Penetration into glass/epoxy [16] Deka. LJ, Bartus. SD, Vaidya. UK "Damage
composite laminates Journal of Composite evolution and energy absorption of E-
Materials , Vol. 28(18), pp.1783-1803, 1994. glass/polypropylene laminates subjected to ballistic
[2] Goldsmith. W, Dharan. CKH, Chang. Hui Quasi- impact" Journal of Materials Science, Vol. 43, pp
static and ballistic perforation of carbon laminates 4399-4410, 2008.
International Journal of Solids and Structures, Vol.
32(1), pp. 89-103, 1995.
[3] Lee. SWR, Sun. CT Dynamic penetration of
graphite/epoxy laminates impacted by a blunt ended
projectile Composites Science and Technology, Vol.
49, pp. 369-380, 1993.
[4] Zhu. G, Goldsmith. W, Dharan. CKH Penetration of
laminated kevlar by projectiles-I. Experimental
investigation International Journal of Solids and
Structures, Vol. 29(4), pp. 399-420, 1992.
[5] Zh.u G, Goldsmith. W, Dharan. CKH Penetration of
laminated kevlar by projectiles- II. Analytical model
International Journal of Solids and Structures, Vol.
29(4), pp. 421-436, 1992.
[6] J. Leppnen Experiments and numerical analyses of
blast and fragment impacts on concrete
International Journal of Impact Engineering, Vol. 31,
pp 843-860, 2005.
[7] GAO. Davies, X. Zhang Impact damage prediction in
carbon prediction composite structures International
Journal of Impact Engineering, Vol. 16, pp 149-170,
1995.
[8] MOW. Richardson, MJ. Wisheart Review of low
velocity impact properties of composite materials
Composites Part A, Vol. 27, pp 11231131, 1996.
[9] WJ. Cantwell, J. Morton Impact perforation of
carbon fiber reinforced plastic Composites Science
and Technology, Vol. 38, pp. 119-141, 1990.
[10] H. Mahfuz, Y. Zhu, et al. "Investigation of high-
velocity impact on integral armor using finite element
method" International Journal of Impact Engineering,
Vol. 24, pp 203-217, 2000.
[11] Z. Hashin "Failure criteria for undirectional fiber
composites" Journal of Applied Mechanics, Vol. 47,
pp 329, 1980.
[12] LS-DYNA Theoretical Manual. Livermore Software
Tech. Corp., 2003.