See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/283991130

Finite element analysis of sandwich panels with stepwise graded aluminum

honeycomb cores under blast loading

Article  in  Composites Part A Applied Science and Manufacturing · January 2016

DOI: 10.1016/j.compositesa.2015.09.025

CITATIONS READS

34 376

6 authors, including:

Li Shiqiang Zhihua Wang

Taiyuan University of Technology Taiyuan University of Technology
11 PUBLICATIONS   115 CITATIONS    180 PUBLICATIONS   2,133 CITATIONS   

SEE PROFILE SEE PROFILE

Guoxing Lu
Swinburne University of Technology
274 PUBLICATIONS   4,415 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

behavior of auxetic materials View project

dynamic response of structures View project

All content following this page was uploaded by Li Shiqiang on 17 February 2019.

The user has requested enhancement of the downloaded file.

 Composites: Part A 80 (2016) 1–12

Contents lists available at ScienceDirect

Composites: Part A
journal homepage: www.elsevier.com/locate/compositesa

Finite element analysis of sandwich panels with stepwise graded

aluminum honeycomb cores under blast loading
Shiqiang Li a,b, Xin Li a,b, Zhihua Wang a,b,⇑, Guiying Wu c, Guoxing Lu d, Longmao Zhao a,b
a
Institute of Applied Mechanics and Biomedical Engineering, Taiyuan University of Technology, Taiyuan, Shanxi 030024, China
b
Shanxi Key Laboratory of Material Strength and Structural Impact, Taiyuan University of Technology, Taiyuan, Shanxi 030024, China
c
College of Mechanics, Taiyuan University of Technology, Taiyuan, Shanxi 030024, China
d
Faculty of Science, Engineering and Technology, Swinburne University of Technology, Hawthorn, VIC 3122, Australia

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

Article history: This paper presents details and brief results of an experimental investigation on the response of metallic
Received 25 July 2015 sandwich panels with stepwise graded aluminum honeycomb cores under blast loading. Based on the
Received in revised form 27 September experiments, corresponding finite element simulations have been undertaken using the LS-DYNA soft-
2015
ware. It is observed that the core compression stage was coupled with the fluid–structure interaction
Accepted 30 September 2015
stage, and the compression of the core layer decreased from the central to the peripheral zone. The blast
resistance capability of sandwich panels was moderately sensitive to the core relative density and graded
distribution. For the graded panels with relative density descending core arrangement, the core plastic
Keywords:
A. Honeycomb
energy dissipation and the transmitted force attenuation were larger than that of the ungraded ones
Sandwich structures under the same loading condition. The graded sandwich panels, especially for relative density descending
B. Plastic deformation core arrangement, would display a better blast resistance than the ungraded ones at a specific loading
C. Finite element analysis (FEA) region.
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction sequent’s, considerably higher transverse velocity of the face sheet

Sandwich panels have been increasingly exploited as blast
resistant structures [1–6]. When they are subjected to impulsive
loading, the core enables large compressive deformation and hence
absorbs a large amount of impact energy, contributing to the supe-
rior blast resistance of the sandwich structure relative to mono-
lithic counterpart with equivalent mass [3,4,7]. Dharmasena [3]
and Deshpande and Uth [7] studied the dynamic mechanical
response of honeycomb core sandwich structures by experiments
and finite element simulations. The honeycomb sandwich struc-
tures significantly outperform monolithic structures of equal mass,
resulting from the superior energy absorption capacity of core
layer and the higher bending strength of the sandwich structure.
However, Langdon et al. [8] studied the blast response of sandwich
panels with PVC foam cores, and found that at a specific loading
region the monolithic panels perform better than the sandwich
panels with PVC foam core. It is attributed to the lower transverse
stiffness of the individual components of the sandwich panel. Con-
⇑ Corresponding author at: Institute of Applied Mechanics and Biomedical
Engineering, Taiyuan University of Technology, Taiyuan, Shanxi 030024, China.
Tel.: +86 3516010560.
E-mail address: wangzh@tyut.edu.cn (Z. Wang).
develops at the beginning of the process causing larger deflections
and therefore larger in-plane stresses in the face sheet, despite the
high energy absorbing capacity of the foam core. Hence, optimiza-
tion between impact resistance and core configurations of sand-
wich panel design is one of the important issues [9]. Tilbrook
et al. [10] used an analytical approach to define several regimes
of behavior of sandwich beams. The FE calculations were also
employed to construct design charts in order to select the optimum
transverse core strength that either minimizes the back face
deflections or the support reactions for a given sandwich beam
aspect ratio or blast impulse. As expected, the optimal core
strength also depends on the level of blast impulse and core
strength [9,10]. A higher strength core will assure a better blast
resistance performance. However, the force transmitted to the
sandwich back face will be increased. In general, soft cores in front
transmit less impulse to the support attached to the main structure
[10,11]. Since the properties of layered graded core structures can
be designed and controlled, they showed great potential to be an
effective core material for absorbing the blast energy and improv-
ing the overall blast resistance of sandwich structures [12].
In recent years, stepwise graded materials, where the material
properties vary gradually or layer-by-layer within the material
itself, were utilized as a core material in sandwich composites

http://dx.doi.org/10.1016/j.compositesa.2015.09.025
1359-835X/Ó 2015 Elsevier Ltd. All rights reserved.
2 S. Li et al. / Composites: Part A 80 (2016) 1–12
Fig. 1. Geometry and dimension of the specimen: (a) geometry and dimension of a single cell; (b) sketch map of the graded sandwich panel; (c) geometry and dimension of
the graded and ungraded sandwich panels. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
[12–19]. The failure modes of the sandwich structures mainly
depend on their geometric parameters, and the graded core sand-
wich structures subjected to bending loads are optimally designed
for minimum weight [20]. Due to the superior core compression
and the wave dissipation characteristics of the graded core layers,
the energy absorption ratio of core layers and the blast resistance
of the sandwich structure can be improved [19,21,22]. Zhang
et al. [11] analyzed the dynamic response of sandwich steel panels
with three kinds of corrugated core arrangements consisting of
identical core density, subjected to dynamic air pressure loading.
It was found that the sandwich plate with relatively smoothly
graded core outperforms the other two arrangements. Liu et al.
[14] investigated the dynamic responses and blast resistance of
all-metallic sandwich-walled panels with graded aluminum foam
cores and compared them with those of conventional ungraded
ones. When graded specimens and ungraded ones are subjected
to identical air blast loadings, the blast resistance of the graded
composites was better than that of the ungraded ones, and the rel-
ative density tapered configuration had advantage over the others.
Li et al. [23] and Apetre et al. [24] numerically investigated the
dynamic responses of metallic sandwich structures with function-
ally graded cores. The particular core design can exhibit better
energy absorption and mitigate or completely prevent impact
damage on the sandwich structures.
Although the graded sandwich components outperform the
ungraded structures of equal mass, the relation between the den-
sity graded distribution of core layers and the blast resistance of
sandwich panels is still not entirely understood. To investigate
the behavior of blast loaded graded metallic sandwich panels, a
large number of experiments have been conducted, and the exper-
imental results were presented and discussed in detail in a sepa-
rate paper [22]. Based on the experiments, in this paper
corresponding finite element simulations are conducted using LS-
DYNA software. After validating the FE model, the structural
response process and the blast resistance of the structure under
different TNT charge conditions are analyzed.
2. Experimental procedure and results
The experimental procedure and results are briefly presented in
the following two sub-sections. The blast resistance of graded
sandwich panels, which were constituted of aluminum alloy face
sheets and triple layered graded honeycomb cores, were tested
on a ballistic pendulum. And the experimental results were com-
pared with conventional ungraded sandwich panels under the
same loading condition.
2.1. Experimental procedure
The specimens used in the tests were divided into two groups:
Graded group (consisting of two face sheets, two interface sheets
and three layers of honeycomb cores), Ungraded group (consist
of two face sheets and a core of honeycomb). The face-sheets and
interface sheets were made from AA2024-O aluminum alloy. The
aluminum honeycomb core was made from AA-5052-H32 alu-
minum alloys. A single regular honeycomb cell includes three crit-
ical geometric parameters, that is, cell side length (a), wall
thickness (s = 0.04 mm) and the expanding angle (h = 30°), as indi-
cated in Fig. 1(a); (b) and (c) shows the geometry and dimension of
sandwich panels, in which the overall side length (l) and thickness
of interface (t0 ) are constant and equal to 300 mm and 0.1 mm,
respectively. Three different cell side length a are adopted for the
core layers: L (a = 2.5 mm), M (a = 2.0 mm) and S (a = 1.5 mm).
 S. Li et al. / Composites: Part A 80 (2016) 1–12 3

(a) Photograph of the ballistic pendulum system

(b) Sketch of the ballistic pendulum system

Fig. 2. Experimental set-up. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

For example, specimen G1-LMS-h7⁄3-20 g-180 mm indicates a
graded sandwich panel with 3 honeycomb layers with a constant
core thickness (h) of 7 mm. The core arrangement is LMS from
front to back face. And the TNT charge mass W equals to 20 g
and the stand-off distance R is 180 mm. Similarly, UG1-h20⁄1-
a1.5-20 g-120 mm stands for a sandwich panel with a single hon-
eycomb layer 20 mm thick, a = 1.5 mm, W = 20 g and R = 120 mm.
The total thickness of the graded and ungraded panel H is
23.2 mm and 22 mm, respectively.
The sandwich panels were peripherally clamped between two
square steel frames. In the experiments, a four-cable ballistic pen-
dulum system was employed to measure the impulse imparted on
the specimen, as shown in Fig. 2. The frames were clamped on the
front face of the pendulum, and the charge was fixed in front of the
center of the specimen using an iron wire. With a TNT charge det-
onated, the impulsive load produced by explosion would push the
pendulum to swing. Based on the oscillation amplitude recorded
by a laser displacement transducer, the impulse on the specimen
can be further estimated [6,22,25,26].
2.2. Experimental results
The specifications of panels as well as the results of deformation
patterns, deflection, impulse and core compressions are given in
Table 1. The permanent deflection of the center point of the back
face of the graded specimens was always smaller than the
ungraded panels, and that of G6-SML (test No. 6) was the smallest
with the same TNT charge condition. It is shown that the core layer
of the all specimens exhibited large inelastic deformation with
local densification. On the other hand, in the graded specimens
the face deflections and core compressions were sensitive to the
core arrangements. Fragmentation occurred when an intensive
blast loading was applied to the structure, i.e. in specimens UG1-
h20⁄1-a1.5-20 g-120 mm and UG1-h20⁄1-a1.5-20 g-150 mm.
When the core layer with smallest cell side length (strongest)
was set as the first layer and the core layer with largest cell side
length as the back layer, the panel had an excellent blast resistance
in the graded specimens. As expected, specimens with the core
layer of smallest cell side length exhibited the best blast resistance
among the ungraded sandwich panels.
3. FE model
3.1. Material property
In the FE model, the face-sheets were simulated with an isotro-
pic hardening plasticity model [2,14,15,19,21,26,27]. The strain
rate effect was accounted for by using the Cowper–Symonds
model, and the yield stress is:

1q
e_ pl
rY ¼ ðry þ Benpl Þ 1 þ ð1Þ
D
B ¼ bEpl ð2Þ
4 S. Li et al. / Composites: Part A 80 (2016) 1–12

Table 1
Specifications of the sandwich panels and experimental results.

Test no. Specimens and deformation patterns Impulse (N s) Deflections (mm) Core compressions (mm)
Front (xF) Back (xO) C1 C2 C3
Graded group
1 20.27 19.8 21.3 6.31 6.21 0.52
G1-LMS-h7⁄3-20 g-180 mm

2 20.06 17.2 18.8 5.91 3.74 1.93

G2-LSM-h7⁄3-20 g-180 mm

3 20.36 20.9 19.1 5.40 5.30 0.58

G3-MLS-h7⁄3-20 g-180 mm

4 20.32 18.4 18.6 5.79 1.50 4.05

G4-MSL-h7⁄3-20 g-180 mm

5 20.23 17.9 17.6 4.31 5.79 1.45

G5-SLM-h7⁄3-20 g-180 mm

6 20.01 22.1 17.1 5.53 4.67 4.41

G6-SML-h7⁄3-20 g-180 mm

7 20.21 21.5 22.6 5.42 1.22 4.04

G6-SML-h7⁄3-25 g-250 mm

8 20.54 24.6 25.3 5.68 5.86 6.00

G6-SML-h7⁄3-30 g-300 mm

Ungraded group
9 20.19 Fragmentation 34.6 Fragmentation
UG1-h20⁄1-a1.5-20 g-120 mm

10 20.03 Fragmentation 21.7 Fragmentation

UG1-h20⁄1-a1.5-20 g-150 mm

11 20.14 15.8 18.6 7.7

UG1-h20⁄1-a1.5-20 g-180 mm

12 15.65 18.0 20.3 8.55

UG2-h20⁄1-a2.0-20 g-180 mm

13 18.23 22.7 26.7 10.88

UG2-h20⁄1-a2.0-25 g-250 mm

14 19.25 24.7 28.6 9.40

UG2-h20⁄1-a2.0-30 g-300 mm

15 15.31 22.0 22.2 13.6

UG3-h20⁄1-a2.5-20 g-180 mm

16 16.41 23.4 26.6 13.5

UG3-h20⁄1-a2.5-25 g-250 mm

17 19.03 27.3 26.1 13.0

UG3-h20⁄1-a2.5-30 g-300 mm

Epl ¼ Etan E=ðE  Etan Þ ð3Þ
where ry is the initial yield stress, B is the hardening constant, and
Epl is the hardening modulus. For isotropic hardening b = 1, epl is the
effective plastic strain, e_ pl is the strain rate, n = 0.25 is the hardening
exponent, D = 6500 s1 is the strain rate constant and q = 4 is the
strain rate exponent.
The density of the honeycomb core is calculated as [28]:
2s=a
q ¼ q ð4Þ
ð1 þ sin hÞ cos h s
where qs = 2700 kg/m3 is the density of the foil material and h = 30°.
The density of the three kinds of honeycomb are (q⁄)2.5 = 0.0246qs,
(q⁄)2.0 = 0.0308qs and (q⁄)1.5 = 0.0411qs, respectively.
The stresses are virtually independent of the impact velocity
because the foil material does not exhibit strain rate sensitivity
[2]. Therefore, a constant plateau stress was used for the whole
range of loading parameters. The stress–strain curves of the three
value of bulk honeycomb cores are shown in Fig. 3.
3.2. Blast loadings
In the experiments, the TNT charge was cylindrical with differ-
ent values of diameter and height, detonated at different stand-off
distance. The Arbitrary Lagrange–Eulerian (ALE) algorithms is suit-
able to analyzing the blasting process in numerical simulation [29].
Eight-node brick elements with ALE are adopted for the TNT
 S. Li et al. / Composites: Part A 80 (2016) 1–12 5

in the tests were modeled by means of varying their diameter and

height, but keeping the density of charge constant [22]. The blast
loading imparted on the front-face-sheet of the sandwich panel
is defined with erosional contact algorithm, which calculates the
interaction between explosion product and structure. As shown
in Fig. 4b, after 150 ls the contact force decreased to zero, and so
the TNT charge part can be deleted after then.

3.3. Validation of FE model

The face sheets were modeled by the shell element with

Fig. 3. The stress–strain curves of the three value of bulk honeycomb cores. (For
interpretation of the references to color in this figure legend, the reader is referred
to the web version of this article.)
charge. Since this ALE approach uses meshes that are imbedded in
material and deform with the material, it combines the best fea-
tures of both Lagrange and Euler methods, and allows the mesh
within any material region to be continuously adjusted in arbitrary
and predefined ways as a calculation proceeds. The TNT charge ele-
ments will be severely distort with its volume expanded [29], and
so the TNT charge part must be deleted when the blast contact
force decreased to zero.
A high explosive material model incorporating the JWL equation
of state was used to describe the material property of the TNT charge
[29,30]. The JWL equation defines the pressure as a function of rela-
tive volume and internal energy per initial volume, namely:



x R1 V x R2 V xE
p¼A 1 e þB 1 e þ ð5Þ
R1 V R2 V V
where p is the blast pressure, E is the internal energy per initial vol-
ume, V is the initial relative volume, x, A, B, R1, and R2 are material
constants, respectively. The key material types and mechanical
properties of sandwich shell and TNT charge, as well as the param-
eters of equations of state are listed in Table 2.
Fig. 4 shows a quarter geometric model of the 20 g explosive
cylinder with a diameter of 25 mm and height of 26 mm, and the
typical contact force history curve. The TNT charge FE model
consists of 8775 elements. The charges with different masses used
Belytschko–Tasy section formulation, while the aluminum honey-
comb cores were modeled by the solid element with full integrated
formulation, whose properties mimic those of the hexagonal cells.
Because of the symmetry of the structure and loading, only a quar-
ter of the panel was modeled, as shown in Fig. 5. The graded model
comprises 244,467 nodes and 198,607 elements, and the ungraded
model comprises 121,675 nodes and 110,000 elements. Mesh sen-
sitivity studies revealed that further refinement does not signifi-
cantly improve the accuracy of the calculations. The bolts used in
the tests to clamp the panels to the fixture were represented by
nodal constraints in the numerical model. Symmetric boundary
conditions about x–z and y–z planes were imposed. The ranges of
clamped and symmetric boundaries are shown in Fig. 5(c) through
the top view. Automatic, surface-to-surface contact options were
generally used for sandwich panel sheets and the core layers.
In the FE simulations, as the TNT charge elements distorting
with its volume expanded, there would be much hourglass energy.
If the hourglass energy was too large, the simulated results would
be unauthentic. In order to validate the energy conservation in the
blast process, the system energy status of test No. 1 is shown in
Fig. 6, which shows that the sum of kinetic energy, internal energy
and hourglass energy is equal to the total energy at any time, and
the maximum hourglass energy was only 5.9% of the total energy.
So it can be indicated that the energy is conservative in the entire
process of structure response.
Comparisons between the experimental results and the numer-
ical predictions are shown in Figs. 7 and 8. Due to the fact that
there was only free contact and no connection between the face
sheet and core layers, delamination modes between the front skin
and core occurred in all the specimens, and the largest displace-
ment and core compression occur at the central area of the panels.
It is evident that the results from the simulations agree reasonably
well with the experimental data. The simulated results of face
sheets deflections and core compression were for fully fixed
boundary condition, which may underestimate the front and back

Table 2
Input data in the numerical simulation (unit = cm, g, ls, Mbar).

Material Part name LS-DYNA material type, material property and EOS input data
⁄
AA-2024-O Face sheets MAT_PLASTIC_KINEMATIC
RO E PR SIGY ETAN
2.78 0.72 0.33 7.5E4 4.62E2

⁄
Aluminum honeycomb (AA5052-H32) Cores MAT_HONEYCOMB
RO E PR SIGY VF LCID
0.1107 8.90E3 0 1.95E3 0.237 Fig. 3 a = 1.5 mm
0.0837 5.60E3 0 1.95E3 0.205 Fig. 3 a = 2.0 mm
0.0675 3.49E3 0 1.95E3 0.196 Fig. 3 a = 2.5 mm

⁄
TNT Charge MAT_HIGH_EXPLOSIVE_BURN
RO D PCJ
1.63 0.67 0.19
⁄
EOS_JWL
A B R1 R2 x E0 V0
3.71 3.23E2 4.15 0.95 0.3 7.0E2 1.0
6 S. Li et al. / Composites: Part A 80 (2016) 1–12

Fig. 4. FE model of the 1/4 charge with mass of 20 g (a) and the typical contact force history curve (b). (For interpretation of the references to color in this figure legend, the
reader is referred to the web version of this article.)

Fig. 5. FE model of the specimens. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

face sheet deflection and over-estimate the core compression due
to possible slippage at the boundary regions in the experiments.
Especially for C3, which was much easier to slide down from the
boundary regions than others.
4. Simulation results and discussion
4.1. Blast and structural response process
Previous researches indicated that the response of the sandwich
structure can be divided into three stages [31]: fluid–structure
interaction stage, core compression stage and retardation stage.
The response process of specimen G1 under the condition of
W = 20 g, R = 180 mm is drawn in Fig. 9. It can be seen that the
TNT charge detonated above its center and the volume expanded
gradually. The propagation of the detonation wave along the axis
was faster than along the radial direction, and so the deformation
model of the explosive expanded with an obvious direction. With
the explosive volume expanding, the blast wave firstly impacted
on the center of the front face sheet and the core layers was com-
pressed sequentially, and the core compression stage was coupling
with the fluid–structure interaction stage.
 S. Li et al. / Composites: Part A 80 (2016) 1–12
Fig. 6. Energy history of the system. (For interpretation of the references to color in
this figure legend, the reader is referred to the web version of this article.)
After the fluid–structure interaction stage, the core layers
would be compressed continually and densified gradually. The
compression of the core layer decreased from the central to the
peripheral zone and the compression region varied in different lay-
ers; the compression area of the first layer was the largest one and
that of the third layer was the smallest.
4.2. Back face deflection for the panels of equal mass
Since personnel or objects shielded from blast attacks are usu-
ally behind the barriers such as sandwich panels, the back-face
deflection of the specimen is herein considered as the main
response of interest. Fig. 10 presents the residual defection
attained at the center of back face sheet as a function of the scaled
distance R:
R stress region under uniaxial compression and most of the energy
R¼p
3
ffiffiffiffiffiffi ð6Þ
W
where W is the TNT charge mass, R is the stand-off distance. All of
the sandwich plates in Fig. 10 have the same mass and the stand-off
distance is 180 mm. In this section, effects of the core density and
density distribution in graded panels are analyzed and discussed
under different scaled distance by varying the TNT charge mass.
Fig. 7. Experimental and numerical results of the deformation model of specimen UG2 (a), G1-LMS (b) and G6-SML (c). (For interpretation of the references to color in this
figure legend, the reader is referred to the web version of this article.)
7
It can be seen that the residual back face deflection decreased
with the scaled distance increasing at a specific stand-off distance.
Over the range of reference distance considered, the panel with lar-
ger cell side length (smaller core relative density) had a superior
blast resistance.
The tapered acoustic impedance material along the wave trans-
mitting direction is conducive to the energy absorption and shock
wave dissipation [19]. In some researches [12,14,19,23,24], the
impact resistance of graded sandwich structures is better than that
of traditional sandwich structures. From Fig. 10, it can also be seen
that the residual back face deflection of G6-SML was smaller than
that of G1-LMS. The back face deflection of G6-SML was the small-
est when the reference distance R P 0:58. However, for the refer-
ence distance R < 0:58, the panel with larger cell side length
(smaller core relative density) had the smallest permanent back
face deflection. It is attributed to the superior compression feature
of the thicker core with smaller relative density.
Sectioned half profiles are plotted in Fig. 11 for the graded sand-
wich panels and the equivalent mass ungraded sandwich plates at
the same TNT charge condition. The difference in back face deflec-
tions along the length (or width) of the sandwich panel gives a
measure of the global structure blast resistance effect of the core
configuration. For the condition of R ¼ 0:58 (W = 30 g,
R = 180 mm), the back face deflection curves of G6-SML and
UG2.5 were almost the same. Under the condition of R ¼ 0:66
(W = 20 g, R = 180 mm) the back face deflection of G6-SML was
smaller than UG2.5 for D < 70 mm, and both were the same for
the rest. When W increased to 35 g (R ¼ 0:55), the back face deflec-
tion curve of G6-SML was larger than UG2.5 for the whole defor-
mation zone. This should be taken into consideration in the
design of protective structures. In-depth theoretical analysis is
required in order to determine the relationship between the
back-face deflection and the reference distance accurately.
4.3. Energy absorption capability
For the bulk honeycomb, there is a stable and long plateau
was absorbed in this region. In order to understand the energy
absorption mechanism of the structure under blast loadings, the
components (front face-sheet, core layers and back face sheet)
plastic energy dissipation histories of two typical graded sandwich
panels (G1-LMS and G6-SML) under the condition of R ¼ 0:58
(W = 20 g, R = 180 mm) are given in Fig. 12. In the early stage of
the response, the aluminum honeycomb core was compressed by
8 S. Li et al. / Composites: Part A 80 (2016) 1–12

Fig. 8. Comparison of experimental and numerical results in terms of the face-sheet deflections and core compressions. (For interpretation of the references to color in this
figure legend, the reader is referred to the web version of this article.)

the front face sheet, resulting in core crushing and significant arrangements had an effective influence on the energy absorption
energy dissipation. Most of the energy is dissipated by the large capacity: the energy absorption of the third core layer (C3) in G6-
deformation of the front face-sheet and core compression. The core SML increased compared to that in G1-LMS.
 S. Li et al. / Composites: Part A 80 (2016) 1–12 9

t=0μs t=5μs t= 20μs

t=40μs t= 60μs t=100μs

t=150μs t=200μs t= 250μs

t=500μs t=1000μs t=1500μs

Fig. 9. A typical process of explosion product-structure interaction. (For interpretation of the references to color in this figure legend, the reader is referred to the web version
of this article.)

Effects of core relative density and core arrangement configura-
tion on the energy absorption of the components of the sandwich
panels are shown in Fig. 13. In general, a high core energy absorp-
tion ratio is beneficial for the blast/impact resistance of structures
[26,6,32]. The energy absorption ratio by the core layer increased
with the core relative density, whereas the plastic energy of the
back face sheet also did. On the contrary, a low relative density
core can effectively weaken the plastic deformation of the back
face and improve the structure blast resistance. Additionally, for
G6-SML specimens, the energy ratio of the graded core layers
was larger than the ungraded core and the plastic deformation of
back face also decreased. Therefore, a reasonable arrangement of
10 S. Li et al. / Composites: Part A 80 (2016) 1–12

Fig. 10. Residual back-face deflection of sandwich panels vs. reference distance. Fig. 13. Energy absorption partition of the sandwich panels (W = 20 g, R = 180 mm).
(For interpretation of the references to color in this figure legend, the reader is (For interpretation of the references to color in this figure legend, the reader is
referred to the web version of this article.) referred to the web version of this article.)

Fig. 11. Back face sandwich panel profile measurements. (For interpretation of the Fig. 14. Contact force–time curves (W = 20 g, R = 180 mm). (For interpretation of
references to color in this figure legend, the reader is referred to the web version of the references to color in this figure legend, the reader is referred to the web version
this article.) of this article.)

Fig. 12. Energy absorption vs. time curves for the graded sandwich panels. (For interpretation of the references to color in this figure legend, the reader is referred to the web
version of this article.)
 S. Li et al. / Composites: Part A 80 (2016) 1–12
Fig. 15. Compression stress of core layers (W = 20 g, R = 180 mm). The solid line is
the plateau stress of the core layers and the dash line is the undulation of the
plateau stress. (For interpretation of the references to color in this figure legend, the
reader is referred to the web version of this article.)
graded core layers would improve the plastic energy dissipation
mechanism and then enhance the blast/impact resistance of
structures.
4.4. Attenuation of the transmitted force
Due to the large compressibility of the aluminum honeycombs,
blast loadings with high intensity cause significant deformations in
the core due to the propagation of shock waves of stress through
the thickness. Different stress pulses are transmitted to the back
face depending on the intensity of the load and core characteristics
[33]. Therefore the attenuation capacity of impact force transmit-
ted to the back face sheet is an important performance factor as
the improvement of the graded panels. The history and maximum
value of contact forces between the back face sheet and the C3 core
layer of the panels are shown in Fig. 14. Soft cores were beneficial
in decreasing the impulse transmitted to the back face sheet
[10,11], and delaying the moment when the impulse arrives at
the back face, which results from the low relative density honey-
comb core with a larger thickness. For the graded core layers, the
contact force was much smaller than that of the ungraded cores.
In particular, G6-SML has the smallest values of the maximum con-
tact force. In all the cases, the maximum contact reaction force
reached after the cores were crushed. At 300 ls, the contact forces
for all the cases are nearly zero.
4.5. Compression stress
Generally, cellular material attenuates impact or blast induced
loads by cell collapse mechanism at low impact velocities or low
pulse pressure intensities, when the stress transmission in a cellu-
lar material is limited by the plateau stress before the densification
stage starts [33]. However, stress enhancement in cellular material
may occur when an intensive loading is applied, which could pro-
duce a higher pressure on the back face sheet [34]. In some exper-
imental studies, shock enhancement has been observed with the
specimens subjected to intense impact loads, leading to severe
damage to the objects to be protected [34–37]. The main point of
interest in this section is the analysis of the effect of the core prop-
erties on the compression stress with the core layers. The maxi-
mum compression stress of the core layers is shown in Fig. 15.
The compression stress of the first core layer in G1, G2 and G3
specimens is much larger than the respective plateau stress, which
11
indicated that the compression capacity of the core has been
exhausted. In specimens G5-SLM and G6-SML, the compression
stresses of the three core layers were close to their respective pla-
teau stress and it decrease from the front to back in G6-SML. This
indicated that the three layers in G5-SLM and G6-SML specimen
were not densified and stress enhancement did not occur. Conse-
quently, under the present loading conditions increasing the rela-
tive density of the first core layer would be better for the objects
to be protected.
5. Conclusions
Experimental investigations have been carried out to study the
impact behavior of layered graded metallic sandwich panels with
honeycomb core under air blast loading. Based on the experiments,
this paper presents a corresponding numerical simulation study
using LS-DYNA software package.
The main interests are in the effect of core arrangement config-
uration on the energy dissipation features, force attenuation char-
acteristic, and stress distribution of blast-loaded sandwich panels
of equal mass. Numerical deformation/failure modes and the
deflection at the back face-sheet are sensitive to the core relative
density and core arrangement configurations. Under the same
charge condition of R ¼ 0:66, the plastic energy dissipated by the
core layers and the contact force attenuation were larger than
those of the ungraded ones, especially for the graded panels with
relative density descending core arrangement (G6-SML). The com-
pressive stresses of the three core layers of G6-SML were close to
their respective plateau stress and decreased from the front to
back.
The graded sandwich panels display a good blast resistance
under a weak blast loading ðR P 0:58Þ. However, with the
increase of blast loading ðR < 0:58Þ, their back face sheet deflec-
tions were larger than the ungraded panels. This should be taken
into consideration in the design of protective structures.
Additionally, it can be noted that the adhesive layers can be
employed in graded configurations in practice. And this may affect
the overall dynamic response behavior and the wave propagation
in the graded materials, which should be further analyzed in
future.
Acknowledgements
This work is supported by the National Natural Science Founda-
tion of China (Nos. 11172196, 11572214 and 11402163), Natural
Science Foundation of Shanxi Province (No. 2014011009-1 and
2013011005-2), Foundation for the Top Young Academic Leaders
of Shanxi, Program for the Homecomings Foundation and the Out-
standing Innovative Teams of Higher Learning Institutions of
Shanxi. The fifth author (G. Lu) thanks the support from the State
Key Laboratory of Explosion Science and Technology (Beijing Insti-
tute of Technology) (project KFJJ13-2Z). The financial contributions
are gratefully acknowledged.
References
[1] Nurick GN, Langdon GS, Chi Y, Jacob N. Behaviour of sandwich panels subjected
to intense air blast – Part 1: experiments. Compos Struct 2009;91(4):433–41.
[2] Karagiozova D, Nurick GN, Langdon GS. Behaviour of sandwich panels subject
to intense air blasts – Part 2: numerical simulation. Compos Struct 2009;91
(4):442–50.
[3] Dharmasena KP, Wadley HNG, Xue Z, Hutchinson JW. Mechanical response of
metallic honeycomb sandwich panel structures to high-intensity dynamic
loading. Int J Impact Eng 2008;35(9):1063–74.
[4] Xue Z, Hutchinson JW. A comparative study of impulse-resistant metal
sandwich plates. Int J Impact Eng 2004;30(10):1283–305.
12 S. Li et al. / Composites: Part A 80 (2016) 1–12
[5] Theobald MD, Langdon GS, Nurick GN, Pillay S, Heyns A, Merrett RP. Large
inelastic response of unbonded metallic foam and honeycomb core sandwich
panels to blast loading. Compos Struct 2010;92(10):2465–75.
[6] Li X, Zhang P, Wang Z, Wu G, Zhao L. Dynamic behavior of aluminum
honeycomb sandwich panels under air blast: experiment and numerical
analysis. Compos Struct 2014;108:1001–8.
[7] Deshpand VS, Uth T. Response of clamped sandwich beams subjected to high-
velocity impact by sand slugs. Int J Impact Eng 2014;69:165–81.
[8] Langdon GS, Karagiozova D, Von Klemperer CJ, Nurick GN, Ozinsky A, Pickering
EG. The air-blast response of sandwich panels with composite face sheets and
polymer foam cores: experiments and predictions. Int J Impact Eng 2013
(54):64–82.
[9] Yahaya MA, Ruan D, Lu G, Dargusch MS. Response of aluminium honeycomb
sandwich panels subjected to foam projectile impact – an experimental study.
Int J Impact Eng 2015;75:100–9.
[10] Tilbrook MT, Deshpande VS, Fleck NA. The impulsive response of sandwich
beams: analytical and numerical investigation of regimes of behaviour. J Mech
Phys Solids 2006;54(11):2242–80.
[11] Zhang L, Hebert R, Wright JT, Shukla A, Kim J. Dynamic response of corrugated
sandwich steel plates with graded cores. Int J Impact Eng 2013;65:185.
[12] Wang E, Gardner N, Shukla A. The blast resistance of sandwich composites
with stepwise graded cores. Int J Solids Struct 2009;46(18):3492–502.
[13] Zenkour AM. A comprehensive analysis of functionally graded sandwich
plates: Part 2 – buckling and free vibration. Int J Solids Struct 2005;42
(18):5243–58.
[14] Liu X, Tian X, Lu T, Liang B. Sandwich plates with functionally graded metallic
foam cores subjected to air blast loading. Int J Mech Sci 2014;84:61–72.
[15] Liu X, Tian X, Lu TJ, Zhou D, Liang B. Blast resistance of sandwich-walled
hollow cylinders with graded metallic foam cores. Compos Struct
2012;94:2485–93.
[16] Zhang X, Zhang H. Optimal design of functionally graded foam material under
impact loading. Int J Mech Sci 2013;68:199–211.
[17] Etemadi E, Afaghi Khatibi A, Takaffoli M. 3D finite element simulation of
sandwich panels with a functionally graded core subjected to low velocity
impact. Compos Struct 2009;89(1):28–34.
[18] Mao YQ, Fu YM, Chen CP, Li YL. Nonlinear dynamic response for functionally
graded shallow spherical shell under low velocity impact in thermal
environment. Appl Math Model 2011;35(6):2887–900.
[19] Li S, Wang Z, Wu G, Zhao L, Li X. Dynamic response of sandwich spherical shell
with graded metallic foam cores subjected to blast loading. Composites Part A
2014;56:262.
[20] Xu G, Zhai J, Zeng T, Wang Z, Cheng S, Fang D. Response of composite sandwich
beams with graded lattice core. Compos Struct 2015;119:666–76.
View publication stats
[21] Li S, Lu G, Wang Z, Zhao L, Wu G. Finite element simulation of metallic
cylindrical sandwich shells with graded aluminum tubular cores subjected to
internal blast loading. Int J Mech Sci 2015;96–97:1–12.
[22] Li S, Li X, Wang Z, Wu G, Lu G, Zhao L. Dynamic behavior of the sandwich
panels with functionally graded aluminum honeycomb cores under air blast:
experimental investigation. Int J Impact Eng 2015.
[23] Li Y, Ramesh KT, Chin ESC. Dynamic characterization of layered and graded
structures under impulsive loading. Int J Solids Struct 2001;38(34):6045–61.
[24] Apetre NA, Sankar BV, Ambur DR. Low-velocity impact response of sandwich
beams with functionally graded core. Int J Solids Struct 2006;43(9):2479–96.
[25] Zhu F, Zhao L, Lu G, Gad E. A numerical simulation of the blast impact of square
metallic sandwich panels. Int J Impact Eng 2009;36(5):687–99.
[26] Li X, Wang Z, Zhu F, Wu G, Zhao L. Response of aluminium corrugated
sandwich panels under air blast loadings: experiment and numerical
simulation. Int J Impact Eng 2014;65:79–88.
[27] Yu T, Hua Y. Introduction to structural dynamics plasticity. Anhui Hefei: China
Science and Technology Press; 1994. p. 87–90.
[28] Gibson LG, Ashby MF. Cellular solids structure and
properties. Cambridge: Cambridge University Press; 1997.
[29] Jing L, Wang Z, Zhao L. Dynamic response of cylindrical sandwich shells with
metallic foam cores under blast loading – numerical simulations. Compos
Struct 2013;99:213–23.
[30] Hallquist JO. LS-DYNA keyword user’s manual. US: Livemore Software
Technology Corporation; 2007.
[31] Tagarielli VL, Fleck NA. A comparison of the structural response of clamped
and simply supported sandwich beams with aluminium faces and a metal
foam core. J Appl Mech 2005(72):408–17.
[32] Jing L, Xi C, Wang Z, Zhao L. Energy absorption and failure mechanism of
metallic cylindrical sandwich shells under impact loading. Mater Des
2013;52:470–80.
[33] Karagiozova D, Nurick GN, Langdon GS, Yuen SCK, Chi Y, Bartle S. Response of
flexible sandwich-type panels to blast loading. Compos Sci Technol 2009;69
(6):754–63.
[34] Li QM, Meng H. Attenuation or enhancement – a one-dimensional analysis on
shock transmission in the solid phase of a cellular material. Int J Impact Eng
2002;27(10):1049–65.
[35] Hanssen AG, Enstock L, Langseth M. Close-range blast loading of aluminium
foam panels. Int J Impact Eng 2002;27(6):593–618.
[36] Reid SR, Peng C. Dynamic uniaxial crushing of wood. Int J Impact Eng 1997;19
(5):531–70.
[37] Zhu F, Chou CC, Yang KH. Shock enhancement effect of lightweight composite
structures and materials. Composites Part B 2011;42(5):1202–11.