1 Errores

Download as pdf or txt
Download as pdf or txt
You are on page 1of 9

Available online at www.sciencedirect.

com

ScienceDirect
Journal of the European Ceramic Society 34 (2014) 2823–2831

Evidence of optimal interfaces in bio-inspired ceramic-composite panels for


superior ballistic protection
Stefano Signetti a , Nicola M. Pugno a,b,c,∗
a Laboratory of Bio-Inspired and Graphene Nanomechanics, Department of Civil, Environmental and Mechanical Engineering, University of Trento,
Via Mesiano 77, I-38123 Trento, Italy
b Centre for Materials and Microsystems, Fondazione Bruno Kessler, Via Sommarive 18, I-38123 Povo (Trento), Italy
c School of Engineering and Materials Science, Queen Mary University of London, Mile End Road, E1 4NS, London, UK

Received 15 November 2013; accepted 19 December 2013


Available online 4 February 2014

Abstract
Ceramic-composite panels are acknowledged to provide effective impact protection even against small fragments and armour piercing projectiles.
Nature shows similar solutions, coupling an hard face and soft backing layers, in dermal animal armours for protection against predators. Finite
element simulations of impact on ceramic-composite panels, to evaluate their energy absorption capability, are presented. The influence of key
parameters, like interface strength and friction, on ballistic limit is studied. We find that a proper set of interface parameters is able to maximize the
specific energy absorption of the panel: although this optimum is variable case by case depending on projectile penetrability and target configuration,
general guidelines are provided. Oblique impact results in a higher ballistic limit also thanks to projectile change in trajectory, providing interesting
spots for future developments. Numerical results are compared with experimental data from literature and forecasts of analytical models.
© 2014 Elsevier Ltd. All rights reserved.

Keywords: Impact; Ceramic-composite panels; Energy absorption; Oblique impact; Finite element simulations

1. Introduction This solution can be found also in nature, for example in Ara-
paima gigas dermal armour, whose scales are made up of an
Laminated composite materials are widely employed in pro- external hard mineral layer on a multilayered collagen backing.6
tective armours, automotive and aerospace applications: these The optimum balance between lightness, thickness, perfor-
homogeneous panels have their weak point in small and high mance, and economic requirements is a challenging engineering
penetrating fragments and armour piercing (AP) projectiles. task. Ceramics are lighter with respect to traditional mono-
Hard faced ceramics with a multilayered composite backing lithic hard-steel panels, while comparable in stiffness, hardness
are widely used in order to solve these problems, for exam- and compressive strength. However, they are characterized by a
ple, in protective body armours. On these heterogeneous plates, higher density (about a factor of two) with respect to composite
impactors are first blunted and weared down by the exterior hard materials. Thus, their use has to be carefully balanced and limited
ceramic which also spreads the load over a larger area; then the in lightweight applications like, for instance, spacecraft or
composite tough backing1–5 deforms and absorbs the residual human body protection from micrometeorites and space debris.
kinetic energy of the decelerated and damaged fragment; the Alumina (Al2 O3 ), Boron Carbide (B4 C) and Silicon Carbide
backing also delays and mitigates the initiation of tensile fail- (SiC) are some of the most widely employed ceramics in the sec-
ure in the ceramic and it is capable to catch both ceramic and tor. For the backing, polyetilene- and aramid-based tough fibres
impactor fragments, preventing them to constitute further injury. are arranged in woven or unidirectional (UD) textiles within a
polymer thermoplastic or thermoset matrix (epoxy or vinylester
resins): they can range from traditional ones, like Kevlar®
∗ Corresponding author at: Laboratory of Bio-Inspired and Graphene (aramid), and more recent like Dyneema® (UHMWPE)7 or
Nanomechanics, Department of Civil, Environmental and Mechanical Engi-
Twaron® (aramid).8 The main advantage of composites is that
neering, University of Trento, Via Mesiano 77, I-38123 Trento, Italy.
Tel.: +39 0461 282525; fax: +39 0461 282599. that their properties can be tailored on the requirements for a
E-mail address: nicola.pugno@unitn.it (N.M. Pugno). specific application. High specific strength, specific stiffness

0955-2219/$ – see front matter © 2014 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.jeurceramsoc.2013.12.039
2824 S. Signetti, N.M. Pugno / Journal of the European Ceramic Society 34 (2014) 2823–2831

and toughness make them an obvious choice for aerospace vehi- into the design of the optimal solution that couples toughness
cles; the resistance against unkind environments (e.g. corrosive, maximization and weigh reduction. Analytical modelling for
UV, extreme temperatures, etc.), enhances the robustness of the the evaluation of impact behaviour of composite targets16,22
structure. A measure of the fibres performance against ballistic generally assumes the laminate resistance σ as a quadratic func-
impact was provided by Cunniff9 for flat targets hit by cylin- tion of the impactor istantaneous velocity V, taking into account
drical projectiles, and was defined as the product of the fibres strain rate effects, σ = σ (V, V2 ). Since the backing layers, due
specific toughness by the strain wave velocity. to impactor deceleration, progressively face a lower velocity,
The design and the performance evaluation of composite the specific absorbed energy Eabs for each ply is expected to
panels undergoing ballistic impact require the understanding of decrease as the number of layers N, i.e. the areal density of the
material properties under high-velocity impact conditions. Recht plate, increases:
and Ipson10 proposed a relatively simple analytical model, based
Eabs
on energy conservation laws, able to determine ballistic curves ∝ Nα (2)
to fit experimental results. In this model the final velocity Vf of N
the impactor is given, as a function of its initial velocity V0 , by: with α < 0. However, this is usually in contrast with experimental
 p tests which show that generally the exponent ␣ can be positive.
p 1/p
Vf = q V 0 − V B (1) Jacobs and Van Dingenen23 showed how the scaling of Eq. (2)
can invert from soft to hard (pressed) panel: however, in the study
where VB is the target ballistic limit, p a fitting parameter, usu- it is not provided a formal explanation of the trend. The obser-
ally equal to 2 in case of rigid projectiles and target resistance vation of scalings of energy absorption and the understanding
independent to impactor velocity, and q a coefficient depending of related mechanisms could lead into optimized panels against
on model assumption (e.g.: q = 1 if assuming that dissipation is high-strain and strain-rate loads.
only due to target deformation, and no projectile damage is con- Our study focuses on these trends in order to find if and
sidered). A comparison between experimental ballistic curves how the failure mechanism of ceramic-composite panels could
and Eq. (1) can be found in Ref. 11 showing that this model be enhanced in order to maximize dissipation. The outcomes
can apply with a good level of approximation for the estima- would be extensible in general to other typologies of multilay-
tion of the final velocity, both for isotropic and heterogeneous ered structures. The effect of oblique impact on the ballistic
composite material. limit is also studied. These scopes require advanced finite ele-
Espinosa et al.12 investigated the response of multilayered ment models with proper material constitutive laws in order to
ceramic–steel targets under high-velocity impact through finite catch the real dynamics of impactor–armour interaction and to
element simulations. A multiple-plane microcracking model to find effective ways of developing optimized solutions. This work
describe the inelastic constitutive behaviour of ceramics under presents a numerical model for ballistic impact simulations in
severe damage was implemented into a finite element code. Their hybrid ceramic-fibre reinforced composite armours. The explicit
analyses showed that the penetration process is highly depend- non-linear finite element solver LS-DYNA®24,25 was used.
ent on the multilayered configuration (stacking sequence) and
the target structural design (geometry and boundary conditions), 2. Modelling of impact
rather than on the type and properties of the ceramic material.
In addition the erosion parameter in simulations, to which the Basing on an energetic approach for modelling impacts, as
residual damage strength is related, plays a key role in predict- widely used in literature,16–23 the variation of projectile kinetic
ing the interaction of the penetrator in the target: thus, a coupled energy in penetrating the plate must balance the amount of dissi-
experimental and numerical study is found to be necessary in a pated energy (Eabs ) in the damaged volume of the target, which
meaningful ceramic-composite armour design. is assumed to be cylindrical and, defined by the projectile radius
Hetherington et al.13,14 developed an analytical model for R and the plate thickness t. Thus, the following relation can be
the analysis of two-component composite armours subjected to written:
normal and oblique impact. They observed that, circular con-
tours of constant deformation which occur in backing plates
1 2
2 mV0 − 21 mVf2 = σπR2 t (3)
under normal impact, tends to be elliptical for oblique strike.
where m is the mass of the projectile, V0 and Vf the initial and
They assumed that the projectile tip deforms into an ellipse as it
final velocity of the impactor respectively and σ the ultimate
impacts the front face of the ceramic under the oblique impact.
compressive strength of the material. Assuming a rigid projec-
It was found that an inclined ceramic composite armour plate is
tile, Eq. (3) yields Eq. (1) for q = 1 and p = 2. A more realistic
more effective, on a thickness basis, than one arranged perpen-
approach consists in considering the velocity as a quadratic func-
dicular to the line of impact; parallel the ballistic limit velocity
tion of the istantaneous impact velocity V; for each layer it can
increases with obliquity. This model is reported in Ref. 15.
be assumed:
Considering the cost related to the ceramic and composite  
materials used in ballistic experiments, the need for developing σ = σ0 a0 + a1 kV + a2 k2 V 2 (4)
accurate predictive simulation tools becomes more important.
Large simulation sensitivity campaigns would let the under- where a0 , a1 , a2 are parameters depending on material behaviour
standing of the influence of each considered parameter, leading and impactor geometry according to Ref. 21, σ 0 is the material
S. Signetti, N.M. Pugno / Journal of the European Ceramic Society 34 (2014) 2823–2831 2825

compressive strength at the reference strain-rate, and k = γ/σ0
the square root of the ratio between the undeformed target den-
sity ␥ and the nominal material strength. Extending Eq. (3) to a
sequence of n layers, the final velocity after the passage through
the (i-1)-th layer would be the initial velocity seen by the i-th
one (V0 = Vi-1 ) and then Vf = Vi . Introducing then the strength
dependency given by Eq. (4) and assuming a1 = 0 (generally
valid for high-velocity impacts), the energy balance can be eas-
ily integrated and the velocity Vi after the passage through the
considered i-th layer obtained:
⎛ ⎞

 2a2,i ki2 πσi R2  
⎜ exp −ti a 0,i + a2,i k 2V 2 − a0,i ⎟
⎜ m i i−1
⎟,
Vi = ⎝ 2 ⎠
a k 2,i i

(5)

valid until the expression under the root is positive, unless the
impactor residual velocity is equal to zero, so it is stopped. Sum-
ming the decelerating contribution of each ply given by Eq. (5),
the final velocity of the impactor can be obtained. Moreover,
the mutual effect between layers under deformation can not be
caught since Eq. (5) is independent from the distance s between
adjacent plies.
Parallel, three-dimensional finite element models of projec-
tile and target were built up using LS-DYNA® finite element
code,24,25 with high non-linear solving capabilities. The projec-
tile was modelled at this stage as rigid body. The thickness ratio
of the two-components composite plate was decided upon the
work of Hetherington,26 where it was concluded upon experi-
mental results that the optimum ratio in order to maximize the Fig. 1. (a) Finite element model of impact on ceramic-composite panel and (b)
energy absorption, i.e. the plate ballistic limit, is given by the example of ceramic-composite target deformation after complete penetration
following relation: (t1 + t2 = 3.8 + 2 mm, V0 = 700 m/s, Vf = 608 m/s).

t1 ρ2
≈4· (6) dimensions of the target representative part were decided
t2 ρ1
upon the fact that both experiments and numerical simulations
where t1 and t2 are the ceramic and composite thicknesses evidence that during high-velocity impact the radius of the load-
respectively, ρ1 and ρ2 the corresponding material densities. influenced area does not exceed six times the radius R of the
Fig. 1 shows the typical finite element model for impact sim- impacting object.
ulation used in our studies. Only a quarter of circular plate The ceramic was implemented through the
with proper boundary conditions was modelled for study- Johnson–Holmquist JH-II material model (MAT 110),27
ing plate energy absorption, half a plate for studying oblique able to describe the response of brittle materials subjected to
impact due to symmetry considerations. The minimum in-plane large deformation at high strain rates. Under quasi-static loading

Table 1
Ceramic material input constants of boron carbide (B4 C) for LS-DYNA® *MAT 110.28
Density [kg/m3 ] G [GPa] A B C M N
2510 197 0.927 0.7 0.005 0.85 0.67
EPSI T [GPa] SFMAX HEL PHEL BETA
1.0 0.26 0.2 19 8.71 1.0
D1 D2 K1 [GPa] K2 [GPa] K3 [GPa] FS
0.005 0.5 233 −593 2800 0

G – elastic shear modulus; A – intact normalized strength parameter; B – fractured normalized strength parameter; C – strength parameter (for strain rate dependence);
M – fractured strength parameter (pressure exponent); N – intact strength parameter (pressure exponent); EPSI – reference strain rate; T – maximum tensile pressure
strength; SFMAX – maximum normalized fractured strength; HEL – Hugoniot elastic limit; PHEL – pressure component at the Hugoniot elastic limit; BETA –
fraction of elastic energy loss converted to hydrostatic energy; D1 – parameter for plastic strain to fracture; D2 – parameter for plastic strain to fracture (exponent);
K1 – first pressure coefficient (equivalent to the bulk modulus); K2 – second pressure coefficient; K3 – third pressure coefficient; FS – failure criteria.
2826 S. Signetti, N.M. Pugno / Journal of the European Ceramic Society 34 (2014) 2823–2831

Table 2
Kevlar-epoxy composite backing properties for LS-DYNA® *MAT 58. For the detailed meaning of each parameter, see Ref. 25.
KEVLAR

Density [kg/m3 ] EA [GPa] EB [GPa] EC [GPa] PRBA TAU1 [GPa] GAMMA1


2130 164 164 18 0.0 0.0 0.0
GAB [GPa] GBC [GPa] GCA [GPa] SLIMT1 SLIMC1 SLIMT2 SLIMC2
18 18 18 0.05 0.50 0.05 0.50
E11C E11T E22C E22T GMS
0.35 0.1 0.35 0.1 0.25
XC [GPa] XT [GPa] YC [GPa] YT [GPa] SC [GPa]
0.950 0.500 0.950 0.500 0.200

EPOXY RESIN

Density [kg/m3] EA [GPa] EB [GPa] EC [GPa] PRBA TAU1 [GPa] GAMMA1


900 0.5 0.5 0.5 0.34 0.0 0.0
GAB [GPa] GBC [GPa] GCA [GPa] SLIMT1 SLIMC1 SLIMT2 SLIMC2
0.187 0.187 0.187 0.5 0.5 0.5 0.5
E11C E11T E22C E22T GMS
0.350 0.05 0.350 0.05 0.025
XC [GPa] XT [GPa] YC [GPa] YT [GPa] SC [GPa]
0.400 0.02 0.400 0.02 0.0075

EA – Young’s modulus in longitudinal direction (A); EB – Young’s modulus in transverse direction (B); EC – Young’s modulus in normal direction (C); PRBA –
Poisson’s ratio; TAU1 – stress limit for the non-linear part of shear stress vs. shear strain curve; GAMMA1 – strain limit for the non-linear part of shear stress vs.
shear strain curve; GAB – shear modulus in AB plane; GBC – shear modulus in BC plane; GCA – shear modulus in CA plane; SLIMT1 – stress limit factor after the
stress maximum (fibre tension); SLIMC1 – stress limit factor after the stress maximum (fibre compression); SLIMT2 – stress limit factor after the stress maximum
(matrix tension); SLIMC2 – stress limit factor after the stress maximum (matrix compression); E11C – strain at longitudinal compressive strength; E11T – strain at
longitudinal tensile strength; E22C – strain at transversal compressive strength; E22T – strain at transversal tensile strength; GMS – strain at shear strength; XC –
longitudinal compressive strength; XT – longitudinal tensile strength; YC – transversal compressive strength; YT – transversal tensile strength; SC – shear strength.

conditions ceramics may be assumed as elastic-brittle materials; input material constant are reported; these were available in
however, when high strain rates are involved (like in ballistic literature28 and calibrated through experimental tests.
impacts) the post-yield response of the ceramic becomes For the Kevlar® composite backing (Table 2), an orthotropic
significant in evaluating its behaviour and then may have non material model (MAT 58) was used, which implements a linear-
negligible influence on ballistic limit. The model allows for elastic branch followed by a non-linear post peak softening
progressive damage, taking into account residual material behaviour: the residual strength in compression, tension and
strength and compressive bulking. The ceramic component was shear can be defined as a fraction of the maximum material stress
constituted in our case by boron carbide (B4 C): in Table 1 the (defined by the SLIM factors). The equations which define the

25

20 Hourglass energy
Internal energy
Kinetic energy E abs
Energy (J)

Contact energy
15
Total energy

10

0
0.00 0.01 0.02 0.03 0.04 0.05

Time (ms)

Fig. 2. Energy balance of one of the performed simulations. The sum of internal, contact (sliding interface) and hourglass energies represents the amount of energy
absorbed at each time by the plate (Eabs ). The values refer to the quartered model.
S. Signetti, N.M. Pugno / Journal of the European Ceramic Society 34 (2014) 2823–2831 2827

Fig. 3. Residual velocity variation with the interlayer distance s. The dashed line represents the output value given by Eq. (5).

failure criteria are presented in detail in the paper by Schweizer- mode) control was set. The energy balance of one of the
hof et al.29 The composite material model was implemented in performed simulations (Fig. 2) confirms the effectiveness of the
the layered thick shell (TSHELL) element, coupled with a user built model, as regards contact modelling, material behaviour,
defined integration rule. This method, more refined than sim- and element performance. Hourglass (spurious modes) energy
ply using the smeared properties of the ply obtained by a rule is within the 15% of total internal energy, value that is generally
of mixture, assigns to each through-thickness integration point acknowledged for not affecting overall response of the system.
the corresponding material (fibre or matrix) properties and con-
stitutive behaviour according to the volumetric fraction of the 3. Energy absorption under ballistic impact
laminate.
The ceramic target and impacting fragment were mod- To first understand the role that layer interaction plays in the
elled with under-integrated solid elements. Hourglass (spurious resistance mechanism under ballistic impact, several target with

0.13-0.15

0.11-0.13

0.09-0.11

0.07-0.09
0.15
0.13 0.05-0.07
0.11
0.09 0.03-0.05
0.07
0.05 0.01-0.03
0.03
α 0.01
-0.01
-0.01-0.01
-0.03
-0.05 -0.03--0.01
-0.07
-0.09 -0.05--0.03
-0.11
-0.13 1.00
-0.15 -0.07--0.05
0.80
1.00 -0.09--0.07
0.60
0.80 -0.11--0.09
0.40
0.60
-0.13--0.11
0.40 0.20
SFLS /σ
NFLS /σ 0.20 0.00 -0.15--0.13
0.00

Fig. 4. Influence of delamination strength parameters on specific energy absorption (α exponent), for given inter-laminar friction (μS = 0.15, μD = 0.12), impactor
velocity and shape. The interface normal failure stress (NFLS) and shear failure stress (SFLS) are normalized with respect to single ply resistance σ. The best response
arises when both parameters reach approximately one half of laminate peak stress. Too high values of these quantities provide “punch shear” failure (i.e. tends to
dissipation by volume, Eabs = σπR2 t ⇒ α ≤ 0).
2828 S. Signetti, N.M. Pugno / Journal of the European Ceramic Society 34 (2014) 2823–2831

0.08
0.07-0.08
0.07
0.06-0.07
0.06
0.05-0.06
0.05
0.04-0.05
0.04 0.03-0.04
0.03
α 0.02
0.02-0.03
0.01-0.02
0.01
0-0.01
0.00
-0.01-0
-0.01
-0.02--0.01
-0.02
1.00 -0.03--0.02
-0.03
0.80
1.00 0.60
0.80
0.40
0.60
0.20 μS
μD 0.40
0.20 0.00
0.00

Fig. 5. Influence of the static (μS ) and dynamic (μD ) coefficient of friction on the scaling exponent ␣ of specific absorbed energy.

equity of materials and stacking sequence but different distance enhancing performances: when the distance between the layers
in between the layer were performed; in this way, any difference tends to be significant (almost over 10 times t), there is saturation
in the behaviour is a function of the mere stacking distance s, in the trend of residual velocity (and so of the absorbed energy).
set up for each configuration as a multiple of the each composite Since the impactor is modelled as rigid body, the decrement of
ply thickness. In Fig. 3 the variation of residual velocity after absorbed energy means a decrease in the internal energy of the
penetration with the distance between layers is shown. The dis- plate.
tance s is normalized with respect to the thickness t of the single Since the stacking distance between ceramic and composite
ply, assumed to be the same for all layers. The absorbed energy layers is seen to be important for overall deformation behaviour
decreases as the distance between layer increases, showing that under impact, the effect of interface parameter was then studied.
when layers are packed together they are able to work in synergy In fact, the discrepancy between the energy absorption provided
by the energy based analytical model and experimental results
(see as example Ref. 23) is intuitively imputable to all those
effects that are not considered in the energy balance: the most
important are delamination and mechanical interaction between
deforming layers. A campaign of simulation was performed in
order to assess the influence on interface parameter on overall
target behaviour and then on the amount of dissipated energy.
The main parameters of interest which define adhesive prop-
erty are, for the specific contact algorithm implemented in the
model: the normal failure stress (NFLS, correlable with Mode
I crack opening), the shear failure stress (SFLS, Mode II), and
friction.
Fig. 4 shows the dependence of the α exponent to inter-
face strength parameters, with equity of impactor characteristics,
velocity and interlaminar friction. An optimum is found. Start-
ing from null interlaminar resistance, both in the in-plane and
out-of-plane directions, the increase in delamination strength let
the layers to work better. Beyond a certain value, depending on
material stiffness, on the stiffness ratio between layers and on
the impactor energy and penetrability, the failure mode tends to
“punch shear” collapse and the α exponent decreases again and
can return negative. The trend is slightly more sensitive to shear
Fig. 6. Scheme of oblique impact configuration on ceramic-composite armour.
adhesive strength. It must be underlined that the failure mode
The damaged zone defined by θ and the possible deviation of the projectile δ
from the initial obliquity β are shown. is extremely dependent on the impactor velocity: punch shear
S. Signetti, N.M. Pugno / Journal of the European Ceramic Society 34 (2014) 2823–2831 2829

Fig. 7. Back view of the ceramic-composite panel with visualization of projectile imprint through contour of normal displacement: (a) oblique impact (β = 60◦ ) with
elliptical imprint; (b) normal impact with circular imprint.

failure may occur also for low values of interface strength when 4. Oblique impact
the impactor is characterized by high penetrability; thus, the
scaling depends also on impactor geometry and velocity. Given Simulations of 15◦ , 30◦ , 45◦ , 60◦ oblique impact were per-
that, an optimum value can be determined with equity of bound- formed and compared with normal impactor incidence. The
ary conditions and the solution that maximize absorption should initial velocity of the impactor, for all cases, is 700 m/s. The
be evaluated case by case. impactor is assumed at this stage to be rigid. In general, the
Fig. 5 shows the results for the same analysis on the value of ballistic limit of the target tends to increase with the increase
static (μS ) and dynamic (μD ) coefficient of friction: the perfor- in angle of obliquity, since, from geometrical consideration the
mance increases with the increase of the two coefficients. The length of the impactor path within the target would be greater in
influence is lower than interface parameter, mostly for two rea- the case of oblique impact (Fig. 6). In case of inclined target the
sons: first, friction arises only after that delamination occurs; projectile imprint on the plate tends to be elliptical with one of
second, it plays a role only for the sliding component of the the two axes that increase its length with the projectile obliquity,
interface displacements. Friction mainly arises from surface as also confirmed by our numerical simulations (Fig. 7). From
morphology. Since the morphology affects in some measure also Heteringhton and Rajagopalan30 , the expression that defines for
the interface strength and given the uncertainty in the experi- the i-th layer the amount of the absorbed energy in the Eq. (3)
mental measuring of these friction coefficients, friction could can be updated, for geometric considerations, as follows:
be considered in some way a minor amount of contribute to the
shear interface strength, and its effect neglected. Eabs (β; θ) = σi πci di ti (7)
2830 S. Signetti, N.M. Pugno / Journal of the European Ceramic Society 34 (2014) 2823–2831

6.0
5.5
5.0
4.5 Analytical solution
4.0 Numerical simulations

E abs /E abs,0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0 10 20 30 40 50 60 70 80

β (°)

Fig. 8. Variation of panel energy absorption with the angle of impact incidence β. Energy values are normalized with respect to the absorbed value in case of normal
impact(Eabs,0 ). Numerical simulations results are compared with the ones from the analytical relation proposed by Hetherington,13 assuming the opening angle of
the cone θ = 45◦ .

where β is the incidence angle of the impactor, θ is the angle of trend provided in Iqbal et al.31 which performed simulations and
diffusion of deformation after impact according to Fig. 7, ci and experiments on metallic targets.
di the axes of the imprint ellipse given for each layer by (Fig. 7): Another important aspect in oblique impact is the projectile
axis obliquity variation with respect to its original configuration.
R  n
ci = + ti tan θ (8.a) This is a direct consequence of the non-symmetry given by the
cos β initial incidence itself. Depending both on the nature of the target
i=1
and the shape of the projectile, this variation can be positive, i.e.

n
increases original obliquity (measured as the angle between the
di = R + ti tan θ (8.b)
normal to the target and the projectile axis, Fig. 7), or negative.
i=1
Fig. 9 shows the trend of the obliquity variation in relation to the
Fig. 8 shows the variation of absorbed energy with different initial inclination of the trajectory: the numerical spots can be
angle of incidence, and θ assumed to be 45◦ . The values are fitted with a quadratic function and the obtained results can be
normalized with respect to the energy absorbed under normal correlated with the analytical previsions provided in Rajagopal
impact condition. The results from numerical simulations are et al.32 for solely composite targets. Notice that for the specific
in good agreement with the analytical model, although the lat- set of simulations performed, in particular for the used impactor
ter is based on very simple hypotheses. The difference seems geometry and the specific plate stacking sequence, the variation
to slightly increase with the angle of incidence, and this can be of obliquity is negative: the projectile, instead of being driven
explained by the fact that with the increase of obliquity the real and incorporated in the thickness, assuring a further mechanism
behaviour tends to diverge with respect to the geometric assump- of protection, is accommodated to the backing. This behaviour
tions of the conoid formation. We underline the similarity with can be explained both with the specific target material behaviour

0
0 10 20 30 40 50 60 70 80
-2

-4

-6

-8
δ (°)

-10

-12

-14

-16

-18

-20

β (°)

Fig. 9. Variation of projectile obliquity δ with the initial angle of impact incidence β. The negative sign of δ means that the impactor tends to realign along the normal
direction, enhancing its penetrability.
S. Signetti, N.M. Pugno / Journal of the European Ceramic Society 34 (2014) 2823–2831 2831

under high-strain compressive loads and the peculiar shape of the 4. Pugno NM. The egg of Columbus for making the world’s toughest fibres;
projectile faces, which helps the penetration instead of driving 2013 (available at arXiv:1304.6658).
5. Simple trick turns commercial polymer into world’s toughest fiber. MIT
the impactor within layers. This suggests further investigation
Technology Review; May/June 2013.
on how these can affect the sign of variation of obliquity, which 6. Yang W, Chen IH, Gludovatz B, Zimmermann EA, Ritchie RO, Meyers MA.
means to discriminate between a high perforating striker or not, Natural flexible dermal armor. Adv Mater 2013;25:31–48.
simply by changing the layer stack-up or the projectile shape. 7. DSM Dyneema; 2013. http://www.dyneema.com
The combined effect of target obliquity and projectile shape as 8. TEIJIN Aramid; 2013. http://www.teijinaramid.com/aramids/twaron/
performed in Iqbal et al.30 for thin steel targets will be extended 9. Cunniff PM. Dimensionless parameters for optimization of textile-based
body armor systems. In: Proceedings of the 18th international symposium
to composite and ceramic-composite targets in a subsequent on ballistics. 1999. p. 1303–10.
paper. 10. Recht RF, Ipson TW. Ballistic perforation dynamics. J Appl Mech
1963;30(3):384–90.
11. Abrate S. Ballistic Impact on Composites. In: 16th international conference
5. Conclusions on composite materials. 2007.
12. Espinosa HD, Dwivedi S, Zavatteri PD, Yuan G. A numerical investigation
of penetration in multilayered material/structure system. Int J Solids Struct
Finite element models of ceramic-composite panels under
1997;35(22):2975–3001.
normal and oblique impact were developed. Our computations 13. Hetherington JG, Lemieux PF. The effect of obliquity on the ballistic
demonstrate good agreement with existing experimental data performance of two component composite armours. Int J Impact Eng
and the prevision of analytical models: thus, we can conclude 1994;15:133–7.
that these simulations can be used for an improved design of 14. Sadanandan S, Hetherington JG. Characterisation of ceramic/steel and
ceramic/aluminium armours subjected to oblique impact. Int J Impact Eng
lightweight ceramic-composite protective panels. We showed
1997;19:811–9.
that the energy absorption of the plate can be maximized for 15. Abrate S. Impact on composite structures. Cambridge, UK: Cambridge
each configuration by setting proper interface parameter: the University Press; 1998.
best behaviour is not given simply by the case of maximum 16. Abrate S. Modeling of impacts on composite structures. Compos Struct
delamination area but from the combination of material frac- 2001;51:129–38.
17. Abrate S. Impact engineering of composite structures, CISM courses and
ture, dissipation through deformation of the plate apart from the
lectures, vol. 526. New York: Springer Wien; 2011.
projectile imprint, and vibrations. Since the impactor penetra- 18. Goldsmith W. Impact – the theory and phisical behaviour of colliding solids.
bility (related to its velocity, geometry, and mass), the laminate Mineola, New York: Dover; 2001.
stacking sequence, and the material constitutive law can affect 19. Ben-Dour G, Dubinsky A, Elperin T. High-Speed penetration modeling and
the overall response, general values of these parameters can- shape optimization of the projectile penetrating. Mech Based Des Struct
2009;37:538–49.
not be provided. However, numerical simulations configure as
20. Banichuk NV, Ivanova SY, Makeyev YV. On penetration of rigid non-
a powerful tool to gain case by case the absorption optimum, axisymmetric bodies into elastic-plastic medium. Probl Strength Plast
maximizing the contribution in adding further layers, saving 2010;70:131–9.
then in weight. Finally, oblique impact was observed to cause 21. Forrestal MJ, Altman BS, Cargile JD, Hanchak SJ. An empirical equation
interesting effect on variation of striker obliquity, suggesting for penetration depth of ogive-nose projectiles into concrete targets. Int J
Impact Eng 1994;15:395–405.
further investigation on this direction, in order to understand,
22. Forrestal MJ, Tzou DY. A spherical cavity-expansion penetration model for
for example, the best arrangement of ceramic and composite concrete targets. Int J Solids Struct 1997;34:4127–46.
components, the worst shape condition in terms of projectile 23. Jacobs MJL, Van Dingenen JLJ. Ballistic protection mechanisms in personal
penetration, and to explain the optimized structure of natural armour. J Mater Sci 2001;36(3127):3142.
dermal armour. 24. Hallquist JO. LS-DYNA theory manual. Livermore: Livermore Software
Technology Corporation; 2006.
25. Hallquist JO. LS-DYNA keyword user’s manual. Livermore: Livermore Soft-
ware Technology Corporation; 2006.
Acknowledgments
26. Hetherington JG. The optimization of two components composite armours.
Int J Impact Eng 1992;12:409–14.
NMP is supported by the European Research Council (ERC 27. Cronin DS, Bui K, Kauffmann C, McIntosh G, Berstad T. Implementation
StG Ideas 2011 BIHSNAM on “Bio-Inspired hierarchical super- and validation of the Johnson–Holmquist ceramic material model in Ls-
nanomaterials”, ERC PoC 2013-1 REPLICA2 on “Large-area Dyna. In: 4th European LS-DYNA users conference. 2003.
28. Johnson GR, Holmquist TJ. Response of boron carbide subjected to
replication of biological anti-adhesive nanosurfaces”, ERC PoC
large strain, high strain rates, and high pressures. JPN J Appl Phys
2013-2 KNOTOUGH on “Super-tough knotted fibres”) and 1999;85(12):8060.
European Commission under Graphene Flagship. 29. Schweizerhof K, Weimar K, Münz T, Rottner T. Crashworthiness Analy-
sis with Enhanced Composite Material Models in LS-DYNA, Merits and
Limits. In: LS-DYNA world conference. 1996.
References 30. Hetherington JG, Rajagopalan BP. Investigation into the energy absorbed
during ballistic perforation of composite armours. Int J Impact Eng
1. Pugno NM. Mimicking nacre with super-nanotubes for producing optimized 1994;11(1):33–40.
super-composites. Nanotechnology 2006;17:5480–4. 31. Iqbal MA, Diwakar A, Rayput A, Gupta NK. Influence of projectile shape
2. Pugno NM. Graded cross-links for stronger nanomaterials – Insight. Mater and incidence angle on the ballistic limit and failure mechanism of thick
Today 2010;13:40–3. shell plates. Theor Appl Fract Mech 2012;62:40–53.
3. Tan PH, Han WP, Zhao WJ, Wu ZH, Chang K, Wang H, et al. The shear 32. Rajagopal A, Naik NK. Oblique ballistic impact behaviour of composites.
mode of multi-layer graphene. Nat Mater 2012;11:294–300. Int J Damage Mech 2013;0(0):1–29.

You might also like