See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/312385384

Numerical study of ultra-high performance

concrete under non-deformable projectile
penetration

Article in Construction and Building Materials March 2017

DOI: 10.1016/j.conbuildmat.2016.12.216

CITATION READS

1 175

3 authors, including:

Jian Liu Chengqing Wu

University of Technology Sydney University of Technology Sydney
6 PUBLICATIONS 1 CITATION 195 PUBLICATIONS 1,508 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Chengqing Wu on 01 February 2017.

The user has requested enhancement of the downloaded file.

 Construction and Building Materials 135 (2017) 447458

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Numerical study of ultra-high performance concrete under non-

deformable projectile penetration
Jian Liu a,, Chengqing Wu a, Xiaowei Chen b
a
Centre for Built Infrastructure Research, School of Civil and Environmental Engineering, University of Technology Sydney, NSW 2007, Australia
b
Institute of Systems Engineering, China Academy of Engineering Physics, Mianyang 621900, China

h i g h l i g h t s

Mat_72R3 is used to characterize UHPC under static and impact loadings.

The DOP is investigated under various compressive strengths of UHPC, striking velocities and CRHs of projectile.

The crater damage is investigated under various compressive strengths of UHPC, striking velocities and CRHs of projectile.

An empirical formula is proposed to determine the DOP for UHPC.

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

Article history: This paper presents a numerical study in evaluating impact response of ultra-high performance concrete
Received 15 August 2016 (UHPC) cylinder targets under ogive-nosed projectile penetration with broad striking velocities from
Received in revised form 14 December 2016 300 m/s to 1000 m/s. Steel ogive-nosed projectiles with an average mass of 360 g are launched to pene-
Accepted 31 December 2016
trate UHPC cylinder targets with 750 mm diameter and 1000 mm length. The Karagozian & Case (K&C)
cementitious material model, namely, MAT_Concrete_Damage_Rel3 (Mat_72R3), is implemented into
finite element package LS-DYNA for UHPC. In order to accurately predict depth of penetration (DOP)
Keywords:
and cratering damage of UHPC cylinder targets, uniaxial compressive and four-point bending testing
UHPC
Projectile
results are used to validate 3D finite element material model. With the validated numerical model incor-
LS-DYNA porating dynamic increase factors (DIF) of UHPC, parametric studies are conducted to investigate effects
DOP of UHPC compressive strength, projectile striking velocity and projectile caliber-radius-head (CRH) ratio
Cratering damage on both DOP and cratering damage of UHPC targets. Moreover, an empirical formula to predict DOP is
derived according to the simulated data.
2017 Elsevier Ltd. All rights reserved.

1. Introduction While attracting great interests from both weapon designers

In recent years, terrorist attacks are becoming more and more
frequent throughout the world. Not only does the terrorism lead
to structural damage, but it seriously threatens residents regular
life. Therefore, investigations on resistance of construction materi-
als against impact loading caused by projectiles have attracted
much attention by researchers and engineers. Ultra-high perfor-
mance concrete (UHPC) is a promising construction material that
contains a large volume of fibers, low water-binder ratio, high
micro-silica content and elimination of coarse aggregate [1]. It
has excellent material qualities, such as ultra-high strength, good
ductility, excellent durability, outstanding abrasion resistance,
self-consolidating workability and low drying shrinkages [25].
Corresponding author.
E-mail address: Jian.Liu-4@student.uts.edu.au (J. Liu).
and structural engineers over the past few decades, impact
response of UHPC under high velocity projectile penetration
remains to be one of the greatest challenges in modern concrete
technology. Experimental investigations on UHPC under high-
velocities projectile impact have been conducted throughout the
world [610]. However, their achievements have been hindered
by rather high cost of facilities for experiments under high-
velocities projectile impact and by a time-consuming nature of
their experimental procedures.
Attempts have been made to build and optimize the material
model and appropriate algorithms for accurately numerical simu-
lations of conventional concrete targets, with the assistance of
explicit finite element software LS-DYNA or AUTODYN to docu-
ment the potential relationship between variables of impact
response, reducing even replacing complicated experiments. Teng
et al. [11] simulated process of high-velocity (2474 m/s) projectile

http://dx.doi.org/10.1016/j.conbuildmat.2016.12.216
0950-0618/ 2017 Elsevier Ltd. All rights reserved.
448 J. Liu et al. / Construction and Building Materials 135 (2017) 447458
penetrating into normal strength concrete (NSC) with compressive
strength of 2833 MPa by using the elastic-plastic hydrodynamics
material model, and it was found that the numerical results show a
good agreement with experimental results in terms of the residual
velocity of projectile and the scabbing diameter of concrete targets.
Nystrm and Gylltoft [12] applied a numerical approach to
describe the penetration into NSC target (40 MPa) by an ogive-
nosed projectile (485 m/s). In this work, modified Riedel, Hier-
mayer and Thoma (RHT) material model was employed in simulat-
ing impact response of NSC targets, and the numerical results show
a good correlation with experimental data with respect to the
depth of penetration (DOP), crater diameter and scabbing diame-
ter. Wang et al. [13] numerically studied the impact response of
high strength concrete (HSC) with the compressive strength rang-
ing from 60 to 140 MPa subjected to high-velocity (400600 m/s)
projectile penetration, and the numerical results are very consis-
tent with experimental results, especially for DOP. Tai [6]
employed the Johnson-Holmquist Concrete (JHC) model to eluci-
date impact response of reactive powder concrete (RPC) with com-
pressive strength of 162193 MPa under low-velocity (35104 m/
s) projectile penetration, indicating that numerical results corre-
late well with experimental data.
There exist a large number of numerical studies to simulate
projectile penetration into the conventional concrete. However,
very limited numerical research had been conducted to investigate
the impact response of UHPC subjected to high-velocity projectile
penetration. Existing numerical studies on conventional concrete
will provide effective research strategies and methodologies for
the present study on UHPC. Furthermore, valid and accurate pre-
diction of impact response of UHPC by numerical simulation will
also provide an alternative for engineers in designing defence
structural members to resist high velocity projectile perforation.
In the present numerical study, a calibrated and validated mate-
rial model, that is, MAT_CONCRETE_DAMAGE_REL3, along with
an equation of state EOS_TABULATED_COMPACTION and erosion
criteria MAT_ADD_EROSION, is applied to simulate the steel
ogive-nosed projectile penetration into UHPC cylinder targets. This
validated material model is then coded into explicit finite element
package LS-DYNA to simulate impact response of UHPC targets
subjected to high-velocity projectile penetration from 300 m/s to
1000 m/s. The effects of UHPC compressive strength, projectile
striking velocities and projectile caliber-radius-head (CRH) ratio
on DOP and cratering damage of UHPC targets are investigated.
Moreover, the empirical formulae to predict DOP and cratering
damage subjected to projectile penetration are derived according
to the simulated data.
2. Finite element model validation
In this paper, uniaxial compression and four-point bending tests
were conducted to validate the accuracy of the material model
used in the numerical simulation of projectile penetration into
UHPC targets. In the numerical simulations of uniaxial compres-
sion and four-point bending tests, LS-DYNA Implicit Solver is used
for the quasi-static analysis.
2.1. Mix design of UHPC samples
All UHPC samples were produced by mixing silica fume, fine
sand and powder materials in a laboratory concrete mixer. They
were primarily dry mixed for 5 min, before the addition of 70%
water and mixing for 3 min. Superplasticizer was then added
before the remaining 30% water was finally mixed. The mixing pro-
cess was undertaken for another 5 min before the addition of steel
fibers. The steel fibers were manually dispersed with the aim to
Table 1
Mix proportions of UHPC (unit:kg).
Constituent Amount
52.5 Cement 750
Silica Fume 225
Silica Flour 190
Sand 1030
Superplasticizer 16
Water 190
Steel Fibers 191
Nanoparticles 63
avoid clumping and to guarantee the uniform distribution and ran-
dom orientation of steel fibers. The steel fiber materials were
employed at a volumetric dosage of 2.5%, and the mix proportions
of UHPC are listed in Table 1.
2.2. Uniaxial compression test
Uniaxial compression test, shown in Fig. 1, was conducted at
Structural Laboratory, with the aims to obtain the material proper-
ties like Youngs modulus, Poissons ratio, compressive strength
and the stress-strain relationship curve, shown in Fig. 2, according
to Chinese Standard GB/T 50081-2002 test method. In the test, a
number of 100 mm  100 mm  100 mm UHPC cubes were mix
designed, and then they were cured at the room temperature
(20 2 C) for 28 days. All specimens were loaded via the hydrauli-
cally controlled constant load rate of 0.2 mm/min until failure. The
data was recorded employing two axial and lateral strain gauges
on each specimen, and four axial LVDTs were placed at each corner
of the loading dial. The average value for each material property
was listed in Table 2.
In order to ensure the sufficient accuracy of the material model
used in the numerical simulation of UHPC targets against projectile
penetration, a numerical simulation of uniaxial compression test is
primarily conducted to compare the average material sample
results. The comparison between the testing and simulated results
of compressive stress-strain curves is illustrated in Fig. 2, which
shows a good correlation between two curves. Therefore, the mate-
rial model accurately characterizes UHPCs material behaviour
under compression.
2.3. Four-point bending test
Four-point bending test was carried out on a number of beam
specimens by using an electromechanical servo hydraulic pressure
testing machine with capacity of 3000 kN according to Chinese
Standard GB/T 50081-2002 test method as shown in Fig. 3. The
dimensions of beam specimen were 100 mm  100 mm 
400 mm. The space between two load points was 1/3 of the clear
span which generated a region of no shear, and two supports were
symmetrically applied to the place at 50 mm from two edges of the
specimen. LVDTs were placed at mid-span and two supports to
measure the deflection of the specimen, and strain gauges were
located at mid-span along the depth of the specimen to produce
an experimental curvature profile.
A numerical simulation of the four-point bending test is carried
out to validate the material model used in the numerical simula-
tion of projectile penetration into UHPC targets. The comparison
between the test results and numerical simulation of typical
force-displacement curve at mid-span is shown in Fig. 4, which
indicates a fair agreement between two curves. Therefore, the
material model will accurately characterize UHPCs material beha-
viour under flexural tension.
 J. Liu et al. / Construction and Building Materials 135 (2017) 447458 449

Fig. 1. Setup of uniaxial compression test (a) real test (b) numerical simulation.

180 3.1. Projectile

160
The scheme of longitudinal sections of projectile, where the
140 CRH ratio value is 3.0, is shown in Fig. 5. The material model
MAT_JOHNSON_COOK coupled with equation of statement
Stress (MPa)

120
100
80
60
40
20 Test result
LS-DYNA result
0 trate the centroid of UHPC cylinder targets.
0 0.005 0.01 0.015 0.02 0.025 0.03
Strain
Fig. 2. Comparison of experimental and numerical compressive stress and strain
curves.
3. Numerical simulation of UHPC targets against projectile
penetration
Three-dimensional numerical models of both UHPC cylinder
targets and steel ogive-nosed projectiles are built and meshed by
using finite element software ANSYS. The process of steel projec-
tiles penetration into UHPC targets is simulated and analysed by
using explicit finite element package LS-DYNA. UHPC targets and
steel projectiles are developed as different parts employing eight-
node solid elements with one-point integration and hourglass con-
trol. In order to reduce the computation, one quarter of UHPC tar-
gets and steel projectiles in consideration of symmetric
configurations, boundary and loading conditions is solved by using
Lagrangian solution technique. CONTACT_ERODING_SURFACE_T
O_SURFACE is employed for the interface between steel projectiles
and UHPC targets. In the contact, the projectile and UHPC target are
defined as master part and slave part, respectively. In the numer-
ical simulation of UHPC targets against projectile penetration, LS-
DYNA Explicit Solver is used for the dynamic analysis.
EOS_GRUNEISEN is used to build the three-dimensional steel
ogive-nosed projectile. Fig. 5(b) shows the casing and filler of pro-
jectile models, where casing density and filler density are 7850 kg/
m3 and 1500 kg/m3, respectively. The casing of projectile was made
of common steel with the yield strength of 800 MPa, and inputs of
the material model are shown in Table 3. During the numerical
simulations, steel ogive-nose projectiles are launched with broad
striking velocities from 550 m/s to 1000 m/s, to horizontally pene-
0.035
3.2. UHPC targets
The UHPC cylinder targets are fully fixed, and the top face of
UHPC targets is placed perpendicular to the level ground. Fig. 6
shows the dimensions of UHPC cylinder targets, where the diame-
ter and thickness are 750 mm and 1000 mm, respectively. The Kar-
agozian & Case (K&C) cementitious material model, namely, MAT_
CONCRETE_DAMAGE_REL3 (Mat_72R3), is used to build the three-
dimensional UHPC cylinder targets. UHPC targets with compres-
sive strengths of 90 MPa, 130 MPa, 170 MPa and 190 MPa are mod-
elled in the simulation.
As the K&C model is only suitable for conventional concrete,
including NSC, HSC and HPC, a K&C model with manually cali-
brated parameters, which has been validated in the simulated sta-
tic compressive test and four-point bending test, is employed in
UHPC in this study.
The exponents, b1 and b2 , respectively govern the compressive
and tensile softening damage evolution, as the stress point moves
from the maximum strength surface to the residual strength sur-
face in stress and strain curve. According to Crawford et al. [14]
and Malvars work [15], expressions for b1 and b2 can be respec-
tively devised as follows:
b1 0:34h 0:79 1

Table 2
Material properties of UHPC.

Density Poissons ratio Compressive strength Youngs modulus Ultimate strain

2400 kg/m3 0.19 170 MPa 19.7 GPa 0.013
450 J. Liu et al. / Construction and Building Materials 135 (2017) 447458

Fig. 3. Setup of four-point bending test (a) real test (b) numerical simulation.

140 inch; f c is the unconfined compressive strength of UHPC in the

unit of ksi. However, in order to apply these equations, the local-
120 ization width wlz should be smaller than the element size h (no
matter what the real aggregate size is).
100 Some other parameters are calibrated to fit for UHPC in this
study. According to the validation of four-point bending test, the
Forces (kN)

80 tensile strength for 170 MPa UHPC input in the K&C model is
18 MPa, which is about 10.6% of compressive strength. Therefore,
60 the tensile strength of other UHPCs can be predicted by following
the certain proportional relation with the corresponding compres-
40 sive strength. Omega, namely, shear dilatancy factor, controls the
volumetric expansion. According to Crawford et al. [14], the pro-
20 Test result posed value for concrete without confinement is 0.75, and for a sit-
LS-DYNA result uation with confinement is 0.90. The key inputs for UHPC
0 employed in the simulation of projectile penetration are listed in
0 0.5 1 1.5 2 2.5 3 3.5 Table 4.
Displacement (mm) An automatic generated equation of state EOS_TABULATED_C
OMPACTION with calibrated bulking unloading modulus is applied
Fig. 4. Comparison of test and numerical load and displacement curves at mid- to reflect the volumetric stress and strain with respect to impact
span. response caused by high velocity projectile penetration. An erosion
algorithm MAT_ADD_EROSION is also employed in the material
model to avoid element distortion during the simulation. Luccioni
2
b2 0:09w2lz  0:98wlz 3:061  0:004f c 0:097f c  0:484 and Aroz [16] summarized the erosion criteria and limited values
2 used by various authors to simulate concrete subjected to blast and
impact loadings, and it has been found that the value of erosion cri-
where h is the characteristic element size in the unit of inch; wlz is teria is highly related to the mesh size of concrete. Teng et al. [11]
the localization width (LocWidth on the input cards) in the unit of and Wang et al. [17] employed the shear strain of 0.4 and the

Fig. 5. Projectiles (a) dimensions (b) numerical model.

 J. Liu et al. / Construction and Building Materials 135 (2017) 447458 451

Table 3
Key inputs for the casing of projectile.

Shear Modulus a b n c m Failure stress

77 GPa .792E + 9 .51E + 9 0.26 0.014 1.03 900 GPa
D1 EOS_GRUNEISEN
C S1 c A
4.0 0.4569 1.49 2.17 0.46

Fig. 6. UHPC targets dimensions.

Table 4 lower than 50 MPa compressive strength, proposed by the CEB-FIP

Key inputs for Mat_72R3. Model Code [24] and Malvar and Crawford [25]. As UHPC is a rel-
Model parameter Value atively new material in terms of its composition and fiber dosage,
Tensile strength 20.0 MPa (190 MPa UHPC) 18.0 MPa (170 MPa UHPC)
formulae proposed above are not acceptable. The DIF proposed
13.8 MPa (130 MPa UHPC) 9.5 MPa (90 MPa UHPC) from previous research [22] demonstrates that UHPC is less sensi-
LocWidth 1.9 mm tive compared with NSC. DIF for UHPC used in the present study is
b1 0.82 (default = 1.6) taken into account by conducting Split Hopkinson Pressure Bar
b2 0.45 (190 MPa UHPC) 1.42 (170 MPa UHPC)
(SHPB) tests [26], and DIF values both in compression and tension
2.76 (130 MPa UHPC) 3.29 (90 MPa UHPC)
Omega 0.75 (default = 0.5) are shown in Fig. 7. The implementation of DIF values measure-
ment will improve the accuracy of material model.

3.4. Simulation data

maximum tensile stress as erosion criteria to simulate the penetra-
tion process, and Wang et al. [18] used the tensile strain of 0.5 and
In the present simulation, 22 shots of projectiles on UHPC tar-
the compressive strain of 0.75. In the present study, two erosion
gets are carried out to investigate DOP and impact crater after pen-
criteria with compressive strain and shear strain corresponding
etration under various scenarios. All the simulation data of DOP are
to material fracture and failure are defined and applied indepen-
recorded in Table 5.
dently. Once any one of the criteria reaches the critical value, the
failed nodes and elements will be immediately deleted from calcu-
3
lations. In this simulation, the value for both compressive and
shear strains is set as 0.2 with the purpose of conservative
Dynamic Increase Factor

2.5
estimate.

2
3.3. Strain rate effects
1.5
Previous research shows mechanical properties of concrete are
very sensitive to strain rate, and mechanical properties of concrete 1
under high loading rate conditions significantly differ from those
under quasi-static conditions [17,1921]. The strain rate effect on 0.5
concrete is due to the combination of material effect and lateral DIF for UHPC
inertial confinement effect [22,23]. Typically, the strain rate effects 0
on both tensile and compressive strength of concrete are defined as -1000 -750 -500 -250 0 250 500 750 1000
a dynamic increase factor (DIF), which is the ratio of dynamic to Strain Rate (1/s)
quasi-static strength versus a range of strain rates. There are two
sets of formulae to approximate the strain rate effects on NSC with Fig. 7. DIF for UHPC in tension and compression.
452 J. Liu et al. / Construction and Building Materials 135 (2017) 447458

Table 5
Penetration tests data of DOP.
0
Test No. f c (MPa) CRH Vs (m/s) DOP (mm)

1 190 3.0 300 60

2 170 62
3 130 71
4 90 82
5 190 3.0 550 115
6 170 121
7 130 142
8 90 171
9 170 2.0 115
10 170 1.0 82
11 190 3.0 675 151
12 170 159
13 130 189
14 90 232
15 190 3.0 800 186
16 170 197
17 130 239
18 90 284
19 190 3.0 1000 235
20 170 265
21 130 317
22 90 414

4. Results and discussions
According to previous research proposed by Chen and Li [27]
and Lundgren [28], penetration mechanics could be classified into
four regimes in terms of the striking velocity of projectile V0 as fol-
lowing: (i) the non-deformable (rigid) penetration (V0 6 1 km=s,
i.e. the traditional Earth Penetration Weapon); the semi-rigid
(mass-abrasive) penetration (1 km=s < V0 6 1:5 km=s, i.e. the
Advanced Earth Penetration Weapon); Semi-hydrodynamic
penetration (1:5 km=s < V0 6 3 km=s, i.e. the armour piercing pro-
jectile); the hydrodynamic penetration (V0 > 3 km/s, i.e. the metal
jet, space debris). In this paper, the non-deformable penetration
analyses for V0 6 1 km=s are discussed, where the projectile is
treated as a rigid body, and the deformation of the projectile during
the penetration process is negligible.
4.1. Compressive stress distribution
Compressive stress distribution within the UHPC target
(130 MPa) against CRH3 projectile penetration at 800 m/s from
90 to 520 ls is shown in Fig. 8. Compressive stress distribution is
generated upon impact, and then propagates inward within the
UHPC target along the movement of the projectile. It can be
observed that high compressive stress, indicated in green fringe
levels, occurs around the interface between the projectile and
UHPC target. Other areas within the UHPC target, indicated in
red fringe level, seem not to be influenced by compressive stress.
The critical compressive stress of the UHPC target during penetra-
tion in this case can be up to around 3500 MPa, which is much
higher than the compressive strength of the UHPC target. There-
fore, the UHPC target will be easily crushed due to high velocity
projectile penetration, and the DOP is probably related to the com-
pressive properties of the UHPC target.
4.2. Effective plastic strain distribution
For the K&C material model, effective plastic strain is character-
ized as d 2k=k km , where k is the modified effective plastic
strain or the internal damage parameter, and km is the critical
value. Effective plastic strain ranges from 0 to 1 to 2 as the failure
surface migrates from initial yield to maximum failure to residual
failure respectively. The effective plastic strain differs from the
strain in a sense, but it can denote the extent of damage. The devel-
opment of effective plastic strain in UHPC target (130 MPa) against
CRH3 projectile penetration at 675 m/s from 90 to 500 ls is illus-
trated in Fig. 9. In this figure, the colour varying from blue to red
means the UHPC targets suffer more serve localized damages.
The localized damage is mostly located in the area around the
interface between the projectile and the UHPC target, and then
develops deeper through the UHPC target with the movement of
the projectile. However, other areas far away from impact point
seem not to experience such serve damage.
4.3. DOP analysis
4.3.1. Effect of compressive strength
Fig. 10 plots the results for DOP of 90 MPa, 130 MPa, 170 MPa
and 190 MPa UHPC targets against CRH3 projectile penetration
with striking velocities at 300 m/s, 550 m/s, 675 m/s, 800 m/s
and 1000 m/s. Taking Tests 11-14 for example, at the striking
velocity of 675 m/s, when continually increasing the compressive
strength of UHPC target from 90 MPa to 190 MPa, the DOP gradu-
ally decreases from 232 mm to 151 mm. For the remaining UHPC
targets with increasing compressive strengths against the same
striking velocity of projectile, the above decreasing phenomenon
in terms of the DOP tends to be the same. Therefore, it can be con-
cluded that the improvement of the compressive strength of UHPC
target tends to resist projectile penetration. The above findings
coincide with the conclusions for UHPC obtained from experiments
by Wu et al. [9] and Zhang et al. [29].
4.3.2. Effect of striking velocity
Shown in Fig. 10, by comparing Test 4, Test 8, Test 14, Test 18
and Test 22, it can be found that when five projectiles penetrate
into the UHPC targets with 90 MPa compressive strength at the
striking velocities of 300 m/s, 550 m/s, 675 m/s, 800 m/s and
1000 m/s, the corresponding DOPs are 82 mm, 171 mm, 232 mm,
284 mm and 414 mm, respectively, showing a significantly upward
tendency. The similar tendency occurred for 190 MPa, 170 MPa
and 130 MPa UHPC targets, so a conclusion can be drawn that as
the increasing of striking velocity of projectile, the DOP signifi-
cantly increases. The above findings are consistent with the con-
clusions for UHPC obtained from experiments by Wu et al. [9]
and Zhang et al. [29].
 J. Liu et al. / Construction and Building Materials 135 (2017) 447458 453

Fig. 8. Development of compressive stress in UHPC (130 MPa) at 800 m/s (a) t = 90 ls (b). t = 240 ls (c) t = 400 ls (d) t = 520 ls.

4.3.3. Effect of CRH ratio
According to the previous research on normal strength concrete
by Wang et al. [18], the CRH ratio of the ogive-nosed projectile is
probably an important factor to influence impact response of the
UHPC target against projectile penetration. Three sets of projectiles
with various CRH ratio values, namely CRH1, CRH2 and CRH3, were
considered to explore the effects of CRH ratio on the DOP, as well
as the cratering damage (analysed in Section 4.4.3). The density
of projectile for CRH2 and CRH1 is increased to ensure the mass,
that is, kinetic energy, of the projectile with various CRH ratios
remains unchanged. By comparing Test 6, Test 9 and Test 10,
shown in Fig. 11, the DOPs of the 170 MPa UHPC target for CRH3,
CRH2 and CRH1 at the striking velocity of 550 m/s are 121 mm,
115 mm and 82 mm, respectively. It indicates that, when reducing
the value of CRH ratio from 3.0 to 2.0, the DOP decreases, and the
above decreasing degree of DOP tends to be increased when con-
tinually reducing the value of CRH ratio from 3.0 to 2.0.
4.3.4. Proposed model to predict DOP
In the present study, a calibrated empirical formula is proposed
to predict the DOP according to modified NDRC formula by the US
National Defence Research Committee [30]. This proposed model
clearly demonstrates that the DOP exponentially increases with
the increase of the CRH ratio and striking velocity of projectile,
but exponentially decreases with the increasing of the compressive
strength of UHPC target, derived as following:
454 J. Liu et al. / Construction and Building Materials 135 (2017) 447458

Fig. 9. Development of effective plastic strain in UHPC (130 MPa) at 675 m/s (a) t = 90 ls (b) t = 240 ls (c) t = 400 ls (d) t = 520 ls.

3:8  105  ;
0:12
 m  V1:865 and abridged one, respectively. S is the ogive curvature radius,
DOP d 0:51 1:76
0
3 and d1 is the projectile abridged diameter. In this study, d1 and l1
fc d
are assumed as zero.
where d is the projectile diameter in the unit of mm; ; is the CRH This proposed formula is compared with other two existing
ratio of projectile, defined as ; S=d; m is the projectile mass in models to predict the DOP proposed by modified NDRC formula
the unit of kg; V0 is the striking velocity of projectile in the unit [30] and Wu et al. [31], shown in Fig. 13. It can be seen that the
of m/s; f c is the compressive strength of UHPC target in the unit modified NDRC formula is conservative to predict the DOP of UHPC
of Pa. Fig. 12 demonstrates the abridged sketch of ogive l1 - in the present study, and the proposed model by Wu et al. [31] is
nosed projectile, which shows the specific parameters used in the over-estimated to predict the DOP of UHPC, especially for
equation above. In detail, l and are the lengths of projectile shank 190 MPa and 170 MPa UHPC targets.
 J. Liu et al. / Construction and Building Materials 135 (2017) 447458 455

450
190 MPa UHPC 170 MPa UHPC
400 130 MPa UHPC 90 MPa UHPC
350
300
DOP (mm)

250
200
150
100
50
0
100 200 300 400 500 600 700 800 900 1000 1100
Striking velocity (m/s)

Fig. 10. DOP versus striking velocities for 90 MPa, 130 MPa, 170 MPa and 190 MPa
UHPCs.

140

120
DOP (mm)

100

80

60
170 MPa UHPC at 550 m/s
40
0.0 1.0 2.0 3.0
CRH ratio
Fig. 11. DOP versus CRH ratios of projectile for 170 MPa UHPC at 550 m/s.
4.4. Cratering damage analysis
4.4.1. Effect of compressive strength
The cratering damages of 90 MPa, 130 MPa, 170 MPa and
190 MPa UHPC targets against CRH3 projectile penetration with
the striking velocity of 300 m/s, presented by the effective plastic
strain, are demonstrated in Fig. 14. The diameter of cratering dam-
age is measured as the average value of the maximum and mini-
mum ones. By comparing Tests 14, for the UHPCs from 90 MPa
to 190 MPa, the diameters of effective plastic strain reaching at
2.0 are 222 mm, 196 mm, 158 mm and 148 mm, respectively,
demonstrating a decreasing tendency with the increasing of com-
pressive strength. Specifically, the decreasing degree is signifi-
cantly restrained when the compressive strength of UHPC
exceeds 170 MPa. Additionally, the same decreasing tendency of
the cratering damage occurs under other four scenarios of
550 m/s, 675 m/s, 800 m/s and 1000 m/s, respectively. Although
the tensile strength of UHPC target is increased as the increasing
of its compressive strength, it will not have any significant influ-
ence on the impact resistance of UHPC target [32]. Therefore, it
can be concluded that there is a decreasing trend, for the cratering
damage with the increasing compressive strength of UHPC target.
4.4.2. Effect of striking velocity
The cratering damages of 190 MPa UHPC targets against various
striking velocities of 300 m/s, 550 m/s, 675 m/s, 800 m/s and
Fig. 12. The abridged sketch of an ogive-nosed projectile.
1000 m/s are compared in Fig. 15. When increasing the striking
velocity from 300 m/s to 550 m/s, the cratering damage signifi-
cantly increases from 148 mm to 233 mm. However, there only
has a slight increase in terms of the crater diameter (from
233 mm to 278 mm) with the increase of striking velocity from
550 m/s to 1000 m/s. The same growing tendency of the cratering
damage occurs under other three scenarios of 90 MPa, 130 MPa
and 170 MPa, respectively. Therefore, a conclusion can be drawn
that the increasing striking velocity of projectile will enlarge the
cratering damage, and the above increasing degree tends to
decrease with the increasing of striking velocity.
4.4.3. Effect of CRH ratio of ogive-nosed projectile
The results for the cratering damage of 170 MPa UHPCs with
CRH1, CRH2 and CRH3 against projectile penetration with striking
velocity of 550 m/s are shown in Fig. 16. Three values of the crater-
ing damage for CRH3, CRH2 and CRH1 are 243 mm, 250 mm and
258 mm, respectively, which demonstrates that the increasing
value of CRH ratio only slightly decreases the cratering damage.
5. Conclusions
The present study adopts the finite element LS-DYNA Explicit
Solver to investigate impact responses of UHPC cylinder targets
with compressive strengths from 90 MPa to 190 MPa subjected
to the steel ogive-nosed projectile penetration with striking veloc-
ities from 300 m/s to 1000 m/s. Two material models,
MAT_JOHNSON_COOK, coupled with EOS_TABULATED_COMPAC
TION, and MAT_CONCRETE_DAMAGE_REL3, coupled with
EOS_GRUNEISEN, are respectively employed in building the pro-
jectile and UHPC target. Prior to the numerical simulation of pro-
jectile penetration into UHPC target, the uniaxial compression
456 J. Liu et al. / Construction and Building Materials 135 (2017) 447458

Simulation data Simulation data

Proposed model by Wu et al. [31] Proposed model by Wu et al. [31]
NDRC formula [30] NDRC formula [30]
Proposed model in the present study Proposed model in the present study
18 18
16 16
14 14
12 12
DOP/d

DOP/d
10 10
8 8
6 6
4 4
2 2
0 0
100 200 300 400 500 600 700 800 900 1000 100 200 300 400 500 600 700 800 900 1000
Vo (m/s) Vo (m/s)
(a) (b)

Simulation data Simulation data

Proposed model by Wu et al. [31] Proposed model by Wu et al. [31]
NDRC formula [30] NDRC formula [30]
Proposed model in the present study Proposed model in the present study
18 18

16 16

14 14

12 12
DOP/d
DOP/d

10 10

8 8

6 6

4 4

2 2
0 0
100 200 300 400 500 600 700 800 900 1000 100 200 300 400 500 600 700 800 900 1000
Vo (m/s) Vo (m/s)
(c) (d)
Fig. 13. Simulation data and prediction of DOP/d versus Vo using proposed model and modified NDRC (a) 190 MPa (b) 170 MPa (c) 130 MPa (d) 90 MPa.

Fig. 14. Cratering damage versus various compressive strength of UHPC under 300 m/s (a) 90 MPa (b) 130 MPa (c) 170 MPa (d) 190 MPa.
 J. Liu et al. / Construction and Building Materials 135 (2017) 447458
Fig. 15. Cratering damage versus various striking velocities of UHPC with 190 MPa (a) 300 m/s (b) 550 m/s (c) 675 m/s (d) 800 m/s (e) 1000 m/s.
Fig. 16. Crater dimensions versus CRH ratio of projectile for 170 MPa UHPC at 550 m/s (a) CRH = 3 (b) CRH = 2 (c) CRH = 1.
test and four-point bending test were used to validate the accuracy
of the material model used in UHPC. From this study, the following
conclusions are drawn herein:
(1) A K&C material model (Mat_72R3) with manually calibrated
parameters can well characterize UHPCs material behaviour
under uniaxial compression and flexural tension;
(2) The localized damage is mostly located in the area around
the interface between the projectile and the UHPC target,
and then develops deeper through the UHPC target with
the movement of the projectile. However, other areas far
away from impact point seem not to experience such serve
damage;
(3) UHPC targets are easily crushed due to high velocity projec-
tile penetration, and the DOP is highly related to the com-
pressive properties of UHPC. The enhancement of the
compressive strength of UHPC target improves its impact
resistance against projectile penetration in terms of the
DOP. In the non-deformable (rigid) penetration regime, the
value of DOP significantly increases with the increasing of
the striking velocity and CRH ratio of projectile;
(4) With the rising magnitude of the compressive strength of
UHPC target, the cratering damage decreases, but the crater-
ing damage will increase as the growing of projectile striking
velocity, and the cratering damage decreases as the increas-
ing of the CRH ratio of projectile;
(5) A calibrated empirical formula is proposed in this present
study to determine the DOP.
Acknowledgements
The research presented in this paper jointly supported by the
ARC Discovery Grant DP160104661 is gratefully acknowledged.
457
References
[1] K. Wille, A.E. Naaman, G.J. Parra-Montesinos, Ultra-High performance concrete
with compressive strength exceeding 150 MPa (22 ksi): a Simpler Way, ACI
Mater. J. 108 (1) (2011).
[2] P. Rossi, A. Arca, E. Parant, P. Fakhri, Bending and compressive behaviours of a
new cement composite, Cem. Concr. Res. 35 (1) (2005) 2733.
[3] B.A. Graybeal, Compressive behavior of ultra-high-performance fiber-
reinforced concrete, ACI Mater. J. 104 (2) (2007) 146.
[4] V.Y. Garas, L.F. Kahn, K.E. Kurtis, Short-term tensile creep and shrinkage of
ultra-high performance concrete, Cement Concr. Compos. 31 (3) (2009) 147
152.
[5] P. Prem, B. Bharathkumar, R.I. Nagesh, Mechanical properties of ultra high
performance concrete, World Acad. Sci. Eng. Technol. 68 (2012) 19691978.
[6] Y. Tai, Flat ended projectile penetrating ultra-high strength concrete plate
target, Theoret. Appl. Fract. Mech. 51 (2) (2009) 117128.
[7] R. Sovjk, T. Vavrink, P. Mca, J. Zatloukal, P. Konvalinka, Y. Song, Experimental
investigation of ultra-high performance fiber reinforced concrete slabs
subjected to deformable projectile impact, Procedia Eng. 65 (2013) 120125.
[8] P. Mca, R. Sovjk, P. Konvalinka, Mix design of UHPFRC and its response to
projectile impact, Int. J. Impact Eng. 63 (2014) 158163.
[9] H. Wu, Q. Fang, X. Chen, Z. Gong, J. Liu, Projectile penetration of ultra-high
performance cement based composites at 5101320 m/s, Constr. Build. Mater.
74 (2015) 188200.
[10] H. Wu, Q. Fang, J. Gong, J. Liu, J. Zhang, Z. Gong, Projectile impact resistance of
corundum aggregated UHP-SFRC, Int. J. Impact Eng. 84 (2015) 3853.
[11] T.-L. Teng, Y.-A. Chu, F.-A. Chang, B.-C. Shen, D.-S. Cheng, Development and
validation of numerical model of steel fiber reinforced concrete for high-
velocity impact, Comput. Mater. Sci. 42 (1) (2008) 9099.
[12] U. Nystrm, K. Gylltoft, Comparative numerical studies of projectile impacts
on plain and steel-fibre reinforced concrete, Int. J. Impact Eng. 38 (2) (2011)
95105.
[13] S. Wang, H.T.N. Le, L.H. Poh, H. Feng, M.-H. Zhang, Resistance of high-
performance fiber-reinforced cement composites against high-velocity
projectile impact, Int. J. Impact Eng. 95 (2016) 89104.
[14] J. Crawford, Y. Wu, J. Magallanes, H. Choi, S. Lan, Use and Validation of the
Release II K&C Concrete Material Model in LS-DYNA, Karagozian & Case,
Glendale, 2012.
[15] Y. Wu, J. Crawford, S. Lan, J. Magallanes, Validation studies for concrete
constitutive models with blast test data, in: Proc., 13th Int. LS-DYNA Users
Conf., LSTC, Livermore, CA, 2014.
[16] B. Luccionia, G. Arozb, Erosion criteria for frictional materials under blast
load, Mecnica Computacional 30 (2011) 18091831.
458 J. Liu et al. / Construction and Building Materials 135 (2017) 447458
[17] Z. Wang, J. Wu, J. Wang, Experimental and numerical analysis on effect of fibre
aspect ratio on mechanical properties of SRFC, Constr. Build. Mater. 24 (4)
(2010) 559565.
[18] Z.-L. Wang, Y.-C. Li, R. Shen, J. Wang, Numerical study on craters and
penetration of concrete slab by ogive-nose steel projectile, Comput. Geotech.
34 (1) (2007) 19.
[19] E. Parant, P. Rossi, E. Jacquelin, C. Boulay, Strain rate effect on bending behavior
of new ultra-high-performance cement-based composite, ACI Mater. J. 104 (5)
(2007) 458463.
[20] Z. Rong, W. Sun, Y. Zhang, Dynamic compression behavior of ultra-high
performance cement based composites, Int. J. Impact Eng. 37 (5) (2010) 515
520.
[21] Z. Rong, W. Sun, Experimental and numerical investigation on the dynamic
tensile behavior of ultra-high performance cement based composites, Constr.
Build. Mater. 31 (2012) 168173.
[22] Y. Hao, H. Hao, Z.-X. Li, Numerical analysis of lateral inertial confinement
effects on impact test of concrete compressive material properties, Int. J.
Protect. Struct. 1 (1) (2010) 145167.
[23] J. Li, C. Wu, H. Hao, An experimental and numerical study of reinforced ultra-
high performance concrete slabs under blast loads, Mater. Des. 82 (2015)
6476.
View publication stats
[24] Code C-FM, Comite Euro-International du Beton. Bulletin dinformation, 1993.
213: p. 214.
[25] L.J. Malvar, J.E. Crawford, Dynamic increase factors for concrete, DTIC
Document, 1998.
[26] Y. Su, J. Li, C. Wu, P. Wu, Z.-X. Li, Effects of steel fibres on dynamic strength of
UHPC, Constr. Build. Mater. 114 (2016) 708718.
[27] X. Chen, Q. Li, Transition from nondeformable projectile penetration to
semihydrodynamic penetration, J. Eng. Mech. 130 (1) (2004) 123127.
[28] R.G. Lundgren, High-velocity penetrators, in: Presented at the American
Institute of Aeronautics and Astronautics Missile Sciences Conference,
Monterey, CA, 79 Nov. 1994, 1994.
[29] M. Zhang, V. Shim, G. Lu, C. Chew, Resistance of high-strength concrete to
projectile impact, Int. J. Impact Eng. 31 (7) (2005) 825841.
[30] R. Kennedy, A review of procedures for the analysis and design of concrete
structures to resist missile impact effects, Nucl. Eng. Des. 37 (2) (1976) 183
203.
[31] H. Wu, Q. Fang, Z. Gong, Experimental and engineering analyses on DOP of HSFRC
target under the impact of rigid projectile, J. Ballistics 24 (3) (2012) 1924.
[32] S. Wang, M. Zhang, S. Quek, Tensile strength versus toughness of cement-
based materials against high-velocity projectile impact, Int. J. Protect. Struct. 2
(2) (2011) 207219.