Composites: Part B 42 (2011) 12851290

Contents lists available at ScienceDirect

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

On the strength and toughness properties of SFRC under

static-dynamic compression
Z.L. Wang a,b,, Z.M. Shi a, J.G. Wang c
a

Department of Geotechnical Engineering, Tongji University, Shanghai 200092, China

Key Laboratory of Geotechnical and Underground Engineering of Ministry of Education, Tongji University, Shanghai 200092, China
c
School of Mechanical and Chemical Engineering, The University of Western Australia, Crawley, WA 6009, Australia
b

a r t i c l e

i n f o

Article history:
Received 14 October 2010
Received in revised form 5 December 2010
Accepted 23 January 2011
Available online 1 February 2011
Keywords:
B. Strength
B. Impact behavior
D. Mechanical testing
Steel ber-reinforced concrete

a b s t r a c t
The addition of steel bers into concrete mix can signicantly improve the engineering properties of concrete. This paper experimentally studies the mechanical behaviors of steel ber-reinforced concrete
(SFRC) through both static and dynamic compression tests. Cylindrical specimens with three different
percentages of short and ne bers 0%, 1.5% and 3% by volume of concrete are rstly fabricated. These
specimens are then tested by MTS for static compression and split Hopkinson pressure bar (SHPB) for
dynamic impact. It is revealed that the failure mode of concrete considerably changes from fragile to ductile with the increase of steel bers. The plain concrete may fail under low strain-rate single impact
whereas the brous concrete can resist high strain-rate repeated impact. Stain-rate exerts great inuence
on concrete strength. Besides, toughness energy is proportional to the ber content in both static and
dynamic compressions.
2011 Elsevier Ltd. All rights reserved.

1. Introduction
Steel ber-reinforced concrete has become a practical alternative construction material in various structures. The steel bers
can be used externally to improve the compressive strength, exure and shear capacities of beams, slabs and wall made by conventional concrete effectively. Also, it can be used internally as
reinforcement replacing common steel reinforcing bars in structures due to its advantages [1,2]. These randomly distributed bers
may bridge microcracks and restrain their widening, thus delaying
the cracks further propagation. Such reinforcements may largely
improve the post-peak ductility and energy absorption capacity
of concrete [3].
Well designed members should be able to avoid catastrophic
failure of a structure [4]. Concrete structures are usually exposed
to various load environments in their service periods. A key design
issue is to fully understand the responses of these structures to
both static and dynamic loads. In conventional concrete members,
crack widths are restrained by the use of steel reinforcing bars.
However, the thicker steel bar has different heat expansion
compared to the surrounding concrete. This may produce microcracks on the interface between bar and concrete. How to maintain
the concrete structure with less or no propagation of microcracks is
a notable issue. Numerous publications can be found in both
Corresponding author at: Department of Geotechnical Engineering, Tongji
University, Shanghai 200092, China. Fax: +86 21 65985210.
E-mail address: cvewzL@tongji.edu.cn (Z.L. Wang).
1359-8368/$ - see front matter 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.compositesb.2011.01.027

experimental studies and in situ applications of steel bers in concrete frame buildings [58].
Strength and toughness have been recognized as two important
characteristics of steel ber-reinforced concrete [8,9]. In general,
toughness or energy absorption capacity (hereafter called toughness
energy) is determined from the area encompassed by the stress
strain curve in compression. This expresses the total energy absorbed by the specimen prior to its complete damage or failure.
The additional load-bearing capacity is in direct proportion to the
toughness that the steel bers impart to the concrete. The
improvement of residual strength of concrete also reects the
capability to carry more loads even after cracking [10].
Over the past several decades, several attempts have been made
in both numerical and experimental methods to understand the
mechanical responses of SFRC. Test methods like servo-controlled
material testing system (MTS) [11], drop-weight tester [8] and split
Hopkinson pressure bar (SHPB) technique [12] have been adopted
so far. The MTS device is typically used for quasi-static tests. For a
higher strain-rate like dozens, even hundreds per second, dropweight tester or pneumatic SHPB are usually employed. In particular, the SHPB is a very popular experimental apparatus for the
study of the dynamic responses of materials. It has been used by
numerous investigators to elucidate the dynamic mechanical properties of solid media [2,13].
This study will experimentally investigate the mechanical responses of the concrete cylindrical specimens by using MTS and
SHPB techniques. The emphasis is on the comparison of strength
and toughness properties between the plain concrete and the steel

1286

Z.L. Wang et al. / Composites: Part B 42 (2011) 12851290

was river sand with a neness modulus of 2.6. Short and straight
steel bers were added in concrete mixes at different volume fractions. The super plasticizer (SP) is a liquor of phenolic aldehyde
which was added to the mix with 1.5% dosage of the cement volume. Fiber shapes are shown in Fig. 1, and their specications
are listed in Table 1. The bers were added to each series of mixes
at 0.0%, 1.5% and 3.0%, by volume of concrete. These values correspond to the steel ber weight of 0, 117 and 234 kg per cubic meter
of concrete. Their volume fraction is denoted by a symbol Vf.

ber-reinforced concrete with volume fractions of 1.5% and 3.0%,

respectively. The mechanical responses of these specimens are
evaluated in terms of strain-rate effect, the ber content by volume
and the failure patterns of the tested specimens.

2. Experiment program
2.1. Materials
Ordinary Portland cement was used as the cementitious material. Dry non-compacted silica fume was provided by Zhongxing
Technology Company of China. The coarse aggregate was crushed
limestone with a maximum size of 10 mm. The ne aggregate

2.2. Specimen preparation

The ingredients of concrete mixes are presented in Table 2. In
the process of specimen-making, steel bers, cement, crushed
stone, sand, and silica fume were rstly mixed for about 5 min.
Water and super plasticizer were then added. The mixture was
mixed until uniform concrete was obtained. From each mix, two
sizes of cylinders (U70 mm  35 mm and U50 mm  100 mm)
were cast in steel moulds. The moulds were oiled and placed on
a vibration table vibrating at low speed to ensure good compaction
while the concrete was poured. The cylindrical specimens were
demoulded 24 h later and cured in lime-saturated water for
28 days at room temperature. The ends of all cylinders are carefully
ground in order to assure the parallelism of the end surfaces.
2.3. Experimental set-up
Quasi-static compression was performed in closed-loop servocontrolled material testing machine (MTS) with a capacity of
1000 kN. A complete uniaxial stressstrain response can be obtained through its measurement system. This experimental process
used displacement control. Two loading rates (e_ 105 and 102)
were tested. An inbuilt electronic data acquisition system was used
to record the axial displacements and corresponding loads. This
data acquisition can measure the total axial displacement over

Fig. 1. Photograph of steel bers tested.

Table 1
Properties of steel bers tested.
Type

Density (kg/m3)

Length (mm)

Diameter (mm)

Tensile strength (MPa)

Appearance

WSF1

7800.0

13.0

0.2

2500.0

Bright/straight

Table 2
Mix proportions (kg/m3).
Vf

w/(c + sf)

Cement

Water

Silica fume

Coarse aggregate

River sand

Steel ber

0.0%
1.5%
3.0%

0.35
0.35
0.35

440
440
440

171.5
171.5
171.5

50
50
50

850
811
770

900
900
900

0
117
234

Fig. 2. Split Hopkinson pressure bar device.

Z.L. Wang et al. / Composites: Part B 42 (2011) 12851290

1287

120
Vf=0.0%
Vf=1.5%
Vf=3.0%

100

(MPa)

80
60
40
Fig. 4. Comparison of toughness under quasi-static compression (e_ 102 ).

20
0

0.005

0.01

0.015

0.02

0.025

0.03

120
Vf=0.0%
Vf=1.5%
Vf=3.0%

100

(MPa)

80
60
40
20
0

0.005 0.01

0.015

0.02

0.025 0.03

0.035

0.04

Fig. 3. Stressstrain curves of SFRC under quasi-static compression.

c0
ls

es

the whole length of the specimen instead of the middle half-length.

It was expected that this would lead to better displacement measurements. The compressive strength and the stressstrain curve
of each specimen were thus determined. Remarkably, the shape
of the uniaxial stressstrain curves is strongly affected by testing
conditions such as the stiffness of the testing machine, the size

(a) Vf =0.0%

and shape of specimens [14], the loading rate, the concrete characteristics (like the water/cement ratio, aggregate type). Particular
care was exercised to avoid variation in the experimental process,
thus minimizing the inuences of testing conditions.
The dynamic compression tests are performed by using a SHPB
technique (see Fig. 2). This device consists of main body, energy
source and measurement system. Main body mainly contains a
launch tube, a projectile (or striker bar), an incident bar, a transmission bar and energy-absorbing parts. The energy source system
has air compressor and pressure vessel. The measurement system
consists of velocity and dynamic strain indicator. The projectile,
incident and transmission bars have the Youngs modulus of
210.0 GPa and the wave velocity of 5190.0 m/s. The compressive
longitudinal incident wave is initiated by the impact of the striker
bar on the incident bar. The incident wave travels along the bar and
is recorded at the strain gauge A. When the stress wave reaches the
interface between the incident bar and the specimen, it is partially
reected by this interface and partially transmitted into the concrete specimen. The reected wave can be also recorded by the
strain gauge A on the incident bar. At the interface between the
specimen and the transmitted bar, the stress wave is again partially reected and partially transmitted. The strain gauge B records the transmitted stress wave in the transmitted bar.
The average strain and stress of the specimen are calculated by
[2,12]:

eI t  eR t  eT tdt

A
EeI t eR t eT t
2As

where A and E are the cross-section area and the Youngs modulus
of the elastic incident and transmitted bars, respectively; c0 is the
wave velocity in the bars. As and ls are instantaneous cross-section

(b) Vf =1.5%

(c) Vf =3.0%

Fig. 5. Comparison of failure patterns under quasi-static compression.

Z.L. Wang et al. / Composites: Part B 42 (2011) 12851290

area and original length of specimen; eI and eR are the incident and
the reected strain waves in the incident bar, and eT is the transmitted strain wave in the transmitted bar.
In the SHPB testing, the accuracy of the experimental results is
determined by the stressstrain uniformity of the specimen. Since
the strain signals are recorded from the incident and transmitted
bars, there are two basic assumptions for the valid application of
Eqs. (1) and (2) to obtain engineering stress, strain and strain-rate
in the specimen [15,16]:

140
21.4/s
33.7/s

120
100

(MPa)

1288

(i) One-dimensional elastic stress wave theory is valid in the

bars.
(ii) Stress and strain states are uniaxial and uniform within the
specimen.

80
60
40
20

3. Experimental results and analyses

0.002

0.004

0.006

3.1. Quasi-static mechanical performance

3.2. Single impact induced failure test

Only the single compression impact is focused in this section. In
order to achieve the failure of the plain concrete, the strain rate of
loading is set as 21.4/s after several attempts. Similarly, the minimum strain rates are 79.7/s and 91.0/s for the brous specimens
containing volume fractions of 1.5% and 3.0%, respectively. These
strain rates can be achieved via impact velocity which is adjusted
by the gas pressure. Fig. 6 presents the inuences of ber content
and strain rate on the dynamic behaviors of SFRC in the single

0.01

(a) Vf =0.0%
140
79.7/s
89.9/s

120

(MPa)

100
80
60
40
20
0

0.005

0.01

0.015

0.02

0.025

(b) Vf =1.5%
140
91.0/s
102.1/s

120
100

(MPa)

The stressstrain curves of specimens are illustrated in Fig. 3 for

the quasi-static tests. For plain concrete, the stressstrain plot
obviously demonstrates the brittle behavior of the material. Its
post-peak strength decreases rapidly with the increase of strain
after the maximum permissible load. When the macrocrack is
localized, its propagation is very rapid and only needs low energy.
For the ber-reinforced concrete, the curve has different shape. The
ascending part is the same as the plain concrete. However, the
post-peak behavior (descending part) is very different. When the
ber volume fraction is 0.0%, the compressive strengths (maximum
stress of the stressstrain curve) are 64.2 MPa for the e_ 105 case
and 66.7 MPa for the e_ 102 case. The corresponding peak strains
are 0.49% and 0.51%, respectively. When the ber volume fraction
is 1.5%, the compressive strengths are 72.8 MPa and 89.2 MPa,
respectively. When the ber volume fraction is 3.0%, the compressive strengths are 89.0 MPa and 102.7 MPa, respectively. The corresponding peak strains now increase up to 0.87% and 0.92%. These
data clearly demonstrate that the strength and corresponding peak
strain signicantly increase with the ber volume fraction.
As the above discussed, the toughness is an index to measure
the energy-absorbing capability of concrete. It is calculated
through the area under the stressstrain curve up to specied
strain. In this research, this area is integrated by a trapezoidal rule.
The change of toughness versus ber volume fraction is presented
in Fig. 4 by a histogram. This gure shows that the toughness signicantly increases with the ber content. The inclusions of 1.5%
and 3.0% ber volume increase the toughness energy by 5.6 and
9.7 times compared to the plain concrete, respectively. Fig. 5 compares typical failure modes of the three specimens under same
strain rate (e_ 102 /s). These photographs indicate that plain concrete is more seriously damaged than the ber reinforced ones. The
specimen of plain concrete takes on an inverted-cone shape failure,
which conforms to the failure characteristic of brittle materials.
The specimen with 3.0% steel bers by volume fraction has no
obvious failure surface, while the Vf = 1.5% specimen exhibits a definite shear plane but their two parts still combine together.

0.008

80
60
40
20
0

0.005

0.01

0.015

0.02

0.025

(c) Vf =3.0%
Fig. 6. Stressstrain curves of SFRC under dynamic compression.

impact compression tests. Firstly, this gure shows that all the
stressstrain relationships have a linear portion until the damage
threshold is reached, then followed by a softening portion.

1289

Z.L. Wang et al. / Composites: Part B 42 (2011) 12851290

(b) Vf =1.5%

(c) Vf =3.0%

Fig. 7. Comparison of failure patterns under dynamic compression.

Secondly, the dynamic response of the SFRC is sensitive to strain

rate. It is observed that the peak stress and corresponding strain increase obviously with the increase of strain rate. Also, the addition
of bers increases peak strain as well as compressive strength. That
is, both ascending and descending branches of the stressstrain
curves are affected by the addition of steel bers, but the effect
in the descending branch is more obvious [2]. Thirdly, steel bers
have an excellent performance on the toughness of SFRC. The
toughness energy also increases with the increase of strain rate
[17,18]. In addition, the volume fraction of bers has vital impacts
on the failure patterns of concrete. Fig. 7 gives the photographs of
post-test specimens after single impact. The plain concrete breaks
into pieces, but for the specimen with the Vf = 1.5% still remains the
overall outline although small segments peel off. The specimen
with ber volume fraction of 3.0% keeps essential integrity.

3.3. Evaluation of strain-rate effect

These static and dynamic test results show that the effect of
strain-rate on concrete strength is distinct. Fig. 8 illustrates the increases of the compression strength with the logarithm of strainrate in the experiments. This gure shows that the strain-rate effect exists in both quasi-static and dynamic tests. This effect is
more obvious for dynamic impact than for quasi-static compression. Besides, larger ber content would imply higher compressive
strength.
The stressstrain curves of the various specimens under single
impact (six independent impacts in all) have been presented in
Fig. 6. It is found that, with increasing the ber content, higher
strain-rate loading is required to completely fail the specimen.

90
16.7/s
21.4/s
23.7/s
30.0/s
33.6/s

1st

80
2nd

70

(MPa)

(a) Vf =0.0%

The strain rates at failure are approximately 20/s, 80/s and 90/s
for the concrete specimens containing 0.0%, 1.5% and 3.0% volume
fraction of bers. Clearly, the brous concrete specimens fail only
under higher strain-rate loading. In order to compare the mechanical response of the specimen with three ber volume fractions, the
Vf = 1.5% and Vf = 3.0% SFRC specimens are specially tested under
low-velocity impact (e_ 20/s). The stressstrain curves of berreinforced concrete specimens under the same striker velocity
and pulse shaper are shown in Fig. 9, where the strain rate in each
impact is calculated and denoted in legend. Owing to the weak
capability of impact resistance, the plain concrete specimen is
completely broken into small pieces after the rst impact (see
Figs. 6a and 7a). At the same strain-rate loading, the specimen with
Vf = 1.5% fails only at the fth impact. The specimen containing volume fraction of 3.0% can resist the ninth impact. Besides, the
decreasing rate of curve peak is well controlled by increasing ber
volume fraction. For example, in the Vf = 3.0% case, the curve peaks
of the 2nd5th impacts are even higher than the peak of the rst
impact. This phenomenon is attributed to the fact that the
hardening effect of strain-rate prevails over the effect of damage
softening. It also exhibits the excellence of brous concrete in
the resistance against repeated impact. Due to the damage

60
3rd

50
40

4th

30
5th

20
10
0

0.002

0.004

0.006

0.008

0.01

0.012

0.014

(a) Vf =1.5%
100

130
Vf=0.0%
Vf=1.5%
Vf=3.0%

3rd 1st

5th
8th

70

110
100

17.3/s
16.4/s
17.0/s
18.5/s
17.9/s
19.9/s
19.1/s
25.4/s
29.1/s

4th

80

(MPa)

120

Strength (MPa)

2nd

90

9th
7th

60
50
6th

40

90

30
20

80

10

70
0

60
-6

-5

-4

-3

-2

-1

Logarithm of strain rate

Fig. 8. Effect of strain-rate on concrete strength.

8
-3

x 10

(b) Vf =3.0%
Fig. 9. Stressstrain curves for SFRC under low strain-rate repeated impact.

1290

Z.L. Wang et al. / Composites: Part B 42 (2011) 12851290

Fig. 10. Comparison of concrete toughness under each dynamic impact.

accumulation, the strain rate increases with the increase of impact

(see Fig. 9).
Fig. 10 shows the toughness energy of the brous specimens
accumulated during the impact process. For the Vf = 1.5% specimen,
the values corresponding to ve impacts are 0.152, 0.244, 0.310,
0.279 and 0.301 MPa, respectively. The Vf = 3.0% specimen has nine
toughness values in all: 0.171, 0.150, 0.137, 0.168, 0.148, 0.176,
0.212, 0.258 and 0.322 MPa, respectively. However, the plain concrete can only experience single impact. Its toughness energy is
0.355 MPa. Their total toughness values are 0.355 MPa, 1.286 MPa
and 1.742 MPa, respectively. These experimental results suggest
that the resistance of concrete against repeated impact is signicantly improved by ber reinforcement.
4. Conclusions
It is important to investigate how SFRC behaves under staticdynamic compression. Through this experimental study, the following conclusions may be drawn:
(1) Uniaxial compression tests show that the failure mode of
ber-reinforced concrete considerably changes from fragile
to ductile. Due to bridging effect of bers, the brous specimens remain essentially intact and only have some spalling
on the sides up to end of the test.
(2) The inclusion of steel bers in concrete mix denitely
improves the mechanical behaviors of concrete. Under the
same loading rate, the plain concrete probably fails by single
impact whereas the brous concrete can resist repeated
impact.
(3) The impact properties of SFRC, including dynamic strength
and toughness energy, exhibit strong strain-rate dependency. The hardening effect of strain-rate in concrete with
high ber content can match or even exceed the effect of
damage softening to a certain extent.

Acknowledgements
This study was supported by NSFC_the Civil Aeronautics Joint
Research Foundation (60776821), the Program for New Century

Excellent Talents in University (NCET-08-0525), the Specialized

Research Fund for the Doctoral Program of Higher Education
(20070358073), the Shanghai Leading Academic Discipline Project
(B308), the Kwang-Hua Fund for College of Civil Engineering
(Tongji University) and the Talent Introduction Fund of Tongji
University (0200144054; 0200144056).
References
[1] Bentur A, Mindess S. Fiber reinforced cementitious composites. Elsevier; 1990.
[2] Wang ZL, Liu YS, Shen RF. Stressstrain relationship of steel bre reinforced
concrete under dynamic compression. Constr Build Mater 2008;22(5):8119.
[3] Altun F, Haktanir T, Ari K. Effects of steel ber addition on mechanical
properties of concrete and RC beams. Constr Build Mater 2007;21(3):65461.
[4] Mindess S, Young F, Darwin D. Concrete. 2nd ed. Prentice Hall; 2003.
[5] Olivito RS, Zuccarello FA. An experimental study on the tensile strength of steel
ber reinforced concrete. Compos Part B: Eng 2010;41(3):24655.
[6] Camps G, Turatsinze A, Sellier A, Escadeillas G, Bourbon X. Steel bre
reinforcement and hydration coupled effects on concrete tensile behaviour.
Eng Fract Mech 2008;75(18):520716.
[7] Wang ZL, Wu J, Wang JG. Experimental and numerical analysis on effect of
bre aspect ratio on mechanical properties of SRFC. Constr Build Mater
2010;24(4):55965.
[8] Marar K, Eren , elik T. Relationship between impact energy and compression
toughness energy of high-strength ber-reinforced concrete. Mater Lett
2001;47(4/5):297304.
[9] Nelson PK, Li VC, Kamada T. Fracture toughness of microber reinforced
cement composites. J Mater Civil Eng 2002;14(15):38491.
[10] Paine KA, Elliott KS, Peaston CH. Flexural toughness as a measure of shear
strength and ductility of prestressed bre reinforced concrete beams. In:
Proceedings of the international congress on challenges of concrete
construction, Dundee; 2002. [p. 20012].
[11] Bhargava P, Sharma UK, Kaushik SK. Compressive stressstrain behavior of
small scale steel bre reinforced high strength concrete cylinders. J Adv Concr
Technol 2006;4(1):10921.
[12] Gray IIIGT. Classic split-Hopkinson pressure bar testing. In: Kuhn H, Medlin D,
editors. Mechanical testing and evaluation. ASM handbook, vol. 8. Materials
Park, Ohio: ASM International; 2000.
[13] Shim JM, Mohr D. Using split Hopkinson pressure bars to perform large strain
compression tests on polyurea at low, intermediate and high strain rates. I J
Impact Eng 2009;36(9):111627.
[14] Brown ET. ISRM suggested methods: rock characterization, testing and
monitoring. Oxford: Pergamon; 1981.
[15] Graff KF. Wave motion in elastic solids. NY Dover Publication; 1991.
[16] Wang ZL, Wu LP, Wang JG. A study of constitutive relation and dynamic failure
for SFRC in compression. Constr Build Mater 2010;24(8):135863.
[17] Li WM, Xu JY. Mechanical properties of basalt ber reinforced geopolymeric
concrete under impact loading. Mater Sci Eng: A 2009;505(1/2):17886.
[18] Rong ZD, Sun W, Zhang YS. Dynamic compression behavior of ultra-high
performance cement based composites. I J Impact Eng 2010;37(5):51520.