Construction and Building Materials 147 (2017) 194–202

Contents lists available at ScienceDirect

Construction and Building Materials

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

Experimental study on dynamic mechanical properties of the basalt fiber

reinforced concrete after the freeze-thaw based on the digital image
correlation method
Yan-Ru Zhao a, Lei Wang a, Zhen-Kun Lei b, Xiao-Feng Han a, Yong-Ming Xing c,⇑
a
School of Civil Engineering, Inner Mongolia University of Technology, Hohhot 010051, China
b
State Key Laboratory of Structural Analysis for Industrial Equipment, Dalian University of Technology, Dalian 116024, China
c
Department of Applied Mechanics, Inner Mongolia University of Technology, Hohhot 010051, China

h i g h l i g h t s

This paper studies the effect of basalt fibers and FTCs on the concrete impact resistance.

Adding fibers could significantly improve the impact resistance of the concrete.

The FTCs made the basalt fibers lose its enhancement effect.

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

Article history: This paper studies the effect of the basalt fibers and the freeze thaw cycles on the concrete impact failure
Received 19 September 2016 mechanism through the concrete beam fall ball impact test. The experiment used the digital image cor-
Received in revised form 23 December 2016 relation method to measure the full field strains and then make a real-time analysis of them. Based on the
Accepted 23 February 2017
number of impacts and the strain curves of the starting crack points, the impact times respectively for
initial cracking and eventual damaging were determined. The test results show that with the basalt fiber
content increasing, the impact times for the concrete specimens’ initial cracking and final failure were
Keywords:
also raised because adding fibers formed a three-dimensional system which could enlarge the transmis-
Dynamic impact
Freeze-thaw cycles
sion range of the impact stress waves and thus improved the concrete’s elastic deformation performance,
Fiber concrete that is, improved the concrete’s initial cracking capability to impact resistance. On the other hand, as the
Full-field strain number of freeze-thaw cycles increased, the impact times for the initial crack and the final failure
Digital image correlation (DIC) declined in that the freeze-thaw damage reduced the elastic deformation performance of the concrete
specimens. The impact energy in the early freeze-thaw period lost the fastest and the most, indicating
that the freeze-thaw cycles would produce damage and defect in the fiber/matrix interface and therefore
made the fiber’s three-dimensional system gradually lose its enhancement effect. The experimental study
provides a new approach to understanding the effect of the fibers and the freeze-thaw cycles on the
impact resistance performance of the concrete.
Ó 2017 Published by Elsevier Ltd.

1. Introduction
The fiber reinforced concrete (FRC) has been developing rapidly
in recent years as a new type of cement base composite and has
been successfully applied in the field of military works, water con-
servancies, construction of buildings, airports, and high ways
owing to its excellent tensile an bending strength, crack and
shrinkage resistance, impact resistance, anti-permeability, and
⇑ Corresponding author.
E-mail addresses: zhaoyanru710523@126.com (Y.-R. Zhao), xym@imut.edu.cn
(Y.-M. Xing).
frost resistance. Fiber reinforced materials include steel fibers, syn-
thetic fibers, basalt fibers, carbon fibers, hybrid fibers, etc.
Researchers have conducted some tests to study the dynamic
mechanical properties of various FRC materials under room tem-
perature [1–6] and have found that fibers can effectively improve
the brittleness of the concrete and thus improve its resistance to
the impact load. However, in cold regions the concrete members
are not only affected by such dynamic loadings as impact, vibration
and collision, but also affected by the loading of freeze-thaw cycles
[7], which may lead to the change of the internal structure of the
material and the decline of its mechanical performance index
[8–9]. Therefore, when calculating the structure design and

http://dx.doi.org/10.1016/j.conbuildmat.2017.02.133
0950-0618/Ó 2017 Published by Elsevier Ltd.
 Y.-R. Zhao et al. / Construction and Building Materials 147 (2017) 194–202
analyzing the damage effect, the freeze-thaw damage degradation
effect has to be considered apart from the impact damage effect of
the dynamic loadings; otherwise the safety problems may arise by
overestimating the bearing capacity of the material. In addition, to
get familiar with the dynamic response properties of the FRC under
freeze-thaw conditions is of great importance for evaluating the
structure damage repair and expanding its application to a much
wider range.
At present, further research needs to be done to represent the
real-time mechanical performance and micro-macroscopic failure
mechanism of the FRC under the freeze-thaw environment. As to
the mechanical performance test under the freeze-thaw cycles,
the material deformation test is one of the most difficult. The tra-
ditional strain gauge method is only able to measure the single
direction deformation and the displacement at specific directions,
but fails to attain the full-field strain information of the sample
surface, and the strain gauge is also easy to fail under freeze-
thaw cycles. The traditional fall ball impact test defines the initial
crack impact times as the impact times that make the specimen
produce microfractures when the strain gauge is placed at the bot-
tom of the specimen, namely, the impact times when the concrete
strain mutation occurs at the specimen’s bottom [10]. However, it
has been found in the experiment that after some specimens are
subject to impacts, the strain value of the gauge gradually
increases but the mutation does not occur even when the specimen
is completely damaged. As a consequence, the number of the initial
crack impact times cannot be measured. This is because, on the one
hand, the crack position cannot be accurately predicted, so the
crack does not go through from where the strain gauge is placed,
and on the other hand, the freeze-thaw and the impact lead to
the decline of the strain gauge bonding strength so the deforma-
tion cannot be accurately transferred to the gauge.
The digital image correlation method (DICM) uses modern opti-
cal technique to measure the mechanical full-field deformation
with non-contact and high precision. It has been widely applied
to engineering due to its simple operation, low demand on the
specimen surface treatment, automatic and precise measurement
[11–13]. In recent years, some researchers have used the DICM
to study the mechanical properties of concrete and rock and have
gained some achievements. Literature [14] applied the DICM to
test the deformation in the concrete fatigue experiment and
obtained the distribution of the displacement field and the strain
field under the fatigue load, and therefore predicted the fatigue
cracking up location and made a qualitative judgment of the per-
formance degradation of the concrete specimen under the fatigue
load. Literature [15] used DICM to study the strain field of the con-
crete precast crack tip region, observed the crack propagation pro-
cess and area, and calculated the fracture toughness of the
material. Literature [16], by means of DICM, measured the crack
tip evolution process of the three-point concrete bending beam
and obtained the crack tip displacement field and strain field of
the specimen under different loads. Literature [17,18] attempted
to use DICM to study the deformation of rocks with microscopic
cracks, and found out that the displacement distribution obtained
was closely related to the distribution of the microscopic cracks,
showing the opening up or closing effect of the microscopic cracks
in the rock damage. The above researches reveal that the DICM
provides an effective experimental method for further study on
the material and structure fracture damage.
Based on the DICM and the three-point bending drop impact
test, this paper conducted a real-time observation of the full field
strain of the concrete beam specimen in the process of impact frac-
ture, and determined the initial crack impact times through the
changing patterns of the strain at the cracking points. It also stud-
ied the mechanical impact properties and the damage evolution
characteristics of the concrete with different basalt fiber contents
195
and different freeze-thaw cycles, and discussed how the fiber
incorporation and the freeze-thaw damage affected the concrete
impact damage
2. Experiment
2.1. Testing method
The digital image correlation method employs CCD camera to
record the respective speckle images before and after the displace-
ment or deformation of the tested object to get two digital gray
level fields by analog digital conversion and then some correlation
calculations are operated to find the extreme value points of the
correlation coefficient and obtain the corresponding displacement
or deformation [16].
In the correlation computation, the gray level of the pixel con-
tains the displacement information, so the correlation method is
based on the gray level matching. Hence, it is necessary to set up
a mathematical model to measure the image matching degree
before and after the deformation. The formula is as follows.
Rðf i ðx; yÞ  f Þ  ðg i ðx ; y Þ  gÞ
C ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð1Þ
2
Rðf i ðx; yÞ  f Þ  Rðg i ðx ; y Þ  gÞ2
Here, C is the correlation coefficient; f (x, y) and g (x⁄ y⁄) repre-
sent the two gray level image subareas before and after the defor-
mation. Due to the random speckle distribution, the spots
distribution in the small area around each point of the object is dif-
ferent and this small area is usually referred to as the subarea. By
searching the peak value of the correlation coefficient C we can
determine the displacement both in direction X and direction Y
for the target point and further obtain the strain change by numer-
ical differential operation. In actual operation, the closer the corre-
lation coefficient is to 1, the more reliable the displacement
computed; otherwise, the closer the correlation coefficient is to
0, the greater the calculation error for the displacement.
2.2. Specimen preparation
The strength of the standard concrete used in the test is C30,
whose mix proportion is listed in Table 1. The size of the specimen
is 40 mm  40 mm  160 mm. In order to study the strengthening
and toughening effects of the basalt fiber on the concrete, with the
standard concrete mix proportion unchanged, different fiber con-
tents were added into the concrete, respectively, 0.0 kg/m3,
1.0 kg/m3, 1.5 kg/m3, 2.0 kg/m3 and 2.5 kg/m3. The physical and
mechanical properties of the basalt fiber are illustrated in Table 2.
In the process of specimen preparation, the dry mix of sand,
stone, cement, and basalt fibers was put into the mixer to stir for
3–5 min so that the fibers could mix well with the cement, sand
and stone. Then water was added, stirring again for a while. In
order to avoid the continuing increase of the intensity during the
freeze-thaw caused by the short curing time, the experiment
adopted the standard long time curing, namely, 120 days.
2.3. Testing method
The freeze-thaw cycling test used the ‘‘quick freeze method”
specified by the national standard GB/T 50082-2009 [19] to freeze
and thaw the specimen. The actual time for one freeze-thaw cycle
lasted 3.5 h and the maximum and minimum central temperature
of the specimen were respectively controlled at 8 ± 2 °C and
17 ± 2 °C. The fall ball impact test was conducted after the spec-
imens had respectively undergone 15, 30, 45, 60 and 75 freeze-
thaw cycles.
196 Y.-R. Zhao et al. / Construction and Building Materials 147 (2017) 194–202

Table 1
Mix proportion of the concrete.

Cement (kg/m3) Coarse aggregate (kg/m3) Sand (kg/m3) Water (kg/m3) Water/cement W/C
385.42 1134.34 695.24 185 0.48

Table 2
Physical and mechanical properties of the basalt fiber.

Material Single fiber length (mm) Fiber diameter (lm) Density (kg/m3) Fracture ductility rate (%) Elasticity modulus (GPa) Tensile strength (MPa)
Basalt fiber 18 15 2650 2.4–3.0 95–115 3300–4500

The fall ball impact test used the self-designed three-point
bending device, with reference to the weight drop method [20]
put forth by the 544 committee of the American Concrete Institute
(ACI). The drop hammer was a solid steel ball which weighs 63 g
and whose impact height was 500 mm. The two ends of the beam
were supported plainly with the clear span of 130 mm. A
40 mm  40 mm steel plate was place on the impacted surface of
the specimen. The fall ball dropped freely from the stipulated
height to hammer the specimen, thus completing an impact cycle.
At the same time, the test also used the DIC method, that is, after
every fall ball impact, when the specimen became stable, the
CCD camera would take a record of it with manual control until
it was damaged. The whole system block diagram is shown in
Fig. 1.
The DIC test used 2 CCD cameras (supplement model) with 2
million pixels and the incandescent lamp was used as the fill light
to avoid the influence that may be caused by the frequent blinking
of the alternating lamp. In order to make the DIC results more
accurate, before the test, the surface of the specimen that faced
the camera lens needs speckle spraying. After adjusting the focal
length and the needed light, the size and location of the specimen
was calibrated by the calibration target whose size and shape had
been recorded in the computer. Then the image acquisition system
collected the images of the impact failure process. The speckle
images were then put in the vic-3D software to compute the dis-
placement and the strain.
3. Results of the test
3.1. Full-field strain
3.1.1. Specimen without basalt fibers
Fig. 2 demonstrated the full field cloud image of the horizontal
strain exx acquired by the DIC method which changed with the
number of impact times. At the beginning period of the loading,
the full field strain was small, and the specimen’s bottom edge dis-
played a strain concentration area which was larger than the other
areas, as is shown in Fig. 2(a) and (b); When the number of impact
times gradually rose to 47, apparent strain clusters appeared at the
bottom of the specimen (Fig. 2(c)); subsequently, cracks occurred
and then extended upward because of instability (Fig. 2
(d) and (e)) until the specimen fractured completely (Fig. 2(f)).
In the process of the impact failure, the maximum strain con-
centration appeared at the crack tip, whose movement trace (the
red area) was almost identical with the direction of the main crack.
The area subject to the strain concentration was reduced radiantly
from the crack tip to the surrounding area and became larger with
the increase of the impact times. Hence, the impact failure process
for the specimen with no basalt fibers can be described as follows.
An increased area of the horizontal tensile strain occurred on the
bottom part of the specimen and then it extended upward and
cracks came up which extended because of instability, and finally
the specimen got fractured.

Preparing concrete specimens

with different basalt fiber
contents and then having them
undergo the freeze-thaw cycle
test
DIC testing system
Light

Fall ball

Image card
Computer
DIC testing area
Light CCD camera
Fall ball impact test

Fig. 1. Scheme of setup.

 Y.-R. Zhao et al. / Construction and Building Materials 147 (2017) 194–202 197

fall ball impact points

‰ ‰ ‰ ‰ ‰ ‰
(a) 5.850
(b) 7.063
(c) 3.520
(d) 5.230
(e) 38.40
(f) 35.00

4.325 5.600 2.415 3.075 28.70 25.40

2.800 3.650 1.450 2.500 19.00 5.400

1.275 1.700 0.175 1.125 9.300 4.300

-0.250 -0.250 -0.140 -0.250 -0.400 -3.400

strain concentration area on the bottom cracking points cracks

Fig. 2. The full field strain ex of the impact failure Impact times: (a) 23, (b) 37, (c) 47, (d) 54, (e) 58, (f) 60.

3.1.2. Specimens with basalt fibers

For the specimens with 2.0 kg/m3 basalt fibers (without freez- 60
ing and thawing), the evolution of the cloud images for the hori-
zontal strain exx, which varied with the number of impact times, 50
is presented in Fig. 3. Like the specimen without fibers, the basalt

Impact times ni
fiber specimens (Fig. 3) experienced a similar impact failure pro- 40 60

cess: the strain concentration area at the bottom of the specimen 50

Impact times ni
became larger with the increase of the number of impacted times 30 40
(Fig. 3(b) and (c)), and then the crack occurred and extended 30 47 times
upward due to instability until the specimen fractured completely 20 20
(Fig. 3(d)–(f)). 10
10
However, there were also some differences. For the fiber speci- 0
men, the strain concentration area not only appeared at the bottom 0.0 0.5 1.0
0
part but also appeared around the impact point at the top. The two Strain (%)
areas approached to each other with the increase of the impact 0.0 0.5 1.0
times and then became connected to form a damage zone (Fig. 3 Strain (%)
(b) and (c)), and eventually, cracking appeared at the bottom of
the specimen and then extended upward (Fig. 3(d)–(f)). Fig. 4. Relationship of the impact times ni and the strain exx for the specimen with
no fibers.

3.2. The critical impact times and the impact times for final failure
Using the DIC method the strain of any point within the range of
observation could be computed, so the initiation and development
of the cracks could be studied through the full-field cloud image
where the horizontal strain exx changed under different loadings.
During the course of the impact damage of the specimens with
no fibers, the developing relationship between the maximum hor-
izontal strain exx at the bottom (i.e., the initial cracking point) and
the impact times ni was illustrated in Fig. 4. It can be seen that the
strain curve could be divided into two stages: the first stage was
the elastic deformation zone, where the strain value changed little
and was on a linear rise with the increase of the impact times; the
second stage was the fracture transition zone, and with the num-
ber of impact times increasing to a critical value, the strain
increased dramatically, indicating that the cracks extended after
crack initiation until the material was completely fractured.
In order to accurately determine the number of impact times for
the initial cracking, we could observe and analyze the changing
pattern of the horizontal strain field at the maximum strain point
(the initial cracking point) on the bottom edge of the specimen

fall ball impact point impact point strain concentration area impact damage zone impact fracture zone
‰ ‰ ‰ ‰ ‰ ‰
(a) 1.899 (b) 0.416 (c) 1.138 (d) 2.520 (e) 23.368 (f) 30.80

1.505 0.010 0.470 0.565 18.550 22.70

0.980 -0.620 -0.420 0.410 12.100 14.60

0.455 -1.250 -1.321 0.356 5.450 6.50

0.061 -1.723 -2.577 -0.156 0.013 -1.60

crack
strain concentration area on cracking point
the specimen bottom
Fig. 3. The full field strain ex of the impact failure with 2.0 kg/m3 of fiber content Impact times: (a) 40, (b) 76, (c) 150, (d) 156, (e) 182, (f) 188.
198 Y.-R. Zhao et al. / Construction and Building Materials 147 (2017) 194–202

by using DIC method. As is shown in Fig. 4, piecewise linear fitting 200

was conducted to the two stages of the elastic deformation zone n0
and the fracture transition zone to make the angular average line
W
intersect with the curves, and the coordinate value of the intersec-
tion point in each phase was taken as the initial crack strain and 150

the critical impact times [21]. It could be seen from Fig. 4 that
the initial crack stain e0 of the concrete specimens with no fibers
and no freeze-thaw was 0.14%, and the critical impact times n0
was 47, whereas the impact times for the final failure W was 57.
Therefore, by analyzing the curves of the horizontal strain ex of
the cracking point at the bottom edge which varied with the
impact times, we could analyze the effect of the fiber content
and the freeze-thaw cycles on the concrete’s impact deformation
properties.
3.2.1. The effect of the fiber content
For specimens with different amounts of fiber, the curves of the
horizontal strain exx at the bottom crack point corresponding with
the impact times were shown in Fig. 5(a), which indicated that
adding fibers could significantly improve the impact resistance of
the concrete. In the elastic deformation stage, increasing the fiber
content could improve the elastic deformation performance of
the concrete whereas in the fracture shifting stage, increasing the
fiber content could improve the anti-impact times.
Furthermore, by extracting the critical impact times n0 and the
impact times for final failure W in Fig. 5(a) and then mapping them
together, we could obtain the impact resistance of the specimens
with different fiber density, as is shown in Fig. 6. It could be found
that both the critical impact times n0 and the impact times for final
failure W increased with the increase of the fiber density from 0 to
2.0 kg/m3, then decreased for fiber density of 2.5 kg/m3, which
indicated that a certain amount of fiber content could improve
the impact resistance of the concrete, or more specific, the cracking
resistance and the damaging resistance, but if the fibers density is
2.5 kg/m3, it woud reduce the impact resistance of the concrete. It
was also found that the ratio between the numbers of impact times
causing the initial crack and that leading to the final failure did not
differentiated much. When the fiber contents were 1.0 kg/m3,
1.5 kg/m3, 2.0 kg/m3 and 2.5 kg/m3, the ratios of the impact times
for the initial cracking and that for the ultimate damaging (n0/W)
were respectively 86%, 75%, 85%, and 75%, which suggested that
adding the basalt fibers mainly improved the impact resistance
to the initial crack of the concrete. Once the specimen has cracked,
the basalt fiber does not have much to do to resist the macroscopic
crack propagation.
Impact times ni
100
50
0.0 0.5 1.0 1.5 2.0 2.5
Fiber density (kg/m3)
Fig. 6. The anti-impact property under different fiber density.
3.2.2. The effect of the freeze-thaw cycles
Taking into account the effects of the freeze-thaw cycles on the
impact resistance of the concrete, specimens without fibers were
selected to be frozen and thawed 0, 15, 30, and 60 times respec-
tively. Fig. 7 showed the relationship between the number of
impact times and the strain at the cracking point under the respec-
tive freeze-thaw cycles. It could be seen from Fig. 7 that the freeze-
thaw reduced the elastic deformation capability of the concrete in
the first phase. When the specimen was frozen and thawed 60
times, the slope of the strain curve in the first phase was only
20% of that of the specimen with no freeze-thaw. In addition, with
the increase of the number of freeze-thaw cycles, the impact times
for final failure came to decrease, but the strain limit increased sig-
nificantly; under the same impact energy, the concrete after the
freeze-thaw deformed to a much larger degree, which indicated
that the freeze-thaw cycles effected much damage to the concrete
and therefore reduced its impact resistance.
Fig. 8 showed the relationship of the impact times and the
strain at the cracking point for the specimen with 2.0 kg/m3 fiber
content after different freeze-thaw cycles. Similar to the specimen
without doped fiber, with the increase of the freeze-thaw cycles,
the elastic deformation performance in the first stage was reduced,
and the number of impact times causing the ultimate damage also
declined, which signified that the freeze-thaw weakened the rein-
forcement of the basalt fiber and thus degraded the impact resis-
tance of the concrete. In addition, under the same freeze-thaw

60
160
Freeze-thaw times
140 50
0 15
120 30 60
40
Impact times ni
Impact times ni

100
30
80

60 20

40 3 3
0.0kg/m 1.0kg/m 10
20 3 3
1.5kg/m 2.0kg/m
3 0
0 2.5kg/m

0.0 0.5 1.0 1.5 2.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
Strain (%) Strain (%)

Fig. 5. Relationship of the impact times ni and the strain ex for the specimens with Fig. 7. Relationship of the impact times ni and the strain ex for the non-fiber
different fiber contents. specimens under different freeze-thaw cycles.
 Y.-R. Zhao et al. / Construction and Building Materials 147 (2017) 194–202 199

170 2.0-0
160
150
140
130
2.0-15
120
110

Impact times ni
100
90
80
70 2.0-30
60
50
40 0.0-0 2.0-45 2.0-60 2.0-75
30
20
10
0 0.0-15 0.0-30 0.0-60
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
Strain (%)

Fig. 8. The full field horizontal strain of the specimens with 0.0 kg/m3 and 2.0 kg/m3 fiber density under different freeze-thaw cycles (the number before‘‘-”refers to the fiber
content and the number after‘‘-”refers to the freeze-thaw cycles).

cycles, the ultimate damaging impact times of the fiber concrete
were larger than those of the non-fiber concrete, suggesting that
under freezing and thawing, the doped fibers still played their rein-
forcement role.
Further, by extracting the critical impact times n0 and the
impact times for final failure W in Fig. 8 and then graphing them
in Fig. 9, a histogram could be attained to illustrate the specimens’
capability for impact resistance with different fiber contents and
under different freeze-thaw cycles. As was shown, with the same
fiber content, n0 and W decreased as the freeze-thaw cycles
increased, which meant the freeze-thaw cycles would reduce the
impact resistance of the concrete. On the other hand, under the
same freeze-thaw cycles (<30), n0 and W reached their peak value
when the fiber density was 2.0 kg/m3, which suggested that a cer-
tain number of fibers could improve the concrete’s impact resis-
tance of the cracking resistance and failure resistance; however,
if the freeze-thaw cycles were beyond 30, the fibers functioned lit-
tle for the improvement of the concrete’s impact resistance.
200
Initial cracking
Failure
150
Impact times
4. Discussion
4.1. The impact energy increase rate of the fiber content
From Fig. 9 it could be seen that when the freeze-thaw cycles
(45) were identical, the impact times of the fiber concrete for
cracking and damaging were increased, compared with those of
the non-fiber concrete. In order to further study the influence of
the fiber content on the concrete’s anti-impact mechanical prop-
erty, the impact energy increase rate of the fiber content could
be defined as follows, in condition that the freeze-thaw cycles
remained the same.
Wm
cm ¼  100% ð2Þ
W0
In the above formula, cm is the increase rate (%) when the fiber
content is m, referring to the improvement of impact resistance of
the fiber concrete compared with the non-fiber one. W0 and Wm
correspond to the number of impact times leading to failure when
the fiber content is 0 and m (kg/m3).
3
The relationship between the impact energy increase rate and
A--0.0kg/m the fiber content was demonstrated in Fig. 10. When the number
B--1.0kg/m
3
of freezing and thawing cycles is small, the effect of fiber incorpo-
3
C--1.5kg/m ration on the impact energy of the concrete is significant, and the
3
D--2.0kg/m value of c first increases and then decreases. When the fiber con-

tent reaches 2.0 kg/m3, c reaches its maximum value. With the
3
E-- 2.5kg/m
increase of freezing and thawing cycles, c value does not have
100
much fluctuation, indicating that the effect of the fiber incorpora-
tion on the increase rate of the impact energy is gradually becom-
ing smaller, and the freeze-thaw damage significantly weakens the
50 fiber enhancement.

4.2. The freeze-thaw loss rate and loss speed of the impact energy
A BC DE A BC DE A BC DE A BC DE A BC DE A BC DE
0 In order to further analyze how the freezing and thawing
0 15 30 45 60 75
affected the concrete’s mechanical property for impact resistance,
Freeze-thaw times
when the concrete was doped with the same amount of fibers,
Fig. 9. The impact resistance of the basalt fiber specimen under different freeze- the freeze-thaw loss rate of the impact energy could be defined
thaw cycles. as follows:
200 Y.-R. Zhao et al. / Construction and Building Materials 147 (2017) 194–202

2.0 Freeze-thaw times

80
0 15
30 45
1.5 60 75
60

1.0
40

0.5
20 3 3
0.0kg/m 1.0kg/m
3 3
1.5kg/m 2.0kg/m
0.0 3
2.5kg/m
1.0 1.5 2.0 2.5 0
Fiber density (%) 0 15 30 45 60 75
Freeze-thaw times
Fig. 10. The impact energy increase rate of the specimens with different fiber
content.
Fig. 11. The impact energy loss rate under different freeze-thaw cycles.

W0  Wi
ai ¼  100% ð3Þ
W0 5
In the formula, ai refers to the freeze-thaw loss rate of the 0.0kg/m
3

impact energy (%) when the concrete was frozen and thawed i 4 1.0kg/m
3

times, or in other words, it means the impact energy loss degree 1.5kg/m
3

of the frozen-thawed concrete compared with the concrete with- 2.0kg/m

3

3 3
out freezing and thawing. The greater the impact energy lost rate, 2.5kg/m
the more damage the freeze-thaw cycles do to the concrete. W0
and Wi respectively refer to the damaging impact times under 0 2
and i freeze-thaw cycles.
Moreover, with the fiber content unchanging, the freeze-thaw
1
loss (decrease) speed of the impact energy could be described in
the following formula
0
W ði15Þ  W i W ði15Þ  W i
bi ¼ ¼ ; ð4Þ
i  ði  15Þ 15 0 15 30 45 60 75

among which, bi is the loss speed of i cycles of freeze-thaw, signify-
ing the degree of the anti-impact energy loss of the fiber reinforced
concrete per freeze-thaw cycle. The impact energy lost speed is
used to indicate how fast the freeze-thaw cycles weaken the impact
resistance energy of the concrete. W(i-15) is the damaging impact
times when the freeze-thaw cycles range from i to 15. The i refers
to the number of freeze-thaw cycles which valued 15, 30, 45, 60
and 75 in the experiment.
Fig. 11 showed the relationship between the impact rate loss
rate and the freeze-thaw cycles, from which it could be found that
with the increase of the freeze-thaw times, the anti-impact energy
loss rate of the concrete, that is, the impact energy loss degree, was
on increase. The relationship between the loss speed of the anti-
impact energy and the freeze-thaw cycles was illustrated in
Fig. 12, which revealed that as the freeze-thaw cycles increased,
the energy loss speed went up and then fell down, in other words,
in the beginning stage of freezing and thawing, the impact energy
loss speed accelerated whereas in the later stage it slowed down.
According to the characteristics revealed in Fig. 12, the process
of energy loss of the fiber concrete under freezing and thawing
cycles could be divided into three stages, namely, the initial stage,
the middle stage and the final stage. In the initial stage of freezing
and thawing, the loss speed was the fastest and the loss rate the
highest. During this stage (the freeze-thaw cycles < 30), the loss
speed took on a quick rising trend (Fig. 12) and the loss rate was
rapidly raised to 53–64% (Fig. 11). In the middle stage (30< the
freeze-thaw cycles <60), the loss speed came to slow down
(Fig. 12) and the loss rate slowly rose to 62–89% (Fig. 11). In the
final stage (60< the freeze-thaw cycles <75), the loss speed kept
low and stable and the loss rate had reached around 90% (Fig. 11).
Freeze-thaw times
Fig. 12. The impact energy loss speed under different freeze-thaw cycles.
4.3. The microscopic reinforcement mechanism of the fiber
The impact load is transmitted in the form of stress wave in the
concrete. In the process of impact, if the stress wave keeps trans-
mitting forward without any transference and remains the acute
crest, only a small volume of the material bears the stress and
therefore brittle fracture occurs when the stress wave is strong
enough. When the basalt fibers are mixed into the concrete, a three
dimensional supporting system would be formed since the fibers
distribute randomly and disorderly. As a result, under the external
force, the fibers are able to have some loose movement and help
the surrounding matrix to sustain the stress, and thus the trans-
mission area of the stress wave becomes larger and larger, which
means only a much stronger stress can make the material fracture,
that is, the tenacity of the concrete has been enhanced. In the full
field strain analysis for the fiber concrete specimens (Fig. 3), it
could be observed that there formed the impact damage zone in
the basalt fiber concrete, which indicated that the microscopic
enhancement mechanism of the fiber in the process of impact
could be attributed to the three dimensional system produced by
the fibers which made the impact stress transmission extend to a
much larger area, and consequently dispersed the impact strain
in a certain range, consumed the impact energy of the fall ball,
and improved the anti-impact times of the specimen.
In addition, in the process of concrete making, the three dimen-
sional supporting system formed by the randomly distributed
 Y.-R. Zhao et al. / Construction and Building Materials 147 (2017) 194–202
fibers in the concrete plays a supporting role which reduces the
sinking of the aggregate and increases the homogeneity of the
aggregate distribution, and thus improves the internal quality of
the concrete such as its homogeneity. Consequently, the strain
concentration is less likely to occur which to a certain extent pre-
vents the occurrence of micro-cracking in the concrete, reduces the
sources and sizes of the micro-cracks. Therefore, with the increase
of the fiber content, the impact resistance performance of the con-
crete is also gradually improved.
In terms of the modulus of elasticity, fibers can be divided into
rigid and flexible fibers. The rigid fiber has a larger elasticity mod-
ulus and a stronger stiffness, so it has a high capability for load
transmission, such as the steel fiber; on the contrary, flexible fibers
have a smaller elasticity modulus and low stiffness, so they are
poor in load transmission but their elongation limit and deforma-
tion resistance behavior are more advantageous, for example, the
Dura fiber, the nylon fiber and the basalt fiber. When the macro-
scopic crack propagates, the rigid fiber undergoes pullout failure
in the cracking area which can provide a strong drawing resistance
or ‘‘bridge” effect to inhibit the macroscopic crack from opening up
and extending; as to the flexible fibers in the cracking area, they
experience pullout and broken failure under the instantaneous
impact, which cannot offer enough drawing resistance to prevent
the opening up and extending of the macroscopic crack. Therefore,
when the basalt fiber content increases, the impact times from the
initial crack to the final failure do not have a significant increase.
In the test, for a particular fiber density, the best performance in
terms of first crack as well as failure impact resistance of the basalt
fiber concrete is given by concrete with 2.0 kg/m3 fiber density,
rather than 2.5 kg/m3 fiber density, which can be explained as fol-
lows: if fibers are made to fully play their enhancing and toughen-
ing role in the concrete, it is necessary for each single fiber to be
wrapped with a certain thickness of the concrete, for only in this
way can the fiber and the cement base be made intensively inte-
grated (connected), making them jointly bear the force. In other
words, an interface with certain strength is needed between the
fiber and the concrete matrix. If excessive fibers are added into
the concrete, it is difficult to mix the concrete homogeneously
and fiber clusters are likely to occur, leading to insufficient bonding
stress and strength because there is no cement between fibers or
the cement layers are too thin. In other words, there is not enough
interface strength, which eventually makes some fibers unable to
play their role, or ‘‘fail”. These ‘‘failed” fibers not only cannot
improve the various performances of the cement matrix but also
may become the defects of the concrete due to their lack of bond-
ing capacity. Therefore, excessive amounts of fibers will yield
adverse effect to the concrete and thus reduce its impact resistance
property.
To sum up, the mechanism of the basalt fiber’s reinforcement
effect on the concrete’s impact resistance lies in two aspects. On
the one hand, the spatial 3-D system resulted from the mixing of
the basalt fibers enables the impact stress waves to transmit to a
much wider area; on the other hand, the fibers can improve the ini-
tial defects of the concrete matrix.
4.4. The microscopic damage mechanism of the freeze-thaw cycle
When the saturated concrete is undergoing the freeze-thaw
cycles, the water in the concrete’s micro-pores under the interac-
tion of plus and minus temperatures, tends to form the fatigue
stress by the joint action of the ice expansion pressure and the
seepage pressure. When the fatigue stress rises to equal the tensile
strength of the concrete, microscopic cracks take place inside the
material. With the increase of the freeze-thaw cycles, the cracks
will run through with one another, resulting in the decrease of
the anti-impact property of the basalt fiber concrete.
201
After the concrete is mixed with the basalt fibers, an interface
with some strength between the fiber and the matrix is needed
to transfer the stress (strain), but the fatigue stress from the joint
action of the ice expansion pressure and the seepage pressure in
the freeze-thaw cycles makes not only the concrete matrix but also
the interface crack. As the freeze-thaw cycles increase, the interfa-
cial damage becomes severe and the interfacial strength comes to
decline, so it cannot transmit the stress any more, making the three
dimensional system gradually fail to work after 30 times of freez-
ing and thawing.
During freezing and thawing, the fibers’ reinforcement effects
on the properties of the concrete matrix keeps on declining until
the number reached 75, when the reinforcement of the fibers
totally disappeared. The spatial 3-D system produced by mixing
fibers can expand the transmission range of the impact stress
waves, but its effect comes to decline and after 30 freeze-thaw
cycles, it failed to work. In other words, before freeing and thawing
30 times, the fibers play two enhancement roles at the same time
though they gradually decrease, but after that, the fibers only have
one role left which continued to decrease. Therefore, in the initial
stage (0–30), the impact energy of the specimen lost the most
and the fastest (Fig. 11 and 12) because of the high porosity and
low strength of the interface, so under the freezethaw damage,
the interface subjected the damage prior to the matrix.
5. Conclusion
By combining the DIC method and the fall ball impact test,
and by using the real time observation to the characteristics of
the full field strain and the evolution of the cracking points in
the process of impact fracture of the concrete specimen beam,
the research studied the effect of the basalt fibers and the
freeze-thaw cycles on the concrete’s mechanical performance
of impact resistance. It was found that the full field strain of
the specimens with different fiber contents could reflect the
formation and evolution of the impact damage zone. The spatial
3-D system formed by the fibers could help to expand the trans-
mission range of the impact stress wave and therefore could
improve the impact resistance performance of the concrete.
However, the freeze-thaw cycles would result in damage and
flaw at the fiber/matrix interface which made the fiber’s 3-D sys-
tem gradually lose its enhancement.
Meanwhile, the research used the horizontal strain exx at the
maximum strain point (the starting cracking point) at the speci-
men’s bottom edge to study the initial impact cracking and the
crack propagation. It wass found that the impact times and the
strain exx curves could be divided into two phases, namely, the
elastic deformation phase and the ductile fracture phase. The elas-
tic deformation phase accounted for about 80% of the total impact
life span. By adding the basalt fibers, the anti-impact elastic defor-
mation property of the concrete could be improved, that is, the
concrete’s crack initiation capacity for impact resistance was
improved. On the other hand, the freeze-thaw cycles weakened
the performance of the concrete in the elastic phase and therefore
reduced the anti-impact property of the concrete.
In addition, the impact energy loss during the freeze-thaw
cycles could be divided into three stages: initial, middle and final.
The loss speed of the impact energy in the initial stage rose very
fast and the loss rate could amount to 53–64%, which indicated
that in the early freeze-thaw period, the impact energy loss of
the concrete specimen is at the fastest speed and highest rate. In
the middle stage, the loss speed fell down quickly and the loss rate
was up to 62–89%. In the final stage, the loss speed kept slow and
stable and the loss rate reached approximately 90%.
202 Y.-R. Zhao et al. / Construction and Building Materials 147 (2017) 194–202
Acknowledgments
The authors are grateful for support from the National Natural
Science Foundation of China - China (Nos. 11362013 and
11062007).
References
[1] N.H. Yi, J.H.J. Kim, T.S. Han, et al., Blast-resistant characteristics of ultra-high
strength concrete and reactive powder concrete, Constr. Build. Mater. 28
(2012) 694–707.
[2] J. Xu, W. Li, F. Fan, et al., Experimental study on impact properties of carbon
fiber reinforced geopolymeric concrete using a SHPB, J. Build. Mater. CN. 13 (4)
(2010) 435–439.
[3] X. Ren, T. Zhou, F. Zhong, Dynamic mechanical behavior of steel-fiber reactive
powder concrete, Explo. Shock. Waves CN. 31 (5) (2011) 540–545.
[4] W.M. Li, J.Y. Xu, Impact characterization of basalt fiber reinforced
geopolymeric concrete using a 100-mm-diameter split Hopkinson pressure
bar, Mater. Sci. Eng. A 513/514 (2009) 145–153.
[5] X. Du, G. Dou, L. Li, Experimental study on dynamic mechanical properties of
fiber reinforced high strength concrete, Eng. Mech. CN. 28 (4) (2011) 138–144.
[6] S.S. Wang, M.H. Zhang, S.T. Quek, Mechanical behavior of fiber-reinforced
high-strength concrete subjected to high strain-rate compressive loading,
Constr. Build. Mater. 31 (2012) 1–11.
[7] M. Hasan, T. Ueda, Y. Sato, Stress-strain relationship of frost-damaged concrete
subjected to fatigue loading, J. Mater. Civ. Eng. 20 (11) (2008) 37–45.
[8] H.S. Shang, Y.P. Song, Experimental study of strength and deformation of plain
concrete under biaxial compression after freezing and thawing cycles, Cem.
Concr. Res. e.36 (10) (2006) l857–l864.
[9] H.W. Micah, F.F. Seamus, W.R. Bruce, Examining the frost resistance of high
performance concrete, Constr. Build. Mater. 23 (2009) 878–888.
[10] P.P. Gu, H.N. Bant, C. Yan, Fiber reinforced wet-mix shotcrete under impact, J.
Mater. Civ. Eng. 12 (1) (2000) 81–90.
[11] H. Xie, Y. Kang, Digital image correlation technique, Opt. Lasers Eng. 65 (2015)
1–2.
[12] D.J. Wu, W.G. Mao, Y.C. Zhou, C. Lu, Digital image correlation approach to
cracking and decohesion in a brittle coating/ductile substrate system, Appl.
Surf. Sci. 257 (14) (2011) 6040–6043.
[13] F. Bernachy-Barbe, L. Gélébart, M. Bornert, J. Crépin, C. Sauder,
Characterization of SiC/SiC composites damage mechanisms using digital
image correlation at the tow scale, Compos. Part A: Appl. Sci. Manufact. 68
(2015) 101–109.
[14] Y. Dai, Z. Sun, G. Wu, X. He, Whole-field deformation measurement of concrete
structure based on digital mi age correlation, J. S.E. Uni. (Nat. Sci. Ed.) CN. 40
(4) (2010) 829–834.
[15] Z.M. Wu, H. Rong, J.J. Zheng, F. Xu, W. Dong, An experimental investigation on
the FPZ properties in concrete using digital image correlation technique, Eng.
Fract .Mech. 78 (17) (2011) 2978–2990.
[16] Q. Pu, H. Zhang, X. Dai, H. Yun, D. Zhang, Measurement of evolutionary process
of concrete crack tip by DSCM, J. XuZhou Insti. Arch. Tech. CN. 3 (9) (2009) 11–
14.
[17] Y. Zhao, H. Liang, C. Xiong, J. Fang, Analysis damage deformation of rock with
digital image technique, CHN. J. Rock. Mech. Eng. CN. 21 (1) (2002) 73–76.
[18] Z. Zhu, Y. Zhang, S. Li, W. Chen, Studies on microcosmic crack damage
properties of a red sandstone with digital image technique, CHN. J. Rock Mech.
Eng. CN. 7 (24) (2005) 1123–1128.
[19] GB/T 50082–2009, Standard for test methods of long-term performance and
durability of ordinary concrete, CHN. Constr. Indus. Press, CN, BeiJing, 2009.
[20] ACI committee 544. Measurement of properties of fiber reinforced cement, ACI
Mater. J. 85(6) (1988) 583–593.
[21] J. Wan, M. Zhou, X.S. Yang, C.Y. Dai, Y. Zhang, W.G. Mao, C. Lu, Fracture
characteristics of freestanding 8 wt%Y2O3–ZrO2 coatings by single edge
notched beam and Vickers indentation tests, Mater. Sci. Eng. A 81 (2013)
140–144.