Construction and Building Materials 152 (2017) 847–855

Contents lists available at ScienceDirect

Construction and Building Materials

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

Dynamic compressive strength of concrete damaged by fatigue

loading and freeze-thaw cycling
Jingzhou Lu ⇑, Kongfeng Zhu, Lizong Tian, Li Guo
School of Civil Engineering, Yantai University, Yantai 264005, PR China

h i g h l i g h t s

The effect of damage history on the dynamic compressive strength was investigated.

The combination of applied fatigue load and freeze-thaw cycles was considered.

Emphasis was placed on the influence of different action sequence.

A variable was used to study the combined effect of load and environment effects.

SEM was used to explain the damage mechanism from the micro perspective.

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

Article history: Uniaxial dynamic compression tests were carried out on 180 concrete prism specimens exposed to
Received 13 April 2017 freeze-thaw cycles combined with fatigue compression loading. The effects of four categories of damage
Received in revised form 15 June 2017 modes, such as fatigue compression loading (Mode F), freeze-thaw cycling (Mode FT), fatigue compres-
Accepted 4 July 2017
sion loading followed by freeze-thaw cycling (Mode F-FT), and freeze-thaw cycling followed by fatigue
compression loading (Mode FT-F), on the dynamic compressive strength of concrete were investigated
systematically. The variable Kc was introduced to characterize the combined effect of fatigue loads and
Keywords:
freeze-thaw damage, and the influence of single factor of fatigue load or freeze-thaw cycles as well as
Plain concrete
Fatigue load
the double factors with different sequence on the dynamic damage characteristics of concrete were
Freeze-thaw cycles discussed. The results showed that, when the number of fatigue cycles and the number of freeze-thaw
Dynamic compressive strength cycles remained constant, just with different sequence, the values of Kc under damage Mode F-FT were
Strain rate obviously higher than that under damage Mode FT-F at the same strain rate. The trends in damaging
Action sequence effects on compressive strength of concrete were justified based on microscopic observation of the trends
in micro crack growth under such damaging effects. In addition, concrete has apparent strain rate effect
after exposed to freeze-thaw cycles combined with fatigue compression loading. The dynamic compres-
sive strength of damaged concrete increases as the strain rate is increased.
Ó 2017 Published by Elsevier Ltd.

1. Introduction
Any kind of concrete materials and structures are subjected to
external loads while experiencing environmental effects [1]. The
hydraulic concrete structures in the severe cold regions are often
subjected to the combined action of fatigue loading and freezing-
thawing cycles. Freeze-thaw cycles and fatigue loading cycles are
both repetitive action and the joint action which may aggravate
the damage of concrete and lead to severe deterioration of the
long-term mechanical properties of concrete. This deterioration
⇑ Corresponding author.
E-mail addresses: lujingzhou@sina.com (J. Lu), 358209310@qq.com (K. Zhu),
ytutlz@163.com (L. Tian), 1263800688@qq.com (L. Guo).
can be broadly divided into the two types of failure: drastic
deterioration of the durability of concrete (i.e. permeability) and
unstable propagation of the major fatigue cracks [2]. The damage
behaviors of concrete under the single effect of fatigue load or
freeze-thaw cycle have been studied by many researchers.
Research of Soroushian et al. has indicated that the fatigue load
increases the number of micro cracks, and the freeze-thaw cycle
increases the width of micro cracks [3]. Changes in concrete prop-
erties and microstructure associated with different damaging
effects were investigated. Most of the existing researches have
studied the static damage performance of concrete under the cou-
pling of fatigue load and freeze-thaw cycles [4,5]. Some studies
were directed to the influence of load history on frost resistance
of concrete [6,8] and freeze-thaw damage on the fatigue

http://dx.doi.org/10.1016/j.conbuildmat.2017.07.046
0950-0618/Ó 2017 Published by Elsevier Ltd.
848 J. Lu et al. / Construction and Building Materials 152 (2017) 847–855
performance of concrete [9,10]. However there are few reports on
the influence of different sequence of fatigue load and freeze-thaw
cycles on the mechanical properties of concrete, especially on
dynamic properties of concrete.
In the course of using, hydraulic structures are inevitably sub-
jected to dynamic loads, such as waves, explosions and earthquakes.
Researchers have carried out a lot of useful exploration on the
dynamic characteristics of concrete after single load history or
freeze-thaw cycles. With regard to the influence of loading history,
the research shows that the load history has significant effects on
the dynamic characteristics of concrete [11–14]. Wang et al. [15]
studied the impacts of freeze-thaw and strain rate on the properties
of fully-graded concrete under uniaxial compressive stress state,
and established the unified failure criterion of fully-graded concrete
considering the strain rate and freeze-thaw cycles. Yi et al. [16]
established the viscoelastic-plastic damage model of porous asphalt
mixtures and analyzed the mechanical behavior of porous asphalt
mixture after freeze-thaw, and carried out uniaxial compression
tests at different temperatures and loading rates to verify the cor-
rectness of the model. Chen et al. [17] studied the effect of freeze-
thaw cycle on mechanical behaviors of ceramsite concrete through
split Hopkinson pressure bar (SHPB) impact test. At present, there
are few researches on the dynamic characteristics of concrete under
different action sequence of fatigue load and freeze-thaw cycles.
Due to the fact that the combined effect of load and freeze-thaw
cycles will bring great uncertainty and risks to the future safety of
concrete structure, relevant research is urgently needed.
In this paper, an experiment was conducted to see the effects of
fatigue load and freeze-thaw cycles on uniaxial dynamic characteris-
tic of concrete. Firstly, 180 concrete prism specimens were subjected
to four kinds of damage modes, namely, the fatigue loading alone,
the freeze-thaw cycle alone, the fatigue loading followed by freeze-
thaw cycle, and the freeze-thaw cycle followed by fatigue loading.
Subsequently uniaxial dynamic compression test of the damaged
specimens was conducted. The effects of the single factor of fatigue
load or freeze-thaw cycles, as well as the double factors with differ-
ent sequence, on the dynamic compressive strength of concrete were
investigated. The reasons for the deterioration of the macro mechan-
ical properties of concrete due to different damage effects were
investigated from the micro perspective. The research results of this
paper can provide experimental data and mechanical properties for
the design of hydraulic concrete structures in cold regions.
2. Experimental details
2.1. Materials and mix proportions
Chinese standard 42.5R ordinary Portland cement supplied by Jidong Cement
Corporation in China was used. Tap water, river sand with fineness modulus of
2.8 and coarse aggregate of crushed limestone with a maximum size of 20 mm were
used. Naphthalene series super-plasticizer with a water-reducing rate of 25% by
weight and air-entraining agent were used. The concrete mix proportion was
shown in Table 1.
Concrete prism specimens were cast in plastic molds of 100  100  300 mm,
removed from the molds 24 h after casting, and cured in the condition of
20 ± 3 °C and 95% of relative humidity for 28 d. The experiment was carried out
after storage for 3 months at room temperature (20 ± 5 °C) in accordance with the
Chinese standard GB/T50082-2009 [18].
2.2. Test procedure
2.2.1. Preset damage history
According to the needs of practical engineering, the effects of the fatigue load,
the freeze-thaw cycle and the combined action with different sequence, on the
Table 1
Mixture composition (kg/m3).
Cement Sand Coarse aggregate
420 633 1175
dynamic compression behaviors of concrete were studied, and the following four
categories of damage modes are considered.
(1) Mode F: Fatigue compression loading. The fatigue compression load was
applied by SDS500 electro-hydraulic servo static and dynamic universal
testing machine. In order to accurately determine the fatigue stress level,
a certain number of specimens were taken to determine the axial compres-
sive strength, with a loading speed of 0.5 MPa per second in accordance
with the Chinese standard GB/T50081-2002 [19], and the average static
compressive strength fc is equal to 51.1 MPa. A constant-amplitude sinu-
soidal waveform loading at a frequency of 5 Hz was applied on the concrete
prism specimens in the fatigue test. The ratio of loading amplitude to static
capacity ranged from 0.1 to 0.5. The number of fatigue loading cycles, n,
were 5000, 10000, 20000, and 40000, separately.
(2) Mode FT: Freeze-thaw cycling. The freeze-thaw test was completed on the
NELD-BFC concrete rapid freezing and thawing test machine in accordance
with the Chinese standard GB/T50082-2009 [18]. All the specimens were
immersed in water for 4 days at a temperature of 15–20 °C before freeze-
thaw test. Each freeze-thaw cycle was completed within 2–4 h, and the
time used for melting was not less than 1/4 of the whole freeze-thaw time.
In each freeze-thaw cycle, the center temperature of the specimen was kept
at 17 ± 2 °C and 8 ± 2 °C separately. The specimens were exposed sepa-
rately to 0, 25, 50, 75 freeze-thaw cycles.
(3) Mode F-FT: Fatigue compression loading followed by freeze-thaw cycling.
Firstly, the fatigue load was applied to specimens for 5000 or 10,000 times,
and then the specimens were exposed separately to 25, 50, 75 freeze-thaw
cycles. The procedure of fatigue test or freeze-thaw test is the same as the
one mentioned above.
(4) Mode FT-F: Freeze-thaw cycling followed by fatigue compression loading.
Firstly, specimens were exposed separately to 25, 50, 75 freeze-thaw cycles,
and then the fatigue load was applied separately to the specimens for 5000
or 10,000 times. The procedure of fatigue test or freeze-thaw test is the
same as the one mentioned above.
2.2.2. Uniaxial dynamic compressive test
Subsequently, a total of 180 concrete prism specimens that have been exposed
to four categories of damage modes were tested under quasi-static and dynamic
loading by using a 200 T electro-hydraulic servo universal testing machine.
Quasi-static test consisted of testing three specimens of each of the four groups
at strain rate of 105s1. The static tests provide the reference basis for the dynamic
values. Dynamic tests were carried out at different strain rates of 104s1, 103s1
within the capability of the test device and specimen combination. All tests were
conducted with controlling displacement increment method. The micromorphology
of the specimens is observed by using the JSM-7610F ultra-high resolution thermal
field emission scanning electron microscope. Non-contact displacement/strain
video measuring instrument was used to real-time monitor the displacements of
the specimens.
3. Results and discussion
3.1. Dynamic compressive strength
3.1.1. Damage Mode F
The dynamic compressive strength of concrete experienced to
damage Mode F, namely under the single effect of fatigue load, is
shown in Table 2. The values in the table are the average values
of 3 concrete prism specimens. F1, F2, F3 and F4, in the specimen
ID, stand for the fatigue cycle number (n) of 5000, 10,000,
20,000, and 40,000 respectively. D1, D2, and D3, in the specimen
ID, stand for the strain rates of 105 s1, 104 s1, and 103 s1
respectively. For example, the specimen ID of F1D1 denotes that
the specimen had been applied 5000 cycles of fatigue load and
the succedent uniaxial dynamic compression test was conducted
at the strain rate of 105 s1. The specimen ID of D3 denotes that
the specimen was not subjected to fatigue loading history and
the succedent uniaxial dynamic compression test was conducted
at the strain rate of 103 s1.
Water Super-plasticizer Air-entraining agent
172 2.94 0.084
 J. Lu et al. / Construction and Building Materials 152 (2017) 847–855 849

Table 2
The average compressive strength of concrete experienced damage Mode F (/MPa).

Specimen ID fc Specimen ID fc Specimen ID fc Specimen ID fc Specimen ID fc

D1 49.17 FID1 51.04 F2D1 51.66 F3D1 50.27 F4D1 48.28
D2 50.99 F1D2 52.69 F2D2 53.50 F3D2 51.98 F4D2 50.31
D3 53.63 F1D3 54.81 F2D3 55.17 F3D3 54.51 F4D3 53.23

strength increased with the increase of strain rate. From the point
60 of view of energy dissipation, the destruction of concrete materials
strain rate is due to the formation and development of cracks, the energy
-5 -1
58 10 s needed for crack formation is relatively much higher than the
-4 -1
10 s energy needed for crack propagation. The higher the strain rate
56 -3 -1 is, the more cracks were generated, and the more energy was
10 s
needed. When the strain rate reaches a higher level, the concrete
fc/MPa

54 material does not have sufficient time to accumulate energy, and

external energy can only be offset by an increase in stress, so the
52 compressive strength increases with the increase of strain rate.
50
3.1.2. Damage Mode FT
48 The dynamic compressive strength of concrete experienced to
damage Mode FT, namely under the single effect of freeze-thaw
46 cycle, is shown in Table 3. T1, T2, and T3, in the specimen ID, stand
0 1 2 3 4 for the number of freeze-thaw cycles of 25, 50, and 75, respec-
4
n/10 tively. When the strain rate remained constant, as shown in
Fig. 3, the compressive strength decreased linearly with an increas-
Fig. 1. Compressive strength versus fatigue cycles.
ing number of freeze-thaw cycles. As shown in Table 3, the greatest
reduction in compressive strength was found in T3D1 series of
specimens, and the average compressive strength decreased by
60
n=0 7.46%. The dynamic compressive strength was observed to increase
58 n=5×10
3 with strain rate in a nearly linear progression as shown in Fig. 4.
4 That means concrete experienced to repeated freeze-thaw damage
n=1×10
56 4 has apparent strain rate effect. As the strain rates become greater,
n=2×10
crack propagation velocities in concrete become faster under
4
n=4×10
fc/MPa

54 freeze-thaw cycles, however, because the crack could not fully

extended through the coarse aggregate, leading to the relative
52 enhancement of strength of concrete.

50
48
46
-5.0 -4.5 -4.0 -3.5
log(strain rate)
Fig. 2. Compressive strength versus strain rates.
Fig. 1 shows the dynamic compressive strength versus the num-
ber of fatigue cycles under different strain rates. When the strain
rate remained constant, the compressive strength increased first
and then decreased with the increase of fatigue cycles, as shown
in Fig. 1. After more than 10000 times fatigue cycles, the compres-
sive strength gradually decreased. When the number of fatigue
cycles reaches 40000 times, the compressive strength of the
specimen was lower than that of the control group without fatigue
loading history. It could be explained by that fewer fatigue cycles
can make the initial crack and defect of concrete become com-
pacted due to extrusion, so the compressive strength can be
improved. However, with the increase of fatigue cycles, the deteri-
oration of concrete cracks will lead to the reduction of compressive
strength. The results are in accordance with Ballatore et al.’s [20]
conclusion which reported that a moderate increase in concrete
static strength after being subjected to low intensity and short
duration cyclic loading.
Fig. 2 shows the dynamic compressive strength versus the
strain rate. The test results revealed that the dynamic compressive
3.1.3. Damage Mode FT-F
After exposure to a combination of freeze-thaw cycling fol-
lowed by fatigue compression loading (Mode FT-F), the dynamic
compressive strength versus the number of fatigue cycles is shown
-3.0 in Fig. 5. It can be seen from Fig. 5(a) that, when the number of
freeze-thaw cycles and the strain rate remained constant, the com-
pressive strength of concrete, after exposure to 25 freeze-thaw
cycles, increased with the increases of the number of fatigue cycles.
However, after exposure to 50 or 75 freeze-thaw cycles, the com-
pressive strength of concrete increases first and then decreases
with the increase of the number of fatigue cycles as shown in
Fig. 5(b) and (c). The difference in trend due to different numbers
of freeze-thaw cycles can possibly be explained by the following
reasons. Concrete prior to being subjected to dynamic loadings
have already withstood initial stress which cause the growth, prop-
agation and joining of micro cracks[21,22]. Less numbers of freeze-
thaw cycles did not have significant impact on the micro cracking
process (micro crack initiation and growth). A moderate increase in
strength can be attributed to compacting effects in the microstruc-
ture due to cyclic loading. Increasing the number of freeze-thaw
cycles, micro cracking occurs in the aggregate, in the matrix and
at the aggregate-matrix interface. The effects of compaction due
to cyclic loading cannot limit the further development of micro
cracks, and the compressive strength decreases.
After exposure to damage Mode FT-F, the dynamic compressive
strength versus the number of freeze-thaw cycles is shown in
Fig. 6. And it can be seen that, when the number of fatigue cycles
and the strain rate remained constant, the compressive strength
850 J. Lu et al. / Construction and Building Materials 152 (2017) 847–855

Table 3
The average compressive strength of concrete experienced damage Mode FT (/MPa).

Specimen ID fc Specimen ID fc Specimen ID fc Specimen ID fc

D1 49.17 TID1 47.43 T2D1 46.59 T3D1 45.50
D2 50.99 T1D2 49.57 T2D2 48.97 T3D2 47.89
D3 53.63 T1D3 52.54 T2D3 51.81 T3D3 50.82

experienced damage Mode FT-F. The dynamic compressive

56
strain rate strength increased with the increase of strain rate when the num-
10 -5 s-1 ber of fatigue and freeze-thaw cycles remained constant.
54
10 -4 s-1
52 10 -3 s-1 3.1.4. Damage Mode F-FT
After exposure to a combination of fatigue compression loading
fc/MPa

followed by freeze-thaw cycling (Mode F-FT), the dynamic com-

50 pressive strength versus the number of fatigue cycles is shown in
Fig. 8. It can be seen that the compressive strength of concrete
48 decreases with the increase of initial number of fatigue cycles
under the damage Mode F-FT. It revealed that the fatigue loading
46 history accelerated the process of freeze-thaw damage of concrete,
which led to the reduction of strength. Soroushian reported that
44 many new micro-cracks can be introduced due to the action of fati-
0 25 50 75 gue loading [3], which cause initial stress damage. Moreover, con-
N crete subjected to repeated freeze-thaw cycles largely results in
Fig. 3. Compressive strength versus freeze-thaw cycles. propagation of existing microcracks which caused by fatigue load,
and an increase in the irreversible tensile strain due to freeze-thaw
cycles causes the reduction in initial stiffness and the compressive
56 strength [9].
N=0 After exposure to damage Mode F-FT, the dynamic compressive
N=25 strength versus the number of freeze-thaw cycles is shown in
54 N=50 Fig. 9. When the number of fatigue cycles and the strain rate
N=75
remained constant, the compressive strength decreased with the
52
increase of the number of freeze-thaw cycles. Other things being
fc/MPa

equal, the compressive strength after 5000 fatigue cycles is greater

50 compared to that after 10000 fatigue cycles, as shown in Fig. 9
(a) and (b). It can be seen from Fig. 9(c), the compressive strength
48 of concrete subjected to initial fatigue loading history decreased
more rapidly than that of concrete without initial loading history.
46 This could indicate that the fatigue loading history accelerated the
process of freeze-thaw damage of concrete, which led to the reduc-
44 tion of strength. Some other researchers [6,7] have also found that
-5.0 -4.5 -4.0 -3.5 -3.0 the initial stress damage can accelerate the process of freeze-thaw
log(strain rate) damage in concrete.
Fig. 10 shows the dynamic compressive strength versus the
Fig. 4. Compressive strength versus strain rates. strain rates. When the numbers of freeze-thaw cycles and fatigue
cycles remained constant, the compressive strength of concrete
decreased with the increase of the number of freeze-thaw cycles. after fatigue increased with the increase of strain rate monotoni-
Fig. 7 shows the dynamic compressive strength versus the strain cally. Moreover, for concrete experienced damage Mode F-FT, the
rates. This figure illustrates that concrete has strain rate effect after reduction of dynamic compressive strength is little bit smaller than

58 58 56 strain rate
strain rate strain rate
10 -5 s-1 10 -5 s-1 10 -5 s-1
56 56 54
10 -4 s-1 10 -4 s-1 10 -4 s-1
54 -3 -1
10 s 54 10 s -3 -1
52 10 -3 s-1
fc/MPa
fc/MPa

fc/MPa

52 52 50

50 50 48

48 48 46

46 46 44
0 0.5 1 0 0.5 1 0 0.5 1
n/10 4
n/10 4 n/104
(a) N=25 (b) N=50 (c) N=75
Fig. 5. Compressive strength versus fatigue cycle number under damage Mode FT-F.
 J. Lu et al. / Construction and Building Materials 152 (2017) 847–855 851

58 58
strain rate strain rate
10 -5 s-1 10 -5 s-1
56 56
10 -4 s-1 10 -4 s-1
54 10 -3 s-1 10 -3 s-1
54
fc/MPa

fc/MPa
52 52

50 50

48 48

46 46
0 25 50 75 0 25 50 75

N N

(a) n=5000 (b) n=10000

Fig. 6. Compressive strength versus freeze-thaw cycle number under damage Mode FT-F.

56 56 n=0 56
n=0 n=0
3 3
n=5×10 n=5×10 n=5×10 3
54 4
54 4 54
n=1×10 n=1×10 n=1×10
4

52 52 52
fc/MPa
fc/MPa

fc/MPa
50 50 50

48 48 48

46 46 46

44 44 44
-5.0 -4.5 -4.0 -3.5 -3.0 -5.0 -4.5 -4.0 -3.5 -3.0 -5.0 -4.5 -4.0 -3.5 -3.0
log(strain rate) log(strain rate) log(strain rate)
(a) N=25 (b) N=50 (c) N=75
Fig. 7. Compressive strength versus strain rate under damage Mode FT-F.

56 54 52
strain rate strain rate strain rate
10 -5 s-1 10 -5 s-1 10 -5 s-1
54 52 50
-4 -1 -4 -1
10 s 10 s 10 -4 s-1
10 -3 s-1 10 -3 s-1
48 10 -3 s-1
52 50
fc/MPa

fc/MPa

fc/MPa

46
50 48
44
48 46
42
46 44 40
44 42 38
0 0.5 1 0 0.5 1 0 0.5 1
4 4 4
n/10 n/10 n/10
(a) N=25 (b) N=50 (c) N=75
Fig. 8. Compressive strength versus fatigue cycle number under damage Mode F-FT.

that of quasi-static compressive strength. It can be explained by (1) Other things constant, the compressive strength of concrete
that, the reduction in strength due to freeze-thaw cycles was even- under damage Mode F-FT is less than that under damage
tually offset, in part, by high strain rate effect. Mode FT-F. And the strength differences increase with the
increase of the number of freeze-thaw cycles, and the differ-
3.1.5. Influence of different action sequences on the dynamic ence can be up to 8 MPa (Specimen T3F2D1 and Specimen
compressive strength of concrete F2T3D1).
A summary of the dynamic compressive strength of concrete (2) When the strain rate and the number of freeze-thaw cycles
experienced damage Mode FT-F and damage Mode F-FT is remained constant, the strength difference increased with
presented in Table 4. There are obvious differences in dynamic the increase of the number of fatigue cycles. It can possibly
compressive strength of concrete due to two categories of damage be explained by that the initial fatigue compressive loading
modes with different action sequences. accelerated the subsequent frosted damage and the
852 J. Lu et al. / Construction and Building Materials 152 (2017) 847–855

56 60 55 n=0
strain rate strain rate
10 -5 s-1 10 -5 s-1 n=5×10
3

10 -4 s-1 55 10 -4 s-1 n=1×10 4

52 50
10 -3 s-1 10 -3 s-1

fc/MPa
fc/MPa

fc/MPa
50
48 45
45
44 40
40

40 35 35
0 25 50 75 0 25 50 75 0 25 50 75
N N N
(a) n=5000 (b) n=10000 (c) Strain rate=10-5s-1
Fig. 9. Compressive strength versus freeze-thaw cycle number under damage Mode F-FT.

60 N=0 65 N=0
N=25 N=25
N=50 60 N=50
55 N=75 N=75
55
fc/MPa

fc/MPa
50 50

45
45
40

40 35
-5.0 -4.5 -4.0 -3.5 -3.0 -5.0 -4.5 -4.0 -3.5 -3.0
log(strain rate) log(strain rate)
(a) n=5000 (b) n=10000
Fig. 10. Compressive strength versus strain rate under damage Mode F-FT.

Table 4
The average compressive strength of concrete experienced combined action of fatigue and freeze-thaw (/MPa).

Mode FT-F Mode F-FT Mode FT-F Mode F-FT

Specimen ID fc Specimen ID fc Specimen ID fc Specimen ID fc
TIF1D1 49.41 F1TID1 46.70 TIF2D1 51.05 F2TID1 45.89
T1F1D2 51.45 F1T1D2 49.01 T1F2D2 53.12 F2T1D2 48.26
T1F1D3 53.88 F1T1D3 51.99 T1F2D3 55.38 F2T1D3 51.23
T2F1D1 48.65 F1T2D1 44.91 T2F2D1 48.50 F2T2D1 43.00
T2F1D2 50.77 F1T2D2 47.35 T2F2D2 50.63 F2T2D2 45.42
T2F1D3 53.71 F1T2D3 50.54 T2F2D3 53.67 F2T2D3 48.50
T3F1D1 47.83 F1T3D1 41.34 T3F2D1 47.14 F2T3D1 39.14
T3F1D2 49.92 F1T3D2 43.70 T3F2D2 49.42 F2T3D2 41.54
T3F1D3 52.85 F1T3D3 47.17 T3F2D3 52.71 F2T3D3 45.20

compressive strength is significantly reduced under damage previously. And the reason will be explained more fully in
Mode F-FT. Whereas, the subsequent fatigue loading has a Section 3.3 with microcosmic method.
certain inhibitory effect on the early freeze-thaw damage
under damage Mode FT-F. For this reason, the differences
in compressive strength exist under two categories of dam- 3.2. Analysis of combined effect of fatigue loads and freeze-thaw cycles
age modes with different action sequences.
(3) When the number of fatigue cycles and freeze-thaw cycles In order to study the combined effect of fatigue load and freeze-
remain constant, the strength difference showed a decreas- thaw cycles on the deterioration of concrete properties, the reduc-
ing trend with the increase of strain rates. To take the spec- tion of compressive strength is used to characterize the compres-
imens in T2F2 and F2T2 series as example, corresponding to sive damage [23], namely
strain rate of 105 s1, 104 s1 and 103 s1, the strength
difference is 5.5 MPa, 5.21 MPa, and 5.17 MPa, respectively. h 0
The reason is that, to a certain extent, high strain rate effect Dc ¼ 1  f c =f c ð1Þ
can make up for the strength reduction of concrete due to In which, Dc is compressive damage of concrete, is compres-
h
fc
fatigue load and freeze-thaw damage histories, as discussed sive strength of concrete subjected to fatigue load combined with
 J. Lu et al. / Construction and Building Materials 152 (2017) 847–855
0
freeze-thaw cycles, f c is initial compressive strength of concrete
without any damage history.
The variable Kc is defined to characterize the combined effect of
fatigue load and freeze-thaw cycle,
K c ¼ Dc3  ðDc1 þ Dc2 Þ
In which, Dc1 is compressive damage of concrete due to fatigue
compression loading, Dc2 is compressive damage of concrete due to
freeze-thaw cycles, Dc3 is compressive damage of concrete due to
combination of fatigue loading and freeze-thaw cycles. When
Kc > 0, it means that the combined effect of fatigue load and
freeze-thaw cycles is positive, namely, the influence of these two
historical actions is mutually reinforcing. When Kc < 0, it means
that the combined effect of fatigue load and freeze-thaw cycles is
negative, namely, the influence of these two historical actions is
mutually inhibitory. Substituted Eq. (1) into Eq. (2), the calculated
Kc is shown in Table 5.
Table 5 shows, under damage Mode FT-F, after exposure to a
combination of 25, 50 and 75 freezing and thawing cycles followed
by 5000 times of fatigue loading, the values of Kc are always less
than zero, no matter how much strain rate is. The experimental
results reveal that, under damage Mode FT-F, the loss of compres-
sive strength of concrete due to combined effect of fatigue load and
freeze-thaw cycles is less than simply adding together that due to
any single factor. When concrete exposed to a combination of 25
freezing and thawing cycles followed by 10,000 times of fatigue
loading, the value of Kc is still negative and is less than that sub-
jected 5000 times of fatigue loading. That means the 10,000 times
of fatigue load has some alleviating effect on the damage of con-
crete after 25 freezing and thawing cycles. However when concrete
exposed to a combination of 50 or 75 freezing and thawing cycles
followed by 10,000 times of fatigue loading, the value of Kc is pos-
itive at strain rates of 105 s1 and 104 s1. That indicates a large
number of subsequent fatigue cycles has certain reinforcing effect
on the severe freeze-thaw damage of concrete. In addition, it is
worth noting that when the strain rates increase to 103 s1, no
matter how many times of fatigue and freeze-thaw cycles, the
value of Kc is always negative. It can be explained by compensation
of high strain rate, in part, as discussed previously.
Table 5 also shows, under damage Mode F-FT, the values of Kc
are always positive. When the number of fatigue cycles and the
number of freeze-thaw cycles remained constant, just with
different sequence, the value of Kc under damage Mode F-FT was
obviously higher than that under damage Mode FT-F at the same
strain rate. The results reveal that the compressive strength reduc-
tion of concrete due to damage history F-FT is greater than that due
to damage history FT-F. When the strain rate and the number of
fatigue loading remained constant, the value of Kc increased with
the increase of the number of freeze-thaw cycles. This implies that
the combined effect of fatigue loading and freeze-thaw cycles will
cause more serious damage to concrete with the increase of freeze-
thaw cycles. When the strain rate and the number of freeze-thaw
cycles remained constant, the value of Kc increased with the
increase of the number of fatigue loading. This implies that, with
Table 5
Kc under the combined action of fatigue loading and freeze-thaw cycles.
Strain rate/s1 Damage Mode The number of fatigue cycles, n/The number of freeze-thaw cycles, N
n = 5000/N = 25 n = 5000/N = 50
105 FT-F 0.22 0.39
F-FT 5.29 7.22
104 FT-F -0.35 0.20
F-FT 4.43 6.51
103 FT-F 0.30 1.34
F-FT 3.23 4.57
853
the increase of initial number of fatigue cycles, the later frozen
damage becomes more serious in concrete. In addition, the value
of Kc decreases with the increase of strain rate. This means that
the high strain rate has a certain inhibitory influence on the com-
bined effect under damage Mode F-FT.
ð2Þ
3.3. Microstructure Comparison and mechanism Analysis
The macroscopic mechanical properties of concrete are largely
determined by their microstructure. Evaluation of the microstruc-
ture of deteriorated concrete provides an invaluable, and in many
instances, the only method of determining the cause of deteriora-
tion. The study reported herein uses scanning electron microscopy
(SEM) to explain the effects of various damage histories on the
compressive strength of concrete, as shown in Fig. 11.
Comparison of Fig. 11(a)–(c) demonstrates that the concrete
specimens without damage history (i.e. control group) has uniform
and dense surface, whereas there are varying degrees of frozen
damage on concrete surface after exposure to fatigue loading and
freezing-thawing. As the number of freeze-thaw cycles is
increased, the number of cracks and pores on the surface of con-
crete increases, which is a good explanation for the compressive
strength reduction of concrete due to damage effects. For instance,
after exposure to a combination of 5000 cycles of fatigue loading
followed by 50 freezing and thawing cycles (i.e. F1T2 series of
specimens), the surface becomes uneven and crisp, accompanied
by a large number of micro cracks and pores, as shown in Fig. 11
(b). After exposure to a combination of 5000 cycles of fatigue load-
ing followed by 75 freezing and thawing cycles (i.e. F1T3 series of
specimens), concrete crack was further expanded and the depth
and width of cracks increase, as shown in Fig. 11(c).
Comparison of Fig. 11(b) and (d) can be seen that, after expo-
sure to a combination of fatigue loading followed by freezing and
thawing cycles, the damage degree of concrete was increased with
the increase of fatigue cycling times. It can be seen from Fig. 11(d)
that, 10,000 times of fatigue loading histories leaves holes and
cracks in the concrete, which become bigger and the gel was crum-
bled and washed away.
Comparison of Fig. 11(e) and (f) can be seen that, there are some
differences in the microstructure of concrete under different action
sequences of fatigue load and freeze-thaw cycles. Damage Mode
F-FT seems to introduce more new micro cracks as shown in
Fig. 11(e), when compared with damage Mode FT-F which largely
causes propagation of existing micro cracks. The microstructure
associated with different damaging effects is matched with the
macro mechanical properties of concrete. As described in Sec-
tion 3.1.5, the compressive strength of concrete under damage
Mode F-FT is less than that under damage Mode FT-F. The reasons
for the deterioration of the macro mechanical properties of concrete
under damage Mode F-FT are revealed from the micro perspective.
The influence of various damage histories upon the properties
of concrete has been reported [11–14,23,24]. Generally, it can be
stated that the process of progressive cracking is the primary cause
of the behavior of concrete observed under compression and
n = 5000/N = 75 n = 10,000/N = 25 n = 10,000/N = 50 n = 10,000/N = 75
0.94 2.30 1.18 1.73
12.26 8.20 12.37 18.00
0.65 2.04 1.67 1.92
11.55 7.49 11.88 17.38
1.58 2.42 0.60 0.65
9.01 5.31 9.04 13.35
854 J. Lu et al. / Construction and Building Materials 152 (2017) 847–855
m m
(a) control group
m 1
(d) F2T2
Fig. 11. SEM images of specimen after fatigue loading and freeze-thaw cycles.
subsequent failure. The density of fine cracking caused by the dam-
age history is therefore a measure of material integrity and can be
defined as damage.
The damage of concrete when subjected to fatigue loading and
freezing-thawing cycles separately or when combined is a complex
physical and chemical process. Concrete is a mixture of aggregate,
and often controlled amounts of entrained air, held together by a
hardened paste made from cement and water. It was well known
that only 1/3 of the water is needed for chemical reaction and more
water (twice than as needed for reaction with cement) is added for
workability. The extra water remains in pores and holes results in
porosity. Subfreezing temperatures lead to shrinkage of the min-
eral particles in the concrete and the volume expansion of free
water in the pores after freezing. The frost force has a destructive
effect on the particles with weak cementation strength, which
leads to local damage of concrete. When temperatures rise above
zero, the ice inside the concrete will melt, and the frost force can
be released. After repeated freeze-thaw cycles, the cracks in the
concrete will pass through each other, and their strength will grad-
ually decrease, and finally even completely lost.
It is well known that some pre-existing cracks exist at the
aggregate–paste interfaces or internal cement within the concrete.
Fatigue loading results in the initiation of new micro-cracks and
propagation of pre-existing cracks, even though the stress level is
lower than the compressive strength of concrete. By increasing
the number of fatigue cycles, joining of micro-cracks occurring in
the aggregate, in the matrix and at the aggregate-matrix interface
leads to the failure. Thus, fatigue is a process of progressive and
permanent internal damage in concrete subjected to repeated
loading. However, low intensity and short duration fatigue loading
in uniaxial compression can compact the material, which displays
a 10% to 15% increase in failure load [20]. Sharma [25] found the
pre-existing cracks have been closed due to the initial application
of compressive load from reconstructed X-ray computed tomogra-
phy (XCT) images. Therefore, the compressive strength of concrete
under damage Mode FT-F increases with the increase of a given
less number of fatigue cycles decreases after being exposed to a
large number of fatigue cycles in Fig. 5.
m
(b) F1T2 (c) F1T3
m 1 m
(e) F2T2 (f) T2F2
As for damage Mode F-FT, on one hand, many new micro-cracks
can be introduced after the initial application of compressive load,
even if the cracks are closed due to the short duration cyclic load.
On the other hand, when the concrete was exposed to cycles of
freezing and thawing after fatigue loading, repeated action of frost
force led to continued expansion and accumulation of cracks as
discussed above, which greatly speed up the damage evolution
process. An increase in the irreversible tensile strain due to
freeze-thaw cycles causes the reduction in initial stiffness and
the compressive strength.
4. Conclusions
The present study was undertaken to systematically investigate
the effects of fatigue compression load, freeze-thaw cycles and the
combined action with different sequence, on the uniaxial dynamic
compressive strength of plain concrete. Although the testing pro-
gram presented here was not intended to perfectly simulate load-
ing and exposure conditions that were representative of those
applied to concrete structures in cold climates, the test results do
give insight into the interaction of the deterioration caused by
freeze-thaw cycles and compressive fatigue loading cycles and
how they may affect the dynamic strength of plain concrete
exposed to such conditions. The following conclusions can be
drawn from the present study.
(1) The compressive strength of concrete subjected to fatigue
compression loading (Mode F) increases first and then
decreases with the increase of initial number of fatigue
cycles. It is mainly due to the fatigue load level. Fewer fati-
gue cycles can make the initial defect in concrete become
compacted due to extrusion, which results in a slight
increase in strength. As the fatigue cycles increase, the dete-
rioration of concrete cracks will lead to the reduction of
compressive strength.
(2) The compressive strength of concrete exposed to freeze-
thaw cycles (Mode FT) decreases linearly with an increasing
number of freeze-thaw cycles. After repeated freeze-thaw
 J. Lu et al. / Construction and Building Materials 152 (2017) 847–855
cycles, the cracks in the concrete will pass through each
other, and their strength will gradually decrease, and finally
even completely lost.
(3) After exposure to a combination of freeze-thaw cycling fol-
lowed by fatigue compression loading (Mode FT-F), less
numbers of freeze-thaw cycles did not have significant
impact on the micro cracking process. A moderate increase
in strength can be attributed to compacting effects in the
microstructure due to cyclic loading. Increasing the number
of freeze-thaw cycles, the cracks in concrete will pass
through each other. The effects of compaction due to cyclic
loading cannot limit the further development of micro
cracks, and the compressive strength decreases.
(4) After exposure to a combination of fatigue compression
loading followed by freeze-thaw cycling (Mode F-FT), the
compressive strength of concrete decreases with the
increase of initial number of fatigue cycles. It revealed that
the fatigue loading history accelerated the process of
freeze-thaw damage of concrete, which led to the reduction
of strength. The reason is that, concrete subjected to
repeated freeze-thaw cycles largely results in propagation
of existing micro cracks which caused by fatigue load, and
an increase in the irreversible tensile strain due to freeze-
thaw cycles causes the reduction in the compressive
strength.
(5) The variable Kc was introduced to characterize the combined
effect of fatigue compression load and freeze-thaw cycles,
and the damage mechanism in concrete was discussed
according to the results of SEM. The results showed that,
when the number of fatigue cycles and the number of
freeze-thaw cycles remained constant, just with different
sequence, the values of Kc under damage Mode F-FT were
obviously higher than that under damage Mode FT-F at the
same strain rate. The reasons for the deterioration of the
macro mechanical properties of concrete under different
damage effects are revealed from the micro perspective.
(6) Test results show that concrete has strain rate effect after
exposure to fatigue compression loading, or freeze-thaw
cycles, or the combination of fatigue load and freeze-thaw
cycles. The dynamic compressive strength increases as the
strain rate is increased. From the point of view of energy dis-
sipation, when the strain rate reaches a higher level, the con-
crete material does not have sufficient time to accumulate
energy, and external energy can only be offset by an increase
in stress, so the compressive strength increases with the
increase of strain rate.
In present paper, the number of fatigue loading as well as the
freeze-thaw cycles is limited, so no absolute conclusion can be
drawn that the damage of concrete deterioration due to Mode
F-FT is more serious than that due to Mode FT-F. In order to fully
understand the combined effect of fatigue load and freeze-thaw
cycle on concrete properties, it is needed to further increase the
number of test items and specimens. In addition, it is necessary
to extend the study on the dynamic mechanical properties of con-
crete with a larger strain rate range.
Acknowledgements
This research is supported by the National Natural Science
Foundation of China (No. 51479174) and Open Research Fund Pro-
gram of State key Laboratory of Hydroscience and Engineering
(sklhse-2017-C-01).
855
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.conbuildmat.
2017.07.046.
References
[1] W. Sun, Y.M. Zhang, H.D. Yan, et al., Damage and its restraint of concrete with
different strength grades under double damage factors, Cem. Concr. Compos.
21 (5) (1999) 439–442.
[2] W. Li, W. Sun, J. Jiang, Damage of concrete experiencing flexural fatigue load
and closed freeze-thaw cycles simultaneously, Constr. Build. Mater. 25 (2011)
2604–2610.
[3] P. Soroushian, M. Elzafraney, Damage effects on concrete performance and
microstructure, Cem. Concr. Compos. 26 (7) (2004) 853–859.
[4] W. Li, W. Sun, J. Jiang, An analytical method to detect the coupling damage
relationship of concrete subjected to bending fatigue and temperature actions,
Journal of Wuhan University of Technology- Mater, Sci. Ed. 26 (3) (2011) 573–
577.
[5] Y. Qiao, W. Sun, J. Jiang, Damage process of concrete subjected to coupling
fatigue load and freeze/thaw cycles, Constr. Build. Mater. 93 (2015) 806–811.
[6] Q.X. Zhao, P.P. Kang, Influence of mechanical damage on frost resistance of
concrete, J. Build. Mater. 16 (2) (2013) 326–334 (in Chinese).
[7] C. Gao, X. Shen, X. Wang, et al., Freeze-thaw resistance of lightweight
aggregate concrete with stress damage, J. Chin. Ceram Soc. 42 (10) (2014)
1247–1252 (in Chinese).
[8] B. Han, J. Cao, J. Dong, Influence of sustainable loading on the freezing and
thawing performance of fly ash concrete, China Railway Sci. 33 (2) (2012) 35–
39 (in Chinese).
[9] M. Hasan, T. Ueda, Y. Sato, Stress-strain relationship of frost-damaged concrete
subjected to fatigue loading, J. Mater. Civil Eng. 20 (1) (2008) 37–45.
[10] J. Hong, C. Miao, W. Huang, et al., Influence of freeze-thaw damage on the
fatigue life of concrete, China Civil Eng. J. 45 (6) (2012) 83–89 (in Chinese).
[11] D. Yan, G. Lin, Influence of initial static stress on the dynamic properties of
concrete, Cem. Concr. Compos. 30 (4) (2008) 327–333.
[12] D. Yan, G. Lin, G. Chen, Dynamic properties of plain concrete in triaxial stress
states, ACI Mater. J. 106 (1) (2009) 89–94.
[13] S. Xiao, P.J.M. Monteiro, H. Li, Influence of strain rates and loading histories on
the compressive damage behaviour of concrete, Mag. Concrete Res. 63 (12)
(2011) 915–926.
[14] S. Xiao, H. Hai, Effects of load histories on dynamic biaxial compressive
damage behaviors of concrete, Mag. Concrete Res. 64 (5) (2012) 445–455.
[15] H. Wang, W. Fan, T. Liu, et al., Experimental study on properties of fully-graded
concrete under uniaxial dynamic compressive stress state after freezing-
thawing cycles, J. Hydraul. Eng. 46 (6) (2015) 732–738.
[16] J. Yi, S. Shen, B. Muhunthan, et al., Viscoelastic-plastic damage model for
porous asphalt mixtures: Application to uniaxial compression and freeze-thaw
damage, Mech. Mater. 70 (1) (2014) 67–75.
[17] J.Y. Chen, X. Qiu, G.Y. Shi, et al., Effect of freeze-thaw cycles on mechanical
behaviors of ceramsite concrete under impact loading, Appl. Mech. Mater. 784
(2015) 508–513.
[18] National Standard of the People’s Republic of China, GB/T50082-2009
Standard for Test Methods of Long-Term Performance and Durability of
Ordinary Concrete, China Architecture and Building Press, Beijing, 2009 (in
Chinese).
[19] National Standard of the People’s Republic of China, GB/T50081-2002
Standard for Test Method of Mechanical Properties on Ordinary Concrete,
China Architecture and Building Press, Beijing, 2002 (in Chinese).
[20] E. Ballatore, P. Bocca, Variations in the mechanical properties of concrete
subjected to low cyclic loads, Cem. Concr. Compos. 27 (3) (1997) 453–462.
[21] D.P. Forgeron, J.F. Trottier, Evaluating the Effects of Combined Freezing and
Thawing and Flexural Fatigue Loading Cycles on the Fracture Properties of FRC,
in: W.P. De Wilde, C.A. Brebbia (Eds.), High performance structures and
materials proceedings No. 2, WIT press, USA, 2004, pp. 177–187.
[22] D.J. Janssen, M.B. Snyder, Resistance of concrete to freezing and thawing,
SHRP-C-391, Strategic Highway Research Program. National Research Council,
Washington DC, 1994.
[23] J.Z. Lu, G. Lin, Z. Wang, et al., Reduction of compressive strength of concrete
due to triaxial compressive loading history, Mag. Concrete Res. 56 (3) (2004)
139–149.
[24] G. Lin, J.Z. Lu, Z. Wang, et al., Study on the reduction of tensile strength of
concrete due to triaxial compressive loading history, Mag. Concrete Res. 54 (2)
(2002) 113–124.
[25] R. Sharma, W. Ren, S. Mcdonald, Micro-mechanisms of concrete failure under
cyclic compression: X-ray tomographic in-situ observations, Int. Conf. Fracture
Mech. Concr. Concr. Struct. (2016) 1–8.