Construction and Building Materials 96 (2015) 404–415

Contents lists available at ScienceDirect

Construction and Building Materials

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

Performance deterioration of corroded RC beams and reinforcing bars

under repeated loading
Junzu Sun ⇑, Qiao Huang, Yuan Ren
School of Transportation, Southeast University, Nanjing 210096, China

h i g h l i g h t s

Failure mode of corroded RC beams under repeated loading was investigated.

Flexural stiffness change of corroded RC beams under repeated loading has two obvious features of stages.

Characteristics of stress–strain curves for reinforcing bars after corrosion fatigue changed significantly.

Modified calculation formula for flexural stiffness of corroded RC beam under repeated loading was presented.

Constitutive relation model for the reinforcing bar after corrosion fatigue was presented.

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

Article history: Based on the accelerated corrosion test and fatigue loading test of RC beam specimens and the static
Received 9 March 2015 tensile test of reinforcement specimens that were removed from the beam specimens after fatigue load-
Received in revised form 4 August 2015 ing test, failure mode and flexural stiffness of corroded RC beams under repeated loading and mechanical
Accepted 9 August 2015
behavior of reinforcing bars after corrosion fatigue were studied. We find that flexural stiffness change of
Available online 14 August 2015
corroded RC beams under repeated loading has two obvious features of stages, namely, slowly decreasing
stage and stable stage. Considering the effects of corrosion and fatigue respectively, stiffness correction
Keywords:
coefficients were introduced to quantitatively present the modified calculation formula for flexural stiff-
Corroded RC beam
Flexural stiffness
ness of corroded RC beam under repeated loading. After corrosion fatigue, characteristics of stress–strain
Corrosion fatigue curves for reinforcing bars changed obviously: the yield strength decreased, percentage of elongation
Mechanical behavior shortened, features of yield plateau changed. Mild steel changed into hard steel to varying degrees.
Constitutive relation Combined with the experimentally obtained stress–strain curves, constitutive relation model for the
reinforcing bar after corrosion fatigue was quantitatively presented. The results of this study can provide
reference for calculation and assessment of corroded RC bridge structures under long-term live load.
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction When damage of repeated loading accumulates to a certain extent,

The widespread occurrence of steel reinforcement corrosion in
existing reinforced concrete (RC) bridge structures is a common
problem with great damage. Steel reinforcement corrosion will
cause decrease in load-carrying capacity and degradation in dura-
bility of bridge structures. Especially, the coastal environment with
high content of chlorine ion as well as the extensive use of chloride
salt snow-melting agent in winter aggravates the corrosion of steel
reinforcing bars in bridge structures. The existing highway and
railway bridge structures in operation are mostly subjected to
repeated action of live vehicle load in addition to dead weight load.
⇑ Corresponding author.
E-mail addresses: sunjunzu@163.com (J. Sun), qhuanghit@126.com (Q. Huang),
magren@126.com (Y. Ren).
it will lead to brittle fatigue failure without any warning.
Particularly, the rapid growth of traffic and widespread overload
of freight make the repeated fatigue effect of vehicle load more
and more significant. Without a doubt, when RC bridge structures
are under dual adverse effects of chloride ion corrosive environ-
ment and repeated vehicle load, namely the steel reinforcement
is under coupling effects of corrosion and fatigue, performance of
RC bridge structures will deteriorate rapidly. Especially, the flexu-
ral behavior of RC beams and mechanical properties of steel rein-
forcing bars will change significantly, which are important
assessment indexes for structural serviceability and safety of RC
bridge structures. Therefore, it has important theoretical value
and practical significance to carry out research on performance
deterioration of corroded RC beams and reinforcing bars under
repeated loading.

http://dx.doi.org/10.1016/j.conbuildmat.2015.08.066
0950-0618/Ó 2015 Elsevier Ltd. All rights reserved.
 J. Sun et al. / Construction and Building Materials 96 (2015) 404–415
A large amount of studies have investigated load-carrying
capacity [1], residual flexural capacity [2], flexural stiffness [3]
and load–deflection curve [4] of corroded RC beams taking into
account the effects of stirrup corrosion [5], different types of rein-
forcing bars [6], size-effect of tensile reinforcement [7] and bond
strength degradation [8]. Flexural behavior of RC beams which
were corroded under constant sustained service loads was
analyzed [9]. Corroded RC beam was simulated through two-
dimensional [10] and three-dimensional [11] finite element
modeling approaches considering material and geometrical non-
linearity [12], material properties change in steel and concrete,
bond-slip deterioration and cracking [13]. Based on numerical
analysis and simplified methodology, crack width prediction [14],
performance assessment [15], assessment of structural safety level
[16] and residual life evaluation [17] of corroded RC beams were
discussed. In addition to static behavior, fatigue performance of
RC beams has caused wide concern. Fatigue life [18], flexural stiff-
ness [19], crack width [20] and crack growth [21] of RC beams
under repeated loading have been investigated by various
researchers. Non-linear behavior of cracked RC beams under vari-
able stress amplitude was numerically simulated [22]. All of the
investigations mentioned above are behavior of RC beams under
the influence of pure corrosion or pure fatigue. There has been little
concern about combined effects of corrosion and fatigue histori-
cally [23]. Fang et al. [24] experimentally investigated flexural
behavior of corroded RC beams under repeated loading regarding
corrosion influence on bond strength as the most important factor
and indicated that a low corrosion level increased the bond
strength between concrete and reinforcing bar. Oyado et al. [25]
conducted fatigue loading test of corroded RC beams and found
that the reduction of fatigue strength was proportionate to the
weight loss of reinforcing bar. Yi et al. [26] found out that corrosion
caused brittle fatigue failure and an increase in degree of corrosion
caused a corresponding reduction of fatigue life. Masoud et al. [27]
experimentally compared the fatigue performance of corroded RC
beams strengthened by FRP wrapping and CFRP wrapping. Above
existing investigations mainly discussed bond strength and flexu-
ral fatigue life of corroded RC beams under repeated loading.
However, for the study of flexural stiffness for corroded RC beams
under repeated loading in serviceability limit state, the research
progress mainly focused on qualitative variation law, few
researches have been carried out in the aspect of quantitative cal-
culation formula currently, and there is no unified understanding.
The effect of corrosion on mechanical properties of reinforcing
bars has been extensively studied through static tensile test. It
was observed that local or pitting corrosion lead to moderate loss
of strength [28] and significant reduction of ductility [29]. High
degree of corrosion caused a brittle failure [30]. Pitting corrosion
was more critical in the case of low corrosion degree and more
likely to increase the chance of brittle fracture failure from ductile
yielding [31]. It was strongly suggested that the spatial variability
of pitting corrosion be considered [32], which lead to a consider-
able reduction in structural reliability [33]. The mechanism of pit
development, the pit depth, the pit area and the effect of pits on
the mechanical properties of embedded steel bar due to chloride
induced corrosion were studied [34]. In addition, results of axial
tensile fatigue test for reinforcing bars were presented in terms
of failure mode, S–N curve [35] and stress concentration [36].
However, relatively little research has dealt with the fatigue per-
formance of corroded reinforcing bars. Apostolopoulos et al. [37]
carried out experimental studies on low cycle fatigue properties
of corroded reinforcing bars to simulate seismic loading condition
and found that load-carrying capacity, available energy and fatigue
life were gradually reduced. Similar conclusions were also reached
by Hawileh et al. [38]. Low cycle fatigue test [39] and finite ele-
ment modeling [40] of nonlinear cyclic response and inelastic
405
buckling for corroded reinforcing bars were carried out. Zhang
et al. [41] investigated fatigue behavior of natural corrosion and
accelerated corrosion induced reinforcing bars and the axial tensile
fatigue test results showed that fatigue life of corroded reinforcing
bars decreased significantly with the increase of corrosion degree
and stress amplitude. The existing related researches mostly
focused on mechanical properties of corroded reinforcing bars in
the case of no fatigue, low cycle fatigue performance of corroded
reinforcing bars and the corrosion influence on axial tensile fatigue
life of reinforcing bars. However, for the research on mechanical
properties of reinforcing bars in concrete after corrosion fatigue,
few studies have been reported so far, and there is no consensus.
In this paper, combined with previous research work, on the
basis of the actual stress level of tensile reinforcement in existing
RC bridge structures, further efforts were made through acceler-
ated corrosion test and fatigue loading test of corroded RC beams
and static tensile test of reinforcing bars after corrosion fatigue.
Change law for flexural stiffness of corroded RC beams under
repeated loading was analyzed, and calculation formula for flexural
stiffness of corroded RC beams under repeated loading was quan-
titatively presented. Degradation law for mechanical properties
of reinforcing bars in concrete after corrosion fatigue was dis-
cussed, and constitutive relation model for reinforcing bars in con-
crete after corrosion fatigue was quantitatively presented. Our
findings are expected to provide reliable theoretical supports and
practical technique methods for the research on fatigue perfor-
mance of corroded RC in-service bridge structures.
2. Experimental procedure
2.1. Details and materials of test specimen
13 RC beam specimens were designed, among which, 1 specimen was subjected
to monotonic static loading without corrosion, 1 specimen was subjected to
repeated fatigue loading without corrosion, the other 11 specimens were subjected
to repeated fatigue loading after accelerated corrosion, as shown in Table 1.
Dimension details and reinforcement arrangement of the beam specimen are
shown in Fig. 1. Strength grade of the concrete was designed to be C25, and the
measured average compressive strength was 31.3 MPa. The tensile reinforcement
used HRB335 of the same factory batch, whose diameter was 16 mm. And the mea-
sured yield strength and tensile strength were 478 MPa and 608 MPa respectively.
The tensile reinforcing bars extended the beam end by 150 mm, which was used for
connecting wires. Before assembling the reinforcements, initial mass of each one of
the tensile reinforcing bars was weighed.
2.2. Accelerated corrosion test
The beam specimens have been maintained for 28 days after casting. Then
direct current was produced and flowed through the tensile reinforcement to con-
duct electrochemical accelerated corrosion test, as shown in Fig. 2. According to
Faraday’s law, by controlling the applied current intensity and conduction time, dif-
ferent degree of corrosion was achieved. As soon as the electrochemical accelerated
corrosion test was completed, namely the expected degree of corrosion was
reached, fatigue loading test was conducted.
2.3. Fatigue loading test
The beam specimens were subjected to four-point bending fatigue loading.
Calculated span is 1500 mm, the length of pure bending section between the load-
ing points is 500 mm, the length of bending shear section between the load point
and the fulcrum is 500 mm. Measuring points of deflection were shown in Fig. 3.
Table 1
Grouping of the beam specimens.
Group Number of the Name of the Accelerated Fatigue
beam specimens beam corrosion or loading or
specimen not not
I 1 J7 No No
II 1 J13 No Yes
III 11 J1–J6, J8–J12 Yes Yes
406 J. Sun et al. / Construction and Building Materials 96 (2015) 404–415

Fig. 1. Dimension details and reinforcement arrangement (unit: mm).

Fig. 2. Principle and method for electrochemical accelerated corrosion test.

Fig. 4. Fatigue loading test system.

Fig. 3. Loading way and measuring point position of deflection.

Fatigue loading test was carried out on the computer controlled two-channel
electro-hydraulic servo static and dynamic loading test system (model: JAW-
500K/4), as shown in Fig. 4. The maximum loading capacity of the actuator is
250 kN. Loading frequency was 4 Hz, control mode of test force control was
adopted, and sine waveform was used. In order to correctly reflect the actual stress
level of simply supported RC girder bridge in service, by means of analysis and cal-
culation, the upper limit value of fatigue load used 70 kN, which was 47.3% of static
failure load of the uncorroded specimen, in order to simulate the short-term effects
of bridge load combination in serviceability state. The lower limit value of fatigue
load used 30 kN, which was equivalent to the effect of bridge dead load. The corre-
sponding upper limit value of fatigue stress for the tensile reinforcement was
221.9 MPa, lower limit value of fatigue stress was 95.1 MPa, and stress range was
126.8 MPa. The fatigue amplitude of the tensile reinforcement is equivalent to
the stress increment caused by the live load of bridge load combination in service-
ability state.
Before the start of fatigue loading, 5–10 kN static load was pre-loaded, and
repeatedly, so that the bearing contact was good. Testing and checking were simul-
taneously done to see if the data of each channel was normal. Then, after the force Fig. 5. Cutting position of tensile reinforcement and prepared reinforcement
of beam specimen was adjusted to the fatigue median 50kN, fatigue loading started. segment specimens (unit: mm).
When number of fatigue cycles reached 0, 1, 1  104, 2  104, 5  104, 10  104,
20  104, 50  104, 100  104, 150  104 and 200  104, fatigue loading test auto-
matically paused and the actuator automatically unloaded. Then after the residual
deformation was stable, the beam specimen was monotonically loaded to the upper water, rinsing with clear water and drying. Mass loss percentage of the two seg-
limit value of fatigue load under step loading. Each step was 10 kN. And the deflec- ments at the fatigue fracture location of the fractured reinforcing bar was regarded
tion data was collected at each step of monotonic loading. When fatigue fracture of as the corrosion degree of corroded RC beam after fatigue failure. If fatigue failure
any one of the tensile reinforcing bars occurred, namely fluctuation range of limited did not occur after cyclic loading to 2 million times, monotonic static loading will
value for test force exceeded 8%, the actuator immediately unloaded for protection continue to be applied until static failure of the beam specimen.
to prevent the beam specimens from being secondary loaded after fatigue failure.
The fatigue loading test ended. Then after fatigue loading test, all the tensile rein- 2.4. Static tensile test
forcing bars were carefully removed out of the beam specimens by chiseling con-
crete to pieces. Each segment of 50 mm long was cut out on both sides of the After fatigue loading test, the removed tensile reinforcing bars were cut into
fatigue fracture location of the fractured reinforcing bar. According to standard segments for static tensile test, as shown in Fig. 5. Then according to the test
measurement method of corrosion degree, the reinforcing bar segment was method of ‘‘Metallic materials – Tensile testing – Part 1: Method of test at room
weighted after pickling with 12% hydrochloric acid solution, neutralizing with lime temperature (GB/T 228.1-2010)” which is equivalent to ‘‘ISO 9892-1: 2009,
 J. Sun et al. / Construction and Building Materials 96 (2015) 404–415 407

MOD”, static tensile test for the reinforcement segment was carried out on com-
puter controlled hydraulic universal testing machine (model: WEW-1000D), as
shown in Fig. 6. Different from the definition of corrosion degree for corroded RC
beam specimens in fatigue loading test, mass loss percentage of full length range
of the corroded reinforcement segment specimen in static tensile test was regarded
as the degree of corrosion.

3. Test results and analysis

3.1. Corrosion induced crack

After a period time of accelerated corrosion of the tensile rein-

forcement, corrosion induced cracks along longitudinal direction of
Fig. 7. Local corrosion induced crack of J2 and J3 beam specimens.
the tensile reinforcement can be observed on either side surface or
bottom surface of the beam specimen around the corroded tensile
reinforcement, as shown in Fig. 7. But the corrosion induced cracks
did not appear on both side surface and bottom surface of the
beam specimen at the same time. The emergence of corrosion
induced cracks rapidly weakened the bond strength between con-
crete and reinforcement. And bond damage gradually accumulated
with the development of corrosion induced cracks.

3.2. Failure mode

Brittle fatigue failure occurred in all the corroded beam speci-

mens after cyclic loading to a certain number of times, as shown
in Fig. 8. It was marked by a sudden fatigue fracture of one of
the tensile reinforcing bars and resulted in a sudden drop in upper
limit value of fatigue load and a sudden increase in mid-span Fig. 8. Fatigue failure of J1 and J6 beam specimens.
deflection. A sudden broadened main crack formed on the location
where tensile reinforcement fractured. The sound of reinforcement
fracture was heard clearly. The fatigue failure of the beam speci-
mens had the risk of sudden failure, which was dangerous and
without any warning. Fig. 9 shows the fracture surface of the frac-
tured tensile bar. From the figure it can be seen that, the fracture
surface did not have tensile plastic deformation, which was signif-
icantly different from the necking deformation characteristics of
static tensile failure.
Fatigue failure did not take place in the uncorroded beam spec-
imen after cyclic loading to 2 million times. And in the subsequent

Fig. 9. Fracture characteristics of the fractured steel bar after fatigue failure.

monotonic static loading test, the uncorroded beam specimen

showed good characteristics of ductile failure. The actual static
failure load was 148 kN, which was almost the same as the static
failure load of the beam specimen without corrosion fatigue. It
was showed that fatigue effect had no substantial influence on
the static performance of RC beams without corrosion.

3.3. Mid-span deflection

The measured mid-span deflection of the beam specimens

Fig. 6. Setup for static tensile test.
under monotonic step loading after cyclic loading to different pre-
determined number of fatigue cycles is shown in Table 2. From
Table 2(a)–(l), it can be found and analyzed as follows.
Cracks appeared after 1 cycle of fatigue loading from the initial
state without any crack. Mid-span deflection of the beam specimen
increased rapidly, and flexural stiffness of the beam specimen sig-
nificantly weakened (N ¼ 0; 1). The significant reduction of flexural
stiffness after cyclic loading for the first time is because that the
concrete in tensile region cracked and was out of work, the neutral
axis moved up, and whole section flexural moment of inertia chan-
ged into cracked section flexural moment of inertia.
408 J. Sun et al. / Construction and Building Materials 96 (2015) 404–415

Table 2
Mid-span deflection under monotonic step loading after cyclic loading to different predetermined number of fatigue cycles.

Cycles Measured value of mid-span deflection (mm)

10 kN 20 kN 30 kN 40 kN 50 kN 60 kN 70 kN
(a) J1 beam specimen g ¼ 9:6%
0 0.29 0.60 0.93 1.26 1.61 1.98 2.36
1 0.32 0.67 1.03 1.40 1.78 2.17 2.58
4
1  10 0.34 0.71 1.10 1.50 1.87 2.29 2.69
2  104 0.34 0.74 1.16 1.56 1.96 2.37 2.78
5  104 0.36 0.74 1.16 1.57 1.97 2.38 2.80
10  104 0.39 0.76 1.18 1.59 1.99 2.41 2.81
20  104 0.37 0.76 1.18 1.59 2.00 2.41 2.82
4
50  10 0.37 0.76 1.18 1.60 2.00 2.41 2.81
(b) J2 beam specimen g ¼ 11:2%
0 0.28 0.59 0.91 1.23 1.58 1.92 2.29
1 0.31 0.64 0.99 1.34 1.70 2.07 2.46
1  104 0.33 0.68 1.05 1.42 1.81 2.17 2.57
2  104 0.34 0.71 1.10 1.48 1.87 2.25 2.63
5  104 0.35 0.72 1.08 1.51 1.90 2.28 2.68
10  104 0.35 0.72 1.12 1.53 1.91 2.31 2.70
4
20  10 0.35 0.73 1.13 1.53 1.92 2.32 2.71
4
50  10 0.35 0.74 1.14 1.55 1.95 2.32 2.73
(c) J3 beam specimen g ¼ 15:6%
0 0.27 0.56 0.86 1.17 1.49 1.83 2.19
1 0.30 0.62 0.96 1.29 1.64 2.02 2.38
1  104 0.32 0.66 1.02 1.37 1.73 2.10 2.48
2  104 0.33 0.68 1.05 1.42 1.78 2.15 2.52
5  104 0.33 0.66 1.06 1.43 1.80 2.18 2.54
4
10  10 0.33 0.70 1.08 1.49 1.83 2.23 2.61
(d) J4 beam specimen g ¼ 5:9%
0 0.29 0.60 0.92 1.25 1.60 1.99 2.37
1 0.29 0.60 0.98 1.40 1.78 2.17 2.58
1  104 0.30 0.63 1.06 1.50 1.88 2.31 2.70
2  104 0.31 0.71 1.14 1.56 1.95 2.37 2.78
5  104 0.32 0.69 1.16 1.59 1.99 2.40 2.82
4
10  10 0.34 0.75 1.18 1.61 2.04 2.44 2.85
4
20  10 0.33 0.77 1.19 1.62 2.04 2.44 2.87
4
50  10 0.31 0.78 1.20 1.63 2.04 2.47 2.90
100  104 0.36 0.78 1.20 1.65 2.07 2.50 2.92
150  104 0.35 0.79 1.23 1.66 2.09 2.51 2.93
(e) J5 beam specimen g ¼ 6:8%
0 0.29 0.59 0.92 1.24 1.59 1.95 2.33
1 0.29 0.66 1.02 1.39 1.77 2.16 2.57
4
1  10 0.29 0.69 1.08 1.50 1.87 2.29 2.69
4
2  10 0.33 0.74 1.15 1.56 1.97 2.37 2.77
5  104 0.35 0.75 1.17 1.60 1.99 2.40 2.82
10  104 0.36 0.76 1.19 1.61 2.03 2.42 2.84
20  104 0.34 0.77 1.19 1.62 2.02 2.39 2.85
50  104 0.35 0.77 1.20 1.62 2.04 2.45 2.88
4
100  10 0.34 0.78 1.20 1.63 2.04 2.47 2.90
4
150  10 0.35 0.78 1.20 1.63 2.06 2.48 2.91
(f) J6 beam specimen g ¼ 9:7%
0 0.29 0.59 0.92 1.24 1.59 1.96 2.34
1 0.31 0.66 1.02 1.38 1.76 2.16 2.56
1  104 0.34 0.71 1.10 1.50 1.87 2.27 2.69
2  104 0.34 0.74 1.15 1.55 1.95 2.35 2.76
4
5  10 0.36 0.75 1.16 1.56 1.96 2.36 2.78
4
10  10 0.36 0.75 1.17 1.59 1.97 2.39 2.79
4
20  10 0.36 0.76 1.13 1.58 1.99 2.39 2.80
4
50  10 0.35 0.76 1.18 1.60 1.99 2.41 2.81
(g) J8 beam specimen g ¼ 18:5%
0 0.26 0.54 0.83 1.12 1.43 1.74 2.09
1 0.29 0.60 0.92 1.24 1.57 1.91 2.27
1  104 0.30 0.63 0.96 1.30 1.65 2.00 2.39
4
2  10 0.28 0.65 1.00 1.38 1.72 2.06 2.41
5  104 0.32 0.66 1.02 1.39 1.74 2.10 2.45
4
10  10 0.32 0.66 1.03 1.40 1.75 2.11 2.46
(h) J9 beam specimen g ¼ 8:9%
0 0.27 0.57 0.88 1.19 1.52 1.87 2.23
1 0.29 0.62 0.99 1.33 1.69 2.07 2.46
1  104 0.30 0.67 1.06 1.46 1.83 2.23 2.62
4
2  10 0.34 0.74 1.15 1.55 1.95 2.35 2.75
4
5  10 0.35 0.75 1.16 1.58 1.93 2.36 2.78
4
10  10 0.34 0.76 1.17 1.59 1.99 2.40 2.80
 J. Sun et al. / Construction and Building Materials 96 (2015) 404–415 409

Table 2 (continued)

Cycles Measured value of mid-span deflection (mm)

10 kN 20 kN 30 kN 40 kN 50 kN 60 kN 70 kN
20  104 0.37 0.76 1.18 1.60 2.01 2.40 2.82
50  104 0.37 0.77 1.19 1.62 2.02 2.43 2.83
(i) J10 beam specimen g ¼ 10:5%
0 0.28 0.58 0.90 1.22 1.56 1.90 2.27
1 0.29 0.64 0.99 1.34 1.70 2.08 2.46
1  104 0.32 0.67 1.05 1.43 1.79 2.19 2.57
2  104 0.32 0.71 1.10 1.48 1.87 2.25 2.65
5  104 0.35 0.73 1.14 1.54 1.92 2.32 2.72
4
10  10 0.36 0.74 1.16 1.57 1.96 2.37 2.76
4
20  10 0.36 0.75 1.16 1.53 1.96 2.37 2.76
50  104 0.36 0.75 1.16 1.58 1.97 2.38 2.78
(j) J11 beam specimen g ¼ 8:7%
0 0.29 0.60 0.92 1.25 1.60 1.97 2.35
1 0.31 0.67 1.03 1.40 1.78 2.18 2.59
1  104 0.32 0.68 1.10 1.50 1.87 2.29 2.68
4
2  10 0.34 0.74 1.15 1.55 1.96 2.36 2.76
4
5  10 0.34 0.75 1.16 1.57 1.96 2.36 2.77
4
10  10 0.34 0.76 1.17 1.59 1.98 2.39 2.78
20  104 0.34 0.76 1.13 1.59 2.00 2.39 2.80
50  104 0.36 0.76 1.18 1.60 2.00 2.40 2.80
100  104 0.35 0.77 1.18 1.60 1.99 2.40 2.82
(k) J12 beam specimen g ¼ 9:6%
0 0.29 0.59 0.91 1.24 1.58 1.95 2.33
1 0.30 0.65 1.01 1.37 1.75 2.14 2.54
4
1  10 0.33 0.69 1.08 1.47 1.84 2.24 2.64
2  104 0.34 0.73 1.13 1.53 1.93 2.32 2.72
5  104 0.36 0.74 1.15 1.54 1.94 2.34 2.75
10  104 0.36 0.75 1.16 1.57 1.95 2.37 2.76
20  104 0.35 0.76 1.17 1.57 1.97 2.37 2.78
4
50  10 0.35 0.76 1.17 1.59 1.92 2.40 2.80
(l) J13 beam specimen g ¼ 0
0 0.37 0.78 1.20 1.63 2.10 2.60 3.14
1 0.41 0.83 1.37 1.88 2.41 2.99 3.58
1  104 0.45 0.99 1.54 2.10 2.65 3.23 3.81
2  104 0.47 1.04 1.62 2.23 2.77 3.35 3.91
5  104 0.48 1.05 1.63 2.23 2.79 3.35 3.94
10  104 0.50 1.05 1.64 2.24 2.83 3.36 3.96
4
20  10 0.50 1.07 1.65 2.26 2.82 3.39 3.95
4
50  10 0.50 1.07 1.67 2.28 2.86 3.39 4.00
4
100  10 0.51 1.07 1.67 2.26 2.86 3.43 4.00
150  104 0.51 1.07 1.67 2.28 2.88 3.45 4.03
200  104 0.51 1.07 1.68 2.29 2.89 3.47 4.05

After concrete cracked, mid-span deflection of the beam speci-
men continued to increase gradually under repeated loading, and
flexural stiffness further decreased slowly (N 6 5  104 ). Rust layer
on the surface of corroded reinforcement damaged and reduced
the chemical gluing force and static friction force between rein-
forcement and concrete. Corrosion of transverse rib further
decreased the mechanical meshing force between reinforcement
and concrete. These lead to an inconsistent deformation between
reinforcement and concrete when subjected to external loads, the
strain between reinforcement and concrete had the tendency of
becoming uniform. Under the effect of fatigue load, the existing
cracks continued to extend and widen slowly, and new vertical
cracks generated. The inconsistent deformation between reinforce-
ment and concrete was then released, so that the strain between
reinforcement and concrete tended to be consistent between
cracks until stabilized cracking phase formed.
With the continual increase of fatigue cycles, mid-span deflec-
tion of the beam specimen remained mostly unchanged, and the
flexural stiffness maintained stable (N > 5  104 ). This is because
that fatigue cracks had almost entirely formed a stabilized cracking
phase.
The above showed that with the increase of fatigue cycles, mid-
span deflection of the beam specimen after cracking exhibited two
obvious features of stages, that is, slowly increasing stage and
stable stage. Accordingly, it indirectly reflected that flexural stiff-
ness change of the beam specimen also had the characteristics of
two stages, namely, slowly decreasing stage and stable stage.
3.4. Stress–strain curve
In the case of four-point bending fatigue loading, stress ampli-
tude in different sections of the tensile reinforcement differs in
size. For the middle 1/3 section of the tensile reinforcement
between loading points, stress amplitude is maximum and keeps
unchanged. For the side 1/3 section of the tensile reinforcement
between loading point and fulcrum, stress amplitude decreases lin-
early from maximum to zero. Stress–strain curves for the rein-
forcement specimens were obtained through static tensile test, as
shown in Fig. 10. Stress–strain curves used nominal stress that is
calculated on the initial cross-section of steel bar.
It can be seen from Fig. 10(m) that, after cyclic loading to 2 mil-
lion times, uncorroded reinforcing bars showed obvious yield pla-
teau and good ductile characteristics; just with the increase of
fatigue stress amplitude, length of yield plateau slightly shortened,
and ultimate tensile strain became somewhat smaller. It can be
found by contrasting Fig. 10(a) with (m) that, under the condition
410 J. Sun et al. / Construction and Building Materials 96 (2015) 404–415
of no reinforcement corrosion, fatigue effect has no substantial
influence on mechanical properties of uncorroded reinforcing bars.
It can be observed from Fig. 10(b)–(l) that stress–strain curves
for the reinforcement specimens after corrosion fatigue showed
the following obvious characteristics: (1) the yield strength
decreased; (2) the ultimate tensile strain decreased; (3) features
of yield plateau changed, that is, yielding stage and strengthening
stage gradually fused, values of yielding stress changed from a
reciprocating fluctuation in the vicinity of a certain value for
uncorroded reinforcing bars into a fluctuating increase. After corro-
sion fatigue, material properties of the reinforcing bars have grad-
ually changed, that is, it appeared a certain degree of hardening
other than the initial properties of mild steel. The degree of this
change is related to the degree of corrosion fatigue, namely is
related to the degree of corrosion, number of fatigue cycles and
size of fatigue stress amplitude.
From Fig. 10(b)–(m), it can be concluded as follows. For the
reinforcing bars in different beam specimens, with the increase
of corrosion degree, both ultimate tensile strain and yield strength
decreased, the higher the degree of corrosion was, the more obvi-
ous the magnitude of reduction was. For reinforcing bars in the
same beam specimens, under the same number of fatigue cycles,
Fig. 10. Stress–strain curves.
ultimate tensile strain decreased with the increase of fatigue stress
amplitude. Under the effect of higher corrosion degree and fatigue
stress amplitude, yield plateau was gradually not obvious and mild
steel became hardening significantly.
Tensile failure occurred suddenly during the yielding stage of
reinforcement segment M in Fig. 10(c). This is because that fatigue
crack formed in the reinforcing bar under the effect of corrosion
fatigue, as shown in Fig. 11. The emergence of fatigue crack lead
to significant reduction of yield strength and almost complete loss
of ductility. The failure showed a brittle fracture without any warn-
ing. The emergence of fatigue crack is a precursor to fatigue failure,
and is the most direct manifestation of corrosion fatigue damage.
The emergence of fatigue crack also indicates that accumulation
of fatigue damage has almost entered the end, the possibility of
sudden failure increased. Under the stress level of serviceability
sate, the emergence of fatigue crack will result in fatigue failure
without any warning with the continual increase of fatigue cycles.
However the static failure mode for ultimate limit state of load-
carrying capacity has also changed as a result of the emergence
of fatigue crack. Tensile failure occurred suddenly in extremely
low percentage of elongation, and characteristics of ductile failure
no longer appeared, which had the risk of sudden failure.
 J. Sun et al. / Construction and Building Materials 96 (2015) 404–415 411

Fig. 10 (continued)

4. Modified calculation formula for flexural stiffness of the calculation formula for mid-span deflection of homogeneous
corroded RC beams under repeated loading elastic beam in mechanics of materials, as follows.
2
Ml
Usually, RC beam cannot be regarded as the isotropic elastic f ¼k ð1Þ
material. Because of working with cracks, the flexural stiffness of B
RC beam varied with the change of load duration and load magni- where k is a coefficient related to the support conditions and load
tude. Considering that the loading time was short, only short-term forms, which can be easily calculated according to the graph multi-
flexural stiffness of the specimens was discussed in order to apply plication method of structural mechanics, for our test, k ¼ 0:1065; l
412 J. Sun et al. / Construction and Building Materials 96 (2015) 404–415

Supposing that a variable stiffness member is equivalent to an

equal stiffness member, according to the principle in structural
mechanics that rotation between two ends of the member is equal
under the same action of bending moment, the equivalent stiffness
Bs of the equal stiffness member can be obtained. For variable sec-
tion member, the rotation h1 between two ends of the member is
given by

a1 lcr Ms a2 lcr Ms
h1 ¼ þ ð4Þ
B0 Bcr
where B0 is the flexural stiffness of transformed (all concrete) sec-
Fig. 11. Fatigue crack in reinforcement after corrosion fatigue. tion, Bcr is the flexural stiffness of cracked transformed (ignoring
concrete in tension zone) section.
For equal section member, the rotation h2 between two ends of
is calculated span, for our test, l ¼ 1:5 m; f, M and B represent mid-
the member is given by
span deflection, mid-span bending moment and mid-span flexural
stiffness respectively. h2 ¼ lcr Ms =Bs ð5Þ
It is conservative to use the flexural stiffness of cracked section
For equilibrium, h1 ¼ h2 , we have
without considering the beneficial effects of uncracked section
when calculating the deflection of RC flexural member. Flexural 1 a1 a2
stiffness of a cracked flexural member is considered to be variable ¼ þ ð6Þ
Bs B0 Bcr
along the length direction of the member, the flexural stiffness is
minimum at the crack and maximum between the cracks, as indi- Substituting Eqs. (2) and (3) into Eq. (6) gives
cated in Fig. 12(b). A cracked flexural member can be divided into B
several flexural crack elements. Of which, A flexural crack element Bs ¼  2  0    ð7Þ
2
B0
is an element with a crack. And a flexural crack element of length
Mcr
Ms
þ 1  MMcrs Bcr
lcr is assumed to be divided into a whole section element of length
a1 lcr and a cracked section element of length a2 lcr , as shown in Considering the change law for flexural stiffness of beam spec-
Fig. 12(c). As discussed in article 6.5.2, ‘‘Code for Design of imen under the dual effects of corrosion and fatigue, corrosion
Highway Reinforced Concrete and Prestressed Concrete Bridges induced correction coefficient gðgÞ and fatigue induced correction
and Culverts (JTG D62-2004)” of China suggests use of the follow- coefficient gðNÞ were introduced to modify the short-term stiffness
ing a1 and a2 to determine the short-term stiffness (Bs) for deflec- of RC beams.
tion calculation of cracked RC flexural members. gðgÞ
B¼ Bs ð8Þ
gðNÞ
a1 ¼ ðMcr =Ms Þ 2
ð2Þ

a2 ¼ 1  ðMcr =Ms Þ2
where Mcr is the moment at first cracking, Ms is maximum moment
in the member at the stage for which deflection is being calculated.
Fig. 12. Cross section equivalence of a cracked flexural member. (a) Cracked
flexural member. (b) Flexural stiffness distribution in elastic range. (c) Equivalent
cross section.
gðNÞ ¼ c1 lg N þ d1 ð9Þ
ð3Þ
gðgÞ ¼ c2 g þ d2 ð10Þ
where g = corrosion rate, N = number of fatigue cycles, c1, c2, d1 and
d2 = undetermined coefficients.
According to Eqs. (1), 7–10, combined with 616 measured val-
ues of mid-span deflection for the specimens under the fatigue
cycle of 1 time, 1  104 times, 2  104 times, 5  104 times,
10  104 times, 20  104 times, 50  104 times, 100  104 times,
etc., and the step load of 10 kN, 20 kN, 30 kN, 40 kN, 50 kN,
60 kN and 70 kN, it was nonlinearly fitted out that c1 ¼ 0:0912,
c2 ¼ 0:0813, d1 ¼ 4:2828 and d2 ¼ 2:1425. Therefore the modified
calculation formula for flexural stiffness of corroded RC beam
under fatigue loading was quantitatively obtained as follows.
0:0813g þ 2:1425
B¼ Bs ð11Þ
0:0912 lg N þ 4:2828
For the specimens of this test, according to Eqs. (1), (7) and (11),
deviations of mid-span deflection between the values calculated by
the formula and the values measured by the test were statistically
analyzed, as shown in Fig. 13. As can be seen from the graph, devi-
ations of the values calculated by the formula from the values mea-
sured by the test are mostly in 0–15% and most of these deviations
are within 10%, which verifies the reliability of the fitted formula.
Such error is satisfactory for engineering applications and proves
that our proposed modified calculation formula for flexural stiff-
ness of corroded RC beam under repeated loading has certain
applicability, which can provide reference for calculation and eval-
uation of deflection for RC bridge structures under long-term live
load.
 J. Sun et al. / Construction and Building Materials 96 (2015) 404–415 413

stage and necking stage for reinforcing bars after corrosion fatigue
respectively.
Combined with the measured values of 360 turning points in
the stress–strain curves mentioned previously of 60 corroded rein-
forcement specimens after fatigue test, the quantitative relation-
ship model for ry , ey , rh , eh , ru and eu was fitted out as follows.

ry ¼ ry0 ð1  0:0124gÞð1 þ 0:0038 lg NÞ ð13Þ

ey ¼ ey0 ð1  0:0026gÞð1  0:0122 lg NÞ ð14Þ

rh ¼ rh0 ð1  0:0048gÞð1 þ 0:0006 lg NÞ ð15Þ

eh ¼ eh0 ð1 þ 0:0192gÞð1  0:0243 lg NÞ ð16Þ

ru ¼ ru0 ð1  0:0105gÞð1 þ 0:0037 lg NÞ ð17Þ

Fig. 13. Deviation distribution of the modified calculation formula.
eu ¼ eu0 ð1  0:0422gÞð1  0:01 lg NÞ ð18Þ

5. Constitutive relation model for reinforcing bars after where ry0 and ey0 are the stress and strain of turning point between
corrosion fatigue elastic deforming stage and yielding stage for reinforcing bars of
original state respectively, rh0 and eh0 are the stress and strain of
Mechanical properties of the reinforcing bars changed after turning point between yielding stage and plastic hardening stage
corrosion fatigue and the constitutive relation changed corre- for reinforcing bars of original state respectively, ru0 and eu0 are
spondingly. On the basis of constitutive relation model for reinforc- the stress and strain of turning point between plastic hardening
ing bars of original state without corrosion fatigue, which was stage and necking deformation stage for reinforcing bars of original
proposed in the literature [42], combined with the features of state respectively, g is degree of corrosion and the range of corro-
stress–strain curves for reinforcing bars after corrosion fatigue, sion degree for the reinforcement specimens is 0–14.6% in the test,
constitutive relation model for reinforcing bars after corrosion N is number of fatigue cycles and the maximum number of fatigue
fatigue was proposed, as shown in Fig. 14. cycles is 200  104 in the test.
According to the above obtained quantitative model for consti-
ry
r¼ e for 0 6 e 6 ey ð12aÞ tutive relation of reinforcing bars after corrosion fatigue, substitut-
ey ing different degree of corrosion (g) and number of fatigue cycles
(N), we can draw a family of curves for constitutive relation of rein-
or
forcing bars with different degree of corrosion fatigue, as shown in
rh  ry Fig. 15. It can be found from the graph that, under the actual stress
r¼ ðe  ey Þ þ ry for ey < e 6 eh ð12bÞ
eh  ey level of tensile reinforcement in existing RC bridge structures, for
the reinforcing bars with corrosion degree of 5% and fatigue cycles
or of 140  104 times, the yield strength and ultimate tensile strain
 3 are reduced by 5.4% and 25.9% compared with the original state
eu  e respectively; when the corrosion degree reaches 14% and fatigue
r ¼ ru  ðru  rh Þ for eh < e 6 eu ð12cÞ
eu  eh cycles are only 10  104 times, the yield strength and ultimate ten-
sile strain are reduced by 17% and 61.1% compared with the origi-
where ry and ey are the stress and strain of turning point between nal state respectively; characteristics of yield plateau also changes
elastic deforming stage and yielding stage for reinforcing bars after obviously from horizontal line into inclined straight line; the yield-
corrosion fatigue respectively, rh and eh are the stress and strain of ing stage and strengthening stage gradually fuse. Under the cou-
turning point between yielding stage and plastic hardening stage pling effects of fatigue and corrosion, constitutive relation of
for reinforcing bars after corrosion fatigue respectively, ru and eu reinforcing bars changes significantly.
are the stress and strain of turning point between plastic hardening

Fig. 15. Family of curves for constitutive relation of reinforcing bars with different
Fig. 14. Constitutive relation model for reinforcing bars after corrosion fatigue. degree of corrosion fatigue.
414 J. Sun et al. / Construction and Building Materials 96 (2015) 404–415
Fig. 16. Deviation distribution of our proposed constitutive relation model.
For the reinforcement specimens after corrosion fatigue,
statistical analysis results of the deviations between the values
calculated by our proposed constitutive relation model and the val-
ues measured by the test were shown in Fig. 16. As can be seen
from the graph, for the deviations of the calculated values from
the measured values, the vast majority of ey , eh and eu are in the
range of 0–10%, and the vast majority of ry , rh and ru are in the
range of 0–5%. For engineering application, the accuracy is satisfac-
tory. The above analysis results validates the reliability of the
calculation formula for turning points of our proposed constitutive
relation model and proves that the model had certain applicability,
which can provide reference for calculation and evaluation of
corroded RC bridge structures under the action of long-term traffic
load.
6. Conclusions
This paper presents the results of an experimental study on fail-
ure mode and flexural stiffness of corroded RC beams under
repeated loading and mechanical behavior of reinforcing bars after
corrosion fatigue. Brittle fatigue failure occurred in all the corroded
beam specimens after cyclic loading to a certain number of times,
which was marked by a sudden fatigue fracture of one of the ten-
sile reinforcing bars. The fatigue failure was dangerous and with-
out any warning. Fatigue failure did not occur in the uncorroded
beam specimen after 2 million times of cyclic loading. And the
uncorroded beam specimen exhibited good ductile failure charac-
teristics in subsequent monotonic static loading test. It was
showed that fatigue effect had no substantial influence on the sta-
tic performance of RC beams without corrosion.
Under repeated loading, with the increase of fatigue cycles, flex-
ural stiffness change of the corroded beam specimens had two
obvious features of stages, namely slowly decreasing stage and
stable stage. Based on the qualitative analysis of flexural stiffness
change for the beam specimens, correction coefficients were intro-
duced considering the effects of corrosion and fatigue respectively.
And the modified calculation formula for flexural stiffness of cor-
roded RC beam under repeated loading was quantitatively pre-
sented. The formula was verified by comparing the calculated
values with the measured values, which can provide reference
for calculation and evaluation of deflection for corroded RC bridge
structures under long-term live load.
After 2 million times of cyclic loading, the uncorroded reinforc-
ing bars like the reinforcing bars of original state without corrosion
fatigue, had obvious yield plateau and showed good ductile failure
characteristics. Fatigue effect has no substantial influence on
mechanical properties of uncorroded reinforcing bars.
Characteristics of stress–strain curves for reinforcing bars after
corrosion fatigue changed obviously: the yield strength decreased,
the ultimate tensile strain decreased, and features of yield plateau
changed. Material properties of reinforcing bars after corrosion
fatigue changed from initial mild steel into exhibiting a certain
degree of hardening. Degree of the change was related to the
degree of corrosion fatigue.
The emergence of fatigue crack in the reinforcing bar makes the
static failure mode for ultimate limit state of load-carrying capacity
change: the yield strength decreased significantly, and tensile fail-
ure suddenly occurred in extremely low percentage of elongation.
The accumulation of fatigue damage has almost entered the end,
which is dangerous.
Constitutive relation of the reinforcing bars after corrosion fati-
gue changed correspondingly. Combined with the experimentally
obtained stress–strain curves for the reinforcing bars after corro-
sion fatigue, constitutive relation model for the reinforcing bars
in concrete after corrosion fatigue was quantitatively presented.
And the model was verified by comparing the calculated values
with the measured values, which can provide reference for calcula-
tion and evaluation of corroded RC bridge structures under long-
term live load.
Acknowledgements
The presented research greatly appreciates the financial sup-
port by National Natural Science Foundation of China (Grant No.
51278119).
References
[1] J. Rodriguez, L.M. Ortega, J. Casal, Load carrying capacity of concrete structures
with corroded reinforcement, Constr. Build. Mater. 11 (4) (1997) 239–248.
[2] A.A. Torres-Acosta, S. Navarro-Gutierrez, J. Terán-Guillén, Residual flexure
capacity of corroded reinforced concrete beams, Eng. Struct. 29 (6) (2007)
1145–1152.
[3] M. Dekoster, F. Buyle-Bodin, O. Maurel, Y. Delmas, Modelling of the flexural
behaviour of RC beams subjected to localised and uniform corrosion, Eng.
Struct. 25 (10) (2003) 1333–1341.
[4] W. Zhu, R. François, Corrosion of the reinforcement and its influence on the
residual structural performance of a 26-year-old corroded RC beam, Constr.
Build. Mater. 51 (2014) 461–472.
[5] H.H. Vu, N.A. Vu, R. François, Effect of corrosion of tensile rebars and stirrups
on the flexural stiffness of reinforced concrete members, Eur. J. Environ. Civ.
Eng. 18 (3) (2014) 358–376.
[6] L. Wang, Y. Ma, W. Ding, J. Zhang, Y. Liu, Comparative study of flexural
behavior of corroded beams with different types of steel bars, J. Perform
Constr. Fac. (2014).
[7] A.K. Azad, S. Ahmad, B.H.A. Al-Gohi, Flexural strength of corroded reinforced
concrete beams, Mag. Concrete Res. 62 (6) (2010) 405–414.
[8] L. Adelaide, B. Richard, F. Ragueneau, C. Cremona, A simplified numerical
approach of global behaviour of RC beams degraded by corrosion, Eur. J.
Environ. Civ. Eng. 16 (3–4) (2012) 414–439.
[9] G. Malumbela, P. Moyo, M. Alexander, Behaviour of RC beams corroded under
sustained service loads, Constr. Build. Mater. 23 (11) (2009) 3346–3351.
[10] A.N. Kallias, M.I. Rafiq, Finite element investigation of the structural response
of corroded RC beams, Eng. Struct. 32 (9) (2010) 2984–2994.
[11] K. Zandi Hanjari, K. Lundgren, M. Plos, D. Coronelli, Three-dimensional
modelling of structural effects of corroding steel reinforcement in concrete,
Struct. Infrastruct. E 9 (7) (2013) 702–718.
[12] F. Biondini, M. Vergani, Deteriorating beam finite element for nonlinear
analysis of concrete structures under corrosion, Struct. Infrastruct. E 11 (4)
(2015) 519–532.
[13] S.F.U. Ahmed, M. Maalej, H. Mihashi, Cover cracking of reinforced concrete
beams due to corrosion of steel, ACI Mater. J. 104 (2) (2007) 153–161.
[14] I. Khan, R. François, A. Castel, Prediction of reinforcement corrosion using
corrosion induced cracks width in corroded reinforced concrete beams, Cem.
Concr. Res. 56 (2014) 84–96.
[15] A.N. Kallias, M.I. Rafiq, Performance assessment of corroding RC beams using
response surface methodology, Eng. Struct. 49 (2013) 671–685.
[16] D. Coronelli, P. Gambarova, Structural assessment of corroded reinforced
concrete beams: modeling guidelines, J. Struct. Eng. 130 (8) (2004) 1214–
1224.
[17] X. Wang, X. Liu, Simplified methodology for the evaluation of the residual
strength of corroded reinforced concrete beams, J. Perform Constr. Fac. 24 (2)
(2009) 108–119.
[18] M. Pimentel, E. Brühwiler, J. Figueiras, Fatigue life of short-span reinforced
concrete railway bridges, Struct. Concr. 9 (4) (2008) 215–222.
 J. Sun et al. / Construction and Building Materials 96 (2015) 404–415
[19] L. Higgins, J.P. Forth, A. Neville, R. Jones, T. Hodgson, Behaviour of cracked
reinforced concrete beams under repeated and sustained load types, Eng.
Struct. 56 (2013) 457–465.
[20] B.H. Oh, S.H. Kim, Advanced crack width analysis of reinforced concrete beams
under repeated loads, J. Struct. Eng. 133 (3) (2007) 411–420.
[21] S. Ray, J.M.C. Kishen, Analysis of fatigue crack growth in reinforced concrete
beams, Mater. Struct. 47 (1–2) (2014) 183–198.
[22] C. Sousa, R. Calçada, A.S. Neves, Numerical evaluation of the non-linear
behaviour of cracked RC members under variable-amplitude cyclic loading,
Mater. Struct. (2014) 1–24.
[23] F.J.O. Coca, M.U.L. Tello, C. Gaona-Tiburcio, J.A. Romero, A. Martínez-Villafañe,
B.E. Maldonado, et al., Corrosion fatigue of road bridges: a review, Int. J.
Electrochem. Sci. 6 (2011) 3438–3451.
[24] C. Fang, S. Yang, Z. Zhang, Bending characteristics of corroded reinforced
concrete beam under repeated loading, Struct. Eng. Mech. 47 (6) (2013) 773–
790.
[25] M. Oyado, M. Hasegawa, T. Sato, Characteristics of fatigue and evaluation of RC
beam damaged by accelerated corrosion, Quart. Rep. RTRI 44 (2) (2003) 72–77.
[26] W. Yi, S.K. Kunnath, X. Sun, C. Shi, F. Tang, Fatigue behavior of reinforced
concrete beams with corroded steel reinforcement, ACI Struct. J. 107 (05)
(2010) 526–533.
[27] S. Masoud, K. Soudki, T. Topper, Postrepair fatigue performance of FRP-
repaired corroded RC beams: experimental and analytical investigation, J.
Compos. Constr. 9 (5) (2005) 441–449.
[28] C.A. Apostolopoulos, M.P. Papadopoulos, S.G. Pantelakis, Tensile behavior of
corroded reinforcing steel bars BSt 500s, Constr. Build. Mater. 20 (9) (2006)
782–789.
[29] C.A. Apostolopoulos, V.G. Papadakis, Consequences of steel corrosion on the
ductility properties of reinforcement bar, Constr. Build. Mater. 22 (12) (2008)
2316–2324.
[30] A.A. Almusallam, Effect of degree of corrosion on the properties of reinforcing
steel bars, Constr. Build. Mater. 15 (8) (2001) 361–368.
415
[31] M.G. Stewart, Mechanical behaviour of pitting corrosion of flexural and shear
reinforcement and its effect on structural reliability of corroding RC beams,
Struct. Saf. 31 (1) (2009) 19–30.
[32] M.G. Stewart, Spatial variability of pitting corrosion and its influence on
structural fragility and reliability of RC beams in flexure, Struct. Saf. 26 (4)
(2004) 453–470.
[33] M.G. Stewart, Q. Suo, Extent of spatially variable corrosion damage as an
indicator of strength and time-dependent reliability of RC beams, Eng. Struct.
31 (1) (2009) 198–207.
[34] C.A. Apostolopoulos, S. Demis, V.G. Papadakis, Chloride-induced corrosion of
steel reinforcement – Mechanical performance and pit depth analysis, Constr.
Build. Mater. 38 (2013) 139–146.
[35] H. Zheng, A.A. Abel, Fatigue properties of reinforcing steel produced by
Tempcore process, J. Mater. Civ. Eng. 11 (2) (1999) 158–165.
[36] H. Zheng, A. Abel, Stress concentration and fatigue of profiled reinforcing
steels, Int. J. Fatigue 20 (10) (1998) 767–773.
[37] C.A. Apostolopoulos, M.P. Papadopoulos, Tensile and low cycle fatigue
behavior of corroded reinforcing steel bars S400, Constr. Build. Mater. 21 (4)
(2007) 855–864.
[38] R.A. Hawileh, J.A. Abdalla, A. Al Tamimi, K. Abdelrahman, F. Oudah, Behavior of
corroded steel reinforcing bars under monotonic and cyclic loadings, Mech.
Adv. Mater. Struct. 18 (3) (2011) 218–224.
[39] M.M. Kashani, A.J. Crewe, N.A. Alexander, Nonlinear cyclic response of
corrosion-damaged reinforcing bars with the effect of buckling, Constr.
Build. Mater. 41 (2013) 388–400.
[40] M.M. Kashani, L.N. Lowes, A.J. Crewe, N.A. Alexander, Finite element
investigation of the influence of corrosion pattern on inelastic buckling and
cyclic response of corroded reinforcing bars, Eng. Struct. 75 (2014) 113–125.
[41] W. Zhang, X. Song, X. Gu, S. Li, Tensile and fatigue behavior of corroded rebars,
Constr. Build. Mater. 34 (2012) 409–417.
[42] R. Park, T. Paulay, Reinforced Concrete Structures, John Wiley & Sons, New
York, 1975.