accelerated corrosion, the test specimens were driedand the length and width of cracks were measured as recent studies showed the potential effects of crack opening to bond performance.1
Trang 1Advances in Mechanical Engineering 1–10
Ó The Author(s) 2015 DOI: 10.1177/1687814015573787 aime.sagepub.com
Experimental study of bond-slip
performance of corroded reinforced
concrete under cyclic loading
Haijun Zhou, Jinlong Lu, Xi Xv, Yingwu Zhou and Feng Xing
Abstract
Reinforced concrete structures exposed to marine environment often sustain high levels of chloride-induced reinforce-ment corrosion, which causes reinforcereinforce-ment corrosion and resulting in degraded performance under cyclic service load-ing This article studied the dynamic bond performance between corroded reinforcing and concrete under force controlled loading Tests were carried out to evaluate the cyclic bond-slip degradation with different reinforcement cor-rosion levels A series of 30 specimens with various corcor-rosion levels of rebar and stirrup were made The specimen was cast as concrete cube with the dimension of 200 mm, and a steel rebar was centrally embedded with two stirrups around The embedded steel rebar and stirrups were corroded using an electrochemical accelerated corrosion tech-nique The corrosion crack opening width and length were recorded after completion of artificial corrosion Then, cyclic loading test was carried out; three different force levels of 24, 36, and 48 kN were adopted The effects of reinforce-ment corrosion rate on crack opening, maximum slip, energy dissipation, and unloading stiffness were discussed in detail
It was found that both reinforcement corrosion rate and crack opening would have significant effects on cyclic bond per-formance Further studies are urgently needed to quantify these effects to the cyclic bond perper-formance
Keywords
Reinforced concrete, bond-slip, corrosion, cyclic loading
Date received: 6 November 2014; accepted: 8 January 2015
Academic Editor: Yu-Fei Wu
Introduction
Corrosion of steel reinforcement embedded in concrete
is an electrochemistry process; it is a major problem
faced by civil engineers and surveyors today as they
maintain an aging infrastructure Studies confirm that
corrosion affected significantly the structural elements
The dynamic and static mechanical performances of
corroded members have been investigated,1–3 and it
was confirmed that their performance was diminished
significantly Now, it has been understood by
engineer-ing professions that the deterioration of performance
was mainly due to the following three factors: loss of
rebar area, spalling of concrete cover, and bond
degra-dation due to steel reinforcement corrosion Earlier
studies found that there existed a limit corrosion rate (mostly smaller than 5% reduction in bar cross sec-tion), and corrosion might have a beneficial effect on bond in deformed bars under this limit.4–7At corrosion levels above this limit, the bond capacity dropped off significantly.5,8
Guangdong Provincial Key Laboratory of Durability for Marine Civil Engineering, Shenzhen University, Shenzhen, China
Corresponding author:
Yingwu Zhou, Guangdong Provincial Key Laboratory of Durability for Marine Civil Engineering, Shenzhen University, Shenzhen 518060, China Email: ywzhou@szu.edu.cn
Creative Commons CC-BY: This article is distributed under the terms of the Creative Commons Attribution 3.0 License
(http://www.creativecommons.org/licenses/by/3.0/) which permits any use, reproduction and distribution of the work without
Trang 2Bond performance is primarily dependent on three
factors: concrete compressive strength, confinement,
and surface of the rebar (deformed or round).9,10
Recent research about the cyclic bond performance for
corroded round and deformed rebar with and without
confinement11showed different performances as stirrup
confinement changed It should be noted that stirrup is
more vulnerable to marine environment corrosion than
rebar as the concrete cover is thinner for stirrup.12
However, little information is available for the effects
of stirrup corrosion on the cyclic bond performance
Furthermore, most of the reported literatures are
mainly focused on the ultimate bond strength by
dis-placement controlled loading,11,13 and little
informa-tion is known about dynamic bond performance under
force controlled loading The cyclic controlled force
loading corresponds to cyclic service loading, such as
fluctuation wind load, repeated running vehicle load,
and so on In this article, the effects of corrosion on
cyclic bond behavior were explored by cyclic loading
tests of 30 accelerated corroded specimens with special
attention to the effects of stirrup corrosion The effects
of reinforcement corrosion on cyclic bond performance
with the correlation of crack opening were discussed
Test setup and procedure
Test specimen
Specimens consisted of deformed steel rebar set in a
concrete prism with two stirrups to provide
confine-ment Polyvinyl chloride (PVC) pipe was used to limit
the bonded length of 80 mm (five times of rebar
dia-meter, Figure 1) The bonded length is much less than
the development length of the rebar, so that the bond
stress along the rebar is relatively uniform Stirrups at
close spacing provide confinement along the bonded
length and help to limit any end effects
Concrete mix design
The concrete was designed to have at least a 30 MPa
compressive strength with a water-cement ratio of 0.41
which conforms to the requirements of Chinese con-crete mix proportion design code.14 The concrete mix content per cubic meter is 187.21 kg water, 456.96 kg ordinary Portland cement, 497.96 kg sand, and 1161.90 kg stone (20 mm) Concrete cubes with dimen-sions of 100 3 100 3 100 mm3were also cast for the compressive strength This mix was found to have a 28-day average compressive strength of 43.2 MPa
Construction
The rebar was dried thoroughly to remove any machin-ing fluid that may come in contact with the surface prior to pouring of the specimens PVC piping was also used to keep the submerged end of the reinforcing dry during the corrosion bath The reinforced rebar was horizontal as concrete was casted in the mold The posi-tion of stirrups was fixed by spacers
Accelerated corrosion
The specimens were corroded using an electrochemical accelerated corrosion technique that involved impress-ing a current through the specimens to accelerate the oxidation process in a 5% NaCl solution Similar setup has been adopted by several researchers.4,8,11,13 To establish different corrosion levels for rebar and stir-rups, rough predictions of the level of corrosion was estimated according to the weight loss of the rebar and stirrup The theoretically calculated amount of corro-sion products, in terms of the duration of electrolytic time, can be expressed by the following Faraday’s law
T =mt3 2 3 F
where T is corrosion duration time, mtis the mass loss,
I is the average electrical current, and F is the Faraday constant
In this study, the current density was set as
150 mA=cm2 in rebar and 300 mA=cm2 in stirrups; the corresponding current was 6 and 77 mA in rebar and stirrups per specimen, respectively The above current density is set according to the reported test results that current density of electrochemical accelerated corrosion
in concrete should not be larger than 500 mA=cm2.15 Specimens were soaked for 5 days prior to applica-tion of the current The accelerated corrosion was actu-ally carried out by two steps: the rebar was corroded first by 10 specimens in series (Figure 2(a)) After the rebar has reached the corrosion time that corresponds
to theoretical corrosion level, the direct-current (DC) power was switched to connect the stirrups for the sec-ond step corrosion by two specimens in series (Figure 2(b)).Figure 2(c) shows the photograph of the acceler-ated corrosion The maximum required artificial Figure 1 Test specimen: (a) front cross-sectional view and (b)
side cross-sectional view (all dimensions in millimeter).
Trang 3corrosion process took approximately 150 days for no.
29 and no 30 specimens
The reinforcement was cleaned using a 12%
hydro-chloric acid solution to remove scale and rust products
and weighed before casting the specimens After the
corrosion process and loading test, the corroded rebar
and stirrup were derived by opening the specimens
Then, they were cleaned and weighed again The real
reduction in cross section was measured as the loss in
weight of stirrup and rebar along the bond length
before corrosion, thereby the loss in weight
represent-ing an average corrosion level for rebar along the bond
length and stirrup The weight loss of reinforcement
could then be derived by the following equation
jR= mR
mR0
jS= mS
mS0
where mR and msare the weight loss of the bond length
rebar and the weight loss of stirrup after removal of the
corrosion products, respectively; mR0 and ms0 are the weight of stirrup and the bond length rebar before cor-rosion, respectively It should be noted that the weight loss of reinforcement could be sometime regarded as the cross-sectional area reduction (corrosion rate) of rebar and stirrup as artificial corrosion was more uni-form compared to natural pitting corrosion
Table 1 lists the tested specimens; the weight loss of reinforcement ranged from 0% to 15% It is found that the measured weight loss of reinforcement is not the same as the theoretical predicted one; however, this has
no effects on the results of this study as only the mea-sured weight loss of reinforcement was used in the fol-lowing studies
Crack opening after artificial corrosion
During accelerated corrosion process, some corrosion products appeared from the corrosion cracks The color
of corrosion products was black green at early stage and turned brown at last This was due to the incom-plete oxidation of corrosion product for lack of oxygen
at the early stage and entirely reacted lately After the
Figure 2 Electrochemical corrosion system: (a) schematic
drawing for rebar corrosion, (b) schematic drawing for stirrup
corrosion, and (c) photograph of corrosion setup.
Table 1 Weight loss ratio of reinforcement, maximum crack width, and total length of tested specimens.
No jR(%) jS(%) Maximum
crack width (mm)
Total crack length (cm)
Trang 4accelerated corrosion, the test specimens were dried
and the length and width of cracks were measured as
recent studies showed the potential effects of crack
opening to bond performance.16 The crack length was
measured by ruler, and the width was measured by
width gauge (3–5 cm reading once and then calculate
the maximum width) It is found that there were mainly
two different kinds of cracks (Figure 3) One was due
to the corrosion of stirrups, and it is in the horizontal
direction and parallel to stirrup location The other was
in the vertical direction, which was due to the
incom-patibility of spacers and concrete, and it was much
shorter than the horizontal one and only near to the
location of spacers The vertical crack has no relation
with corrosion, so they were not included in the data of
total crack length and maximum crack width (Table 1)
in the following analysis
Figure 4 shows the stirrup corrosion rate versus total
crack length and maximum crack width It clearly shows
increasing of both maximum crack width and total
crack length as stirrup corrosion rate increases It was
confirmed that the crack opening is strongly related to
stirrup corrosion.17 This correlation of concrete
crack-ing with stirrup corrosion rate has significant influence
on the bond performance as discussed in the following
Studies also showed that the rebar corrosion rate had
no such relation with maximum crack width and total
crack length for the cover depth of rebar is much larger
than that of stirrup for this specially designed specimen
Loading and measuring instruction
The corroded specimens were tested in a loading
machine with a specially designed and fabricated
loading frame Figure 5(a) shows the schematic draw-ing of the loaddraw-ing and measurdraw-ing system, and Figure 5(c) shows the system photograph Load force was measured through the load cell, and the free-end slip was measured using an extensometer with precision of
60.001 mm The extensometer was attached between the rebar and the bottom surface of concrete as shown
in Figure 5(b) and (d) The cyclic loading was incre-mentally increased or decreased to the maximum or minimum force of 6 24, 6 36, 6 48, and 6 24 kN, with three complete cycles performed for the first three force levels, and two cycles for the end 6 24 kN The last 6 24 kN-level cycles were to compare the perfor-mance of bond-slip behavior after large force ampli-tude cyclic loading All the loads were applied through the computer using displacement–control The loading speed was set as 0.4 mm/min during cyclic loading
Test results General observations
Figure 6 shows typical cyclic bond stress–slip curves of
no 2, no 7, and no 10 specimens The bond stress could be derived from the measured load force by the following formulation
t= P
where l is the bond length, d is the diameter of rebar, and P is the measured load force It should be noted that the rebar corrosion rate of the above three speci-mens is only increased from 1.1% to 2.79%, but stirrup corrosion rate is increased from 4.97% to 11.93% Here, rebar slip in the direction of pulling out is defined
as positive, while that in the opposite direction is defined as negative The three bond-slip loops show the similar ‘‘S’’ shaped curves The observed loops have asymmetrical shapes, especially when the loading force
Figure 3 Cracks after corrosion (jS= 6:30%), specimen no 26.
Figure 4 Stirrup corrosion rate versus total crack length and maximum crack width.
Trang 5becomes large; this may be due to the mixed nature of
concrete The bond stress–slip curve starts at a slip
value of 0 and ascends toward a peak value of the bond
stress For the first three cycles corresponding to load
force level of 6 24 kN, the maximum slip value of
sec-ond and third cycles is close to that of the first cycle,
which means that the degradation of bond is trivial for
the force level of 6 24 kN However, the maximum slip
value increases rapidly after the fourth cycle for the
loading force level of 6 36 kN, and it increases further
rapidly for the force level of 6 48 kN The maximum
slip value for the last two circles of 6 24 kN is much
larger than that of the first three circles of 6 24 kN
The unloading stiffness seems degraded significantly
for the first loading cycle and the last loading cycle,
and this will be discussed later
These above observations illustrated degradation of
bond between reinforcing steel bar and concrete under
cyclic loading It could be confirmed that the corrosion
damaged bond degradation is strongly related to the
maximum loading force level The maximum bond-slip
values of the same force level are far different from
each other for different loading histories This indicates that the bond performance depended on the load his-tory.9 And most importantly, Figure 6 also indicates that different reinforcement corrosion levels have sig-nificant influence on hysteretic loops as discussed in the following
Maximum slip value
Effects of rebar corrosion As the loading force is deter-mined in each cycle, it is interesting to evaluate the cor-responding maximum slip value Certainly, the larger the maximum slip value, the severe the degradation of bonding performance for the tested specimens Figure 7(a) shows the maximum slip versus rebar corrosion levels under different loading cycles for all the tested specimens Here, the maximum slip value is the average
of the maximum and the absolute minimum slip value
of the last loading cycle for each applied force level A tendency of decreasing slip value as rebar corrosion level increases could be observed for rebar corrosion level smaller than 6%; this well accords with other Figure 5 Loading and measuring system: (a) schematic drawing, (b) detail of extensometer installation, (c) photograph of system, and (d) photograph of extensometer.
Trang 6reported results that a slight rebar corrosion will
enhance the bond behaviors However, it is a pity that
most specimens have rebar corrosion level smaller than
6%, and the results of higher rebar corrosion level
could not be confirmed from this test It should be
noted that the points in Figure 7(a) are a bit scattered,
especially when the load force and stirrup corrosion
rate are higher; which may suggest that other factors
may also have effects on the maximum slip value
Effects of stirrup corrosion Figure 7(b) shows the
maxi-mum slip value versus stirrup corrosion level under
dif-ferent loading force cycles for all the tested specimens
It clearly shows that the maximum slip values increase
slightly as the stirrup corrosion levels increase for the
first three 6 24 kN loading force cycles However, as
the cyclic loading force increases, the maximum slip
value increases rapidly as the stirrup corrosion levels
increase And this tendency also exists for the last two
loading cycles of 6 24 kN force level It is confirmed
that the stirrup corrosion level has great effects on bond
performance, and it seems that the bond performance
will be greatly degraded after combined corrosion and
large force loading cycles
To further assess the combined effects of both rebar and stirrup corrosion, Figure 7(c) gives the maximum slip versus stirrup corrosion level by two groups of spe-cimens with rebar corrosion levels of (2.62%–2.95%,
no 6, 10, 13, 14, and 26) and (3.68%–3.98%, no 9, 12,
23, 25, and 28) It clearly shows a linear increasing max-imum slip value with stirrup corrosion rate for the first groups of specimens, and the slope also increases rap-idly with load force levels For the second group, there
is also an increasing tendency for the maximum slip value with stirrup corrosion rate; however, the maxi-mum slip is much smaller and more scattered than that
of the first group The reason might be due to the fact that the second group has higher rebar corrosion rate and thus gives higher restriction to bond-slip as stated above Another reason might be due to the fact that the stirrup corrosion rate of the second group is more scat-tered than that of the first group It could be confirmed that both rebar and stirrup corrosion had significant influence on the maximum slip The above finding is interesting as slight rebar corrosion could increase the cyclic bond performance; on the contrary, stirrup cor-rosion would decrease the cyclic bond performance; this seems to be contradiction evidence However, it is found that most of the corroded specimens have
(c)
Figure 6 Typical bond stress–slip loops: (a) specimen no 2, (b) specimen no 7, and (c) specimen no 10.
Trang 7corrosion crack opening due to stirrup corrosion
(Table 1), and actually, the maximum slip was the
smal-lest for the specimens without crack opening as shown
in Figure 7(b) and (c) The following addresses this
point from the view of crack opening
Effects of crack opening Figure 8 shows the maximum
slip versus maximum crack width for different cyclic
loading force It can be found that the maximum slips
are only slightly changed as stirrup corrosion level
increases for the first 6 24 kN loading force level As
loading force and cycle number increases, although the
points are more scattered than those of Figure 7(a), it
still can be found that there is an increasing tendency
of maximum slip value as crack width increases for
each loading cycle forces It should be noted that wider
corrosion crack means that the crack penetrates deeper
into concrete, which would inevitably reduce the
con-finement and thus reduce bonding performance
Similar results of maximum slip versus total crack
length could also be derived and were not further shown in this article
0.00
0.05
0.10
0.15
0.20
0.25
24kN 36kN 48kN 24kN (2nd)
0.00 0.05 0.10 0.15 0.20 0.25
0.30
24kN 36kN 48kN 24kN (2 nd )
(b)
0.00 0.05 0.10 0.15 0.20 0.25
0.30
No.6,10,13,14 and 26 24kN 36kN 48kN 24kN (2 nd
)
No.9,12,23,25 and 28 24kN 36kN 48kN 24kN (2nd)
(c)
Figure 7 Maximum slip versus reinforcement corrosion ratio at different force levels: (a) rebar, (b) stirrup (all specimens), and (c) stirrup (two groups of specimens).
Figure 8 Maximum slip versus maximum crack width for different cyclic loading forces.
Trang 8It should be noted that the crack opening is related
to stirrup corrosion level (Figure 4) The above
phe-nomena shows the complexity of relationship between
reinforcement corrosion and corrosion crack opening
on bond performance, and the detailed mechanism of
the degradation of bond performance by rebar
corro-sion, stirrup corrocorro-sion, and crack opening still needs
further investigation
Energy dissipation
The energy dissipation in one cycle of loading could be
derived by calculating the area of hysteresis loop of
force–slip curve Figure 9(a) shows energy dissipation
for the first and ninth loops (corresponding to the first
624 kN and the last 6 48 kN loading force cycles)
with different rebar corrosion levels For the first cycle,
the energy dissipation of different rebar corrosion levels
is almost the same For the ninth cycle (Figure 9(b)),
the difference in energy dissipation between rebar
cor-rosion levels is not so obvious, and the points are
scat-tered However, more points corresponding to rebar
corrosion between 3% and 5% are observed with energy dissipation lower than others
Figure 9(b) shows energy dissipation for the first and ninth loops (corresponding to the first 6 24 kN and the last 6 48 kN loading force cycles) with different stirrup corrosion levels It clearly shows that the energy dissipa-tion of different stirrup corrosion levels is almost the same for 6 24 kN force level However, the energy dis-sipation shows an increasing tendency as the stirrup corrosion level increases for the 6 48 kN force level To further assess the combined effects of both rebar and stirrup corrosion, Figure 9(c) gives the energy dissipa-tion versus stirrup corrosion level by the above two groups of specimens It clearly shows a similar phenom-enon observed in Figure 7(c) and confirmed that both rebar and stirrup corrosion had significant influence on the energy dissipation
It should be noted that the energy dissipation in one cycle is related to the damping capacity of structures The above-observed phenomena may verify the observed test results that the corroded reinforced con-crete members have higher damping than that of
(c)
Figure 9 Energy dissipation versus reinforcement corrosion level: (a) rebar, (b) stirrup (all specimens), and (c) stirrup (two groups
of specimens).
Trang 9corroded when loading force increases as stirrup is
more vulnerable to corrosion.18
Unloading stiffness
The unloading stiffness is defined as
k = Pm
dm d0
ð5Þ
where Pm is the maximum or minimum loading force,
dm is the slip value corresponding to the maximum or
minimum loading force, and d0 is the slip value when
loading force is unloaded to 0
Figure 10(a) shows the corresponding unloading
stiffness of each cycle for no 2, no 7, and no 10
speci-mens Figure 10(a) shows that there are only slight
dif-ferences of the forward and reversed unloading stiffness
for a specimen It also clearly shows a decreasing
ten-dency of unloading stiffness as the loading cycles
increase However, the unloading stiffness does not
decrease monotonously, especially when the loading
force level changes (cycle no 7 and no 4) The
unload-ing stiffness of specimen no 2 is obviously larger than
that of specimen no 7 and no 10 It should be noted
that the stirrup corrosion rate of specimen no 2
(4.97%) is the smallest among the three specimens, and the stirrup corrosion rate of specimen no 7 (9.05%) is obviously smaller than that of no 10 (11.93%) However, there is only slight difference between the unloading stiffness of specimen no 7 and no 10 Further study shows that the maximum crack width and total crack length of specimen no 7 and no 10 are nearly the same, and this will be discussed in the following
Figure 10(b) shows the forward unloading stiffness
of specimen no 1, no 2, and no 8 The three specimens have the rebar corrosion rate from 0.30% to 5.57% and nearly the same stirrup corrosion level (Table 1) The reversed unloading stiffness is very close to forward unloading stiffness and is not shown here It clearly shows that the unloading stiffness of specimen no 1 is obviously larger than that of specimen no 2, no 8, no
7, and no 10 Here, the specimen no 1 is one of the spe-cimens that has no crack opening; and the stirrup cor-rosion rate of specimen no 1 (3.77%) is the smallest among the five specimens The rebar corrosion rate of specimen no 2 (1.1%) is obviously smaller than that of
no 8 (5.57%) However, there is only slight difference between the unloading stiffness of specimen no 2 and
no 8 It should be noted that although the rebar corro-sion rate of specimen no 2 is smaller than that of speci-men no 8, the maximum crack width and total crack length of specimen no 2 are a bit larger than that of spe-cimen no 8 The above analysis also shows a complex relation of the bond performance between the reinforce-ment corrosion rate and crack opening as discussed above
Conclusion
This article reported a preliminary study of reinforce-ment corrosion effects on bond performance under force controlled cyclic loading It is found that
(1) Stirrup corrosion will reduce the confinement
on the concrete and thus degrade bond perfor-mance; however, slight rebar corrosion seems
to increase the bond performance;
(2) Maximum crack width and total crack length are strongly related to stirrup corrosion level for this test It is also confirmed that there are also strong relevance of crack opening to degradation of bond performance;
(3) Bond performance degraded more severely when cyclic loading amplitude becomes large for members with highly corroded stirrups; (4) When the cyclic loading force is relatively small, the energy dissipation shows little differ-ence for the specimens with different reinforce-ment corrosion rate However, the energy
(a)
(b)
Figure 10 Unloading stiffness versus cycle number: (a)
specimen no 2, no 7, and no 10; and (b) specimen no 1, no 2,
and no 8.
Trang 10dissipation increases rapidly when stirrup
cor-rosion increases as cyclic loading force becomes
large;
(5) Unloading stiffness degraded more severely for
specimens with crack opening
This preliminary study suggested the importance
and complexity of reinforcement corrosion to bond
performance under cyclic loading Both stirrup and
rebar corrosion would have significant effects on
bond performance, and crack opening, which is
related to the reinforcement corrosion in nature, also
has effects on bond performance And further
investi-gation to quantify these effects will be carried out in
the near future
Declaration of conflicting interests
The authors declare that there is no conflict of interest.
Funding
The work described in this article was financially supported
by the Ministry of Science and Technology for the
973-proj-ect (no 2011CB013604), the National Natural Science
Foundation of China (grant nos 51378313, 51378314), the
Education Department of Guangdong Province (grant no.
2013KJCX0157), and the Science and Creativity Committee
of Shenzhen (grant no JCYJ20120614085454232), to which
the writers are grateful.
References
1 Biondini F, Camnasio E and Palermo A Seismic
perfor-mance of concrete structure in aggressive environments in
Life-Cycle of Civil Engineering Systems In: Proceedings
of the 2nd international symposium on Life-Cycle Civil
Engineering, Taipei, Taiwan, 27–31 October 2010 Taipei,
Taiwan: Taiwan Tech, National Taiwan University of
Science and Technology.
2 Cairns J, Du Y and Law DW Structural performance of
corrosion-damaged concrete beams Mag Concr Res
2008; 60(5): 359–370.
3 Ou YC, Tsai LL and Chen HH Cyclic performance of
large-scale corroded reinforced concrete beams
Earth-quake Eng Struct Dynam 2012; 41: 593–604.
4 Almusallam AA, Al-Gahtani AS, Aziz AR, et al Effect
of reinforcement corrosion on bond strength Construct Build Mater 1996; 10(2): 123–129.
5 Bhargava K, Ghosh A, Mori Y, et al Suggested empiri-cal models for corrosion-induced bond degradation in reinforced concrete J Struct Eng ASCE 2008; 134: 221–230.
6 Mangat PS and Elgarf MS Bond characteristics of cor-roding reinforcement in concrete beams Mater Struct 1999; 32: 89–97.
7 Yalciner H, Eren O and Sensoy S An experimental study
on the bond strength between reinforcement bars and concrete as a function of concrete cover, strength and corrosion level Cement Concr Res 2012; 42: 643–655.
8 Fang CQ, Lundgren K, Chen L, et al Corrosion influ-ence on bond in reinforced concrete Cement Concr Res 2004; 34(11): 2159–2167.
9 Harajli MH Bond stress–slip model for steel bars in unconfined or steel, FRC, or FRP confined concrete under cyclic loading J Struct Eng ASCE 2009; 135(5): 509–518.
10 Li J, Gao X and Zhang P Experimental investigation on the bond of reinforcing bars in high performance concrete under cyclic loading Mater Struct 2007; 40: 1027–1044.
11 Fang CQ, Gylltoft K, Lundgren K, et al Effect of corro-sion on bond in reinforced concrete under cyclic loading Cement Concr Res 2006; 36(3): 548–555.
12 Xue X, Seki H and Song Y Shear behavior of RC beams containing corroded stirrups Adv Struct Eng 2014; 17(2): 165–177.
13 Kivell A Effects of bond deterioration due to corrosion on seismic performance of reinforced concrete structures Master Thesis, University of Canterbury, Christchurch, New Zealand, 2012.
14 Specification for mix proportion design of ordinary con-crete (JGJ 55-2011) Beijing, China: China Architecture
& Building Press, 2011 (in Chinese).
15 Coronelli D, Hanjari K and Lundgren K Severely cor-roded RC with cover cracking J Struct Eng ASCE 2013; 139: 221–232.
16 Law DW, Tang D, Molyneaux TKC, et al Impact of crack width on bond: confined and unconfined rebar Mater Struct 2011; 44: 1287–1296.
17 Firouzi A and Rahai AR Prediction of extent and likeli-hood of corrosion-induced cracking in reinforced con-crete bridge decks Int J Civ Eng 2011; 9(3): 183–192.
18 Abdul Razak H and Choi FC The effect of corrosion on the natural frequency and modal damping of reinforced concrete beams Eng Struct 2001; 23: 1126–1133.