Construction and Building Materials 176 (2018) 50–59

Contents lists available at ScienceDirect

Construction and Building Materials

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

Study on the compression performance of small eccentric degradation

columns strengthened with TRC in a chloride environment
Yin Shi-ping a,b,⇑, Hu Xiang-qian b, Hua Yun-tao b
a
State Key Laboratory for Geomechanics & Deep Underground Engineering, School of Mechanics & Civil Engineering, China University of Mining and Technology, Xuzhou
221116, China
b
Jiangsu Key Laboratory of Environmental Impact and Structural Safety in Engineering, School of Mechanics & Civil Engineering, China University of Mining and Technology,
Xuzhou 221116, China

h i g h l i g h t s

TRC can provide better confinement under chloride environment.

Change of failure mode for TRC confined degradation columns due to the increase in reinforcement layers.

The carrying capacity of degradation columns can be increased to a maximum of 28.85%.

Calculation model agrees well with the experimental results.

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

Article history: To explore the mechanical properties of small eccentric compression degradation columns strengthened
Received 16 January 2018 with textile-reinforced concrete (TRC), the effects of using different numbers of textile layers and the
Received in revised form 27 April 2018 degree of corrosion mass loss on the TRC reinforcement were studied. The results showed that compared
Accepted 1 May 2018
with non-corroded columns, the brittle characteristics of the electrified corrosion column were more
Available online 7 May 2018
obvious, and the strengthening effect of the TRC was improved. These characteristics provided good resis-
tance to chloride corrosion and restraint effects for strengthening the columns. The bearing capacity of
Keywords:
the corroded column increased after being reinforced with a TRC layer. However, as the number of rein-
Textile-reinforced concrete (TRC)
Small eccentric compression degradation
forcing layers increased, the increase in the bearing capacity decreased. When the number of TRC rein-
columns forcement layers increased to three layers, the damage pattern of the corroded column changed, and
Number of textile layers the damage position moved from the upper part of the column to the central region of the column.
Degree of corrosion mass loss Finally, the ultimate bearing capacities of the small eccentric compression columns strengthened with
Calculation model TRC in dry-wet environments with chloride salt were calculated, and the calculation model was given.
Ó 2018 Elsevier Ltd. All rights reserved.

1. Introduction damage in the RC structure causes more damage than the other
components. Therefore, it is very important to study RC structures
Reinforced concrete (RC) structures are used in coastal build- in marine environments, especially regarding reinforcing RC
ings, bridges, ports and other structures. However, corrosive mar- columns.
ine environments lead to steel corrosion in the concrete Currently, few studies have focused on repairing and strength-
structures, especially when subjected to dry-wet cycles and ening corroded RC columns. These studies focused on reinforce-
chloride-induced corrosion in tidal areas. The cross-sectional area ment with carbon fiber sheets, in which the steel bars were
decreases and the rust expansion causes the concrete cover to subjected to quick electrified corrosion [2,3]. Tapan et al. [4] noted
crack or peel, causing serious damage and hindering the normal that corroding longitudinal steel bars in the compression zone
use and safety of the structure [1]. More importantly, the column reduced the effective compression zone height of the cross-
section, and the corrosion influenced the bearing capacity of the
eccentric column. Maaddawy [5] and Lee et al. [6] studied the
⇑ Corresponding author at: State Key Laboratory for Geomechanics & Deep mechanical properties of reinforced concrete columns strength-
Underground Engineering, School of Mechanics & Civil Engineering, China ened with carbon fiber sheets after electrified corrosion. In a study
University of Mining and Technology, Xuzhou 221116, China.
on strengthening corroded reinforced concrete columns with a
E-mail address: yinshiping7808@aliyun.com (Y. Shi-ping).

https://doi.org/10.1016/j.conbuildmat.2018.05.003
0950-0618/Ó 2018 Elsevier Ltd. All rights reserved.
 Y. Shi-ping et al. / Construction and Building Materials 176 (2018) 50–59 51

fiber-reinforced polymer (FRP), Maaddawy et al. [7] found that h = 120 mm  150 mm. The height was 800 mm. The thickness of
FRPs have a delayed effect on crack development in columns. the concrete cover was 20 mm. The longitudinal reinforcement of
Although FRPs strengthen corroded reinforced concrete columns, the column was symmetrical and made of four HRB400 steel bars
it is difficult to apply FRPs to strengthen and repair RC columns with a measured yield stress of 427 MPa. The stirrups with 100
in marine environments because they are unsuitable for humid mm spacing in the test region were HPB300 steel bars with a 6.5
environments. mm diameter. To prevent local compression failure at the end of
To provide a new form of corrosion-resistant protection for the column during the compression process, a closed hoop was
steel bars, cement-based materials can be used as an economical needed at the end of the column (the spacing of the hoop reinforce-
solution. Textile-reinforced concrete (TRC) is a new composite ment was 50 mm), and a 10 mm thick steel plate was added to the
material that is composed of multi-axis fiber braids and fine- top of the column. Additionally, the stigma was wrapped with
grained concrete. TRC has a good bearing that restricts cracking, three layers of fiber cloth. The column section sizes and the rein-
is impermeable and provides corrosion resistance. Additionally, forcement configurations are shown in Fig. 1.
TRC exhibits multi-slot cracking and micro-fracture self-healing In this experiment, the chloride environment of dry-wet cycles
abilities [8,9]. The research results from previous studies on refers to putting the specimens into a steel tank with a 5% sodium
strengthening concrete columns with TRC in conventional environ- chloride solution. After soaking for 12 h, the sodium chloride solu-
ments indicated that this is an effective reinforcement method tion in the steel tank was drained, and the sample was left for 12 h.
[10–14]. Ludovico et al. [10] studied the axial compression perfor- This process was one chloride cycle.
mance of concrete columns strengthened with TRC made of basalt
fibers. They found that compared with FRP reinforcement, the TRC 2.2. Test materials and tests
improved the compressive strength and ductility of the columns,
the increase in the compressive strength of the concrete columns 2.2.1. Concrete
strengthened with TRC was similar to that achieved with FRP, In this paper, five reinforced concrete columns were made of
but the ductility performance was better than that with the FRP concrete with a strength grade of C40. The mix proportion was
reinforcement. Additionally, axial compression experiments were as follows: 42.5R ordinary Portland cement: water: sand: stone:
conducted by Ombres [11], and Colajanni et al. [12] using concrete water: reducing agent = 415: 161: 643: 1181: 2.85. The average
columns strengthened with TRC. The results showed that TRC rein- strength of the standard concrete test block was 45.5 MPa after it
forcement significantly improved the strength and ductility of the was cured for 28 days.
concrete columns. Fewer studies have been conducted regarding
the strengthening of eccentric compression columns with TRC.
Trapko [13,14] conducted experiments with 15 elongated rein-
forced concrete columns to investigate the reinforcement effect
under different eccentricities and different methods of strengthen-
ing (vertical or horizontal).
However, fewer studies have been conducted regarding TRC-
strengthened corroded RC columns in marine environments.
Therefore, such studies are required to study the mechanical prop-
erties of deteriorated columns strengthened with TRC under the
action of dry-wet cycles in chloride environments. In this paper,
experiments were conducted considering the influence of different
numbers of textile layers and corrosion on the reinforcement
effect. A model for calculating the normal section bearing capacity
of TRC-strengthened corroded columns in chloride environments is
also proposed. This model provides theoretical guidance for practi-
cal engineering applications.

2. Experimental program

2.1. Experimental design

For these experiments, five columns were designed. X0 was a

non-corroded and unreinforced column, X1 was a corroded and
unreinforced column, and the other columns were corroded and
then reinforced. The parameters of the specimens are shown in
Table 1. The eccentric compression column section size was b  Fig. 1. The sizes and reinforcement of the columns.

Table 1
Parameters of the test specimens.

Specimen Column height Longitudinal reinforcement Number of TRC Number of dry-wet Eccentricity Nominal degree of corrosion
number (mm) (mm) layers cycles (mm) mass loss
X0 800 10 0 180 35 Non-corrosion
X1 800 10 0 180 35 5%
X2 800 10 1 180 35 5%
X3 800 10 2 180 35 5%
X4 800 10 3 180 35 5%
52 Y. Shi-ping et al. / Construction and Building Materials 176 (2018) 50–59

Table 2 applying the field, the specimen was soaked for three to four days
Mechanism properties of fine-grained concrete. to ensure that the concrete in the corroded reinforcement region of
Mechanism properties 1 day 7 days 28 days the column specimen was fully wetted. The positive electrode of
Compressive strength /MPa 18 40 52 the direct current power supply was connected to the electrical
Bending strength /MPa 3.8 6.5 7.8 reinforcement in the column specimen, and the negative electrode
Elastic modulus /MPa 12,600 29,200 32,100 was connected by wires to a copper sheet that was placed in the 5%
Shrink /(mm/m) – 0.5 0.6 NaCl solution. The impressed current density was approximately
Fracture energy GF/(N/m) 40
Fracture toughness KIC/(N/mm3/2) 19
0.6 mA/cm2. The column specimen corroded by the applied electric
field is shown in Fig. 2. To ensure that the current in the specimen
Notes: 1 mm/m = 0.012 in./ft; 1 N/m = 0.0685 lbf/ft; 1 N/mm3/2 = 28.778 lbf/in.3/2. was constant during the test, the electric field-induced corrosion
process was monitored every 12 h, so that error caused by changes
2.2.2. Textile in the current could be adjusted with time.
In this experiment, the textile was a two-dimensional knitted
fabric that was made of carbon fiber bundles and alkali-free glass 2.4. The scheme for strengthening the specimen
fiber bundles. The fibers were perpendicular to each other [15].
The weft direction was the carbon fiber bundles that were used The steps for strengthening the specimen were as follows. First,
to bear loads. The wrap direction was the alkali-free glass fiber the specimen corroded using electric corrosion was chiseled. Then,
bundles that did not bear force aided in fixing the carbon fiber bun- the surface was cleaned with water while keeping the column sur-
dles. The grid spacing of the two materials was 10 mm, and details face moist. Second, the substrate material (fine-grained concrete)
of its mechanical properties and geometric parameters were pro- was applied from the bottom to the top of the column to create a
vided in the literature [15]. reinforcement area with a thickness of 2  3 mm. Third, the textile,
which was pre-cut, was tiled on the plate, and the textile was
2.2.3. Fine-grained concrete painted with a thin layer of mortar (the purpose was to increase
The fine-grained concrete was provided in a previous reference the cohesiveness between the textile and fine-grained concrete
[8]. The mix proportion of cement: fly ash: silica fume: water: fine and to simplify the process for construction). When laying the tex-
sand: coarse sand: water reducer, for fine-grain concrete used, was tile on the column, the reinforced fiber direction (the direction of
475:168:35:262:460:920:9.1. The materials included 52.5R Port- the force) of the textile was required to be parallel to the circum-
land cement, first-class fly ash (FA I), a high-performance water ferential direction of the column specimen. The textile was laid
reducer of a polycarboxylic series, 32-mesh to 64-mesh common smoothly and superimposed well to ensure good contact with
quartz sand (i.e., fine sand), and 26-mesh to 32-mesh common the column. Fourth, the fine-grained concrete was applied. The
quartz sand (i.e., coarse sand). The compressive strength of the thickness was approximately 2 mm, completing the reinforcement
cubes with dimensions of 70.7  70.7  70.7 mm was measured of a layered textile. The above reinforcement steps were repeated
at 52.9 MPa at 28 days. According to reference [16], relevant exper- to add strength with multiple textile layers. Fifth, the fine-
iments have proven that fine-grained concrete possess fluidity grained outer concrete was leveled, and the TRC was maintained
ability, which ensured good adhesive properties with the textile. at a certain age with water. The total thickness of the TRC was
Additionally, the relative slippage phenomenon between the fine- approximately 10 mm, and the scheme for strengthening the col-
grained concrete and the textile did not occur. The hardened prop- umn is shown in Fig. 3.
erties of the fine-grained concrete are shown in Table 2.
2.5. Test points and loading methods
2.3. Corroding the specimen using an applied electric field
Along the side of the small eccentric column, the displacement
First, the specimen was corroded by applying an electrical meters were arranged at the end of the column, the middle part of
field in a 5% NaCl solution that was prepared in advance. Before the column (at 1/2 of the column height) and 1/4 of the column

Fig. 2. Schematic diagram of the electric corrosion.

 Y. Shi-ping et al. / Construction and Building Materials 176 (2018) 50–59 53

Fig. 3. Scheme for strengthening the column: (a) artificial chiseling; (b) applying fine-grained concrete; (c) laying the textile; (d) applying upper fine-grained concrete.

height and were used to test the lateral deflection deformation of started, a preloading step was performed to eliminate the influence
the column. Three 100 mm concrete strain gauges were set at of the bearing offset or settlement, and the instrumentation equip-
equal spacing along the height direction along the column section, ment was checked to determine that it was working properly. The
and two 100 mm concrete strain gauges were set on the surface of instrumentation equipment was returned to zero before the formal
the concrete on the left and right force sides. These gauges were loading step. The loading method adopted monotonic loading step
used to measure the strain law of the column cross-section. Steel by step. The loading rate was 10 kN/min. Each level of loading did
strain gauges were attached to the surface of the mid-span position not exceed 10% of the ultimate load, and when the load reached
of the longitudinal reinforcement of the column to measure the 90% of the calculated bearing capacity, it was increased by less
longitudinal reinforcement strain development under the eccentric than 5% of the ultimate load. After finishing the loading process
compression tests. The displacement meters were indexed as for each level, the load was maintained for eight minutes, and
DM1–DM5, and the strain gauges test points were indexed as the relevant test values were recorded after reading the test instru-
S1–S11. The arrangement of test points is shown in Fig. 4. ment was stable.
Before the loading tests, the specimen was placed into position
using a geometrical pair, and the displacement meter and the other
instrumentation equipment were installed. Before the tests 3. Experimental results and discussion

3.1. Rust phenomenon and crack distribution

Fig. 5 shows the corrosion condition of the specimen after

applying the electrical field. During the corrosion process using
rapid electrification, we found that rust overflowed from the sur-
face of the column specimen as time progressed while applying
the electrical field. Furthermore, the color of the rust deepened
gradually. Brown and red rust can be seen in Fig. 5. At the stirrup
position, the rust on the upper surface of the eccentric compression
column was a horizontal distribution.
Fig. 6 shows the crack distribution of the specimen. After the
specimen was subjected to rapid electrochemical corrosion, the
surface cracks of the specimen columns were caused by rust
expansion on the longitudinal reinforcement. Additionally, the cor-
rosion crack distributions that appeared on the surfaces of the
specimen were not the same on each side. It was found that there
were more cracks on some sides and fewer on other sides.

3.2. Summary of test results

The test results of the specimens are shown in Table 3. After

completing the destruction tests, the damaged specimens were
demolished, and the corroded steel bars in the corrosion specimen
were removed. Then, the steel bars were soaked in 12% dilute
hydrochloric acid solution for 35 min and rinsed with water. After
Fig. 4. Arrangement of test points. drying, the bars were weighed using an electronic scale with a
54 Y. Shi-ping et al. / Construction and Building Materials 176 (2018) 50–59

precision of 0.01 g. The degree of corrosion mass loss was defined

according to Formula (1).
Dm m  m0
q¼ ¼ ð1Þ
m m
where q is the real degree of corrosion mass loss, m is the mass of
the uncorroded bars, and m0 is the mass of the corroded bars.

3.3. Failure modes and analysis

Fig. 7 shows the failure modes of the eccentric compression col-

umns. For unreinforced corrosion column X1, obvious changes
were not observed on the surface of the column when it was sub-
jected to a lighter load. This result is similar to the unreinforced
column X0. However, as the load increased, when approaching
90% of the ultimate load, multiple cracks appeared on one side of
the cross-section that was close to the applied vertical pressure,
and the cracks further extended and developed on the foundation
of the corrosive cracks. The load continued to increase, and cracks
caused by degumming of the fiber cloth, which was used to
strengthen the stigma, continued to form. On one side of the
cross-section that faced away from the applied vertical pressure,
transverse cracks appeared in the upper part of the column and
extended gradually. When the failure load was applied, the con-
crete near the side of the applied vertical pressure was not com-
Fig. 5. Corrosion condition of the column after the electric corrosion. pletely crushed, and it did not burst suddenly. However, a larger
block of the concrete cover peeled off at the upper part of the col-
umn. In contrast, the sound caused by the damage was not the
same as the noise caused during the destruction of the non-
corroded and unreinforced column X0. After strengthening with
TRC, the failure modes of the specimens slightly changed. The
TRC in the compression zone near the vertical side was broken,
and the sizes of the internal concrete fragments were larger than
the non-corroded reinforced eccentric column. The difference
was observed because when three textile layers were used, e.g.,
column X4 as shown in Fig. 7, the damage of the strengthened cor-
roded eccentric column was transferred from the upper part of the
column to the central region.
The damage in column specimen X1 occurred in the upper
region of the column. More concrete blocks fell from X1, and the
ductility was worse than the unreinforced column X0. The different
behavior was attributed to the corrosion of the reinforcement
reducing the overall stiffness and strength of the concrete struc-
ture. The stirrup was denser in the upper region of the column
and exhibited corrosion earlier than the longitudinal reinforce-
ment. The amount of corrosion and the degree of corrosion mass
loss were larger in X1 so that cracking and damage increased in
the upper part of the column. The cracks were more conducive to
erosion by the chloride ions during the process of the dry-wet cycle
in the chloride environment, leading to further concrete deteriora-
tion and cracking [8]. For the reinforced corrosion column X2, the
TRC acted as a hoop constraint after the corroded column was
strengthened with the TRC. Therefore, the crack development of
Fig. 6. Distribution of the cracks. the corrosion column decreased. However, the damage was serious
in the upper region of the column after corrosion, and the

Table 3
Test results of specimens.

Research factors Specimen Ultimate bearing Increment in the bearing Nominal degree of corrosion Real degree of corrosion
number capacity (kN) capacity mass loss mass loss
The number of textile X0 580 41.81% 0% –
layers X1 409 Contrast column 5% 4.01%
X2 489.6 19.71% 5% 4.07%
X3 513.1 25.45% 5% 3.98%
X4 527 28.85% 5% 4.20%
 Y. Shi-ping et al. / Construction and Building Materials 176 (2018) 50–59 55

Fig. 7. Failure modes of columns X0, X1, X2, and X4.

constraint was inadequate. Therefore, damage occurred in the

upper part of the column. When the number of textile layers was
increased to three layers, the constraint capacity of the TRC was
enhanced, and it restrained the damage of the upper part of the
column effectively.

3.4. Load-strain relationship

Fig. 8 shows the cross-section strain distributions of the

strengthened columns. The strain of the concrete in the core area
of the reinforced corrosion column increased as the load increased,
and the strain of the concrete in the core area of the corroded col-
umn was consistent with the plane-section assumption.

3.5. Load-deflection curves at the middle part of the column

Fig. 9 shows the load-deflection curves at the middle part of the

columns. Compared with the unreinforced columns, the bearing Fig. 9. Load-deflection curves of the corroded columns.
capacity of the strengthened columns increased, and the deflection
decreased. During the early loading phases, the curves of the three
reinforced corrosion columns exhibited overlapping sections when changed slowly. At the same time, the differences between the
the eccentric loading was small. When the load increased, the three curves were small after reinforcement was applied. This
curves separated, and the strengthening effect of the TRC became result indicates that the improvement in the bearing capacity
apparent. The deflection of the small eccentric compression col- was limited as the number of layers increased. This was due to
umn strengthened with more textile layers was small. When the the applied electric field accelerating the steel bar corrosion,
number of textile layers increased, the bearing capacity increased, resulting in the eccentric compression column deteriorating and
and the deflection decreased, but the magnitudes of both factors multiple corrosive cracks forming. The crack formation enabled

a X2 b X3
Fig. 8. Cross-section strain distribution of the corroded column.
56 Y. Shi-ping et al. / Construction and Building Materials 176 (2018) 50–59

the chloride ions to corrode the inside of the column more easily, a of the TRC environment was proposed in this paper, and the cal-
resulting in further damage and decreasing the column stiffness. culation model was simplified. The compressive strength of the
However, after the column was strengthened with the TRC, the confined concrete under the action of the dry-wet cycle in the chlo-
cracks caused by the corrosion were repaired, and a protective ride environment is described by formula (2).
layer formed on the surface that effectively restricted the develop-
ment of column cracks and improved the column stiffness. How- f c ¼ f c0 þ 4af el ð3Þ
ever, the efficiency of the improvement diminished after the Using the formula, the influence coefficient (a) of the TRC envi-
column was strengthened continuously. ronment was calculated as 0.2476 after 180 chloride dry-wet
cycles according to the formula in reference [20].
4. Calculating the ultimate bearing capacity of the normal Meanwhile, the combination of compressive stress and cor-
section roded expansion force resulting from the corrosion of the steel
reinforcement places the concrete in a complex state of stress
The principle of strengthening the columns with TRC is similar [21]. Therefore, the compressive strength of the concrete decreased
to using carbon fiber. The reinforcement layers of TRC are thin, and as tensile stress increased. The stiffness also decreased. The actual
therefore, have little influence on the section size of eccentric com- degree of corrosion mass loss caused by the applied electric field
pression columns. Thus, their thickness can be ignored [17]. For was approximately 4% after the test. Corrosive cracks were
TRC reinforcement, fine-grained concrete is used as a binder and observed in the concrete column, and parts existed in which the
the textile aids in restricting the column. For CFRP reinforcement, tensile stress caused by corrosion expansion was not completely
epoxy resin is used as a binder and fiber cloth aids in restricting released at the same time. However, corrosive cracks reduced the
the column. Therefore, in this text, based on the model for effective sectional height of the eccentrically rectangular column.
CFRP-confined concrete, which has been extensively studied and To simplify the calculation, after a comprehensive analysis, the
characterized, the constraint model for the TRC-confined concrete concrete strength grade was decreased by a grade to C35, which
column was determined by analogy. considered the influence of the corrosion cracks on the concrete
strength and the weakened sectional effective height.
4.1. Basic assumptions
4.2.2. The equivalent constraint strength provided by the TRC
(1) The tensile stresses of the fine-grained concrete and con- wrapping
crete were not calculated. Because the fiber material bears the force during the process of
(2) The strain of the cross-section of the degradation column restraining the concrete with FRP or TRC, the restraint effects are
strengthened with TRC meets the plane-section assumption similar [17]. Using the force of the rectangular column restrained
during the compression process. For the convenience of cal- with FRP as a reference [18], the force diagram of the concrete col-
culation, the non-effective constraint by fiber bundles was umn restrained with TRC is shown in Fig. 10. The textile in the TRC
not considered. Only the effective constraint was considered was affected by the section size during the compression process, so
[18]. the constraints provided for the rectangular section were not uni-
(3) The reinforcement layer of the TRC was thin, so the increase form. Additionally, effective and non-effective restraining regions
in the section thickness was ignored to simplify the were present. The non-effective restraining regions are shaded in
calculation. Fig. 10.
(4) The constitutive relation of the corroded steel followed Considering that the section shape and column angle weaken
Hooke’s law. Relative slippage between the concrete and the textile materials, referring to the simplified method in refer-
steel bars, and between the fine-grained concrete and the ence [22], an equivalent circle was introduced to simplify the cal-
textile was not considered. When the column was broken, culations. The diagonal length of the rectangular section is the
the TRC experienced tensile failure, and the reinforcement diameter of the equivalent circle. The section shape coefficient
of the compression zone yielded. (ks ) was also introduced. This coefficient accounted for the weak-
ening effect of the constraint caused by the shape of the section.
4.2. Proposed constraint model

4.2.1. The compressive strength of the confined concrete

According to the test results of the concrete cylinder under tri-
axial compression, the cylinder was still subjected to compressive
stress all around, although the section of the concrete column was
rectangular. The strength of the concrete in the weak constraint
area was not considered. The hoop constraint provided in the
weakest damage zone was considered, namely, the binding
improved by the column angle. The compressive strength f c of
the confined concrete was calculated according to modified for-
mula [19].

f c ¼ f c0 þ 4f el ð2Þ

where f c0 is the compressive strength of the unconfined concrete,

and f el is the equivalent constraint strength provided by the TRC
wrapping.
According to reference [17], it was found that the effect of dry
and wet chloride on the strength of concrete in this age is relatively
small. Therefore, this section did not consider its effect on the com-
pressive strength f c . From reference [20], the influence coefficient Fig. 10. Rectangular column constrained with TRC.
 Y. Shi-ping et al. / Construction and Building Materials 176 (2018) 50–59 57

Additionally, the weakening coefficient (kf ) of the tensile strength convenience during the calculations, we used the section diameter
of the textile was introduced because of the concentration of stress of the zonal fiber bundles as the thickness of a single textile layer:
at the column angle. The tensile strength of the lateral restraint can rffiffiffiffiffiffiffiffi
be converted into the following [23]: 4Af
tef ¼ d ¼ ð7Þ
p
2f ef t ef In the formula: Af is the sectional area of the zonal fiber bun-
f el ¼ ks kf pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð4Þ
2
h þb
2 dles, which is the ratio of the liner density and density of the zonal
fiber bundles. Thus,Af ¼ ð801  105 =1:78Þ  102 ¼ 0:45 mm2 :
The variables in the formula are as follows:
ks is the sectional influence coefficient. kf is the weakening coef- 4.2.3. Yield stress of the corroded reinforcement
ficient of the tensile strength of the textile. f ef is the ultimate ten- The study conducted by Ou et al. [25] showed the degradation
sile strength of the textile, in which the ultimate tensile strength of formula of the mechanical properties when the longitudinal rein-
the weft fiber bundles is 0.75 times that of the monofilament fibers forcement was corroded, and the relationship between the nomi-
[20]. tef is the thickness of the textile. h is the height of the rectan- nal yield strength ratio and the corroded rebars was as follows:
gular section of the concrete column. b is the width of the rectan-
f ysc
gular section of the concrete column. ¼ 1  bfy qð%Þ; bfy ¼ 0:0127 ð8Þ
f ys0
(1) The influence coefficient of the rectangular section (ks )
The variables in the formula are as follows: f ysc is the yield stress of
the corroded reinforcement. f ys0 is the yield stress of the un-
According to references [22,23], the ratio of the area of the
effective constraint area to the entire area restrained with fibers corroded reinforcement. qð%Þ is the degree of corrosion mass
is defined as the shape coefficient of the rectangular section. lossð%Þ. The strain of the steel bar in the compression zone was
measured during the test, and the analysis of the strain confirmed
2 2
b bh  0:86r 2  ð1=3Þ½ðh  2rÞ þ ðb  2rÞ  that the steel bar yielded. Therefore, as shown in Fig. 11, according
ks ¼  ð5Þ to the equilibrium conditions of the force and torque, the formulas
h bh  0:86r2
were determined as follows:
where r is the chamfering radius of the rectangular section, and b 8
N ¼ a1 f c bx þ f y A0s  rs As
0
and h are respectively the width and height of the rectangular sec- >
>
<   0
tion of the concrete column. Nue ¼ a1 f c bx h0  2x þ f yA0s ðh0  a0s Þ ð9Þ
>
>
: rs ¼ nnb 1
fy  f y 6 rs 6 f y
0
(2) The weakening coefficient of the tensile strength of the tex- b b1

tile caused by the concentration of stress at the column The variables in the formula are as follows:
angle (kf ) N is the normal section bearing capacity of the small eccentric
columns, and Nu is the ultimate value of the normal section bear-
Based on references [23,24], the concentration of stress at the ing capacity of the small eccentric columns. rs is the stress value of
column angle section decreases the tensile strength of the fiber the reinforcement on the tensile side that is far away from the ver-
cloth, which is related to the chamfer r of the section and the width tical load. n and nb are the relative height of the compression zone
b of the concrete as follows. and the relative height of the limiting compression zone respec-
 0:38 tively. a1 is the ratio of the stress value of the equivalent rectangu-
2r
kf ¼ 0:3 þ 0:35 ð6Þ lar stress block of the compressive concrete to the axial
b compressive strength of the concrete [26]. b1 is the ratio of the
compression zone depth of the equivalent rectangular stress block
(3) The thickness of a single textile layer (t ef ) to the neutral axis depth [26]. f c is the compressive strength of the
confined concrete. f y is the yielding stress of the steel. x is the
The textile provided the constraint force. Only the constraint height of the concrete compression zone. When x > h, x is equal
force provided by the zonal fiber bundles was considered. For to h. h0 is the effective height of the section, h0 ¼ h  as . e is the

Fig. 11. Calculation diagram of the eccentric column strengthened with TRC.
58 Y. Shi-ping et al. / Construction and Building Materials 176 (2018) 50–59

Table 4
Calculation and test values of the corroded columns.

Specimens Textile layers Diameter of the steel bar (mm) Degree of corrosion mass loss Calculation values (kN) Test values (kN) Error
X2 1 10 4.07% 447.2 489.6 8.7%
X3 2 10 3.98% 521.0 513.1 1.5%
X4 3 10 4.20% 594.4 527.0 12.8%

Table 5
Calculation values and experimental results.

Author Label Cross-section (mm) Height (mm) Eccentricity (mm) Number of Calculation Test values (kN) Error
PBO layer values (kN)
Trapko [13] C_1H_16 200  200 1500 16 1 2099.03 1956.77 7.27%
C_1H_32 32 1 1717.10 1596.00 7.59%
C_2H_16 16 2 2150.34 2043.65 5.22%
C_2H_32 32 2 1758.73 1812.19 2.95%

Ombers and Verre [27] RC-II-1-e30 150  150 1200 30 1 670.51 706.85 5.14%
RC-II- 2-e30 30 2 714.43 734.10 2.68%

distance between the action point of the vertical load and the resul- (2) With the increase in the number of TRC layers, the bearing
tant point of the tension reinforcement (As ), and e ¼ ei þ 2h  as . ei capacity increased gradually, and the amplitude of the
is the initial eccentricity, and as is the distance between the resul- increase decreased. However, when the number of TRC rein-
tant point of the tension reinforcement and the tensile edge of the forcement layers increased to three layers, the damage form
section. of the corroded column changed significantly, and the dam-
age occurred in the central area of the column.
4.3. Model validation and analysis (3) We compared the mechanical properties of the rectangular
section of the concrete restrained with FRP with our sam-
After combining formulas (2)(9), the bearing capacities of the ples. Based on the calculation model of the normal section
TRC-strengthened corrosion columns in the chloride environment bearing capacity of the small eccentric compression col-
with different numbers of textile layers were obtained and com- umns, the influence of dry-wet cycles in a chloride environ-
pared with the test values. We found that the calculation model ment was also considered. Then, we deduced the calculation
accurately calculated the ultimate bearing capacity of the com- model of the normal section bearing capacity of the small
pressive normal section of the small eccentric corrosion columns eccentric corrosion column strengthened with TRC in the
strengthened with one to two layers of TRC. The results are shown chloride environment after multiple dry-wet cycles. The dif-
in Table 4. For the X3 column, the error between the test value and ference between the theoretical calculations and test values
the calculated value was larger, which was attributed to the failure was small.
to make full use of the tensile strength of the fiber bundles.
The accuracy of the model was also demonstrated by comparing
Conflict of interest
the existing test data, as [13,27] for the FRCM-confined concrete
column. As shown in Table 5, the proposed model is accurate
We declare that there are not have any commercial or associa-
and consistent with the experimental results. Most calculated val-
tive interest that represents a conflict of interest in connection
ues for the ultimate load of the specimens [13] was larger than the
with the work submitted.
experimental value because the second-order eccentricity was not
considered. Based on the comparison of error, it was found that
this model more effectively predicted the behavior of columns in Acknowledgments
[27]. The occurrence of this situation may have been related to
the radius of the chamfer radius and the slenderness ratio. The authors gratefully acknowledge the financial support from
the Fundamental Research Funds for the Central Universities
5. Conclusions (2017XKZD09). The experimental work described in this paper
was conducted at the Jiangsu Key Laboratory of Environmental
The experiment results show that TRC has good enhancement Impact and Structural Safety in Civil Engineering in the China
effect on degradation columns and provides good resistance to University of Mining and Technology. Helps during the testing
chloride corrosion and restraint effects for strengthening the col- from staffs and students at laboratory are greatly acknowledged.
umns. But, the limited number of specimens in this paper made
it difficult to provide sufficient data for the proposed model. In
References
the future, more experimental studies will be needed to improve
the model to expand its application. [1] J.A. González, S. Feliú, P. Rodríquez, E. Ramírez, C. Alonso, C. Andrade, Some
questions on the corrosion of steel in concrete. Part II: corrosion mechanism
and monitoring, service life prediction and protection methods, Mater. Struct.
(1) After the dry-wet cycles in the chloride environment, the
29 (2) (1996) 97–104.
strengthening effect of the corroded columns strengthened [2] J. Revathy, K. Suguna, P.N. Raghunath, Effect of corrosion damage on the
with TRC was very good. The TRC provided better anti- ductility performance of concrete columns, Am. J. Eng. Appl. Sci. 2 (2) (2009)
erosion ability and better confinement. It also slowed the 324–327.
[3] X.H. Wang, X.L. Liu, B.R. Deng, Effects of length and location of steel corrosion
brittleness and improved the ductility of the corroded on the behavior and load capacity of reinforced concrete columns, J. Shanghai
column. Jiaotong Univer. (Sci.) 17 (4) (2012) 391–400.
 Y. Shi-ping et al. / Construction and Building Materials 176 (2018) 50–59 59

[4] M. Tapan, R.S. Aboutaha, Effect of steel corrosion and loss of concrete cover on [16] S.P. Yin, Y. Li, Z.Y. Jin, P.H. Li, Interfacial properties of textile-reinforced
strength of deteriorated RC columns, Constr. Build. Mater. 25 (5) (2011) 2596– concrete and concrete in chloride freezing-and-thawing cycle, ACI Mater. J.
2603. 115 (2) (2018) 197–208.
[5] T.E. Maaddawy, Behavior of corrosion-damaged RC columns wrapped with FRP [17] S.P. Yin, C. Peng, Z.Y. Jin, Research on mechanical properties of axial-
under combined flexural and axial loading, Cem. Concr. Compos. 30 (6) (2008) compressive concrete columns strengthened with TRC under a conventional
524–534. and chloride wet-dry cycle environment, J. Compos. Constr. 21 (1) (2017)
[6] H.S. Lee, T. Kage, T. Noguchi, F. Tomosawa, An experimental study on the 04016061.
retrofitting effects of reinforced concrete columns damaged by rebar [18] Z. Wang, D. Wang, S.T. Smith, M. Asce, D. Lu, CFRP-confined square RC
corrosion strengthened with carbon fiber sheets, Cem. Concr. Res. 33 (4) columns. I: experimental investigation, J. Compos. Constr. 16 (2) (2012) 150–
(2003) 563–570. 160.
[7] T.E. Maaddawy, A. Chahrour, K. Soudki, Effect of fiber-reinforced polymer [19] R. Carrazedo, J.B.D. Hanai, Concrete prisms and cylinders wrapped by FRP
wraps on corrosion activity and concrete cracking in chloride-contaminated loaded in compression with small eccentricities, J. Compos. Constr. 21 (4)
concrete cylinders, J. Compos. Constr. 10 (2) (2006) 139–147. (2016) 04016115.
[8] S.P. Yin, M.W. Na, Y.L. Yu, J. Wu, Research on the flexural performance of RC [20] S.P. Yin, Y. Li, H.D. Li, C. Peng, S.X. Ai, Axial compression performance of
beams strengthened with TRC under the coupling action of load and marine reinforced concrete columns strengthened with TRC under chloride dry-wet
environment, Constr. Build. Mater. 132 (2017) 251–261. cycle, China J. Highway Trans. 30 (6) (2017) 230–238 (in Chinese).
[9] A. Pourasee, A. Peled, W.J. Weiss, Fluid transport in cracked fabric-reinforced- [21] R. Capozucca, Detection of damage due to corrosion in prestressed RC beams
cement-based composites, J. Mater. Civ. Eng. 23 (8) (2011) 1227–1238. by static and dynamic tests, Constr. Build. Mater. 22 (5) (2008) 738–746.
[10] M.D. Ludovico, A. Prota, G. Manfredi, Structural upgrade using basalt fibers for [22] L. Lam, J.G. Teng, Compressive behavior of carbon fiber reinforced polymer-
concrete confinement, J. Compos. Constr. 14 (5) (2010) 541–552. confined concrete in elliptical columns, J. Struct. Eng. 128 (12) (2002) 1535–
[11] L. Ombres, Concrete confinement with a cement based high strength 1543.
composite material, Compos. Struct. 109 (6) (2014) 294–304. [23] Y. Pan, S.Y. Cao, D.H. Jing, D.B. Chen, Test and analysis of the axial stress-strain
[12] P. Colajanni, F.D. Domenico, A. Recupero, N. Spinella, Concrete columns relationship of square section concrete columns confined by CFRP under
confined with fibre reinforced cementitious mortars: experimentation and preload, China Civ. Eng. J. 42 (1) (2009) 23–29 (in Chinese).
modelling, Constr. Build. Mater. 52 (2) (2014) 375–384. [24] Y. Pan, R. Guo, H. Li, H. Tang, J. Huang, Analysis-oriented stress–strain model
[13] T. Trapko, Effect of eccentric compression loading on the strains of FRCM for FRP-confined concrete with preload, Compos. Struct. 166 (2017) 57–67.
confined concrete columns, Constr. Build. Mater. 61 (7) (2014) 97–105. [25] Y.C. Ou, Y.T.T. Susanto, H. Roh, Tensile behavior of naturally and artificially
[14] T. Trapko, Behaviour of fibre reinforced cementitious matrix strengthened corroded steel bars, Constr. Build. Mater. 103 (2016) 93–104.
concrete columns under eccentric compression loading, Mater. Des. 54 (2) [26] S.P. Yin, S.L. Xu, H.L. Lv, Flexural behavior of reinforced concrete beams with
(2014) 947–954. TRC tension zone cover, J. Mater. Civ. Eng. 26 (2) (2014) 320–330.
[15] S.P. Yin, S.L. Xu, F. Wang, Investigation on the flexural behavior of concrete [27] L. Ombres, S. Verre, Structural behaviour of fabric reinforced cementitious
members reinforced with epoxy resin-impregnated textiles, Mater. Struct. 48 matrix (FRCM) strengthened concrete columns under eccentric loading,
(1–2) (2015) 153–166. Compos. B Eng. 75 (2015) 235–249.