Materials and Structures (2008) 41:1251–1262

DOI 10.1617/s11527-007-9323-0

ORIGINAL ARTICLE

Preloaded RC columns strengthened with high-strength

concrete jackets under uniaxial compression
Adilson Roberto Takeuti Æ Joao Bento de Hanai Æ
Amir Mirmiran

Received: 4 April 2007 / Accepted: 16 October 2007 / Published online: 7 November 2007

RILEM 2007

Abstract Although use of high-strength reinforced Nomenclature

concrete (RC) jackets has become common practice Ac,inner Inner concrete area of the jacket, limited by
worldwide, there are still two unresolved issues its transverse reinforcement
regarding the contribution of the original concrete Ac,prim Area of the original column cross section
and the effects of existing loads. Twelve RC-jacketed As,bar Area of the longitudinal reinforcing bars in
columns were tested with and without preloading the original column
under uniaxial compression. Tests showed the entire As,mesh Area of the longitudinal welded wire mesh
core to contribute to the capacity of the jacketed f0 cc Concrete’s compressive strength (uniaxial
column, as long as adequate confinement is provided. test in cylindrical sample)
Also, preloading does not adversely affect the fc,core Compression strength of the core’s
capacity of the jacketed column, while it may concrete
increase its deformability, especially in square sec- fc,jack Compression strength of the jacket’s
tions. Transverse reinforcement in the jacket directly concrete
improves ductility of the strengthened column, espe- fc,jacket Concrete jacket’s compressive strength
cially in circular sections. (cylindrical sample)
fc,prim Compression strength of the primary
Keywords Column
Concrete
High-strength
column’s concrete
Confinement
Jacket
Preloading
Fu Ultimate load of the column
Reinforcement
Strengthening fy,bar Longitudinal steel bars’ yield strength
fy,mesh Tensile strength of longitudinal wires of
welded mesh

1 Introduction
A. R. Takeuti
J. B. de Hanai (&)
Department of Civil Engineering, University of
Sao Paulo, Sao Carlos, Brazil Repair and retrofit of reinforced concrete (RC)
e-mail: [email protected] columns is usually achieved by providing a jacket
made of concrete, steel, or more recently, fiber
A. Mirmiran
Department of Civil and Environmental Engineering, reinforced polymer (FRP) materials. Although not a
Florida International University, Miami, FL 33174, USA new technique, concrete jacketing continues to be one
1252 Materials and Structures (2008) 41:1251–1262

of the most widely used techniques for strengthen- column section may be considered for the analysis, as
ing of concrete columns around the world. The long as adequate confinement is provided.
method has in recent years benefited from the use Another important consideration in RC jacketed
of high-strength concrete (HSC), as shown by columns is the confining effect of the transverse
Takeuti and Hanai [1]. The use of HSC leads to reinforcement in the jacket in comparison to the
thinner jackets, without adversely affecting the effects of the transverse reinforcement in the primary
ultimate load capacity and ductility of the retrofitted column. Ong et al. [8] investigated the suitability of
member. Thinner HSC jackets are also expected to two existing confinement models, based on the
result in less interruption in the functionality of the concept of an effectively confined concrete core, to
structure. predict the behavior of jacketed columns under axial
Research on concrete jacketing has focused on loads. Both full-length and partial-length RC jackets
several aspects of the strengthening process. Dritsos were considered. Also, consideration was given to the
[2] presented an interesting technique to retrofit RC difference between concretes used in the jacket and
columns by pre-tensioning transverse ties in the the primary column. Existing experimental data show
jacket. Alcocer [3], on the other hand, focused on the that RC jacketing enhances the axial load-carrying
effectiveness of RC jacket for rehabilitating different capacity, strain at peak load and post-peak ductility.
levels of damages in an existing column. Testing four However, available confinement models underesti-
sets of jacketed slab–beam–column joints under mated the peak load and the corresponding strain. For
bidirectional cyclic loading, he concluded that RC columns subjected to partial-length jacketing with
jackets could be designed for different levels of prior only one set of stirrups, the predicted peak loads
damage to the structure. showed good agreement with experimental results.
Rodriguez and Park [4] tested four RC column However, the corresponding strains were overesti-
assemblies with pre-1970 arrangements of transverse mated by as much as 20%.
steel reinforcement. Test assemblies represented the This paper presents results from a rigorous exper-
column region between the mid-heights of successive imental program by Takeuti [9] to evaluate the
stories. Two of the columns were jacketed as part of confinement and preloading effects on RC jacketed
the repair process after an initial load test on the columns. In addition, test results are compared
column, while the other two were jacketed before any against three available confinement models to assess
load test on the column. The repair and retrofit of the effect of confining action on the load carrying
columns using RC jackets were both found to be capacity of the jacketed columns.
quite successful.
In practice, it is very difficult to eliminate or
reduce pre-existing loads before strengthening, espe- 2 Experimental work
cially in the case of columns. Therefore, in most
cases the strengthening is carried out on preloaded A total of 12 RC columns were tested as part of a
columns. On the other hand, majority of research thorough experimental plan to study the effect of
reports on RC jacketing refer to non-preloaded preloading and the contributions of the original
concrete columns, and very limited data is available concrete core to the capacity of RC-jacketed
on the effect of preloading on the behavior of columns. Test parameters included preloading of
retrofitted column. Ersoy et al. [5] tested a single the primary column, two types of cross section
pair of columns, with and without preloading, and (square or circular), two levels of compressive
showed 10% reduction in the ultimate capacity of the strength of concrete in the jacket, and three different
preloaded column. types and sizes of transverse reinforcement in the
There are also questions as to the use of effective jacket.
cross section of the existing column that would Figures 1, 2 show the reinforcement configurations
actually resist the loads. Cánovas [6] recommended of the two series of tested columns with square and
neglecting the load-bearing capacity of the primary circular cross sections, respectively. Tables 1, 2
column as a safety measure. On the other hand, tests present mechanical properties of the materials used
by Takeuti and Hanai [7] have verified that the entire in the original sections and the jackets of the square
Materials and Structures (2008) 41:1251–1262 1253

and circular columns, respectively. The specimen Figure 5a illustrates the preloading system, which
name consists of three characters; the first refers to consisted of prestressing steel strands anchored in
the type of cross section (S for square, and C for two double I-beams at the top and bottom of the
circular columns), the second indicates the type of specimen. Figure 5b shows the preloading operation.
transverse reinforcement in the jacket (1 for 2.5 mm One load-cell was placed at one of the anchorages to
diameter welded steel wire mesh at 50 mm spacing, measure the effective load applied onto the primary
and 2 and 3 for 5 mm diameter steel ties and hoops at column.
70 and 50 mm spacing, respectively), and the last Table 3 summarizes the test sequence, which
character identifies the preloading condition (P for consisted of preloading, jacketing, and axial com-
preloaded and N for non-preloaded). pression loading of each specimen to failure. The
predicted capacity of each primary column in the
table was calculated at the preloading age of the
2.1 Instrumentation and test procedure specimen. Preload level varied between 44 and 87%
of the predicted capacity of the primary columns. The
The instrumentation consisted of four (4) Kyowa preloads shown in the table drifted from their initial
displacement transducers with 0.01 mm sensitivity, values, as losses due to creep and shrinkage of
placed at mid-height of the columns, one on each concrete accumulated throughout the entire test
side of the square sections, and diametrically sequence. Figure 6 shows the loss in the load of
opposed on circular sections. Additionally, Kyowa each preloaded column. The losses varied between a
mod.KGF-10-120 strain gages were installed on the minimum of 13% and a maximum of 20% of the
longitudinal and transverse steel reinforcement, as initial value for different columns.
shown in Figs. 3, 4 for the square and circular Pairs of preloaded and non-preloaded primary
columns, respectively. columns were kept in a weather controlled chamber

Fig. 1 Reinforcement primary column jacket

configuration for square
columns PRIMARY COLUMN
120mm
90 m m
Longitudinal reinforcement
15mm

90 mm
90mm

25 mm
4 Ø 8mm length=870mm
Ties
6.3mm Ø @ 90mm
15mm

STRENGTHENED COLUMN
2 0 0 mm
REINFORCEMENT RATIO 1
120mm

90mm

Longitudinal reinforcement
900mm

4 Ø 8mm length=870mm

Welded wire mesh

160mm 1 layer - Ø 2.5mm spacement=50mm

REINFORCEMENT RATIO 2 REINFORCEMENT RATIO 3

160 mm 1 60 mm
160 mm

160 mm

50 mm 50 mm

Longitudinal reinforcement Longitudinal reinforcement

4 Ø 8mm length=870mm 4 Ø 8mm length=870mm
Ties Ties
120mm 5.0mm Ø @ 80mm 5.0mm Ø @ 50mm
40mm 4 0mm
1254 Materials and Structures (2008) 41:1251–1262

Fig. 2 Reinforcement primary column jacket

configuration for circular

15mm
columns PRIMARY COLUMN

Longitudinal reinforcement

120mm
90mm
6 Ø 8mm length=870mm

15mm
Ties
5.0mm Ø @ 50mm

STRENGTHENED COLUMN
190mm
REINFORCEMENT RATIO 1

120mm

100mm
Longitudinal reinforcement
900mm

6 Ø 8mm length=870mm

Welded wire mesh

160mm 1 layer - Ø 2.5mm spacement=50mm

REINFORCEMENT RATIO 2 REINFORCEMENT RATIO 3

Longitudinal reinforcement Longitudinal reinforcement

6 Ø 8mm length=870mm 6 Ø 8mm length=870mm
Ties Ties
5.0mm Ø @ 70mm 5.0mm Ø @ 50mm
120mm
35mm 35mm

Table 1 Properties of square columns

Column Compressive strengtha Transverse reinforcement
Core concrete Jacket concrete Core Jacket
(MPa) (MPa)
Type Size (mm) Spacing (mm) Type Size Spacing
(mm) (mm)

S1P 32.7 54.0 Tieb 6.3 90 Meshc 2.5 50

b
S1N 32.7 54.0 Tie 6.3 90 Meshc 2.5 50
S2P 32.7 80.0 Tieb 6.3 90 Tied 5.0 70
S2N 32.7 80.0 Tieb 6.3 90 Tied 5.0 70
S3P 24.8 81.9 Tieb 6.3 90 Tied 5.0 50
b
S3N 24.8 81.9 Tie 6.3 90 Tied 5.0 50
a
Compressive strength of concrete based on average test results of three cylinders
b
Yield strength = 652 MPa
c
Yield strength = 634 MPa
d
Yield strength = 724 MPa

for 29–64 days. The temperature and relative humidity test. Afterwards, each primary column was strength-
inside the chamber were maintained at 32C and 40%, ened by an RC jacket. For preloaded columns, the
respectively. Both temperature and humidity were strengthening was carried out without removing the
controlled and recorded by a computer throughout the preloading force, as illustrated in Fig. 5c.
Materials and Structures (2008) 41:1251–1262 1255

Table 2 Properties of circular columns

Column Compressive strengtha Transverse reinforcement
Core concrete Jacket concrete Core Jacket
(MPa) (MPa)
Type Size (mm) Spacing (mm) Type Size (mm) Spacing (mm)
b c
C1P 31.4 74.0 Hoop 5.0 50 Mesh 2.5 50
C1N 31.4 74.0 Hoopb 5.0 50 Meshc 2.5 50
b d
C2P 31.4 63.3 Hoop 5.0 50 Hoop 5.0 70
C2N 31.4 63.3 Hoopb 5.0 50 Hoopd 5.0 70
C3P 24.8 77.9 Hoopb 5.0 50 Hoopd 5.0 50
C3N 24.8 77.9 Hoopb 5.0 50 Hoopd 5.0 50
a
Compressive strength of concrete based on average test results of three cylinders
b
Yield strength = 724 MPa
c
Yield strength = 634 MPa
d
Yield strength = 724 MPa

T1

7
3
1 5
T4 T2

2
6
4
8

T3
Fig. 3 Instrumentation plan for square columns (mid-height of the columns). 1, 2, 3, 4—longitudinal strain gages; 5, 6, 7 and 8—
transverse strain gages; T1, T2, T3 and T4—displacement transducers

The specimens were then left in the weather- loads, displacements and strains were recorded using
controlled chamber for an additional period of 56– the high speed Vishay System 5000 of Measurements
171 days. Subsequently, each pair of preloaded and Group.
non-preloaded columns was taken out, and one by
one placed in a servo-controlled universal testing
machine (INSTRON 8506/Custom), as shown in 2.2 Test results and discussions
Fig. 7. After positioning the assembly of the original
column and the preloading system, and placing the Table 4 presents the ultimate load capacity of the
instrumentation, test proceeded with applying the strengthened columns. It is clear that all preloaded
axial load at a rate of 0.005 mm/s in displacement columns, except for one, had higher strength than
control. For the preloaded columns, the load was their non-preloaded counterparts. The ultimate load
applied directly onto the double I-beams. Data for ratio of the preloaded to non-preloaded columns in
1256 Materials and Structures (2008) 41:1251–1262

T1

1
9 10 7 8 11 12 T4 T2
3 2
6 5

T3
Fig. 4 Instrumentation plan for circular columns (mid-height of the columns). 1, 2, 3, 4, 5 and 6—longitudinal strain gages; 7, 8, 9,
10, 11 and 12—transverse strain gages; T1, T2, T3 and T4—displacement transducers

JACKS

STEEL TUBES
LOA D CELL
261.7 mm
976.2 mm

MODEL

D OU B L E- I
BEAMS
261.7 mm

(a) (b) (c)

Fig. 5 Preloading and strengthening procedures. (a) Preloading system, (b) Preloading operation, and (c) Strengthening

each pair varied between 0.98 and 1.14, even for Figure 8a–c show comparisons of the axial load–
preloading at as high a level as 86% of the predicted displacement response for each pair of preloaded and
ultimate capacity of the primary column. Therefore, it non-preloaded square columns. In each figure, the
can be concluded with adequate confidence that level of preload and the axial capacity of the primary
preloading does not affect the strengthening process column are also shown for comparison. Also shown
nor does it adversely affect the load-bearing capacity in each figure are the average longitudinal strains in
of the retrofitted column. RC columns. Figure 8d compares load–deflection
Materials and Structures (2008) 41:1251–1262 1257

Table 3 Preloading and test sequence

Specimen pair Preload Capacity of Age at preloading (Days) Age at jacketing Age at loading
primary columna (Days) (Days)
Initial (kN) Final (kN) (kN)

S1 165.9 143.8 377.0 16 49 106

S2 172.6 138.1 375.2 20 49 105
S3 207.6 178.7 247.1 9 43 214
C1 213.2 184.7 266.5 14 76 139
C2 220.0 183.0 267.3 14 78 137
C3 171.2 148.8 196.8 9 44 214
a
Predicted at the age of preloading

170 210
S1P 180 S3P
165 S2P 205

170
200
160

Load (kN)
Load (kN)

Load (kN)

160 195
155
190
150
150
185

145 140
180

140 130 175

0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 120 140 160 180 200 220
Time (days) Time (days) Time (days)
(a) (b) (c)
220 175
C2P C3P
C1P 220
215 170
210
210 165
Load (kN)

205
Load (kN)
Load (kN)

200 160
200
195
155
190 190
150
185

180 180 145

0 20 40 60 80 100 120 140 0 20 40 60 80 100 120 140 0 20 40 60 80 100 120 140 160 180 200 220
Time (days) Time (days) Time (days)
(d) (e) (f)
Fig. 6 Preloading force loss in tested columns: (a) S1P, (b) S2P, (c) S3P, (d) C1P, (e) C2P, and (f) C3P

curves of all square columns, with the loads for each This similarity is more pronounced in the case of
column being normalized with respect to the corre- circular columns, possibly due to the better
sponding peak load. distribution of the confinement pressure around
Figure 9a–c show load–deflection curves for each the section;
pair of preloaded and non-preloaded circular col- • In some cases, such as S1 and S2, the non-
umns. Figure 9d compares load–deflection curves of preloaded column demonstrated higher ductility
all circular columns, again with the loads for each than its preloaded counterpart. This may be
column being normalized with respect to the corre- attributed to the higher level of damage in
sponding peak load. concrete core of the primary columns, and also
The following observations could be drawn from to a lower efficiency of square ties in providing
the above referenced figures: confinement pressure;
• As the transverse reinforcement ratio increases,
• Behavior of preloaded and non-preloaded col- the column develops a higher initial stiffness.
umns is quite similar, especially before reaching After the peak load, on the other hand, higher
the predicted capacity of the primary column. transverse reinforcement ratios result in an
1258 Materials and Structures (2008) 41:1251–1262

Fig. 7 Axial load test

procedure

6.5 m
Table 4 Rupture test
Specimen pair Peak load (kN) Ratio of peak load of preloaded
results
to non-preloaded columns
Non-preloaded Preloaded

S1 1,557 1675.9 1.07

S2 1,650 1623.7 0.98
S3 1,684 1822.2 1.08
C1 1251.8 1429.6 1.14
C2 1291.5 1436.6 1.11
C3 1303.3 1385.9 1.06

enhanced ductility. For the square columns, the and circular columns. The following observations
ties are not as efficient in providing adequate may be made from these figures:
confinement;
• Circular columns have a better ductility than
• In general, welded steel wire meshes proved
square columns, due to a uniform confinement
efficient, even with small diameter wires.
around the section.
Figure 10a–d show the load–strains for the longi- • For non-preloaded columns, the reinforcement in
tudinal steel reinforcement in the core and the jacket the jacket is stressed similar to the reinforcement
for specimens S2N, S2P, C2N, and C2P, respectively. in the core. In circular columns with no preload-
These figures represent typical behavior of the square ing, this trend remains in effect for the entire
Materials and Structures (2008) 41:1251–1262 1259

o o
Strain ( /oo) Strain ( /oo)
0,00 2,22 4,44 6,66 8,88 11,10 13,32 0,0 0 1 ,11 2 ,2 2 3 ,33 4 ,4 4 5, 5 5 6 ,66 7 ,77 8 , 88 9 ,99 1 1 ,1 0
1800 45 1800 45

1600 S1N 40 1600

S2N 40
S1P S2P
1400 35 1400 35
compressive strength of the primary column concrete compressive strength of the primary column concrete
1200 30 1200 30

Stress (MPa)

Stress (MPa)
Load (kN)

Load (kN)
1000 25 1000 25

800 20 800 20

600 15 600 15

400 10 400 10

200 5 200 preload 5

preload
0 0 0 0
0 2 4 6 8 10 12 0 2 4 6 8 10
Longitudinal displacement (mm) L o n gi t u d i n a l d is p la ce m e n t ( m m )
(a) (b)
o
Strain ( /oo)
0,00 2,22 4,44 6,66 8,88 11,10
2000 50 1,0
1800 S3N 45
S3P

(Load/Ultimate load)
1600 40 0,8
1400 35
Stress (MPa)

1200 30 0,6
Load (kN)

1000 25

800 20 0,4
600 15
compressive strength of the primary column concrete
400 10 0,2
S1N S2 N S3N
200 preload 5
S1P S2P S3 P
0 0 0,0
0 2 4 6 8 10
0 2 4 6 8 10 12
Longitudinal displacement (mm) Longitudinal displacement (mm)
(c) (d)
Fig. 8 Load–deflection response for square columns: (a) S1N and S1P, (b) S2N and S2P, (c) S3N and S3P, and (d) normalized load–
deflection for all square columns

loading, whereas in square columns with no concrete inside the transverse reinforcement, when
preloading, the jacket reinforcement becomes less high-strength concrete is used. This is because
effective after the peak load is reached. This concrete cover does not fully contribute up to failure.
difference may be attributed to the better con- This approach, as expressed below, neglects the
finement effect in circular columns. contribution of the concrete cover of the jacket in the
• In preloaded columns of both square and circular ultimate capacity of the column:
sections, there is a distinct difference between the
Fu ¼Ac;inner fc;jacket þ Ac;primary fc;primary
strains in the steel reinforcement of the jacket and ð1Þ
the core. The difference can be explained by the þ As;bar fy;bar þ As;mesh fy;mesh
value of the preload which is applied only on the where Fu is the ultimate capacity of the column, Ac,inner
core. is the area of concrete jacket inside the transverse
reinforcement, fc,jack is the compressive strength of
3 Analytical investigation concrete in the jacket, Ac,prim is the area of the primary
column, fc,prim is the compressive strength of concrete
The ultimate load of strengthened columns can be in the primary column, As,bar is the area of longitudinal
predicted from the equilibrium of axial forces in the steel in primary column and the jacket, fy,bar is the yield
section assuming strain compatibility and perfect strength of the longitudinal steel, As,mesh is the area of
bond between concrete and the reinforcement. Previ- longitudinal wires of welded mesh, and fy,mesh is the
ous studies have only considered the area of core yield strength of longitudinal wires of welded mesh.
1260 Materials and Structures (2008) 41:1251–1262

o o
Strain ( /oo) S t r a i n ( /o o )
0,00 1,11 2,22 3,33 4,44 5,55 6,66 0,00 1,11 2,22 3,33 4,44 5,55 6,66 7,77
1500 52,90 1500 52,90

C1N C2N
C1P C2P
1200 42,32 1200 42,32

Stress (MPa)
Load (kN)

Stress (MPa)
900 31,74 900 31,74

Load (kN)
compressive strength of the primary column concrete
600 21,16 600 21,16

compressive strength of the primary column concrete

300 10,58 300 10,58

preload preload

0 0,00 0 0,00
0 1 2 3 4 5 6 0 1 2 3 4 5 6 7
Longitudinal displacement (mm) Longitudinal displacement (mm)
(a) (b)
o
Strain ( /oo)
0,00 2,22 4,44 6,66 8,88
1500 52,90 1,0
C3N
C3P

(Load/Ultimate load)
1200 42,32 0,8

Stress (MPa)
Load (kN)

900 31,74 0,6

compressive strength of the primary column concrete

600 21,16 0,4

300 10,58 0,2

C1N C2N C3N
preload
C1P C2P C3P
0 0,00 0,0
0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8
Longitudinal displacement (mm) Longitudinal displacement (mm)
(c) (d)
Fig. 9 Load–deflection response for circular columns: (a) C1N and C1P, (b) C2N and C2P, (c) C3N and C3P, and (d) normalized
load–deflection for all circular columns

The above equation does not include the effect of f 0cc ¼ f 0co
ð1:125 þ 1:25a-n Þ a-w  0:1 ð4Þ
confining action of the transverse reinforcement on the
f 0cc ¼ f 0co
ð1 þ 2:5a-w Þ a-w  0:1 ð5Þ
load carrying capacity of the column. A number of
investigators have developed theoretical models to For all columns tested in this study, a-w  0:1; and
account for such effects. Of these, three are evaluated therefore, Eq. 5 does not apply. Table 5 compares
in this paper. Cusson and Paultre [10] suggested the predictions using Eqs. 1–4 with the test data. The
following relation to predict the compressive strength table shows the average and standard deviation values
of confined concrete, f 0 cc for each of the four equations. It is quite clear that the
"
0:7 #
fle use of confinement models results in better predic-
0 0
f cc ¼ f co
1 þ 2:1
0 ð2Þ tion, while remaining conservative.
f co

where f 0 cu is the compressive strength of unconfined

concrete, and fle is the lateral confinement pressure 4 Conclusions
exerted by the transverse reinforcement. Razvi and
Saatcioglu [11], on the other hand, suggested an A rigorous test program was developed to address
additive relation, as follows two major issues for RC columns strengthened with a
high-strength RC jacket; (a) should the strength
f 0cc ¼ f 0co þ 6:7ðfle Þ0:83 ð3Þ
contribution of the primary be fully considered, and
Finally, EUROCODE 8 [12] suggests the following (b) does the preloading force on the primary column
relation for the compressive strength of confined affect the capacity and ductility of the jacketed
concrete; column.
Materials and Structures (2008) 41:1251–1262 1261

1800
1800
1600
1600 S2N
Core - longitudinal reinforcement 1400
1400 Jacket - longitudinal reinforcement
1200
1200

1000

Load (kN)
1000
Load (kN)

800
800

600 600

400 400 S2P

Core - longitudinal reinforcement
200 200 Jacket - longitudinal reinforcement

0 0
0 2 4 6 8 10 12 14 16 0 1 2 3 4 5 6 7 8
0 0
Strain ( /00) Strain ( /00)
(a) (b)
1600
1400
C2P
1400 Core - longitudinal reinforcement
1200 Jacket - longitudinal reinforcement

1200
1000
C2 N
Core - longitudinal reinforcement 1000
Jacket - longitudinal reinforcement
800 Load (kN)
Load (kN)

800

600
600

400
400

200 200

0 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
0 0
Strain ( /00) Strain ( /00)
(c) (d)
Fig. 10 Load–strain response for longitudinal steel reinforcement in the core and the jacket: (a) S2N, (b) S2P, (c) C2N, and (d) C2P

Table 5 Comparison of confinement equations

Column series Specimen Ratio of theoretical to experimental peak load
Eq. 1 Eq. 2 Eq. 3 Eq. 4

Square columns S1N 0.79 0.82 0.84 0.80

S1P 0.84 0.87 0.90 0.86
S2N 0.97 1.01 1.03 0.99
S2P 0.96 1.00 1.02 0.98
S3N 0.82 0.87 0.90 0.85
S3P 0.89 0.95 0.97 0.92
Circular columns C1N 0.94 0.95 0.94 0.95
C1P 1.07 1.08 1.07 1.08
C2N 0.87 0.94 0.88 1.02
C2P 0.97 1.01 0.97 1.05
C3N 0.94 1.01 0.95 1.15
C3P 1.00 1.07 1.01 1.19
Average 0.92 0.97 0.96 0.99
Standard deviation 0.08 0.08 0.07 0.12
1262 Materials and Structures (2008) 41:1251–1262

A total of 12 RC-jacketed columns were there- References

fore, tested in two series of circular and square
sections, with half of the specimens preloaded 1. Takeuti AR, Hanai JB (1999) Strength and ductility of
reinforced concrete columns strengthened with high-per-
between 44 and 87% of the predicted capacity of formance concrete jackets. In Holand I, Sellevold EJ (eds)
the primary columns. Three different types and Proceedings of international symposium on utilization of
levels of transverse reinforcement were used in the high strength/high performance concrete, Sandefjord,
jacket. The jacketing for preloaded columns was Norway, pp 646–655
2. Dritsos SE (1997) Jacket retrofitting of reinforced concrete
carried out without affecting the preloading system. columns. Construction Repair 11(4):35–40
All columns were then axially loaded to failure. The 3. Alcocer SM (1993) RC frame connections rehabilitated by
following conclusions may be drawn from this test jacketing. J Struct Eng ASCE 119(5):1413–1431
program: 4. Rodriguez M, Park R (1994) Seismic load tests on rein-
forced concrete columns strengthened by jacketing. ACI
Struct J 91(2):150–159
• The entire core section may be considered to 5. Ersoy U, Suleiman R, Tankut T (1993) Behavior of jack-
calculate the load-bearing capacity of the jacketed eted columns. ACI Struct J 90(3):288–293
column, as long as adequate confinement is 6. Cánovas MF (1988) Pathology and rehabilitation of rein-
forced concrete. Pini, Sao Paulo, Brasil (in Portuguese)
provided; 7. Takeuti AR, Hanai JB (2000) Reinforced concrete columns
• Preloading does not adversely affect the capacity strengthened with high performance concrete jackets In:
of the jacketed column, while it may reduce its Johal LSP (ed) Proceedings of the PCI/FHWA/FIB inter-
deformability. The effect is more pronounced in national symposium on high performance concrete: the
economical solution for durable bridges and transportation
square sections due to their less effective structures. PCI, Orlando, FL, pp 439–448
confinement; 8. Ong KCG, Kog YC, Yu CH, Sreekanth APV (2004) Jac-
• The transverse reinforcement in the jacket directly keting of reinforced concrete columns subjected to axial
affects ductility of the strengthened column, espe- load. Mag Concrete Res 56(2):89–98
9. Takeuti AR (2003) Immediate and long-term behavior of
cially in circular section where confinement is reinforced concrete columns strengthened with high-per-
uniform around the section. formance concrete jackets. PhD thesis, University of Sao
• The paper also compared three different confine- Paulo, Sao Carlos, Brazil (in Portuguese)
ment models, and showed that accounting for 10. Cusson D, Paultre P (1994) Confinement model for high-
strength concrete tied columns. Report SMS-93/02, Uni-
confinement effects leads to a better prediction of versity of Sherbrooke, Canada, 54 pp
the load-carrying capacity of the jacketed column. 11. Razvi SR, Saatcioglu M (1994) Strength and deformability
of confined high-strength concrete columns. ACI Struct J
91(6):678–687
Acknowledgments Financial support for this study was 12. EUROCODE 8 (1988) Structures in seismic regions –
provided by the FAPESP—Sao Paulo State Foundation for design part 1, general and building, EC 8, Report EUR
Research Support. Additional support in the form of materials 12266 EN. Industrial Processes, Building and Civil Engi-
was provided by the MBT Brazil Industrial and Commerce neering, Commission of the European Communities
Ltd., Weiler-C. Holzberger & Cia. Ltd., Holcim Brasil S.A.,
Gerdau S.A., and Silmix/Camargo Correa Cimentos S.A.
Findings and opinions expressed in this paper, however, are
those of the authors alone, and do not necessarily reflect the
views of the sponsoring agencies.