Construction and Building Materials 23 (2009) 15081520

Contents lists available at ScienceDirect

Construction and Building Materials

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

Interaction diagram methodology for design of FRP-conned reinforced

concrete columns
Silvia Rocca a,*, Nestore Galati a, Antonio Nanni b,c
a

Structural Group Inc., Strengthening Division, Hanover, MD, USA

Department of Civil, Architectural and Environmental Engineering, University of Miami, Coral Gables, FL, USA
c
Department of Structural Engineering, University of Naples Federico II, Naples, Italy
b

a r t i c l e

i n f o

Article history:
Received 8 March 2007
Received in revised form 22 October 2007
Accepted 25 June 2008
Available online 8 August 2008
CE database subject headings:
Bending
Columns
Compression
Connement
Design
Ductility
FRP
Interaction diagram
Reinforced concrete
Strengthening

a b s t r a c t
This paper presents a procedure that allows the construction of a simplied axial load bending moment
interaction diagram for FRP-wrapped Reinforced Concrete (RC) columns of circular and non-circular
cross-sections for practical design applications. In the proposed methodology, the analysis of FRP-conned columns is carried out based on principles of equilibrium and strain compatibility equivalent to that
of conventional RC columns, the primary difference being the use of the stressstrain model for FRP-conned concrete developed by Lam and Teng. Based on the consideration that the strength enhancement is
of signicance in members where compression is the controlling failure mode, only the portion of the
interaction diagram corresponding to this type of failure is the focus of the methodology. Experimental
evidence from RC specimens with a minimum side dimension of 300 mm (12 in.) and subjected to combined axial compression and exure was collected and compared to the theoretical interaction diagrams.
Even though limited experimentation has been conducted in the compression-controlled region for such
type of members, data points appear to be consistent with the analytical predictions. A design method for
RC members is therefore proposed following the principles of the ACI building code.
2008 Elsevier Ltd. All rights reserved.

1. Introduction
Columns are structural members subjected to combinations of
axial compression and bending moment, rather than pure axial
loading. The exural effect may be induced by different factors,
such as unbalanced moments at connecting beams, vertical misalignment, or lateral forces resulting from wind or seismic
activity.
Connement of reinforced concrete (RC) columns by means of
ber reinforced polymers (FRP) jackets is a technique being frequently used to seek the increment of load carrying capacity
and/or ductility of such compression members. The need for improved strength results from higher load capacity demands because of design/construction errors, change in the facility use, or
revisions of code requirements. Improving ductility stems from
the need for energy dissipation, which allows the plastic behavior
of the element and, ultimately, of the structure. Ductility enhancement is typically required in existing columns that are subjected to
a combination of axial load and bending moment because of reasons similar to those listed for strengthening. Among these rea* Corresponding author.
E-mail address: silvia.rocca@gmail.com (S. Rocca).
0950-0618/$ - see front matter 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.conbuildmat.2008.06.010

sons, seismic upgrade and correction of detailing defects (i.e.,

improper splicing of the longitudinal reinforcement or lack of
transverse ties) are most common.
Several theoretical and experimental studies have been conducted on the behavior of FRP-wrapped concrete columns subjected to axial force and exure and have produced theoretical
axial forcemoment (PM) interaction diagrams [16]. Other
works have been conducted in the context of concrete-lled
FRP tubes [79]. Fam et al. [9] in particular performed experimental and analytical modeling considering the variability of
FRP-connement effectiveness as a result of the presence of exural moments in a column. They proposed an analytical procedure to determine a PM diagram for circular cross-sections
accounting for the variability of concrete connement as a result
of the gradual change of the biaxial state of stress developed in
the FRP shell as the eccentricity of the axial force changes. As
an overall observation, all the previous studies conclude that signicant enhancement due to FRP-connement is expected in
compression-controlled RC members.
This paper presents a methodology for the construction of a
simplied PM diagram following principles of equilibrium and
strain compatibility, and it is limited to the compression-controlled region of the interaction diagram. The procedure is based

S. Rocca et al. / Construction and Building Materials 23 (2009) 15081520

1509

Nomenclature
Ac
Ae
Ag
As
Asi
b
CE
dsi
D

e
E2
Ec
Ef
fc0
0
fcc

ffu
ffu
fl
fsi
fy
H
h
Mmax
Mn
n
Pa
Pmax
Pn
r
tf

cross-sectional area of concrete in column = Ag(1  qg)

effectively
conned
area
=
Ag  h  2r2
2
b  2r =3  qg Ag
total cross-sectional area
area of steel reinforcement = Agqg
cross-sectional area of ith layer of longitudinal steel
reinforcement
short side dimension of a non-circular cross-section
environmental reduction factor
distance from position of ith layer of longitudinal steel
reinforcement to geometric centroid of the cross-section
diameter of circular cross-section or diameter of equivalent circular column for non-circular cross-sections =
p
2
2
b h
eccentricity of axial load
slope of linear portion of conned stressstrain curve =
0
 fc0 =eccu
fcc
initial modulus of elasticity of concrete
tensile modulus of elasticity of FRP
characteristic concrete compressive strength determined from standard cylinder
maximum compressive strength of conned concrete
ultimate tensile strength of FRP

design ultimate tensile strength of FRP = C E ffu
connement pressure due to FRP jacket
stress in ith layer of longitudinal steel reinforcement
yield strength of longitudinal steel reinforcement
height of column
long side dimension of a non-circular cross-section
maximum bending moment
nominal bending moment capacity
number of FRP plies composing the jacket
constantly applied axial load
maximum applied axial load
nominal axial load capacity of a RC column
corner radius of non-circular cross-sections
FRP nominal ply thickness

on ACI protocols and limiting values (i.e., Pn and Mn multiplied by

strength reduction factor /; ultimate axial strain of unconned
concrete ecu equal to 0.003). Additionally, the paper presents
experimental evidence on FRP-wrapped RC columns subjected to
combined compression and exure and having a size representative of real scale members. The experimental results are compared
to the theoretical simplied PM diagrams obtained following the
proposed methodology. Data points appear to be consistent with
the analytical predictions. The proposed simplied PM diagram
could be used by practitioners as a design tool and for a clear illustration of the procedure; an example application is presented in
this paper.
The symbols used throughout the paper are dened in a notation list. Some of these symbols are also dened in the text for
clarity.

a1
b1
e0c

eccu
ecu
efe
efu
efu
esy
e0t

ja
jb
je

qf
qg
wf

factor relating the uniform compressive stress of the

equivalent rectangular block in the compression zone
to the compressive strength fc0 0:85
factor relating depth of equivalent rectangular compressive stress block to neutral axis depth
axial compressive strain corresponding to fc0
0:002 mm=mm
ultimate axial compressive strain of conned concrete
ultimate axial compressive strain of unconned concrete = 0.003 mm/mm
FRP effective strain (strain level reached at failure)
ultimate tensile strain of the FRP
design ultimate tensile strain of the FRP = CEefu
strain corresponding to the yield strength of steel reinforcement
transition strain in stressstrain curve of FRP-conned
concrete. It corresponds to point of change between initial parabola and straight line = 2fc0 =Ec  E2
strength reduction factor
efciency factor for FRP reinforcement in the determi0
(based on the geometry of the cross-secnation of fcc
tion)
efciency factor for FRP reinforcement in the determination of eccu (based on the geometry of the cross-section)
efciency factor for FRP strain proposed to account for
the difference between the actual rupture strain observed in FRP conned specimens, and the FRP material
rupture strain determined from tensile coupon testing
volumetric
ratio
of
FRP
reinforcement
=

4ntf =D Circular
2ntf b h=bh Non-circular
ratio of the area of longitudinal steel reinforcement to
the cross-sectional area of a compression member =
As/Ag
additional FRP strength reduction factor

considers an appropriate stressstrain model for FRP-conned concrete in the compression zone.
3. Experimental behavior of FRP-conned RC columns

2. Research signicance

The experimental evidence on real-size type specimens,

although not extensive, consists of RC columns of circular and
non-circular cross-sections subjected to a combination of axial
compressive load and exure [4,1018]. The collected experiments
from the available literature mainly featured an applied constant
axial load and a moment induced by a lateral force. Only one study
[4] considered RC columns axially loaded at various eccentricities
covering cases of pure compression to pure bending. Additionally,
the FRP jacket of this study was characterized by bidirectional oriented bers (0/90), and it was applied along the entire height of
the specimens. The database was assembled under the following
restrictions:

The purpose of this study is to provide a methodology for the

construction of a simplied axial forcebending moment interaction diagram for FRP-wrapped RC columns for practical design
applications. The proposed method is based on principles of equilibrium and strain compatibility. The analysis of the FRP-conned
specimens is equivalent to that of conventional RC columns but

 RC columns of circular and non-circular cross-sections with one

minimum dimension of the cross-section set at 300 mm (12 in.),
with a maximum side-aspect ratio (h/b) of 2.0 in the case of noncircular cross-sections.
 FRP-wrapped RC columns with FRP bers oriented perpendicular to the longitudinal axis of the column.

1510

S. Rocca et al. / Construction and Building Materials 23 (2009) 15081520

 Concrete-lled FRP tubes, RC columns part of environmental

research studies, and RC columns that were damaged or loaded
prior to strengthening were not considered in this study.
 RC columns exhibiting a shear type of failure were not included.

4. PM interaction diagram
The analysis of the FRP-conned specimens is equivalent to that
of customary RC columns [20,21] with the primary difference
being the use of a selected stressstrain model for FRP-conned
concrete in the compression zone. The following assumptions are
considered for the analysis: (a) plane sections remain plane, (b)
the tensile strength of concrete is neglected, and (c) complete composite action is assumed between both steel reinforcement-concrete and FRP-concrete.
For ease of calculation, rather than a continuous curve, a conservative PM diagram was constructed as a series of straight lines
joining the axial load and moment values corresponding to ve
characteristic points (Fig. 1) [20]:

Table 1 presents a total of 13 RC columns of circular cross-section (including control and strengthened units) divided into 3 sets
of experiments [11,13,16]. The information presented in the table
is organized as follows: the rst and second columns in the table
show the reference to each experimental set and the specimen
codes, respectively. The type of FRP jacket, when applicable, is given in the third column. Geometrical and material properties are
shown in the columns (4) to (14) in this order: diameter of the
cross-section (D), height of the specimen (H), ratio of the area of
longitudinal steel reinforcement to the cross-sectional area of the
member (qg), volumetric ratio of transverse steel reinforcement
to concrete core (qt), volumetric ratio of FRP reinforcement (qf),
unconned concrete compressive strength fc0 , yield strength of
longitudinal steel reinforcement (fy), tensile modulus of elasticity
of FRP (Ef), ultimate tensile strength of FRP (ffu), ultimate tensile
strain of the FRP (efu), and FRP nominal ply thickness (tf). The last
two columns in the table ((15) and (16)) report the maximum axial
load (Pa or Pmax) and maximum moment (Mmax), respectively. The
specimens included in this table have unconned concrete compressive strength varying from 36.5 MPa (5.29 ksi) to 44.8 MPa
(6.5 ksi), and the level of constantly applied axial load in terms of
Pa =Ag fc0 varies from 0.12 to 0.65.
Table 2 presents a total of 48 RC columns of non-circular crosssection including control and strengthened units featuring side-aspect ratios (h/b) of 1.0, 1.75, and 2.0. These experiments are divided
into eight sets [4,10,12,1418]. The information corresponding to
each set of experiments is presented as in Table 1, with the only
difference being on the cross-section dimension parameters. Specically, Table 2 shows the short side dimension (b), long side
dimension (h), and the side-aspect ratio (h/b) of the non-circular
cross-section instead of diameter (D) from Table 1. In the cases
of specimens from three sets [1517], the radius of the chamfered
corner of the strengthened specimens was not reported in the
source publications. Therefore, a minimum corner radius of
13 mm (0.5 in.) as recommended by the current ACI440.2R-02
[19] was assumed in these columns for the purpose of this study.
The specimens included in this table have unconned concrete
compressive strength varying from 17.9 MPa (2.6 ksi) to 44.2 MPa
(6.41 ksi), and the level of applied axial load in terms of P a =Ag fc0
varies from 0.12 to 0.65.

 Point A: uniform axial compressive strain of conned concrete

eccu (or ecu for the case of an unstrengthened cross-section).
 Point B: strain distribution corresponding to a maximum compressive strain eccu (or ecu) and zero strain at the layer of longitudinal steel reinforcement nearest to the tensile face.
 Point C: strain distribution corresponding to the balanced failure
with a maximum compressive strain eccu (or ecu) and a yielding
tensile strain esy at the layer of longitudinal steel reinforcement
nearest to the tensile face.
 Point D: strain distribution corresponding to the limiting tension-controlled failure having a maximum compressive strain
eccu (or ecu) and a tensile strain of 0.005 as per ACI318-05 [22]
at the layer of longitudinal steel reinforcement nearest to the
tensile face.
 Point E: point corresponding to the pure bending moment and
zero axial force.
In the interaction diagram, point A represents the pure compression case (zero bending moment). For points B, C, and D, the
position of the neutral axis c is directly computed by similar triangles in the strain distribution corresponding to each case (Fig. 1).
Point E, as representing the pure bending moment case (zero axial
force), the neutral axis position is obtained by following conventional RC beam theory. Assuming no contribution of the FRP-connement for control and strengthened specimens, the procedure
for computing Mmax does not vary [5]. In the case of the experimental set by Chaallal and Shahawy [4] where specimens featured FRP
jackets with bidirectional bers, point E is computed accounting
for the FRP in the longitudinal direction and its contribution to
the exural capacity according to ACI440.2R-02 [19].

Table 1
RC columns of circular cross-section
Specimen
code

Fiber
type

D
(mm)

H
(m)

qg

qt

qf

(%)

(%)

fc0
(MPa)

fy
(MPa)

Ef
(MPa)

ffu
(MPa)

efu

(%)

(%)

tf
(mm)

Pa or Pmax
(kN)

Mmax
(kN m)

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

(13)

(14)

(15)

(16)

Saadatmanesh et al. [2]

SA1
SA2
SA3
SA4

None
GFRP
None
GFRP

305
305
305
305

1.62
1.62
1.62
1.62

2.47
2.47
2.47
2.47

0.19
0.19
0.19
0.19

0.00
6.30
0.00
6.30

36.5
38.3
36.6
36.5

358
358
358
358

NA
18,600
NA
18,600

NA
532
NA
532

NA
2.86
NA
2.86

NA
0.80
NA
0.80

445
445
445
445

83
116
102
124

Sheikh and Yau [13]

SY1-a
SY2-a
SY3-a
SY4-a
SY1-b
SY2-b
SY3-b

None
GFRP
GFRP
CFRP
None
CFRP
GFRP

356
356
356
356
356
356
356

1.47
1.47
1.47
1.47
1.47
1.47
1.47

3.01
3.01
3.01
3.01
3.01
3.01
3.01

0.30
0.30
0.30
0.30
0.30
0.30
0.30

0.00
1.40
1.40
1.12
0.00
0.56
1.40

39.2
40.4
40.4
40.4
39.2
44.8
40.8

500
500
500
500
500
500
500

NA
20,800
20,800
152,000
NA
152,000
20,800

NA
416
416
1900
NA
1900
416

NA
2.00
2.00
1.25
NA
1.25
2.00

NA
1.25
1.25
0.50
NA
0.50
1.25

2547
2600
2600
2600
1273
1398
1309

220
180
300
290
225
280
280

Elnabelsy and
Saatcioglu [16]

ES1-C
ES2-C

CFRP
CFRP

508
508

1.73
1.73

1.78
1.78

0.33
0.33

1.42
2.83

38.0
38.0

400
400

60,000
60,000

700
700

1.17
1.17

0.90
0.90

944
944

710
500

Reference

Note: 1 mm = 0.04 in.; 1 MPa = 0.145 ksi; 1 kN = 0.225 kip; 1 kN m = 0.738 kip ft; NA = Not applicable.

Table 2
RC columns of non-circular cross-section
Reference

Specimen
code

Fiber type

b (mm)

h (mm)

h/b

H (m)

r (mm)

qg (%)

qt (%)

qf (%)

fc0 (MPa)

fy (MPa)

Ef (MPa)

ffu (MPa)

efu (%)

tf (mm)

Pa or Pmax (kN)

Mmax
(kN m)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

(13)

(14)

(15)

(16)

(17)

(18)

(19)

Nosho [10]

NO1
NO2
NO3
NO4

None
CFRP
CFRP
CFRP

280
280
280
280

280
280
280
280

1.00
1.00
1.00
1.00

2.36
2.36
2.36
2.36

13
13
13
13

1.02
1.02
1.02
1.02

0.25
0.25
0.25
0.25

0.00
0.94
0.16
0.08

40.6
41.6
41.6
42.7

407
407
407
407

NA
261,484
267,358
267,358

NA
4165
4206
4206

NA
1.60
1.57
1.57

NA
0.17
0.11
0.11

1062
1087
1088
1116

140
147
137
140

Walkup [12]

WA1
WA2
WA3
WA4

None
CFRP
CFRP
CFRP

458
458
458
458

458
458
458
458

1.00
1.00
1.00
1.00

3.05
3.05
3.05
3.05

51
51
51
51

1.48
1.48
1.48
1.48

0.21
0.21
0.21
0.21

0.00
0.86
0.58
0.29

24.6
22.7
24.7
24.1

460
460
460
460

NA
230,303
230,303
230,303

NA
3515
3515
3515

NA
1.50
1.50
1.50

NA
0.17
0.17
0.17

1299
1399
1352
1334

559
581
537
556

Chaallal and Shahawy [4]a

CS1-a
CS2-a
CS1-b
CS2-b
CS1-c
CS2-c
CS1-d
CS2-d
CS1-e
CS2-e
CS1-f
CS2-f

None
CFRP
None
CFRP
None
CFRP
None
CFRP
None
CFRP
None
CFRP

200
200
200
200
200
200
200
200
200
200
200
200

350
350
350
350
350
350
350
350
350
350
350
350

1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75

2.10
2.10
2.10
2.10
2.10
2.10
2.10
2.10
2.10
2.10
2.10
2.10

25
25
25
25
25
25
25
25
25
25
25
25

1.62
1.62
1.62
1.62
1.62
1.62
1.62
1.62
1.62
1.62
1.62
1.62

1.57
1.57
1.57
1.57
1.57
1.57
1.57
1.57
1.57
1.57
1.57
1.57

0.00
1.57
0.00
1.57
0.00
1.57
0.00
1.57
0.00
1.57
0.00
1.57

25.5
25.5
25.5
25.5
25.5
25.5
25.5
25.5
25.5
25.5
25.5
25.5

414
414
414
414
414
414
414
414
414
414
414
414

NA
45,000
NA
45,000
NA
45,000
NA
45,000
NA
45,000
NA
45,000

NA
540
NA
540
NA
540
NA
540
NA
540
NA
540

NA
1.20
NA
1.20
NA
1.20
NA
1.20
NA
1.20
NA
1.20

NA
0.50
NA
0.50
NA
0.50
NA
0.50
NA
0.50
NA
0.50

1878
2412
1228
1335
712
828
310
441
240
356
0
0

1
1
105
145
139
159
114
158
119
179
94
145

Iacobucci et al. [14]

IA1-a
IA2-a
IA3-a
IA4-a
IA5-a
IA6-a

None
CFRP
CFRP
CFRP
CFRP
CFRP

305
305
305
305
305
305

305
305
305
305
305
305

1.00
1.00
1.00
1.00
1.00
1.00

1.47
1.47
1.47
1.47
1.47
1.47

16
16
16
16
16
16

2.58
2.58
2.58
2.58
2.58
2.58

0.61
0.61
0.61
0.61
0.61
0.61

0.00
1.31
2.62
1.31
3.93
2.62

31.4
36.5
36.9
36.9
37.0
37.0

465
465
465
465
465
465

NA
76,350
76,350
76,350
76,350
76,350

NA
962
962
962
962
962

NA
1.26
1.26
1.26
1.26
1.26

NA
1.00
1.00
1.00
1.00
1.00

1168
1290
2231
2231
2237
1308

180
229
233
218
260
246

Bousias et al. [15]

BO1-a
BO2-a
BO3-a
BO4-a
BO1-b
BO2-b
BO3-b
BO4-b

None
CFRP
CFRP
GFRP
None
CFRP
CFRP
GFRP

250
250
250
250
500
500
500
500

500
500
500
500
250
250
250
250

2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00

1.60
1.60
1.60
1.60
1.60
1.60
1.60
1.60

13
13
13
13
13
13
13
13

0.83
0.83
0.83
0.83
0.83
0.83
0.83
0.83

0.51
0.51
0.51
0.51
0.51
0.51
0.51
0.51

0.00
0.31
0.78
1.02
0.00
0.31
0.78
1.02

18.3
18.1
17.9
18.7
17.9
18.1
17.9
18.7

560
560
560
560
560
560
560
560

NA
230,000
230,000
70,000
NA
230,000
230,000
70,000

NA
3450
3450
2170
NA
3450
3450
2170

NA
1.50
1.50
3.10
NA
1.50
1.50
3.10

NA
0.13
0.13
0.17
NA
0.13
0.13
0.17

862
830
865
858
843
830
865
858

304
358
379
326
110
112
122
120

Elnabelsy and Saatcioglu [16]

ES1-S
ES2-S

CFRP
CFRP

500
500

500
500

1.00
1.00

1.73
1.73

13
13

1.44
1.44

0.33
0.33

2.16
3.60

33.0
38.0

400
400

60,000
60,000

700
700

1.17
1.17

0.90
0.90

1002
1128

640
640

Harajli and Rteil [17]

HR1-a
HR2-a
HR3-a
HR1-b
HR2-b
HR3-b

None
CFRP
CFRP
None
CFRP
CFRP

150
150
150
150
150
150

300
300
300
300
300
300

2.00
2.00
2.00
2.00
2.00
2.00

1.12
1.12
1.12
1.12
1.12
1.12

13
13
13
13
13
13

1.50
1.50
1.50
3.00
3.00
3.00

0.80
0.80
0.80
0.80
0.80
0.80

0.00
0.26
0.39
0.00
0.26
0.39

20.3
21.1
21.7
20.3
21.1
21.7

534
534
534
565
565
565

NA
230,000
230,000
NA
230,000
230,000

NA
3500
3500
NA
3500
3500

NA
1.50
1.50
NA
1.50
1.50

NA
0.13
0.13
NA
0.13
0.13

196
196
196
231
231
231

59
65
70
88
94
94

Memon and Sheikh [18]

MS1
MS2
MS3
MS4
MS5
MS6

None
GFRP
GFRP
GFRP
GFRP
GFRP

305
305
305
305
305
305

305
305
305
305
305
305

1.00
1.00
1.00
1.00
1.00
1.00

1.48
1.48
1.48
1.48
1.48
1.48

16
16
16
16
16
16

2.58
2.58
2.58
2.58
2.58
2.58

0.61
0.61
0.61
0.61
0.61
0.61

0.00
3.28
6.56
3.28
1.64
4.92

42.4
42.5
42.7
43.3
43.7
44.2

465
465
465
465
465
465

NA
19,754
19,754
19,754
19,754
19,754

NA
450
450
450
450
450

NA
2.28
2.28
2.28
2.28
2.28

NA
1.25
1.25
1.25
1.25
1.25

2454
1449
2467
2493
1479
2532

195
250
252
232
235
286
1511

Note: 1 mm = 0.04 in.; 1 MPa = 0.145 ksi; 1 kN = 0.225 kip; 1 kN m = 0.738 kip ft.
a
Specimens axially loaded at various eccentricities.

S. Rocca et al. / Construction and Building Materials 23 (2009) 15081520

(1)

1512

S. Rocca et al. / Construction and Building Materials 23 (2009) 15081520

Fig. 1. Simplied interaction diagram (circular and non-circular cross-sections).

The nominal axial load Pn corresponding to point A can be found

using Eq. (1) (Mn(A) equals zero).

PnA 0:85fcc0 Ag  Ast fy Ast 

The nominal axial load Pn and the nominal bending moment Mn at

points B, C, and D are found by integration of the stresses over the
cross-section, and they are given by Eq. (2) for the case of circular
cross-sections and by Eq. (3) for the case of non-circular crosssections.
For design purposes, to replace the unconned concrete stress
distribution, ACI318-05 [22] allows the use of an equivalent rectangular stress block distribution with an average stress of a1 fc0
(a1 equals 0.85) and a depth of b1c (b1 equals 0.65 for
fc0 P 55 MPa [8 ksi], 0.85 for concrete with fc0 6 28 MPa [4 ksi],
and 0.05 less for each 6.9 MPa [1 ksi] of fc0 in excess of 28 MPa)
for the calculation of the resultant compressive force in the
cross-section. However, these values of a1 and b1 are no longer valid for FRP-conned concrete stressstrain curve since they were
based on experimental observations of unconned concrete columns and prisms [20,22].

PnB;C;D

0 s1
 2 

2
X
D
D
0
@2
Afc ydy
 y
Asi fsi
d
2
2
2a

M nB;C;D

0 s1
 2 

2 

D
D
0
@2
A D  c y fc ydy

 y
d
2
2
2
0
X

Asi fsi dsi

Z

2b
PnB;C;D

Asi fsi

3a



X
h
 c y fc ydy
b
Asi fsi dsi
2

3b

bfc ydy

by Lam and Teng [23,24]. This model was selected based on the
evaluation of FRP-conned models with a database composed of
solely RC columns of a minimum side dimension of 300 mm
(12 in.) and subjected to pure axial compressive loading [25]. The
model by Lam and Teng as shown in Fig. 2 proved to be the most
suitable for predicting the maximum conned compressive
strength and strain for RC columns of circular and non-circular
cross-sections.
The stressstrain curve of this model is composed of a parabolic
rst portion followed by a linear second portion (Fig. 2). The curve
and the line meet smoothly at a transition strain e0t . The linear
second portion ends at a point where both the maximum compressive
strength fcc0 and the ultimate axial strain of conned concrete eccu are
reached. Based on experimental observations of small-plain concrete specimens with sufcient amounts of FRP connement, Lam
and Teng suggested a minimum ratio of FRP connement pressure
to unconned concrete compressive strength fl =fc0 of approximately
0.08 to ensure an ascending second branch in the stressstrain
curve. This limitation was later conrmed for circular cross-sections
by Spoelstra and Monti [26] using their analytical model.
In the case of unconned concrete, the model by Lam and Teng
reduces to a stressstrain curve composed of an initial parabola
followed by a horizontal straight line, which has been adopted
by design codes such as BS 8110 and Eurocode 2 [23]. This model
is used in this study to compute the stressstrain curve of the FRPconned concrete. It is given by the following expressions:

(
fc

e0t

2
ec 2
Ec ec  Ec E
4f 0

0 6 ec 6 e0t

fc0 E2 ec

e0t 6 ec 6 eccu

4a

2fc0
Ec  E2

4b

fcc0  fc0
eccu

4c

E2

where, fc and ec are the axial stress and the axial strain of conned
concrete, respectively. Ec is the elastic modulus of unconned concrete, e0t is the transition strain, E2 is the slope of the second linear
portion, and eccu is the ultimate axial strain of conned concrete.
The maximum FRP-conned concrete compressive strength is
given by Eq. (5):

fcc0 fc0 3:3ja fl

The parameter ja is a geometry efciency factor and is discussed later in the text. The FRP conning pressure (fl) in the case
of circular cross-sections is determined by equilibrium and strain
compatibility. However, in the case of non-circular cross-sections,
the diameter D is replaced by an equivalent diameter, which consists of the diagonal of the rectangular cross-section (Eq. (6)).

M nB;C;D

Z
0

In the expressions above, c is the distance from the neutral axis

position to the extreme compression ber in the cross-section. Asi
and fsi are the cross-sectional area and the normal stress, respectively, of the ith layer of longitudinal steel reinforcement, respectively. The parameter dsi is the distance from the position of the
ith layer of longitudinal steel reinforcement to the geometric
centroid of the cross-section.
In Eqs. (2) and (3), y is the variable of integration within the
compression zone. The concrete stress fc corresponds to the model

Fig. 2. Stressstrain model for FRP-conned concrete by Lam and Teng [23].

S. Rocca et al. / Construction and Building Materials 23 (2009) 15081520

fl

8
< 2ntfDEf efe

Circular cross-section

2ntf Ef efe

Non-circular cross-section
: p
2
2

b h

The effective strain efe in Eq. (6) is computed as the product of

an efciency factor je and the ultimate tensile strain efu. The FRP
strain efciency factor je to account for the difference between
the actual rupture strain observed in FRP-conned concrete specimens and the FRP material rupture strain determined from tensile
coupon testing is considered in the determination of the effective
strain in FRP reinforcement attained at failure efe (i.e., efe = jeefu).
Based on experimental calibration with CFRP-wrapped, smallscale, plain concrete specimens, an average value of 0.586 was
computed for je by Lam and Teng [23]. Similarly, Carey and Harries
[27] computed a value of 0.58, while experimental tests on medium- and large-scale columns resulted in values of 0.57 and 0.61,
respectively [28]. In this study, for practical design applications,
the parameter je is taken equal to 0.55.
Based on the work by Priestley et al. [29] and as recommended
by ACI Committee 440 [19], the effective strain in the FRP at failure
efe in members subjected to combined axial compression and
bending moment is limited to the minimum value between 0.004
and jeefu in order to ensure the shear integrity of the conned
concrete.
The ultimate axial strain of the FRP-conned concrete compressive stressstrain curve is given as follows:

eccu

e0c

 0:45 !
fl efe
6 0:01
1:5 12jb 0 0
f c ec

where the axial strain at the compressive strength of unconned

concrete e0c is taken equal to 0.002. The model of Lam and Teng
[23,24] originally features a value of 1.75 instead of 1.5, which for
the case of unconned concrete yields an ultimate axial strain ecu
equal to 0.0035. This change to 1.5 is necessary to limit the axial
strain of unconned concrete ecu to 0.003, a value consistent with
ACI318-05 [22].
Additionally, based on recommendations from the Concrete
Society in the Technical Report 55 [30], a limiting value of ultimate
axial strain of 0.01 in the case of conned concrete is introduced

1513

only in members under pure compression to prevent excessive

cracking and the resulting loss of concrete integrity. When this
limit is applicable, the corresponding maximum value of fcc0 is
recalculated from the stressstrain curve.
The efciency factors ja (Eq. (5)) and jb (Eq. (7)) can be computed according to Eqs. (8) and (9) as they account for the geometry of the cross-section. In the case of circular cross-sections, they
are taken as 1.0, and in the case of non-circular cross-sections, they
depend on two parameters: the effectively conned area ratio Ae/Ac
and the side-aspect ratio h/b. These factors are given by the following expressions:

ja

 2
Ae b
Ac h

jb

 0:5
Ae h
Ac b

where
2
2
Ae 1  b=hh  2r h=bb  2r =3Ag  qg

Ac
1  qg

10

Simplied theoretical interaction diagrams composed of a series

of straight lines connecting points A, B, C, D, and E using the equations and procedures described above were built for each set of
experiments presented in Tables 1 and 2. These diagrams along
with the experimental results, are shown in Figs. 35 for the cases
of circular and non-circular RC columns. No environmental (CE)
and strength reduction factors (/, wf) were considered in these
interaction diagrams.
To express the diagrams independently of the different crosssectional area sizes (Ag) and unconned concrete compressive
strength fc0 , the diagrams are shown as non-dimensional. This is
done by dividing the nominal axial load values Pn by Ag fc0 , and
0
dividing the nominal moment values Mn by Ag hf c . In each of
the plots, the legend indicates the percentage quantity corresponding to the FRP volumetric ratio used in each of the specimens.
Every gure shows the analytical curve and experimental data
at the same scale: 0 to 0.5 for the x-axis and 0 to 2.0 for the y-axis.

Fig. 3. Interaction diagrams and experimental results (circular cross-sections).

1514

S. Rocca et al. / Construction and Building Materials 23 (2009) 15081520

Fig. 4. Interaction diagrams and experimental results (non-circular cross-sections).

Fig. 5. Interaction diagrams and experimental results (non-circular cross-sections).

1515

S. Rocca et al. / Construction and Building Materials 23 (2009) 15081520

0

All plots are drawn with x-axis equal to Mn =Ag hf c and y-axis
equal to P n =Ag fc0 .
Due to the difculty of conducting experiments with real-size
cross-sections, Figs. 35 report a limited number of data points
that fall in the compression-controlled region: 2 (from a total of
13 units) and 13 (from a total of 48 units) for the cases of circular
and non-circular specimens, respectively. From this total, for only
one data point among the circular specimens [13], its corresponding theoretical PM diagram appears to be not conservative. However, the original source of publication [13] notes that such
specimen experienced a premature failure as a result of the stress
concentrations on the jacket induced by one reinforcing bar used to
afx measurement instrumentation. For all other cases, the interaction diagrams built on the Lam and Teng [23,24] model for the
FRP-conned concrete stressstrain curve appear to be conservative with respect to the experimental data points.
5. Proposed design methodology
Strength enhancement can only be considered when the applied ultimate axial force and bending moment (Pu and Mu) are
such that a point having these coordinates falls above the line connecting the origin and the balanced point in the interaction diagram for the unconned member (compression-controlled
region) (Fig. 6). This limitation stems from the consideration that
the enhancement is only of signicance in members where compression failure is the controlling mode [3].
For simplicity, the portion of the unconned and conned interaction diagrams corresponding to compression-controlled failure
can be reduced to two bilinear curves passing through the previously dened points A, B, and C (Fig. 6). Eq. (1) can be used to
compute the Pn corresponding to point A of a member with any
cross-sectional shape (Mn(A) equals zero). The coordinates of points
B and C can be computed using Eqs. (2) and (3) for the case of
members of circular and non-circular cross-sections, respectively.
To provide a closed form solution, this paper focuses only on the
case of members with rectangular cross-sections. Substituting the
stressstrain relationship for FRP-conned concrete (Eq. (4)) in Eq.
(3) and expressing the axial compressive strain ec at any point in
the compression region in terms of the integrating variable y (similar triangles) (Fig. 7), Eqs. (11) and (12) are obtained for the axial
loads and moments at points B and C, respectively:

PnB;C

#
e
 E  E 2 e
2
ccu
c
2
ccu
bdy
y 
y
c
4fc0
c
0
Z ch
e
i
X
ccu

y bdy
Asi fsi
fc0 E2
c
yt

yt

"

Ec

Fig. 7. Strain distributions of points B and C for simplied interaction diagram.

MnB;C

#

e
 E  E 2 e
2  h
ccu
c
2
ccu
b

c

y
dy
y 
y
2
c
4fc0
c
0

Z ch
e
i h
X
ccu

fc0 E2
 c y dy
Asi fsi dsi
y b
2
c
yt
Z

yt

"

Ec

12
In the expressions above, c represents the distance from the extreme compression ber to the neutral axis and is given by Eq. (13).
The parameter yt represents the vertical coordinate within the
compression region measured from the neutral axis position, and
it corresponds to the transition strain e0t (Eq. (14)) (see Fig. 7).

8
<d
:d

yt c

for point B
eccu
esy eccu

13

for point C

e0t
eccu

14

After integrating and reordering terms, Eqs. (11) and (12) can be
reduced to expressions (15) and (16), respectively, where the coefcients A, B, C, D, E, F, G, H, and I, are given by Eq. (17).

h
i X
PnB;C Ayt 3 Byt 2 Cyt D
Asi fsi
11

MnB;C Eyt 4 Fyt 3 Gyt 2 Hyt I

Asi fsi dsi

16

bEc  E2 2 eccu 2
12fc0
c

17a

bEc  E2 eccu 
2
c

17b

C bf c
0

D bcf c

Fig. 6. Representative simplied interaction diagram.

15

17c
bcE2
eccu
2

bEc  E2 2 eccu 2
16fc0
c



h Ec  E2 2 eccu 2 bEc  E2 eccu 

F b c
2
3
12fc0
c
c

17d

17e

17f

1516

S. Rocca et al. / Construction and Building Materials 23 (2009) 15081520




b
h Ec  E2 eccu 
G  fc0 b c 
2
2
2
c

17g



h
0
H bf c c 
2

17h





2
2
bc 0
h
bc E2
bcE2
h
0

eccu
fc  bcf c c 
eccu 
c
2
2
2
3
2

17i

The values of fsi are calculated by similar triangles from the

strain distribution corresponding to points B and C, and depending
on the neutral axis position c, the sign of fsi will be positive for
compression and negative for tension.
According to ACI design philosophy, the nominal axial load
capacity under pure compression loading condition (point A in
the interaction diagram) of an RC column featuring ties as steel
transverse reinforcement is reduced to 80% to account for a
possible accidental eccentricity [21]. In the example application
presented in Appendix A and Fig. 6, this point is denoted as
A0 .

/PnA0 /0:80:85fcc0 Ag  Ast fy Ast 

18

For a tied RC column, the strength reduction factors / as recommended by ACI318-05 [22] must also be considered. Since the portion of the interaction diagram corresponds to compressioncontrolled members, the value of / is 0.65 (Fig. 6).
The proposed design method consists of closed form equations
for the determination of the compression-controlled region of a
PM diagram for the case of a rectangular FRP-wrapped RC column.
This portion of the interaction diagram is conservatively represented by two bilinear curves joining the previously dened points
A0 , B, and C. Based on ACI design philosophy, the vertical coordinate
(axial force) of point A0 can be found with Eq. (18) (bending moment is zero). The coordinates for points B and C can be obtained
with Eqs. (15) and (16). The contribution of concrete within the
compression region of the cross-section is represented by polynomials of third and fourth order for the case of the nominal axial
capacity (Eq. (15)) and the nominal exural capacity (Eq. (16)),
respectively. These algebraic expressions result from mathematical
substitution and re-arrangement of terms resulting from the integrals given in Eqs. (11) and (12).
A ow-chart illustrating the application of the proposed methodology for design purposes is shown in Fig. 8, and a practical
example application is presented in Appendix A.

Fig. 8. Flow chart for application of methodology.

S. Rocca et al. / Construction and Building Materials 23 (2009) 15081520

6. Summary and conclusions

This paper responds to the need for a simple and yet rational
design approach to FRP-wrapped RC columns subjected to combined axial force and bending. The proposed design approach that
follows ACI principles requires further validation with experimental test results of RC columns having geometry and properties similar to the ones found in practice.
The design method presented in the paper is based on the construction of a simplied interaction diagram for FRP-wrapped RC
columns stemming from a process that included the four steps
listed below:
(a) Collection of experimental data on FRP-wrapped circular
and non-circular RC columns with minimum diameter or
side dimension of about 300 mm (12 in.), maximum sideaspect ratio of 2, and featuring FRP ber orientation perpendicular to the longitudinal axis of the column.
(b) Construction of a theoretical P-M diagram based on principles of equilibrium and strain compatibility, equivalent to
that of conventional RC columns but considering the
stressstrain model for FRP-conned concrete developed
by Lam and Teng.
(c) Partial validation of the theoretical PM diagrams with the
experimental results.
(d) Simplication of the PM diagram to a series of straight lines
joining characteristic points only in the compression-controlled region of the diagram. This limitation recognizes that
capacity enhancement is signicant only if the combined
external actions are such that the FRP wrapping inuences
the properties of the concrete.
The paper illustrates the proposed design method for the case of
a RC column of rectangular cross-section as an application example
(see Appendix A).
Acknowledgements
The authors would like to acknowledge the funding and support
received from the National Science Foundation Industry/University
Cooperative Research Center on Repair of Buildings and Bridges
with Composites (RB2C) (supplement Grant No. 0453808).

umn are shown in Fig. A.1. A method of strengthening the column

is sought.
Since the column is located in an interior environment, and
CFRP material will be used, according to Table 8.1 in the
ACI440.2R-02 guide [19], an environmental-reduction factor CE
equal to 0.95 will be used. Additionally, the application of a reduction factor wf equal to 0.95 as recommended by ACI440.2R-02 [19]
is taken into consideration. However, in this example application,
instead of applying the latter factor on the resulting conned concrete compressive strength fcc0 , it is applied to the sole contribution of the FRP (i.e., the connement pressure fl) as is the
approach that the revised version of the guide is considering.
The properties of the FRP system, as reported by the manufacturer, are shown in Table A.1.
Solution:
Step 1: Determine the simplied curve for the unstrengthened
column (n = 0 plies).Points A0 , B, and C can be obtained
by well-known procedures:

/P nA0 9281 kN; /M nA0 0 kN m

/P nB 8266 kN; /M nB 874 kN m
/P nC 4127 kN; /M nC 1198 kN m
Step 2: Compute the design FRP material properties.

ffu C E ffu 0:95  3792 MPa 3602 MPa

efu C E efu 0:95  0:0167 mm=mm 0:0159 mm=mm
Step 3: Determine the simplied curve for the strengthened column.A wrapping system composed of six plies will be
the starting point to construct the curve ABC and then
be compared with the position of the required Pu and Mu.
Points A0 , B, and C can be computed using Eqs. (18), (15)
and (16), where the coefcients A, B, C, D, E, F, G, H, and
I are given by Eq. (17).
Point A0 :
Design axial capacity:

uPnA0 u0:80:85fcc0 Ag  Ast fy Ast 

uPnA0 0:65  0:80:85  56:96 MPa  37; 612  9; 832 mm2
414 MPa  9832 mm2 

uPnA0 11; 224 kN

Appendix A. Example application

Given:
A 610  610 mm (24  24 in.) RC column is subjected to an ultimate axial compressive load Pu = 8451 kN (1900 kip) and to an ultimate bending moment Mu = 515 kN m (380 kip ft) (e = 0.1h). It is
required to increase the demand by 30% at constant eccentricity
(Pu = 10,987 kN, Mu = 670 kN m). Concrete and steel reinforcement
material properties as well as details of the cross-section of the col-

1517

where

flfcc0 fc0 3:3ja fl 44:8 MPa 3:3  0:425  8:67 MPa

 2
Ae b
Ae
56:96 MPaja
0:42512 0:425
Ac h
Ac

1  b=hh  2r2 h=bb  2r2 =3Ag  qg

1  qg

Fig. A.1. Column cross-section details and materials properties.

1518

S. Rocca et al. / Construction and Building Materials 23 (2009) 15081520

Ae 1 

Ac

1610225 mm2 1610225 mm2 

 2:65  102
337;612 mm2
2

yt c

1  2:65  10

c 559 mm

0:425
wf 2nt f Ef efe

fl p
2
2
b h
fl

e0t

0:95  2  6  0:33 mm  227; 527 MPa  0:55  0:0159 mm=mm

q
2  610 mm2

E2

fcc0  fc0
50:38  44:8 MPa
1315 MPa

eccu
4:23  103 mm=mm

fcc0 44:8 MPa 3:30:4253:97 MPa 50:38 MPa

 0:45 !
fl efe
eccu e0c 1:5 12jb 0 0
f c ec

Point B:
Design axial capacity:

uP nB u Ayt 3 Byt 2 Cyt D

Asi fsi



0:45 !
3:97 MPa 0:004 mm=mm
eccu 0:002 mm=mm 1:5 12  0:425
44:8 MPa 0:002 mm=mm


kN
0:65 6:003  105
389 mm3
mm3
70:14 MPa389 mm2  27:32 kN=mm389 mm

16; 215 kN 3277 mm2 414 MPa
2

where

bEc  E2 2 eccu 2
12fc0
c

610mm31; 685  1315MPa2 4:23  103 mm=mm

559 mm
12  44:8 MPa
5

6:003  10

ja 0:425
jb

uP nB 9829 kN

eccu 4:23  103 mm=mm

 0:5
Ae h
0:42510:5 0:425
Ac b
wf 2ntf Ef efe

fl p
2
2
b h
0:95  2  6  0:33 mm  227; 527 MPa  0:004 mm=mm
q
fl
2  610 mm2

1639 mm 414 MPa 1639 mm 257 MPa

2fc0
2  44:8 MPa

2:95  103 mm=mm

Ec  E2 31; 685  1315 MPa

fcc0 fc0 3:3ja fl

8:67 MPa

uP nB

e0t
2:95  103 mm=mm
389 mm
559 mm 
eccu
4:23  103 mm=mm

!2

kN
mm3

3:97 MPa
efe min0:004; je efu 0:55  0:159 mm=mm
0:004 mm=mm
Checking the minimum connement ratio:

fl =fc0 3:97=44:8 0:09 P 0:08 Ok!

bEc  E2 eccu 
2
c

!
610 mm31; 685  1315 MPa 4:23  103 mm=mm
B
2
559 mm
70:14 MPa
0

The strains in each layer of steel are determined by similar triangles in the strain distribution (Fig. 7). The corresponding stresses
are then given as follows:

fs1 es1 Es 0:0038 mm=mm  200; 000 MPa 414 MPa

fs2 es2 Es 0:0026 mm=mm  200; 000 MPa 414 MPa

C bf c

fs3 es3 Es 0:0013 mm=mm  200; 000 MPa 257 MPa

C 610 mm  44:8 MPa 27:32 kN=mm

fs4 es4 Es 0 mm=mm  200; 000 MPa 0 MPa

bcE2
eccu
2
D 610 mm  559 mm  44:8 MPa
0

D bcf c

Design bending moment:


610mm  559 mm  1315 MPa 
4:23  103 mm=mm

2
D 16; 215 kN
For the calculation of the coefcients it is necessary to compute
key parameters from the stressstrain model:

uMnB u Eyt 4 Fyt 3 Gyt 2 Hyt I

Thickness per ply, tf


Ultimate tensile strength, ffu
Rupture strain, efu
Modulus of elasticity, Ef

0.33 mm
3792 MPa
0.0167 mm/mm
227,527 MPa

0.013 in.
550,000 psi
0.0167 in./in.
33,000,000 psi

Asi fsi dsi

h

uMnB 0:65 4:5  105 kN=mm3 389 mm4

62 MPa389 mm3  31:48 kN=mm389 mm2

6939 kN389 mm 500; 175 kN mm
3277 mm2 414 MPa254 mm

Table A.1
Manufacturers reported FRP material properties example application

1639 mm2 414 MPa85 mm

i
1639 mm2 257 MPa85 mm

uMnB 924 kN m

1519

S. Rocca et al. / Construction and Building Materials 23 (2009) 15081520

where

Table A.2
Summary of results example application

bEc  E2 eccu 2
E
16fc0
c
2

Point

!2
610 mm31; 685  1315 MPa2 4:23  103 mm=mm
E
559 mm
16  44:8 MPa
4:5  105 kN=mm3


h Ec  E2 2 eccu 2 bEc  E2 eccu 
Fb c

2
3
12fc0
c
c

A
A0
B
C

n = 0 plies (unstrengthened member)

n = 6 plies

/Pn (kN)

/Mn (kN m)

/Pn (kN)

/Mn (kN m)

11,601
9281
8266
4127

0
0
874
1198

14,030
11,224
9829
5870

0
0
924
1345

Note: 1 kN = 0.225 kip; 1 kN m = 0.738 kip ft.

31; 685  1315 MPa2

12  44:8 MPa
!2
4:23  103 mm=mm
610 mm31; 685  1315 MPa


559 mm
3
!
4:23  103 mm=mm
62 MPa

559 mm




b
h Ec  E2 eccu 
G  fc0 b c 
2
2
2
c
F 610 mm559  305 mm

"
G  44:8 MPa  305 mm 610 mm559  305 mm


!#


31; 685  1315 MPa 4:23  103 mm=mm
2
559 mm

G 31:48 kN=mm


h
0
H bf c c 
2
H 44:8 MPa  610 mm559  305 mm 6939 kN




2
2
bc 0
h
bc E2
bcE2
h
0
I

eccu
fc  bcf c c 
eccu 
c
2
2
2
3
2
2

559 mm
 44:8 MPa559
2
 305 mm559 mm610 mm 1315 MPa  610 mm

559 mm2 

4:23  103 mm=mm  1315 MPa  610 mm
3
559 mm
559  305 mm4:23  103 mm=mm

2
500; 175 kN mm

I 44:8 MPa  610 mm 

The distances from each layer of steel reinforcement to the geometric centroid of the cross-section are shown below:

ds1 254 mm
ds2 ds3 85 mm
Point C:
Following the same procedure as for point B, the design axial
capacity is as follows:

uPnC 5870 kN
where

A 1:33  104 kN=mm3

B 104:4 MPa

Fig. A.2. Simplied interaction diagram example application.

yt 262 mm
c 375 mm
The strains in each layer of steel are determined by similar triangles in the strain distribution (Fig. 7). The corresponding stresses
are then given:

fs1 es1 Es 0:0037 mm=mm  200; 000 MPa 414 MPa

fs2 es2 Es 0:0018 mm=mm  200; 000 MPa 350:2 MPa
fs3 es3 Es 1:59  104 mm=mm  200; 000 MPa 31:8 MPa
fs4 es4 Es 0:0021 mm=mm  200; 000 MPa 414 MPa
The design bending moment: uMn(C) = 1345 kN  m where

E 9:98  105 kN=mm3

F 79 MPa
G 21:03 kN=mm
H 1928 kN
I 1; 315; 459 kN mm
Note:
The designer should bear in mind that for the case of pure compression, the effective strain in the FRP efe is limited by jeefu, and in
the case of combined axial and bending by efe = minimum (0.004,
jeefu).
Step 4: Comparison of simplied partial interaction diagram with
required Pu and Mu.
Table A.2 summarizes the axial and bending design capacities for the unstrengthened and strengthened members
at points A, A0 , B, and C. These points are plotted in
Fig. A.2.

C 27:32 kN=mm
D 10; 892 kN
For the calculation of the coefcients it is necessary to compute
key parameters from the stressstrain model:

References
[1] Fardis MN, Khalili H. Concrete encased in berglass-reinforced plastic. ACI J
1981;78(6):4406.

1520

S. Rocca et al. / Construction and Building Materials 23 (2009) 15081520

[2] Saadatmanesh H, Ehsani MR, Li MW. Strength and ductility of concrete

columns externally reinforced with ber composite straps. ACI Struct J
1994;91(4):43447.
[3] Nanni A, Norris M. FRP jacketed concrete under exure and combined exurecompression. Constr Build Mater 1995;9(5):27381.
[4] Chaallal O, Shahawy M. Performance of ber-reinforced polymer-wrapped
reinforced concrete column under combined axial-exural loading. ACI Struct
J 2000;97(4):65968.
[5] Teng JG, Chen JF, Smith ST, Lam L. FRP strengthened RC structures. West
Sussex, UK: John Wiley and Sons; 2002.
[6] Di Nardo A, Faella C, Realfonzo R. A design procedure of FRP conning systems
for upgrade R/C columns. In: Mirmiran A, Nanni A, editors, Third international
conference on FRP composites in civil engineering (CICE), Miami, USA, CDROM; 2006.
[7] Mirmiran A, Shahawy M. A new concrete-lled hollow FRP composite column.
Compos Part B 1996;27B:2638.
[8] Mirmiran A, Shahawy M, Samaan M. Strength and ductility of hybrid FRPconcrete beam-columns. ASCE J Struct Eng 1999;125(10):108593.
[9] Fam A, Flisak B, Rizkalla S. Experimental and analytical modeling of concretelled ber-reinforced polymer tubes subjected to combined bending and axial
loads. ACI Struct J 2003;100(4):499509.
[10] Nosho KJ. Retrot of
rectangular reinforced concrete columns using
carbon ber. Masters Thesis. Seattle, WA, USA: University of Washington;
1996.
[11] Saadatmanesh M, Ehsani M, Jin L. Seismic strengthening of circular bridge Pier
models with ber composites. ACI Struct J 1996;93(6):63947.
[12] Walkup S. Rehabilitation of non-ductile reinforced concrete building columns
using ber reinforced polymer jackets. Masters Thesis. Bethlehem, PA, USA:
Lehigh University; 1998.
[13] Sheikh S, Yau G. Seismic behavior ofconcrete columns conned with steel and
ber-reinforced polymers. ACI Struct J 2002;99(1):7280.
[14] Iacobucci R, Sheikh S, Bayrak O. Retrot of square concrete columns with
carbon ber-reinforced polymer for seismic resistance. ACI Struct J
2003;100(6):78594.
[15] Bousias S, Triantallou T, Fardis M, Spathis L, ORegan B. Fiber-reinforced
polymer retrotting of rectangular reinforced concrete columns with or
without corrosion. ACI Struct J 2004;101(4):51220.

[16] Elnabelsy G, Saatcioglu M. Seismic retrot of circular and square bridge

columns with CFRP jackets, advanced composite materials in bridges and
structures (CD-ROM 8p). Calgary, Alberta, Canada; 2004.
[17] Harajli M, Rteil A. Effect of connement using ber-reinforced polymer or
ber-reinforced concrete on seismic performance of gravity load-designed
columns. ACI Struct J 2004;101(1):4756.
[18] Memon M, Sheikh S. Seismic resistance of square concrete columns retrotted
with glass ber-reinforced polymer. ACI Struct J 2005;102(5):77483.
[19] American Concrete Institute, ACI 440.2R, Guide for the design and construction
of externally bonded FRP systems for strengthening of concrete structures.
Farmington Hills, MI, USA: American Concrete Institute; 2002.
[20] MacGregor J. Reinforced concrete mechanics and design. 3rd ed. Upper Saddle
River, NJ, USA: Prentice Hall Inc.; 1997.
[21] Bank LC. Composites for
construction: structural design with FRP
materials. Hoboken, N.J. NJ, USA: John Wiley and Sons; 2006.
[22] American Concrete Institute, ACI 318-05. Building code requirements for
structural concrete. Farmington Hills, MI, USA: American Concrete Institute;
2005.
[23] Lam L, Teng J. Design-oriented stressstrain model for FRP-conned concrete.
Constr Build Mater 2003;17(6&7):47189.
[24] Lam L, Teng J. Design-oriented stressstrain model for FRP-conned concrete
in rectangular columns. J Reinf Plast Compos 2003;22(13):114986.
[25] Rocca S. Experimental and analytical evaluation of FRP-conned large size
reinforced concrete columns. Ph.D. Dissertation. Rolla, MO, USA: University of
Missouri-Rolla; 2007.
[26] Spoelstra MR, Monti G. FRP-conned concrete model. ASCE J Compos Constr
1999;3(3):14350.
[27] Carey S, Harries K. The effects of shape, Gap, and Scale on the behavior and
modeling of variably conned concrete. Report No. ST03-05. Columbia, SC,
USA: University of South Carolina; 2003.
[28] Carey SA, Harries KA. Axial behavior and modeling of small-, medium- and
large-scale circular sections conned with CFRP jackets. ACI Struct J
2005;102(4):596604.
[29] Priestley M, Seible F, Calvi G. Seismic design and retrot of bridges. New York,
USA: John Wiley and Sons; 1996.
[30] Concrete Society. Design guidance for strengthening concrete structures using
bre composite material. Technical Report 55. Crowthorne, UK; 2004.