Fourth International Conference on FRP Composites in Civil Engineering (CICE2008)

22-24July 2008, Zurich, Switzerland

Effective bond length of externally bonded FRP sheets

M. Ben Ouezdou
Engineering Laboratory, National School of Engineering of Tunis, Tunisia

S. Bae, & A. Belarbi

Department of Civil, Architectural, and Environmental Engineering, Missouri University of Science and
Technology, Missouri, USA

ABSTRACT: Strengthening and repair of concrete structures using externally bonded fiber reinforced polymer (FRP) composite sheets has been popular around the world during the last two
decades. However, premature failure due to debonding of the FRP is one of the important issues remaining to be resolved. Numerous research studies have dealt with the debonding problem in terms of bond strength, bond stress-slip relationship, effective bond length, and the interfacial fracture energy. Determination of the effective bond length is still not completely
understood. This paper summarizes previous work on effective bond length and proposes a new
relationship for effective bond length with FRP stiffness based on existing experimental data
collected in this study
1 INTRODUCTION
Strengthening and repair/rehabilitation of concrete structures using externally bonded fiber reinforced polymer (FRP) sheets has demonstrated its effectiveness through many laboratory tests
and field applications performed in the last two decades. However, premature failure due to
debonding of the FRP sheets does not take advantage of its full effectiveness and may also reduce the level of safety of the strengthened structures. Hence the debonding problem has become a growing concern for both engineers and researchers. Numerous researches have been
conducted on this subject in terms of bond strength, bond stress-slip relationship, effective bond
length, and the interfacial fracture energy.
In the determination of the effective bond length it is important to evaluate the maximum
load to be carried by the strengthened structure. The effective bond length can be defined as a
length over which the majority of the bond stress is maintained. Currently, many design
codes/guidelines/specifications around the world present equations to evaluate the effective
bond length. Figure 1 presents the effective bond length versus FRP stiffness (Ef tf) calculated
by several code equations (refer to Table 1), using the same material properties (i.e. a concrete
strength of 40 MPa, an elastic modulus of GFRP and CFRP of 72 and 228 GPa respectively, and
thicknesses of 0.353 and 0.17 mm for GFRP and CFRP respectively. As shown in Figure 1, the
effective bond length is quite different depending on which equations are used for calculations.
It was found that some of the code equations typically show a decrease in the effective bond
length with increase in stiffness of the FRP (Eftf) while others show the exactly opposite trend.
The main reason for the difference in results can be attributed to the fact that the equations were
derived using a very limited experimental data base, from which the mechanism involved in the
debonding of FRP can not be completely understood.
In this paper, the existing research on effective bond length and debonding mechanism are
presented along with a simple but practical equation proposed for determining the effective
bond length.

-1-

250
ACI 440.2R-02 (USA)
ISIS Design Manual 4 (Canada)
FIB Bulletin 14 Appendix A1 (Europe)
FIB Bulletin 14 Appendix A2 (Europe)
CS TR 55 (UK)
CNR-DT 200/04 (Italy)

Effective Bond Length (mm)

200

150

GFRP 1 to 5 layers

100

CFRP 1 to 5 layers

50

0
0

50,000

100,000
E f t f (kN/mm)

150,000

200,000

Figure1. Effective bond length calculated using the current codes/specs/guidelines

Table 1.Summary of the effective bond length as specified by various FRP codes.
Code

year

ACI 440.2R-02
(USA)

2002

Expression
23300
Le =
0.58
(n E f t f )

Le =

ISIS & CSA S806-02

(Canada)

2002

FIB B14
Appendix A1
(Europe)

2001

Le =

FIB B14
- Appendix A2 (Europe)

2001

Le = c2

CS TR55
(UK)

(E t )
f

2004

CNR-DT 200/04
(Italy)

2005

Eurocode 8-3
(Europe)

2004

CIDAR [8]
(Australia)

25350

2006

0.58

Ef tf

c2 = 2

c2 f ctm
Ef tf
f ck f ctm

Le = 0.7

Le =

c2 =1.44

Ef tf
f ctm

Ef tf
2 f ctm
Ef tf

Le =

4 max

Le =

Ef tf
f c'

Reference applied
Maeda et al.
6.134 0.58 ln ( E f t f )

Le = e

Maeda et al.
6.134 0.58 ln ( E f t f )

Le = e

Neubauer and Rostsy

Ef tf
Le =
2 f ctm
Neubauer and Rostsy
Ef tf
Le =
2 f ctm
FIB- B14 - Apx A1
Ef tf
c2 = 2
Le =
c2 f ctm
Chen and Teng
Ef tf
Le =
f c'

E f = elastic modulus of FRP, Le = effective bond length, f c' =concrete strength, f ck = characteristic

strength of concrete, f ctm = mean tensile strength of concrete, n = number of layers of FRP, t f = thickness of FRP, max = maximum bond strength of FRP onto concrete surface

-2-

2 THE CONCEPT OF EFFECTIVE BOND LENGTH

Strain in FRP

Local bond stresses

Tension in the concrete is transferred to the FRP mainly through shear stresses in the adhesive
in a short length near the applied load. As the load increases, cracking near the applied load
shifts the active bond zone to a new area further away from the loading point, indicating that
only part of the bond is effective. This part is called the effective bond length. Hence, the effective bond length is the active bonding zone along which most of the interfacial stress is
transmitted into the concrete.
Effective Bond Length
When the bonded length of
FRP along the FRP-concrete
0.9999 Pu
interface exceeds the effec0.999 Pu
tive bond length, no further
increase in failure load can
0.96 Pu
be achieved (Figure 2a).
The concept of effective
0.8 Pu
bond length is also defined
0.5 Pu
through the strain distribution for which the effective
0.2 Pu
bond length is the distance
required for the strain to
Free end
Loaded end
vanish (Figure 2b). The efP
Concrete
fective bond length of the
FRP
FRP takes the entire load to
(a)
a certain level at which localized debonding occurs,
Effective Bond Length
causing the effective bond
0.9999 Pu
length to shift to another active bonding zone.
This
0.999 Pu
shifting continues until the
FRP is completely debonded
0.96 Pu
from the concrete. In other
0.8 Pu
words, when debonding occurs in the vicinity due to
0.5 Pu
fracture of the concrete sur0.2 Pu
face, the active zone is
shifted to a new zone. This
Loaded end
Free end
phenomenon is repeated un(b)
til debonding propagates
Figure 2. Concept of the effective bond length (EBL): (a): in
completely. At any stage of
terms of stress distribution (after Ueda and Dai (2005)); (b) in
loading, the effective bond
terms of strain distribution (after Ueda and Dai (2004))
length is the length at which
the FRP resists the entire
load through its bond stress. However, it is also important to numerically quantify the effective
bond length. Although extensive research has been conducted to investigate the bond behavior
between FRP and concrete, there are no commonly accepted analytical models to predict the effective bond length as previously shown in Figure 1. This issue is discussed in more detail in
this paper and a practical model is proposed based on analysis of the data obtained from literature review.
3 EXPERIMENTAL DATA ON THE EFFECTIVE BOND LENGTH
The problems associated with debonding of FRP were studied through many experimental
works. Several test set-ups were employed to study the bond mechanism of FRP sheets exter-

-3-

nally bonded to concrete, such as single shear tests, double shear tests, and bending tests. The
set-up usually consisted of two concrete elements on which an FRP sheet was adhered (on both
sides) and data was taken from strain-gages glued to one side of the FRP in order to obtain the
graphs shown in Figure 2b.
Data gathered from the literature on experimental studies of effective bond strength has been
summarized in Table 2 in terms of concrete strength, elastic modulus of concrete and FRP, and
thickness of FRP, since it has been known from previous research studies that these parameters
affect the bond of FRP sheets to concrete surfaces.
Table. 2. Summary of data on the experimental results of the effective bond length ( Le ) gathered from literature
Reference
Sato et al.
(1997)
Bizindavyi
and Neale
(1999)

Specimen

Types of

f c' ,

Ec ,

Ef ,

tf ,

Le ,

ID

FRP

(MPa)

(GPa)

(GPa)

(mm)

(mm)

CFRP

37.6

25.5

236

0.115

45.2

GFRP

29.2

1.00

75

GFRP

29.2

2.00

100

CFRP

CFRP

CFRP

C5-ARF

Aramid

C5-SCF

CFRP

42.5

33.5

75.7

0.33

55

75.7

0.66

70

227

0.16

93

124.5

0.193

De Lorenzis
et al.

47.3

32.5**

(2001)

Nakaba et al.
(2001)

Foster and
Khomwan
(2005)

65.9
95.7

261.1

0.167

C5-SCFL

CFRP

C5-SCFH

CFRP

C5-HCF

H-CFRP

425.1

0.165

M5-ARF

Aramid

124.5

0.193

M5-SCF

CFRP

57.6

29

63.5
133.5
120.3
70.3
96.6

47.1

24.5

M5-SCFL

CFRP

M5-SCFH

CFRP

M5-HCF

H- CFRP

C2-SCF

CFRP

23.8

22.0

BS37

CFRP

37

25.5

160

1.4

270

BS53

CFRP

53

29.2

160

1.4

240

2 C2 a

390

0.33

112

2 C1 a

230

0.33

85

390

0.33

115
115

0.167

134.1
121.2

58

2 C2 c

67.0
261.1

36**

99.1

3 C2 a

390

0.495

Boshetto et al.

1 C1 b

230

.165

80

(2006)

2 C2 d

390

0.33

130

CFRP

40

29.9

1 C2 c

390

.165

95

3 C2 b

390

0.495

130

230

0.495

115

230

0.495

106

58

3 C1 a

36**

3 C1 b
CS- 1

130

Iwashita et al.

CS-2

(2007)

CS-3

95

CF-20-1

120

PBO* FRP

36.8

30

Poly-p-phenylene-BenzobiOxazole.
Ec is estimated according to ACI 318-05 as Ec = 4730 fc'

**

-4-

235

0.128

125

4 PROPOSED EQUATIONS TO DETERMINE THE EFFECTIVE BOND LENGTH

Data shown in Table 1 was plotted in Figure 3 in order to find the relationship between the effective bond length and the above mentioned parameters. When normalized by the FRP thickness, the effective bond length is presented as a function of the dimensionless ratio of the elastic
modulus of the FRP to that of the concrete (Figure 3(a)). The best fit curve is a linear correlation, which is expressed as:
Le =

tf Ef

20 Ec

(1)

where, Le = effective bond length (mm), t f = thickness of FRP (mm), E f = elastic modulus of
FRP (MPa), and Ec = elastic modulus of concrete.
1.2

1.2

Experimental Results

Experimental Results
1

Proposed

0.8

Linear
Proposed
(Proposed)

0.8

(Le / tf ) x 10-3

(Le / tf ) x 10

-3

0.6

0.4

0.2

0.6

0.4

0.2

10

12

14

16

10

20

30

40

50

60

70

E f / (f' c )0.5

E f /E c

(a)

(b)

Figure. 3. Relationship between the effective bond strength and the affecting parameters such as FRP
thickness, elastic modulus of concrete and FRP, and concrete strength.

This relationship takes into account the stiffness of the FRP (i.e. thickness and elastic
modulus of FRP) and the elastic modulus of the concrete. However, the most measured property of concrete is the concrete strength, not the elastic modulus. Thus, a relationship between
the effective bond length and the concrete strength was also obtained as shown in Figure 3(b),
which can be expressed as:
Ef

Le = 0.012 t f
f'
c

(2)

where, fc' is the concrete strength (MPa). It can be clearly shown from the proposed relationships that the effective bond length increases with an increase in FRP stiffness (i.e. thickness
and elastic modulus of FRP). Therefore, the equations from some of the current design
codes/guidelines showing the opposite trend should be revised.
5 CONCLUSIONS
Premature failure of concrete structures strengthened with externally bonded FRP due to
debonding is an important problem to be resolved for ensuring this emerging technology. In order to account for debonding problems, determination of the effective bond length must proceed. However, the effective bond lengths calculated by the equations adopted in the current
codes/guidelines/specifications were found to be inconsistent. This is mainly due to the equations being developed based on limited experimental data. Thus, the experimental data available in the literature were collected and analyzed in this study, which showed that the effective

-5-

bond length is proportional to the stiffness of the FRP. Based on the collected data, simple but
practical equations were proposed.
REFERENCES
ACI 440.2R-02. 2002. Guide for the design and construction of externally bonded FRP systems for
strengthening concrete structures. ACI Committee 440, American Concrete Institute, Farmington
Hills, MI, USA.
ACI 318-05. Building code requirements for structural concrete (318-05) band commentary (318R-05).
American Concrete Institute, Farmington Hills, MI, USA; 2005
Bizandavyi, L., and Naele, K. W. 1999. Transfer lengths and bond strengths for composites bonded to
concrete, ASCE J of Comp Constr, l3(4), 153-160.
Boschetto, G., Pellegrino, C., Tinazzi, D., and Modena, C. 2006. Bond behaviour between FRP sheets and
concrete: an experimental study. Proc, 2nd Int Cong, FIB, Naples, Italy, (CD-Rom).
CNR-DT 200/04. 2004. Instructions for the design, execution and control of strengthening measures
through fibre-reinforced composites, Italian Society Research Society, Rome, Italy.
CS-TR-55-UK, The Concrete Society. 2004. Design guidance for strengthening concrete structures using
fibre composites materials, Concrete Society Technical Report No. 55, 2nd ed., Report of a concrete
society committee, Berkshire, UK.
De Lorenzis, L., Miller, B., and Nanni, A. 2001. Bond of fiber-reinforced polymer laminates to concrete,
ACI Mat J, 98(3), 256-264.
FIB Task Group 9.3. 2001. Externally bonded FRP reinforcement for RC structures, Technical Report.
Fib Bulletin 14, CEB-FIP, Lausanne, Switzerland.
Foster, S. J., and Khomwan, N. 2005. Determination of bond stress versus slip for externally bonded FRP
from standardized bond strength tests. Proc, Int Sym on Bond behavior of FRP in Structures, Hong
Kong, 85-90.
ISIS Canada. 2001. Design Manual No. 4 Strengthening Reinforced Concrete Structures with
Externally-Bonded Fibre Reinforced Polymers (FRPs). Intelligent Sensing for Innovative Structures,
Winnipeg, Canada.
Iwashita, K., Wu, Z., Ishikiwa, T., Hamagushi, Y., and Suzuki, T. 2007. Bonding and debonding behavior
of FRP sheets under fatigue loading, Adv Comp Mat, 16(1), 31-44.
Nakaba, L., Kanakubo, T., Furuta, T., and Yoshizawa, H. 2001. Bond behavior between fiber-reinforced
polymer laminates and concrete, ACI Str J, 98(3), 359-367.
Sato, Y., Kimura, K., and Kobatake, Y. 1997. Bond behavior between CFRP sheet and concrete (Part 1),
J of Str Constr Eng of AIJ, Japan, 500, 75-82. (in Japanese).
Ueda, T., and Dai, J. G. 2004. New shear bond model for FRP-concrete interface - from modeling to application. Proc, 2nd Int Conf on FRP Comp in Civil Eng, CICE 2004, Adelaide, Australia, 69-81.
Ueda, T., and Dai, J. G. 2005. Interface bond between FRP sheets and concrete substrates: properties,
numerical modeling, and roles in member behaviour, Proc, Str Eng and Mat, 7(1), 27-43.

-6-