CICE 2010 - The 5th International Conference on FRP Composites in Civil Engineering September 27-29, 2010, Beijing, China

Bond Strength of FRP Rebar to Concrete: Effect of Concrete Confinement

Shahriar Quayyum & Ahmad Rteil (ahmad.rteil@ubc.ca) School of Engineering, The University of British Columbia, Kelowna, BC, Canada

ABSTRACT According to ACI 440.1R-06, the presence of transverse reinforcement does not affect the bond strength of fiber reinforced polymer (FRP) rebar to concrete. This conclusion was based on a limited number of data (19 confined beam tests) available in literature at the time. In the present study, 177 beam bond test data, failed by concrete splitting, was collected to investigate the effect of concrete confinement on the bond strength with FRP rebar. Of these 177 beam-type specimens, 105 specimens had transverse reinforcement. It was observed that the presence of transverse reinforcement increased the bond strength of FRP rebar to concrete by 10%-15%, which eventually decreases the development length needed for attaining the tensile strength of FRP rebar. A linear regression was performed on the collected data to develop an equation to determine the bond strength of FRP rebar to concrete in presence of transverse reinforcement and the equation was compared with the ACI 440.1R-06 equation and the experimental results. Based on the analysis, it was found that the proposed equation is in good agreement with the experimental results and it yields a better estimate of bond strength than the ACI 440.1R-06 equation. KEY WORDS

1 INTRODUCTION
Studies on bond behavior of steel reinforcement have demonstrated that the presence of transverse reinforcement confines developed and spliced bars by limiting the progression of splitting cracks and, thus, increases the bond force required to cause failure (Tepfers 1973; Orangun et al. 1977; Darwin and Graham 1993a, b). ACI 408R-03 proposed an equation for estimating the bond strength of steel rebar based on the study of Orangun et al. (1975) which takes into account the effect of transverse reinforcement. Fiber reinforced polymer (FRP) rebars have been recently used in concrete to increase its durability. However, there are only few literature studies available on the effect of concrete confinement of FRP rebar. Due to the limited availability of experimental data, no definite relationship can be established between transverse reinforcement and bond strength for FRP rebar. ACI 440.1R-06 proposed the following bond strength equation of FRP rebar to concrete (in SI units):
u fc c 0.33  0.025 d c b  8.3 d l b embed

(1)

where u is the bond strength of FRP rebar to concrete, fc is the compressive strength of concrete, c is the lesser of the cover to the center of the bar or one-half of the center-on-center spacing of the bars being developed, db is the bar diameter and lembed is the embedment length of

the rebar in concrete. This equation was developed from the study by Wambeke and Shield (2006) in which a linear regression was performed on 67 beam-type specimen data which failed by concrete splitting. Forty eight (48) of these specimens were unconfined and 19 were confined. From this study, it was found that the bar surface did not appear to affect the results, nor surprisingly did the presence of confining reinforcement. Darwin et al. (1996) found that confining steel used in beams that had steel reinforcing bars with a high relative rib area had more of a beneficial increase in the bond force over the same-size steel bars with moderate rib area. The counterargument was proposed in Wambeke and Shields (2006) study. The GFRP bars have a very low relative rib area and, therefore, the presence of confinement may not increase the average bond stress. Additional research into the effect of confining reinforcement on bond of FRP bars, however, was recommended. The study reported herein provides an investigation on the effect of the presence of transverse reinforcement on the bond strength of FRP rebar to concrete based on 177 beam test data (105 confined and 72 unconfined) which included carbon, glass and aramid FRP rebars with different surface texture (sand coated, spiral wrap, helical lug) and which failed by concrete splitting.

2 DESCRIPTION OF THE DATABASE

In this study, a database of 177 beam-type specimen was

L. Ye et al. (eds.), Advances in FRP Composites in Civil Engineering Tsinghua University Press, Beijing and Springer-Verlag Berlin Heidelberg 2011

582

Proceedings of the 5th International Conference on FRP Composites in Civil Engineering

created from the available literature. Only beam-type specimens that failed by concrete splitting were considered. These specimens consisted of beam end specimens, beam anchorage specimens, and splice specimens. The data included glass, carbon and aramid FRP rebars with different surface textures such as sand coated, spiral wrap and helical lugs. Table 1 shows the breakdown of the database with respect to type of FRP and surface texture. A complete listing of all the data for each test can be found in Quayyum (2010).
Table 1 Classification of the specimens with respect to type of FRP and surface texture of rebar GFRP Spiral wrap Helical lug Sand coated Total 56 62 10 128 CFRP 1 12 28 41 AFRP 3 3 2 8 Total 60 77 40 177

u c fc c

0.03  0.14

d c b  9.0 d l b embed

(2)

where uc is the bond strength of unconfined FRP rebars to concrete (i.e. due to concrete cover only). When the predicted values were plotted with the experimental values (Figure 1), it was found that the bond strength values obtained from the proposed equation are very close to the actual test results. The ACI equation is also in close proximity to the proposed equation. The average of the ratio of experimental to predicted values was found to be 0.92 with a standard deviation of 0.22 which is quite reasonable. Thus, Equation (2) can provide a reasonable estimate of bond strength of FRP rebar to concrete.

The nominal diameter of the rebars ranged from 8 mm (0.3 in) to 29 mm (1.2 in). The compressive strength of the concrete for the beam tested ranged from 27 MPa (3900 psi) to 66 MPa (9600 psi). Eighty five percent (85%) of the specimens had concrete cover between one and three bar diameters and the embedment lengths of all the specimens ranged between 4 to 116 bar diameters. The database included both confined and unconfined beam tests on top and bottom bars (105 confined bottom, 50 unconfined bottom and 22 unconfined top bars). There were 105 beam tests that contained transverse reinforcement. The nominal diameter of the steel stirrups used in the specimens varied between 8 mm (0.3 in) to 11.3 mm (0.6 in) with a spacing between 78 mm (3 in) and 150 mm (6 in) and all of the tests were performed on bottom bars. Thus the database contains an adequate representation of the different parameters that appear to influence bond performance of FRP rebar to concrete and therefore, the data can be thought to be sufficient to perform statistical analysis.

Figure 1 Test vs predicted average bond stress for unconfined bottom bar specimens failed by concrete splitting

4 CONFINED BEAM TESTS WITH SPLITTING FAILURES

In this study, there were 105 beam-type specimens which had transverse reinforcement and failed by concrete splitting. From the accumulated data, it was found that the presence of transverse reinforcement had certain positive impact on the bond behavior of the specimens. Figure 2 shows the normalized average bond stress plotted against normalized embedment length for both confined and unconfined specimens which failed by concrete splitting. Figure 2 clearly shows that the presence of transverse reinforcement increased the overall bond strength of FRP rebar to concrete. The increase in bond strength can be regarded as the strength of an unconfined rebar plus the strength contributed by the transverse reinforcement (Orangun et al. 1975). The transverse reinforcement contribution (utr) to bond stress was calculated by subtracting uc, as determined from equation (2), from the total bond stress achieved in a confined splice test, uconfined i.e. utr= uconfined - uc. The value of utr normalized by square root of fc was plotted against Atr/sndb for the bars considered

3 UNCONFINED BEAM TESTS WITH SPLITTING FAILURES

There were 72 unconfined beam tests that failed by splitting of the concrete. Of these 72 tests, 22 tests were performed on specimens where the bars were cast as top bars. These 22 tests were not used to develop the bond strength equation. Using the same approach as Orangun et al. (1975), a linear regression analysis on the normalized cover (cover to the center of the bar divided by the nominal bar diameter) and the inverse of the normalized embedment length was used to develop Equation (2) from the 50 beam tests on bottom bars.

September 2729, 2010, Beijing, China

583

required development length) by 10-15%. This will reduce the amount of FRP material to be used in construction and hence, decrease the overall cost of construction. This in turn will encourage the use of FRP rebars in construction projects.

Figure 2 Normalized average bond stress plotted against normalized embedment length for bottom bar specimens failed by concrete splitting

(Figure 3). The straight line fit proposed led to the following equation:
u tr f' c A 2.9 tr snd b

(3)

Figure 3 Effect of transverse reinforcement for 105 confined tests with splitting failures

Where Atr is the area of transverse reinforcement normal to the plane of splitting through the bars, s is the center to center spacing of the transverse reinforcement and n is the number of bars being developed along the plane of splitting. The total bond strength of a FRP bar with transverse reinforcement was determined by combining Equations (2) and (3) as follows:
u f cc 0.03  0.14 d A c  9.0 b  2.9 tr db lembed sndb

(4)

Equation (4) gives bond strength values for FRP rebar to concrete for splitting mode of failure. Figure 4 presents comparison of the proposed equation against the ACI 440.1R-06 equation and the test results. It was observed that ACI 440.1R-06 equation underestimates the bond strength of FRP rebars to concrete in presence of transverse reinforcement which has not been taken into consideration for the development of the ACI 440.1R-06 equation. On the other hand, the proposed equation takes into account the effect of transverse reinforcement and hence, it gives less conservative estimate of bond strength than the ACI 440.1R-06 equation. Also, the proposed equation shows good agreement with the test results and the perfect fit line which gives indication about the satisfactoriness of the equation. The proposed equation results in shorter development length required to achieve the tensile strength of FRP rebar because it takes advantage of the presence of confinement. It was found that the proposed equation can increase the bond strength (or reduce the

Figure 4 Test vs predicted average bond stress for confined bottom bar specimens failed by concrete splitting

5 CONCLUSION
The study reported herein provides an investigation on the effect of transverse reinforcement on the bond strength of FRP rebar to concrete. It was observed that the presence of transverse reinforcement increases the bond force of FRP rebar to concrete. The increase in bond force was about 10-15% which eventually will decrease the development length needed to acquire the desired tensile strength of the rebar. Hence, the presence of transverse reinforcement should be taken into consideration while calculating the development length of FRP rebar. Based on the analysis, an equation was developed for evaluating the bond strength of FRP rebar to concrete and it was found that the proposed equation

584

Proceedings of the 5th International Conference on FRP Composites in Civil Engineering height and spacing on bond strength of reinforcing bars. SL Report 93-1, University of Kansas Center for Research, Lawrence, Kans., 68 pp. Orangun, C. O., Jirsa, J. O., and Breen, J. E. 1975. The strength of anchor bars: a reevaluation of test data on development length and splices. Research Report 154-3F, Center for Highway Research, The University of Texas Austin, Austin, Tex., 78 pp. Orangun, C. O., Jirsa, J. O., and Breen, J. E. 1977. Reevaluation of test data on development length and splices. ACI Journal, Proceedings V. 74, No. 3: 114-122. Quayyum, S. 2010. Bond-slip modeling of FRP rebars in concrete. MS Thesis (in progress). The University of British Columbia, Kelowna, BC, Canada. Tepfers, R. 1973. A theory of bond applied to overlapping tensile reinforcement splices for deformed bars. Publication 73:2. Division of Concrete Structures, Chalmers University of Technology, Goteborg, Sweden, 328 pp. Wambeke, B. W., and Shield, C. K. 2006. Development length of glass fiber-reinforced polymer bars in concrete. ACI Structural Journal, V. 103, No. 1: 11-17.

is in good agreement with the actual test results and it gives less conservative estimate of bond strength than the equation proposed by ACI 440.1R-06 in presence of transverse reinforcement. In short, the proposed equation can provide a more cost effective design than the equations reported in the current literature.

REFERENCES
ACI Committee 408, 2003. Bond and development length of straight reinforcing bars in tension (ACI 408R-03). American Concrete Institute, Farmington Hills, MI, 49 pp. ACI Committee 440, 2006. Guide for the design and construction of structural concrete reinforced with FRP bars (ACI 440.1R-06). American Concrete Institute, Farmington Hills, MI, 44 pp. Darwin, D., Zuo, J., Tholen, M., and Idun, E. 1996. Development length criteria for conventional and high relative rib area reinforcing bars. ACI Structural Journal, V. 93, No. 3: 347-359. Darwin, D., and Graham, E. K. 1993a. Effect of deformation height and spacing on bond strength of reinforcing bars. ACI Structural Journal, V. 90, No. 6: 646-657. Darwin, D., and Graham, E. K. 1993b. Effect of deformation