FORUM

Carbon Fiber Reinforced Polymer Strengthening of Reinforced Concrete Beams: Experimental Study Camille A. Issa, P.E., F.ASCE Professor of Civil Engineering, Lebanese American Univ., Byblos, Lebanon. E-mail: [email protected]

Abdo AbouJouadeh Former Student, Dept. of Civil Engineering, Lebanese American Univ., Byblos, Lebanon.

428 N / mm2 for 14 mm diameter, 582 N / mm2 for 12 mm diameter, and 275 N / mm2 for 6 mm diameter. Fig. 1 shows the cross section of a typical beam. Because one of the goals of the experiment was to determine how to strengthen concrete reinforced members that might exist on any site, a normal weight concrete mix design was used with a target strength of 30 MPa. The concrete mix design had the ratios of 1: 1.5: 2.5 of cement, sand, and aggregates, respectively, with a water-cement ratio equal to 0.50.

Introduction Composite Materials Poststrengthening of a structure becomes necessary if its stability and/or fitness for its functionality under stipulated conditions of use can no longer be guaranteed. More and more of our buildings during their service life will have to undergo structural transformations due to change of utilization or new code imposed load provisions. Many different strengthening methods are available, such as adding of unstressed or prestressed steel, installation of external prestressed reinforcement, increasing crosssections, and many others. In this experimental study, carbon fiber reinforced polymers (CFRP) materials were used for structural strengthening. CFRP materials do not corrode because they are a combination of carbon fibers and an epoxy resin matrix that have very high strength and rigidity in the fiber direction and outstanding fatigue characteristics. The first time the method of CFRP was used outside the laboratory was on the Ibach Bridge at Lucerne in 1991. Since that time, this method of strengthening has been used and developed for a variety of structural repair and rehabilitation projects (An et al. 1991; Dublois et al. 1992; Highway 1993; M’Bazaa et al. 1996; Meier and Kaiser 1991; Meier et al. 1992; Norris and Saadatmanesh 1994; Ritchie et al. 1991; Saadatmanesh and Ehsanni 1991; Shahawy et al. 1995; Triantafillou and Plevris 1991). This study was an attempt to explore the correct application procedure that would result in an increase in flexural capacity, shear strength, and stiffness compared to a reference substrate by varying one CFRP parameter at a time. This procedure demonstrated the percentage increase/decrease in substrates strengthened by the different types and methods of applying CFRP materials.

Two types of CFRP materials were used. The CFRP used for flexural strengthening is composed of SIKA CardoDur (rigid plates) (Sika AG, Baar, Switzerland) laminates S512, and the CFRP used for shear strengthening and anchorage is composed of SIKA Wrap HEX 230C. The properties of these materials are shown in Table 1. The bonding agent used was SIKAdur 30 normal adhesive for the bonding of the CFRP rigid plates and SIKAdur 330 adhesive was used for the CFRP wrap flexible sheets. The technical data are shown in Table 2. The CFRP materials applied to two of beams are shown in Figs. 2 and 3.

Casting of Beams The steel reinforcements were cut on-site to the required length and assembled in cages ready for concrete casting. Plastic spacers were used in formwork as well as corner chamfers to provide beam specimens typical to those found on-site. The concrete was mixed using a small rotary mixer and shoveled into the formwork, and a vibrator was used to minimize air voids in concrete members. The formwork consisted of plywood to provide good finishing of substrates. Two beams were cast at a time, and three cylinders 150 by 300 mm were retained as samples for compressive strength testing.

Carbon Fiber Reinforced Polymer Strengthening Application Five beams of the same reinforced concrete characteristics as previously described were constructed. Beam 1 was used as the con-

Construction Materials Portland cement type I was used in the concrete mix design. The maximum size of the coarse aggregates (crushed limestone) was about 13 mm. The fine aggregates were of natural sand with finess modulus equal to 1.80. Two 14 mm diameter grade 60 bars were used as bottom longitudinal reinforcement and two 12 mm diameter grade 60 bars were used as top longitudinal reinforcement. Grade 40 plain 6 mm diameter stirrups were placed at spacing of 15 cm. The experimental yield stresses (fy) were determined to be

Fig. 1. Typical beam cross section

JOURNAL OF ARCHITECTURAL ENGINEERING © ASCE / DECEMBER 2004 / 121

Table 1. Properties of Carbon Fiber Reinforced Polymer Materials (SIKA 2002) Cardo Dur S512 Strip dimensions Cross section Fiber orientation Density Elastic modulus Tensile strength Strain Breaking strain Tensile failure Shelf life

Wrap HEX 230C 50 mm 1.2 mm 60 mm2 Unidirectional 1.5 g / cm3 165, 000 N / mm2 ⬎2800 N / mm2 1.7% 1.85% 3050 N / mm2 Unlimited

Fiber type Real weight Fabric design thickness Tensile strength of fibers Tensile modulus of fibers Elongation at break Fabric length / roll Fabric width Shelf life

trolling beam. Beam 2 had a CFRP strip applied to the bottom. Beam 3 was the same as Beam 2 except that CFRP warps were applied at the supports. Beam 4 was similar to Beam 3, except that CFRP warps were applied under the location of the point load application. Beam 5 was similar to Beam 4, except that CFRP strips were applied to the top. The layout of CFRP strips and CFRP wrap will typically be as indicated in the application procedure. Figs. 2 and 3 present a better view for the actual application.

Testing A Sheinder UTM D7940 (Reidlingen-West German) four column, closed loop servo-hydraulic testing machine with a 1,000 KN dynamic capacity actuator (see Fig. 8) was used. Mechanical gauges were placed at the midspan, and the load was transferred into two point loads by using a steel I-beam. Electric resistance strain gauges were placed on the CFRP materials, and a computerized data acquisition system was utilized (Fig. 4). The following measures and characteristics were observed before actual testing: • Concrete minimum strength was 15 MPa and cured for a minimum 28 days • Bonding or application of CFRP for 7 to 10 days for proper curing • Electrical strain gauges were connected to CFRP plates • Mechanical dial gauges were used for measuring center and under point load deformation • Supports set up correctly as well as loads at L / 3 apart The following procedures were followed during testing:

High strength carbon fibers 225 g / m2 0.13 mm 3500 N / mm2 230, 000 N / mm2 1.5% 45.7 m 305/ 610 mm Unlimited

• Beams were loaded in 4 point bending through two symmetrical concentrated loads applied at one-third the span length • Load applied at a rate of 10 KN/ min • Deflection was measured and cracks were traced at each load increment • Strain in CFRP measured at different stages of loadings • Experimental data was tabulated

Concrete Properties Compressive strengths were obtained by crushing three 150 by 300 mm cylinders for each concrete batch. The results revealed concrete compressive strengths at testing ages were slightly higher than the mix design target strength. The average compressive strength at 45 days after casting was about 34.04 MPa with an average concrete density of 2 , 321 kg/ m3 (Table 3).

Experimental Results Beam 1 (Controlling Beam without Carbon Fiber Reinforced Polymer) Beam 1 was the controlling beam. This beam was used as a reference for the members strengthened with CFRP materials. From results, one can deduce that the maximum load capacity of the controlling beam is 123 KN. The deflection corresponding to this load is 8.80 mm. Central cracks propagated starting at a 60 KN load and continued until a major failure in the shear zone occurred as expected. Yielding of steel was at a second stage untill total crushing took place (Fig. 5).

Table 2. Technical Data for Bonding Agent (SIKA 2002) Technical data

SIKAdur 30

Appearance Application temperature Shelf life Mix ratio Density Pot life Static E-modulus Adhesive strength Shear strength Consumption

SIKAdur 330 Comp A: white paste Comp B: black paste

10 to 35°C 12 months A:B=3:1 1.77 kg/ l 40: 100 min= 35 to 15° C 12, 800 N / mm2 4 N / mm2 15 N / mm2 0.34 kg/ lm

122 / JOURNAL OF ARCHITECTURAL ENGINEERING © ASCE / DECEMBER 2004

15°C to 35°C 18 months A:B=4:1 1.31 kg/ l 30: 90 min= 35: 15° C 3800 N / mm2 30 N / mm2 1.0 kg/ m2

Fig. 5. Beam 1 during the loading process

Fig. 2. SIKA Cardodur applied to beam

Fig. 3. Beam wrapped with SIKA HEX 230C laminates

Fig. 6. Beam 2 debonding under stress concentration at point load

Fig. 7. Beam 3 cracks traced Fig. 4. Typical testing setup

Table 3. Plain Concrete Properties f ⬘c共MPa兲

Cylinder 1

Cylinder 2

Cylinder 3

Average

Beam 1,2 Beam 3,4 Beam 5,6 Weight (grams) Beam 1, 2 Beam 3, 4 Beam 5, 6

33.00 32.90 31.00 Sample 1 12,400 12,560 12,500

40.70 40.30 24.50 Sample 2 12,460 12,600 12,505

32.00 35.00 37.00 Sample 3 12,430 12,580 12,525

35.23 36.07 30.83 Average 12,430 12,580 12,510

Fig. 8. Beam 4 crushing at failures of bondage and delamination of laminates

JOURNAL OF ARCHITECTURAL ENGINEERING © ASCE / DECEMBER 2004 / 123

Table 4. Summary of Results for the Experimental Study

Fig. 9. Beam 4 failure of bonds and cracking due to shear

Beam 2 (with One Rigid Strip of Carbon Fiber Reinforced Polymer in Flexure Extending to Supports) The second beam is similar to the first but having a Sika Carbodur strip (S512) applied to the bottom of the beam for flexure. The results indicate that Beam 2 can handle a 164 KN load with a deflection of 7.50 mm at failure. This beam test revealed premature debonding (anchorage failure) of the strip, especially at support areas, due to the high shear stress. The precracking stage finished at a 75 KN load, and concrete defects were observed under the point loads areas at 145 KN (Fig. 6).

Beam 3 (Similar to Beam 2 Plus Anchorage at Supports with Flexible Sheets) Beam 3 was similar to Beam 2 except that anchoring at supports is added using flexible sheets. These flexible sheets are expected to prevent debonding especially at supports and serve in increasing shear capacity. Load capacity reached 210 KN with a deformation of 6.89 mm. Cracks propagation started at 109 KN, and concrete defects started under point loads at 190 KN. No debonding occurred due to the anchorage provided by the wrapping sheets at the supports. The high load that was reached induced the specimen to fail in shear and to bulge the concrete under the point loads (Fig. 7).

Beam 4 (Similar to Third Beam Plus Flexible Wrapping Sheets under Point Loads) For the fourth beam, flexible sheets were used under point loads in order to increase shear capacity and to provide better confinement of the concrete under the point loads. Load capacity and deflection at failure did not vary from Beam 3. The improvement was in the precracking stage, rigidity of the member, and the confinement at the high stresses of the point loads. Cracks were observed at the compression zone between the point loads. The

Fig. 10. Beam 5 beginning of delamination of strips at 225 KN

Deflection (mm)

Beam 1

Beam 2

Beam 3

Beam 4

Beam 5

123 KN 164 KN 210 KN 224 KN Failure Load (KN)

8.80

4.67 7.50

2.24 3.66 6.89

3.20 4.50 7.65

123

164

210

208

2.92 4.21 6.07 7.75 224

specimen failed due to the delamination of the rigid strip at the bottom and bondage failure between the beam and wrapped flexible sheets. The ultimate failure in this beam occurred in shear (Figs. 8 and 9).

Beam 5 (Similar to Fourth Beam Plus Flexible Sheets in Compression Zone) Beam 5 was the last to be tested. It differed from Beam 4 in that flexible sheets were applied to the compression zone. The compression fibers were placed perpendicular to the shear fibers. Load capacity increased up to 223.5 KN with a deflection of about 7.75 mm. This beam acted as the previous one but showed better rigidity, debonding of wrapped flexible sheets occurring at a higher loads, as well as delamination and failure (Fig. 10).

Discussion of Test Results The percentage of increase in load capacity is significant, as shown in Table 4. The load increased by 33, 70, 69, and 82% between Beams 2 to 5, respectively, when compared to the controlling beam. An increase in load capacity between each type of beam when compared to each other is also shown. Data collected for deflection shows a 300% improvement in serviceability of strengthened members between the fifth and first beam at different loads. This is similar to all beams when compared with each other at specific loads. In addition, when conducting the test, a gain in stiffness of members was one of the most important factors that varied from one specimen to another. All the specimens showed three stages: precracking, cracking to yield, and yield to crushing. Cracking stages improved with improving application of CFRP laminates in various locations. The more anchorage and wrapping used, the better the specimen performed in terms of shear, bondage, flexural strength, and duration for total destruc-

Fig. 11. Comparison of load deflection curves for five beams

124 / JOURNAL OF ARCHITECTURAL ENGINEERING © ASCE / DECEMBER 2004

tion of reinforced strengthened members. This can be seen in the Fig. 11 comparison chart.

Conclusion and Recommendations It can be concluded from experimental results that the use of CFRP materials is one of the most powerful techniques for strengthening concrete structural reinforced members. Strengthening of concrete with CFRP results in an increase in load capacity as well as an increase in stiffness. Better performances and serviceability measures are encountered when anchorage is taken into consideration. Stiffness and rigidity of members progress with an increased application of CFRP laminates, which avoids crushing or the total destruction of members without warning. These results indicate that the application of CFRP laminates whenever needed, taking into consideration anchoring, rigidity, and stiffness, does actually results in an increase of strength of beams and provides additional load carrying capacity.

Acknowledgments These tests were conducted in the Materials Testing Laboratories at the Beirut Arab University (BAU). The author would like to thank Sika Near East S.A.L for providing the CFRP materials.

References An, W., Saadatmanesh, H., and Ehsani, M. (1991). “RC beams strengthened with GFRP plates II: Analysis and parametric study.” J. Struct. Eng., 117(11). 3434–3455. Dublois, M., Picard, A., and Beaulieu, D. (1992). “Reinforcement de

putres en beton arme a l’aide de materiaux composites: Etudes theorique et experimentale.” Proc., 1st Int. Conf. on Advanced Composite Materials in Bridges and Structures, Sherbrooke, 265–275. “Highway Bridge Replacement and Rehabilitation Program.” (1993). “11th Rep. of the Secretary of Transportation to the Congress of the United States.” Bridge Div., Federal Highway Administration, Washington, D.C. M’Bazaa, I., Missihoun, M., and Labossiere, P. (1996). “Strengthening of reinforced concrete beams with CFRP sheets.” Proc. 1st Int. Conf. on Composites in Infrastructure, 746–759. Meier, U., and Kaiser, H. (1991). “Strengthening of structures with CFRP laminates.” Proc., Advanced Composites Materials in Civil Engineering Structures, ASCE, New York, 224–232. Meier, U., Deuring, M., Meier, H., and Schweger, G. (1992). “Strengthening of structures with CFRP laminates: Research and applications in Switzerland.” Proc. 1st Int. Conf. on Advanced Composite Materials in Bridges and Structures, 243–251. Norris, T., and Saadatmanesh, H. (1994). “Improving the serviceability of concrete beams using carbon fiber reinforced polymer (CFRP) sheets.” Grant No. DDEGRF-93-P-05.82, Federal Highway Administration, Washington, D.C. Ritchie, P., Thomas, D., Lu, W. L., and Connelly, G. (1991). “External reinforcement of concrete beams using fiber reinforced plastics.” ACI Struct. J., 88, 490–500. Saadatmanesh, H., and Ehsani, M. (1991). “RC beams strengthened with GFRP plates I: Experimental study.” J. Struct. Eng., 117(11), 3417– 3433. Shahawy, M., Arockiasamy, M., Beitelman, T., and Sowrirajan, R. (1995). “Reinforced concrete rectangular beams strengthened with CFRP laminates.” Composites, 27B, 225–233. SIKA, (2002). “Data book 2002.” Sika Near East, S.A.L, Beirut, Lebanon. Triantafillou, T., and Plevris, N. (1991). “Post-strengthening of R/C beams with epoxy-bonded fiber composite materials.” Proc., Advanced Composite Materials in Civil Engineering Structures, ASCE, New York, 245–256.

JOURNAL OF ARCHITECTURAL ENGINEERING © ASCE / DECEMBER 2004 / 125