GFRP-STEEL HYBRID REINFORCED CONCRETE BRIDGE DECK SLABS IN QUEBEC, CANADA

Transcription

1 Fourth Asia-Pacific Conference on FRP in Structures (APFIS 213) December 213, Melbourne, Australia 213 International Institute for FRP in Construction GFRP-STEEL HYBRID REINFORCED CONCRETE BRIDGE DECK SLABS IN QUEBEC, CANADA E. Ahmed and B. Benmokrane Department of Civil Engineering The University of Sherbrooke, Sherbrooke, Quebec, Canada. ABSTRACT In GFRP-Steel hybrid reinforced concrete (RC) bridge deck slabs, the non-corrodible glass fiber-reinforced polymer (GFRP) bars are used in the top reinforcement mat and the steel bars are used in the bottom reinforcement mat. This technique is being used by the Ministry of Transportation of Québec (MTQ) in collaboration with the University of Sherbrooke (UdeS) as it is expected to yield cost-effective and durable concrete bridge decks. This paper highlights the recent hybrid-reinforced bridge deck applications on the extension of Hwy 4 in Sherbrooke (Québec, Canada). So far, the project included four hybrid bridges where three of them were single-span simply supported bridges (P-1, P-2N, and P-2S) and the fourth one was three-span continuously supported bridge (P-64). In addition, this paper presents and discusses the typical results of the live load field tests conducted on the P-2S bridge using three calibrated trucks. KEYWORDS Reinforced concrete, deck slab, bridge, GFRP, hybrid, wheel load, load test, fiber optic sensor (FOS). INTRODUCTION In typical slab-on-girder bridge deck, the top reinforcing mat is closer to the concrete surface and consequently, is susceptible to chloride and chemical exposure and consequently accelerated corrosion. On the other hand, the bottom reinforcement mat is not susceptible to such exposure. In addition, the design of these bridge decks controls the crack width which limits the penetration of the chloride from the top surface to the bottom reinforcement layer. Thus, design engineers and municipalities proposed the use of the non- corrodible GFRP bars in the top reinforcement mat and maintain the steel bars in the bottom reinforcement mat (hybrid-reinforced bridge deck slabs). This technique is expected to yield cost-effective and durable concrete bridge decks. Recently, the Ministry of Transportation of Québec (MTQ) initiated an extensive research project in collaboration with the University of Sherbrooke (UdeS) aimed at investigating the structural performance of hybrid-reinforced concrete bridge decks through some new concrete bridges that are being constructed on the extension of Hwy 4 (Sherbrooke, Québec, Canada). The first bridge in these series was the 4 overpass bridge (P-1) (Ahmed and Benmokrane 212). The second and third ones were the twin bridges over Ste- Catherine (P-2N and P-2S) (Ahmed and Benmokrane 213) and the fourth one was the three-span continuously supported bridge (P-64). This paper highlights the recent hybrid bridge deck applications on the extension of Hwy 4 in Sherbrooke (Québec, Canada). So far, the project included four hybrid bridges where three of them were single-span simply supported bridges (P-1, P-2N, and P-2S) and the fourth one was three-span continuously supported bridge (P-64). In addition, this paper presents and discusses the typical results of the live load field tests conducted on P-2S (one of a twin bridges). These applications and field tests will lead to introducing the hybrid reinforcement technique to the Canadian Highway Bridge Design Code (CHBDC) (CSA S6 2) which permits using glass fiber-reinforced polymer (GFRP) as main reinforcement in bridge deck slabs. DESCRIPTION OF THE PROJECT The extension of the Hwy 4 (Sherbrooke, Québec, Canada) is about 13 km. This extension include seven bridges. So far, four of those bridges were constructed using the hybrid-reinforced bridge decks technique where

2 the bottom mat was maintained as steel bars and the top mat was replaced with GFRP bars. Figure 1 shows the locations of these bridges. Figures 2 to 4 shows the layouts of these bridges. This paper will present the Ste- Chaterine Overpass twin bridges (P-2N and P-2S) and the field test results conducted on one of them (P-2S). Layout of one of the twin bridges is shown in Figure 3. Each of twin bridges has five steel girders simply supported over a single span of m (Figure a). The deck slab is a 2-mm thickness (Figure b) concrete slab continuous over four spans of 2.6 m each with an average overhang of about 1. m on each side (measured in the perpendicular direction to axis of the girders). P-1 4 Overpass P-2N P-2S P-64 Twin Bridges Continuous Bridge Figure 1. Locations of the bridges on the Hwy 4 extension. Figure 2. Layout of the P-1 bridge. Figure 3. Typical layout of the twin bridges (P-2N and P-2S). Figure 4. Layout of the continuous bridge (P-64).

3 4 1 Fixed Side Additional GFRP No. 14 (Top) Additional GFRP No. 14 (Top) spacing GFRP No. 14 (Top) Steel (Bottom) Base Additional 126 BEAM BEAM 4 96 for GFRP 6 for Steel BEAM 3 BEAM 2 BEAM 1 (a) Plan view GFRP No.2@2 mm (top) Steel M@2 mm (bottom) Moving Side JOINT C.L 4 GFRP No. 2@2 mm Add. GFRP No. 2@14 mm GFRP No. 2@14 mm Slope 4.1% 2 Steel 2M@ mm Steel M@2 mm B3 B2 B1 Beams 4 26 = 6 (b) Cross-section 4 free For GFRP 7 free for steel Figure. Geometry and reinforcement details of the twin bridges (P-2N and P-2S). DESIGN AND INSTRUMENTATION OF P-2N AND P-2S BRIDGES Material Properties Normal-strength concrete (Type V MTQ) with a 28-day concrete compressive strength of 3 MPa was used for the bridge deck slab. Sand-coated GFRP bars were used for the top mat of the bridge deck slab (No. 2, 19.1 mm diameter) while galvanized steel bars M and 2M were used for the bottom mat of the bridge deck slab. The GFRP bars were made of high strength E-glass fibers with a fiber content of 76.8% (by weight) in a vinyl ester resin using the pultrusion and an in-line coating processes (Pultrall Inc., Thetford Mines, Québec, Canada). The average ultimate tensile strength and the tensile modulus of elasticity of the GFRP bars were 139±8 MPa and 4.3±.8 GPa, respectively. Design of the Concrete Bridge Deck Slab The bridge deck slab was designed according to the flexural design method of the CHBDC (CSA S6 2). The design bending moments were based on a maximum wheel load of 87. kn (CL-62 Truck). The design service load for the deck slab was taken as =1.2 kn, where 1.4 is the impact coefficient, and.9 is the live load combination factor, while the design factored load was taken as =28.2 kn, where 1.7 is the live load combination factor. The deck slab was designed based on serviceability and ultimate limit states. The crack width of the concrete slab and allowable stress limits were the controlling design factors. The MTQ has selected to limit the maximum allowable crack width to. mm and the stresses in the GFRP bars to 2 and % of the ultimate strength of the material under service and sustained loads, respectively. Based on this design approach, the bridge deck slab was entirely reinforced with two reinforcement mats using No. 2 GFRP bars and 2M and M steel bars. For the bottom reinforcement mat, 2M steel bars spaced at and M steel bars at 2 mm in the transverse and longitudinal directions, respectively. For the top reinforcement mat, No. 2 GFRP bars spaced at 14 and 2 mm in the transverse and longitudinal directions, respectively, were used. Top and bottom clear concrete cover of and 38 mm, respectively, were used. Additional No. 2 GFRP bars spaced at 14 mm were placed in the top transverse layer at the two cantilevers (Figure b), as well as in the top longitudinal layer at the ends of the deck slab. It should be mentioned that the construction of the bridge deck slab started on April 212 and the two bridge deck slabs were cast on May 31, 212 starting by the Southern Bridge and ending with the Northern one (instrumented one). Figure 6 shows the reinforcement of the bridge and the concrete casting, respectively. (a) Reinforcement of the bridge deck (b) Concrete casting Figure 6. Bridge construction: (a) Reinforcement of the bridge deck; (b) Concrete casting.

4 Instrumentation The bridge was instrumented at the mid-span section to capture the strains in the reinforcing bars, as shown in Figure 7. Fiber optic sensors (FOSs) were glued on the transverse GFRP reinforcing bars of the top mat and the transverse steel bars of the bottom mat. The FOS were glued on the GFRP bars (T1 T) at locations of support girders (maximum negative moment) and on the steel bars (B1 B4) at centerlines between the support girders (maximum positive moment) (Figure 7). The GFRP and steel bars were instrumented at the structural laboratory of the University of Sherbrooke. Thereafter, the bars were shipped to the construction site where they were installed in the designated locations. The objective of using FOS was to allow for the long-term monitoring and future field tests of the bridge. During static tests, deflection of the steel girders was measured (D1 D) with a theodolite and a system of rulers installed at the bridge midspan (Figure 7). Fixed Side 43 4 Moving Side Stn. 3 Stn. 2 Stn. 1 D T B4 D4 T4 Bottom Bar (Steel) B3 D3 Top Bar (GFRP) T3 B2 D2 T2 B1 D1 T1.7 L. L.2 L Figure 7. Instrumentation for strain and deflection measurements LIVE LOAD TESTING OF BRIDGE Trucks and Paths The bridge was tested on October 3, 212 for service performance as specified by the CHBDC (CSA S6 2) using three three-axle calibrated trucks. Trucks No. 1 to 3 had loads of about 72 kn on the front axle and approximately 9 kn on each back axle. The bridge deck was tested using six truck paths using one, two, and three trucks at three different stations (truck stops) in the longitudinal direction of the bridge (Stn 1 Stn 3). The stations were selected at the quarters and at the middle of the bridge to capture the variation in the straining actions with the trucks location on the bridge. The six paths shown in Figure 8 were marked on the bridge deck as well as the three stations. Readings were recorded at a truck(s) station when the midpoint of the truck s second and third axle was directly over the station. Figure 9 shows the trucks on the bridge during testing. (1) (2) (1) Path 1 Path 2 Path (2) (1) (2) (1) (2) (3) (1) Figure 8. Truck loads and paths of the trucks during testing. Figure 9. Trucks during testing.

5 Live Load Test Results After each path the bridge deck slab was visually observed for any signs of cracking over the girders (negative moment areas). No cracks were observed at any location in the top surface of the bridge deck at girders locations. Strain measurements Figure presents the strains in the bottom transverse steel bars and top transverse GFRP bars. The strains were very low in the bottom transverse steel bars and top transverse GFRP bars. The strain in the bottom transverse steel bars at its maximum location was microstrains. Similarly were the strains in the top transverse GFRP bars where its maximum value was 2 microstrains. It should be noted that strains in steel reinforcing bars of the same order as those measured here were reported for similar bridges that were constructed using FRP and steel bars in their concrete deck slabs (El-Salakawy et al. 23, 2; Ahmed and Benmokrane 2). The maximum measured tensile strains in the GFRP bars are less than 1% of the GFRP s ultimate strains (139/43=29 microstrains). The design service load of 1.2 kn (specified by the CHBDC is greater than the maximum wheel load of the trucks used in the field test which equals 4 kn by approximately 2. times. However if the maximum values of the strains measured in the field are linearly extrapolated (multiplied by 2.) the resulting values of the tensile strains will be about 63 microstrains in the top transverse GFRP bars. These values are still less than 1% of the ultimate strains of the GFRP bars. According to the CHBDC (CSA S6 2), the allowable stress or strain limit for GFRP bars in concrete slabs is 2% of the material s ultimate stress or strain values. Similar extrapolation of the strains in the bottom transverse steel bars yielded an strain value of about 2% of the ultimate strain capacity of the steel bars (based on yield strain of 2 microstrains) which is also very low. The very small measured strains (Figure ) indicate that the arch action exists in the restrained hybridreinforced concrete bridge decks slabs. Although, the bridge deck was designed according to the flexural design method of the CHBDC (CSA S6 2), the slabs did not show a real flexural response due to the arch action effect. Furthermore, since the bridge deck meets the requirements of the CHBDC for the empirical method, it was possible to design it using the empirical method. The design of this bridge deck based on the empirical method yields M steel bars@3 mm as bottom transverse reinforcement (minimum reinforcement), M steel bars@3 mm (minimum reinforcement) as bottom longitudinal reinforcement, and GFRP bars mm (minimum reinforcement) as top transverse and longitudinal reinforcement. This design saves a significant amount of the transverse reinforcement (steel and GFRP bars) compared to the design using the flexural method. Strain in the bars (microstrain) Path 3 T1 T2 T3 T4 T B1 B2 B3 B4 Strain in the bars (microstrain) Path 3 T1 T2 T3 T4 T B1 B2 B3 B4 T1 T2 T3 T4 T FOS Sensor Identification (a) Strains in GFRP bars B1 B2 B3 B4 FOS Sensor Identification (b) Strains in steel bars Figure. Strains in the GFRP and steel bars. Deflection Measurements Figure 11a shows the variation of the measured deflection of the steel girders with the trucks location on the bridge while Figure 11b presents the deflection of the steel girders at the bridge midspan due to trucks located at midspan (Stn 2) for the different paths. The figure indicates that the truck loading was not evenly distributed on the steel girders. The girder closest to the truck path deforms more than those further away. This was more obvious when the truck traveled over or near the edge girder. As shown in Figure 11b, the single truck following Path 1 produced the peak deflection in Girder of 12. mm (L/3617). The peak deflection with the two calibrated trucks traveling simultaneously along Paths 4 and was 14. mm (L/31) in Girder 2. Furthermore, as noted from Paths 4 and, when the load is applied onto two lanes the deflection distribution was better and evidenced from Figure 11b.

6 Deflection (mm) 2 2 Deflection (mm) 2 2 Path 1 Path 2 Path 3 Beam Truck location along the bridge (%L) (a) Variation with the truck location Beam Number (b) Deflection of all girders CONCLUSIONS Figure 11. Maximum Measured deflection of steel girders (trucks at midspan-stn 2). Based on the details presented herein and the results of the field-loading test, the following conclusions can be drawn: The maximum tensile strain in the top transverse GFRP bars was less than 1% of the ultimate tensile strain of the GFRP bars. Nevertheless, it is lower than the expected strains using the flexural design method. This result suggests that the CHBDC (CSA S6 2) flexural design method overestimates the calculated design moments. The successful bridge applications using hybrid-reinforced concrete bridge deck techniques provides a step forward towards introducing this technique the CHBDC (CSA S6) and the other FRP design codes and guides. The long-term monitoring of these bridge decks, which is in progress, will enable understating the long-term performance in real environmental and service conditions. ACKNOWLEDGMENTS The financial support of the Natural Science and Engineering Research Council of Canada (NSERC), the Fonds québécois de la recherche sur la nature et les technologies (FQRNT), and Ministry of Transportation of Quebec (MTQ). The authors are also grateful to the Eng. Nadeau Dominique from exp. (former Teknika-HBA) (Sherbrooke, Quebec). REFERENCES Ahmed, E., and Benmokrane, B. (212). Design, Construction, and Testing of Hybrid-Reinforced Concrete Bridge Deck: 4 Overpass Bridge, Proceedings of the 6 th Int. Conference on FRP Composites in Civil Engineering, Rome, Italy, June 13-, 8p. Ahmed, E.A. and Benmokrane, B. (213). Construction and Testing of Hybrid Reinforced Concrete Bridge Deck Slabs: Ste-Catherine Overpass Twin Bridges, Proceedings of the 11 th International Symposium on Fiber-Reinforced Polymer Reinforcement for Concrete Structures (FRPRCS-11), Guimarães, Portugal, June 26-28, p. Canadian Standards Association (CSA). (2). Canadian Highway Bridge Design Code (CAN/CSA S6), Rexdale, ON, Canada. El-Salakawy, E.F., Benmokrane, B., and Desgagné, G. (23). FRP Composite Bars for the Concrete Deck Slab of Wotton Bridge, Canadian Journal of Civil Engineering, 3(), El-Salakawy, E.F., Benmokrane, B., El-Ragaby, A., and Nadeau, D. (2) Field Investigation on the First Bridge Deck Slab Reinforced with Glass FRP Bars Constructed in Canada, Journal of Composites for Construction, ASCE, 9(6),