Fiber Reinforced Polymer Composites for Infrastructure Strengthening - From Research to Practice

Transcription

1 Fiber Reinforced Polymer Composites for Infrastructure Strengthening - From Research to Practice Antonio Nanni University of Missouri Rolla and University of Naples Federico II Abstract This paper focuses on the use of emerging technologies for the upgrade of the transportation infrastructure. A number of bridge rehabilitation projects conducted in the State of Missouri in the last three years were selected for presentation. The review is limited to bridge superstructure strengthening to correct flexure and shear deficiencies in non-seismic applications. Strengthening of concrete members with externally bonded fiber reinforced polymer (FRP) laminates or near surface mounted (NSM) bars has received remarkable attention. The design and construction principles for use in practice have been finalized by the American Concrete Institute (ACI). On the application side, FRP materials have been used in some multi-million dollar projects for strengthening bridges, parking garages, multi-purpose convention centers, office buildings and silos. The drivers for this technology are several, but perhaps the most relevant ones are the ease and speed of installation. In the repair/upgrade arena, one of the most important unresolved questions remains that of durability. Addressing this issue will increase the degree of confidence in the technology and allow for its full exploitation. Keywords: Bridges, construction, design, flexure, externally bonded reinforcement, fiber reinforced polymers, near surface mounted reinforcement, reinforced concrete, repair, shear, strengthening. 1 Introduction Fiber reinforced polymer (FRP) materials are composites consisting of high strength fibers embedded in a polymeric resin (Figure 1) (Nanni 1999). Fibers in an FRP composite are the load carrying elements, while the resin maintains the fibers alignment and protects them against the environment and possible damage. Among commercially available fibers, carbon ones exhibit the highest strength and stiffness when compared with steel. The type of fiber is selected based on mechanical properties and durability requirements, while the type of resin depends upon environmental and constructability needs. Perhaps the most relevant property of carbon FRP (CFRP) composites for construction use is their resistance to corrosion that allows having them installed on the concrete surface. FRP laminates have been used worldwide to strengthen, repair or add ductility to existing concrete bridges and buildings in the last fifteen years. Currently, the Federal Highway Administration (FHWA) and State Departments of Transportation (DOT) are trying to advance FRP technology to contribute to the rebuilding of the US transportation infrastructure. The idea is to avoid, whenever possible, replacement by rehabilitating and maintaining the existing and deficient bridges (Alkhrdaji et al 1999).

2 Polymer (Resin) Fiber Reinforcement Figure 1 : Representation of FRP Material In recent years, the technology of strengthening reinforced concrete (RC) and prestressed concrete (PC) members with externally bonded FRP laminates has been widely investigated and reported. Similar to steel plate bonding, FRP laminate bonding involves adhering thin flexible fiber plies to the concrete surface with a thermoset resin. This technique, known as manual lay-up, may be used to increase the shear and flexural capacity of beams and slabs and to provide confinement in columns. The advantages of this technology include speed and ease of installation, durability of the material system, light weight, and performance. The upgrade approach used for the structures presented in this paper is available in a technical document published by the American Concrete Institute (ACI) (ACI Committee 440, 2002), which provides guidance for the selection, design, and installation of FRP systems for external strengthening of concrete structures as an emerging technology. This document can be used to select an FRP system for increasing the strength and stiffness of RC and PC beams or the ductility of wrapped columns. Conditions are also identified where FRP strengthening is beneficial and where its use may be limited. 2 Design Principles According to ACI 440 (2002), it is recommended that the increase in load-carrying capacity of RC or PC members strengthened with an FRP system be limited. The philosophy is that a loss of FRP reinforcement should leave a member with sufficient capacity to resist at least 1.2 times the design dead load and 0.85 times the design live load. Design recommendations are based on limit-states-design principles. This approach sets acceptable levels of safety against the occurrence of both serviceability and ultimate limit states (e.g., deflections, cracking, stress rupture, and fatigue). In determining the ultimate strength of a member, all possible failure modes and resulting strains and stresses in each material should be assessed. For evaluating the serviceability of an element, engineering principles, such as modular ratios and transformed sections, can be used. In addition to conventional strength-reduction factors required by code for conventional materials, other reduction factors applied to the contribution of the FRP reinforcement are recommended to reflect the limited body of knowledge of FRP systems compared with steel RC and PC. For the design of FRP systems for the seismic retrofit of a structure, it may be appropriate to use established capacity design principles, which assume a structure should develop its full elastic capacity and require that members be capable of resisting the associated shear demands. The guaranteed tensile strength of FRP is defined as the mean tensile strength of a sample of test specimens minus three times the standard deviation. The design tensile strength that should be used in all design equations is given by the guaranteed value multiplied by a knock down factor to account for the service environment and is dependent on fiber type and exposure conditions of the structure. The design rupture strain should be determined similarly, whereas the design modulus of elasticity is the same as the value reported by the manufacturer.

3 For both flexural and shear strengthening, it is recognized that the FRP cannot attain its design rupture strain as a result of potential debonding. For this, the strain level in the FRP reinforcement at the ultimate-limit state needs to be determined and limited to an upper value to prevent debonding or delamination. This term recognizes that laminates with greater stiffness are more prone to delamination. To avoid plastic deformations, the existing steel reinforcement should be prevented from yielding at service load levels. The stress in the steel at service should be limited to 80% of the yield stress. Similarly, to avoid failure of an FRP-reinforced member due to creep rupture of the FRP, stress limits for these conditions should be imposed on the FRP reinforcement. In the case of shear strengthening, the effective strain cannot be larger than 0.4% to avoid the loss of aggregate interlock of the section before FRP delamination or rupture. Several of the unresolved issues remain the objective of on-going research, for example: durability (including fire) and test methods. Further research into the mechanics of bond of FRP reinforcement is of critical importance. New experimental evidence and analytical tools should yield more accurate methods for predicting delamination at the interface and in the concrete. Further developments will likely account for the stiffness of the laminate, the stiffness of the member to which the laminate is bonded, and the influence of the adhesive thickness and properties. The interim recommendations to limit the strain in the FRP to prevent delamination need to be revisited and confirmed. In addition to the above, efforts need be made to allow for other forms of FRP strengthening. Although not directly addressed by ACI 440 (2002), the use of near-surface mounted (NSM) FRP bars is a promising technology for increasing flexural and shear strength of deficient RC and PC members (Parretti and Nanni, 2004). The advantages of NSM FRP bars compared with external FRP laminates are the possibility of anchoring the reinforcement into adjacent RC members, and minimal surface preparation work and installation time. For installation, a groove is cut in the desired direction into the concrete surface, the groove is then filled half-way with adhesive paste, and the FRP bar is placed in the groove and lightly pressed. This forces the paste to flow around the bar and fill completely between the bar and the sides of the groove. Finally, the groove is filled with more paste and the surface is levelled. As this technology emerges, the structural behavior of RC and PC elements strengthened with NSM FRP bars needs to be fully characterized. 3 Examples of Bridge Superstructure Applications The selection of the case studies reported in this paper is intentionally limited to bridge structures where a deficient superstructure was upgraded in shear or flexure. All projects were undertaken in the State of Missouri for system bridges (i.e., bridges owned by the Department of Transportation (MoDOT) or offsystem bridges ((i.e., bridges owned by Counties or Municipalities) made of concrete. The selection includes a variety of bridge superstructures (i.e., precast RC channels, PC birders, cast-inplace RC slabs, and cast-in-place RC deck-girder systems) strengthened with different types of composites (i.e., manual lay-up FRP laminates, pre-cured FRP laminates, NSM FRP bars, mechanically fastened FRP laminates, and steel reinforced polymer (SRP) laminates) to correct various deficiencies (i.e., increased flexural and shear demand, construction errors, and vehicular impact). 3.1 Strengthening of Precast Channels (Schiebel et al. 2002) Three bridges in Boone County, Missouri were selected for strengthening with CFRP laminates in both shear and flexure to remove the 15 tons (13.6 metric tons) load posting that had been imposed on each of the bridges. In order to verify the results of the upgrade, load tests were performed before and after strengthening on two of the bridges. The three bridges were all simply supported and composed of precast RC channel sections. The bridge spans varied from ft (5.92 m) to ft (11.84 m) and were composed of seven or eight channel

4 sections (see Figure 2). The existing capacity of the bridges was determined using basic principals of RC theory and the existing factored moment and shear capacities (φm n and φv n ) for a single RC channel section in each bridge were computed. Visual inspection of the bridges revealed no significant deterioration of the concrete channels apart from some hairline cracks. Design loads were based on a standard truck load as given by current design guidelines (AASHTO 1996) for precast multi-girder bridge decks. The imposed moment and shear demand (M u and V u ) and the corresponding deficiencies were computed and from this, the level of required strengthening ( M and V) was determined Shear Strengthening (exterior channel) 24 - in. wide Spaced at 34 - in c/c, 10 - in clear edge/edge Sheets U - wrapped on web. Sheets are terminated in from channel interface. 6 Flexural Strengthening 4-in. wide Run the entire span Centered on 6-in web bottom. Shear Strengthening (interior channel) 24-in. wide Spaced at 34-in c/c, 10 - in clear edge/edge Sheets attached to web side and continuing on web bottom. Sheets terminated 0.5-in from channel interface ½ Figure 2 : Strengthening Configuration for Coats Lane Bridge (Note: 1 in=25.4 mm; 1 ft= m) Flexural strengthening was attained by applying strips of CFRP laminate to the soffit of the channel legs with the fibers oriented in the longitudinal direction (see Figure 2 for a schematic view). Strengthening the bridges for shear was achieved by applying strips of CFRP laminate (with fiber perpendicular to the direction of the bridge) to the sides of the web on each channel leg. Due to the way sections are joined side-by-side, the CFRP strips could only be applied to one side of each channel leg. Each strip would extend from the channel web across the soffit of the leg, covering the longitudinal FRP strip (see Figure 2). Figure 3 shows details of the completed installation for one of the bridges. The evaluation included review of the bridge construction drawings, visual inspection, and use of established state and federal guidelines (AASHTO 1994). According to MoDOT s current load rating guidelines (MoDOT 1996), any bridge built, rehabilitated, or re-evaluated is to be rated using the Load Factor Method rating at two load levels, the maximum load level (Operating Rating) and a lower load level (Inventory Rating). The Operating Rating is the maximum permissible load that should be allowed on the bridge. Exceeding this level could damage the bridge. The Inventory Rating is the load level the bridge can carry on a daily basis without damage and is taken as 60% of the Operating Rating. From the results of the analysis, the strengthening effectively increased the Rating Factor for all bridges in both

5 flexure and shear to a value greater than 1. This result allowed for the recommendation to remove the load posting of the three bridges. A load-testing program was carried out on two of the three bridges before and after strengthening to verify performance under service load conditions. Under known loads, the flexural response of the bridges was measured in terms of steel and CFRP reinforcement strains, concrete strains, and member deflections. The recorded deflections for the pre-strengthened test compare well with theoretical results. Branson s classical procedure (Branson 1977) was used in which the effective moment of inertia, I e, was calculated for a channel section based on an applied moment as given by the AASHTO wheel load distribution factor. The mid-span deflection was then calculated based on this effective moment of inertia. From a comparison of the moment-deflection results from both before and after strengthening, the deflection for any applied load decreased after the strengthening was applied. This reflected the increased stiffness of the member due to the addition of the FRP laminates. Based on the results of the analytical calculations of the structural components as applied to the load rating equation and the validation by load testing, the recommendation to remove the load posting could be substantiated. Figure 3 : Completed Installation 3.2 Strengthening of Impacted PC Girder (Parretti et al. 2003) The objective of this project was to restore the original ultimate flexural capacity of an accidentally impact-damaged PC girder of a bridge over the Gasconade River, Missouri. Two prestressing tendons in the central girder of the North span of the bridge were fractured due to the impact (Figure 4). Similar projects involving the strengthening of damaged bridge girders have been conducted (Nanni 1997), including two in the State of Missouri (Nanni et al. 2001; Schiebel et al. 2001). In these projects as well as the one described here, CFRP laminates were used for installation by manual lay-up (Figure 4) to restore the original ultimate capacity of the impacted girder. The damaged girder was prestressed by 38 low-relaxation steel strands with a tensile strength of 270 ksi (1862 MPa). It was assumed that a portion of the bridge deck with dimensions of 8 x 106 in (20 x 270 cm) provided composite action with the girder. The design of the concrete cross-section of the girder was carried out according to ACI 440.2R-02. In this instance, the factored moment capacity of the member after strengthening was imposed to be larger or equal to the value of the original member capacity having a composite (PC plus deck) cross-section. The rehabilitation of this impact damaged girder called for concrete repair and application of CFRP laminates as shown in Figure 5. The flexural strengthening consisted of three 24 in (60 cm) wide plies

6 with lengths of 10 ft, 11 ft, and 12 ft (3.00, 3.35, and 3.65 m), respectively, applied to the bottom of the girder with fibers aligned along its longitudinal axis. The triple-ply laminate was centered over the damaged area. Ten strips, 8 in (20 cm) wide and spaced at 16 in (40 cm) on centers, were then U- wrapped around the bulb of the girder over the previous installation (see Figure 5). The purpose of the U- wrap was to prevent delamination of the FRP plies applied to the bottom surface of the girder. Figure 4 : Severed Tendons and CFRP Repair Damaged Area 8" 8" 8" 8" Damaged Area 10' 11' 12' (a) Elevation View 1st CFRP Ply 2nd CFRP Ply 3rd CFRP Ply 2' 12' 3 CFRP Plies (a) Elevation View 10 U-Wrapped CFRP Strips 8" ocs (b) Plan View (b) Cross-Section Figure 5 : CFRP Flexural Strengthening (1 in=2.54cm) FRP reinforcement was no longer needed at the cross section where the damaged strands became fully effective again. For the damaged strands, it is reasonable to assume a linear transition between the point of zero stress and full prestress transfer over their development length. The development length for the strand was calculated equal to 5.5 ft (1.70 m). The length of the first and longest FRP ply was then set equal to 12 ft (3.65 m), which corresponded to 11 ft (3.35 m) necessary to account for the tendons development length plus 1 ft (0.3 m), corresponding to 6 in (15 cm) at each end, deemed necessary for the FRP development length as suggested by ACI 440.2R-02. The remaining two plies were set to have

7 lengths of 11 and 10 ft (3.35 and 3.0 m). A two-man crew with a cherry-picker was able to complete the repair over a period of three half-days. The alternative would have been an onerous, expensive, and timeconsuming solution. 3.3 Strengthening of Cast-in-place RC Slab (Casadei et al. 2003) The bridge selected for this project is located in Phelps County, Missouri that was commissioned in 1926 and was originally on a gravel road. Load posting of this bridge to 13 tons (11.79 metric tons) 15 mph (24.14 km/hr) was approved around 1985 and had a significant impact on the local economy. This bridge is a three-span simply supported RC slab. The total bridge length is 66 ft (20.12 m) and the total width of the deck is 22.5 ft (6.86 m). Based on visual and Non Destructive Testing (NDT) evaluation, it was determined that: the superstructure is 14 in (35.56 cm) thick, running from pier to pier, the longitudinal reinforcement is made of #8 ( 25.4 mm) bars spaced at 5 in (12.7 cm) on centers, and no transverse reinforcing is present. From cores (cylinders 3 in 6 in, 7.62 cm cm), the average compressive strength of the concrete was measured to be 4100 psi (28.27 MPa); the yield of the steel was also tested on one bar sample, and resulted to be 32 ksi ( MPa). The lack of transversal reinforcement and the presence of very rigid parapets caused the slab to crack along the mid-line in all spans. The cracks were approximately 1 in (2.54 cm) wide (see Figure 6). There was no significant cracking in any other portion of the slab and only minor corrosion of the bars crossing the crack. Figure 6 : Soffit Longitudinal Crack and Epoxy Injection Based on conventional RC theory, the computed nominal flexural and shear capacities φm n and φv n were larger than M u and V u respectively. No flexural and shear strengthening was therefore required in the longitudinal direction. The strengthening design had the purpose of giving the bridge a moment capacity in the transversal direction equal to or greater than the cracking moment computed, in order to avoid further crack openings and deterioration of the concrete due to water percolation through the cracks. Two different FRP strengthening techniques were adopted: (1) externally bonded CFRP laminates installed by manual wet lay-up, and (2) NSM CFRP rods embedded in pre-made grooves and bonded in place with an epoxy-based paste. Before surface preparation for FRP application, the central crack was repaired in order to re-establish material continuity and assure no water percolation through the crack. The crack was first sealed using an epoxy-paste and then injected with a very low viscosity resin as shown in Figure 6. The design for externally bonded laminates called for a total of six, 12 in (30.48 cm) wide, single ply CFRP strips overlapping at center span for 10 ft (3.05 m). The strips were evenly spaced over the width of 20 ft (6.09 m) and ran the entire width of the slab (Figures 7 and 8). The required number of NSM bars was determined to be two CRFP tapes per slot on a 9 in (22.86 cm) groove spacing. The bars were

8 embedded in 17 ft (5.18 m) long, ¾ in (19.05 mm) deep, and ¼ in (6.35 mm) wide grooves cut onto the soffit of the bridge deck (Figures 7 and 8). 17'-0" 4'-3" 22'-0" 20'-0" 2'-0" 2'-0" 1'-0" 1'-6"1'-6" 1'-0" 1'-0" 1'-6" 2'-0" 1'-0" 1'-0" 1'-0" 1'-6" 2'-0" 0.75" 0.25" Groove dimensions Groove partially filled with epoxy adhesive Tapes insertion and groove completely covered with epoxy Note: Before installation place tapes together after buttering surfaces 4'-3" a) Plan View b) Section View Figure 7 : Strengthening with NSM Bars and CFRP Laminate on Span 2 (1 in= 2.54 cm) a) Primer application b) Completed installation c) Insertion of NSM Bar d) Application Completed Figure 8 : Phases of Externally Bonded and NSM Reinforcement Application In order to clarify the behavior of the structure, load tests were performed and a finite element method (FEM) analysis undertaken. It was proven that the FRP strengthening designed to avoid further cracking

9 and such that the transverse flexural capacity be higher than the cracking moment, was attained. Both FRP techniques were easily implemented and showed satisfactory performance. 3.4 Five-Bridge Program (Nanni and Lopez 2004) A five-bridge strengthening program, using five variations of composite strengthening techniques, was undertaken to offer a comprehensive case study, addressing several aspects of installation and long-term monitoring. The project is intended to validate the use of FRP materials, and CFRP in particular, to strengthen structurally-deficient concrete bridges. As part of a five-year long monitoring program, NDT research was concentrated on one structure, namely bridge P-0962 in Dallas County, Missouri. The five bridges are located over three Missouri Department of Transportation (MoDOT) Districts. Four structures, including P-0962, are of the T-beam type and one is of the solid slab type. Of the four T-beam bridges, one has a four-girder supported deck, while the others, including P-0962, have a three-girder supported deck. In this group, P-0962 is the most recent structure in terms of age (1956 construction). Four FRP system technologies and a steel fiber reinforced polymer (SRP) system were used for the flexural strengthening of the five structures. Shear strengthening, when necessary, was always in the form of externally bonded laminates installed by manual lay-up. The four FRP technologies were as follows: Externally bonded FRP laminates installed by manual wet lay-up Adhered pre-cured FRP laminates Near surface mounted FRP bars Mechanically fastened FRP laminates This investigation covered aspects such as NDT measurement of concrete surface roughness, evaluation of bond properties, investigation of fiber alignment, and detection of delaminations and concrete cracking. For this purpose, some strengthening was conducted with intentional defects at non-critical locations. Additionally, in-situ load testing prior to and after the strengthening was implemented on each of the five bridges. The intention of these load tests is to determine possible degradation of stiffness over time. The overall performance of a strengthened bridge superstructure is evaluated by comparing the behavior before and immediately after the strengthening to that over a reasonably long period of time. The information and data collected during all phases of the upgrade, that is: initial assessment, design, construction, inspection and monitoring are to be used for the development of specifications and guidelines written in technical language for future strengthening projects. 4 Conclusions Even with some unresolved issues that should be addressed in future research, it can be concluded that the availability of design and construction guides developed by ACI for the use of FRP external reinforcement for the upgrade of existing structures should allow the construction industry in North America and in the rest of the world to take full advantage of this emerging technology. Applications for strengthening of existing structures with externally bonded laminates and NSM bars have already captured a significant market share. Projects in the field of transportation infrastructure are becoming numerous and the State of Missouri has taken a leadership role in this endeavor. 5 Acknowledgements The author would like to thank all who have participated in the selected projects and UMR students in particular for their invaluable contributions in all phases of the projects. The UMR based University

10 Transportation Center together with the NSF Industry/University Cooperative Research Center have provided part of the funding. 6 References AASHTO, Manual for Condition Evaluation of Bridges, 2 nd Edition, pp , AASHTO, Standard Specifications for Highway Bridges, 16 th Edition, 760 pp., ACI 440.2R-02, Guide for the Design and Construction of Externally Bonded FRP Systems for Strengthening Concrete Structures, Reported by ACI Committee 440, American Concrete Institute, Farmington Hills, Alkhrdaji, T., Nanni, A., Chen, G., & Barker, M. "Upgrading the Transportation Infrastructure: Solid RC Decks Strengthened with FRP," Concrete International: Design and Construction, Vol. 21, No.10, Oct. 1999, pp Alkhrdaji, T., Nanni, A., Chen, G., & Barker, M., Destructive and Non-Destructive Testing of Bridge J857, Phelps County Missouri, Center for Infrastructure Engineering Studies, University of Missouri-Rolla, Rolla, Missouri, pp , Branson, D.E. Deformation of Concrete Structures, McGraw-Hill, New York, Casadei, P., N. Galati, R. Parretti & A. Nanni, Strengthening of a Bridge Using Two FRP Technologies, Field Application of FRP Reinforcement: Case Studies, ACI Convention, Boston, Sept. 27-Oct. 1, 2003, Rizkalla, S. & A. Nanni, Editors, ACI Special Publication No. 215, American Concrete Institute, Farmington Hills, MI, pp Missouri Department of Transportation, MoDOT Bridge Load Rating Manual, Jefferson City, Missouri, pp , Nanni, A., Carbon FRP Strengthening: New Technology Becomes Mainstream, Concrete International: Design and Construction, Vol. 19, No. 6, June, pp , Nanni, A., Composites: Coming on Strong, Concrete Construction, Jan. 1999, Vol. 44, No. 1, pp Nanni, A., Huang, P.C., & Tumialan, G., "Strengthening of Impact-Damaged Bridge Girder Using FRP Laminates," 9th Int. Conf., Structural Faults and Repair, London, UK, July 4-6, M.C. Forde, Ed., Engineering Technics Press, CD-ROM version, 7 pp., Nanni, A. & A. Lopez Validation of FRP technology Through Field Testing 16 th World Conference on Nondestructive Testing, Montreal, Canada, Aug. 30- Sept. 3, 2004 (in print). Parretti, R. & A. Nanni, Strengthening of RC Members Using Near-Surface Mounted FRP Composites: Design Overview, Advances in Structural Engineering An International Journal, 2004 (in print). Parretti, R., A. Nanni, J. Cox, C. Jones, & R. Mayo, Flexural Strengthening of Impacted PC Girder with FRP Composites, Field Application of FRP Reinforcement: Case Studies, ACI Convention, Boston, Sept. 27-Oct. 1, 2003, Rizkalla, S. & A. Nanni, Editors, ACI Special Publication No. 215, American Concrete Institute, Farmington Hills, MI, pp Schiebel, S., Parretti, R., & Nanni, A., (2001): "Repair and Strengthening of Impacted PC Girders on Bridge A4845," Final Report RDT01-017/RI01-016, Nov. 2001, Missouri DOT, Jefferson City, Missouri, 21 pp. Schiebel, S., R. Parretti, A. Nanni, & M. Huck, Strengthening and Load Testing of Three Bridges in Boone County, Missouri, ASCE Practice Periodical on Structural Design and Construction, Nov. 2002, Vol. 7, No. 4, pp