<%BANNER%>

Durability of Carbon Fiber Reinforced Polymer (CFRP) Strengthening Systems Used to Repair Corrosion Damage in Reinforced...


PAGE 1

DURABILITY OF CARBON FIBER REINFORCED POLYMER (CFRP) STRENGTHENING SYSTEMS USED TO REPAIR CORROSION DAMAGE IN REINFORCED CONCRETE By MARKUS PETER KUTARBA A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ENGINEERING UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Markus Peter Kutarba

PAGE 3

ACKNOWLEDGMENTS Completion of my thesis and the research associated with it would not have been successful without the help and guidance of a number of individuals. I thank the chair and members of my committee (Dr. H.R. Hamilton, Dr. John Lybas, and Dr. Andrew J. Boyd) for their guidance and assistance. The Department of Civil and Coastal Engineering provided financial support throughout my graduate studies. The National Science Foundation (NSF) gave financial support to my study. The Fyfe Cooperation donated all materials. Sarah Witt is gratefully acknowledged for arranging this contribution. Others deserving thanks for contributions of resources and knowledge are John Levar, Jeff Brown, and Chuck Broward; and Mario Paredes Cruz and Mitch Langley (Florida Department of Transportation, Materials Office). Lastly, I thank all of my close friends and family. Their enduring support proved to be of the utmost value. iii

PAGE 4

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF TABLES.............................................................................................................vi LIST OF FIGURES..........................................................................................................vii ABSTRACT.......................................................................................................................ix CHAPTER 1 INTRODUCTION........................................................................................................1 Background...................................................................................................................1 Objective of Study........................................................................................................2 2 LITERATURE REVIEW.............................................................................................4 3 EXPERIMENTAL PROGRAM...................................................................................6 Introduction...................................................................................................................6 Test Specimen Description.....................................................................................6 Initial Corrosion......................................................................................................6 Concrete Repair......................................................................................................7 Concrete strengthening...........................................................................................8 Post-Repair Accelerated Corrosion..............................................................................9 Load Testing.................................................................................................................9 4 TEST RESULTS AND DISCUSSION......................................................................15 Crack Pattern due to Corrosion...................................................................................15 Repaired Only Specimens.....................................................................................16 Specimens Strengthened by Scheme I and II.......................................................16 Specimens Strengthened by Scheme III...............................................................16 Corrosion Damage Rating..........................................................................................17 Category I.............................................................................................................18 Category II............................................................................................................18 Category III...........................................................................................................18 Category IV..........................................................................................................18 iv

PAGE 5

Structural Behavior.....................................................................................................18 Control Beams......................................................................................................19 Repaired Beams....................................................................................................20 Scheme I Specimens.............................................................................................20 Scheme II Specimens............................................................................................21 Scheme III Specimens..........................................................................................22 5 CONCLUSIONS........................................................................................................37 LIST OF REFERENCES...................................................................................................39 BIOGRAPHICAL SKETCH.............................................................................................41 v

PAGE 6

LIST OF TABLES Table page 1 Material properties for strengthening system........................................................13 2 Test specimen description.....................................................................................14 3 Half-cell readings from accelerated corrosion......................................................35 4 Results load test after initial corrosion..................................................................35 5 Results load test after PR corrosion......................................................................36 vi

PAGE 7

LIST OF FIGURES Figure page 1 Application of FRP on bridge Girder (Chaffee Rd, Jacksonville)..........................3 2 Potential problems with repaired and strengthened reinforced concrete................3 3 Specimen construction details...............................................................................10 4 Ponding area on beam...........................................................................................10 5 Beam 7 B-S...........................................................................................................11 6 Beam repair details................................................................................................12 7 Repair schemes......................................................................................................12 8 Plot of half-cell readings vs. strengthening scheme..............................................13 9 Typical crack pattern after initial accelerated corrosion process..........................24 10 Corrosion induced cracking on beam 6 A-D.........................................................24 11 Crack pattern on beam 3 A-S................................................................................24 12 Debonded areas on beam 3 B-D............................................................................25 13 Corrosion on beam 4 A-S......................................................................................25 14 Pictures taken of category I corrosion-damage.....................................................26 15 Pictures taken of category II corrosion-damage..................................................276 16 Pictures taken of category III corrosion-damage on control beam.......................27 17 Pictures taken of category IV corrosion-damage on repaired beam.....................28 18 Half-cell readings on beams strengthened by scheme III......................................29 19 Ring anode corrosion at interface between old concrete and repair material.......29 20 Load-deflection diagram for beam C 2.................................................................30 vii

PAGE 8

21 Load-deflection diagram for beams C 2 and 2 B-S...............................................30 22 Load-deflection diagram for beams C 2, 1 A-S, and 4 A-D.................................30 23 Load-deflection diagram for beams C 1 and 2 A-D..............................................31 24 Load-deflection diagram for beams C 2, 5 A-D, and 7 B-S..................................31 25 Load-deflection diagram for beams C 2 and 3 B-S...............................................31 26 Load-deflection diagram for beams C 2 and 7 A-S..............................................32 27 Load-deflection plot for beams C2, 5 B-S and 4 B-D...........................................32 28 Debonded area on beam 5 B-S..............................................................................32 29 Failure mode for beam 4 B-D at mid-span, view from the north side..................33 30 Load-deflection diagram for beams C 2 and 4 B-S...............................................33 31 Load-deflection diagram for beams 4 A-S, 5 B-D, and C2...................................33 32 Load-deflection diagram for beams C2 and 3 B-D...............................................34 33 IR pictures of beam 3 B-D....................................................................................34 viii

PAGE 9

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Engineering DURABILITY OF CARBON FIBER REINFORCED POLYMER (CFRP) STRENGTHENING SYSTEMS USED TO REPAIR CORROSION DAMAGE IN REINFORCED CONCRETE By Markus Peter Kutarba May 2004 Chair: H.R. Hamilton Major Department: Civil and Coastal Engineering We studied the use of carbon fiber reinforced polymer (CFRP) systems combined with concrete repair as a rehabilitation technique for corroded reinforced concrete beams. Thirty reinforced concrete test specimens were constructed. Twenty-six of them were subjected to an aggressive environment for 18 weeks. After being corroded, the beams were repaired and strengthened, with CFRP using three different schemes. A group of eight beams was load tested after the repair and strengthening process, while the remaining beams were subjected to an aggressive environment for another 22 weeks. After the post-repair (PR) corrosion, a second group of eight beams was load tested. The results were compared to the first group regarding load capacity, stiffness, and degree of corrosion damage. It was found that the application of CFRP significantly decreases corrosion activity. The load capacity of PR corroded beams strengthened with CFRP was about 30% higher than for control beams, and the stiffness was restored to the original ix

PAGE 10

value. Based on the results and observations, a corrosion damage rating system for reinforced concrete beams was developed. x

PAGE 11

CHAPTER 1 INTRODUCTION Background The United States transportation infrastructure continues to age and deteriorate. Approximately 15% of the nations 583,000 bridges are structurally deficient because of corrosion of steel sections and steel reinforcement. The annual direct cost for corrosion damage on bridges is estimated at $8.3 billion (Virmani 2002). Until recently, the most common repair and strengthening methods were replacement of the damaged element, steel plate bonding, or installation of additional supports. The introduction of fiber reinforced polymer (FRP) composite systems into construction is a promising alternative for the rehabilitation of weakened and deficient concrete members. The advantages of FRP applications over traditional methods include a high strength-to-weight ratio, ease of material handling and installation, and resistance to attack in aggressive environments. Examples of current repair methods include bonded systems in which a plate, saturated fabric, or spray-up system is applied to the surface of the concrete element. The cured FRP composite provides additional shear reinforcement, flexural reinforcement, or both to existing beam and slab elements. shows FRP applied to an existing bridge structure. The bridge girder was strengthened for flexure and shear using a wet lay-up system. Figure 1 A large number of experimental studies have demonstrated the effects of moisture, aggressive environments, fire, UV-light, creep, and fatigue on externally bonded FRP systems (Karbhari 1996). Results of these studies indicated that bond behavior between 1

PAGE 12

2 FRP and concrete is the most critical component. However, few studies have investigated the long-term performance of FRP systems used as a strengthening material for corrosion-damaged concrete. Figure 2 shows a possible repair and strengthening scenario on a concrete member. Before applying the strengthening system, the damaged steel and concrete must be properly repaired. If concrete was patched under the FRP, then the bond between existing concrete and patch material must be studied. Cracks due to shrinkage or a faulty repair along the interface can cause further corrosion and possibly result in a bond failure between repair material and existing concrete. If the FRP fully covers the repair, then the usual outward signs of corrosion (such as cracking, staining, and spalling) will not be visible. Visual inspection is currently the primary form of bridge and building evaluation and monitoring. If corrosion of the steel reinforcement continues after repair, the FRP composite may prevent early detection and intervention. Consequently, it is important that the effect of post-repair corrosion on the structural behavior be better understood. In addition, it is important that the outward signs of continued degradation of the beams be characterized so that adequate evaluations of the post-repair status can be made. Objective of Study My objective was to study the long-term behavior of repaired reinforced concrete that has been strengthened with FRP. The durability of the composite beam was evaluated by examining the bond between the FRP strengthening system and the concrete; and also the bond between patch material and concrete. We also developed a damage-rating system, based on visual inspection of surface and reinforcement damage. The first step was to corrode the reinforcing steel in the beams. The second step was to repair the beams, and apply an FRP strengthening system. The beams were then

PAGE 13

3 exposed to another corrosion cycle. After a sufficient amount of corrosion has taken place in the repaired beams, the effects of the corrosion on the repair and strengthening system can be determined. Results were then used to create the corrosion-damage rating system. Figure 1. Application of FRP on bridge girder (Chaffee Rd, Jacksonville). original concrete patch debonding FRP s ystem accumulation of rust original steel patch material cracking due to corrosion FRP debonding Figure 2. Potential problems with repaired and strengthened reinforced concrete.

PAGE 14

CHAPTER 2 LITERATURE REVIEW The effect of corrosion on reinforced concrete is well documented. It is known that concrete provides a high alkaline environment that forms a passive film on the steel surface, which prevents steel from corroding. Chlorides disrupt this protective layer, and steel starts to corrode. Expansive forces, which are a result of corrosion, will crack and spall the surrounding concrete. The consequence is less-protected steel causing additional corrosion, and a greater loss in steel cross-section (Jones 1996, ACI 1996). Depending on the degree of damage, a concrete repair in conjunction with strengthening method (such as bonded FRP composites) can improve the strength and serviceability (ACI 2001). FRP composites can be designed to be resistant to the aggressive environments that attack reinforced concrete, which makes them an attractive material for use in repair and strengthening of corrosion-damaged structures. While the behavior of the FRP composite in aggressive environments is well understood, researchers have only recently attempted to characterize the performance of FRP composites bonded to corrosion damaged-reinforced concrete. Testing has shown that corrosion-damaged beams strengthened with an FRP system are stiffer, and have a higher load-carrying capacity than unstrengthened beams (Bonacci and Maalej 2000, Soudki et al. 2003). It has also been shown that FRP reduces corrosion rates of reinforcing steel in concrete by minimizing the diffusion of chlorides (Berver et al. 2001, Soudki, et al. 1999). However, very little information is available on the durability of repaired corrosion-damaged 4

PAGE 15

5 structures. Questions remain concerning the synergistic effects of continued corrosion on repaired system performance. This thesis highlights the research findings of a multiphase experimental study aimed to evaluate the durability and structural behavior of corrosion-damaged beams repaired and strengthened with FRP.

PAGE 16

CHAPTER 3 EXPERIMENTAL PROGRAM Introduction Thirty test specimens were fabricated to investigate the long-term behavior of repaired reinforced concrete beams strengthened with FRP. First, the beams were subjected to a corrosion process for initial damage. They were then repaired and strengthened, and exposed to a second corrosion process. At this point, 16 beams were tested in four-point bending. Eight beams continued to be exposed to an accelerated corrosion process, and the six remaining beams were reserved for control specimens. Test Specimen Description Each rectangular beam was 203 mm x 305 mm (8 in x 12 in) in cross-section and 2900 mm (9 ft-6 in) long. The internal longitudinal steel reinforcement consisted of two 19 mm (#6), 414 MPa (60 ksi) bars at the bottom, and two 9.5 mm (#3), 276 MPa (40 ksi) bars at the top of the beam. The reinforcing steel was extended 76 mm (3 in) beyond the end of the concrete, for the purpose of making external electrical connections. The clear concrete cover was 51 mm (2 in) on all aspects of the specimen. Three different batches of concrete mixtures were used to construct the beams. The average compression strength ranged between 29.6 and 37.9 MPa (4300 to 5500 psi). Figure 3 shows a typical beam section. Initial Corrosion Twenty-six of the thirty beams were subjected to accelerated corrosion. Before starting the initial accelerated corrosion, the beams were subjected to center-point loading 6

PAGE 17

7 necessary to cause a minimum crack width of 0.33 mm (0.013 in). Next, ponding areas were constructed around the cracked regions, and a salt solution (5% NaCl by weight) was applied. The ponding was cycled at 1 week wet and 1 week dry. A 5-volt constant potential was applied, to accelerate the corrosion process. Half-cell potential readings were taken at the beginning and end of each cycle, to determine the corrosion activity. The corrosion process lasted 18 weeks (9 wet and dry cycles), and resulted in severe corrosion damage to the beams. Figure 4 shows a beam in the initial corrosion process, with a single ponding area. Concrete Repair ACI repair procedures (ACI 2001) were followed in repairing the corrosion damage. All damaged material was carefully separated from sound concrete without causing further damage. Undamaged concrete behind the reinforcing steel was removed to create a clear space of 25.4 mm (1 in). At the perimeter of the removal area 25.4 mm (1 in) deep saw cuts were made. This prevented the formation of feathered edges and provided adequate spacing for repair material. After the concrete was removed, all corroded areas on the reinforcement were sandblasted to white metal. The repair and cleaning process is shown in shows the repair layout. Figure 5 Figure 6 To restore the beams original shape after the corrosion repair a concrete mixture with a 46.2 MPa (6700 psi) 28 day-strength was used. The maximum aggregate size was limited to 10 mm (0.375 in). In addition, a super plasticizer was mixed with the concrete to reduce the water-cement ratio. The concrete performance evaluation report is shown in the Appendix.

PAGE 18

8 Concrete strengthening After preparing the concrete surface by a light sandblasting, three strengthening schemes were applied to the beams. The first scheme (scheme I) involved strengthening the specimen with separate FRP sheets for flexure and shear. A typical flexural sheet was 2743 mm (108 in) long and 203 mm (8 in) wide. The shear strips were 152 mm x 305 mm (6 in x 12 in). On each side eight sheets were used with a center-to-center spacing of 305 mm (12 in). At mid-span a 305 mm (12 in) wide field was left un-strengthened to allow for free development of corrosion cracks. Beams strengthening by the second scheme (scheme II) used the same FRP layout as scheme I except that the shear sheets were used as anchors for the flexural sheet. This sheet was fixed either at the ends only or over the entire length using all shear FRP sheets. The third group was strengthened with a single FRP sheet used as a full wrap (scheme III). This sheet was 2743 mm x 610 mm (108 in x 24 in) and covered the bottom of the specimens as well as 203 mm (8 in) of each side. The different strengthening configurations are shown in Figure 7. A unidirectional carbon fabric with glass veil backing for added fabric stability during installation was used for strengthening. The matrix, Tyfo S Epoxy, was a two component epoxy material. The composite material properties as given by the manufacturer are shown in Table 1. After the specimens were strengthened, eight beams were load tested to failure. These tests were done to study the beam behavior under loading conditions and to collect reference data for future testing.

PAGE 19

9 Post-Repair Accelerated Corrosion For the post-repair (PR) corrosion process 16 beams, 4 repaired only (R), 10 repaired and strengthened (RS) and 2 unrepaired and strengthened (URS) specimens were pre-cracked under a load of 44.48 kN (10 kip) which simulated service load conditions for the control beams. Crack widths could not be measured due to the FRP composite. The beams were then ponded with a 3.5% salt solution for one week under constant 5-volts followed by one week of drying. Half-cell readings were taken after each wet and dry cycle. After 22 weeks (11 wet and dry cycles) eight beams were taken from the PR corrosion process and load tested. The last eight beams remained in the corrosion process. Load Testing After each corrosion process, eight beams were tested in four-point loading to failure. All beams were simply supported; the distance between the load point and the end supports was set at 1219 mm (48 in) except for beams C 1 and 2 A-D. The distance between the load point and the end supports for beams C 1 and 2 A-D was 914 mm (36 in). This resulted in a constant moment zone at mid-span of 914 mm (36 in) for C 1 and 2 A-D, and 305 mm (12 in) for all other beams. The beams were tested in three or four stages of loading. Deflection was measured with two Linear Variable Displacement Transducers (LVDT) at mid-span. A hydraulic jack applied the load on the test specimens at a load rate of 222 N/sec (50 lbf/sec). The data were collected using an automated data acquisition system. Two sets of beams have been tested thus far. The first group consisted of control, repaired only (R), and repaired and strengthened (RS) beams. These tests were performed after the initial accelerated corrosion, repair and strengthening process.

PAGE 20

10 Beams from the second group were tested after PR accelerated corrosion. All beams except for one (3 B-D) have been repaired as described earlier. Beam 3 B-D was strengthened without a corrosion repair (URS) by scheme III. The test grouping can be seen in Table 2; Figure 8 shows the plots of half-cell readings vs. strengthening scheme. 305mm(12in)(8in)203mm(2)#3(2)#6 2900mm(114in)51mm203mm(2in)(2in)(8in)(2)#3(2)#6 51mm A B Figure 3. Specimen construction details. A) Cross section. B) Elevation. Figure 4. Ponding area on beam.

PAGE 21

11 A B C Figure 5. Beam 7 B-S. A) Splitting caused by corrosion. B) After removal of damaged concrete; saw cuts can be seen at the perimeter edges. C) Corroded reinforcement before sandblasting.

PAGE 22

12 (8in)(114in)(12in)(8in)(2in)(2in) #6#3 patch material, patch sizedepends on concrete crack 203mm305mmpatch material 2900mm51mm203mm51mm saw cut 203mm(8in) a Figure 6. Beam repair details. Scheme IScheme IIScheme III Figure 7. Repair schemes

PAGE 23

13 -700-600-500-400-300-200-1000mV (vs. Cu/CuSO4)controlscheme Ischeme IIscheme IIIafter 4 weeks A -700-600-500-400-300-200-1000mV (vs. Cu/CuSO4)controlscheme Ischeme IIIafter 22 weeksscheme II B Figure 8. Plot of half-cell readings vs. strengthening scheme. A) After 4 weeks. B) After 22 weeks. Table 1. Material properties for strengthening system Tensile strength MPa (ksi) Max. elongation Tensile modulus GPa (ksi) Thickness mm (in) Tg 876 (127) 1.2% 72.4 (10500) 1.0 (0.04) 82C (180F)

PAGE 24

14 Table 2. Test specimen description Beam Strengthening scheme Date of testing Repair information C 1 Control After initial corrosion Not corroded C 2 Control After initial corrosion Not corroded 1 A-S Control After re-corrosion Repaired 2 A-D Control After initial corrosion Repaired 2 B-S Control After initial corrosion Repaired 3 B-D Scheme III After re-corrosion Not repaired 3 B-S Scheme I After re-corrosion Repaired 4 A-D Control After re-corrosion Repaired 4 A-S Scheme III After re-corrosion Repaired 4 B-D Scheme II After re-corrosion Repaired 4 B-S Scheme III After initial corrosion Repaired 5 A-D Scheme I After initial corrosion Repaired 5 B-D Scheme III After re-corrosion Repaired 5 B-S Scheme II After re-corrosion Repaired 7 A-S Scheme II After initial corrosion Repaired 7 B-S Scheme I After initial corrosion Repaired Numbers indicate the set of test specimen, the letter A or B the position in the corrosion racks. The letters S or D describe the number of ponding areas (single or double) during the initial corrosion process.

PAGE 25

CHAPTER 4 TEST RESULTS AND DISCUSSION Crack Pattern due to Corrosion Both corrosion programs were aimed to cause concrete cracks due to corrosion of the main reinforcing steel. The initial corrosion process was stopped after 18 weeks. Horizontal cracks had formed at the surface along the level of the main reinforcement with rust leaching out. Crack lengths depended on the ponding area lay-out; cracks on beams with two ponding areas were an average 2337 mm (92 in) long and cracks on beams with a single ponding area were an average 1600 mm (63 in) long. Figure 9 shows a typical picture of a horizontal corrosion crack on beam after 18 weeks of wet and dry cycles. PR corrosion was conducted on the selected 16 beams which were described earlier. These beams were exposed for 22 weeks to wet and dry cycles before the test was stopped. The crack pattern and the level of damage on the PR corroded beams depended on the type of strengthening system; un-strengthened specimen, specimen strengthened by scheme I or scheme II, and specimens strengthened by scheme III. summarizes the accelerated corrosion process of all beams. Half-cell readings were made using a copper/copper sulfate electrode on the concrete surface. Although the current ASTM Standard (ASTM C 876) requires concrete contact for half-cell measurements, readings on beams strengthened by scheme III were made on FRP. Knowing that epoxy is not conductive, half-cell readings taken on the FRP laminate were expected to be zero. However, it was possible to get stable readings, which will be Table 3 15

PAGE 26

16 compared with the degree of corrosion to verify the measurement. Readings smaller than -350 mV indicate corrosion activity. Repaired Only Specimens First signs of corrosion were observed after four weeks of wet and dry cycles. Small cracks with rust leaching out were found along the level of the main reinforcing steel and at the interface between old concrete and patch material. The crack size and the amount of rust increased over the entire length of the test. By the end of the PR corrosion period all specimen had the same crack pattern; one major crack at the level of the reinforcement and a smaller secondary crack at the interface. Figure 10 shows an unstrengthened beam after the re-corrosion process. Cracks at the interface and the main reinforcement can be seen. Specimens Strengthened by Scheme I and II Beams strengthened by scheme I and II showed no visible signs of corrosion before the sixth week. After the sixth week small cracks without any rust traces were visible. By the end of the PR corrosion process, small cracks along the level of the main reinforcement and at the interface had formed on all beams with very little, or no rust leaching out. It was also observed that the shear FRP hindered corrosion cracks to grow in vertical direction but not in horizontal. However, these cracks were much smaller than the cracks on un-strengthened beams. Typical corrosion-damage for specimens strengthened by scheme I can be seen in Figure 11. Specimens Strengthened by Scheme III This group consisted of two different types of beams, RS beams (4 A-S and 5 B-D) and URS beams (2 B-D and 3 B-D). No visible signs of corrosion were spotted over the entire length of testing in both groups. However, half-cell readings verified corrosion

PAGE 27

17 activity on the un-repaired beams. A comparison of infrared (IR) pictures taken after curing and at the end of the corrosion process showed the development of large debonded areas on these beams. The debonded regions are a result of tension forces at the concrete surface caused by steel corrosion. The development of debonded regions can be seen in b denoted by light-colored regions. Figure 12 No debonded areas or other signs of corrosion were detected on RS beams during the entire PR corrosion. An investigation of the reinforcement after testing verified the results from the visual inspection, except for beam 4 A-S. On beam 4 A-S some corrosion was found at the interface between old and new concrete on the main reinforcing steel. This corrosion spot was caused by the so called "ring anode corrosion" or "halo effect". When repairs are completed on corrosion-damaged structures, the new concrete used to form the patch creates a corrosion potential with the existing concrete, which initiates new corrosion sites in the contaminated concrete near the patch interface. But since this spot was small and at the end of the beam it did not affect the structural behavior. shows this effect on beam 4 A-S. Figure 13 Corrosion Damage Rating After investigating the corrosion damage on the test specimens, each beam was assigned to one of four categories. Each category represents a specific level of damage based on visual observations of concrete cracking and reinforcement corrosion. Category I represents the lowest degree of corrosion damage, category III the highest. Category IV represents beams with a full FRP wrap on which half-cell measurements indicated corrosion activity.

PAGE 28

18 Category I Description: Very small horizontal cracks on the concrete surface with little or no traces of rust. No damage on the reinforcing steel visible. shows typical category I corrosion damage. Figure 14 Category II Description: Visible horizontal cracks with rust traces on concrete surface. Crack width is less than 1 mm (0.04 in). Very few and very small pitting corrosion spots on steel reinforcement. Figure 15 shows typical category II corrosion damage. Category III Description: Visible horizontal cracks with large rust stains on concrete surface. Crack width is less than 3 mm (0.12 in). Severe pitting corrosion and uniform surface corrosion damage on steel reinforcement. A loss of cross section area is visible. and show typical category III corrosion damage. Figure 16 Figure 17 Category IV Description: The typical outward signs of corrosion were not visible on beams strengthened by scheme III due to the FRP layout. However, half-cell measurements can be an indicator of corrosion activity. Figure 18 shows half-cell readings from two RS beam at mid-span and at the interface between old concrete and repair material. Even though the readings fluctuate over the period of 20 weeks, the measurements on beam 4 A-S (Figure 18a) at the interface tend to indicate corrosion activity. An investigation after the test verified ring anode corrosion (Figure 19). Structural Behavior After each corrosion process, a group of eight beams was tested in flexure to failure. Typically, all corroded R beams failed at lower loads than the control beam.

PAGE 29

19 RS and URS beams failed at higher loads. Post-cracking stiffness was always less than the control beams stiffness, except for beams strengthened by scheme III. Those beams stiffness were almost identical with the control beam stiffness. All beams failed in the form of concrete crushing except for beams C 1 and 2 A-D. At maximum load, RS and URS specimens showed a brittle failure mode in the form of concrete crushing. R beams failed in a more ductile mode. It can be seen on the load-deflection diagrams for R beams (, ) that the plot plateaus at approximately 89 kN (20 kip) indicate that the main reinforcing steel has yielded. This was confirmed by strain measurements on the reinforcement. RS and URS beams, however, do not exhibit as lengthy a plateau even though strain gauge data indicated that steel had yielded. Figure 21 Figure 22 Control beams C 1 and 2 A-D were tested in a different test set-up as described earlier. Both beams failed in shear and therefore they were not compared in this study. After testing C1 and 2 A-D the free length at mid-span was reduced to 305 mm (12 in). displays the results from tests after the initial corrosion and Table 5 displays the results from tests after PR corrosion. The post cracking slope (PC) was calculated using load and deflection values between 44.5 kN (10 kip) and 66.7 kN (15 kip) and was used as a tool to describe the stiffness. Table 4 Control Beams Beam C 2 was tested to be used as a control beam for all test specimens. This beam was not corroded. The maximum load capacity was 96 kN (21.6 kip) at a deflection of 20 mm (0.79 in). The reinforcement started to yield at approximately 62.3 kN (14 kip). Figure 20 shows the load versus mid-span deflection plot. Beam C 2 failed by concrete crushing.

PAGE 30

20 Repaired Beams Beam 2 B-S was tested to failure after the initial corrosion and repair process. The ultimate load capacity was 8% lower than the control beams capacity. The PC stiffness was reduced by about 20%. Figure 21 shows the load-deflection diagram for beam C 2 and 2 B-S. The load deflection plots for PR corroded specimens (1 A-S and 4 A-D) are shown in Figure 22. Both beams were tested after the initial corrosion, repair, and PR corrosion process. It can be seen that the load capacities and stiffness are less than that of the control beam. The ultimate capacities were 8% (4 A-D) and 15% (1 A-S) lower; the post-cracking stiffness was 29% (4 A-D) and 11% (1 A-S) lower. It is likely that the lower stiffness and load capacities were caused by the corrosion process. Typically, steel cross section area loss and bond failure between the reinforcement and concrete occur during corrosion. A reduced steel cross section area results in a lower load carrying capability, the loss of bond allows the reinforcing steel to yield without any restraint and results in a lower stiffness as well as load capacity. The concrete repairs performed well in the load tests with no significant cracking debonding, or spalling of the repair material noted. Beams C 1 and 2 A-D both failed in shear with diagonal cracks forming in the existing concrete and passing through the interface into the repair material. Figure 23 shows the load deflection diagram for both beams. It can be seen that beam 2 A-D has the same post corrosion loss in stiffness as those of the other specimen that had been through the accelerated corrosion. Scheme I Specimens The load-deflection response for strengthened beams (5 A-D and 7 B-S) tested after the initial corrosion process is shown in Both beams have been repaired Figure 24

PAGE 31

21 before FRP application and testing. The flexural FRP sheet on beam 5 A-D covered the tension face completely, while the sheet on beam 7 B-S was 102 mm (4 in) wide and covered only 50%. The increase in capacity over the control beam was 14% for beam 7 B-D and 30% for beam 5 A-D. The PC stiffness increased slightly (13%) for beam 5 A-D; for beam 7 B-S it did not change. Beam 3 B-D was tested un-repaired after the re-corrosion process. The ultimate capacity increased by 27%; the stiffness did not change within the limits. Figure 25 shows the load-deflection plot for the re-corroded beam compared to the control beam. During testing, prior to failure, signs of debonding were observed on the beams. Several loud cracking sounds indicated that epoxy was cracking at multiple locations. However, epoxy cracking was limited to very small areas and was not seen at all during IR inspections. Concrete crushed on beams strengthened by scheme I before FRP debonding became the failure mechanism. The change in post cracking stiffness was very small for beam 5 A-D and 7 B-S compared to the control beam. The same performance was observed on beam 3 B-S. Since the degree of corrosion damage was small too, this behavior was expected. Scheme II Specimens Figure 26 shows the load-deflection diagram for beam 7 A-S. The flexural sheet on this beam was anchored with U-wraps at multiple points leaving an un-strengthened length of 305 mm (12 in) at mid-span. Beam 7 A-S was load tested after the initial corrosion and a repair procedure. The load capacity was 47% higher than the capacity of the control beam; the post cracking stiffness increased only slightly.

PAGE 32

22 Beam 5 B-S was strengthened by the same FRP lay-out as beam 7 A-S. The load capacity after the re-corrosion process was 30% higher than the control beams capacity. The stiffness decreased at the same time by about 20% (see ). Figure 27Figure 27 The flexural FRP sheet on beam 4 B-D was anchored at each end by a single U-wrap. The load capacity increased by 31%, while the stiffness dropped by 17%. shows the load-deflection diagram for beams 5 B-S and 4 B-D compared to beam C 2. Cracking sounds were observed on all three beams during testing. At ultimate load, shortly after concrete crushing started, FRP debonded from the beams. On beams 7 A-S and 5 B-S debonding started at mid-span and was limited on the area between the first anchors (U-wraps). Figure 28 shows the debonded area at mid-span on beam 5 B-S. It can be seen that the U-wrap prevented the FRP sheet from debonding. The white colored spot indicate the debonded area; the dark colored spot left of it indicate the U-wrap. FRP on beam 4 B-D debonded at the east end and ripped out the concrete cover over the reinforcement at mid-span. No corrosion damage was spotted before testing in this area; however, an investigation after testing found some rust traces. shows beam 4 B-D after testing; concrete delaminated can be seen at the level of corrosion cracks. Figure 29 Scheme III Specimens Figure 30 shows the load-deflection diagram for beam 4 B-S tested after the initial corrosion compared to the control beam. It can be seen that the load capacity increased by 44% without any loss in stiffness. The PR corroded beams 4 A-S and 5 B-D showed an almost identical load deflection behavior during testing. The ultimate load capacities were 30% (4 A-S) and

PAGE 33

23 32% (5 B-D) higher than the control beams capacity. No loss in stiffness was observed. The load-deflection diagram for these beams can be seen in Figure 31 Throughout all tests several cracking sounds were heard. Like on the other strengthened beams, the source was epoxy cracking. However, no debonded areas were found at any point of testing. At ultimate load, all beams strengthened by scheme III failed suddenly by concrete crushing without any previous warning. Beam 3 B-D was strengthened by scheme III without a concrete repair after the initial corrosion (URS). The strengthening FRP sheet was used to cover all corrosion cracks. After the PR corrosion program several large debonded areas at multiple points on the specimen were found. Figure 32 shows the load-deflection diagram for beam 3 B-D compared to the control beam. The load capacity was 18% higher, while the stiffness did not change. Like the other beams strengthened by scheme III this beam failed suddenly by concrete crushing without any previous warning. IR scans of the tension face were taken during loading of a large debonded area caused by corrosion. At maximum load the beam failed by concrete crushing at mid-span. The failure mode indicated that the previously debonded area did not adversely affect the beam capacity. In addition, the two investigated debonded areas did not expand towards mid-span during loading, but rather grew towards each other forming one large debonded area (see Figure 33).

PAGE 34

24 crack at main reinforcement Figure 9. Typical crack pattern after initial accelerated corrosion process. Figure 10. Corrosion induced cracking on beam 6 A-D. The lower crack is at the patch interface, while the upper crack is at the level of the main reinforcing bar. crack at main reinforcement Figure 11. Crack pattern on beam 3 A-S. Crack formed along the level of the main steel reinforcement.

PAGE 35

25 A B Figure 12. Debonded areas on beam 3 B-D. IR picture was taken A) before the corrosion process and B) after the corrosion process. interface between old concrete and repair material Figure 13. Corrosion on beam 4 A-S.

PAGE 36

26 A B Figure 14. Pictures taken of category I corrosion-damage. A) Horizontal crack. B) Corresponding damage on reinforcing steel. A Figure 15. Pictures taken of category II corrosion-damage. A) Horizontal crack. B) Corresponding damage on reinforcing steel.

PAGE 37

27 B Figure 15. Continued. A B Figure 16. Pictures taken of category III corrosion-damage on control beam. A) Horizontal crack. B) Corresponding damage on reinforcing steel.

PAGE 38

28 A B Figure 17. Pictures taken of category IV corrosion-damage on repaired beam. A) Horizontal crack. B) Corresponding damage on reinforcing steel.

PAGE 39

29 readings on FRP-800-700-600-500-400-300-200-10000102030weekmV (vs. Cu/CuSO4) mid-span interface A readings on FRP-800-700-600-500-400-300-200-10000102030weekmV (vs.Cu/CuSO4) mid-span interface B Figure 18. Half-cell readings on beams strengthened by scheme III. A) Beam 4 A-S (ring anode corrosion at interface). B) Beam 5 B-S (no corrosion). Figure 19. Ring anode corrosion at interface between old concrete and repair material.

PAGE 40

30 0510152025303500.511.5Deflection [in]Load [kip]0204060801001201400102030Deflection [mm]Load [kN] Figure 20. Load-deflection diagram for beam C 2. 0510152025303500.511.5Deflection [in]Load [kip]0204060801001201400102030Deflection [mm]Load [kN] C 2 2 B-S Figure 21. Load-deflection diagram for beams C 2 and 2 B-S. 0510152025303500.511.5Deflection [in]Load [kip]0204060801001201400102030Deflection [mm]Load [kN] C2 4 A-D 1 A-S Figure 22. Load-deflection diagram for beams C 2, 1 A-S, and 4 A-D.

PAGE 41

31 0510152025303500.511.5Deflection [in]Load [kip]0204060801001201400102030Deflection [mm]Load [kN] C1 2 A-D Figure 23. Load-deflection diagram for beams C 1 and 2 A-D. 0510152025303500.511.5Deflection [in]Load [kip]0204060801001201400102030Deflection [mm]Load [kN] C 2 5 A-D 7 B-S Figure 24. Load-deflection diagram for beams C 2, 5 A-D, and 7 B-S. 0510152025303500.511.5Deflection [in]Load [kip]0204060801001201400102030Deflection [mm]Load [kN] C 2 3 B-S Figure 25. Load-deflection diagram for beams C 2 and 3 B-S.

PAGE 42

32 0510152025303500.511.5Deflection [in]Load [kip]0204060801001201400102030Deflection [mm]Load [kN] C 2 7 A-S Figure 26. Load-deflection diagram for beams C 2 and 7 A-S. 0510152025303500.511.5Deflection [in]Load [kip]0204060801001201400102030Deflection [mm]Load [kN] C 2 5 B-S 4 B-D Figure 27. Load-deflection diagram for beams C2, 5 B-S, and 4 B-D. Figure 28. Debonded area on beam 5 B-S.

PAGE 43

33 Figure 29. Failure mode for beam 4 B-D at mid-span; view from the north side. 0510152025303500.511.5Deflection [in]Load [kip]0204060801001201400102030Deflection [mm]Load [kN] C 2 4 B-S Figure 30. Load-deflection diagram for beams C 2 and 4 B-S. 0510152025303500.511.5Deflection [in]Load [kip]0204060801001201400102030Deflection [mm]Load [kN] C 2 4 A-S 5 B-D Figure 31. Load-deflection diagram for beams 4 A-S, 5 B-D, and C2.

PAGE 44

34 0510152025303500.511.5Deflection [in]Load [kip]0204060801001201400102030Deflection [mm]Load [kN] C 2 3 B-D Figure 32. Load-deflection diagram for beams C2 and 3 B-D. A B Figure 33. IR pictures of beam 3 B-D. A) Before load testing. B) After 15 kip.

PAGE 45

35 Table 3. Half-cell readings from accelerated corrosion Beam Type Half-cell reading after 4 th week Half-cell reading after 22 nd week Damage category 1 A-S control -376 mV (conc) -582 mV (conc) III 1 B-D control -382 mV (conc) -563 mV (conc) III 2 B-D scheme III -505 mV (FRP) -593 mV (FRP) IV 3 A-D scheme II -434 mV (conc) -494 mV (conc) I 3 A-S scheme I -309 mV (conc) -552 mV (conc) I 3 B-D scheme III -338 mV (FRP) -440 mV (FRP) IV 3 B-S scheme I -407 mV (conc) -461 mV (conc) I 4 A-D control -440 mV (conc) -545 mV (conc) III 4 A-S scheme III -80 mV (FRP) -80 mV (FRP) IV 4 B-D scheme II -368 mV (conc) -502 mV (conc) I 5 A-S scheme II -280 mV (conc) -312 mV (conc) I 5 B-D scheme III -357 mV (FRP) -63 mV (FRP) 0 5 B-S scheme II -130 mV (conc) -260 mV (conc) 0 6 A-D control -600 mV (conc) -504 mV (conc) III 6 A-S scheme I -496 mV (conc) -467 mV (conc) I 6 B-S scheme I -518 mV (conc) -514 mV (conc) I Readings made on concrete (conc) or FRP (FRP) Damage categories are explained in chapter 4 0 no signs of corrosion activity Table 4. Results load test after initial corrosion FRP configuration Beam Test result in kN (kip) PC slope in kip/in Capacity change to beam C 2 control* C 1 104.3 (23.4) 37 not compared control* 2 A-D 128 (28.8) 35 not compared control C 2 96.1 (21.6) 36 0% control 2 B-S 88.1 (19.8) 29 -8% scheme I 5 A-D 124.5 (28) 41 +30% scheme I (4) 7 B-S 111.2 (25) 36 +14% scheme II 7 A-S 141.4 (31.8) 42 +47% scheme III 4 B-S 137.9 (31) 37 +44% beam failed in shear

PAGE 46

36 Table 5. Results load test after PR corrosion FRP configuration Beam Test result in kN (kip) PC slope in kip/in Capacity change to beam C 2 control 1 A-S 81.8 (18.4) 31 -15% control 4 A-D 87.6 (19.7) 25 -8% scheme I 3 B-S 121.9 (27.4) 36 +27% scheme II 4 B-D 125.4 (28.2) 30 +31% scheme II 5 B-S 124.5 (28) 29 +30% scheme III 3 B-D 113.4 (25.5) 40 +18% scheme III 4 A-S 125.4 (28.2) 36 +31% scheme III 5 B-D 127.2 (28.6) 36 +32%

PAGE 47

CHAPTER 5 CONCLUSIONS The experimental set-up was designed to evaluate the long-term durability of externally bonded FRP reinforcement for concrete structures and to develop models to predict their service life. Thus far, two test groups, with a total of 16 beams, were investigated after one and two cycles of accelerated corrosion. The first group consisted of eight specimens tested after the initial corrosion process. Beams out of the second group were repaired after the initial corrosion process, exposed to the second cycle of accelerated corrosion, and then load tested. The following conclusions can be withdrawn from the test program: Un-strengthened test specimen lost between 8% and 15% of their load capacity due to corrosion compared to the control beam. Beams strengthened by scheme I had an increase in load capacity after PR-corrosion of 27% over the control beam. The post-repair performance of specimen strengthened by scheme I tested before and after PR-corrosion showed almost no decrease of ultimate load capacity (-3%). Beams strengthened by scheme II had an increase in load capacity after PR-corrosion of 30% and 31% over the control beam. The post-repair performance of specimen strengthened by scheme II tested before and after PR-corrosion showed a decrease in ultimate load capacity of 12%. Beams strengthened by scheme III had an increase in load capacity after PR-corrosion by 30% and 32% over the control beam. The post-repair performance of specimen strengthened by scheme III tested before and after PR-corrosion showed a decrease in ultimate load capacity between 9% and 12%. Un-repaired beams strengthened by scheme III had an increase in load capacity after continuing corrosion by 18% over the control beam. The post-repair performance of un-repaired specimen compared to repaired specimen decreased by 9%. 37

PAGE 48

38 Corrosion caused a loss of stiffness on test specimen. Stiffness can be regained by FRP application. FRP applications decreased corrosion rates during post-repair corrosion by minimizing chloride diffusion.

PAGE 49

LIST OF REFERENCES American Concrete Institute (ACI) Committee 222 (1996). Corrosion of metals in concrete. American Concrete Institute, ACI 222R-96, Farmington Hills, Michigan. American Concrete Institute (ACI) Committee 440 (2002). Guide for the design and construction of externally bonded FRP systems for strengthening concrete structures. American Concrete Institute, ACI 440.2R-02, Farmington Hills, Michigan. American Concrete Institute (ACI) Committee 546 (2001). Concrete repair guide. American Concrete Institute, ACI 556R-96 (Reapproved 2001), Farmington Hills, Michigan. American Society for Testing and Materials (ASTM) (1999). Standard test method for half-cell potentials of uncoated reinforcing steel in concrete. C 876-91 (Reapproved 1999), West Conshohocken, Pennsylvania. Berver, E. W., Fowler, D. W., and King, J. J. (2001). Corrosion in FRP-wrapped concrete members. 9 th International Conference and Exhibition Structural Faults and Repair 2001, London, UK. Bonacci, J. F. and Maalej, M. (2000). Externally bonded fiber-reinforced polymer for rehabilitation of corrosion damaged concrete beams. ACI Structural Journal, September/October 2000, 703-711. Jones, D. A. (1996). Principles and prevention of corrosion. Prentice Hall, Upper Saddle River, NJ, 2 nd Edition 1996. Karbhari, V. M. (1996). Effects of environmental exposure on the external strengthening of concrete with composites short term bond durability. Journal of Reinforced Plastics and Composites, Vol. 15, 1194-1216. Soudki, K. A., Zeng, J. S., Meise, B. and Sherwood, E. G. (1999). Durability of repaired concrete beams with CFRP laminates subjected to wet-dry cycles. Proceedings of the first International Conference Durability of FRP Composites for Construction (CDCC 98), 499-512 Soudki, K. A., Sherwood, T., and Masoud, S. (2003). FRP repair of corrosion-damaged reinforced concrete. University of Waterloo, Canada, Department of Civil Engineering. 39

PAGE 50

40 Virmani, P. Y. (2002). Corrosion costs and preventive strategies in the United States. Federal Highway Administration, Report FHWA-RD-01-156

PAGE 51

BIOGRAPHICAL SKETCH The author was born June 18, 1975, in Malapane, Poland, to Henrik and Margarete Kutarba. He graduated from KSL High School in Berlin, Germany, in 1994, and then started his college career at the Berufsakademie Berlin, Germany. After receiving the degree of Bachelor of Science in Civil Engineering, he enrolled in the graduate program of the University of Florida, Department of Civil Engineering to pursue a Master of Engineering with a focus in structural design. His graduate career was focused on durability problems of fiber reinforced polymer (FRP) repair methods. After receiving his Master of Engineering degree in May 2004, he will pursue a career in structural design. 41


Permanent Link: http://ufdc.ufl.edu/UFE0004766/00001

Material Information

Title: Durability of Carbon Fiber Reinforced Polymer (CFRP) Strengthening Systems Used to Repair Corrosion Damage in Reinforced Concrete
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0004766:00001

Permanent Link: http://ufdc.ufl.edu/UFE0004766/00001

Material Information

Title: Durability of Carbon Fiber Reinforced Polymer (CFRP) Strengthening Systems Used to Repair Corrosion Damage in Reinforced Concrete
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0004766:00001


This item has the following downloads:


Full Text












DURABILITY OF CARBON FIBER REINFORCED POLYMER (CFRP)
STRENGTHENING SYSTEMS USED TO REPAIR CORROSION
DAMAGE IN REINFORCED CONCRETE















By

MARKUS PETER KUTARBA


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ENGINEERING

UNIVERSITY OF FLORIDA


2004



































Copyright 2004

by

Markus Peter Kutarba















ACKNOWLEDGMENTS

Completion of my thesis and the research associated with it would not have been

successful without the help and guidance of a number of individuals. I thank the chair

and members of my committee (Dr. H.R. Hamilton, Dr. John Lybas, and Dr. Andrew J.

Boyd) for their guidance and assistance. The Department of Civil and Coastal

Engineering provided financial support throughout my graduate studies. The National

Science Foundation (NSF) gave financial support to my study. The Fyfe Cooperation

donated all materials. Sarah Witt is gratefully acknowledged for arranging this

contribution.

Others deserving thanks for contributions of resources and knowledge are John

Levar, Jeff Brown, and Chuck Broward; and Mario Paredes Cruz and Mitch Langley

(Florida Department of Transportation, Materials Office).

Lastly, I thank all of my close friends and family. Their enduring support proved to

be of the utmost value.
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S .................................................................... ......... .............. iii

LIST OF TABLES ............. ........... ............... .......... vi

L IST O F FIG U R E S .... ...... ................................................ .. .. ..... .............. vii

ABSTRACT ........ .............. ............. ...... .......... .......... ix

CHAPTER

1 IN TR OD U CTION ............................................... .. ......................... ..

B a c k g ro u n d ................................................................. .................................... 1
Objective of Study ........................................................ ................ .2

2 LITER A TU R E REV IEW ............................................................. ....................... 4

3 EXPERIM EN TAL PROGRAM .............................................................. ...............

Introdu action ............................................................................................ ............. 6
T est Specim en D description .......................................................... ...............6.
Initial C orrosion ................................................... 6
Concrete Repair ......................................... ........................ ............
Concrete strengthening .......................................... .. .. .... ................ .8
Post-R epair A accelerated Corrosion ........................................ ......................... 9
L oad T testing ............................................................................ 9

4 TEST RESULTS AND DISCUSSION ........................................ ...............15

Crack Pattern due to C orrosion........................................................ ............... 15
R paired O nly Specim ens.............................................. ........................... 16
Specimens Strengthened by Scheme I and II ..................... .................. .......... 16
Specimens Strengthened by Scheme III ...........................................................16
C orrosion D am age R ating ................................................ .............................. 17
Category I ..................................... ................................ ......... 18
C category II ................................................................ .. .... ......... 18
C ate g o ry III.................................................................... .. 1 8
C ateg ory IV ........................................................................18










S tru ctu ral B eh av io r ................................................................................................ 18
Control Beam s ..................................... .. ............... ....... 19
R paired B eam s ................ ... .............. ............. ....... ....... ..... 20
Schem e I Specim ens ........... ........... .. .......................................... ......20
Schem e II Specim ens........... ...... ................................................... ... .... .... .. 2 1
Schem e III Specim ens .................................................. .............................. 22

5 CON CLU SION S ........................................................ ..............37

LIST OF REFEREN CE S ...................................................................... ............... 39

B IO G R A PH IC A L SK E T C H ...................................................................... ..................41












































v
















LIST OF TABLES

Table page

1 Material properties for strengthening system ................................................... 13

2 Test specimen description. ................... ......... ....................... 14

3 Half-cell readings from accelerated corrosion. ............................................. 35

4 Results load test after initial corrosion. .............. .................................... ........ 35

5 Results load test after PR corrosion. ....................................................... 36
















LIST OF FIGURES


Figure page

1 Application of FRP on bridge Girder (Chaffee Rd, Jacksonville) ..................... 3

2 Potential problems with repaired and strengthened reinforced concrete ............... 3

3 Specim en construction details ......................................................................... 10

4 Ponding area on beam .. .................................................... ...... .............. 10

5 Beam 7 B-S. .......................................... 11

6 B eam repair details................... .......................... .. ................. ...... ....... ...... 12

7 R epair schem es............................. .............. ...... 12

8 Plot of half-cell readings vs. strengthening scheme..................................... 13

9 Typical crack pattern after initial accelerated corrosion process ...................... 24

10 Corrosion induced cracking on beam 6 A-D......... ................................... 24

11 C rack pattern on beam 3 A -S. ............................................................................... 24

12 D ebonded areas on beam 3 B -D .................................................. ... ................. 25

13 Corrosion on beam 4 A -S ............................................................... ..... 25

14 Pictures taken of category I corrosion-damage ............................................. ...... 26

15 Pictures taken of category II corrosion-damage............................................. 276

16 Pictures taken of category III corrosion-damage on control beam. .................. 27

17 Pictures taken of category IV corrosion-damage on repaired beam. ................ 28

18 Half-cell readings on beams strengthened by scheme III................. ......... 29

19 Ring anode corrosion at interface between old concrete and repair material. ...... 29

20 Load-deflection diagram for beam C 2. ............. ............................... ....... ....... 30









21 Load-deflection diagram for beams C 2 and 2 B-S....................................... 30

22 Load-deflection diagram for beams C 2, 1 A-S, and 4 A-D. ............................... 30

23 Load-deflection diagram for beams C 1 and 2 A-D........................................ 31

24 Load-deflection diagram for beams C 2, 5 A-D, and 7 B-S................................ 31

25 Load-deflection diagram for beams C 2 and 3 B-S.............................................. 31

26 Load-deflection diagram for beams C 2 and 7 A-S. .......................................... 32

27 Load-deflection plot for beams C2, 5 B-S and 4 B-D........................................ 32

28 D ebonded area on beam 5 B -S. ............................................................................. 32

29 Failure mode for beam 4 B-D at mid-span, view from the north side. .............. 33

30 Load-deflection diagram for beams C 2 and 4 B-S............................ ........... 33

31 Load-deflection diagram for beams 4 A-S, 5 B-D, and C2.................................. 33

32 Load-deflection diagram for beams C2 and 3 B-D. ......................................... 34

33 IR pictures of beam 3 B -D ................................................................................... 34















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Engineering

DURABILITY OF CARBON FIBER REINFORCED POLYMER (CFRP)
STRENGTHENING SYSTEMS USED TO REPAIR CORROSION
DAMAGE IN REINFORCED CONCRETE

By

Markus Peter Kutarba

May 2004

Chair: H.R. Hamilton
Major Department: Civil and Coastal Engineering

We studied the use of carbon fiber reinforced polymer (CFRP) systems combined

with concrete repair as a rehabilitation technique for corroded reinforced concrete beams.

Thirty reinforced concrete test specimens were constructed. Twenty-six of them were

subjected to an aggressive environment for 18 weeks. After being corroded, the beams

were repaired and strengthened, with CFRP using three different schemes. A group of

eight beams was load tested after the repair and strengthening process, while the

remaining beams were subjected to an aggressive environment for another 22 weeks.

After the post-repair (PR) corrosion, a second group of eight beams was load tested. The

results were compared to the first group regarding load capacity, stiffness, and degree of

corrosion damage. It was found that the application of CFRP significantly decreases

corrosion activity. The load capacity of PR corroded beams strengthened with CFRP was

about 30% higher than for control beams, and the stiffness was restored to the original









value. Based on the results and observations, a corrosion damage rating system for

reinforced concrete beams was developed.














CHAPTER 1
INTRODUCTION

Background

The United States transportation infrastructure continues to age and deteriorate.

Approximately 15% of the nation's 583,000 bridges are structurally deficient because of

corrosion of steel sections and steel reinforcement. The annual direct cost for corrosion

damage on bridges is estimated at $8.3 billion (Virmani 2002).

Until recently, the most common repair and strengthening methods were

replacement of the damaged element, steel plate bonding, or installation of additional

supports. The introduction of fiber reinforced polymer (FRP) composite systems into

construction is a promising alternative for the rehabilitation of weakened and deficient

concrete members. The advantages of FRP applications over traditional methods include

a high strength-to-weight ratio, ease of material handling and installation, and resistance

to attack in aggressive environments. Examples of current repair methods include

bonded systems in which a plate, saturated fabric, or spray-up system is applied to the

surface of the concrete element. The cured FRP composite provides additional shear

reinforcement, flexural reinforcement, or both to existing beam and slab elements.

Figure 1 shows FRP applied to an existing bridge structure. The bridge girder was

strengthened for flexure and shear using a wet lay-up system.

A large number of experimental studies have demonstrated the effects of moisture,

aggressive environments, fire, UV-light, creep, and fatigue on externally bonded FRP

systems (Karbhari 1996). Results of these studies indicated that bond behavior between









FRP and concrete is the most critical component. However, few studies have

investigated the long-term performance of FRP systems used as a strengthening material

for corrosion-damaged concrete.

Figure 2 shows a possible repair and strengthening scenario on a concrete member.

Before applying the strengthening system, the damaged steel and concrete must be

properly repaired. If concrete was patched under the FRP, then the bond between

existing concrete and patch material must be studied. Cracks due to shrinkage or a faulty

repair along the interface can cause further corrosion and possibly result in a bond failure

between repair material and existing concrete. If the FRP fully covers the repair, then the

usual outward signs of corrosion (such as cracking, staining, and spelling) will not be

visible. Visual inspection is currently the primary form of bridge and building evaluation

and monitoring. If corrosion of the steel reinforcement continues after repair, the FRP

composite may prevent early detection and intervention. Consequently, it is important

that the effect of post-repair corrosion on the structural behavior be better understood. In

addition, it is important that the outward signs of continued degradation of the beams be

characterized so that adequate evaluations of the post-repair status can be made.

Objective of Study

My objective was to study the long-term behavior of repaired reinforced concrete

that has been strengthened with FRP. The durability of the composite beam was

evaluated by examining the bond between the FRP strengthening system and the

concrete; and also the bond between patch material and concrete. We also developed a

damage-rating system, based on visual inspection of surface and reinforcement damage.

The first step was to corrode the reinforcing steel in the beams. The second step

was to repair the beams, and apply an FRP strengthening system. The beams were then









exposed to another corrosion cycle. After a sufficient amount of corrosion has taken

place in the repaired beams, the effects of the corrosion on the repair and strengthening

system can be determined. Results were then used to create the corrosion-damage rating

system.


Figure 1. Application of FRP on bridge girder (Chaffee Rd, Jacksonville).


original
concrete

FRP system

original steel

patch
material


patch
debonding

accumulation
of rust

cracking due
to corrosion

FRP debonding


Figure 2. Potential problems with repaired and strengthened reinforced concrete.














CHAPTER 2
LITERATURE REVIEW

The effect of corrosion on reinforced concrete is well documented. It is known that

concrete provides a high alkaline environment that forms a passive film on the steel

surface, which prevents steel from corroding. Chlorides disrupt this protective layer, and

steel starts to corrode. Expansive forces, which are a result of corrosion, will crack and

spall the surrounding concrete. The consequence is less-protected steel causing

additional corrosion, and a greater loss in steel cross-section (Jones 1996, ACI 1996).

Depending on the degree of damage, a concrete repair in conjunction with strengthening

method (such as bonded FRP composites) can improve the strength and serviceability

(ACI 2001).

FRP composites can be designed to be resistant to the aggressive environments that

attack reinforced concrete, which makes them an attractive material for use in repair and

strengthening of corrosion-damaged structures. While the behavior of the FRP composite

in aggressive environments is well understood, researchers have only recently attempted

to characterize the performance of FRP composites bonded to corrosion

damaged-reinforced concrete. Testing has shown that corrosion-damaged beams

strengthened with an FRP system are stiffer, and have a higher load-carrying capacity

than unstrengthened beams (Bonacci and Maalej 2000, Soudki et al. 2003). It has also

been shown that FRP reduces corrosion rates of reinforcing steel in concrete by

minimizing the diffusion of chlorides (Berver et al. 2001, Soudki, et al. 1999). However,

very little information is available on the durability of repaired corrosion-damaged






5


structures. Questions remain concerning the synergistic effects of continued corrosion on

repaired system performance.

This thesis highlights the research findings of a multiphase experimental study

aimed to evaluate the durability and structural behavior of corrosion-damaged beams

repaired and strengthened with FRP.














CHAPTER 3
EXPERIMENTAL PROGRAM

Introduction

Thirty test specimens were fabricated to investigate the long-term behavior of

repaired reinforced concrete beams strengthened with FRP. First, the beams were

subjected to a corrosion process for initial damage. They were then repaired and

strengthened, and exposed to a second corrosion process. At this point, 16 beams were

tested in four-point bending. Eight beams continued to be exposed to an accelerated

corrosion process, and the six remaining beams were reserved for control specimens.

Test Specimen Description

Each rectangular beam was 203 mm x 305 mm (8 in x 12 in) in cross-section and

2900 mm (9 ft-6 in) long. The internal longitudinal steel reinforcement consisted of two

19 mm (#6), 414 MPa (60 ksi) bars at the bottom, and two 9.5 mm (#3), 276 MPa (40

ksi) bars at the top of the beam. The reinforcing steel was extended 76 mm (3 in) beyond

the end of the concrete, for the purpose of making external electrical connections. The

clear concrete cover was 51 mm (2 in) on all aspects of the specimen. Three different

batches of concrete mixtures were used to construct the beams. The average compression

strength ranged between 29.6 and 37.9 MPa (4300 to 5500 psi). Figure 3 shows a typical

beam section.

Initial Corrosion

Twenty-six of the thirty beams were subjected to accelerated corrosion. Before

starting the initial accelerated corrosion, the beams were subjected to center-point loading









necessary to cause a minimum crack width of 0.33 mm (0.013 in). Next, ponding areas

were constructed around the cracked regions, and a salt solution (5% NaCl by weight)

was applied. The ponding was cycled at 1 week wet and 1 week dry. A 5-volt constant

potential was applied, to accelerate the corrosion process. Half-cell potential readings

were taken at the beginning and end of each cycle, to determine the corrosion activity.

The corrosion process lasted 18 weeks (9 wet and dry cycles), and resulted in severe

corrosion damage to the beams. Figure 4 shows a beam in the initial corrosion process,

with a single ponding area.

Concrete Repair

ACI repair procedures (ACI 2001) were followed in repairing the corrosion

damage. All damaged material was carefully separated from sound concrete without

causing further damage. Undamaged concrete behind the reinforcing steel was removed

to create a clear space of 25.4 mm (1 in). At the perimeter of the removal area 25.4 mm

(1 in) deep saw cuts were made. This prevented the formation of feathered edges and

provided adequate spacing for repair material. After the concrete was removed, all

corroded areas on the reinforcement were sandblasted to white metal. The repair and

cleaning process is shown in Figure 5. Figure 6 shows the repair layout.

To restore the beams' original shape after the corrosion repair a concrete mixture

with a 46.2 MPa (6700 psi) 28 day-strength was used. The maximum aggregate size was

limited to 10 mm (0.375 in). In addition, a super plasticizer was mixed with the concrete

to reduce the water-cement ratio. The concrete performance evaluation report is shown

in the Appendix.









Concrete strengthening

After preparing the concrete surface by a light sandblasting, three strengthening

schemes were applied to the beams.

The first scheme (scheme I) involved strengthening the specimen with separate

FRP sheets for flexure and shear. A typical flexural sheet was 2743 mm (108 in) long

and 203 mm (8 in) wide. The shear strips were 152 mm x 305 mm (6 in x 12 in). On

each side eight sheets were used with a center-to-center spacing of 305 mm (12 in). At

mid-span a 305 mm (12 in) wide field was left un-strengthened to allow for free

development of corrosion cracks. Beams strengthening by the second scheme (scheme

II) used the same FRP layout as scheme I except that the shear sheets were used as

anchors for the flexural sheet. This sheet was fixed either at the ends only or over the

entire length using all shear FRP sheets. The third group was strengthened with a single

FRP sheet used as a full wrap (scheme III). This sheet was 2743 mm x 610 mm (108 in x

24 in) and covered the bottom of the specimens as well as 203 mm (8 in) of each side.

The different strengthening configurations are shown in Figure 7.

A unidirectional carbon fabric with glass veil backing for added fabric stability

during installation was used for strengthening. The matrix, Tyfo S Epoxy, was a two

component epoxy material. The composite material properties as given by the

manufacturer are shown in Table 1.

After the specimens were strengthened, eight beams were load tested to failure.

These tests were done to study the beam behavior under loading conditions and to collect

reference data for future testing.









Post-Repair Accelerated Corrosion

For the post-repair (PR) corrosion process 16 beams, 4 repaired only (R), 10

repaired and strengthened (RS) and 2 unrepaired and strengthened (URS) specimens were

pre-cracked under a load of 44.48 kN (10 kip) which simulated service load conditions

for the control beams. Crack widths could not be measured due to the FRP composite.

The beams were then ponded with a 3.5% salt solution for one week under constant 5-

volts followed by one week of drying. Half-cell readings were taken after each wet and

dry cycle. After 22 weeks (11 wet and dry cycles) eight beams were taken from the PR

corrosion process and load tested. The last eight beams remained in the corrosion

process.

Load Testing

After each corrosion process, eight beams were tested in four-point loading to

failure. All beams were simply supported; the distance between the load point and the

end supports was set at 1219 mm (48 in) except for beams C 1 and 2 A-D. The distance

between the load point and the end supports for beams C 1 and 2 A-D was 914 mm (36

in). This resulted in a constant moment zone at mid-span of 914 mm (36 in) for C 1 and

2 A-D, and 305 mm (12 in) for all other beams. The beams were tested in three or four

stages of loading. Deflection was measured with two Linear Variable Displacement

Transducers (LVDT) at mid-span. A hydraulic jack applied the load on the test

specimens at a load rate of 222 N/sec (50 lbf/sec). The data were collected using an

automated data acquisition system.

Two sets of beams have been tested thus far. The first group consisted of control,

repaired only (R), and repaired and strengthened (RS) beams. These tests were

performed after the initial accelerated corrosion, repair and strengthening process.






10


Beams from the second group were tested after PR accelerated corrosion. All beams

except for one (3 B-D) have been repaired as described earlier. Beam 3 B-D was

strengthened without a corrosion repair (URS) by scheme III. The test grouping can be

seen in Table 2; Figure 8 shows the plots of half-cell readings vs. strengthening scheme.

(2)#3 (2)#6 E
E
I- E
(2)#3 n
LO\E I I
E ................ .. -
o i\ E C'



iqi E
E
203mm 2900mm 1
(8in) (114in)
A B

Figure 3. Specimen construction details. A) Cross section. B) Elevation.


Figure 4. Ponding area on beam.






11
















A















B















C

Figure 5. Beam 7 B-S. A) Splitting caused by corrosion. B) After removal of damaged
concrete; saw cuts can be seen at the perimeter edges. C) Corroded
reinforcement before sandblasting.







patch material, patch size
depends on concrete crack


Figure 6. Beam repair details.


Figure 7. Repair schemes


Scheme I

S1 1 111 1111
Scheme II






Scheme III

I6-












0
-100
-200
-300
-400
-500
-600
-700


0
S-100
O
) -200
0 -300
S-400
-500
E -600

-700


control scheme I scheme II


after 4 weeks

A


control scheme I scheme II scheme III


Figure 8. Plot of half-cell readings vs. strengthening scheme. A) After 4 weeks. B)
After 22 weeks.


Table 1. Material properties for strengthening system
Tensile strength Max. Tensile modulus Thickness
MPa (ksi) elongation GPa (ksi) mm (in)
876 (127) 1.2% 72.4 (10500) 1.0(0.04) 820C (1800F)


..... ..... ..... ....


scheme III









Table 2. Test specimen description
Beam Strengthening scheme
C 1 Control
C 2 Control
1 A-S Control
2 A-D Control
2 B-S Control
3 B-D Scheme III
3 B-S Scheme I
4 A-D Control
4 A-S Scheme III
4 B-D Scheme II
4 B-S Scheme III
5 A-D Scheme I
5 B-D Scheme III
5 B-S Scheme II
7 A-S Scheme II
7 B-S Scheme I


Date of testing
After initial corrosion
After initial corrosion
After re-corrosion
After initial corrosion
After initial corrosion
After re-corrosion
After re-corrosion
After re-corrosion
After re-corrosion
After re-corrosion
After initial corrosion
After initial corrosion
After re-corrosion
After re-corrosion
After initial corrosion
After initial corrosion


Repair information
Not corroded
Not corroded
Repaired
Repaired
Repaired
Not repaired
Repaired
Repaired
Repaired
Repaired
Repaired
Repaired
Repaired
Repaired
Repaired
Repaired


Numbers indicate the set of test specimen, the letter A or B the position
in the corrosion racks. The letters S or D describe the number of
ponding areas (single or double) during the initial corrosion process.














CHAPTER 4
TEST RESULTS AND DISCUSSION

Crack Pattern due to Corrosion

Both corrosion programs were aimed to cause concrete cracks due to corrosion of

the main reinforcing steel. The initial corrosion process was stopped after 18 weeks.

Horizontal cracks had formed at the surface along the level of the main reinforcement

with rust leaching out. Crack lengths depended on the ponding area lay-out; cracks on

beams with two ponding areas were an average 2337 mm (92 in) long and cracks on

beams with a single ponding area were an average 1600 mm (63 in) long. Figure 9 shows

a typical picture of a horizontal corrosion crack on beam after 18 weeks of wet and dry

cycles. PR corrosion was conducted on the selected 16 beams which were described

earlier. These beams were exposed for 22 weeks to wet and dry cycles before the test

was stopped. The crack pattern and the level of damage on the PR corroded beams

depended on the type of strengthening system; un-strengthened specimen, specimen

strengthened by scheme I or scheme II, and specimens strengthened by scheme III. Table

3 summarizes the accelerated corrosion process of all beams. Half-cell readings were

made using a copper/copper sulfate electrode on the concrete surface. Although the

current ASTM Standard (ASTM C 876) requires concrete contact for half-cell

measurements, readings on beams strengthened by scheme III were made on FRP.

Knowing that epoxy is not conductive, half-cell readings taken on the FRP laminate were

expected to be zero. However, it was possible to get stable readings, which will be









compared with the degree of corrosion to verify the measurement. Readings smaller than

-350 mV indicate corrosion activity.

Repaired Only Specimens

First signs of corrosion were observed after four weeks of wet and dry cycles.

Small cracks with rust leaching out were found along the level of the main reinforcing

steel and at the interface between old concrete and patch material. The crack size and the

amount of rust increased over the entire length of the test. By the end of the PR corrosion

period all specimen had the same crack pattern; one major crack at the level of the

reinforcement and a smaller secondary crack at the interface. Figure 10 shows an

unstrengthened beam after the re-corrosion process. Cracks at the interface and the main

reinforcement can be seen.

Specimens Strengthened by Scheme I and II

Beams strengthened by scheme I and II showed no visible signs of corrosion before

the sixth week. After the sixth week small cracks without any rust traces were visible.

By the end of the PR corrosion process, small cracks along the level of the main

reinforcement and at the interface had formed on all beams with very little, or no rust

leaching out. It was also observed that the shear FRP hindered corrosion cracks to grow

in vertical direction but not in horizontal. However, these cracks were much smaller than

the cracks on un-strengthened beams. Typical corrosion-damage for specimens

strengthened by scheme I can be seen in Figure 11.

Specimens Strengthened by Scheme III

This group consisted of two different types of beams, RS beams (4 A-S and 5 B-D)

and URS beams (2 B-D and 3 B-D). No visible signs of corrosion were spotted over the

entire length of testing in both groups. However, half-cell readings verified corrosion









activity on the un-repaired beams. A comparison of infrared (IR) pictures taken after

curing and at the end of the corrosion process showed the development of large debonded

areas on these beams. The debonded regions are a result of tension forces at the concrete

surface caused by steel corrosion. The development of debonded regions can be seen in

Figure 12b denoted by light-colored regions.

No debonded areas or other signs of corrosion were detected on RS beams during

the entire PR corrosion. An investigation of the reinforcement after testing verified the

results from the visual inspection, except for beam 4 A-S.

On beam 4 A-S some corrosion was found at the interface between old and new

concrete on the main reinforcing steel. This corrosion spot was caused by the so called

"ring anode corrosion" or "halo effect". When repairs are completed on

corrosion-damaged structures, the new concrete used to form the patch creates a

corrosion potential with the existing concrete, which initiates new corrosion sites in the

contaminated concrete near the patch interface. But since this spot was small and at the

end of the beam it did not affect the structural behavior. Figure 13 shows this effect on

beam 4 A-S.

Corrosion Damage Rating

After investigating the corrosion damage on the test specimens, each beam was

assigned to one of four categories. Each category represents a specific level of damage

based on visual observations of concrete cracking and reinforcement corrosion. Category

I represents the lowest degree of corrosion damage, category III the highest. Category IV

represents beams with a full FRP wrap on which half-cell measurements indicated

corrosion activity.









Category I

Description: Very small horizontal cracks on the concrete surface with little or no

traces of rust. No damage on the reinforcing steel visible. Figure 14 shows typical

category I corrosion damage.

Category II

Description: Visible horizontal cracks with rust traces on concrete surface. Crack

width is less than 1 mm (0.04 in). Very few and very small pitting corrosion spots on

steel reinforcement. Figure 15 shows typical category II corrosion damage.

Category III

Description: Visible horizontal cracks with large rust stains on concrete surface.

Crack width is less than 3 mm (0.12 in). Severe pitting corrosion and uniform surface

corrosion damage on steel reinforcement. A loss of cross section area is visible. Figure

16 and Figure 17 show typical category III corrosion damage.

Category IV

Description: The typical outward signs of corrosion were not visible on beams

strengthened by scheme III due to the FRP layout. However, half-cell measurements can

be an indicator of corrosion activity. Figure 18 shows half-cell readings from two RS

beam at mid-span and at the interface between old concrete and repair material. Even

though the readings fluctuate over the period of 20 weeks, the measurements on beam 4

A-S (Figure 18a) at the interface tend to indicate corrosion activity. An investigation

after the test verified ring anode corrosion (Figure 19).

Structural Behavior

After each corrosion process, a group of eight beams was tested in flexure to

failure. Typically, all corroded R beams failed at lower loads than the control beam.









RS and URS beams failed at higher loads. Post-cracking stiffness was always less

than the control beam's stiffness, except for beams strengthened by scheme III.

Those beams stiffness were almost identical with the control beam stiffness.

All beams failed in the form of concrete crushing except for beams C 1 and 2 A-D.

At maximum load, RS and URS specimens showed a brittle failure mode in the form of

concrete crushing. R beams failed in a more ductile mode. It can be seen on the load-

deflection diagrams for R beams (Figure 21, Figure 22) that the plot plateaus at

approximately 89 kN (20 kip) indicate that the main reinforcing steel has yielded. This

was confirmed by strain measurements on the reinforcement. RS and URS beams,

however, do not exhibit as lengthy a plateau even though strain gauge data indicated that

steel had yielded.

Control beams C 1 and 2 A-D were tested in a different test set-up as described

earlier. Both beams failed in shear and therefore they were not compared in this study.

After testing Cl and 2 A-D the free length at mid-span was reduced to 305 mm (12 in).

Table 4 displays the results from tests after the initial corrosion and Table 5 displays the

results from tests after PR corrosion. The post cracking slope (PC) was calculated using

load and deflection values between 44.5 kN (10 kip) and 66.7 kN (15 kip) and was used

as a tool to describe the stiffness.

Control Beams

Beam C 2 was tested to be used as a control beam for all test specimens. This

beam was not corroded. The maximum load capacity was 96 kN (21.6 kip) at a

deflection of 20 mm (0.79 in). The reinforcement started to yield at approximately 62.3

kN (14 kip). Figure 20 shows the load versus mid-span deflection plot. Beam C 2 failed

by concrete crushing.









Repaired Beams

Beam 2 B-S was tested to failure after the initial corrosion and repair process. The

ultimate load capacity was 8% lower than the control beam's capacity. The PC stiffness

was reduced by about 20%. Figure 21 shows the load-deflection diagram for beam C 2

and 2 B-S.

The load deflection plots for PR corroded specimens (1 A-S and 4 A-D) are shown

in Figure 22. Both beams were tested after the initial corrosion, repair, and PR corrosion

process. It can be seen that the load capacities and stiffness are less than that of the

control beam. The ultimate capacities were 8% (4 A-D) and 15% (1 A-S) lower; the

post-cracking stiffness was 29% (4 A-D) and 11% (1 A-S) lower.

It is likely that the lower stiffness and load capacities were caused by the corrosion

process. Typically, steel cross section area loss and bond failure between the

reinforcement and concrete occur during corrosion. A reduced steel cross section area

results in a lower load carrying capability, the loss of bond allows the reinforcing steel to

yield without any restraint and results in a lower stiffness as well as load capacity.

The concrete repairs performed well in the load tests with no significant cracking

debonding, or spelling of the repair material noted. Beams C 1 and 2 A-D both failed in

shear with diagonal cracks forming in the existing concrete and passing through the

interface into the repair material. Figure 23 shows the load deflection diagram for both

beams. It can be seen that beam 2 A-D has the same post corrosion loss in stiffness as

those of the other specimen that had been through the accelerated corrosion.

Scheme I Specimens

The load-deflection response for strengthened beams (5 A-D and 7 B-S) tested after

the initial corrosion process is shown in Figure 24. Both beams have been repaired









before FRP application and testing. The flexural FRP sheet on beam 5 A-D covered the

tension face completely, while the sheet on beam 7 B-S was 102 mm (4 in) wide and

covered only 50%. The increase in capacity over the control beam was 14% for beam 7

B-D and 30% for beam 5 A-D. The PC stiffness increased slightly (13%) for beam 5 A-

D; for beam 7 B-S it did not change.

Beam 3 B-D was tested un-repaired after the re-corrosion process. The ultimate

capacity increased by 27%; the stiffness did not change within the limits. Figure 25

shows the load-deflection plot for the re-corroded beam compared to the control beam.

During testing, prior to failure, signs of debonding were observed on the beams.

Several loud cracking sounds indicated that epoxy was cracking at multiple locations.

However, epoxy cracking was limited to very small areas and was not seen at all during

IR inspections. Concrete crushed on beams strengthened by scheme I before FRP

debonding became the failure mechanism.

The change in post cracking stiffness was very small for beam 5 A-D and 7 B-S

compared to the control beam. The same performance was observed on beam 3 B-S.

Since the degree of corrosion damage was small too, this behavior was expected.

Scheme II Specimens

Figure 26 shows the load-deflection diagram for beam 7 A-S. The flexural sheet on

this beam was anchored with U-wraps at multiple points leaving an un-strengthened

length of 305 mm (12 in) at mid-span. Beam 7 A-S was load tested after the initial

corrosion and a repair procedure. The load capacity was 47% higher than the capacity of

the control beam; the post cracking stiffness increased only slightly.









Beam 5 B-S was strengthened by the same FRP lay-out as beam 7 A-S. The load

capacity after the re-corrosion process was 30% higher than the control beam's capacity.

The stiffness decreased at the same time by about 20% (see Figure 27).

The flexural FRP sheet on beam 4 B-D was anchored at each end by a single U-

wrap. The load capacity increased by 31%, while the stiffness dropped by 17%. Figure

27 shows the load-deflection diagram for beams 5 B-S and 4 B-D compared to beam C 2.

Cracking sounds were observed on all three beams during testing. At ultimate load,

shortly after concrete crushing started, FRP debonded from the beams. On beams 7 A-S

and 5 B-S debonding started at mid-span and was limited on the area between the first

anchors (U-wraps). Figure 28 shows the debonded area at mid-span on beam 5 B-S. It

can be seen that the U-wrap prevented the FRP sheet from debonding. The white colored

spot indicate the debonded area; the dark colored spot left of it indicate the U-wrap.

FRP on beam 4 B-D debonded at the east end and ripped out the concrete cover

over the reinforcement at mid-span. No corrosion damage was spotted before testing in

this area; however, an investigation after testing found some rust traces. Figure 29 shows

beam 4 B-D after testing; concrete delaminated can be seen at the level of corrosion

cracks.

Scheme III Specimens

Figure 30 shows the load-deflection diagram for beam 4 B-S tested after the initial

corrosion compared to the control beam. It can be seen that the load capacity increased

by 44% without any loss in stiffness.

The PR corroded beams 4 A-S and 5 B-D showed an almost identical load

deflection behavior during testing. The ultimate load capacities were 30% (4 A-S) and









32% (5 B-D) higher than the control beams capacity. No loss in stiffness was observed.

The load-deflection diagram for these beams can be seen in Figure 31.

Throughout all tests several cracking sounds were heard. Like on the other

strengthened beams, the source was epoxy cracking. However, no debonded areas were

found at any point of testing. At ultimate load, all beams strengthened by scheme III

failed suddenly by concrete crushing without any previous warning.

Beam 3 B-D was strengthened by scheme III without a concrete repair after the

initial corrosion (URS). The strengthening FRP sheet was used to cover all corrosion

cracks. After the PR corrosion program several large debonded areas at multiple points

on the specimen were found. Figure 32 shows the load-deflection diagram for beam 3 B-

D compared to the control beam. The load capacity was 18% higher, while the stiffness

did not change. Like the other beams strengthened by scheme III this beam failed

suddenly by concrete crushing without any previous warning.

IR scans of the tension face were taken during loading of a large debonded area

caused by corrosion. At maximum load the beam failed by concrete crushing at mid-

span. The failure mode indicated that the previously debonded area did not adversely

affect the beam capacity. In addition, the two investigated debonded areas did not

expand towards mid-span during loading, but rather grew towards each other forming one

large debonded area (see Figure 33).















crack at main
reinforcement


















Figure 9. Typical crack pattern after initial accelerated corrosion process.


i*-~---
r-
:.1 ..:.
i,,I

; .~*~ II~)
I:
ra~? ;.I. !;


:-' ''




~*E-
~.. .,cr~* ~i~.


*' ; -i~
~ II.
I
I


Figure 10. Corrosion induced cracking on beam 6 A-D. The lower crack is at the patch
interface, while the upper crack is at the level of the main reinforcing bar.







crack at main

reinforcement


Figure 11. Crack pattern on beam 3 A-S. Crack formed along the level of the main steel

reinforcement.











































Figure 12. Debonded areas on beam 3 B-D. IR picture was taken A) before the
corrosion process and B) after the corrosion process.


interface
between old
concrete and
repair
material


Figure 13. Corrosion on beam 4 A-S.








































Figure 14. Pictures taken of category I corrosion-damage. A) Horizontal crack. B)
Corresponding damage on reinforcing steel.


Figure 15. Pictures taken of category II corrosion-damage. A) Horizontal crack. B)
Corresponding damage on reinforcing steel.



























Figure 15. Continued.


91'

B


Figure 16. Pictures taken of category III corrosion-damage on control beam. A)
Horizontal crack. B) Corresponding damage on reinforcing steel.










































Figure 17. Pictures taken of category IV corrosion-damage on repaired beam. A)
Horizontal crack. B) Corresponding damage on reinforcing steel.












0
. -100
8 -200
o -300
o -400
6 -500
> -600
E -700
-800


readings on FRP
-mid-span
-interface


Figure 18. Half-cell readings on beams strengthened by scheme III. A) Beam 4 A-S
(ring anode corrosion at interface). B) Beam 5 B-S (no corrosion).


Figure 19. Ring anode corrosion at interface between old concrete and repair material.


week
10


readings on FRP
--mid-span
- interface


week
10


0
-100
-200
-300
-400
-500
-600
-700
-800


I Z













Deflection [mm]
0 10 20 30






r I-


35
30
S25
~ 20
5 15
0 10
5
0


Deflection [in]


Figure 20. Load-deflection diagram for beam C 2.


Deflection [mm]
0 10 20


35
30
- 25
20
S15
. 10
5
0


140
120
100
80
60
40
20
0


0.5 1
Deflection [in]


Figure 21. Load-deflection diagram for beams C 2 and 2 B-S.

Deflection [mm]
0 10 20 30


35
30
" 25
i.20
a 15
. 10
5
0


--C2
-- -- C2 -


-/ / 4 A-D
/ -- 1A-S

3 0.5 1 1.
Deflection [in]


Figure 22. Load-deflection diagram for beams C 2, 1 A-S, and 4 A-D.


140
120
100
80
60
40
20
0










Deflection [mm]
0 10 20 30


35
30
S25
i20

0
j 10
5
0


0.5 1
Deflection [in]


Figure 23. Load-deflection diagram for beams C 1 and 2 A-D.

Deflection [mm]
0 10 20 30
35
30 140
S25 120
2... -.81 0 0 z
20 80
'o 15 60 -
-J 10 _--C2 40 -
5 / 5 A-D
5 7 B-S 20
0 0
0 0.5 1 1.5
Deflection [in]

Figure 24. Load-deflection diagram for beams C 2, 5 A-D, and 7 B-S.


Deflection [mm]
0 10 20


0.5 1
Deflection [in]


Figure 25. Load-deflection diagram for beams C 2 and 3 B-S.











Deflection [mm]
0 10 20 30


35
30
S25
S20
" 15
-- 10
5
0


0.5 1
Deflection [in]


Figure 26. Load-deflection diagram for beams C 2 and 7 A-S.

Deflection [mm]
0 10 20 30


35
30
' 25
" 20
cc 15
0
- 10
5
0


0.5 1
Deflection [in]


Figure 27. Load-deflection diagram for beams C2, 5 B-S, and 4 B-D.


Figure 28. Debonded area on beam 5 B-S.


- -
-C2

5 B-S
4 B-D























Figure 29. Failure mode for beam 4 B-D at mid-span; view from the north side.


Deflection [mm]
0 10 20 30


/



-C2
4 B-S


0.5 1
Deflection [in]


Figure 30. Load-deflection diagram for beams C 2 and 4 B-S.


Deflection [mm]
0 10 20 30

--4--
-4"


-C2
4A-S
--5B-D


0.5 1
Deflection [in]


Figure 31. Load-deflection diagram for beams 4 A-S, 5 B-D, and C2.










Deflection [mm]
0 10 20


0.5 1
Deflection [in]


Figure 32. Load-deflection diagram for beams C2 and 3 B-D.


Figure 33. IR pictures of beam 3 B-D. A) Before load testing. B) After 15 kip.









Table 3. Half-cell readings from accelerated corrosion


Beam


Type
control
control
scheme III
scheme II
scheme I
scheme III
scheme I
control
scheme III
scheme II
scheme II
scheme III
scheme II
control
scheme I
scheme I


Half-cell reading
after 4th week
-376 mV (conc)
-382 mV (conc)
-505 mV (FRP)
-434 mV (conc)
-309 mV (conc)
-338 mV (FRP)
-407 mV (conc)
-440 mV (conc)
-80 mV (FRP)
-368 mV (conc)
-280 mV (conc)
-357 mV (FRP)
-130 mV (conc)
-600 mV (conc)
-496 mV (conc)
-518 mV (conc)


Readings made on concrete (conc) or FRP (FRP)
Damage categories are explained in chapter 4
0 no signs of corrosion activity


Table 4. Results load test after initial corrosion
Test result in PC slope in Capacity change to
FRP configuration Beam .
kN (kip) kip/in beam C 2
control* C 1 104.3 (23.4) 37 not compared
control* 2 A-D 128 (28.8) 35 not compared
control C 2 96.1 (21.6) 36 0%
control 2 B-S 88.1 (19.8) 29 -8%
scheme I 5 A-D 124.5 (28) 41 +30%
scheme I(4") 7 B-S 111.2 (25) 36 +14%
scheme II 7 A-S 141.4 (31.8) 42 +47%
scheme III 4 B-S 137.9 (31) 37 +44%
beam failed in shear


Half-cell reading
after 22nd week
-582 mV (conc)
-563 mV (conc)
-593 mV (FRP)
-494 mV (conc)
-552 mV (conc)
-440 mV (FRP)
-461 mV (conc)
-545 mV (conc)
-80 mV (FRP)
-502 mV (conc)
-312 mV (conc)
-63 mV (FRP)
-260 mV (conc)
-504 mV (conc)
-467 mV (conc)
-514 mV (conc)


1 A-S
1 B-D
2 B-D
3 A-D
3A-S
3 B-D
3B-S
4 A-D
4A-S
4 B-D
5 A-S
5 B-D
5 B-S
6 A-D
6A-S
6B-S


Damage
category
III
III
IV
I
I
IV
I
III
IV
I
I
0
0
III
I
I






36


Table 5. Results load test after PR corrosion
Test result in PC slope in Capacity change to
FRP configuration Beam .
kN (kip) kip/in beam C 2
control 1 A-S 81.8(18.4) 31 -15%
control 4 A-D 87.6(19.7) 25 -8%
scheme I 3 B-S 121.9 (27.4) 36 +27%
scheme II 4 B-D 125.4 (28.2) 30 +31%
scheme II 5 B-S 124.5 (28) 29 +30%
scheme III 3 B-D 113.4 (25.5) 40 +18%
scheme III 4 A-S 125.4 (28.2) 36 +31%
scheme III 5 B-D 127.2 (28.6) 36 +32%














CHAPTER 5
CONCLUSIONS

The experimental set-up was designed to evaluate the long-term durability of

externally bonded FRP reinforcement for concrete structures and to develop models to

predict their service life. Thus far, two test groups, with a total of 16 beams, were

investigated after one and two cycles of accelerated corrosion. The first group consisted

of eight specimens tested after the initial corrosion process. Beams out of the second

group were repaired after the initial corrosion process, exposed to the second cycle of

accelerated corrosion, and then load tested. The following conclusions can be withdrawn

from the test program:

* Un-strengthened test specimen lost between 8% and 15% of their load capacity due
to corrosion compared to the control beam.

* Beams strengthened by scheme I had an increase in load capacity after PR-
corrosion of 27% over the control beam. The post-repair performance of specimen
strengthened by scheme I tested before and after PR-corrosion showed almost no
decrease of ultimate load capacity (-3%).

* Beams strengthened by scheme II had an increase in load capacity after PR-
corrosion of 30% and 31% over the control beam. The post-repair performance of
specimen strengthened by scheme II tested before and after PR-corrosion showed a
decrease in ultimate load capacity of 12%.

* Beams strengthened by scheme III had an increase in load capacity after PR-
corrosion by 30% and 32% over the control beam. The post-repair performance of
specimen strengthened by scheme III tested before and after PR-corrosion showed
a decrease in ultimate load capacity between 9% and 12%.

* Un-repaired beams strengthened by scheme III had an increase in load capacity
after continuing corrosion by 18% over the control beam. The post-repair
performance of un-repaired specimen compared to repaired specimen decreased by
9%.






38


* Corrosion caused a loss of stiffness on test specimen. Stiffness can be regained by
FRP application.

* FRP applications decreased corrosion rates during post-repair corrosion by
minimizing chloride diffusion.














LIST OF REFERENCES


American Concrete Institute (ACI) Committee 222 (1996). "Corrosion of metals in
concrete." American Concrete Institute, ACI 222R-96, Farmington Hills, Michigan.

American Concrete Institute (ACI) Committee 440 (2002). "Guide for the design and
construction of externally bonded FRP systems for strengthening concrete
structures." American Concrete Institute, ACI 440.2R-02, Farmington Hills,
Michigan.

American Concrete Institute (ACI) Committee 546 (2001). "Concrete repair guide."
American Concrete Institute, ACI 556R-96 (Reapproved 2001), Farmington Hills,
Michigan.

American Society for Testing and Materials (ASTM) (1999). "Standard test method for
half-cell potentials of uncoated reinforcing steel in concrete." C 876-91
(Reapproved 1999), West Conshohocken, Pennsylvania.

Berver, E. W., Fowler, D. W., and King, J. J. (2001). "Corrosion in FRP-wrapped
concrete members." 9th International Conference and Exhibition "Structural Faults
and Repair 2001", London, UK.

Bonacci, J. F. and Maalej, M. (2000). "Externally bonded fiber-reinforced polymer for
rehabilitation of corrosion damaged concrete beams." ACI Structural Journal,
September/October 2000, 703-711.

Jones, D. A. (1996). "Principles and prevention of corrosion." Prentice Hall, Upper
Saddle River, NJ, 2nd Edition 1996.

Karbhari, V. M. (1996). "Effects of environmental exposure on the external strengthening
of concrete with composites short term bond durability." Journal of Reinforced
Plastics and Composites, Vol. 15, 1194-1216.

Soudki, K. A., Zeng, J. S., Meise, B. and Sherwood, E. G. (1999). Durability of
repaired concrete beams with CFRP laminates subjected to wet-dry cycles."
Proceedings of the first International Conference "Durability of FRP
Composites for Construction" (CDCC 98), 499-512

Soudki, K. A., Sherwood, T., and Masoud, S. (2003). "FRP repair of corrosion-
damaged reinforced concrete." University of Waterloo, Canada, Department
of Civil Engineering.






40


Virmani, P. Y. (2002). "Corrosion costs and preventive strategies in the United
States." Federal Highway Administration, Report FHWA-RD-01-156















BIOGRAPHICAL SKETCH

The author was born June 18, 1975, in Malapane, Poland, to Henrik and Margarete

Kutarba. He graduated from KSL High School in Berlin, Germany, in 1994, and then

started his college career at the Berufsakademie Berlin, Germany. After receiving the

degree of Bachelor of Science in Civil Engineering, he enrolled in the graduate program

of the University of Florida, Department of Civil Engineering to pursue a Master of

Engineering with a focus in structural design. His graduate career was focused on

durability problems of fiber reinforced polymer (FRP) repair methods. After receiving

his Master of Engineering degree in May 2004, he will pursue a career in structural

design.