Journal of Composites for Construction

The scope of this paper is to describe the development of an anchor system for tension testing of unidirectional fiber-reinforced plastic bars of large diameter used as reinforcement in concrete structures. The main objective was to develop a simple, economical, and effective system to permit tensile loading of the bars up to failure in a universal testing machine. In the system suggested, each end of the bar is embedded in a conical polymeric head that fits a conical hole inside the anchoring device. In the experimental tests performed, the bars broke between the anchorages, while the resin heads remained always undamaged. These outcomes strengthen the confidence in the proposed anchor system.

Reinforcing concrete with a combination of steel and glass fiber-reinforced polymer (GFRP) bars promises favorable strength, serviceability, and durability. To verify its promise and to support design of concrete structures with this hybrid type of reinforcement, we have experimentally and theoretically investigated the load-deflection behavior of concrete beams reinforced with hybrid GFRP and steel bars. Eight beams, including two control beams reinforced with only steel or only GFRP bars, were tested. The amount of reinforcement and the ratio of GFRP to steel were the main parameters investigated. Hybrid GFRP/steel-reinforced concrete beams with normal effective reinforcement ratios exhibited good ductility, serviceability, and load carrying capacity. Comparisons between the experimental results and the predictions from theoretical analysis showed that the models we adopted could adequately predict the load carrying capacity, deflection, and crack width of hybrid GFRP/steel-reinforced concrete beams.

The effectiveness of a new structural material, namely, textile-reinforced mortar (TRM), was investigated experimentally in this study as a means of confining oldtype reinforced concrete (RC) columns with limited capacity due to bar buckling or due to bond failure at lap splice regions. Comparisons with equal stiffness and strength fiber-reinforced polymer (FRP) jackets allow for the evaluation of the effectiveness of TRM versus FRP. Tests were carried out on nearly full scale nonseismically detailed RC columns subjected to cyclic uniaxial flexure under constant axial load. Ten cantilevertype specimens with either continuous or lap-spliced deformed longitudinal reinforcement at the floor level were constructed and tested. Experimental results indicated that TRM jacketing is quite effective as a means of increasing the cyclic deformation capacity of oldtype RC columns with poor detailing, by delaying bar buckling and by preventing splitting bond failures in columns with lap-spliced bars. Compared with their FRP counterparts, the TRM jackets used in this study were found to be equally effective in terms of increasing both the strength and deformation capacity of the retrofitted columns. From the response of specimens tested in this study, it can be concluded that TRM jacketing is an extremely promising solution for the confinement of reinforced concrete columns, including poorly detailed ones with or without lap splices in seismic regions.

A strengthening technique, combining carbon fiber-reinforced polymer (CFRP) laminates and strips of wet layup CFRP sheet, is used to increase both the flexural and the energy dissipation capacities of reinforced concrete (RC) columns of square cross section of low to moderate concrete strength class, subjected to constant axial compressive load and increasing lateral cyclic loading. The laminates were applied according to the near surface mounted technique to increase the flexural resistance of the columns, while the strips of CFRP sheet were installed according to the externally bonded reinforcement technique to enhance the concrete confinement, particularly in the plastic hinge zone where they also offer resistance to the buckling and debonding of the laminates and longitudinal steel bars. The performance of this strengthening technique is assessed in undamaged RC columns and in columns that were subjected to intense damage. The influence of the concrete strength and percentage of longitudinal steel bars on the strengthening effectiveness is assessed. In the groups of RC columns of 8 MPa concrete compressive strength, this technique provided an increase of about 67% and 46% in terms of column’s load carrying capacity, when applied to undamaged and damaged columns, respectively. In terms of energy dissipation capacity, the increase ranged from 40%–87% in the undamaged columns, while a significant increase of about 39% was only observed in one of the damaged columns. In the column of moderate concrete compressive strength (29 MPa), the technique was even much more effective, since, when compared to the maximum load and energy dissipation capacity of the corresponding strengthened column of 8 MPa of average compressive strength, it provided an increase of 39% and 109%, respectively, showing its appropriateness for RC columns of buildings requiring upgrading against seismic events.

According to the available experimental work, the most common failure in existing structures strengthened by plate bonding is the laminate peeling off. In the last few years, an important effort in the development of mathematical models to avoid premature peeling failures has been made. However, a suitable and reliable design method to predict debonding due to the shear flow between crack discontinuities or at the laminate end is still not available. This paper describes a new design procedure for structures strengthened by plate bonding to avoid peeling failure at any location. After calculating the laminate area required for flexural strengthening, a two-step procedure to prevent peeling failure is proposed. The first step, to avoid peeling failure along the span, is based on a shear-bending interaction diagram associated with the theoretical maximum transferred force between laminate and support along the crack spacing before laminate debonding. This interaction diagram can be obtained through the application of nonlinear fracture mechanics. The second step consists of checking for peeling failure at the laminate end. The bonded length between the laminate end and the nearest crack should be enough to transfer the laminate tensile force acting on this crack. The proposed method has been verified with available experimental results assembled in a database. A good agreement between the experimental and predicted failure load has been obtained. Finally, an application example is presented to show the applicability of the method.

Reinforced concrete (RC) beams shear-strengthened with fiber-reinforced polymer (FRP) fully wrapped around the member usually fail due to rupture of FRP, commonly preceded by gradual debonding of the FRP from the beam sides. To gain a better understanding of the shear resistance mechanism of such beams, particularly the interaction between the FRP, concrete, and internal steel stirrups, nine beams were tested in the present study: three as control specimens, three with bonded FRP full wraps, and three with FRP full wraps left unbonded to the beam sides. The use of unbonded wraps was aimed at a reliable estimation of the FRP contribution to shear resistance of the beam and how bonding affects this contribution. The test results show that the unbonded FRP wraps have a slightly higher shear strength contribution than the bonded FRP wraps, and that for both types of FRP wraps, the strain distributions along the critical shear crack are close to parabolic at the ultimate state. FRP rupture of the strengthened beams occurred at a value of maximum FRP strain considerably lower than the rupture strain found from tensile tests of flat coupons, which may be attributed to the effects of the dynamic debonding process and deformation of the FRP wraps due to the relative movements between the two sides of the critical shear crack. Test results also suggest that while the internal steel stirrups are fully used at beam shear failure by FRP rupture, the contribution of the concrete to the shear capacity may be adversely affected at high values of tensile strain in FRP wraps.

The present paper shows and discusses some of the results obtained within an experimental investigation carried out on 15 reinforced concrete (RC) beams strengthened in shear by externally bonded fiber-reinforced plastics (FRP) sheets. The aim of the study is to analyze the influence that the geometrical percentage of transverse steel reinforcement could have on the FRP resisting action. In particular, the objectives of the experimental campaign are to explore the possible interaction between FRP and steel transverse reinforcement resisting actions, analyzing the deformation behavior of the shear resisting system (FRP, transverse steel, and concrete) and the modes of failure of the strengthened and not strengthened beams. The results of the tests in terms of shear capacity are compared to the design formulations provided by the American Concrete Institute and the National Research Council of Italy code-format recommendations.

This paper presents experimental results of reinforced concrete beams strengthened using near-surface mounted (NSM) carbon fiber-reinforced polymer (CFRP) reinforcement. A total of nine beam specimens were tested under fatigue loads. In addition, two specimens were tested for monotonic capacity. The beams were 3,500 mm long with a cross section of 254 mm deep by 152 mm wide. Different load ranges were considered in the fatigue tests to construct the fatigue life curves. The test results showed that under monotonic loading, the beam strengthened with NSM CFRP rod exhibited increases of 26 and 50% in the yield and ultimate load over the control beam, respectively. Under cyclic loading, the fatigue life for the strengthened beams was 24% higher than that of the control unstrengthened beams. An analytical model using sectional analysis and strain-life approach was developed to estimate the fatigue life of the specimens at various cyclic load ranges. A good agreement between the experimental results and analytical prediction of the fatigue life was obtained.

An approximate analysis method is proposed to determine the blast resistance of fiber-reinforced polymer (FRP)-strengthened masonry walls. The method relates the static to dynamic response by incorporating the strain rate effect on the material strength and a dynamic load factor for the applied peak load. Based on the method, 18 full-scale masonry walls reinforced with three different FRP systems were designed and subjected to field explosions, using charges of 27-ton TNT in one test and 5-ton TNT in the other. For each test, the walls were placed at three different standoff distances and orientations to the blast source. The response of the strengthened walls under blast was monitored by high-speed data acquisition systems. Post-test observations indicated no visible damage, crack, or debonding in any of the walls, thus confirming the effectiveness of the FRP retrofit technique in blast protection. The data presented are valuable for validation of analytical or numerical models.

This paper presents an analytical approach for the elastic stress analysis of monolithic circular arches strengthened with externally bonded fiber-reinforced polymer (FRP) strips. Emphasis is placed on the interfacial stresses between the existing structure and the supplemental reinforcement layers. Two analytical models are presented: The first model formulates the governing equations in terms of the displacements in the arch and the FRP strip and the tangential distribution of the shear stresses in the adhesive layer as unknowns without involving any assumptions on the stress and displacement fields in the adhesive layer. The second model uses the functional form of the displacement field derived in the first model yielding a formulation in terms of displacement unknowns only. Two numerical examples that examine the capabilities of the analytical approach are discussed. The first example focuses on the stress analysis of a strengthened arch under a localized load. The second example studies the elastic response of a partially strengthened arch to a symmetric load and a horizontal support settlement. The numerical study quantifies the interfacial shear and peeling stresses between the old and the new components underlining the stress concentrations. Finally, conclusions are presented and directions for future research on the application of the theory to masonry arches are outlined.

This paper introduces a dilatancy-based analytical model of the response of an axially loaded concrete cylinder, confined with a fiber-reinforced polymer (FRP) composite jacket. Model construction is based on the experimentally based observation that the relation between axial secant stiffness and the lateral (dilatancy) strain is effectively unique for cylinders with the same unconfined concrete strength, although the confinement levels may differ. Model development incorporates strength degradation of the concrete with dilatancy (lateral dilation); this feature allows one to demonstrate that the performance of FRP-confined concrete is consistent with the strength envelope obtained from triaxial tests. Model validation is accomplished by comparisons with existing test database and the new results on large-scale concrete cylinders. The results of the validation reveal good agreement with key response functions and parameters. The present study illustrates basic constitutive equations to model FRP-confined concrete in a more rational manner.

This paper presents a new girder consisting of a trapezoidal pultruded glass fiber-reinforced polymer (GFRP) hat-shaped section commercially available as a sheet pile, but used in this study as a structural form for concrete. It can also offer continuity in the transverse direction through a pin-and-eye connection. Five $610\text{mm}\times 325\text{mm}$ and 3,300-mm-long girders were tested in flexure to examine different bond systems, voided and solid concrete cores, and the performance in positive and negative bending. Bond systems were wet adhesive bond to freshly cast concrete, adhesively bonded coarse aggregates, and mechanical shear studs. No slip was observed between concrete and the GFRP section until delamination failure occurred within a thin layer of cement mortar that remained attached to GFRP. The studs failed by pull out from the concrete flange. In general, 47–75% of the full strengths of concrete and GFRP were reached at ultimate bond failure. Wet adhesive bonding was the simplest and quickest to apply, while resulting in a comparable strength to other systems. A “moment-curvature” analytical model, incorporating a robust bond failure criterion, was developed, validated, and used in a parametric study. It showed that varying the concrete compressive strength or thickness of the GFRP section has insignificant effect on the bond failure load. Also, there are critical values for shear span-to-depth ratio, shear strength of cement mortar, concrete strength, and width of the top GFRP flange, beyond which, the desired flexural failure mode would precede bond failure.