Journal of Composites for Construction

This paper evaluates the effectiveness of a new type of shape memory alloy (SMA)/fiber-reinforced polymer (FRP) patch for repair of fatigue-sensitive steel elements. Twenty-seven single edge-notched steel coupons with different configurations of reinforcement were tested under tension-tension fatigue loading at three different stress ranges up to failure. Test results indicated that the coupons that were reinforced with the SMA/carbon FRP (CFRP) composite exhibited average fatigue lives that were 26.4 and 15.3 times those of the unreinforced coupons at stress ranges of 155 and 217 MPa, respectively. In comparison, coupons that were reinforced with CFRP only and those reinforced with only SMA wires exhibited average fatigue lives that were 8 and 1.7 times those of the unreinforced coupons, respectively, at a stress range of 155 MPa. This suggests that there is a synergistic effect between the prestressing forces provided by the SMA wires and the crack bridging provided by the CFRP. The results indicate that these SMA/FRP patches are a promising technology for rehabilitation of fatigue-sensitive steel structures.

Fiber-reinforced polymer (FRP) laminates used as externally bonded reinforcement (EBR) to strengthen or repair concrete structural members have proven to be an economical retrofit alternative. However, when strengthening long-span members with limited access (e.g., over waterways and freeways), labor and equipment demands may hinder the use of continuous EBR FRP. Recently, carbon FRP (CFRP) rod panels (CRPs) have been developed and used to overcome these limitations. Each CRP is made of several small-diameter CFRP rods placed at predefined spacing. Several CRPs are brought together and made continuous to achieve the strengthening length by means of an overlap (or finger joint). This study experimentally investigated the bond behavior between CRP and concrete. Eighteen double-lap shear specimens were tested under pull-off loading to evaluate the bond strength, development length, transfer mechanism, and bond-slip relation. The bond strength and development length were established for two CRPs, CRP 070 [generated from rods of 2.00 mm (0.08 in.) in diameter, spaced at 6.25 mm (0.25 in.)], and CRP 195 [generated from rods of 4.00 mm (0.16 in.) in diameter, spaced at 9.35 mm (0.38 in.)]. The development length measured 100 mm (4.00 in.) and 119 mm (4.75 in.) for CRP 070 and CRP 195, respectively. The bond strength per unit width of CRP was $563\mathrm{kN}/\mathrm{m}$ ($38.5\mathrm{kip}/\mathrm{ft}$) for CRP 070 and $712\mathrm{kN}/\mathrm{m}$ ($48.8\mathrm{kip}/\mathrm{ft}$) for CRP 195.

Fiber-reinforced polymer (FRP) wrapping has been extensively studied for strengthening reinforced concrete columns. However, few studies were found on the behavior under small eccentric loads, a frequent situation in real-world applications. In this paper an experimental, numerical, and theoretical study on the influence of small eccentricity of loading in FRP-confined prisms and cylinders is shown. Thirty-six short concrete specimens of different cross sections were tested under concentric and eccentric loads. Unwrapped specimens as well as specimens wrapped by one or two layers of FRP were tested. All specimens showed increases in axial strength and ultimate strain due to wrapping with FRP. The lateral displacement was also measured for all specimens until failure. A simple finite-element model was used to predict the axial strength and axial and lateral displacement. Simplified theoretical procedures were also used to generate axial load–bending moment interaction diagrams. The accuracy of both numerical and theoretical procedures to predict the experimental behavior depended significantly of the confinement model used.

A major cause underlying the collapse of reinforced concrete (RC) moment-resisting frames observed in past earthquakes is the shear failure of nonseismically detailed exterior beam–column joints. A number of studies have investigated the external application of fiber-reinforced polymer (FRP) composites for the seismic rehabilitation of these members. However, the main challenge facing this technique is the premature debonding of FRP composites off the concrete substrate posed. The grooving method (GM), as an alternative to the conventional externally bonded reinforcement (EBR) technique, has yielded promising results in postponing, or in some cases eliminating altogether, the undesirable debonding failure mode in the flexural/shear strengthening of beams. The present experimental study was conducted to investigate the effects of GM in the seismic behavior of exterior RC beam–column deficient joints strengthened with carbon-FRP (CFRP) sheets taking advantage of the special technique of externally bonded reinforcement on grooves (EBROG). Another facet of the study is the investigation of the performance of FRP fans in preventing the splitting failure of the concrete cover. For the purposes of this study, six half-scale RC beam–column subassemblies with no transverse reinforcement in the joint region were constructed. Results revealed that the EBROG technique, coupled with FRP fans at the termination point of FRP sheets, was able to eliminate altogether the debonding. It was, further, able not only to delay the brittle shear failure of the beam–column joints but also to relocate the beam plastic hinge away from the column interface under specific circumstances. Moreover, it was found that application of CFRP composites via the EBROG method was able to produce a significant enhancement in the seismic capacity of the test specimens in terms of their strength, stiffness, ductility, pinching width ratio, and energy dissipation.

This paper presents the experimental results of five full-scale glass fiber–reinforced polymer (GFRP) reinforced concrete (RC) columns with lap splices under axial and cyclic-reversed loads. One additional reference column was reinforced with steel bars and stirrups for comparison purposes. The column specimens had a 350-mm square cross section with 1,650-mm shear span. Each specimen represented the lower segment of a first-story column between the footing and the point of contraflexure. The test parameters included type of reinforcement, lap splice length of longitudinal reinforcement and transverse reinforcement spacing. Test results indicated that a splice length of 60 times the diameter of the longitudinal column bar was adequate in transferring the full bond forces along the splice length. In addition, lap-spliced GFRP-RC columns with closely spaced transverse reinforcement showed stable hysteresis response and achieved high levels of deformability, which far exceeded the limitations of the North American building codes.

The use of fiber-reinforced polymers (FRPs) has become common in many civil engineering applications, including prestressing applications. Glass FRP (GFRP) bars are less expensive than carbon FRP (CFRP) bars, but they have high relaxation properties and are susceptible to creep rupture, which has limited their use in prestressed applications. A new generation of GFRP bars has been recently manufactured with improved mechanical properties that make them suitable for prestressed applications. Much research has been conducted on the short-term behavior of concrete members prestressed/reinforced with GFRP tendons, but relatively little is known about the long-term behavior of these members. This paper presents the results of testing 12 beams as part of an ongoing program to investigate the long-term behavior of GFRP prestressed concrete beams. Three beams were tested to failure in a monotonic four-point loading test, and nine beams were subjected to sustained loading for 300 days. The results showed that the sustained loading for 300 days did not affect the beams’ ultimate capacity, which suggests that creep or relaxation of the GFRP bars was not significant and did not affect the strength of the beams. It was concluded that the prestressing decreased the total and long-term midspan deflection and the total and the long-term concrete strains, mainly at the lower sustained load levels.

To address the issue of corrosion of steel in concrete structures, glass fiber reinforced polymer (GFRP) reinforcement is slowly gaining acceptability as a replacement of steel. This paper aims to evaluate the feasibility of using GFRP as internal lateral reinforcement in columns and presents the experimental results of seven circular columns that were tested under simulated earthquake forces which included constant axial load and cyclic lateral displacement excursions. In addition to the moment versus curvature response and shear versus deflection behavior, a number of ductility parameters related to curvature and displacement, and work and energy dissipation are used to evaluate the performance of specimens. Due to their linear elastic behavior until rupture at a strain of approximately 0.02, GFRP spirals provided continuous confinement to the columns, which displayed large deformability. The application of GFRP as lateral reinforcement in concrete columns minimizes the corrosion concerns and delivers a strong and ductile column response.

Fiber-reinforced polymer (FRP) jacketing/wrapping has become a widely accepted technique for strengthening/retrofitting reinforced concrete (RC) columns. Although extensive research has been conducted on FRP-confined concrete columns under concentric compression, leading to many stress–strain models, the applicability of these concentric-loading models in the analysis of columns under eccentric loading has not been properly clarified. This paper presents the development and application of a reliable three-dimensional (3D) finite element (FE) approach for an in-depth investigation into this problem. In the proposed FE approach, an accurate plastic-damage model recently developed by the authors’ group for concrete under a general state of multiaxial compression is employed. The accuracy of the proposed FE approach is demonstrated through comparisons with available tests. Numerical results from the verified FE approach are then presented to gain an improved understanding of the behavior of confined concrete in such columns (e.g., distributions of axial stresses and confining pressures), leading to the conclusion that the direct use of a concentric-loading stress–strain model in the analysis of an eccentrically-loaded column may lead to significant errors.

This paper focuses on multibolted pultruded fiber reinforced polymer (FRP) plates in all FRP beam-to-beam connections under both static and cyclic loads. Thanks to the experience accumulated from a previous study concerning the behavior of joints under monotonic loading, the goal of this work was to investigate the behavior of plates connected through bolts to a vertical and a horizontal beam. The experimental tests demonstrated that multibolted FRP pultruded plates have a dominant influence on the response of connections and, in particular, on their collapse. In detail, damage only affected the multibolted plate during the tests, especially near bolt holes, whereas structural profiles remained substantially intact. As a consequence, the nonlinear behavior of joints was affected strongly by progression of the damage in the multibolted plate. The innovative use of multibolted pultruded plates, instead of web and flange cleats, for the realization of bolted beam-to-beam connections is a challenge, and the aim of this work was to understand the real behavior and failure modes of these FRP plates in the ultimate limit state.

Existing design codes and guidelines do not adequately address the design of concrete columns reinforced with fiber-reinforced polymer (FRP) bars. Accordingly, a number of research studies investigated the behavior of FRP bar-reinforced concrete columns. However, the previous studies were limited to FRP bar-reinforced normal-strength concrete (NSC) columns. In this study, the behavior of glass fiber-reinforced polymer (GFRP) bar-reinforced high-strength concrete (HSC) specimens under different loading conditions was investigated in terms of axial load-carrying capacity, confinement efficiency of the GFRP helices, as well as the ductility and post-peak axial load-axial deformation response. The effects of the key parameters such as the type of the reinforcement (steel and GFRP), the pitch of the transverse helices, and the loading condition (concentric, eccentric, and four-point loading) on the performance of the specimens were investigated. It was observed that the GFRP bar-reinforced HSC specimen sustained similar axial load under concentric axial compression compared to its steel counterpart, but the efficiency of GFRP bar-reinforced HSC specimens in sustaining axial loads decreased with an increase in the axial load eccentricity. Direct replacement of steel reinforcement by the same amount of GFRP reinforcement in HSC specimens resulted in about 30% less ductility under concentric axial load. However, it was found that the ductility and post-peak axial load-axial deformation behavior of the GFRP bar-reinforced HSC specimens can be significantly improved by providing closely spaced helices.

This paper presents an experimental study carried out on unbonded posttensioned (PT) segmental precast concrete columns. In total, eight cylindrical columns were posttensioned and tested under incrementally increasing reverse cyclic loading. The columns consisted of four concrete cylindrical segments that were assembled on top of each other with dry joints. PT bars were passed through a hole cast into the center of the segments, left unbonded, and anchored to the column loading head on the top and footing on the bottom. The variables in this study were the level of posttensioning, concrete material type, and concrete confinement. Two different levels of posttensioning force of 50 and 100 kN, corresponding to an average axial stress on the concrete of 2.83 and 5.77 MPa, were applied to the columns. Conventional concrete and crumb rubber concrete (CRC) were considered to investigate the effect of concrete material on the behavior of the segmental column. The concrete confinement effect was investigated by confining the bottommost segment using fiber-reinforced polymer (FRP) wraps. The results of this study showed that FRP-confined CRC segmental columns could provide an environmentally-friendly alternative to conventional concrete columns in structural applications. The effect of the confinement on the strength of rubberized concrete was higher than that of the conventional concrete. The negative effect of the rubber particles on the strength reduction in the concrete at the structural level was much lower than that in the concrete at the material level. Increasing the column posttensioning force increased the measured peak load and decreased the ultimate drifts for both unconfined and FRP-confined specimens. This paper also concluded that in unbonded PT columns in order to ensure a fully self-centering behavior, the level of axial stress ratio must be limited.

This study presents the shear behavior of reinforced concrete (RC) T-beams with innovative steel stirrups hybridized in the longitudinal direction with a fiber-reinforced polymer (FRP) composite. Six beams were experimentally tested: three beams were reinforced with the hybrid steel-FRP stirrups, and the others served as control samples and were reinforced with conventional steel stirrups. Furthermore, a two-dimensional finite-element (FE) model was created and executed using FE analysis software to examine the effect of several influential parameters, including the type and amount of FRP used in producing the hybrid stirrups. Large-scale beams reinforced with carbon FRP (CFRP) stirrups were numerically simulated before and after replacing the transverse CFRP reinforcement with steel-FRP stirrups. Compared with conventional RC beams, concrete beams reinforced with steel-FRP stirrups successfully showed a considerable increase in the beam shear strength and deformability. Moreover, hybrid steel-FRP stirrups can provide design engineers with a new, flexible design to control both the structural response and the construction cost.

The bond performance of fiber-reinforced polymers (FRPs) at elevated temperature depends upon the glass transition temperature of the matrix material. Included herein is an overview of the results of an experimental program in which postcuring of carbon fibre reinforced polymer (CFRP)-strengthened reinforced concrete beams was evaluated to observe changes in performance at elevated temperatures. Fourteen beams were tested in four-point bending. The testing schemes included 1 unstrengthened control beam, 2 room-temperature strengthened beams, and 11 high-temperature tests. The results of this experimental program have demonstrated that postcuring an FRP system can be an effective means of increasing the performance of FRP-strengthened beams when subjected to elevated temperatures, even those well in excess of the glass transition temperature of the FRP.

Throughout this paper, a new system for connecting composite sandwich wall panels is proposed. The relevant structural components are investigated with the aim of utilizing these panels as insulated wall elements in building applications or prefabricated modular systems. The adopted sandwich wall panels are composed of hand-layup glass fiber–reinforced polymer (GFRP) outer skins and low-density closed polyurethane (PU) foam core. The sandwich wall panels present an overall geometry of $\mathrm{2,880}\times 960\times 64{\mathrm{mm}}^3$. One challenge of the proposed new system that was examined included joining the panels in the longitudinal direction (along their height) and transversally connecting (along their width) to other structural elements, similar to beams at the bottom and top. The structural performance of the sandwich wall panels was experimentally tested and thereafter analytically assessed in two cases: (1) single wall panels; and (2) two jointed wall panels. Outward localized GFRP wrinkling, followed by global buckling was observed as the dominant failure mode in both cases. Further, the capability of the proposed connection system to increase the axial load capacity of the jointed panels was evaluated. The study illustrates that axial capacity of two jointed sandwich wall panels, compared to the single sandwich wall panel, increased substantially from 91 to 152% depending on the failure modes.

Despite their advantage of creating stronger fiber-reinforced polymer (FRP)-concrete bonds, methods such as the externally bonded reinforcement (EBR) often pose the serious problem of premature debonding of FRP sheets off the concrete substrate. Externally bonded reinforcement on grooves (EBROG) has been recently introduced as a more efficient alternative to combat such drawbacks by either delaying or totally eliminating the debonding of FRP sheets off the concrete substrate under normal conditions. However, the impacts of adverse environmental conditions on FRP-concrete bond strength have not ever been investigated in FRP-concrete bond through EBROG technique. This study was especially designed to investigate the effects of environmental conditions, including three alkaline solutions at temperatures of 23, 40, and 60°C on FRP-concrete bond strength. For the purposes of this study, specimens were strengthened with either the EBR or the EBROG method. Both types of specimens were subsequently maintained in the experimental alkaline conditions for 3,000 h. Single-shear tests were conducted to evaluate the FRP-concrete bond behavior using the particle image velocimetry technique. Experimental results showed that the specimens strengthened by the EBROG method and subjected to the adverse conditions of an alkaline environment resisted higher bond loads by up to 50% their ultimate bond loads compared to those strengthened via the EBR method.

Restoring the capacity of deficient structural members to meet original design requirements or new higher demands is common. For that purpose, research efforts to use composite materials in retrofitting steel structures have been gradually gaining ground in recent years, with more focus on inhibiting buckling modes by increasing local and/or global member stiffness. In this paper, the authors experimentally investigate retrofitting alternatives for shear-deficient steel structures. The proposed alternatives fall on what is referred to as strengthening by stiffening (SBS), in which thin-walled steel beams are retrofitted by bonding pultruded glass fiber reinforced polymer (GFRP) sections to the buckling-prone web panels. Different GFRP stiffener orientations, types, and adhesives were investigated in addition to conventional welded steel stiffeners, and bonded biaxial carbon fiber reinforced polymer (CFRP) sheets were tested in this study. The experimental results showed that gains in shear strength of up to 81% compared to non-retrofitted beams were achieved with bonded CFRP sheets, whereas an increase of about 50% in load capacity was achieved for bonded GFRP and welded steel stiffeners. The energy ductility of the retrofitted beams was found to be higher when a ductile epoxy type was used. Recommendations regarding the number of GFRP stiffeners, their orientation, type, and adhesive properties were drawn based on the findings from this paper. The results also showed that it may be possible to use SBS in new construction to eliminate conventionally welded steel stiffeners, which is known to introduce stress concentrations and their associated fatigue problems.

This paper focuses on the bonding performance of the interface between fiber-reinforced polymer (FRP) profiles and concrete. The bond shear capacity and the bond shear strength of the wet-bond (WB) interface, the shear-key (SK) interface, and the shear-key combined with wet-bond (SK-WB) interface were investigated via double-lap push shear tests with 15 specimens. Varying numbers of shear keys were adopted in the SK and SK-WB specimens to study the influence of the bonding area of the shear keys. The slip between the FRP profiles and the concrete was measured, and the bond stresses at different points of the test zone were calculated with differential operations. The results show that the average bond strength and the maximum slip of the SK-WB interface are significantly higher than those of the SK interface, whereas the average bond strength and maximum slip of the SK interface are significantly higher than those of the WB interface. A much greater concentration of stress was discovered in the SK interface compared with the SK-WB interface. The capacity load value and the maximum slip value of the double-SK interface are higher than those same values of the single-SK interface. The comparison of bond strength was performed between the FRP profiles-concrete interface and the FRP plate-concrete interface. The results show that the bond strength of the WB interface between the concrete and the FRP profile by the double-lap push shear method is much lower than the bond strength of the WB interface between the concrete and the FRP plate when using the double-lap pull shear method. Finally, the significant difference in the bond strengths is explained using the stress analysis method.

Glass fiber-reinforced polymer (GFRP) composite fender systems, including FRP columns, protect vehicles from colliding with bridge piers. The FRP columns are either hollow (flexible) or filled with concrete (rigid). This paper presents the experimental and analytical results of tests with hollow and concrete-filled GFRP tubes under impact loads. The impact responses are evaluated as a function of fiber volume fraction and energy absorption, in addition to other common structural parameters. The test results reveal that increasing the longitudinal fiber volume fraction up to approximately 60% decreases deformation or failure of both the hollow and concrete-filled GFRP tubes and has an insignificant effect on the response of peak impact and energy absorption. This may be attributed to an inadequate bond strength between concrete and FRP at failure, or failure may be driven by local effects. Hollow GFRP tubes were damaged more seriously than concrete-filled GFRP tubes, and the peak impact loads of hollow tubes were lower by approximately 80% than those of concrete-filled GFRP tubes. However, the energy absorption capacities of hollow GFRP tubes were nearly identical to those of concrete-filled GFRP tubes under the same magnitude of impact energy. A three-dimensional finite-element model was developed to simulate the impact behavior of two types of GFRP tubes, and the numerical results agreed well with the test data. Analytical solutions from the Euler–Bernoulli beam equation for the lateral displacement of beams under sinusoidal impact loads were obtained for comparison with experimental results. These results showed that the maximum displacement could be accurately predicted by the proposed theoretical model.

This paper presents an experimental study on concrete cylinders confined by fiber-reinforced polymer (FRP) and loaded under constant axial load and increasing bending moment. A total of 55 cylinders were tested with FRP thickness and axial force ratio being the variables. Compared with another load path of constant load eccentricity and increasing axial load, the response of the cylinders under this loading scheme was very different. It is concluded that load path has a significant effect on the stress-strain behavior of FRP-confined concrete and different stress-strain models are required for FRP-confined concrete under different load paths. Based on the test results, a stress-strain model is proposed for FRP-confined concrete under the relevant load path.

The main objective of this paper is to develop closed-form equations for estimating the elastic local buckling capacity of pultruded fiber-reinforced polymer (FRP) structural shapes. This includes I-sections, channels, angles, and Z-sections that are loaded either in bending or axial compression. A comprehensive parametric study is conducted to investigate the effect of each of the parameters of the problem. The results of the parametric study were used to conduct a regression analysis to develop closed-form equations for estimating the elastic local buckling capacity for each of the shapes considered. The developed equations were validated by using them to estimate the local buckling capacity of a number of sections and comparing the results with the exact elastic analysis. Furthermore, the equation results were compared with some reported experimental data. The developed equations were successful in estimating the elastic local buckling capacity with error values that did not exceed 2%, and their results were generally in good agreement with the experimental data. A numerical example is given to illustrate the use of the proposed equations.

Strengthening of unreinforced masonry walls is essential to mitigate their brittle failure during natural hazards like earthquakes. An experimental program was conducted to evaluate the diagonal shear and out-of-plane flexural strength of masonry walletes strengthened using glass fabric–reinforced cementitious matrix (FRCM). The performance of walletes strengthened in two modes of wet lay-up application: direct and sandwich, which differ in the sequence of placing the fabric, was studied. Also, the contribution of mechanical anchors was investigated in enhancing the performance of masonry subassemblages. Test results showed that with the provision of FRCM, the strength and ductility of masonry could be significantly improved under both shear and flexural loads, and the brittle collapse could be mitigated. Both methods of FRCM application showed comparable results, and it was observed that the mechanical anchors were helpful in attaining higher deformability and in containing the broken fragments of masonry. Finally, the strength parameters obtained from these tests were compared with those predicted using existing analytical models for strengthened masonry walls.

In this paper, the mechanical performance of concrete beams strengthened by an impregnated carbon fabric–reinforced cementitious matrix (CFRCM) composite is investigated. The study is aimed at the rehabilitation of the Finale Emilia hospital roofs, which were severely damaged by the 2012 northern Italy earthquake. An 8-m-long concrete beam was taken from the building for reinforcement and testing in a beam test setup. The composite is designed to be externally applied to the existing thin clay tile layer bonded to the concrete beam intrados. Two lamination cycles, which differ by the way in which the partially organic adhesion promoter is applied to the fabric, are considered. It was found that impregnation through fabric immersion provides a 1.5-fold increase in the ultimate strength of the strengthened beam compared to expedited impregnation with a brush and that clay tiles make a very good supporting substrate, to the extent that cohesive fracture at the tile–concrete interface takes place on the verge of concrete failure near the hinge zone. Conversely, expedited impregnation of the carbon fabric with the adhesion promoter was unable to provide adequate fabric–matrix adhesion and led to delamination failure. Estimates of the adhesion strength, optimal bonded length, and of the composite, as well as of the concrete strain at failure, are provided.