Journal of Composites for Construction

This paper presents the test results of an experimental investigation to evaluate the buckling behavior of built-up columns of pultruded profiles, subjected to axial compression. Specimens are assembled by using four (off the shelf) channel shaped profiles of $E$-glass fiber-reinforced polymer (FRP), having similar detailing to strut members in a large FRP structure that was executed in 2009 to start the restoration of the Santa Maria Paganica church in L’Aquila, Italy. This church had partially collapsed walls and no roof after the April 6, 2009, earthquake of 6.3 magnitude. A total of six columns are characterized with two different configurations for the bolted connections joining the channel sections into a built-up strut. Test results are discussed and a comparison is made with closed-form equation predictions for flexural buckling resistance, with buckling resistance values established from both eigenvalue and geometric nonlinear finite element analyses. Results show that there is a significant role played by the end loading condition, the composite action, and imperfections. Simple closed-form equations overestimate the flexural buckling strength, whereas the resistance provided by the nonlinear analysis provides a reasonably reliable numerical approach to establishing the actual buckling behavior.

The flexural strengthening of concrete structures by the near-surface-mounted (NSM) method has proven to be a reliable alternative to the existing externally bonded (EB) method using carbon fiber–reinforced polymer (CFRP) composites. However, the NSM method has not been used as widely as the EB method, and many factors influencing the effectiveness of NSM-CFRP bars still need to be investigated and clarified. Therefore, a total of eight concrete beams were constructed and tested. The effects of different parameters, including the number of NSM-CFRP bars, prestressing in NSM-CFRP, and the type of end treatment—for reduced/delayed end bond failure—on the performance of the strengthened beams were investigated. The test results demonstrated that the use of NSM-CFRP bars is effective in increasing the flexural capacity of concrete beams. However, when the number of CFRP bars was increased from one to two, only the postyielding cross-section stiffness of the beam displayed a better performance. The prestressed beams exhibited a higher cracking load and a higher yield load compared with the nonprestressed beams. The prestressing considered in this regard was 40% of the ultimate capacity. Three end treatments were considered: $U$-shaped steel-plate hoops, expanded groove at the ends of the bar, and reduced resin modulus at the bar ends when prestress is released. Of these, the end treatment using a $U$-shaped steel-plate hoop developed in this article was simple and effective and was capable of increasing the ductility of both prestressed and nonprestressed beams. By addressing the effects of the aforementioned factors on the NSM strengthened beams, this paper offers valuable information for the increased implementation of this technique.

Fiber-reinforced polymer (FRP) overlays have been successfully used in the aerospace industry to repair fatigue damage in aluminum plates. With this success there is potential for use of similar FRP overlays to repair fatigue damage in aging steel bridge structures. This study investigated the effectiveness of repairing fatigue damage in steel plate with adhesively bonded carbon fiber–reinforced polymer (CFRP) overlays. A total of 15 steel plate specimens with preexisting fatigue cracks were repaired with varying thicknesses of CFRP overlays to evaluate the effect of the ratio of axial stiffness of the composite to that of the underlying steel, the axial stiffness ratio (SR), on increased fatigue life and decreased applied stress. The results showed that increasing the axial stiffness ratio from 0 to 0.4 could increase the fatigue life by a factor of 10 for the most extreme conditions, and with an optimal axial stiffness ratio infinite fatigue life may be reached. Fatigue life of the steel specimens in this study was found to be dependent on both axial stiffness and applied stress range. Results from finite-element analyses validated the use of axial stiffness as a design parameter and correlated to the experimental results discussed.

The prestressed fiber-reinforced polymer (FRP) strengthening technique has been demonstrated to be an effective method to improve the effectiveness of FRP materials and the performance of the strengthened members under normal service conditions. Resin-impregnated FRP plates have been the primary form of prestressed strengthening material in many applications. However, FRP plates are somewhat inflexible, which makes it difficult to control the bond quality between the FRP plates and imperfect concrete surfaces in field applications. Using dry fiber sheets instead of FRP plates is an effective solution to this problem. However, the prestressing control stress of dry fibers is much lower than that of the corresponding FRP laminates with impregnated resin. A method using partially impregnated carbon-basalt hybrid fiber sheets (CBHFS) is proposed in this paper to improve the tensile capacity of dry fiber sheets. The following parameters of the fiber sheets were tested: fiber hybridization ratio, partial impregnation, and specimen length. The test results show that the tensile capacity of dry fiber sheets can be enhanced effectively and that it is not influenced by the specimen length when fiber hybridization and partial impregnation are applied together. Furthermore, the strengthening effects of prestressed CBHFS were verified by externally bonded RC and prestressed concrete (PC) T-beams. Vacuum-assisted resin transfer molding (VARTM) technique was applied in the tests to improve the impregnation and bond quality of the fiber sheets. Test results indicate that the prestressed strengthening of externally bonded CBHFS can significantly improve the flexural performance of RC beams under normal service conditions compared with traditional FRP strengthening techniques.

Two different prerepair loading histories were simulated in $\mathrm{3,000}\times 300\times 150\mathrm{mm}$ reinforced concrete beams, namely cracking within the elastic range, and overloading in the plastic range. After unloading, the beams were repaired with either high or ultrahigh modulus (210 or 400 GPa) carbon fiber–reinforced polymer (CFRP) plates, or a hybrid system, and then reloaded to failure. It was shown that the level of preexisting damage has insignificant effect on strengthening effectiveness and failure mode at ultimate. The 210 and 400 GPa CFRP of reinforcement ratio ${\rho }_f=0.17\%$ increased the yielding load by 33 and 60%, respectively, whereas the ultimate strength was increased by up to 29 and 51%, respectively. Doubling ${\rho }_f$ of the 400 GPa CFRP to 0.34% resulted in a 63% overall gain in flexural strength, which is only an 8% increase over ${\rho }_f=0.17\%$, because of change in failure mode from rupture to concrete cover delamination. The beam retrofitted by hybrid CFRP showed remarkable pseudoductility and warning signs. The parametric study revealed a critical balance in proportioning the areas of hybrid CFRP to achieve reliable pseudoductility. A maximum of 30% ${\rho }_f$ of 400 GPa CFRP is recommended. A higher percentage would lead to a final ultimate load lower than the first failure peak load.

The paper presents experimental work to determine the load-bearing capacity, strain responses up to failure, and failure mode of pultruded carbon fiber–reinforced polymer (CFRP) strips with different deviation saddle geometries. Several tests with different deviation angles and deviation saddle radii were carried out. The results show that the tensile strength of the CFRP strip is reduced significantly by deviation, even if the deviation radius is comparatively large in relation to the thickness of the CFRP strips. Further tests were carried out to determine the influence of other important parameters on the load-bearing capacity of the strip such as surface properties of the saddle. With the help of a polyethylene layer (PE-HD) between the CFRP strip and the deviation saddle, friction could be reduced. This leads to an increase in the breaking strength of the CFRP strip. The effect of these parameters on the load-bearing capacity of the CFRP strip is presented in this article.

External application of fiber-reinforced polymers (FRPs) for the seismic retrofit/repair of reinforced concrete (RC) beam–column joints has been extensively investigated in the last decade. However, the majority of the suggested FRP schemes follow the application of the composite sheets on the webs of the beam and the joint core which would be impeded in real three-dimensional structures due to the presence of cross beams and slabs. In addition, upgrading code-compliant RC joints using a practical FRP scheme needs to be scrutinized in detail. A desirable retrofitting scheme for this type of joints may follow the application of composite sheets on the top and bottom sides (flanges) of the beam and columns aimed at increasing the flexural strength of beam–column joints. However, the major issue of concern for a flange-bonded FRP is providing adequate development length from the critical section to transfer the FRP tensile forces from beam to columns and vice versa. To overcome this concern, a novel anchorage system for flange-bonded carbon FRP (CFRP) is developed in this study and its efficiency is evaluated through a comprehensive experimental program. In total, nine small-scale ($1/2.85$) beam–column test specimens, including two control specimens, one damaged specimen, and six retrofitted specimens were considered. Both monotonic and cyclic loading regimes were implemented to estimate, albeit approximately, the seismic responses of the test specimens. The load-carrying capacity, initial stiffness, displacement ductility, and dissipated energy of beam–column joint subassemblies are compared before and after the application of the CFRP retrofits. The results showed a remarkable improvement in the load-carrying capacity and elastic stiffness of CFRP-retrofitted specimens, thus confirming the efficiency of the suggested anchorage system. In addition, and subject to specific circumstances, the plastic hinge can also be relocated away from the beam–column interface.

The use of unreinforced masonry (URM) walls is common practice in different types of construction, such as interior or exterior walls. URM walls are typically deficient in flexural capacity when subjected to out-of-plane loading caused by high wind pressure or earthquakes. Retrofitting masonry walls with novel materials, such as fiber-reinforced polymer (FRP) composites, has shown enhancement of the flexural capacity and pseudoductility. This study evaluates the feasibility of fabric-reinforced cementitious matrix (FRCM) as an alternative external strengthening technology to improve out-of-plane behavior of URM walls. This paper reports experimental results on flexural capacity of nine clay brick walls of which six were strengthened with two different amounts of FRCM, namely one and four reinforcement fabrics. Experimental evidence shows significant improvements in the structural performance in terms of flexural capacity and stiffness of the strengthened walls. An analysis, disregarding arching effect, is conducted, and its results are compared with the experimental database. From here, the design limits are discussed. Finally, experimental data from other research programs using FRP are reported to show that when flexural capacity is related to a calibrated reinforcement ratio, the two technologies are equivalent.

It is well understood that confining concrete with fiber-reinforced polymer (FRP) composites can significantly enhance its strength and deformability. However, the confinement demand of concrete increases proportionally with its strength, resulting in higher confinement requirements for high-strength concrete (HSC). This paper reports on a study on the axial compressive behavior of FRP-confined HSC. A large experimental test database, which consists of 237 axial compression tests results for FRP-confined HSC, was assembled from the published literature and presented in this paper. This database was augmented with another database of FRP-confined normal-strength concrete (NSC), which consists of 739 test results. The combined database of 1063 test results, which cover specimens with unconfined concrete strengths ranging from 6.2 to 169.7 MPa, was used to investigate and quantify the factors that influence the compressive behavior of FRP-confined HSC. Analysis of the test results reported in the database indicates that the confinement requirement increases significantly with an increase in concrete strength, which adversely affects the observed strength enhancement through confinement. In addition, it was also observed that the hoop rupture strain of the FRP shell decreases as the concrete strength increases. Many existing stress-strain models developed for FRP-confined concrete were assessed by using the HSC database. A close examination of the results of the model assessment led to many important conclusions regarding the strengths and weaknesses of existing stress-strain models. Finally, a novel design-oriented model for FRP-confined concrete is presented that was developed on the basis of the database summarized in the paper. It is shown that the proposed model performs significantly better than any of the existing stress-strain models of FRP-confined concrete in predicting the ultimate conditions of FRP-confined HSC.

This experimental study focuses on the design and spacing of splices of fiber-reinforced polymer (FRP) stay-in-place structural forms used in concrete bridge decks. The ribbed FRP panels, which span the gap between girders, are typically of limited width, and therefore are spliced by overlapping. A total of 23 auxiliary lap splice tension tests addressed various bond systems including adhesive and/or mechanical fasteners of various diameters and spacing. Seven scaled bridge deck systems were built with FRP panels of various widths, some with bonded splices and some with no bond at all at the overlap region. The study showed that using narrow FRP plates of small width-to-deck thickness ($b/t$) ratio significantly reduces deck capacity when no splice bond is provided because narrow panels may force the punching shear crack to terminate at a splice, at a steeper angle, thereby reducing punching shear capacity. The deck with $b/t=1.9$ exhibited a 24% lower capacity than that with $b/t=7.2$. In bonded-splice decks, plate width has no effect on strength. Generally, ultimate failure loads ranged from 5.7 to 8.8 times the equivalent service load. For an overlap splice of a length equal to 4.3 times plate thickness, the optimal splice was by combined adhesive and fasteners of diameter equal to plate thickness, spaced at 1.8 times the overlap length. It developed 68% of plate tensile strength, whereas fasteners alone developed only 14–45% strength depending on their diameter and spacing.

Premature brittle failures because of seismic actions strongly affect the behavior of existing reinforced concrete (RC) structural systems. They commonly involve exterior beam-column joints of structures designed without transverse reinforcement. This paper investigates the behavior of unconfined joints that do not conform to current seismic codes and the effectiveness of externally bonded fiber-reinforced polymers (FRPs) as a strengthening technique. It presents an experimental program carried out on six full-scale RC corner joints under constant axial load and transverse cyclic loading in the as-built and FRP-strengthened configuration. After describing the specimen design strategy and test setup, seismic performance is compared. Particular attention is paid to comparing the experimental capacity of as-built joints and the capacity, which can be predicted on the basis of models available in the literature. Finally, a discussion on the effectiveness of different FRP-strengthening layouts is reported.

The confinement effect of fiber-reinforced polymer (FRP) jacket on the plastic hinge length of square RC columns is studied through experimental testing and analytical study. Seven half-scale RC square columns with different confinement ratios were tested. The strain of longitudinal reinforcement bars and the extent of yielding were measured by strain gauges mounted inside the reinforcement bars. The variation of the strain field of the external column face was recorded continuously by digital image correlation (DIC) during a test. More rational approaches of data analysis are used for identification of plastic hinge length from measured strain fields. The obtained results show that compared with unconfined RC columns, FRP jacketing increases the plastic hinge length when the confinement level is low, but reduces it when the confinement level is high. The analytical study shows that all existing models of plastic hinge length are inadequate in one way or another because no model includes all important factors. The proposed model can correctly capture the trends but it needs to be further improved for accuracy when more data are available.

A technique for reinforcing stone slabs with prestressed near-surface mounted (NSM) carbon fiber–reinforced polymer (CFRP) bars is proposed and investigated in this research. Six stone slabs were tested under monotonic loading. Among these slabs, one served as a control slab and was not reinforced, and two were reinforced with one or two NSM CFRP bars, and each bar was prestressed to 30% of the bar’s ultimate strength. The remaining three were reinforced with three NSM CFRP bars that had been prestressed to 0, 15, or 30% of the bar’s ultimate strength. The test results indicate that reinforcement with either prestressed or nonprestressed NSM CFRP bars shifted the failure mode of the stone slabs from abrupt fracture to a more ductile flexural failure. A remarkable improvement in the flexural response was obtained when the stone slabs were reinforced with prestressed NSM CFRP bars. In addition to exhibiting a higher cracking load, the stone slabs that had been reinforced with prestressed NSM CFRP bars exhibited cracks with a smaller width, and the strength utilization of the CFRP reinforcement was much higher compared to that of the nonprestressed reinforced slabs. An analytical model was developed based on a sectional analysis method to predict the flexural response of stone slabs reinforced with prestressed NSM CFRP bars. The predicted values show good agreement with the experimental results.