Journal of Composites for Construction

Existing strength models for square concrete columns confined with fiber-reinforced polymer (FRP) laminates are almost all proposed on the basis of test data from relatively small-scale specimens. The possibility of extrapolating the models to represent the behavior of real-scale square columns still needs further studies. In particular, the quantitative evaluation of size effect is still a subject that requires systematic research. To address this knowledge gap, this paper reports the results of testing on 42 relatively large-scale square concrete columns ($300\times 300\times 600\mathrm{mm}$) under axial compression. The testing demonstrated that the carbon FRP (CFRP) wrapping strain around the perimeter exhibits a certain regularity, even under very high axial load. The strain efficiency factor at corner from the large-scale columns is lower than those from small specimens, and its value shows an increasing trend as the corner radius ratio increases. Based on the available data, confinement effectiveness provided by FRP wraps was quantified as a comprehensive function considering the influence of lateral confinement level, corner radius ratio, and size effect. A database of FRP-confined square columns was assembled and used to assess the predictability of typical existing models. The results show that the existing models may significantly overestimate the strength of real-size structural columns wrapped by FRP. A modified strength model for FRP-confined square columns is also presented and is shown to have good correlation with experimental results.

This article develops an analysis-oriented stress-strain model for rubberized concrete (RuC) passively confined with fiber-reinforced polymer (FRP) composites. The model was calibrated using highly instrumented experiments on 38 cylinders with high rubber content (60% replacement of the total aggregate volume) tested under uniaxial compression. Parameters investigated include cylinder size ($\mathrm{diameter}\times \mathrm{height}$: $100\times 200\mathrm{mm}$ or $150\times 300\mathrm{mm}$), as well as amount (two, three, four, or six layers) and type of external confinement (carbon or aramid FRP sheets). FRP-confined rubberized concrete (FRP CRuC) develops high confinement effectiveness ($f_{\mathrm{cc}}/f_{\mathrm{co}}$ up to 11) and extremely high deformability (axial strains up to 6%). It is shown that existing stress-strain models for FRP-confined conventional concrete do not predict the behavior of such highly deformable FRP CRuC. Based on the results, this study develops a new analysis-oriented model that predicts accurately the behavior of such concrete. This article contributes toward developing advanced constitutive models for analysis/design of sustainable high-value FRP CRuC components that can develop high deformability.

The seismic behavior of reinforced concrete (RC) circular columns rehabilitated with sprayed-glass fiber–reinforced polymer (GFRP) composites was investigated in this study. Six large-scale RC circular columns were constructed and tested under simultaneous reversed-cyclic lateral loads and a constant axial load. All columns had a diameter of 305 mm with 1,525-mm shear span and were deficient in lap-splice length at the footing-column joint. The six specimens were divided into two identical series. In each series, one specimen was rehabilitated after being fully damaged, one was rehabilitated after being partially damaged (up to 2.5% drift ratio), and one was strengthened without prior loading. The difference between the two series was the thickness of the sprayed-GFRP confinement: 3.0 and 6.0 mm. It was concluded that the 3-mm-thick sprayed GFRP was inadequate for rehabilitation of the fully damaged column; however, it can be used for the rehabilitation of undamaged and partially damaged columns. Sprayed GFRP of 6-mm thickness not only increased the drift capacity of the column but also significantly increased the lateral load capacity of seismically deficient RC circular columns.

Although the use of glass fiber–reinforced polymer (GFRP) bars as compression reinforcement in RC columns has been extensively investigated, no research has been conducted on spliced GFRP bars under compression. This study presents an experimental program aimed at providing more insight into the effects of confinement on the behavior of spliced GFRP bars under compression. The test variables were confinement and compression-bar splice length. The test matrix included 14 circular concrete columns that were tested under concentric compression loading. Nine of the specimens were confined with GFRP spirals at two different spacings (40 and 80 mm), whereas the rest were reinforced longitudinally without transverse reinforcement. The test results are presented and discussed in terms of load–displacement behavior, failure mode, splice strength, and load–strain behavior. The postpeak behavior of the columns improved significantly with increasing levels of confinement. The bond and end-bearing contributions are described in detail. A regression analysis of the experimental results is presented, along with an analytical model proposed for predicting the strength of spliced GFRP bars under compression. The model estimates splice strength with satisfactory accuracy.

Although the seismic behavior of moment-resisting frames (MRFs) reinforced with conventional steel has been extensively studied, the performance of MRFs that are reinforced with alternative materials, such as glass fiber–reinforced polymers (GFRPs) is not fully explored. The presence of floor slabs, integrally cast with beams, is one of the main factors that affects the seismic behavior of GFRP-reinforced concrete (RC) MRFs, which is still uncovered. To fill this gap, three full-scale assemblies, one beam-column and two beam–column–slabs, were constructed and tested to failure under reversal quasi-static cyclic loading. The main test parameters were the presence of slabs and type of reinforcement (steel and GFRP). Moreover, a series of finite-element models were developed and used to investigate the effect of slab width and lateral beam size on the contribution of slabs. Test results indicated that GFRP-RC slabs, when in tension, significantly contribute to the moment capacity of beams. It was also observed that the presence of slabs increases the initial lateral stiffness of GFRP-RC MRFs, which can reduce lateral deformations of the frames during earthquakes.

Numerous investigations have addressed the size effect on the shear behavior of conventional reinforced concrete (RC) beams. In contrast, very few studies have been dedicated to the size effect of RC beams strengthened in shear with externally bonded (EB) fiber-reinforced polymer (FRP) sheets. This paper presents the results of an experimental investigation conducted on 18 RC T-beam specimens of three different sizes to study the size effect on concrete shear strength at failure of RC beams strengthened in shear with EB carbon FRP (CFRP) sheets. The specimens were divided into two groups, with and without internal transverse steel, and six series, each consisting of small, medium, and large beams as follows: two series of control (unstrengthened) specimens, two series of specimens strengthened in shear with one-ply EB-CFRP sheets in a U-wrap configuration, and two series of specimens strengthened in shear with two plies of EB-CFRP sheets in a U-wrap configuration. The results clearly revealed the presence of a size effect on both the concrete and CFRP sheet contribution to shear resistance. For strengthened specimens without internal transverse steel, the size effect was found to be magnified as the CFRP sheet stiffness increased. This was not the case for beams strengthened with transverse steel, where the size effect behavior reduced considerably due to the interaction between internal transverse steel and EB-CFRP sheets.

The flexural response of 12 2-m-long and 152-mm-diameter fiber-reinforced polymer (FRP) skinned members with foam cores was investigated. Nine cylinders with various flax-FRP (FFRP) laminate arrangements and foam core densities were fabricated and tested in flexure; these were compared with three glass-FRP (GFRP) skinned specimens. Varying the foam density from 32 to $65\mathrm{kg}/{\mathrm{m}}^3$ was found to be insignificant in FFRP skinned beams because their failure was governed by fiber tensile rupture. However, increasing the foam density increased the capacity of the GFRP skinned members that failed in compression. The most effective ways to increase strength-to-weight and stiffness-to-weight ratios of FFRP skinned cylinders are to increase the number of longitudinal fiber layers and decrease the foam core density. A FFRP cylinder with five longitudinal and three transverse layers was able to develop over 80% the strength-to-weight and stiffness-to-weight ratios as a reference GFRP cylinder, substantially higher than the 30%–50% values seen in coupon tests. Although the FFRP cylinder performed fairly well, further work on the member axial, impact, and durability resistance is recommended.

The debonding phenomenon between fiber-reinforced polymer (FRP) composites and masonry is influenced by the presence of mortar joints that could reduce the bond capacity of the FRP-brick interface (i.e., in the absence of the joints). The actual estimate of the maximum load transferable at the FRP-masonry interface is cumbersome. Thus, the aim of this paper is twofold. First, the paper provides a set of useful key remarks to understand how the presence of joints influence the behavior of the FRP-masonry interface with respect to the FRP-brick interface. This is an important aspect of the paper because it summarizes previous work and provides guidance for designers and researchers. Secondly, this paper proposes a step-by-step procedure to obtain an approximate estimate of the maximum transferable load at the FRP-masonry interface provided that the characteristics of the FRP-brick and FRP-mortar interfaces are known or can be derived from codes and guidelines, and the thickness of brick and joints is assigned. To the best knowledge of the authors, an approximate formulation for the FRP-masonry interface is presented in this paper for the first time together with guidance to evaluate when an approximate formulation might be necessary based on the desired acceptable error. The comparison between the maximum load transferable at the FRP-masonry interface and its approximation presented in this paper provides a relative error of roughly 0.4% for case of practical interest. Finally, the paper provides an example application of the proposed procedure.

This paper presents a high-capacity anchorage system for multitendon carbon fiber–reinforced polymer (CFRP) cables and the associated construction technology. Four groups of cable anchorage systems used 37 CFRP tendons, each with a diameter of 4 mm, and extensively investigated their behavior through static and fatigue tests. In the static test, the synchronous performance of the cables was studied. The mechanical properties of the cables in the anchor zone were compared with the results from a finite-element (FE) analysis, and the anchoring efficiency also was evaluated. Furthermore, the fatigue behavior of cable anchorage system was verified. The relative displacement of each component, distribution of the shear stress, retention rate of the stiffness, and strength of the cables were discussed. The results indicate that the cable exhibits excellent synchronous tensile performance. The stress and displacement distributions of the cable in the anchor zone were in good agreement with the results from the FE analysis. The stress concentration was greatly reduced by the variable stiffness of the load transfer medium via layered casting technology, and the anchoring efficiency reached 97%. The cable anchorage system sustained 2 million cycles without any macroscopic cracks forming in either the cable or anchorage at a maximum stress of $0.45f_s$ (the standard strength value) and a stress range of 200 MPa. The residual strength and stiffness of the cable reached 95% and 96%, respectively, compared with the initial behavior.

An American Concrete Institute (ACI) design guide for fiber-reinforced polymer (FRP)–reinforced concrete recommends that, in order to avoid creep-rupture failure, service-load sustained stress should not exceed 0.20 times the design tensile strength $f_{fu}$ of glass-fiber-reinforced-polymer (GFRP) reinforcement, where $f_{fu}$ is equal to the guaranteed tensile strength multiplied by an environmental reduction factor. The 0.20 value, which was put in place prior to 2001, was conservatively chosen due to both limited test data and a lack of standards for both materials and test methods. In this study, values for the creep-rupture strength at a 1,000,000-h (114-year) endurance time were extrapolated from recent creep-rupture tests of GFRP bars based on regression analysis of rupture stress versus the logarithm of endurance time in hours. The bars were made of continuous E-CR or E-glass fibers in a vinyl ester resin and manufactured by the pultrusion process. The analyses indicate that, given a 114-year service life and a 1.67 safety factor as is used by the ACI design guide for FRP-reinforced concrete, a sustained stress limit of $0.30f_{fu}$ is a conservative value to avoid a creep rupture of GFRP reinforcing bars that conform to ASTM standards.