Carlos Chastre

Reinforced concrete (RC) beams externally strengthened with glass fiber-reinforced polymer (GFRP) strips bonded to the soffit may see their load-carrying capacity reduced due to environmental conditions—especially due to the deterioration of bond between the adhesively bonded laminates and concrete, causing premature failure.
More research has been published on the detachment of the laminate progressing from the anchorage zone than on failure induced by the formation of flexural or shear-flexural cracks in the midspan followed by fiber-reinforced polymer (FRP) separation and failure designated as intermediate crack (IC) debonding. An experimental program to study degradation of the GFRP laminate beam specimens after accelerated temperature cycles, namely: 1) freezing-and-thawing type; and 2) cycles of the same amplitude (40°C [104°F]) and an upper limit approximately 70% of the glass vitreous transition temperature of the resin, Tg, is described.
Effects on the bond stress and ultimate capacity are reported. Substantial differences between shear and bending-induced failure and a decrease of bond stresses and engagement of the laminates on the structural response are analyzed.

Despite the fact that FRP composites are a reliable structural material with reasonable durability performance, the environment to which the strengthened structure is exposed can make the strengthening system vulnerable. In this study, the effectiveness of Externally Bonded Reinforcement (EBR) systems when external compressive stresses are applied to glass (G) FRP-to-concrete interfaces in several aggressive environments is analysed. The compressive stress imposed on the GFRP-to-concrete interface intends to simulate, for instance, the effect produced by a mechanical anchorage system applied to the EBR system. The design and the region to set those mechanical anchorage systems are not yet well understood and are mostly applied without really knowing how they will behave. This work shows an exhaustive experimental programme based on several double shear tests subjected to salt fog cycles, dry/wet cycles and two distinct temperature cycles: from -10ºC to +30ºC and +7.5ºC to +47.5ºC. The Mohr-Coulomb failure criterion was found to provide a good representation of the performance of the GFRP-to-concrete interface, and changes of cohesion and internal friction angle of those interfaces during the hours of exposure to the aggressive environments are reported.

The retrofit of reinforced concrete columns with FRP jackets has received considerable attention in recent years. The advantages of this technique compared to other similar techniques include the high strength-weight and stiffness-weight ratios of FRP (Fibre Reinforced Plastics), the strength and ductility increase of RC columns confined with FRP jackets as well as the fact that FRP external shells prevent or mitigate environmental degradation of the concrete and consequent corrosion of the steel reinforcement. Furthermore, this method also reduces the column transversal deformation and prevents the buckling of longitudinal reinforcement. Twenty five experimental tests were carried out on reinforced concrete columns confined with CFRP composites, and subjected to axial monotonic compression. In order to evaluate the influence of several parameters on the mechanical behavior of the columns, the height of the columns was maintained, while changing other parameters: the diameter of the columns, the type of material (plain or reinforced concrete), the steel hoop spacing of the RC columns and the number of CFRP layers. Predictive equations, based on the experimental analysis, are proposed to estimate the compressive strength of the confined concrete, the maximum axial load and the axial or the lateral failure strain of circular RC columns jacketed with CFRP. A stress-strain model for CFRP confined concrete in compression, which considers the effect of the CFRP and the transversal reinforcement on the confined compressive strength of the column is also proposed. The curves, axial load versus axial or lateral strain of the RC column, are simulated based on the stress-strain model and include the longitudinal reinforcement effect. The results demonstrate that the model and the predictive equations represent very well the axial compression behavior of RC circular columns confined with CFRP. The applicability of this model to a large spectrum of RC column dimensions is its main advantage.

The durability of adhesively bonded fiber-reinforced polymers (FRP) and concrete substrates has been the subject of recent studies. The degradation of bonded interfaces conjugated with other factors that affect the interface strength may compromise the potentialities of using FRP in externally bonded reinforced (EBR) concrete structures. However, the estimation of the effects of degradation on these bonded interfaces and the analytical methodologies to quantify them are not fully understood. The present work focuses on a local bond-slip model characterized by two parameters for which the values are obtained experimentally. Then, the determination of the local bond-slip relationship of a glass (G) FRP-to-concrete interface can be estimated. The assessment of the degradation of the bonded interface when subjected to cycles of (1) salt fog; (2) wet-dry environments with salt water; (3) temperatures between −10°C and +30°C; and (4) temperatures between +7.5°C and +47.5°C is presented. The results obtained using the proposed bond-slip model led to the conclusion that after 10,000 h of exposure to temperature cycles between −10°C and +30°C, there was a small change in the GFRP-to-concrete interface performance, whereas the effects on the bonded interface for the specimens subjected to temperature cycles between +7.5°C and +47.5°C were far more most severe.

The influence of compressive stresses exerted on FRP-concrete joints created by external strengthening of structural members on the performance of the system requires better understanding especially when mechanical devices are used to anchor the externally bonded reinforcement (EBR). The numerical modelling of those systems is a tool that permits insight into the performance of the corresponding interfaces and was used in the present study, essentially directed to analyse the effectiveness of EBR systems under compressive stresses normal to the composite surface applied to GFRP-to-concrete interfaces. The compressive stresses imposed on the GFRP-to-concrete interface model the effect produced by a mechanical anchorage system applied to the EBR system. An experimental program is described on which double-lap shear tests were performed that created normal stresses externally applied on the GFRP plates. A corresponding bond-slip model is proposed and the results of its introduction in the numerical analysis based in an available 3D finite element code are displayed, showing satisfactory agreement with the experimental data. The results also showed that lateral compressive stresses tend to increase the maximum bond stress of the interface and also originate a residual bond stress which has significant influence on the interface strength. Also, the strength of the interface increases with the increase of the bonded length which have consequences on the definition of the effective bond length.

The paper analyses numerical solutions for the process leading to debonding failure of fiber reinforced polymers (FRP)-to-concrete interfaces in shear tests with the FRP plate subjected to a tensile load at one end. Any realistic local nonlinear bond-slip law can be used in the numerical analysis proposed in the present study. However, only a Popovics’ type expression is employed in the numerical process due to its use in different studies found in the literature. Effective bond length (Leff) is discussed and an expression depending on the Popovics’ constant (nP) is proposed to calculate it. Assuming a fracture in pure Mode II, the debonding process is analyzed in detail and distributions of bond stresses and strains in the FRP plate along the interface are presented. The load-displacement behaviour is also presented and the influence of the local bond-slip law on the debonding process is discussed.