Durability of adhesive joints between concrete and FRP reinforcement in aggressive environments
The durability of the bondline between concrete and its fiber-reinforced polymer reinforcement was characterized at various temperature and humidity levels. The bondline consisted of an epoxy primer, an epoxy putty and an epoxy saturant. In principle, fracture could occur anywhere in this bondline, but attention was focused on the concrete/primer interface in this study because preliminary experiments indicated that this was the dominant failure mechanism. The initial part of the constitutive modeling of the epoxy primer was conducted using linear viscoelastic experiments. Confined compression experiments determined two linear material functions simultaneously. Because this was a relatively new experiment, the results were validated by conducting bulk compliance experiments. The viscoelastic region of the bulk modulus was as wide as that of the tensile and shear relaxation moduli. This result contradicts previous conceptions but is agreement with some other recent observations. Thermal and hygral expansions were also measured and used in a hybrid nonlinear viscoelastic constitutive model. The hybrid model captured the hygrothermal nonlinear viscoelastic deformation of the epoxy primer. This model is a combination of Schapery’s model and Popelar’s shear modified free volume model. Torsion tests were conducted and used to calibrate the distortional parameters in the free volume model. Tension experiments were performed at four different temperature and humidity levels and were used to calibrate the dilatational, thermal and hygral parameters in the hybrid model. The linear and hygrothermal nonlinear viscoelastic constitutive models were used in the analysis of time-dependent interfacial fracture between concrete and epoxy primer. A generalized time-dependent J integral was used as a fracture parameter for characterizing the time-dependent interfacial fracture. This was used instead of the strain energy release rate and the stress intensity factor because of the nonlinear viscoelastic deformation of the primer. Schapery’s pseudo stress model was calibrated using tension data at various temperature and humidity levels because it is required for the generalized J integral. An instrumented wedge test was conducted in order to determine the interfacial fracture energy at several loading rates and various temperature and humidity levels. The crack length was measured as a function of wedge speeds during steady state crack growth. The generalized J integral and cohesive zone size or failure zone size were computed using finite element analyses that incorporated the pseudo stress model. The pseudo stress model, cohesive zone size and the generalized J integral were all used to compute the work input into the failure zone, which was then equated to the fracture energy. The loading rate, temperature and humidity level all affected the fracture energy, which decreased with increasing temperature and humidity levels.