Mesh-Independent Framework for the Bidimensional Analysis of CFRP–Concrete Debonding Shear Tests with Discrete Fracture

Abstract

The performance of concrete structures strengthened with carbon fiber–reinforced polymer (CFRP) systems can depend heavily on the bond strength of the interface between the concrete and the reinforced polymer. Even though experimental testing can be used to derive suitable constitutive models, their interpretation and analysis is often limited by the reliability of available numerical/analytical models. The debonding in shear tests can be controlled by the highly nonlinear interaction of the bonded interface with the microcracks developing in the substrate. This process cannot be efficiently predicted by simplifying assumptions, which is why robust models accounting for those features, while relying only on material parameters that can be easily measured and interpreted, need to be developed. This paper introduces a framework for developing such models based on the discrete representation of fracture that can be easily deployed into existing finite-element codes. The substrate bond failure, in addition to the interface bond failure and any combination thereof, are automatically accounted for, and the cracks are not prespecified to the underlying finite-element mesh, which means that the results are mesh-insensitive and discretization-independent. A validation of the proposed framework was performed using modified double-shear bond tests between CFRP and concrete. An in-depth analysis was carried out to assess the influence of bond length and CFRP reinforcement area on the debonding behavior and ductility of the connection.