Search

This paper investigates the role of dynamic stress concentrations, and of fracture mechanics-driven growth of critical clusters of fibres, on the longitudinal tensile failure of fibre-reinforced composites. For this purpose, we developed a semi-analytical fibre bundle model to simulate the longitudinal tensile failure of large composite bundles of continuous fibres. The model uses shear-lag to calculate the stress recovery along broken fibres, and an efficient field superposition method to calculate the stress concentration on the intact fibres, which has been validated against analytical and Finite Element (FE) results from the literature. The baseline version of the model uses static equilibrium stress states, and considers fibre failure driven by strength of materials (stress overload) as the only damage theory which can drive bundle failure. Like other models in the literature, the baseline model fails to capture the correct size effect (decreasing composite strength with bundle size) shown by experimental results. Two model variants have been developed which include dynamics stress concentrations and a fracture mechanics failure criterion respectively. To the knowledge of the authors, it is the first attempt in the literature to investigate these two effects in a fibre bundle model by direct simulation of large composite bundles. It is shown that, although the dynamic stress concentration significantly decreases the predicted bundle strength, it does not allow to predict the correct trend of the size effect. Finally, the results suggest that fracture mechanics may be the physical mechanism which is necessary to include to correctly predict the decreasing composite strength with bundle size shown by experimental results.

The translaminar fracture toughness of fibre-reinforced composites is a size-dependent property which governs the damage tolerance and failure of these materials. This paper presents the development, implementation and validation of an original analytical model to predict the tensile translaminar (fibre-dominated) toughness of composite plies and bundles, as well as the associated size effect. The model considers, as energy dissipation mechanisms, debonding and pull-out of bundles from quasi-fractal fracture surfaces; the corresponding lengths are stochastic variables predicted by the model, based on the respective bundle strength distributions and fracture mechanics. Parametric studies show that composites are toughened by stronger fibres with large strength variability, and intermediate values of interfacial toughness and friction. Predictions are validated against four different composite ply systems tested in the literature, proving the model’s ability to capture not only size effects, but also the influence of different fibres and resins.