This thesis discusses methodologies for the prediction of stable delamination development in fiber-reinforced composites in pure fracture modes using analytical and numerical methods.

A new test method, named internal-notched-flexure (INF) test, has recently been proposed to quantify mode II interlaminar fracture toughness of fibre composites. Previous investigation has shown that unlike any of the existing test methods, the INF test generates unconditionally stable delamination growth. This thesis discusses a follow-up study that revises the analytical expressions for compliance (C) of INF specimen and its energy release rate for delamination (G). The main improvement in the current approach is to take into account load in the overhanging section of the specimen;​ while in the previous approach, the overhanging section was assumed to be load-free. Validation of the revised expressions is through comparison of the initial specimen stiffness with that from a finite element (FE) model of the INF specimen. The virtual INF specimen has cohesive elements in the interlaminar region to simulate the delamination growth, from which extent of damage in front of the crack tip can be quantified. Results from FE model suggest that an extensive damage exists at the crack tip before delamination growth commences. Therefore, the use of a physical crack length in the analytical expression for G may severely overestimate the interlaminar fracture toughness. Instead, an effective crack length should be used. Expression for G based on the effective crack length yields a value that is very close to the input critical energy release rate (Gc) for the cohesive elements. The study concludes that load in the overhanging section should be considered for deriving the analytical expressions for C and G of the INF specimen, and an effective crack length should be used to calculate the Gc value from the analytical expression.

In addition to the above work, the study also touches on a finite element approach based on continuum solid elements with an elastic-plastic damage material property. The approach was proposed to simulate crack growth in the interlamianr region of FRP, but should also be applicable to other crack growth phenomena. In this approach, solid elements are used to simulate crack growth, based on criteria that are a combination of all stresses, in order to take into account the effect of in-plane normal stress on the damage initiation. The criterion for delamination propagation is defined based on critical strain energy. The approach was implemented in a finite element code and was applied to precracked composites to illustrate its feasibility to simulate the crack development.