Debonding and friction at fibre-polymer interfaces. II: Macroscopic model experiments

Abstract

Experiments in which steel rods were embedded in epoxy resins contained within steel and polymer tubes, and in model composites consisting of six similar rods equispaced at a constant distance from the pulle-out rod, indicated that debonding of the rods is a brittle fracture process, with an energy of about 200 Jm −2. The debonding force, F A , in most of the experiments fitted an expression of the form F A = B√ AL tanh ( AL), where A and B are constants and L is the embedded length of rod. For small values of L ( AL < 0·3), the expression reduces to F A ⋍ ABL , suggesting an approximately constant shear stress governed by the fibre Young's modulus and diameter, and the interfacial work of fracture, together with matrix elastic constants, and a geometrical factor. For relatively large values of L ( AL >3), F A ⋍ B√AL . The pull-out method is potentially the most powerful technique for determining interface properties, since it can yield post-debonding frictional constants as well as information on the debonding process itself. However, it has to be carried out under very carefully controlled conditions if it is to yield absolute rather than comparative values. In particular, the container for the polymer should ideally be a polymer, to obviate tensile stresses at the interface (arising from the cure) and F A should be plotted as a function of L over as wide a range of L values as possible.