The effect of matrix yield strain on the data reduction technique of the single-filament fragmentation test

Abstract

Abstract The effect of the elastic modulus and shear yield strength of the matrix on the interfacial sheat strength obtained from the fragmentation test has been the subject of numerous studies. Little attention has been paid to the effect of the matrix yield strain. An axisymmetric finite element model has been used to study the effect of matrix yield strain on the shear stress at the fibre—matrix interface and the tensile stress in the fibre and, consequently, the value of interfacial shear strength obtained from the fragmentation test data using the constant shear model of Kelly and Tyson. It was observed that saturation in fibre fragmentation during the test depends on the matrix yield strain rather than the fibre failure strain. Therefore, the matrix should be carefully selected to have the yield strain necessary for saturation. The maximum tensile stress that develops in the fragment depends on the tensile yield strain of the matrix. For a matrix with high tensile yield strain, a higher tensile stress in the fibre will exist than in a matrix of low tensile yield strain. Therefore, at saturation, the fragment lengths will be shorter so that an erroneously high value of the interfacial shear strength will be estimated from the constant shear model. This particular observation can lead to the hypothesis of a rigid interphase in the case of very ductile or elastomeric matrices even though such an interphase is not present.