Journal of Adhesion Science and Technology, Vol.28, No.1, 65-84, 2014
Analysis of a pull-out test with real specimen geometry. Part II: the effect of meniscus
This paper continues our study on the platelet model of the pull-out specimen, in which the matrix droplet shape is approximated by a set of thin parallel disks with the diameters varying along the embedded fiber. Using this model, the fiber tensile stress and the interfacial shear stress profiles were calculated for real-shaped matrix droplets, including menisci (wetting cones) on the fibers, taking into account residual thermal stresses and interfacial friction. Then, these profiles were used to numerically simulate the processes of crack initiation and propagation in the pull-out test and to obtain theoretical force-displacement curves for specimens with different embedded lengths and wetting cone angles. Our simulations showed that the interfacial crack in real-shaped droplets initiated at very small (practically zero) force applied to the fiber, in contrast to the popular equivalent cylinder' approximation. As a result, the equivalent cylinder approach underestimated the interfacial shear strength (IFSS) value determined from the pull-out test and at the same time overestimated the interfacial frictional stress; the smaller was the wetting cone angle, the greater the difference. We also investigated the effects of the embedded fiber length and interfacial frictional stress in debonded areas on the calculated IFSS. The simulated force-displacement curves for the real-shaped droplets showed better agreement with experimental curves than those plotted using the equivalent cylinder approach.
Keywords:pull-out test;local interfacial shear strength;specimen geometry;meniscus on fiber;equivalent cylinder;crack propagation;force-displacement curves;interfacial friction