Mechanisms of fatigue crack initiation and growth in particulate-reinforced aluminum metal matrix composites

The high cycle fatigue crack initiation and small crack growth processes of a particulate-reinforced aluminum alloy metal matrix composite (MMC) system have been evaluated. The research is based on a model Al-Si-Mg system in which Si particles are dispersed within an age-hardenable aluminum alloy matrix. The resulting microstructure is systematically controlled with regard to matrix flow behavior, particulate volume fraction (10 to 20 vol. %), and particulate size (4 and 8 {dollar}mu{dollar}m). The influence of these microstructural variables on the fatigue crack initiation process is examined in detail. The research identifies two competing mechanisms by which a fatigue microcrack can initiate: (a) intense deformation within a thin ligament ({dollar}le{dollar}0.1 particle diameters) of matrix material separating a subsurface particle from a free surface and (b) cracking of the reinforcement particles. Computational analysis using finite element modelling shows that thermally induced, tensile residual stresses exist in both the particle and the matrix provided the particle is well bonded to the matrix and located {dollar}le{dollar}0.1 particle diameter from a surface. It is proposed that this residual stress state assists crack nucleation in both the particle-cracking and the matrix-ligament fatigue crack initiation mechanisms.; While the residual stress state can assist both types of crack initiation, the MMC microstructure determines the preferred mechanism. The experimental results show that for a given loading condition, larger particulate sizes, higher volume fractions, and/or artificial aging tends to favor the particle-cracking mechanism. Furthermore, for the cracked particle mechanism, the responsible particle is much larger than average, and it generally intersects the specimen surface while remaining predominantly subsurface. As such, the responsible mechanism of initiation is a statistical process sensitive to microstructural parameters such as the range of particle sizes and spatial distributions (i.e., particle clustering) which create the "weakest" microstructural location.