Differences Between Composite and Metal Fatigue

Compare and contrast fatigue of metals and composites.

From a certain point of view, fatigue in composites and fatigue in metals are similar: both begin with damage initiation, followed by damage propagation, and end in ultimate failure. Fatigue life N f , or number of cycles to failure, for both can be thought of as the sum of the cycles during damage initiation N i and the cycles during damage propagation N p .

One difference is in the relative time spent in each phase. For metals, a significant portion of the fatigue process is spent propagating a single crack. Damage initiation is usually ignored because generally, many defects such as grain boundaries and dislocations exist in the material that can replicate new defects. The propagation phase is longer because metals strain harden. As a crack attempts to propagate through the metal, plasticity occurs at the crack tip causing crack blunting and strain hardening. The process of crack blunting, strain hardening, and crack progress can be repeated for many thousands of cycles. So in the case of metals, Equation (34) is simplified to

The amount of crack growth during each cycle of the propagation phase is often described with an empirical law, such as the Paris Law.

For a composite, such as a unidirectional carbon-epoxy laminate, strain hardening is negligible. This makes the propagation phase of the fatigue life much shorter than the damage initiation phase. This happens because damage progresses very quickly to ultimate failure once a defect of sufficient size is nucleated. Thus, for fiber-reinforced polymer (FRP) composites, Equation (34) is simplified to

The initiation of damage in a FRP is governed by a kinetic process of microcrack accumulation. When a critical density of microcracks is achieved, a macroscopic crack forms. This type of fatigue failure can be modeled with the kinetic theory of fracture (KTF) [29-34]. In the case of a FRP composite material, stresses in the polymer matrix are not the same as the composite stresses. To apply KTF to the polymer, a methodology for determining matrix stresses from composite level stresses must be implemented. Therefore, we employ multicontinuum theory (MCT) to extract these polymer matrix stresses from composite stresses, as described earlier (The MCT Decomposition), and use KTF to predict the matrix (and therefore composite) fatigue life.