3D cohesive modeling of nanostructured metallic alloys with a Weibull random field in torsional fatigue

•CFEM with MCS is developed for 3D fatigue crack propagation in metals.•The model can predict realistic crack patterns and reasonable fatigue life.•Random fields and loads have significant effects on crack initiation and pattern.•Crack initiation mode varies and fatigue life is prolonged as NGL is thicker.

The cohesive finite element method together with Monte Carlo simulation for nanostructured metallic alloys with random fracture properties is developed to study the 3D fatigue crack propagation and torsional fatigue life. Three-parameter Weibull distribution is used to characterize the spatially random cohesive strength and fracture energy. The proposed model also considers the effects of thickness and different treatment of the nanograined layer (NGL) on the fatigue life. It is shown that the model can predict realistic crack patterns and reasonable fatigue life. The simulated fatigue cracks are mainly circumferential or oblique at an angle and they are in good agreement with the experimentally observed fracture patterns. Both different random fields and loads have significant effects on the crack initiation, crack pattern, and fatigue life. It is found that this layer plays a very important role in improving the fatigue life. As the layer thickness increases, the torsional fatigue life of the nanostructured metal also increases. The increase is particularly pronounced at high stress levels. We find that the major source of this increase is due to the increased probability for fatigue cracks to initiate from the interior surface of the tubular specimen and then propagate toward the exterior surface. This process has a profound beneficial effect on the fatigue life.