Atomistic Simulation of Short Crack Growth in Correlation with Fatigue Indicator Parameter

Abstract

Under time-varying extreme dynamic environmental conditions, fatigue damage could gradually accumulate with increased stress cycles at critical details of steel or other metallic structures. As a promising tool, a fatigue indicator parameter (FIP) has been widely used to evaluate the crack initiation and propagation process in a mesoscale model. However, the correlation between the FIP and crack growth rate has not been carefully investigated, especially in the micromechanics level. In the present study, to calculate FIP for nanoscale fatigue crack growth, molecular dynamics simulation is performed. Two cyclic loading regimes are applied to the specimen with a preexisting central crack. In each loading regime, the relationship between the FIP and crack growth rate da/dN is calculated based on the FIP distribution contour plot. In the increasing maximum strain loading regime, cracks propagate linearly with the increase of the numbers of stress cycles, and the FIP distribution is similar to the distribution of the disordered lattice. However, lattices with maximum FIP values are mainly located at the boundary of the disordered lattice due to the dislocation and slip concentration. In the constant maximum strain loading regime, a void forming near the crack tip is observed, and a crack propagates by linking these voids and crack tip. With the increase of loading cycles, the crack growth rate decreases to nearly zero while the maximum FIP decreases slowly, implying a lagging effect in this process. The atomistic simulations in both loading regimes demonstrate a linear correlation between the FIP value and the crack growth rate da/dN.