Mechanical Behavior and Life Prediction of a Low Nickel/High Nitrogen Austenitic Steel under Pure Torsion, Uniaxial Tension-Compression, and Multiaxial Fatigue

Abstract

Abstract Fatigue life prediction, crack growth, and propagation mechanisms are important details to consider for a structure to be used in service without fear of early failure. This is especially important for a new material for which very limited information is available. In this study, austenitic stainless steel P558 with low nickel content and high nitrogen content was subjected to fatigue testing at room temperature in multiaxial (static compression and cyclic torsion), pure torsion, and uniaxial tension-compression modes. The material generally depicts cyclic softening in the three loading conditions; although the multiaxial experiments experienced some initial hardening, those with a high shear stress and axial stress combination furthermore experienced a secondary hardening before softening. The Fatemi-Socie model predicts the pure torsion and multiaxial fatigue well but fails to predict fatigue lives for uniaxial tension-compression. On the other hand, the Smith-Watson-Topper model predicts fatigue lives for the uniaxial fatigue and pure torsion sufficiently but is poor in predicting fatigue lives for the multiaxial fatigue loading conditions. The cracking behavior was observed to be dependent on the loading condition and the strain amplitude with Mode II cracks or shear-oriented cracks observed for the high strain amplitudes (resulting in shorter lives) whereas Mode I cracks, or tensile-oriented cracks, were observed from low strain amplitudes (resulting in higher lives). A mixed mode was also observed for specimens in the midrange strain amplitudes, which exhibited a branching from shear to tensile cracks. The static compressive stress is believed to have a negative influence on the cracking behavior and the fatigue life by the rubbing of the crack surfaces, which in turn reduces the load-holding capability of the material. The same friction at the crack tip could also be the reason for the heating up of the specimens.