Modelling and experimental study of the tertiary creep stage of Grade 91 steel

This article addresses experimental studies and analytical simulations of the tertiary creep stage of Grade 91 steel tested at various stresses and temperatures between 500°C (up to 160 × 103 h) and 600°C (up to 94 × 103 h). The strain rate increases after its minimum mainly because of the softening of the material which microstructure evolves strongly during creep deformation. An interrupted creep test shows that necking significantly affects the acceleration of the reduction in cross-section only during the last 10% of the creep lifetime. The Hoff model based on homogeneous reduction of cross-section correctly predicts lifetimes only for high applied stress. The Hart necking model using the Norton power-law allows fair predictions of lifetimes up to 60 × 103 h at 500°C. The necking model using a modified Norton power-law combined with a material softening term allows predictions of lifetimes for all creep tests, differing from the experimental results by less than 50%, which is consistent with the experimental scatter. The evolution of the cross-section predicted by this model is in agreement with measurements carried out during the interrupted creep test. Two prediction rules for the lifetime prediction are deduced from the necking model that takes into account the material softening. For a large number of tempered martensitic steels, these two criteria bound the experimental lifetimes up to 200 × 103 h at 500–700°C.