Power-law and viscous creep in advanced 9% Cr steel

Short-term creep tests were performed on a 9%Cr (P-91 type) steel at temperatures from 823 Ii to 923 K;and wide range of stresses from 1 MPa to 300 MPa. The conventional constant stress tensile creep tests were performed for stresses above 100 MPa, while the helicoid spring specimens technique was used for low strain rates at lower stresses. Current knowledge of the creep mechanisms is based mainly on short-term laboratory experiments in power-law creep regime with creep rates higher than 10(-10) s(-1). However, service loading of real high temperature components may lead to rates lower than 10(-12) s(-1), which are typical for viscous creep. Unfortunately, experimental data describing viscous creep regime in structural materials are very rare, because their limited microstructural stability does not enable experiments to be run at very high temperatures, where the viscous creep can also be observed at higher strain rates. The steady state creep rates correspond to viscous behaviour under stresses below approximately 100 MPa at 600 degrees C, characterised by the apparent stress exponent close to 1. Since the stress exponent at higher stresses is around 10, the change in the deformation mechanism at lower stresses is proved. The deformation mechanisms map resulting from the presented data shows that the service loading conditions respond to the viscous creep. Extrapolation from power-law creep regime to low stresses can cause serious underestimation of predicted deformation rates. The service loading conditions for the steel under consideration lay close to the transition boundary from viscous to power-law creep regime. None of both deformation mechanisms is then negligible at the service conditions. Thus, any realistic model describing creep properties of the steel must take into account both power-law and viscous mechanisms. The effect of the viscous deformation mechanism on the creep life remains questionable. Since the materials behaves like Newtonian viscous fluid, it is possible that the viscous deformation mechanism has only little effect on the creep damage development. Unfortunately, only long term creep experiments (tens or hundreds of years) can shed light into this problem. Nevertheless, viscous creep should be taken into consideration, mainly for the parts where the dimension stability is critical.