Assessment of creep damage models in the prediction of high-temperature creep behaviour of Alloy 617

This study aims to predict the creep response of Alloy 617 and assess the accuracy of three established creep damage models in predicting the creep time-to-failure (t(f)) and strain-to-failure (epsilon(f)) in the temperature range of 800-1000 degrees C. Idaho National Laboratory (INL) creep data were used to calibrate (1) Ductility Exhaustion (DE), (2) Stress-Modified Ductility Exhaustion (SMDE), and (3) Strain-Energy Density (SED) creep damage models using multiple linear regression (LR) and reverse damage (RD) approaches. In order to predict the creep response of Alloy 617 these creep damage models are then coupled with (secondary) creep strain model calibrated using Idaho National Laboratory (INL), Korean Atomic Energy Research Institute (KAERI), and Argonne National Laboratory (ANL) experimental data. Direct comparisons made between the investigated creep damage models revealed that the SED model with parameters calibrated through RD approach captures the creep response of Alloy 617 the most accurately and it thus produces the most accurate prediction of time-to-failure (t(f)) and strain-to-failure (epsilon(f)) across different temperature/stress (creep) conditions. However, it is also shown that none of the employed creep damage models are able to fully capture the material creep response at 1000 degrees C. This is attributed the strong oxidation of Alloy 617 at 1000 degrees C (tested in air) leading to the formation of a thick oxidation layer, which might affect the failure mechanism of the alloy at this temperature.