Revisiting copper as a case study of creep in pure metals: Prediction of creep response in pure Cu in half-hard and friction-stir-processed states

A physical model was used to obtain a pre-assessment of the dependence of minimum creep rate on stress and temperature for ETP copper under two different initial conditions, i.e., in the R240 half-hardened state and after friction stir processing. Although the R240 samples contained a high dislocation density owing to pre-straining, the material that underwent friction stir processing (FSP) had a much finer grain size and a far lower dislocation density. The original Sandström physically based model developed for copper was modified to take into account the strengthening role of the grain boundaries, and its declining importance with increasing temperature and/or decreasing applied stress. In addition, the grain growth effect was estimated using appropriate empirical equations. The model curves obtained by introducing a few initial information, such as the grain size and the ultimate tensile strength at room temperature, in the resulting set of constitutive equations, largely overlapped. This finding was somewhat in conflict with the traditional view, which assumes that a fine grain size is an ineffective source of creep strengthening, if not detrimental. Subsequent creep experimental testing carried out between 178 and 355 °C substantially confirmed the picture provided by the pre-assessment, and the minimum creep rate values for the two materials largely overlapped on the same curves. Only at 355 °C did the FSP samples exhibit a somewhat higher creep rate, when compared with the R240 samples and the pre-assessment curve. The experimental data obtained for R240 copper at 475 °C were not properly described by the physical model, as expected on the basis of the previous literature reports.