The low stress creep of aluminium near to the melting point: the influence of oxidation and substructural changes

Publisher Website
Google Scholar
Abstract
Creep rates of pure metals measured at high temperatures and low stresses often, agree well with those predicted by the Nabarro-Herring theory. Aluminium has been regarded as an exception since reported creep rates are about three orders of magnitude faster than predicted and other observations could not be accounted for by the theory. Various explanations of these discrepancies have been attempted in the literature involving the effects of oxidation and of substructural changes. In this paper experiments are described in which the low stress creep of aluminium is re-examined with special reference to the influence of these variables. The first part describes experiments on the creep of aluminium in vacuum at 913°K and in the stress range 0-1 MN/m2. At a stress level of ∼ 650 kN/m2a transition occurred in creep behaviour. Above this transition creep rate varied with applied stress σ raised to the power of ∼ 5 typical of a process controlled by dislocation climb and results were in good agreement with previous work. Below the transition creep rates were much lower than reported previously and in this case could be reasonably described by the modified form of the Nabarro-Herring equation: = (BΩD/dt kT) σE, where σE is an effective stress given by: σE = σ—σ0, where σ0= 145 kN/m2, Ω is the atomic volume, D the lattice self-diffusivity, d the grain size in the specimen whose thickness is t and kT has its usual meaning. B is a numerical constant estimated theoretically to be ∼ 10. Further experiments on the influence of oxidation show that this causes inhibition rather than enhancement of creep. It leads to the development of a large threshold stress for deformation associated with the restraining effect of the oxide layer. For the specimen to extend oxide fracture or shear on the Al/Al2 O3 interface must occur and the relative importance of these two processes is assessed. Substructural changes introduced by a period of fast straining did not influence subsequent creep in the low-stress range, demonstrating that such changes are unimportant, probably due to the ease at which dislocations anneal out at these high temperatures.