Superplastic deformation without relative grain translation?

It is now almost universally accepted that the mechanism by which superplastic deformation occurs involves relative grain translation, by grain boundary sliding, together with dislocation creep and/or diffusional creep. While there has been much debate on the fine points of the process - principally the so-called "accommodation mechanisms", the primary role of the relative translation of grains via grain boundary sliding is virtually undisputed. Although this mechanism appears to be consistent with several aspects of microstructural evolution in superplastic deformation, the direct evidence for a dominant grain translation mechanism is mainly drawn from work using surface markers, not withstanding the fact that the extra degree of freedom at the surface of a superplastically deformed sample may make it somewhat unrepresentative of the bulk material. Research carried out recently on Ti-6Al-4V and AA5083+Cr, two well known superplastic materials, using both internal and surface markers throws into some doubt the importance of relative grain translation in superplastic deformation. The behaviour of aligned grain structures acting as internal markers - is also difficult to reconcile with relative grain translation mechanisms. The fact that such microstructures occur in materials which give a high strain rate sensitivity is inconsistent with mechanisms in which grain boundary sliding is a principal, rather than a subservient or even accommodating, mechanism. Evidence from transmission electron microscopy and plastic anisotropy suggest that intragranular slip may have a fundamental role in superplasticity. Certainly, some consideration must be given to mechanisms other than relative grain translation, though these must include the role of grain boundaries in leading to high strain rate sensitivity.