Superplastic deformation of inorganic glasses is known since early times. More recently, it has been reported in microcrystalline and sub-microcrystalline metallic and ceramic materials, intermetallics, metal-based and ceramic-based composites and nanostructured materials. In materials with grain sizes in the lower end of the micrometer range and the sub-micron range, one observes this phenomenon at strain rates in excess of 10-2 s-1, when it is referred to as high strain rate superplasticity. Superplasticity is also present in bulk metallic glasses. It is desirable that an explanation be found for the underlying deformation mechanism which applies to these different classes of materials uniformly. The easiest way-out is to suggest that the phenomenon is the result of a host of mechanisms which combine to produce the observed results. Predictability is lost in such an approach. Grain boundary diffusion control and dislocation control are two other mechanisms that have been suggested, but both per se flounder because they cannot account for the varying value of the strain-rate sensitivity index, m, with strain rate in an isothermal test without the use of elaborate schemes. Analysis of the phenomenon using a generic equation, proposed to understand creep deformation when the stress sensitivity, n (inverse of m), was 5 and the activation energy for the rate controlling mechanism was equal to that for bulk diffusion, as applicable to superplastic flow is often resorted to. But, the present author finds the reservations of Ashby and coworkers and Pharr about this approach convincing. In our view, the simplest approach is to suggest that grain boundary sliding, which develops to a mesoscopic scale during superplastic deformation and for whose dominance there is incontrovertible experimental evidence, is the mechanism that controls the rate of flow during optimal superplastic deformation in all the above classes of materials. Developing a physical picture can follow two lines: treating sliding in terms of the glide of grain boundary dislocations along the grain boundaries and the movement of disclinations at triple junctions or an approach in which boundary sliding results from the accumulation of localized, atomic scale shears in basic sliding units of a few atomic diameters size. The latter approach does not require the presence of grain boundary dislocations and disclinations for which there is no experimental evidence. The present author will interpret the observations concerning superplasticity from the latter point of view. Some of the predictions of the analysis that have already been verified and those which can be verified by simple experiments will be listed.
<< back