Superplasticity and cavitation in an aluminum-magnesium alloy

Fundamental issues related to the forming performance of superplastic metals include the mechanisms of flow and cavitation occurring during the forming process. Cavitation beyond a critical amount is damaging to the mechanical behavior of fabricated parts. Therefore, the role of process parameters which influence cavitation must be precisely documented and understood. In this study, (1) the mechanism of deformation, (2) cavity formation and growth, and (3) the effect of forming parameters on cavitation are systematically investigated in a fine grain Al-4.7%Mg-0.8%Mn-0.4%Cu alloy.; The mechanical flow response of the alloy is characterized by a new type of step strain-rate test which preserves the initial microstructure of the alloy. Under isostructural condition, sigmoidal log $s$ vs. log relationship is determined and then analyzed by using a grain-mantle based quantitative model1 for superplastic flow. The activation energies in both grain-mantle creep and core creep are analyzed, and the overall controlling mechanism is found to be dislocation glide and climb. Grain-mantle creep rate in the low strain-rate region is found to be enhanced many times due to a high concentration of vacancies near grain boundaries.; Cavitation caused by superplastic straining under uniaxial tension is evaluated by the SEM (for < 0.5μm size) and the number and size of cavities are monitored by image analysis through optical microscopy. Growth of pre-existing cavities and nucleation and growth of new cavities at grain boundary particles are monitored with increasing strain. Cavity nucleation and growth occur in two stages: crack-like growth along the particle-matrix interface by a constrained growth process, and beyond complete debonding growth via plastic deformation of the matrix which is modeled here. Stresses and strain-rates near the void are intensified due to the perturbed flow field near the void, and not relaxed during the time frame associated with superplastic deformation. In the model, faster cavity growth is predicted for lower m and for smaller cavity density when cavity stress fields are not overlapping. Observed cavitation quantitatively agrees with the present model, but diffusional growth is found to be too slow, which cannot explain the observed nanoscale void growth behavior. Another parameter affecting the degree of cavitation is the imposed stress-state. Cavity growth rate as well as cavity nucleation rate increase with the level of mean hydrostatic tension. For a fixed cavitation volume fraction, V, the principal surface strains, $31$ and $32$, for the various stress-states can be empirically given by: $31=aVb-a32$, where a and b are constants determined from $31$ values for plane-strain $32=0$. The value of b is found to be 0.2 ∼ 0.3, and α is 0.4 ∼ 1.; 1 A. K. Ghosh, Mat. Sci. Forum, Trans. Tech. Publications, Switzerland, 170–172, 39 (1994).