S. Agarwal, C. Briant, A. F. Bower, Brown University, Providence, RI; P. E. Krajewski, GM Research and Development Center, Warren, MI; E. Taleff, University of Texas, Austin, TX

Summary:

We have recently developed a finite element method designed to model the mechanisms that cause superplastic deformation (J. Mech. Phys. Solids, 52#6(2004) 1289). Our computations idealize the solid as a collection of two-dimensional grains, separated by sharp grain boundaries. The grains may deform plastically by thermally activated dislocation motion, which is modeled using a conventional crystal plasticity law. The solid may also deform by sliding on the grain boundaries, or by stress driven diffusion of atoms along grain boundaries. The governing equations are solved using a finite element method, which includes a front-tracking procedure to track the evolution of the grain boundaries and surfaces in the solid. Our goal is to validate these computations by systematically comparing its predictions to experimental measurements of the elevated temperature response of aluminum alloy AA5083 (Adv. Superplasticity and Superplastic Forming (2004) 127). This experimental work had previously shown that a transition occurs from grain boundary sliding to dislocation (solute drag) creep at approximately 0.001/s for temperatures between 425C and 500C. In addition, increasing the grain size from 7 to 12 microns decreased the transition to significantly lower strain rates. Predictions from the finite element method accurately predict both the effect of temperature and grain size on the transition in deformation mechanisms.