Monte Carlo simulation of grain refinement during friction stir processing

A procedure combining computational fluid dynamics modeling/Monte Carlo simulation was implemented to predict grain refinement during friction stir processing (FSP) of an Al–Mg alloy. Based on the critical parameters during FSP treatment such as rotational tool speed (w), and traverse velocity (v), the thermal and strain rate contours were simulated, and used as inputs for a statistical model of dynamic recrystallization. Afterward, the simulated grain structures were verified experimentally by electron backscattering diffraction analysis. FSP generated equiaxed grains with average sizes in the range of 3–10 µm depending on the heat input index in terms of w/v ratios in the range of 4–28 rev.min/mm. A correlation between simulated and experimentally validated grain structures is observed, with crystallographic textures consistent with shear strain induced preferred orientations with a dominant {112} <110> component.