Experimental and computational investigation of Mg AZ31 grain refinement by shear-enhanced rolling

Abstract

Magnesium (Mg) alloys have been actively researched to replace various steel and aluminum alloys in the automotive, aerospace, and defense industries. However, their inferior formability needs improvement to compete with these alloys. For instance, in the case of rolling, conventional grain refinement produces surface layers of a finer grain structure while the core of thick rolled sections remains substantially coarser. This work introduces shear-enhanced rolling (SER) to overcome this issue and correspondingly enhance formability. SER applies a transverse load (TL) between two conventional rolling stations to induce high shearing stresses at the section core. An experimental study is done here on an Mg AZ31B alloy using an in-house prototype to prove the concept of SER. Experimental specimens are then tested to identify corresponding grain sizes across the SER section and extract Johnson-Cook model constants for computational optimization. Knowing that micro-grained Mg alloys obey the Hall-Petch relationship, the simulations could be used to predict and optimize grain refinement and uniformity by adjusting SER process parameters numerically. The computational results indicate that SER can reduce the grain sizes of AZ31B by up to 42% relative to conventional rolling of equivalent reduction ratio, in addition to the corresponding non-uniformity value of 0.0914. Also, results revealed that the contact length (L) between the rolled section and the traditional rollers correlates strongly with grain refinement and uniformity. L is directly relatable to the controllable process parameters of TL, the coefficient of friction (µ), and the roller radius (Rr). Our multi-objective optimization suggests increasing (L) to its maximum practical value to obtain the best grain refinement and uniformity with SER. Limitations of SER are also discussed.