The role of bimodal grain size distribution in nanocrystalline shape memory alloys

We report on a phase field simulation study of the effects of grain size and grain size distribution on martensitic transformations in shape memory alloys (SMAs). Recent experimental studies of polycrystalline SMAs revealed interesting grain size effects when grain sizes are reduced to tens of nanometers. While small grain sizes yield superior mechanical strength, the material loses its shape memory property if grain sizes are smaller than some critical size. To reintroduce shape memory properties, and at the same time retain strong mechanical properties, we propose to introduce several large grains (much larger than the critical transformation grain size) amid the nano-sized grains, which yields a grain microstructure with a bimodal grain size distribution. A phase field model is adopted to simulate the martensitic transformation in polycrystals, to which excess grain boundary energy is introduced that suppresses martensitic transformation in the grain boundaries. We observe that simulated as-quenched microstructures of the bimodal grain size distribution may in a special case equilibrate as a phase composite (austenite martensite). The spatial distribution of these phases and alloy functionalities (shape memory) may be controlled by the design of the grain size distribution. Loading and unloading simulations of the equilibrium martensitic microstructures show that the large grains enable martensitic detwinning to occur, hence reintroducing shape memory functionality to the nanocrystalline SMA.