Modelling of polycrystalline nial shape memory alloys and investigation of its structural properties

Modelling of Polycrystalline NiAl Shape Memory Alloys and Investigation of Its Structural Properties The effect of the number of atoms and composition on martensitic phase transformations in a single crystal NiAl shape memory alloys were investigated structurally. The effect of grain size and number on the martensitic phase transformations for the polycrystalline shape memory NiAl alloys were studied thermodynamically and structurally. Molecular dynamics simulations were used in the calculations. 12 single crystals and 18 polycrystals models, totally 30 models, were produced with the aid of the LATGEN software and the Voro++ library. All of the calculations were realized using the NPT molecular dynamics algorithm. The physical interactions between atoms were represented by the embedded atom method. All of the simulation works were performed through LAMMPS software package. The single crystal NiAl models having B2 superlattices and totally 3456 atoms were set up to be 8 different models in nickel-rich phase zones, 68, 69, 70, 71, 72, 73, 74, 75 Ni in atomic percentages. System temperatures for each model were reduced from 800 K to 25 K, as to 25 K interval at every 100,000 MD steps. It was found that there is a strong relation between atomic composition and Ms temperatures. To investigate the effect of the model size on the Ms temperature, 5 different models with 2000, 3456, 16000, 54000, 128000 atoms of Ni-at.25%Al composition were produced. The system temperature was also reduced from 800 K to 400 K in 25 K interval at each 100000 MD steps. It was observed that the Ms temperatures decrease curvilinearly with increasing model size. To determine the effect of model size and grain size on the Ms temperature in the polycrystalline structures, 18 different models with two-, three- and four-grained and Ni-at.%25Al composition were established. The geometry of the grains was kept constant in all models. 6 Different sizes of models were made for each of two-, three- and four-grained models. The average number of atoms in these models vary from 9000 to 995000. It was observed that the Ms temperatures and the phase transformation rates increase with the increasing of the number of grains. As increasing model size, it was found that the Ms temperatures increase and the phase transformation speeds are affected. Keywords: Polycrystalline alloys, Shape memory alloys, Nanostructures, Potential energy functions, Molecular dynamics simulation.