Atomic ordering in Cu-Al-Ni: Point approximation and Monte Carlo simulations

Abstract

Long range ordering in the bcc phase of the Cu-Al-Ni alloy is modelled through the analytical Bragg-Williams approximation and by means of numerical Monte Carlo simulations. The interchange energies that govern the ordering reactions are determined by fits to experimental ordering temperatures. A satisfactory agreement with the experimental data is obtained within both models, using slightly different sets of interchange energies. It is found that ordering in first neighbours is driven by the Ni-Al interactions, whereas the ordering in next nearest neighbours occurs by a reordering of Cu-Al pairs. Monte Carlo simulations enable a reinterpretation of the experimentally observed ordering reactions. Further details of the ordering process, such as the existence of tricritical points as the Ni content is reduced, and the evolution of sublattice occupancies as the ordering proceeds are also discussed: the site occupation probabilities at low temperatures agree with the experimental values.