Abstract
Despite numerous investigations, all previous efforts on thermodynamic modeling of Al–Sr have suffered from inaccurate energetics of either the solid-state compounds or of the liquid alloy. Here, we demonstrate a method yielding simultaneously, accurate solid-state and liquid energetics, as given by first-principles density functional calculations and experimental measurements, respectively. Via first-principles methods, we have investigated the T=0 K energetics of not only the reported ground state compounds in Al–Sr (“the usual suspects”), but also of a wealth of other possible crystal structures observed in isoelectronic alloy systems (somewhat more “unusual suspects”). We find: (i) LDA calculations surprisingly show that Al 2Sr in the C15 structure is slightly lower in energy than the observed CeCu 2-type structure. However, GGA predicts the opposite order, consistent with the observed CeCu 2-type/ C15 stability. (ii) An as-yet-unreported Al 5Sr 4 compound (observed in Al–Ba) is found to be on the T=0 K ground state hull. (iii) An Al 3Sr 8 phase, isostructural with the recently discovered Al 3Ca 8 compound, is predicted to lie above the ground state hull and is not a T=0 K ground state. Using the first-principles formation enthalpies along with experimental thermodynamic and phase stability information, we have performed a new CALPHAD modeling of Al–Sr, including the three observed intermediate compounds as well as a hypothetical compound Al 3Sr 8. Two different models of the liquid phase were considered: an associate and a random solution model. The descriptions resulting from the two liquid models are critically evaluated with respect to experimental data in the literature and the present first-principles results.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have