The phase stability of superstructures based on the fcc lattice in the Au-Pd and Ag-Pt alloy systems are examined from the fully relativistic electronic density functional theory. The electron-ion interaction is described by the projector augmented-wave ͑PAW͒ method and the exchange-correlation effects are treated in the generalized gradient approximation ͑GGA͒. The cluster expansion method is used to obtain effective cluster interactions on the fcc lattice and is used also to guide a systematic ground state search for both alloy systems. The ground state analysis reveals a multitude of ground states in Au-Pd, especially at the Au-rich side. Possibly long-period super-structures occur near the Au 70 Pd 30 composition. The ground state analysis indicates a uniquely stable AgPt compound with the L1 1 structure ͑CuPt prototype͒ and it also suggests a marginally stable ordered compound for Ag 3 Pt. However, our ab initio study rules out the existence of the remarkably stable Ag 3 Pt phase with L1 2 structure, reported first more than half a century ago and since then included in many assessments. We also find no indication for a stable ordered state at the AgPt 3 composition. The cluster variation method ͑CVM͒ with a large maximal cluster is used to compute the enthalpy of mixing of the disordered solid solutions and the solid portion of the Au-Pd and Ag-Pt phase diagrams. These results are critically compared with experimental data and phase diagram assessments. It is shown that cluster expansions cannot account for the high-temperature miscibility gap in the Ag-Pt system when the effective cluster interactions do not reach beyond the second nearest neighbor. Only when third nearest neighbors are included in the cluster expansion is it possible to obtain a phase diagram that agrees qualitatively with the assessed Ag-Pt phase diagram.

We present a study of the thermodynamic and physical properties of Ta 5 Si 3 compounds by means of density functional theory based calculations. Among the three different structures (D8 m , D8 l , D8 8 ), the D8 l structure (Cr 5 B 3 -prototype) is the low temperature phase with a high formation enthalpy of -449.20kJ/mol, the D8 m structure (W 5 Si 3 -prototype) is the high temperature phase with a formation enthalpy of -419.36kJ/mol, and the D8 8 structure (Mn 5 Si 3 -prototype) is a metastable phase. The optimized lattice constants of the different Ta 5 Si 3 compounds are also in good agreement with the experimental data. The electronic density of states (DOS) and the bonding charge density have also been calculated to elucidate the bonding mechanism in these compounds and the results indicate that bonding is mostly of covalent nature.The elastic constants of the D8 m and D8 l structures have been calculated together with the different moduli. Finally, by using a quasiharmonic Debye model, the Debye temperature, the heat capacity, the coefficient of thermal expansion and the Grüneisen parameter have also been obtained in the present work. The transformation temperature (2303.7K) between the D8 m and the D8 l structures has been predicted by means of the Gibbs energy, and this predicted temperature (2303.7K) is close to the experimental value (2433.5K).

Phase stability in the Ni-Al binary system is investigated using linear muffin-tin orbitals total energy (LMTO) calculations. They provide total energies for the different existing compounds and, using Connolly-Williams inversion, the many-body interactions occurring in the FCC and BCC lattices. These interactions are used in conjunction with the cluster variation method (CVM) to calculate the phase diagram. The computed phase diagram agrees very well with the experimental one.

No abstract

No abstract

We report an ab initio study of the semiconducting Mg(2)X (with X = Si, Ge) compounds and in particular we analyze the formation energies of the different point defects with the aim of understanding the intrinsic doping mechanisms. We find that the formation energy of Mg(2)Ge is 50% larger than that of Mg(2)Si, in agreement with the experimental tendency. From a study of the stability and the electronic properties of the most stable defects, taking into account the growth conditions, we show that the main cause of the n doping in these materials comes from interstitial magnesium defects. Conversely, since other defects acting like acceptors such as Mg vacancies or multivacancies are more stable in Mg(2)Ge than in Mg(2)Si, this explains why Mg(2)Ge can be of n or p type, in contrast to Mg(2)Si. The finding that the most stable defects are different in Mg(2)Si and Mg(2)Ge and depend on the growth conditions is important and must be taken into account in the search for the optimal doping to improve the thermoelectric properties of these materials.

No abstract

No abstract