Thermodynamics of pyrope–majorite, Mg3Al2Si3O12–Mg4Si4O12, solid solution from atomistic model calculations

Optimization of chemical ordering in AgAu nanoalloys

The determination of optimal chemical ordering in nanoalloys, i.e. of the most stable pattern in which atoms are arranged in bi- or multicomponent metallic clusters, is quite complex due to the enormous number of different possible configurations. This problem is very difficult to tackle by first-principle methods except for very small systems. On the other hand, the treatment at the atomistic potential level is complicated in many cases (such as AgAu) by charge transfer effects between atoms of different species in different coordination environments. Here an empirical atomistic model is developed to take into account such effects. The model is used to determine the optimal chemical ordering in AgAu nanoalloys. Charge transfer between atoms is taken into account by a modification of the charge equilibration method of Goddard and Rappé [J. Phys. Chem., 1991, 95, 3358], in which a coordination-dependent electronegativity and hardness are introduced. The model is applied to the determination of chemical ordering in AgAu nanoalloys. It is shown that the inclusion of charge transfer effects is important for improving the agreement of the atomistic model with density-functional calculations, leading to the determination of lower-energy chemical ordering patterns.

Coupled Al/Si and O/N order/disorder in BaYb[Si4–x Al x O x N7–x ]sialon: neutron powder diffraction and Monte Carlo simulations

The fractions of aluminium, [Al]/[Al + Si], and oxygen, [O]/[O + N], in crystallographically distinct sites of BaYb[Si4–x Al x O x N7–x ] oxonitridoaluminosilicate (space group P63 mc, No. 186) were refined based on the results of neutron powder diffraction for a synthetic sample with the composition of x = 2.2(2) and simulated as functions of temperature for the compositions x = 2 and x = 2.3 using a combination of static lattice energy calculations (SLEC) and Monte Carlo simulations. The SLEC calcu lations have been performed on a set of 800 structures differing in the distribution of Al/Si and O/N within the 2 × 2 × 2 supercell containing 36 formula units of BaYb[Si4–x Al x O x N7–x ]. The SLEC were based on a transferable set of empirical interatomic potentials developed within the present study. The static lattice energies of these structures have been expanded in the basis set of pair-wise ordering energies and on-site chemical potentials. The ordering energies and the chemical potentials have been used to calculate the configuration energies of the oxonitridoaluminosilicates (so-called sialons) using a Monte Carlo algorithm. The simulations suggest that Al and O are distributed unevenly over two non-equivalent T(Si/Al) and three L(N/O) sites, respectively, and the distribution shows strong dependence both on the temperature and the composition. Both simulated samples exhibit order/disorder transitions in the temperature range 500–1000 K to phases with partial long-range order below these temperatures. Above the transition temperatures the Si/Al and N/O distributions are affected by short-range ordering. The predicted site occupancies are in a qualitative agreement with the neutron diffraction results.