Concentration Fluctuations In Alloys Research Articles

Liquid alkali metals and alloys as electron-ion plasmas

Linear screening theory yields expressions for the structure of liquid metals and alloys in terms of the structure of a bare ionic plasma on a rigid background, of the dielectric function of the homogeneous electron gas, and of electron-ion pseudopotentials. The results of such an electron-ion plasma model are examined against experimental data on the structural and thermodynamic properties of the liquid alkali metals and their alloys. The calculations rely on a very accurate theoretical determination of the structure factor of the classical ionic plasma, which embodies its known equation of state and thermodynamic consistency. The model gives a rather accurate account ofthe compressibility of the liquid metal. The main empirical adjustments that are included at finite wavenumber are a choice of the plasma parameter at freezing and a large momentum cutoff of the electron-ion coupling, as suggested by melting-curve data and by a discussion of dynamic criteria for the solid-liquid transition in the alkali metals. The model is then shown to yield very good agreement with experiment for the detailed shape of the structure factor of the liquid metal near freezing, and for its dependence on pressure, temperature and density. A discrepancy with experiment is found in the description of long-wavelength concentration fluctuations in alloys and is attributed to electronic charge transfer between the constituents, which is not included in the model. An estimation of the spinodal curve for the Li-Na alloy is presented in support of this conjecture.

Concentration fluctuations and structure factors of alloys

Structure factors for small wave numbersk are determined for alkali alloys by using perturbation on the hard-sphere potential. At zerok, formulae are proposed which link this model with the previous work on concentration fluctuations in alloys. By using these formulae, the phase separation curve for the Li−Na system is determined. In all cases, the agreement with experiment is good.