Direct and Reverse Separation of Binary Metallic Melts in Capillaries

Abstract

The main purpose of the work is to refine the mechanism of diffusion in melts and its relation to the mechanism of metal separation in capillaries. When a binary melt is separated in a capillary at a given temperature, a stable equilibrium component concentration distribution over its height is reached in 1–2 h and does not change further. A number of studies have shown that this distribution is related to the barometric distribution of component clusters of a certain size in the terrestrial gravitational field, and the latter distribution strongly depends on temperature. As the temperature increases, the depth of separation decreases. It is obvious that, in the presence of concentration gradients inside a capillary, diffusion fluxes of both components directed opposite to the process of separation also exist; therefore, the achieved equilibrium state is likely to be stationary. This assumption is tested in an experiment, which consists in creating special conditions in a capillary with a melt, under which separation is excluded but the diffusion conditions remain unchanged. For this purpose, capillaries with the Sn + 40 wt % Pb melt are heated to a high temperature after reaching equilibrium. In the experiment, the dynamics of the transition of a concentration curve along the capillary height to a new equilibrium state is studied. The experimental results confirm that both processes, namely, diffusion concentration equalization and separation directed in opposite directions, occur simultaneously in the state of stationary equilibrium. The results of processing the experimental data with a mathematical model of separation, which is based on the concept of component motion in the form of a monatomic film, suggest that diffusion also proceeds via the flow of the melt components in the form of a film between melt clusters. This finding points to the same carrier for these processes. The method of conducting the experiment and processing the experimental data additionally allow us to determine the diffusion coefficient of lead in tin, which turns out to correspond to the reported data. Therefore, this method can be recommended for practical use.