Solidification of ternary melts with a two-phase layer.

This review is concerned with the nonstationary solidification of three-component systems in the presence of two moving phase transition regions-the main (primary) and cotectic layers. A non-linear moving boundary problem has been developed and its analytical solutions have been defined. Namely, the temperature and impurity concentration distributions were determined, the solid phase fractions in the phase transition regions and the laws of motion of their boundaries were found. It was shown that variations in the initial impurity concentration affect significantly the ratio between the lengths of the two-phase layers. A non-linear liquidus surface equation is theoretically taken into account as well.

Microstructure and morphology of Si crystals grown in pure Si and Al-Si melts.

Microstructure of Al-40 wt%Si samples solidified in electromagnetic levitation furnace is studied at high melt undercooling. Primary Si with feathery and dendritic structures is observed. As this takes place, single Si crystals either contain secondary dendrite arms or represent faceted structures. Our experiments show that at a certain undercooling, there exists the microstructural transition zone of faceted to non-faceted growth. Also, we analyze the shape of dendritic crystals solidifying from liquid Si as well as from hypereutectic Al-Si melts at high growth undercoolings. The shapes of dendrite tips grown at undercoolings >100 K along the surface of levitated Al-40 wt%Si droplets are compared with pure Si dendrite tips from the literature. The dendrite tips are digitized and superimposed with theoretical shape function recently derived by stitching the Ivantsov and Brener solutions. We show that experimental and theoretical dendrite tips are in good agreement for Si and Al-Si samples.

Dendrite tips as elliptical paraboloids.

This review article summarizes current theories of the steady-state growth mode of dendrites in the form of elliptical paraboloids. The shape of dendrite tips is analyzed, temperature and solute concentration distributions are described in its vicinity, and a solution of the hydrodynamic problem of a viscous incompressible fluid flowing against a dendrite tip is developed. A significant difference in analytical solutions describing a dendrite tip as an elliptic paraboloid as compared to an axisymmetric morphology is shown. The system of nonlinear equations for determining the stationary velocity of dendrite growth and the radii of curvature of the dendrite tip along the major and minor axis of the ellipse, respectively, is derived. The developed theory is compared with experimental data on the growth of ice crystals consisting of D2O or H2O.

Dendritic growth of ice crystals: a test of theory with experiments.

Motivated by an important application of dendritic crystals in the form of an elliptical paraboloid, which widely spread in nature (ice crystals), we develop here the selection theory of their stable growth mode. This theory enables us to separately define the tip velocity of dendrites and their tip diameter as functions of the melt undercooling. This, in turn, makes it possible to judge the microstructure of the material obtained as a result of the crystallization process. So, in the first instance, the steady-state analytical solution that describes the growth of such dendrites in undercooled one-component liquids is found. Then a system of equations consisting of the selection criterion and the undercooling balance that describes a stable growth mode of elliptical dendrites is formulated and analyzed. Three parametric solutions of this system are deduced in an explicit form. Our calculations based on these solutions demonstrate that the theoretical predictions are in good agreement with experimental data for ice dendrites growing at small undercoolings in pure water.

Modeling of dendrite growth from undercooled nickel melt: sharp interface model versus enthalpy method.

The dendritic growth of pure materials in undercooled melts is critical to understanding the fundamentals of solidification. This work investigates two new insights, the first is an advanced definition for the two-dimensional stability criterion of dendritic growth and the second is the viability of the enthalpy method as a numerical model. In both cases, the aim is to accurately predict dendritic growth behavior over a wide range of undercooling. An adaptive cell size method is introduced into the enthalpy method to mitigate against 'narrow-band features' that can introduce significant error. By using this technique an excellent agreement is found between the enthalpy method and the analytic theory for solidification of pure nickel.