For the seeding process of oxide Czochralski crystal growth, influence of the crucible bottom shape on the heat generation, temperature and flow field of the system and the seed-melt interface shape have been studied numerically using the finite element method. The configuration usually used in a real Czochralski crystal growth process consists of a crucible, active afterheater, induction coil with two parts, insulation, melt, gas and seed crystal. At first, the volumetric distribution of heat inside the metal crucible and afterheater inducted by the RF-coil was calculated. Using this heat generation in the crucible wall as a source the fluid flow and temperature field of the entire system as well as the seed-melt interface shape were determined. We have considered two cases, flat and rounded crucible bottom shape. It was observed that using a crucible with a rounded bottom has several advantages such as: (i) The position of the heat generation maximum at the crucible side wall moves upwards, compared to the flat bottom shape. (ii) The location of the temperature maximum at the crucible side wall rises and as a result the temperature gradient along the melt surface increases. (iii) The streamlines of the melt flow are parallel to the crucible bottom and have a curved shape which is similar to the rounded bottom shape. These important features lead to increasing thermal convection in the system and influence the velocity field in the melt and gas domain which help preventing some serious growth problems such as spiral growth.

The motivation for this study is the need for accurate numerical models of melt flow instabilities during Czochralski growth of oxides. Such instabilities can lead to undesirable spiralling shapes of the bulk crystals produced by the growing process. The oxide melts are characterized by Prandtl numbers in the range 5

No abstract

Rare earth scandate crystals ReScO 3 (Re=La, Ce, Pr, Nd, Sm, Eu, Gd, Tb and Dy) can be grown from the melt at temperatures of about 2100 °C. The needs of thermal insulation of the whole system are very high in order to reach the melting point and to control the thermal gradients which are required by the Czochralski (Cz) method. The consequence is that in-vivo system observations are practically almost impossible or very hard to perform. Therefore numerical investigations using a mathematical model of the real system can be very helpful. However, numerical models need some physical properties of the considered real system (e.g. density, viscosity, thermal expansion coefficients, thermal conductivity). Most material properties of high melting oxides are not referenced in the literature or they are incomplete and inaccurate. Because the accuracy of qualitative and quantitative results of numerical simulations depend on the used physical properties, we performed corresponding measurements in an adapted Cz configuration at a temperature around the melting point of DyScO 3 (2060 °C). The results are presented and discussed in this work.

No abstract