Modeling of transport processes during solution, melt and colloidal crystal growth.

University of Minnesota Ph.D. dissertation. August 2008. Major: Chemical Engineering. Advisor: Jeffrey J. Derby. 1 computer file (PDF); xvi, 196 pages. Ill. (some col.)In this thesis, numerical models are developed and applied to study systems used for the growth of crystals from both solution and the melt. Additionally, numerical models are employed to study the convective self-assembly of microspheres within solution. Solution crystal growth can be visualized in real-time through the application of atomic force microscopy (AFM) within a fluid cell. We apply a three-dimensional finite element method on a parallel supercomputer to determine the continuum transport of momentum and mass in an AFM fluid cell during crystal growth, using data acquired from calcium oxalate monohydrate crystal growth measurements as a comparison. Simulations quantify mass transfer resistances to crystal growth inherent to the fluid cell geometry, and examine influences on growth via high-frequency cantilever oscillations. The melt growth of single crystal cadmium zinc telluride (CZT), a high-value crystal used in radiation detectors, has posed a serious challenge for crystal growers for over three decades. We employ a two-dimensional finite volume method to simulate CZT growth in a vertical Bridgman furnace used by our collaborators at Pacific Northwest National Laboratories. Models couple the continuum transport of mass, momentum and radiation, and track the interface shape between the melt and crystal. Results provide insight into the thermal behavior of two crucibles to be used for CZT growth by our collaborators. Three-dimensional computations of steady flows directed toward the (1 1 1) plane of a face-centered cubic (fcc) packing of microspheres are carried out to assess the convective steering hypothesis, which posits that solvent flow could play a role in the assembly of colloidal crystals. The computations clearly show the kinematics of flows into and through the packing and clarify the influences of fluid inertia and particle arrangement. Results from the computations accurately describe the outcome of macroscopic experiments