Divecha, Mia2018-03-142018-03-142017-12https://hdl.handle.net/11299/194619University of Minnesota Ph.D. dissertation. 2017. Major: Chemical Engineering. Advisor: Jeffrey Derby. 1 computer file (PDF); 179 pages.Cadmium zinc telluride (CZT) is a semiconductor used for gamma ray detection in applications ranging from nuclear weapons monitoring to medical imaging. The production of CZT via the vertical gradient freeze (VGF) method is plagued by tellurium rich inclusions in the crystal that significantly decrease the detector performance. The accelerated crucible rotation technique (ACRT) is a promising approach for improving the production of detector-grade CZT crystals. ACRT, which repeatedly spins the crucible at varying rotation rates, was designed by Scheel and Schulz Du-Bois in the early 1970s as a means of controlling nucleation of new grains and reducing the size and density of inclusions. Despite the many improvements made in crystalline quality with the application of ACRT, little is understood with regard to how ACRT works to reduce inclusions, and few guidelines exist to help aid the selection of rotation schedules. Toward these ends, we have developed a realistic, comprehensive model based on the experimental set up at Washington State University (WSU). Finite-element methods are employed to solve the coupled phenomena of fluid mechanics, heat transfer, and solute transport in the VGF-ACRT system. We track the transport of tellurium and incorporate its thermodynamic effects on solidification phenomena. We show for the first time that the segregation of tellurium during growth drives constitutional supercooling. This supercooling is an indicator that a morphological instability has the potential to form and yield inclusions. Through a series of transient calculations, we aim to reduce the amount of supercooling via assessment of the transport phenomena and solidification dynamics. We present a thermodynamically based metric, which employs the classic instability criterion developed by Mullins and Sekerka, that represents the effect of ACRT on the stability along the solid-liquid interface. This metric is utilized for the comparison and optimization of rotation schedules. We find that, in contrast to conventional wisdom, slower rotation schedules that promote disruption of the solute field without disruption of the local interface velocities are found to be most favorable for the WSU system. Preliminary experimental evidence is presented that supports these findings.enThesis or Dissertation