Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
4642636 | Journal of Computational and Applied Mathematics | 2007 | 14 Pages |
Optimization is the ultimate goal of numerical modeling of crystal growth processes. We present the most important physical phenomena to establish a numerical process model for an industrial vertical gradient freeze (VGF) process of barium fluoride (BaF2BaF2). After reviewing the governing equations we describe an adaptive finite element approach for a phase change problem with internal radiation. We describe an interface to a commercial software package for industrial crystal growth processes to define a numerical process model in a highly automatized way. In this way we can apply an external optimization algorithm to our process model. We use a modified quasi-Newton method to propose an improvement of an industrial VGF growth process of BaF2BaF2.