Multi-scale modeling on select chemical vapor deposition processes

Chemical vapor deposition (CVD) is a process for producing solid particles from volatile precursors and depositing the particles on a suitable substrate. Many factors influence the performance of a CVD reactor such as operating conditions, chemistry, species concentration, and reactor configuration. In general, high purity, crystallinity, uniformity, and low cost deposition are the desired factors in CVD processes. A common problem with CVD reactors is non-uniform deposition (i.e. higher deposition rates at the edges of the substrate). Despite many efforts in changing the reactor configuration, (i.e. rotating, tilting, and changing the orientation of the substrate), achieving uniform deposition remains a challenge in field of chemical vapor deposition. Ample understanding of the CVD process in both macroscopic and microscopic levels is necessary to design and optimize a CVD reactor in order to increase the purity, uniformity, and decrease the final cost. Multi scale modeling of chemical vapor deposition aims at relating macroscopic process conditions (such as species concentration, flow rate, and operating pressure and temperature) to macroscopic and microscopic film properties (such as purity, morphology, crystallinity, thickness uniformity, and chemical composition). An important issue with modeling of chemical vapor deposition is the lack of experimental data (i.e. gas phase and surface kinetic data, physical properties, etc) for new processes. Molecular modeling fills this void by making it possible to estimate the necessary parameters with high fidelity.^ In this dissertation, we have developed a multi-scale computational model in order to investigate the deposition process, particle dynamics, and effect of substrate geometry on the deposition pattern in chemical vapor deposition reactors. The kinetic parameters were estimated using density functional theory (DFT) and transition state theory (TST). A dynamic model has been developed in order to optimize the substrate geometry based on the desired deposition pattern. Using a global optimization technique, we optimized the substrate configuration for zinc sulfide production.