Semi-empirical and numerical modeling of metal-organic chemical vapor deposition

The purpose of this research was to develop numerical and semi-empirical models for metal-organic chemical vapor deposition (MOCVD) growth, with the growth of titanium dioxide films serving as an application vehicle. This work utilizes 2- and 3-dimensional numerical simulation to gain a more complete understanding of the fundamental relationships underlying the epitaxial growth process.; To augment the numerical modeling, semi-empirical models which utilize hybrid neural networks (HNN) were investigated as a means to estimate unknown physical parameters such as reaction rate constants and activation energies. The numerical model predicts gas velocity, temperature, concentration of reactants and growth rate. The effects of radiation, external heat transfer, thermal diffusion, susceptor surface temperature distribution, and reaction rate order are considered, and a 3-D model is developed to explain experimentally observed off-center film deposition. The effects of growth parameters, gravity, and susceptor rotation on growth rate and thickness uniformity of TiO{dollar}sb2{dollar} films are also studied. In addition, an optimum reactor design is proposed for mass flow without circulations, which results in excellent uniformity and high growth rate. Finally, the relative influence of specific experimental conditions on film growth and uniformity is been found by sensitivity analysis.