| Gas-phase etching and deposition are among the most important processes in modern integrated circuit manufacturing and Micro-Electro-Mechanical-Systems (MEMS) fabrication. Developing simulation tools driven by improved etching and deposition models can facilitate an in-depth understanding of various processes. It can also help bring down the development cost of new processes.; The focus of this work was on developing analytical self-consistent feature scale models, which were incorporated into our etch and deposition profile simulator, SPEEDIE. A self-consistent approach, in both flux transport and surface velocity calculations, enabled us to analytically model more complex surface kinetics such as chemical reactions. It also gave the models more physical meaning and empowered us with tools to discover phenomena, which would otherwise go unnoticed.; Our approach was based on Langmuir adsorption model and mass/site balance equations. Using this approach, surface recombination of species was recognized as an important surface reaction mechanism, especially for predicting Aspect Ratio Dependent Etching (ARDE). In addition, a model was developed for the calculation of ion flux reflection from feature sidewalls in order to simulate micro-trenching phenomenon in silicon etching.; A similar self-consistent approach was used to incorporate chemical reaction rates in analytical feature scale simulations. This reaction rate method was verified for tungsten chemical vapor deposition. The site balance approach was used to develop a self-consistent model for ion-enhanced deposition of CFx polymer in C4F8 plasma. This deposition is an important part of the Bosch deep trench etch process. |
