Multiscale Modeling and Control of Microstructural Defects and Surface Morphology in Thin Film Deposition.

摘要

Thin film deposition is a key process used in the manufacturing of microelectronic devices as well as in the manufacturing of thin film solar cells. Improving the ability to operate and control thin film deposition such that the deposited films have desired levels of internal microstructure and surface morphology is an issue of major technological significance, with significant implications in the efficiencies of microelectronic devices and thin film solar cells. While significant progress has been made over the last ten years on modeling and control of thin film surface roughness and porosity, there is a number of important unresolved practical issues with respect to our ability to implement the existing solutions for porosity control in practice as well as with respect to control of surface morphology metrics that directly regulate light trapping properties of thin films.;This dissertation presents a unified and practical framework for multiscale modeling and control of thin film internal microstructure and surface morphology in thin film deposition processes. Multiscale modeling provides the link between microscopic film properties like film porosity and surface morphology and macroscopically-controlled process variables like temperature and precursor flow rate and concentration. Within this multiscale modeling framework, model predictive control is used to develop novel control problem formulations and manipulated input trajectories which account for control actuator constraints and lead to a balanced trade-off in the closed-loop system between the possibly conflicting control objectives of film porosity and film surface roughness and/or slope. The proposed multiscale modeling and control methods are applied to a series of complex thin film deposition processes and extensive simulation studies are carried out to evaluate the resulting closed-loop system performance and robustness in terms of achieving key film quality metrics such as amount of internal defects and light trapping efficiency of film surface.