Simulation Based Design Optimization And Control Of Silicon Carbide And Gallium Nitride Thin Film Chemical Vapor Deposition Reactor Systems

ABSTRACT: The design and development of a reactor to make this process controlled and repeatable can be accomplished using theoretical and empirical tools. Fluid flow modeling, reactor sizing, low-pressure pumping and control are engineering concepts that were explored. Work on the design and development of an atmospheric pressure cold-wall CVD (APCVD) reactor will be presented. A detailed discussion of modifications to this reactor to permit hot-wall, low-pressure CVD (LPCVD) operation will then be presented. The consequences of this process variable change will be discussed as well as the necessary design parameters. Computational fluid dynamic (CFD) calculations, which predict the flow patterns of gases in the reaction tube, will be presented. Feasible CVD reactor design that results in laminar fluid flow control is a function of the prior mentioned techniques and will be presented.

Semiconductor equipment modeling has in recent years become a field of great interest, because it offers the potential to support development and optimization of manufacturing equipment and hence reduce the cost and improve the quality of the reactors. This book is the result of two parallel lines of research dealing with the same subject - Modeling of Tungsten CVD processes -, which were per formed independently under very different boundary conditions. On the one side, Chris Kleijn, working in an academic research environment, was able to go deep enough into the subject to laya solid foundation and prove the validity of all the assumptions made in his work. On the other side, Christoph Werner, working in the context of an industrial research lab, was able to closely interact with manufacturing and development engineers in a modern submicron semiconductor processing line. Because of these different approaches, the informal collaboration during the course of the projects proved to be extremely helpful to both sides, even though - or perhaps because - different computer codes, different CVD reactors and also slightly different models were used. In spite of the inconsistencies which might arise from this double approach, we feel that the presentation of both sets of results in one book will be very useful for people working in similar projects.

"Chemical Vapour Deposition: An Integrated Engineering Design for Advanced Materials" focuses on the application of this technology to engineering coatings and, in particular, to the manufacture of high performance materials, such as fibre reinforced ceramic composite materials, for structural applications at high temperatures. This book aims to provide a thorough exploration of the design and applications of advanced materials, and their manufacture in engineering. From physical fundamentals and principles, to optimization of processing parameters and other current practices, this book is designed to guide readers through the development of both high performance materials and the design of CVD systems to manufacture such materials. "Chemical Vapour Deposition: An Integrated Engineering Design for Advanced Materials" introduces integrated design and manufacture of advanced materials to researchers, industrial practitioners, postgraduates and senior undergraduate students.

In December 1996, a project was initiated at the Institute for Systems Research (ISR), under an agreement between Northrop Grumman Electronic Sensors and Systems Division (ESSD) and the ISR, to investigate the epitaxial growth of silicon[germanium (Si[Ge) heterostructures in a commercial rapid thermal chemical vapor deposition (RTCVD) reactor. This report provides a detailed account of the objectives and results of work done on this project as of September 1997. The report covers two main topics - modeling and model reduction. Physics[based models are developed for thermal, fluid, and chemical mechanisms involved in epitaxial growth. Experimental work for model validation and determination of growth parameters is described. Due to the complexity and high computational demands of the models, we investigate the use of model reduction techniques to reduce the model complexity, leading to faster simulation and facilitating the use of standard control and optimization strategies.

This work describes the development of a mathematical model of ahigh-pressure chemical vapor deposition (HPCVD) reactor and nonlinearfeedback methodologies for control of the growth of thin films in thisreactor. Precise control of the film thickness and composition is highlydesirable, making real-time control of the deposition process veryimportant. The source vapor species transport is modeled by the standardgas dynamics partial differential equations, with species decomposition reactions, reduced down to a small number of ordinary differential equationsthrough use of the proper orthogonal decomposition technique. This systemis coupled with a reduced order model of the reactions on the surfaceinvolved in the source vapor decomposition and film deposition on thesubstrate wafer. Also modeled is the real-time observation technique usedto obtain a partial measurement of the deposition process. The utilization of reduced order models greatly simplifies the mathematical formulation of the physical process so it can be solved quickly enough to beused for real-time model-based feedback control. This control problem isfairly complicated, however, because the surface reactions render the modelnonlinear. Several control methodologies for nonlinear systems are studiedin this work to determine which performs best on test examples similar tothe HPCVD problem. One chosen method is extended to a tracking control toforce certain film growth properties to follow desired trajectories. Thenonlinear control method is used also in the development of a stateestimator which uses the nonlinear partial observation of the nonlinearsystem to create an estimate of the actual state, which the feedback controlformula then can use to guide the HPCVD system. The nonlinear trackingcontrol and estimator techniques are implemented on the HPCVD model and theresults analyzed as to the effectiveness of the reduced order model andnonlinear control.