Publications

A physics-based understanding of material aging mechanisms helps to increase reliability when predicting the lifetime of mechanical and electrical components. This report examines in detail the mechanisms of atmospheric copper sulfidation and evaluates new methods of parallel experimentation for high-throughput corrosion analysis. Often our knowledge of aging mechanisms is limited because coupled chemical reactions and physical processes are involved that depend on complex interactions with the environment and component functionality. Atmospheric corrosion is one of the most complex aging phenomena and it has profound consequences for the nation's economy and safety. Therefore, copper sulfidation was used as a test-case to examine the utility of parallel experimentation. Through the use of parallel and conventional experimentation, we measured: (1) the sulfidation rate as a function of humidity, light, temperature and O{sub 2} concentration; (2) the primary moving species in solid state transport; (3) the diffusivity of Cu vacancies through Cu{sub 2}S; (4) the sulfidation activation energies as a function of relative humidity (RH); (5) the sulfidation induction times at low humidities; and (6) the effect of light on the sulfidation rate. Also, the importance of various sulfidation mechanisms was determined as a function of RH and sulfide thickness. Different models for sulfidation-reactor geometries and the sulfidation reaction process are presented.

Simulation of chemical vapor deposition (CVD) in submicron features typical of semiconductor devices has been facilitated by extending the EVOLVE thin film etch and deposition simulation code to use thermal reaction mechanisms expressed in the Chemkin format. This allows consistent coupling between EVOLVE and reactor simulation codes that use Chemkin. In an application of a reactor-scale simulation code providing surface fluxes to a feature-scale simulation code, a proposed reaction mechanism for TEOS pyrolysis to deposit SiO{sub 2}, which had been applied successfully to reactor-scale simulation, is seen not to predict the low step coverage over trenches observed under short reactor residence time conditions. An apparent discrepancy between the mechanism and profile-evolution observations is a reduced degree of sensitivity of the deposition rate to the presence of reaction products, i.e., the byproduct inhibition effect is underpredicted. The cause of the proposed mechanism's insensitivity to byproduct inhibition is investigated with the combined reactor and topography simulators first by manipulating the surface to volume ratio of a simulated reactor and second by calibrating parameters in the proposed mechanism such as the calculated free energies of surface molecules. The conclusion is that the byproduct inhibition can not be enhanced to fit profile evolution data without comprising agreement with reactor scale data by simply adjusting mechanism parameters. Thus, additional surface reaction channels seem to be required to reproduce simultaneously experimental reactor-scale growth rates and experimental step coverages.