We are studying the relationship between structure and properties of grain boundaries and surfaces of transition metals. Using molecular dynamics and similar computer simulation techniques, we can calculate the positions of atoms at grain boundaries and surfaces; our results agree with experiment to within about 0.1A, the limit of experimental accuracy. We are studying other properties of surfaces, including diffusion and thin-film growth. For grain boundaries, we are studying diffusion, mechanical properties, and crack formation and growth.

The structure and bond strength of polymer-metal interfaces is investigated using first principles quantum mechanics. Molecular dynamics simulations are carried out to determine the optimal positions of polymer chains adsorbed on metal and metal oxide surfaces. The force required to pull the polymer chains off the substrates can be calculated to determine the adhesive strength. A variety of functional groups are being investigated to determine their effect.

We are investigating the atomic-level reactions relevant to thin-film growth by chemical vapor deposition for several materials, including diamond, amorphous Si, and crystalline Si. For example, we investigate how the reactive species in the gas (such as SiH₃) land on the surface, diffuse about, and then either desorb or chemisorb, resulting in growth. We are also investigating the structure of surfaces in amorphous Si, which is useful for solar cell and other applications.

We are studying how metallic clusters catalyze chemical ractions relevant to reducing automotive engine exhaust. For example, we study how gases such as deadly CO plus O₂ react to form harmless CO₂, thereby reducing automotive pollution. Our work involves computer simulations of the structure of the clusters and of the catalytic reactions of the clusters on the clusters to determine how their catalytic properties can be enhanced in a cost-effective way.

We are developing empirical models for high-temperature intermetallics, such as Ti-Al alloys, and using those models to study the mechanical properties of these alloys. We are especially concerned with the low ductility (at room temperature) of these otherwise useful materials.

We are simulating the implantation of high-energy ions into Si wafers to determine how much damage is initially created. Then, using some macroscopic simulation methods, we study how far the dopants diffuse when the damage is annealed out.

We are studying how metallic clusters catalyze chemical reactions relevant to reducing automotive engine exhaust. For example, we study how gases such as deadly CO plus O₂ react to form harmless CO₂, thereby reducing automotive pollution. Our work involves computer simulations of the structure of the clusters and simulations of the catalytic reactions of the clusters on the clusters to determine how their catalytic properties can be enhanced in a cost-effective way. This work involves a close collaboration with Ford Motor Co.

We are developing large-scale models of polymer chains based on treating polymers as a set of connected segments. Using these models, we are simulating the fracture of adhesively bonded structures to determine how the fracture mechanisms and strength depend on type of stress, surface treatment, polymer thickness, bulk polymer properties, and environmental effects. This project is funded by several major companies and involves a close collaboration with them.

Using a common set of computational methods, we are investigating a variety of phenomena involving metallic and semiconductor surfaces and interfaces. Our three main projects include: (1) segregation: modeling the grain boundary structure in Al-Mg alloys to determine the amount of segregation of Mg and its effect on mechanical properties; (2) catalysis: Rh is used in automotive catalytic converters to reduce NO emissions, and we are studying how the NO adsorbs and reacts; (3) CVD growth: we are studying the structure of a-Si:H and the growth mechanisms involved during chemical vapor deposition.

Computer simulations of time-dependent processes can b e written using discrete systems of lattice sites whose values synchronously change in discrete time steps as a result of local interactions with other sites. We are developing cellular automata programs of a wide variety of physical, chemical, biological, engineering, and social phenomena (e.g., spinoidal decomposition, heterogeneous catalysis, predator-prey ecosystems, traffic flow, social cooperation).

This research aims to elucidate the relationship between structure and elementary diffusion mechanisms in noncrystalline substances. It addresses practical problems encountered with molten electrolytes, fast ion conductors, and mixed alkali glasses. Molecular dynamic simulations are used to model the diffusive displacements of structural components in amorphous materials. The spectral analysis of the diffusion trajectories reveals information concerning correlation coefficients associated with the elementary diffusion jump and the relaxation times that characterize the activated processes. The diffusion mechanisms of mobile species in amorphous structures with various degrees of complexity are systematically investigated.

This research focuses on modeling the removal of volatile species from particulate ceramic films during drying. Specifically, 2-D and 3-D cellular automata models have been developed that contain both random walk and invasion percolation algorithms. These models are being used to determine the removal kinetics, effective diffusivities, and geometry of the developing pore front as a function of relative volatility and composition of the multicomponent fluid, volatile to particle size ratio, and particle size distribution and loading.