Pretty Pictures! 
Cool Videos 
Research Group
Pretty Pictures! Cool Videos Research Group
Atomic-Scale Surface Dynamics: Surface processes such as adsorption, diffusion (along the surface and into the bulk), reaction, and desorption are crucial steps in heterogenous catalysis, nucleation and growth of thin films, corrosion, and etching. Our research focuses on determining the detailed mechanisms of these surface processes by direct observation with atomic-scale microscopy.
The development of scanning tunneling microscopy (STM) revolutionized the study of surfaces. Recently, STM has been used to observe dynamic processes on surfaces on an atomic-scale; in our work we observed the reorientation of fullerene molecules adsorbed on Au and Ag surfaces. Since virtually all STMs operate only at room temperature, it has not been possible to obtain kinetic data such as activation energies from these observations. Therefore, we have built a novel high-speed variable-temperature ultra-high vacuum STM specifically to study the dynamics of surface processes on an atomic-scale. We are using this instrument along with standard surface science techniques to study the mechanisms of etching reactions of metal surfaces, to determine the importance of surface restructuring on the nucleation and growth of thin metal films, and to determine the structure and stability of reactive sites on oxide catalyst surfaces. These research areas are described in more detail below.
Atomic-Scale Mechanisms of Metal Etching, Oxidation and Corrosion: Etching reactions of semiconductors and metals are extensively used in the microelectronics industry. For metals, etching reactions are used to define patterns for gates and on-chip interconnects. Currently, there is intense interest in substituting Cu for Al in on-chip interconnects to reduce interconnect resistance. However, the introduction of commercial devices with Cu interconnects is being delayed by the lack of suitable etching processes.
The development of etching processes for metals has been hindered by a lack of detailed understanding of the elementary reaction steps occurring at the surface. Previous studies have shown that etching of metals generally proceeds in two steps: 1) formation of a growing compound layer (usually a halide), and 2) removal of the compound by sublimation or reaction to form a volatile species. Neither of these steps have been well characterized and unusual phenomena such as reaction rates decreasing with temperature have been reported, thus motivating the present study. Our approach is to independently characterize the kinetics of the two steps on a macroscopic scale using temperature programmed desorption/reaction (TPD/R) and to relate these kinetics to atomic-scale events observed in STM "movies." The objective of this project is to develop a mechanistic kinetic model to predict etching behavior from the atomic-scale measurements. The longer term goal is to use the model as a basis for developing etching processes that allow new materials to be used in integrated circuits and to develop etching schemes that allow direct patterning of fine metal lines.
More recently we have been working on the related oxidation reactions. Specifically, we have been interested in Pd oxidation because of interest in Pd as an oxidation catalyst. Under reaction conditions the Pd can be repeatedly oxidized and reduced thus systematic catalyst development dictates that we understand how these changes impact the activity of the catalyst. Thus far we have shown that Pd oxidation is a complex process that proceeds through three distinct stages involving as many as four surface phases. We are now working on characterizing the catalytic activity of the different surface phases/
Nucleation and Growth of Thin Films: Thin film growth has been considered to proceed through three mechanisms: 1) layer-by-layer growth; 2) three- dimensional cluster growth; and 3) layer-by-layer growth followed by cluster growth. Recently, a fourth growth mode has been observed: substrate atom displacement resulting in surface intermixing. This has been observed even for systems that are immiscible in the bulk. Analyses suggest that intermixing may always be favored suggesting that the growth mode is determined by the kinetics of adatom-substrate exchange processes. Therefore we have been studying adatom-substrate exchange mechanisms with the goal of developing a model to predict the conditions that lead to surface intermixing during film growth. We have also begun to explore ion beam modification of metal surfaces with the objective of altering the surface morphology to remove the driving force for intermixing, and to explore the use of surfactants to alter growth modes, and the use of surfactants to manipulate the growth mechanism.
The Structure and Stability of Reactive Sites on Oxide Catalyst Surfaces: Metal oxides represent an industrially important class of heterogeneous catalysts. Despite the importance of oxide catalysts, the structure of the reactive sites on oxide catalyst surfaces remains poorly understood. For many catalytically important oxides it has been shown that low densities of defects can dominate the adsorption properties and may be responsible for the catalytic activity. To a first approximation, one might assume that the defects are Lewis acid sites or isolated oxygen vacancies that expose metal ions. However, many of the catalytically important oxides, display complex reduction chemistry and isolated surface oxygen vacancies may not be stable at reaction temperatures.
Since the defect concentrations can be low and it is unlikely that ordered arrays of defects form, real-space techniques are required to study the electronic, structural, and chemical properties of defects on oxide surfaces. Therefore, STM and related techniques are the ideal tools to study catalytic sites on oxide surfaces. By varying the sample temperature, we will obtain kinetics for specific surface sites. In addition, the STM tip can also be used as a source of low energy electrons for surface modification. The reactivity and stability of these "modified" surfaces can then be studied by STM.
Thus far we have completed a detailed study of surface structure and reactivity of tungsten oxide surfaces. We have shown that the adsorption sites for alcohols are fully oxidized, exposed W6+ cations on the surface, that alkoxides bound to these sites only undergo dehydration reactions to yield olefins, and that olefin desorption temperature decreases as the alkoxide becomes easier to dehydrate suggesting that C-O bond scission is the rate limiting step in the process.
The objective of this study is to identify the key structural features required for oxide catalysis and the processes that are important in determining the types of sites that will be available under reaction conditions. These results will help guide development of new oxide materials with stable catalytic sites.
Scanning Tunneling Microscopy (STM) Images
Click on any of the pictures for a full-size version and description.
Br2 Induced Step Faceting of a Cu Surface
Br2 Etching of Cu(100)
C60 Layers on Ag(111).
Scanning Tunneling Microscopy (STM) Movies
Click any of the links to view the movie and a description.
Reaction of Br2 with Cu(100) to form CuBr
Vacancy diffusion in a layer of adsorbed Br atoms on Cu(100).
Thermal fluctuations at step edges in an adsorbed layer of Br atoms on Cu(100).
Kink diffusion at a peninsula formed by the reaction of Br2 with Cu(100).
Step faceting during exposure of Br2 to a stepped Cu surface.
Step consumption during Cu halide formation
CuBr cluster restructuring.
12/23/97
(1,238k avi)
(1,911k avi)
(293k avi)