Particle nucleation, growth, and sintering of metallic films on oxide substrates

X-Ray Photoelectron Spectroscopy (XPS) and Low Energy Ion Scattering Spectroscopy (LEIS) have been used to characterize the growth of metal films on oxide substrates. During the vapor deposition of Cu onto the ZnO (0001)-Zn surface and Au onto the TiO2 (110) 1 x 1 surface, it is seen that two-dimensional (2D) islands are initially formed up to a certain average critical coverage (thetacr). At coverages above theta cr, added metal mainly goes on top of the existing islands, forming three-dimensional (3D) islands. Even though thermodynamically the islands would prefer to form 3D islands from the onset of growth, kinetic limitations initially constrain the growth to 2D islands. The dependence of this critical coverage with substrate temperature, metallic vapor flux, and surface defect density are discussed within a kinetic model that incorporates various energetic parameters of metal adatom migration.; Au particles on TiO2 (110) and Pt particles on ZnO (0001)-Zn have been probed using a new experimental technique called Temperature Programmed Low Energy Ion Scattering Spectroscopy (TP-LEIS). This novel technique monitors the area fraction of the surface that metal islands cover during a linear temperature ramp. It is seen that the particles thicken such they conceal less of the substrate as the temperature is increased. The temperature range over which these particles thicken is much larger than predicted using an individual island thickening model or an Ostwald ripening model as outlined by Wynblatt and Gjostein. Microcalorimetry data of Pb/MgO(100) by others in the Campbell group suggest that the surface free energy of particles is not constant, and that it depends on the radii of the individual islands. Since Wynblatt and Gjostein's model assumed a constant surface free energy, their model is revised to incorporate the strong decrease of the surface energy with the island radii. Using this modified Ostwald ripening model, a much larger temperature range of sintering is predicted and is compared directly to experimental TP-LEIS data.