Starting in the 1970s, considerable work was done on dimers and trimers and their surface diffusion, but there were no experimental studies of larger clusters, containing twenty or more atoms, since they were assumed to be immobile at the surface. This changed in 1984, with the work of Fink using the FIM, in which he assembled a cluster of twenty or more palladium atoms on the (110) plane of tungsten. At 390 K, this large cluster moved over the surface as a unit, as shown in Fig. 9.1, demonstrating its diffusivity. Large clusters turn out to be mobile at relatively low temperatures and their movement needed to be investigated, since it influences the stability of nanostructures and thin film growth kinetics. With the invention of the scanning tunneling microscope, large clusters were rediscovered a few years later, and work began to unravel how diffusion occurred, many of the studies focusing on the dependence of diffusivity on cluster size. This effort will be surveyed, arranged according to the type of the surface. Study of large clusters began with the examination of movement on a bcc surface, on W(110), but this work was not continued later; instead fcc surfaces were investigated in detail.

Large clusters on fcc(100) surfaces

Theoretical investigations of large clusters on fcc(100) surfaces started in 1980 with the work of Binder and Kalos, which initiated a number of discussions of how the cluster diffusivity D was affected by the size and the specific mechanism of diffusion.

Surface diffusion studies on single adsorbed entities, the focus of our presentation, had to await the development of techniques capable of revealing atoms. This was first accomplished by Müller roughly fifty years ago in 1956, with his invention of the field ion microscope (FIM). The natural extension of FIM was the development of the Atom Probe which allowed identification of chemical identities and control of composition for surfaces, but there also were earlier investigative methods, such as field emission microscopy, helium scattering, contact potential measurements and so on, which provided useful information about surface diffusion. Today there are newer techniques that have been shown to have the capability of revealing atoms. The scanning tunneling microscope (STM), devised by Binnig and Rohrer in 1983 is one of them. Less frequently used techniques, such as measurements of work function changes, perturbed angular correlation, or atomic beam scattering will also be mentioned, if only very briefly. Insights into diffusion phenomena on the atomic scale gained with the scanning tunneling microscope are certain to grow in number and importance. Both field ion and scanning tunneling microscopy have been covered extensively in the literature, and will also be described here in reference to diffusion studies. It should be noted that for examination of diffusion phenomena on clean surfaces by any of these techniques, good vacuum conditions are crucial.