The role of vacancies in surface diffusion

From the numbers that are mentioned above it is obvious that the direct observation of surface vacancies is a formidable, if not an impossible, experimental challenge. The very combination of a low density 

Two alternative approaches exist. The first one involves significantly lowering the temperature to values where the diffusion of vacancies can be observed with a technique like STM. At lower temperatures a surface vacancy can then be artificially created by ion bombardment or direct removal of an atom by the tip. This approach has been applied successfully to several semiconductor surfaces [29-31]. For metal surfaces, although vacancy creation at a step by direct tip manipulation of the surface has been demonstrated [32], to our knowledge, no studies have been published where the diffusion of artificially created vacancies in a terrace has successfully been measured. The second approach involves the addition of small amounts of appropriate impurities that serve as tracer atoms in the first layer of the surface [20-24]. The presence and passage of a surface vacancy is indirectly revealed by the motion of these embedded atoms. If one seeks to measure both the formation energy and the diffusion barrier of surface vacancies explicitly, a combination of these two approaches is needed. 

In the remainder of this chapter we review measurements that were based on the dilute surface alloy approach for two different systems. We first discuss experimental results of the In/Cu(00 1) [21-23] and Pd/Cu(00 1) [24] systems. After this we discuss the theoretical framework that was developed to analyze the diffusion measurements [33], We then discuss how to use this framework to interpret the measurements and discuss the differences between these two material systems. We also discuss the effect of steps on our measurements. Finally, in Section 6, we put the measurements in a broader context and discuss the relevance of these results for surfaces in general. 

---