Skip to Main Content
Your organization might have access to this article on the publisher's site. To check, click on this link:http://dx.doi.org/+10.1116/1.575365
We have incorporated a scanning tunneling microscope (STM) into an UHV analytical chamber containing low‐energy electron diffraction (LEED), reflection high‐energy electron diffraction, Auger, and scanning electron microscopy facilities. A second chamber has facilities for surface cleaning together with an entry lock for loading specimens. The STM incorporates a transfer system for both sample and tip, and fine control for the specimen to tip approach is achieved mechanically. X–Y scanning and data acquisition are fully computerized. This instrument has been used to study the effect of cooling rate after annealing on the surface reconstruction of Si(111). The silicon is cleaned by Ar ion bombardment and then annealed at 920 °C. When the specimen is cooled slowly over several minutes, a sharp 7×7 LEED pattern is obtained, and the STM images show areas of well‐ordered 7×7. In this slow cooling regime, small areas of well‐ordered 2×2 reconstruction have also been seen. As the silicon cooling rate is increased, the 7×7 LEED spots become less distinct. In contrast, the STM images show small areas of well‐ordered 2×1 reconstruction typically a few hundred angstroms across, and very similar to the larger areas of 2×1 seen on cleaved Si(111). These results suggest that with rapid cooling (typically a few hundred degrees per minute), the high‐temperature 1×1 surface is unable to form the energetically favorable 7×7 surface. The 2×1 structure is most easily formed, as is seen on cleaved surfaces. As the cooling rate decreases, the surface is more likely to attain the 7×7 structure, but may be trapped in the 2×2 structure which forms the basic building blocks of the 7×7 reconstruction. To be consistent with these results we conclude that the high‐temperature 1×1 surface must be an ordered structure.