Absolute coverage of the saturated cesium silicon interfaces

Metal/semiconductor interfaces are of great interest for a variety of reasons. They shed light on surface metal/semiconductor transitions, and they form Schottky barriers, which are of scientific as well as significant technological importance (primarily for the production of high speed, low-power logic circuitry). The cesium/silicon interfaces are of particular interest since the oxide forms a Negative Electron Affinity state. Cesium does not mix with the bulk silicon, so the interface is very abrupt, and the electronic structure of cesium is easier to understand than that of the transition metals. Further, cesium (like other alkali metals at room temperature) forms a single atomic layer on the various silicon faces and then the coverage saturates (i.e. atoms stop sticking to the surface). This makes the cesium/silicon interfaces ideal model systems of the metal/semiconductor interface. In spite of their importance, the detailed structures of the cesium saturated silicon faces are still uncertain. Numerous structural models have been proposed and many of them have quite different absolute coverages. Thus absolute coverage measurements can effectively distinguish between the various models. Rutherford Backscattering Spectrometry (RBS) provides an ideal measurement of absolute coverage since its results can be directly interpreted without dependence upon any structural model. A new beam line has been set up on the Laboratory for Research on the Structure of Matter\u27s tandem accelerator. The Ultrahigh Vacuum system is equipped with an Auger Electron Spectrometer, a Low Energy Electron Diffraction system, a retarding field method work function analyzer, a cesium doser, a Medium Energy Ion Scattering two dimensional toroidal analyzer and a Rutherford Backscattering: Spectrometry (RBS) solid state ion detector. It has been used to manufacture saturated Cs/Si(100)-2 x 1 and Cs/Si(111)-7 x 7 interfaces and measure their absolute coverage via RBS. The coverage for the Si(111)-7 x 7 surface was found to be 4.0 ± 0.2 × 1015 atoms/cm2 , rejecting the structural model of Magnusson et al. and supporting the model of Eteldäniemi et al. The coverage for the Si(100)-2 × 1 surface was found to be 3.7 ± 0.1 × 1015 atoms/cm2, rejecting the Abukawa and Kono model of the interface, and supporting the Levine (or a similar) model