Mechanisms of metal—semiconductor interface interactions

It is evident from the preceding sections that, with a few exceptions, adsorption of a metal on a clean semiconductor surface leads to pronounced solid state reactivity. We may reasonably ask, therefore, whether it is possible to establish a general mechanism for this interface instability. Specifically, an explanation is required for solid state reactions and transport processes which are occurring at comparatively high rates at room temperature, where they might have been expected to be negligible. Two plausible models have been advanced, one by Tu et al. [324, 325] and the other by the Spicer group [298, 326, 327]. 

At this stage, there is insufficient evidence to decide which, if either, of these models is correct, or indeed whether a unified explanation is possible. We should point out that Tu s model was derived principally for Si, whilst Spicer s proposals were based on work on Group III—V compounds, but it is quite evident that there is a common feature of high interface mobility. 

As an aside, it is interesting to note that these interface effects make the classical theory of Schottky barriers basically untenable. A complete 

The corresponding explanation for silicon, advanced by Ottaviani et al. [325], is that the specific reactive interface (i.e. silicide—silicon) determines the barrier height, but again it is the result of a metal—semiconductor interaction. 

---