Some electrical properties of the silicon-silicon dioxide system

Abstract

As a result of its importance in the planar process for transistor fabrication the silicon-silicon dioxide interface is probably the most well characterised solid- solid interface. Over the past decade extensive investigations of the electrical properties of this interface have been carried out. These investigations have shown that, after taking into account impurity redistribution during oxidation and metal- silicon work function differences in metal-oxide-silicon (MOS) structures, thermally grown silicon dioxide always has a positive charge associated with it. Thus the underlying silicon has a negative charge induced in it, and the surface of “p”-type silicon is either depleted or inverted, while the surface of “n”-type silicon is accumulated. This positive surface charge is associated with at least four types of centres; these are: fixed surface state or interface charge, which does not vary with the silicon surface potential, mobile charge in the oxide probably due to contamination with alkali ions or protons, fast surface states whose charge occupancy does depend on the silicon surface potential, and traps within the oxide which can be ionised by incident radiation. In this paper an attempt is made to summarise some of these electrical characteristics of the silicon-silicon dioxide interface and the techniques used to control them.