Elektrochemische Prozesse an Halbleitern

Abstract

Electrochemical reactions at semiconductors are a great deal different from those at metals. In the case of a metal, the rate of an electrochemical process is controlled by the high electric field (in the order of 10⁷ volt cm^-1) on the surface. On semiconductors, however, the rate is mainly controlled by the concentration of electronic charge carriers in the surface. Therefore, the type of majority carriers in the semiconductor (electrons or holes) can influence the reaction rate to a very large extent.
In many systems investigated so far the presence of holes has been found necessary for anodic dissolution (oxidation) processes. Some details of the anodic oxidation mechanism are discussed for Germanium and a review on other semiconductors with increasing band gaps is given for the same process. In all cases where holes cannot be generated thermally anodic oxidation of the semiconductor requires either illumination where holes are generated by absorption of light, or very high electric fields in the semiconductor surface where holes can be generated by internal field emission.
The role of holes for the anodic oxidation of a semiconductor crystal can be explained in general terms as an effect of bond weakening in the presence of holes. If one of the two electrons in a bonding orbital between two neighboring atoms has been excited to an antibonding orbital, the resulting electronic defect in a crystal surface represents a place where an oxidizing reaction can start much easier than at an intact bond. Thus, in the presence of a hole, the reaction can start from a higher energy state which reduces the activation energy.
A specific classification of the electrons involved is possible for redox reactions, where the semiconductor only exchanges electrons with a redox couple in solution. Electrons can come from or be incorporated into the conduction or the valence band. The latter case can better be described as an exchange of holes, since the electron leaving an energy state in the valence band leaves a hole behind and vice versa, an electron can only enter the valence band if a hole is available.
It is described in what way this distinction can be realized experimentally by various techniques and some results are reported. It has been found that normally only either exchange of conduction electrons or of holes occurs, depending on the redox system and the type of semiconductor.
Under the assumption that the electron exchange in such redox processes, since it is a radiationless transition, has its greatest probability for electronic energy states of equal energies, a model can be derived which gives a good understanding for kinetic correlations.
This model further allows to predict whether a redox couple should interact preferentially with the conduction or the valence band.