A series of electrochemical measurements have been performed on a variety of n-and p-type Si wafers of{100} and {111} orientation in aqueous K~)H. It has been determined that {100} surfaces have similar, systematic features regardless of carrier type and doping level below 102~ -3, while {I 11 } surfaces have a different set of similar systematic features. These results suggest the presence of a different prepassive layer On each of these surfaces. In addition, ellipsometric measurements have been carried out in the voltage range beyond the passivation potential to clarify the anodization process.The importance of orientation dependent (OD) etching in silicon technology has prompted us to study the etch mechanisms with the a~m of improving the OD etching of silicon sufficiently to meet the needs of the microstructure electronics of the future and to devise similar type, equally reliable etchants for other materials (in particular the III-V intermetaUic compounds) which show promise. There are several etchants that are orientation dependent not only for Si (1-6) but also for Ge (7-9), for many of the III-V intermetallic compounds (10),-and for some other semiconductors and a few metals (11). At present most of these are not, or at least have not been shown to be strongly enough OD for uses similar to those of Si. The success of the OD etching in Si device technology is due to the strong orientation dependence of aqueous KOH (and of a few nonaqueous combinations of organic oxidizing and complexing agents).Studies on the kinetics of chemical etching, in the absence of an electric field (open-circuit conditions in electrochemical studies), have been carried out on a number of materials (12-15). For Ge, it was shown (15) that the oxidizing agent controlled the type of attack while the complexing agent controlled the rate of attack; however, neither this study nor any of the other studies could explain why an etch was orientation dependent. Chemical etching has been referred to as a competition between the rate of oxidation by the oxidizing agent and the rate of dissolution of the oxide. In this respect, it is generally considered that there must be a "thick" layer of reaction products to obtain polishing. Attempts have been made to identify the chemical makeup of the reaction layer at the etching interface with little success. Izidinov et al. (16) reported that the gas evolved at the silicon surface during etching with aqueous KOH was hydrogen and suggested some form of silicate was produced (and presumably made up the layer) and went into solution. Neither the composition nor the structure of the reaction ("oxide") layer itself has been identified, although variations of SiOx and/or other silicates have been suggested (1, 3, 17, 18). In our previous electrochemical studies on the action of aqueous KOH on silicon (19), we showed strong evidence to suggest that the "etch-stop" found at high carrier concentrations (> 7 • 1019 am -3) was due to the formation of a prepassive layer. At higher carrier concentrations, this p...