Doping inhomogeneous

But even in a homogeneously doped material an etch stop layer can be generated by an inhomogeneous charge carrier distribution. If a positive bias is applied to the metal electrode of an MOS structure, an inversion layer is formed in the p-type semiconductor. The inversion layer passivates in alkaline solutions if it is kept at the PP using a second bias [Sm5], as shown in Fig. 4.16b. This method is used to reduce the thickness variations of SOI wafers [Og2]. Illuminated regions... 

Fig. 6.3 Optical micrographs of edges of cleaved Si wafers showing different crystal planes anodized at 100 mA cm"2 in ethanoic HF (1 1). (a) The growth rate of meso PS formed on a highly doped n-type substrate (2xl018 crrT3, 2 min) shows a clear dependence on crystal orientation, (b) An orientation dependence is not observed for micro PS formed on moderately doped p-type samples (1.5 xlO16 cm-3, 4 min) but the PS thickness becomes inhomogeneous because of local variations in the current density caused by the edge geometry.

A lateral variation of the anodization current will produce different growth rates and consequently an interface roughness for porous layers. Note that this is not the case for stable macro PS formation on n-type, because here the growth rate is independent of current density. An inhomogeneous current distribution at the O-ring seal of an anodization cell or at masked substrates produces PS layer thickness variations, as shown in Fig. 6.6. Inhomogeneities of the current distribution become more pronounced for low doped substrates, as shown in Figs. 6.3 b and 6.5 d [Kr3]. 

Fig. 6.6 PS layer thickness inhomogeneities as a result of different kinds of masking layers and doping densities, (a) While underetching is minimal for a silicon nitride mask, (b) a resist mask shows severe under-etching.

The electron mobilities at 296 and 420°K are given for several Cr-doped and -doped samples in Table II. The data for the Cr-doped crystals should be considered less accurate since a mixed-conductivity analysis was necessary in most cases (Look, 1980). However, the temperature dependences are not unlike those of conductive GaAs samples with similar impurity concentrations (1016—1018 cm-3). At least two of the crystals (MA 287/80 and MOR 56/76) appeared to be inhomogeneous, as evidenced by nonlinear Arrhenius plots. However, it is doubtful that the bulk of the data require a percolation-type conduction mechanism to be operative, as has been suggested (Robert et ai,... 

It is important to point out that the observed two-component EPR spectra are an intrinsic property of the lightly doped LSCO and are not due to conventional chemical phase separation. We examined our samples using x-ray diffraction, and detected no impurity phases. Moreover, the temperature dependence of the relative intensities of the two EPR signals rules out macroscopic inhomogeneities and points towards a microscopic electronic phase separation. The qualitatively different behavior of the broad and narrow EPR signals indicates that they belong to distinct regions in the sample. First we notice that the broad line vanishes at low temperatures. This... 

---