Measuring with a spectroscopic ellipsometer (SE) in the 1.845 eV photon energy region we determined the complex dielectric function (E = e1 + ieZ> of different kinds of amorphous silicon prepared by self-implantation and thermal relaxation (500 "C, 3 h) . These measurements show that the complex dielectric function (and thus the complex refractive index) of implanted a-Si (i-a-Si) differs from that of relaxed (annealed) a-Si (r-a-Si) . Moreover, its E differs from the E of evaporated a-Si (e-a-Si) found in the handbooks as E for a-Si. If we use this E to evaluate SE measurements of ion implanted silicon then the fit is very poor. We deduced the optical band gap of these materials using the Davis-Mott plot based on the relation: (E&'>~'~ -(E -E8). The results are: 0.85 eV (i-a-Si), 1.12 eV (e-a-Si), 1.30 eV (r-a-Si) . We attribute the optical change to annihilation of point defects.Understanding the structure and properties of amorphous silicon (a-Si) is a scientific challenge of some complexity.' The problem is that there is not a single type of a-Si. There are several measurements which indicate that the properties of implanted amorphous silicon (i-&i) differ significantly from those of well relaxed (annealed) a-Si (r-u-Si) .&' Additionally, we must know the optical constants of different kinds of a-Si to evaluate well the optical measurements. Earlier measurements' showed that the E (complex dielectric function, c = e1 + ieZ) of i-&i and r-a-Si differ from the E of evaporated a-Si (e-a-Si) found in the handbooks as E for aSi.' If we use this E to evaluate spectroscopic ellipsometric (SE) measurements of ion implanted silicon then the fit is very poor.' A similar thing is described by McMarr" and Vedam, McMarr, and Narayan" who also measured self-implanted fully amorphous silicon and tried to evaluate the spectra modeling the sample as a mixture of voids and a-Si prepared by low-pressure chemical vapor deposition (LPCVD) . l2 The model calculations resulted in a surprising -9% void fraction. This fact also indicates that the E of LPCVD a-Si must not be used for i-a-Si.For the experiments Wacker made, p-type Si (100) wafers of 4-8 Sz cm resistivity were implanted with Si ions at room temperature. The implantation conditions are shown in Table I. After implantation an adequate plasma stripping procedure13 was applied to remove a possible hydrocarbon deposition. The SE measurements were followed by annealing (500 "C, 3 h, N2 ambient) to achieve the well relaxed state.2'3'7 (Well relaxed means that longer or higher temperature annealing does not change the optical properties of the layer.)The ellipsometric measurements were performed with a rotating analyzer type ellipsometer in the 270-700 nm wavelength (4.6-1.8 eV photon energy) region at 70", 73", and 75" angle of incidence at the Twente Technical University. From these multiple-angle-of-incidence measurements we could take into account the thickness of the native oxide. The measurement error increased at the red end of the spectrum because of the light source (...

The influence of cold work on the initially formed oxide layer on the stainless steels AISI 304 and Incoloy 8OOH has been studied by XPS. Oxidations were performed at pressures of 10-6-10-4Pa and temperatures of 300-800 K. All samples showed a similar oxidation behaviour. The oxidation rates of iron and chromium are of the same order of magnitude at temperatures below 650 K. Subsequent oxidation results in an iron oxide on top of a chromium oxide layer. At temperatures above 650 K the metal surface becomes enriched in chromium, which is preferentially oxidized at these temperatures and pressures. Even prolonged oxidation does not result in an iron-rich oxide surface. Nickel has never been found in its oxidized form. The binding energy of oxygen, in the various oxide layers, is independent of the extent of oxidation and is 530.6 eV.

A several-parameter fitting of spectroscopic ellipsometry data is developed to characterize near-surface layers in semiconductors damaged by implantation. The damage depth profiles are described by either rectangular, trapezoid-type, or coupled half-Gaussian (realistic) optical models. The rectangular model has three parameters: the average damage level, the effective thickness of the implanted layer, and the thickness of the native oxide. The trapezoid-type model is enhanced with a fourth parameter, the width of the amorphous/crystalline interface. The realistic optical model consists of a stack of layers with fixed and equal thicknesses. The damage levels are determined by a depth profile function (presently coupled half-Gaussians). Five parameters are used: the position of the maximum, the height, and two standard deviations of the profile, plus the thickness of the native oxide. The complex refractive index of each layer is calculated from the actual damage level by the Bruggeman effective medium approximation. The optical models were tested on Ge-implanted silicon samples and cross checked with high-depth-resolution Rutherford backscattering spectrometry and channeling.

No abstract

No abstract