Determination of ion track and shapes with damage simulations on the base of ellipsometric and backscattering spectrometric measurements

Abstract

On the base of geometrical and statistical considerations a damage simulator was created in order to determine the ion track‐radius and ‐shape of ion‐implantation caused damage in single‐crystalline Si. Damage vs. dose curves calculated by spectroscopic ellipsometry (SE) and Rutherford backscattering/channeling spectrometry (RBS/C) measurements, using different doses of 100 keV Xe implantation, gave information about the damage profile in depth. Both methods are required, because of dose‐dependent discrepancies of SE compared with RBS/C [Fried et al., Thin Solid Films 455/456, 404 (2004)]. Different kinds of damage models were investigated to calculate the ion track‐radius and to describe the damages in depth and the shape of ion track. Comparing directly the simulated and the measured damage vs. dose curves, the damage function and the other simulation parameters were optimized and hence the ion track size and even the shape can be determined. The dose dependent mean size of the unchanged crystalline regions, obtained from the simulation was correlated with the complex dielectric functions, obtained from the SE analysis. The results clearly show the effect of decreasing size of the unchanged crystalline regions. (© 2008 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)

In Situ Characterization of Biomaterials at Solid‐Liquid Interfaces Using Ellipsometry in the UV‐Visible‐NIR Wavelength Range

“…Due to the spectroscopic capability and the resulting large number of measured data, complex models with many parameters can and have to be constructed. To find the global minimum in such a system, sophisticated methods and algorithms have to be applied, but even a random parameter search before starting the gradient method is a very important technique to avoid local minima.…”

(Invited) High Sensitivity Optical Characterization of Thin Films with Embedded Si Nanocrystals

“…The first one is a survey of the parameterizations of the dielectric function [1,2,3,4] from the effective medium approximation to model dielectric functions, also in combination. The second one is the use of parameter search algorithms that helps to avoid or at least to minimize the probability of getting into local minima during the fitting of numerous model parameters [4,5,6,7]. Using oscillator models for the description of the dielectric function the number of fit parameters is relatively high.…”