QUANTITATIVE-ANALYSIS OF ELECTRON SPECTROSCOPIC IMAGING SERIES

Citation
J. Mayer et al., QUANTITATIVE-ANALYSIS OF ELECTRON SPECTROSCOPIC IMAGING SERIES, Micron, 28(5), 1997, pp. 361-370
Citations number
27
Journal title
MicronACNP
ISSN journal
09684328
Volume
28
Issue
5
Year of publication
1997
Pages
361 - 370
Database
ISI
SICI code
0968-4328(1997)28:5<361:QOESIS>2.0.ZU;2-Q
Abstract
We have developed new methods to quantify the data acquired by electro n spectroscopic imaging (ESI) which are based on recording series of e nergy filtered images across inner shell loss edges or in the low loss region. From the series of ESI images, electron energy loss (EEL) spe ctra can be extracted for any given image area, i.e. each individual p ixel or an array of pixels over which the signal is integrated. The EE L spectra can subsequently be analysed using standard EELS quantificat ion techniques. This makes it possible to obtain a measure of the abso lute amount (area density) of an element in the given sample area or o f the concentration ratios of one element with respect to other elemen ts. From a series of ESI images in the low-loss region, accurate value s of the specimen thickness can be obtained, provided the mean free pa th for inelastic scattering is known. As examples, results obtained on Si2N2O ceramics and thin Al2O3 films, which were grown by anodical ox idation, will be shown. The number densities of the atoms and concentr ation ratios can be measured with an accuracy of 10-15% using calculat ed inelastic scattering cross-sections. Similar accuracies can be obta ined for the measurements of the specimen thickness, as will be shown for the thin Al2O3 films and for a wedge shaped silicon crystal. In th e case of Si, convergent beam electron diffraction was employed to det ermine the mean free path for inelastic scattering. For the same sampl e, the thickness of the carbon contamination layers and the amorphous surface oxide layers were measured. The results of the latter measurem ents may provide important information for the quantitative evaluation of high resolution images or CBED patterns. (C) 1997 Elsevier Science Ltd. All rights reserved.