Quantitative energy-filtering transmission electron microscopy in materials science

Citation
W. Grogger et al., Quantitative energy-filtering transmission electron microscopy in materials science, MICROS MICR, 6(2), 2000, pp. 161-172
Citations number
65
Categorie Soggetti
Multidisciplinary,"Spectroscopy /Instrumentation/Analytical Sciences
Journal title
MICROSCOPY AND MICROANALYSIS
ISSN journal
14319276 → ACNP
Volume
6
Issue
2
Year of publication
2000
Pages
161 - 172
Database
ISI
SICI code
1431-9276(200003/04)6:2<161:QETEMI>2.0.ZU;2-V
Abstract
Energy-filtered transmission electron microscopy (EFTEM) can be used to acq uire elemental distribution images at high lateral resolution within short acquisition times. In this article, we present an overview of typical probl ems from materials science which can be preferentially solved by means of E FTEM. In the first example, we show how secondary phases in a steel specime n can be easily detected by recording jump ratio images of the matrix eleme nt under rocking beam illumination. Secondly, we describe how elemental map s can be converted into concentration maps. A Ba-Nd-titanate ceramics serve s as a typical materials science example exhibiting three different compoun ds with varying composition. In order to reduce diffraction and/or thicknes s variation effects which may be a problem for quantification of crystallin e specimens, we calculated atomic ratio maps by dividing two elemental maps and subsequent normalizing by the partial ionization cross-sections (or k- factors). Additionally, the atomic ratio maps are correlated using the scat ter diagram technique thus leading to quantitative chemical phase maps. Fin ally, we show how the near-edge structures (electron energy-loss near edge fine structures, or ELNES) can be used for mapping chemical bonding states thus differentiating between various modifications of an element. In order to distinguish between diamond and non-diamond carbon in diamond coated mat erials, we have investigated a diamond layer on a substrate with the help o f ELNES mapping utilizing the pi*-peak of the C-K ionization edge.