Portal de Periódicos da CAPES

Quantitative chemical phase analysis of EFTEM elemental maps using scatter diagrams

1998; Elsevier BV; Volume: 29; Issue: 1 Linguagem: Inglês

10.1016/s0968-4328(97)00061-9

1878-4291

Werner Grogger, Ferdinand Hofer, Gerald Kothleitner,

Machine Learning in Materials Science

Energy-filtered transmission electron microscope (EFTEM) images can yield elemental maps of very high lateral resolution (1–2 nm) in a short time (typically less than 3 min). Additionally, correlation techniques such as scatter diagrams yield information about the intensity relationships between the various elemental maps, thus leading to the calculation of chemical phase maps. The application of such techniques reduces the amount of data (i.e. number of images) by compressing information of different images into only a few. This makes interpretation a lot easier and clearer, since information in chemical phase maps is restricted to the chemical composition of the specimen. In this work we used a Gatan Imaging Filter (GIF) to acquire EFTEM images of a BaNd-titanate specimen. The scatter diagram technique together with quantification procedures was then applied to these images in order to show the distribution of chemical phases within the specimen. First, we quantified the elemental maps using atomic ratio images. Then we applied the scatter diagram technique on the atomic ratio images and calculated chemical phase maps.