Electron energy-loss spectrum-imaging

Electron energy-loss spectroscopy (EELS) in the scanning transmission electron microscope (STEM) is a powerful method for analyzing elemental, chemical, dielectric, and other information. The technique can be extended to produce quantitative images useful for interpreting the spatial distribution of this information. This paper describes an implementation of a new method of quantitative mapping termed “spectrum-imaging”. The resultant “spectrum-images” consist of complete spectra stored at each pixel in a scanned image. This allows not only the data in each spectrum to be analyzed a posteriori using time-consuming methods, but also permits the spatial statistics of the collective spectra to be exploited. Spectrum-image processing is fundamentally different from previous quantitative image processing that has been performed “on-the-fly”. The methods used to acquire and process EELS spectrum-images are here described, along with a range of applications and examples of the technique.