Towards large-area EELS mapping of precipitates in a steel matrix: Comparing low loss and high loss quantification

It is demonstrated that the low loss region of the electron energy loss spectrum can successfully map (V,Ti,Nb)C precipitates in the matrix of a steel including determining their thicknesses, volume, number density, volume fraction and size distribution. A comparison is made of the results obtained using both the low loss (mainly plasmon-type losses and low-lying semi-core-loss edges) and high loss signals (principally more classic core-loss edges) from the same dataset. The agreement between the two sets of results is excellent. While the high loss results are more element specific, the data takes much longer to acquire. When acquiring only the low loss data, the acquisition time is much shorter and would therefore allow mapping of much larger areas of a specimen, with obvious potential for making statistically significant measurements of precipitate size distributions and volume fractions. Provided that an initial study which includes high loss data is made, such large area maps can be quantified. The residual diffraction contrast in the low loss signal also provides a link connecting the precipitates and the microstructure of the matrix. While the approach here is applied to steels, it has potential for much wider applicability.