Linking TEM Analytical Spectroscopies for an Assumptionless Compositional Analysis. (6th March 2014)

• Record Type: Journal Article
• Authors: Kothleitner, Gerald; Grogger, Werner; Dienstleder, Martina; Hofer, Ferdinand
• Is Part Of: Microscopy and microanalysis. Volume 20(2014)Supplement 3
• Page Start: 678
• Page End: 686
• Publication Date: 2014-03-06
• Subjects: X-ray analysis, -- EELS analysis, -- elemental quantification, -- ζ-factors, -- absorption, -- EELS cross-sections, -- chemical shifts; Microscopy -- Periodicals; Microchemistry -- Periodicals; 502.82
• DOI: 10.1017/S1431927614000130
• Languages: English
• ISSNs: 1431-9276

Abstract

Abstract: The classical implementation for putting quantitative figures on maps to reveal elemental compositions in transmission electron microscopy is by analytical methods like X-ray and energy-loss spectroscopy. Typically, the technique in use often depends on whether lighter or heavier elements are present and—more practically—which calibrations are available or sample-related properties are known. A framework linking electron energy-loss spectroscopy (EELS) and energy-dispersive X-ray (EDX) signals such that absolute volumetric concentrations can be derived without assumptions made a priori about the unknown sample, is largely missing. In order to combine both techniques and harness their respective potentials for a light and heavy element analysis, we have set up a powerful hardware configuration and implemented an experimental approach, which reduces the need for estimates on many parameters needed for quantitative work such as densities, absolute thicknesses, theoretical ionization cross-sections, etc. Calibrations on specimens with known geometry allow the measurement of inelastic mean free paths. As a consequence, mass-thicknesses obtained from the EDX ζ -factor approach can be broken up and quantities like concentrations and partial energy-differential ionization cross-sections become accessible. ζ -factors can then be used for conversion into EELS cross-sections that are hard to determine otherwise, or conversely, connecting EDXS and EELS in a quantitative manner quite effectively.