Can anyone throw some light to understand the degree of quantitative data that can be generated by employing (SEM)-EDS run, as compared to other quantitative technique? Let us assume the situation where the sample constituents are uniformly distributed.
EDS is quite accurate method provides quantitative data, but it depends on the elements that you are going to to measure (elements with Z
EDS is quite accurate method provides quantitative data, but it depends on the elements that you are going to to measure (elements with Z
As Evgeny already pointed out, there is no general question. It is in principle comparable to other 1D signals like X-ray diffraction. The more overlapping you have the less accurate/precise the measurements become. If the characteristic lines are far away from each other precision and accuracy are absolutely sufficient. If you spend some time results can be better than 1%. Problematic are small concentrations and all elements which have radiation components smaller than 2 keV. There the chance for bigger errors increases. WDS has a clearly higher sensitivity, i.e. the precision is much better, but the accuracy can be disappointing! You need to have comparable standards to describe the specific bonding which generates a chemical shift of the energy. This is not detectable with EDS since the widths are too big to see these small shifts. For WDS it is essential.
I agree with the above views that WDS will give more accurate quantitative data than the EDS which is comparatively less sensitive than WDS.
I guess most of your question has already been answered above. One point I would add is about the surface preparation and the voltage used in the SEM. To achieve quantitative results in EDS analysis in the SEM, the software needs to calculate the X-ray generation and absorption in the specimen. There are several published methods out there to do that, but they all assume a flat specimen and a homogeneous composition over the volume from which X-rays are excited. This assumption may not hold if you are looking at small particles (not flat) or acquire and X-ray spectrum over a large specimen area which includes several different phases (not compositionally homogeneous). You can reduce the interaction volume by reducing the acceleration voltage and therefore get better quant from smaller areas as long as the X-ray lines you are using for quantification are still excited.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science
- Material Characterization