Local phase identification in scanning electron microscopy would be a wonderful tool if electron backscatter diffraction (EBSD) would be more sensitive regarding lattice parameters. In uncertain cases the EDX (energy dispersive X-ray spectroscopy) signal is used as a discriminator that works nicely for simple phases but has limitations regarding the comparability of the information volumes. BSE has a better lateral resolution than EDX but reflects only an average value of the atomic number. However, especially with respect to such light elements as H, Li, or OH-groups or crystal water BSE should be quite beneficial since the related signal cannot (or only with a limited reliability) be collected with present EDX detectors. Couldn't we use BSE as phase discriminator, or proof for the correctness of a locally assigned phase? Where do you see the limitations?
The limitations of BSE is to be able to differentiate between polymorphic phases of the same composition, or phases with similar average Z.
As far as I know, there is little you can do in electron microscopy about detecting light elements. BSE will simply show you an "absence" of signal, which might be good enough, depending on what you're after.
If you are more serious about the light elements, then maybe wavelength-dispersive x-ray spectroscopy (WDS) with proper reference standards can be used to get a quantitative analysis of "visible" elements, and then the light elements could be deduced from stoichiometry.
In principle you are right, but how big is the chance that two phases are identical in their chemical composition as well as in their diffraction signal? Does this not mean that these phases are then de-facto the same phase :-)?
I mean if you have two polymorphs with the same chemical composition, e.g. quartz, tridymite, and cristoballite, their diffraction signal should be different enough to seperate them by crystallographic "features" only. In these cases you don't need BSE at all (although it might be a nice opportunity to have the signal, only in case of...).
I guess that practically only a few phases exist which perhaps make problems. The only examples I have in mind are magnetite and wuestite - but even these phases generate a quite different BSE signal - and gamma and gamma' phase for Ni-based superalloys. Also the segregation as reason for the existence of gamma' causes a sufficient BSE difference.
Yes, you are absolutely right. If your composition (EDX/WDS) and your diffraction patterns match, then you pretty much have the same phase. (Not sure, can one get chirality information from EBSD? Because that's the one case where I can imagine similar composition and crystallography AND possible multiple phases.)
If you use diffraction and BSE imaging alone, then it's conceivable to have different compositions with similar average Z and similar diffraction properties, although, as you had probably pointed out, it is very unlikely. (The latter situation could be a special case for insoluble glassy phases, where your diffraction is of limited use, and you have to rely more heavily on elemental analysis.)
You can see chirality in EBSD patterns. Hard to find if you don't know where you have to search for since only a few zone axis are affected in a way that you can see it. Similar to polarity which is also not always detectable.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science