Hi, I am trying to identify what phases are present in these pictures, can someone help where can I learn about it?
I will really appreciate if someone can give me comments about these pictures.
I suggest you to go for XRD for phase Identification.
SEM is actually not a technique for phase identification, unless you can go for EBSD. Secondary electron (SE) mode will give you just the surface image (from where you can get idea of phase distribution and can also calculate the amount of phases by linear intercept or other method), whereas back scattered electron (BSE) mode will give the image contrast on the basis of atomic no. of elements present in the phases.You can use EDS to know approximately the amount of elements present in different phases. but you can't tell with 100% surety that this phase is present.
You can’t do that using SEM in SE mode alone. Either you go for EBSD to obtain SEM images overlapped by diffraction details or XRD to get diffraction information from you sample. Check the detailed difference between both in the following thesis (Chapter 4 section 4.3 and 4.4):
Thesis Characterization and Modification of Solar Energy Water Spli...
Only short comment: phase identification by EBSD is practically impossible with the standard tools, and even in case of dynamically simulated patterns a good fit is in maximum an indication only. Current EBSD analysis is comparable to XRD 100 years ago, where from the diffraction signal mainly the peak positions were used (and we know that the Bragg angles are very bad, as well as the calibration which is incomparable with the accurate manufacturing of a diffractometer). With EBSD you only get a rough indication for an orientation description which uses the phases imcluded in your phase list, i.e., you only let select the system betwen canditates you previously selcted. However, you can index many cubic phases with Si or Fe...and because of the limited understanding and missing tools any evaluation of band withs (Bragg angle) or band intensities is highly questionable or totally impossible. Moreover, you have to ask yourself, how high-resolving (spatially) EBSD really is? How probable it will be that you will get patterns from these plates, and how you can make sure that you do not get a superposition with the matrix which possibly shows a distinct orientation relationship? For an educated guess you need noise-poor pattern and the EDX information of grains which are bigger that about 5µm so that you can exclude any interference. I believe that the estimation about the physical resolution from S. Zaefferer was quite correct. You should assume that you need at least 100x300nm (for EBSD, and a material like steel). In case of lower scattering phases the size becomes claerly bigger...and I am not yet talking about the volume represented by the EDS signal.
There is a software called EBSDL (from Li and Han) which derives at least the lattice parameters from unknown phases with an unexpectedly high precision (not comparable with XRD but much better than assumed for EBSD). This is a relieble approach but still the first step for an only geometrically approved lattice description. It does not tell you anything about ordered structures or other pseudosymmetric effects (please compare the numerous SiC polytypes where only a few can be discriminated by EBSD).
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Structural Equation Modeling