In the jones' paper, if we look at fig 15 (after MEK sample), we clearly see a correlation between Si and O...
Is it silica (SiO2) ???
There is correlation between the two and the most likely compound is SiO
2, but there is also a degree of Carbon associated too, not just with the Si and O, but also with Al.
I don't know what their eyes are telling them, but when they say,
Furthermore, the data indicate that wherever silicon or iron is concentrated, oxygen is also concentrated. On the other hand, there also exist regions where the aluminum is concentrated but where the oxygen may not accompany it commensurately.
yet they don't mention Carbon when there is far more correlation between Carbon and Oxygen (and Aluminium) in the sample. (In Fig 15 there are two distinct "holes" in the maps.) This is ambiguous, but instead of presenting data on that sample with higher SEM magnification, redoing an XEDS map as they did before (Fig 10) as well as showing distinctly, which elements belong to which particle, they decide to do XEDS on areas of high concentration of Si, Al and Fe of their sample at a much lower magnification. This is worthless without closer inspection, because you cannot say exactly what is present and what particle it's present in. After MEK then we would expect a much "cleaner" sample. Why are they not providing us with evidence showing that the Zinc and Chromium in the pre MEK soaked sample was a contaminant?
Note that this XEDS map is taken at 50µm and the other at 1µm. I don't understand the difference if the idea was separation of aluminium particles. They don't even show any consistency and and have different magnifications in post-MEK - Fig 15 (a) and Fig 12 (b) and pre MEK - Fig 12 (a). If they claim Al separation then they should show the particles!
Remember that we don't have any idea what particles where present before or after the soaking in MEK, because none has been presented, all we have is Fig 15 (a) and 12 (a). We don't have the before XEDS map of this sample either so we can't say if anything
has been separated.
Bearing in mind the massive discrepancy between the red layers of samples a,b,c,d and the sample that was soaked in MEK, then we can't say that the two are the same material. This is a huge inconsistency. So a test on one cannot be cross-read to the other.
I think that the MEK test is very dodgy, not only because of all of the above, but also the fact that they performed the same test on a paint sample yet didn't include details (or pictures for comparison) of that sample in the paper. I still cannot understand why anyone would want to try to separate out particles of Al when they have the very tool infront of them to see Aluminium particles. Doesn't make sense.
Why didn't they perform a proper characterisation of the samples they had? I'm of the opinion that they think that anything separated from the dust by magnet which has a "red layer" is thermite, irrespective of whether the material shows different compositions for different layers; red as well as grey.
Ooops - sorry Moorea34 - got carried away.