Benjamin,
I will address your points in reverse order:
Finally, does anything significant hang on this issue?
Harrit's conclusion is that the red layer of certain red-gray chips contains nano-thermite of the Al+Fe2O3 persuasion that's still active and reacts to form tiny iron balls.
This claim has several distinct elements:
#1 There is iron oxide
#2 There is elemental Al
#3 These come in nano-sized bits
#4 They react, resulting in elemental (or at least reduced) Fe
Now, #1 is uncontroversial - there is good evidence for iron oxide whatever way you look at the red stuffs.
But:
#2 has only been shown (and dubiously so) for one chips, the MEK-soaked one, by virtue of Fig. 17
#3 has only been shown for chips a-d, by virtue of SEM-images showing Al-rich platelets and Fe-rich faceted grains
#4 has been shown for chips that are neither the MEK-soaked one nor chips a-d
To form the conclusion, one can only conflate #2 to #4 if one assumes, or proves, that all these chips are essentially the same material.
Once you allow that some may really be just paint and contain no thermite, then the conclusion becomes invalid.
What is the basis for claiming that (a) - (d) is different from the MEK chip?
Here are the EDS-spectra for the red and gray layers of chips a-d:
Red layer a-d
Gray layer a+b,
Gray layer c+d
These spectra, plus the identical visual appearance of the microstructure under SEM, are good evidence that these four chips are indeed the same material: They show basically only five elements: C, O, Al, Si and Fe; and only tiny traces of other stuff. An important hint is that in all four red layers, the peaks for Al and Si are almost identical in height, which does in fact mean nearly identical mass- and atomic proportions. This is particularly important when one looks at
Fig. 10, in which we can see that the spacial distributions of Al and Si are practically identical, and they coincide with the platelets in 10a. Which means these platelets are made of something that has equal Al and Si, and there is no significant Al or Si elsewhere in these red layers.
Contrast this with
Figure 14:
Here you have very significant Ca, very significant S - both are absent from Chips a-d. The peaks for Zn and Cr are pretty significant, too, and, perhaps most importantly, there is very little Al compared to Si.
Harrit et al fail to show the microstructure of this red layer.
So there is zero evidence that this MEK-soaked chip is the same material as chips a-d, but there are several indications that it probably is different.
The burden of evidence is on Harrit et al: Remember, the whole point of showing this MEK-soak is to "prove" that there is Al separate from Si, and in fact elemental. This finding of elemental Al can only be applied to the other chips - those that burn vigorously; those that have nano-sized pigments - if you show that it's the same material. But they haven't shown that. We don't have to prove it's a different material (although we are very certain).
Also, this [XEDS] is apparently not a good method of identifying materials, since Millette went on to do a great deal more.
Half true.
In isolation, it wouldn't be sufficient to identify.
But it does help a lot positively - much better than DSC. A large peak around 3.69 keV in an XEDS spectrum when there is none at 3.44 keV is positive proof for significant Ca (the 3.44 controls for Sn, which has a signal very near the 3.69 Ca-peak, but would also have a larger signal at 3.44).
This in contrast to DSC, where an exotherm at 430 °C could mean a thousand things.
However, looking at the various spectra across the two papers, it seems that there is a great degree of variability in the relative heights of peaks between the samples, with no clear tendency to group into two readily identifiable sub-groups (as far as these lay eyes can tell).
Looking at Millette's data, you are absolutely right - but why limit this to just
two readily identifiable sub-groups? We already know that the twin towers had their steel primed with
at least two different red primers, but there may have been more. And we can be pretty sure that Building 7, having been built 15 years later or so, had a another primer still. There are thousands of other things painted red in those collapsed buildings.
I recently looked over the Bentham paper again (and the Millette paper), and I have a question about interpreting the XEDS results. I seem to recall members here claiming that the chip soaked in MEK is a different material from chips (a) - (d), and that this could be determined from XEDS.
Do you see now how this can indeed be done, but that the reverse question - "
are they the same?" - is implicitly claimed, but unproven, by Harrit et al?