The 9/11 Forum

Intelligent and evidence-based discussion of 9/11 issues

Skip to content



Our vision is to provide a home to sincere 9/11 researchers free from biased moderation and abusive tirades from other members.

You are currently viewing our boards as a guest, which only gives you access to view the discussions. New registration has been temporarily enabled; take advantage of it!

Some simulations of XEDS spectra

Discussions and analysis of thermite theories

Some simulations of XEDS spectra

Postby Oystein » Sun Oct 30, 2016 8:58 pm

I create this thread as a repository for myself, to comment on XEDS spectra I created using simualtion software.

For starters: Here is the simulation software: ... index.html
It's free and public domain.

All spectra in this thread are run at 20 keV, unless otherwise noted.
Posts: 531
Joined: Fri Sep 30, 2011 2:00 pm


Re: Some simulations of XEDS spectra

Postby Oystein » Sun Oct 30, 2016 9:40 pm

LaClede Primer, and variations

My blog has the LaClede primer recipe, and a break-down by mass of the chemical elements in it: ... ndard.html

The elemental composition is, as best as I can determine:
  • C: 48% by weight
  • O: 21%
  • Fe: 11%
  • H, N: 7% each
  • Si: 2.5%
  • Al: 2.4%
  • Sr: 0.5%
  • Cr: 0.3%

I did new simulations today:
  • LaClede as is
  • LaClede with 0.3 wt-% Pb (lead)
  • LaClede with 0.3 wt-% Ti (titanium)
  • LaClede with 2.5% Ge (germanium) instead of 2.5% Si
The purpose of the latter is to show the Sr (strontium) peak that is otherwise hidden underneath the Si-peak. Picking Ge was arbitrary - it is chemically similar to Si, but that has no effect on the XEDS readings.

Here goes
LaClede as is:

LaClede + 0.3% Pb:

LaClede + 0.3% Ti:

The three specra above overlayed (scaled such that the area under the graph, i.e. the total X-ray count, is equal):

The comparison shows that as little as 0.3% of either Ti or Pb would very clearly show in the graph - Ti at 4.5 keV (the "K-alpha" line of Ti), Pb at 2.35 keV (M-alpha1).

The reason I show the Ti- and Pb peaks is a note by Steven Jones on 911Blogger: ... p-analyses
"Jeff notes that in his TEM analyses he observed “very small (nanometer-scale) Pb particles in the TEM samples” as well as strontium and chromium in small amounts. (Much of the TEM analysis was performed at higher magnification than used in the SEM analysis done in the paper.) Thus, red/gray chips which match ours will show these same elements under TEM analysis.
I (Dr. Jones) have searched Millette's plots and see no indication of strontium (Sr) or lead (Pb) in his samples, but he does report titanium (Ti) which we do not see. Thus, his samples do not appear to be the same material as what we reported on."

I note that Harrit et al themselves, when preparing their 2009 paper, did NOT show that any of the chips used to form their conclusions contained Pb. However, they do have some Ti in Figures 7d (uncertain), 25, 28.

LaClede plus Germanium minus of Silicon:

Comparison of LaClede with LaClede plus Germanium minus of Silicon:

The thing to note here is that there is a clear peak for Sr (strontium) when the Si is removed. This is the reason why Harrit et al Fig 7 doesn't show Sr, even if it were present.
Posts: 531
Joined: Fri Sep 30, 2011 2:00 pm

Re: Some simulations of XEDS spectra

Postby DGM » Sun Oct 30, 2016 9:57 pm

Great thread.

It will be nice to put to bed the claims made in the Harrit paper on this evidence based forum.
Posts: 374
Joined: Thu Jun 09, 2016 9:33 pm

Re: Some simulations of XEDS spectra

Postby Oystein » Wed Nov 09, 2016 3:56 pm

According to Harrit's white paper "Why the Red/Gray chips aren't paint" ( ... Harrit.pdf), the composition of Tnemec Red as per 1967 was as follows:

- 35.9% Iron Oxide (Fe2O3)
- 20.3% Zinc Yellow (ZnCrO4)
- 33.7% Proprietary Tnemec pigment (???)
- 10.1% Diatomaceous silica (SiO2)
These add up to 100%
However, Pigments are only a smaller fraction of the wet and the dry paint, under 50%.

The VEHICLE (organic binder) consists of several organic materials:
- 19.3% Soya alkyd resin solids, Hard resin (I am guessing: C7HxO1)
- 41.5% Raw and bodied linseed oil (roughly C18H32O2, +4 O when dried)
- 6.9% Others (I am guessing: C6HxO)
These add up to 67.7%. The remaining 32.3% are thinners, which I assume evaporate upon drying.

Harrit found a 2003 Material Data Sheet, which he erroneously believed to be describing the "Proprietary Tnemec pigment", when in fact it describes the wet paint.
This data sheet has 22.71% mineral spirits - if these are equal to the 32.3% thinners in the veicle, it follows that the vehicle is 22.71/32.3 = 70.3% of the wet paint, with the Pigments being the remaining 29.7%.
The MDS most certainly does not reflect the 1967 paint composition; it tells us there is 21-30% Talc (Mg3Si4O12H2) in the paint, for which there is no room in the 1967 recipe. Also, the MSD has some "Calcium silicates and aluminates".

Now, I played around with these ingredients and numbers, made up some "Proprietary Tnemec pigment" as a mix of Ca-aluminates, talc and iron oxide, and also threw in a dash of gypsum as surface contamination, in an effort to mimic Fig 14 of Harrit et al (the chip later soaked in MEK). Here is my own private Tnemec + Gypsum mix:

92% dried Tnemec, consisting of
... 22% Pigment, consisting of
... ... 35.9% Fe2O3
... ... 20.3% ZnCrO4
... ... 33.7% Tnemec pigment, mixed in molar proportions as
... ... ... 1 mol Mg3Si4O12H2 (Talc)
... ... ... 3 mol Ca3Al2O3 (Tricalcium aluminate)
... ... ... 7 m
ol Fe2O3
... ... 10.1% SiO2
... 78% organic binder, having C and O in the following mass proportion:
... ... 39.5 parts C
... ... 13.3 parts O

8% CaSO4-2H2O (Gypsum)

(The blue part is my best guessing: In the case of Tnemec pigment, this resulted from iterative trying of XEDS sims, in the case of the binder, this is mostly the result of studying all sorts of paint binder substances, including linseed oil and alkyd resin; exact proportions of C:O are impossible to come by).

By multiplying these percentages with atomic weights, this results in the following elemental mass proportions of my simulated gypsum-contaminated MEK-chip:
C: 48.5%
O: 30.0%
Mg: 0.2%
Al: 0.7%
Si: 1.5%
S: 1.5%
Ca: 3.6%
Cr: 1.6%
Fe: 10.4%
Zn: 2.0%

(Note at this point that Harrit, after several steps of progressive stupidity, determined the Zn-content in the dry paint to be 12.3% instead of 2.2%, which further get diluted by adding some gypsum).

At long last, here is my simulated XEDS for this recipe:


Compare this to Harrit et al's Figure 14 (without scales and labels; I normalized the above graph such that both images have the same scale on the x-axis, and the Si-peaks are of equal height)


My C and O are lower, but these two elements are pretty sensitive to a lot of things, including parameters of the XEDS equipment, sample preparation etc, so that difference means little.
My Ca peak is also lower - and I have no good idea how to fix that: If I increase the gypsum proportion, my sulfur gets too large, if I increase Ca-aluminates, Al gets too large, if I add Ca-Silicates, Si gets too large. Conceivably, Harrit et al's beam energy was somewhat larger than the nominal 20 keV.

Anyway, this exercise merely serves to demonstrate that this allegedly "thermitic" chip can very well be seen as paint with close to 10% gypsum. This paint contains less than 1% by weight Al and thus cannot be thermitic, as purethermite is 25% by weight Al. The Fe-content, at 11.3%, is abozt 14 times the the Al-content, where the factor ought to be just 2. Even IF the Al were elemental and did react with iron oxide, it would reduce only 1/7th of that iron oxide, and release no more than 0.125 kJ/g of energy, based on the mass of the red material.
Posts: 531
Joined: Fri Sep 30, 2011 2:00 pm

Return to Thermite and Nanothermite

Who is online

Users browsing this forum: No registered users and 1 guest