Tnemec Red 99 primer paint - composition and properties

This thread is a call for research and input of solid facts.

I do NOT want to debate anything with debunkers and would very much appreciate if you could resist the urge to feed trolls. Thank you.


Initial sources for this post are:

NIST NCSTAR 1-3C (Damage and Failure Modes of Structural Steel Components. Federal Building and Fire Safety Investigation of the World Trade Center Disaster)
NIST NCSTAR 1-3C Appendices A-G
Steven Jones presentation Science and Society (part 6/8) (December 2009 in Sydney)
- Some Screenshots of that presentation
supplementary perhaps: Harrit N. H.: Why The Red/Gray Chips Are Not Primer Paint. Open Letter, May 2009
____________

Here's why I am starting this thread:
NIST claims in Chapter 6 of NCSTAR 1-3C, page 219:

"The paint coating was not a paint in the traditional sense, but was actually a ceramic coating (Tnemic) containing no organic binders. Thus, when the coating was exposed to high temperatures, the coating did not burn as no organic binder exists to combust."

This claim is reinforced in the Appendices to NCSTAR 1-3C, namely Appendix D, page 433:

"Most paints contain organic materials in the form of binders or agents that allow them to be spread or sprayed. Upon drying, depending on the particular paint formulation, some of these organic compounds are left in the coating and burn upon exposure to high temperatures.

The structural steel components for the WTC were painted with a coating consisting of several oxide pigments suspended in a volatile liquid (Sramek 1967). The constituents of the paint are found in Table D–1. The paint was applied by the fabricators at the factory and given a low temperature (approx. 120 °C) bake to cure the paint and evaporate the suspending liquid. The paint was essentially a ceramic coating, consisting of predominantly of iron oxides with small additions of other oxides that provide color, and silica sand.

Under the assumption that this was a typical paint with an organic binder, initial burn tests were conducted at relatively low temperatures, looking for the temperature at which the paint will combust."

They then describe how the paint behaves, mud cracking and all, but apparently no obvious decomposition or ignition of the binder.

I must admit I never read Chapter 6, and never was cleat that NIST really claims the Tnemec Red 99 coating of the perimeter columns is free of organic binder!

Can this be true?

It seems to be contradicted by both the Tnemec 99 recipe and by Steven Jones' analysis of WTC paint.

The paint recipe is listed in Appendix D on page 438:
Pigments:

Iron oxide	35.9 %
Zinc yellow	20.3 %
Tnemec pigment (proprietary
composition)	33.7 %
Diatomaceous silica	10.1 %

Vehicle:

Soya alkyd resin solids	16.5 %
Hard resin	2.8 %
Raw linseed oil	35.1 %
Bodied linseed oil	6.4 %
Suspension agents	2.2 %
Driers and anti-skin	4.8 %
Thinners	32.3 %

The reason I say this contradicts the "no organic binder" claim is that neither linseed oil not alkyd resin are, to the best of my knowledge, volatile. They crosslink and harden and form a matrix that embeds the pigments; they don't disappear. Only the thinners and perhaps the auxiliary contents go away when drying at 120 °C.

And Steven Jones famously showed the elemental (XEDS) spectrum of the perimeter column paint, at 2:43 minutes in the presentation linked above. Here's a screenshot:

This has a very prominent peak for carbon - this has to be a binder.


So explain to me: Why did the primer coating behave in the NIST test the way it did?
Did they not observe carefully?
Is the recipe wrong?
?


For extra credit: On page 435 of the Appendices, NIST describes a DTA experiment on weld material. They dicuss the the result thus:

"Figure D–5 [D–6 actually; Oy] shows two superimposed DTA scans. The initial heating scan at 20 K/min exhibits a broad exothermic peak centered at 400 °C. The cooling scan at 2.5 K/min shows no significant transformations. The second heating scan at 20 K/min shows that the broad exothermic peak is absent. In general, DTA scans exhibit endothermic peaks on heating for normal 'equilibrium' transformations. On the other hand, exothermic peaks on heating usually indicate the return of a non-equilibrium state formed during prior processing to an equilibrium state. The interpretation of the present experiment was that the broad exothermic peak during first heating resulted from the annealing (exothermic heat release) of the weld microstructure."

It is, of course, tempting to see the exotherm peak around 400 °C and think "hey, Harrit et al had an exotherm peak there, too!"
Could they have been so stupid as to leave paint on the tested weld??
(Be very careful though: You can't put DTA and DSC plots next to each other and compare them 1:1. Also, I am confused about the y-axis label "DSC /uV/mg" - should that be µW/mg, or W/mg, and why DSC?? The exotherm hump exceeds its baseline by close to 0.03 "uV/mg" - or 30 W/g? Or 0.00003 W/g??)