• Quick note - the problem with Youtube videos not embedding on the forum appears to have been fixed, thanks to ZiprHead. If you do still see problems let me know.

Origin of the paint that was found as red-gray chips - any ideas?

Burning of epoxide under air - a macroscopic approach

Introduction:
I have visited two big hobby markets this morning. It seems that no red epoxy primers are easily available here, such specialty paints can be probably found in some big paint shops only. Anyway, I have no electrophoretic facility to apply it to the steel…

Meanwhile, I bought some epoxy adhesive/sealant (clear, i.e. without any filler), with the amine hardener (as can be judged from its fishy smell). Here is a description of my very first “experiment” on its burning. I decided to use aluminum oxide as filler here since this non-reactive stuff should not significantly change the epoxy thermal/oxidative behavior. I added 20 wt % of this powder to the epoxy.

Experimental:
Epoxy used: Pattex Repair Universal Epoxy (Henkel Co, 5 min hardening time)
Filler: Aluminum oxide, basic, very fine particles (20 wt%)

Preparation of “epoxy macrochip”: both components of epoxy were weighed and thoroughly mixed with a weighed amount of Al2O3. The resulting viscous mixture was cast on polyethylene substrate. After hardening (1h), the white layer of the filled epoxy resin was stripped out from the substrate.

Burning experiment: the macrochip (irregular shape, ca 5x7 mm, thickness about 0.2 mm) was placed on the microscopic slide and ignited with an ordinary lighter (manufacturer unknown).

Filming device: mobile phone Nokia E52. Method of picture stabilization: trying to keep my old fingers not trembling too much.


Results and discussion:
You can see the video of this burning here http://www.youtube.com/watch?v=zfCclPpYSvU. I am therefore proud that I also added one more blurred video to the countless row of them in Youtube…
As you can see, the epoxide chip was easily ignited and burned with some bright flame for some seconds. Horrible smell was evolved as well.

Conclusion:
Chip of this cured epoxide resin filled with aluminum oxide (20 wt%) easily burns under air when ignited with an open flame. Dark (charred, brittle) residue can be observed as a result of this burning.
 
Last edited:
bill smith said:
Thanks Ivan. Would you say that the amount of oxygen present in the grey layer is surprisingly high ?
Click to expand...
What do you mean by high? What level is just right or expected? What level is high? Please show your calculations. Why are you even asking this? :confused::confused::confused:
 
Sunstealer said:
What do you mean by high? What level is just right or expected? What level is high? Please show your calculations. Why are you even asking this? :confused::confused::confused:
Click to expand...


I'm just following a harmless train of thought Sunstealer. I know that the oxygen levels were very high. I want to find out if there is any reason why such high levels might be unexpected.
 
bill smith said:
I'm just following a harmless train of thought Sunstealer. I know that the oxygen levels were very high. I want to find out if there is any reason why such high levels might be unexpected.
Click to expand...
How high were they? How high are they in rust? Looks like you don't know what you are talking about? Please how high were the oxygen levels? Compare that to rust, Fe/iron rust for us. What conclusion do you make?
 
bill smith said:
I'm just following a harmless train of thought Sunstealer. I know that the oxygen levels were very high. I want to find out if there is any reason why such high levels might be unexpected.
Click to expand...
You know that the oxygen levels were very high do you? OK. How do you know this? What is your baseline? How are you working out that oxygen levels were high? Show your working. Show your evidence for high oxygen levels. I expect some work not a one word answer.
 
Sunstealer said:
You know that the oxygen levels were very high do you? OK. How do you know this? What is your baseline? How are you working out that oxygen levels were high? Show your working. Show your evidence for high oxygen levels. I expect some work not a one word answer.
Click to expand...

From the peer reviewed paper...

''..indicate that the gray layers are consistently characterized
by high iron and oxygen content including a smaller amount
of carbon.''
 
Why do you allow Bill to derail this thread so easily? It is obvious that he is trolling! There is no point to his question. Do not reply, unless he manages to reasonably link his question, or work, to the topic of this thread, which is still "can we identify the type of paint?"
 
Structural steel contains about 0.5% carbon. The gray magnetic material is iron oxide from the surface of the steel . Jones separated based on magnetic properties for some reason.
The solvent he used, MEK, is a poor solvent for cured paint and he didn't think to look at what was in paint stripper [CH2Cl2] to do a proper job. The aluminum and silicon elemental maps overlap [Fig 10 in his paper, as I remember] which says to me that there is no elemental aluminum to speak of. What he neglects to mention is that standard SEM stages are aluminum and shine through of sample is always a possibility. A carbon stage is used if one wants to see aluminum metal.
A simple XRD will show the truth but apparently that is what Jones wants to avoid.
 
Oystein said:
Why do you allow Bill to derail this thread so easily? It is obvious that he is trolling! There is no point to his question. Do not reply, unless he manages to reasonably link his question, or work, to the topic of this thread, which is still "can we identify the type of paint?"
Click to expand...
I hope you didn't direct this totally at me. I was trying to make a subtly dig.


:boxedin:
 
pteridine said:
Structural steel contains about 0.5% carbon. The gray magnetic material is iron oxide from the surface of the steel . Jones separated based on magnetic properties for some reason.
The solvent he used, MEK, is a poor solvent for cured paint and he didn't think to look at what was in paint stripper [CH2Cl2] to do a proper job. The aluminum and silicon elemental maps overlap [Fig 10 in his paper, as I remember] which says to me that there is no elemental aluminum to speak of. What he neglects to mention is that standard SEM stages are aluminum and shine through of sample is always a possibility. A carbon stage is used if one wants to see aluminum metal.
A simple XRD will show the truth but apparently that is what Jones wants to avoid.
Click to expand...

Hi pteridine,
I don't think I have seen you around here, so happy to see you in my thread, especially since you seem to have a good grasp on the topic!
The information about Al stages is news to me. Not sure if anybody has thought of this before, and if and how we can eliminate that as a problem. I think I remember that C coating of ...whatever... poses a problem for determining C content. Does that ring a bell?

Anyway, the goal here is not to identify all the flaws that Jones, Harrit and co committed on their way to a moronic conclusions, but to make the best of their work and identify the paint. I think we are already quite certain that the gray layer is oxidized, flaked-off steel. If we could identify the type of steel ... but I think data resolution is too bad for that. As a second best option, we should formulate a theory about the gray layer ("it's Axx steel that oxidized when... and spalled off such and such..."), then, using literature, make predictions about how that would look like in the experiments that were in fact done: EDS spectra, microscopic appearance, electrical resistance, magnetism, ... If we find our predictions are a good fit with experiment, our theory is viable - and the only theory out there, afaik.
 
Burning of Laclede primer paint imitation and other epoxide samples under air – still macroscopic approach

Introduction:
I found yesterday (see post #256) that chip of the cured epoxy resin “Pattex Repair Universal Epoxy” filled with 20 wt% of aluminum oxide can be ignited very easily and burns with a bright flame for some seconds.

Then, I decided to prepare some imitation of Laclede primer paint used for the corrosion protection of WTC1 and WTC2 floor joists. Such imitation can serve not only for simple macroscopic burning tests, but also for: a) measurements of its TGA, DSC properties; b) microscopic study of its appearance during/after heating/burning; c) as a sample of typical epoxy resin suitable for the determination of a typical carbon, nitrogen and hydrogen content in epoxides by elemental analysis. Other uses of this paint imitation might be found later.
All these experiments could serve as a supporting info for the hypothesis that chips (a) to (d) in the paper of Harrit et al. were particles of Laclede primer.

In this contribution, my intention was not only to prepare such “Laclede imitation”, but also to compare its (macroscopically observed) burning with burning of previously prepared epoxy resins filled with aluminum oxide (“Al2O3 resin”) and with the pristine (non-filled, “pristine resin”) epoxy as well.

Declared composition of Laclede primer paint was (NCSTAR1-6b report of NIST, Appendix B):

Pigment: (21 % of total weight of dry/cured paint)
Iron oxide 55 % (wt%)
Aluminum Silicate 41 %
Strontium Chromate 4 %
Total Pigment 100 %

Vehicle:
Unmodified Epoxy Amine 45 %
Deionized Water and Amine 55 %
Total Vehicle 100 %

I gathered powdered iron oxide and a form of aluminosilicate called Nanoclay. Since I was not able to get strontium chromate, I decided to replace it with potassium chromate. Although there was not very much of chromate in the Laclede paint, the paint imitation should also contain some chromate, since chromates are generally strong oxidizing agents and can influence thermal/oxidative/burning behavior of the original paint as well as of its imitation.

Experimental:
Preparation of “Laclede imitation” layer:
Used chemicals
- Epoxy: Pattex Repair Universal Epoxy (Henkel, 5 min hardening time), 2.45 g
Fillers:
- Iron oxide, particle size between 1 and 3 μm; dark red-brownish powder (Lachema), 0.55 g
- Nanoclay, an aluminosilicate (hydrophilic bentonite) in the very fine platelet form (Sigma Aldrich), whitish powder, 0.41 g.
- Potassium chromate, yellow powder (Lachema), 0.04 g (molar masses of strontium chromate and potassium chromate are similar – 203.6 vs 194.2, so this amount of potassium chromate is adequate).

The sum of component weights in this epoxy composite is 2.45 + 0.55 + 0.41 + 0.04 g = 3.45 g and its composition should be in a good accordance with the composition of the real Laclede paint. The wt. ratio between epoxy resin and inorganic fillers is 2.45 : 1.

You can see the collection of starting chemicals on this Fig. 1: http://bobule100.rajce.idnes.cz/epoxides#chemicals_used_for_Laclede_paint_imitation.jpg .

Potassium chromate was first thoroughly grinded in a grinding mortar in order to achieve fine particles of this chemical.

Then, all chemicals were thoroughly mixed with a spatula (as usual in the preparation of epoxy adhesives/sealants) and the resulting viscous dark red matter was casted by spatula on polyethylene substrate. After curing (1 h) the tough (but flexible) dark red layer of the “Laclede imitation” was stripped from the substrate.


Preparation of “Al203 resin” layer
It was described in the post #256, but new layer (thicker) was prepared for better comparison here.

Preparation of “pristine resin” layer
It was prepared like in other cases but no filler was added.

All layers were about 0.5-0.8 mm thick.


Macroscopic burning experiments:
The epoxy “macrochips” (irregular shapes, ca 6-8x10-12 mm), were placed on the microscopic glass slides and ignited with an ordinary lighter. Some typical burning experiments were filmed and links to videos are given in the next section.
After burning, chips were again filmed and video is available (see later).


Results and discussion:
You can see typical burning of epoxy chips under study here:
http://www.youtube.com/watch?v=9SGc8HlEP9Q&feature=mfu_in_order&list=UL (pristine resin);
http://www.youtube.com/watch?v=Uhh2wlqinAs&feature=mfu_in_order&list=UL (Al2O3 resin);
http://www.youtube.com/watch?v=Pdzr2CJuxHE&feature=mfu_in_order&list=UL (Laclede imitation).

I can judge from this:
All epoxy samples are easily ignited with an ordinary lighter and burn with the bright flame for ca 30 s to 90 s.
The most flammable is “pristine resin” – the flame was the brightest and lasted for the longest time.
“Laclede imitation” burns better, longer and with the brighter flame than “Al203 resin”. This could be caused (at least partially) by the presence of chromate as oxidizing agent. Also iron oxide might play some role (?).

Chips after burning can be seen here:
http://www.youtube.com/watch?v=ET4gs0J0Bso&feature=mfu_in_order&list=UL
From the left to right:
“Pristine resin”: molten dark matter as a result, almost whole chip was burned.
“Al2O3 resin” some minor part was burned, this part was dark and brittle.
“Laclede imitation”: about half of the chip was burned, dark brittle matter resulted.

Conclusion:
Generally, fillers in epoxy resins seem to “quench” the burning to some extent, but even both filled resins ignite easily and burn with the bright flame.
In the case of Laclede imitation, it can serve as some hint that the real Laclede primer paint can be easily oxidized, e.g., during DSC measurements under air (as performed by Harrit et al.). This oxidation can be a source of exotherms observed by Harrit et al. for chips (a) to (d).
Further measurements on Laclede imitation (TGA, DSC, some microscopy on burned/heated resin, elemental analysis) will follow soon (I hope).

Thank you for your kind attention:o)
 
Last edited:
(Remark: sorry, I accidentally removed the first video of epoxy burning from yesterday in Youtube. Although I have added this video again to Youtube, the original link in the post #256 probably will not work. But does not matter so much, since the burning of this kind of chip is presented again in the post #277 just above)
Anyway, a question for administrator: can I edit again the post #256 somehow?
 
Last edited:
This thread is a good example of the obvious reasons why Jones doesn't want independent verification of his experiments...
 
Ivan Kminek said:
Some (much) more thorough calculations are welcome:rolleyes: Btw, we expect so far that the Laclede primer was stripped out from the steel joists during collapses to the quite high extent, since it can explain the abundance of paint particles in the dust.
Click to expand...

Given that the floor elements all took the express route down the inside of the towers, you shouild expect nearly all of it to have been ground away. This was the devil's own rock tumbler, with multi-ton pieces of grit slamming into the metal elements. I doubt that much paint was left on any of them by the time they reached the end of that maelstrom.

Looking again to some photos of the floor joist steel (post #206, link http://www.drjudywood.com/articles/JJ/JJ5.html Fig. 14) this steel looks more rusty than painted.
Click to expand...

I am certain that it is mostly rust. If you look at the end of that top rail next to the bunched-up laces on the left edge, you will see that there is a dark area with slightly rusty margins. The paint has been stripped completely away, exposing the black oxide that coated the steel before it was painted. Heat and steam converted most of this black oxide to the red form (will some chemist help me out here? I am looking at it from the background of a casual construction and foundary worker.)

But, this rusty color can be to some extent also a result of mixing of colours of iron oxide (red) and strontium chromate (yellow) in the paint. I would like to possess a piece of this widely twisted steel...
Click to expand...

On the other hand, it is also a good match with the sort of colors you will see in steel recovered from a really hot fire scene days afterward.

I really doubt that any great amount of the pigment remains on the steel.
 
You must log in or register to reply here.

Back
Top Bottom