I am still not clear on two points.
(1)At what temperatures could we expect a layer of epoxy coating to react with an underlying layer of iron oxide to begin reducing the iron oxide?
(2) Would epoxy mixed with iron oxide reduce the iron oxide at that same temperature?
I know that smelting iron by the bloomery process takes place at temperatures far lower than those required for the production of steel or cast iron. It seems to me that a large office fire with a good draught should produce nearly those temperatures.
I'm trying to find some info again,
Lefty, with no success so far.
If you allow me again one "self-promoting post" (a post reffering to my own simple experiments): my Laclede epoxy paint imitation has a composition close to the real WTC paint. Micrographs of the pristine imitation and imitation after heating up to 700 degrees C under air can be seen
here and
here.
In principle, XEDS analysis of the burned Laclede paint imitation could show us
experimentally, if some iron oxide used in this formulation was reduced to iron.
Consider this:
- Oxygen in this paint imitation is present in a) epoxy binder (70 wt %); b) inorganics: iron oxide (12 %), aluminosilicate (16.5 %) and potassium chromate (1.2 %).
- Oxygen content in all inorganics is given by their formulae and oxygen content in epoxy used (Henkel brand) can be determined using e.g. elemental analysis (this method precisely determines carbon, hydrogen and nitrogen content in any organic material, "the rest" to 100 % must be oxygen for such an epoxy resin).
- Paint imitation after burning is still basically red, which means that (almost) all epoxy binder was "burned out" (no organics left). This is clear also from thermogravimetric (TGA) curve measured under air, see
here.
- Aluminosilicate used in this imitation should not be transformed to anything else during heating to 700 degrees C
(but I can't say this for sure). As regards chromate, we can neglect it for any calculations (with its 1.2 % content).
- Let me say that I will measure oxygen to iron ratio in pristine and burned paint imitation by XEDS
(and I can do it in principle, since I have SEM microscopes with XEDS probes "in house" (in my institute). By subtracting the oxygen from epoxy binder (which "disappeared" during burning), I can get some idea if oxygen content in inorganics is lower than in unburned sample (i.e. some iron oxide was reduced to iron or some iron compounds with lower oxidation number).
But, there are some factors which make precise comparison of the oxygen/iron ratio in unburned and burned samples difficult or even impossible (?). Namely, XEDS can't give reliable quantitative results, as concerns determination of oxygen (and carbon), as explained by Almond. Moreover, aluminosilicate can be dehydrated at high temperatures, which can lower the overall oxygen content (and overall mass) of inorganics.
In summary: such experiments would be easily just a waste time and all this my elaborate contribution is a waste of time as well - but still it is a kind of fun for me
