Thanks for the clarity Oystein. I hate myself for saying this, but getting this explained in 'lay mans' terms is really the easiest way I can understand this. When it comes to physics, mechanics etc I am more than competent at understanding in depth, however chemistry and in depth experiments like this one is a nightmare for me to understand easily, so any interpretation is much appreciated
yw
On another note, one of the main arguments for thermitic material was that these red/gray chips were highly reactive in heat. IIRC, Harret et al heated the chips to 400C and recorded a violent reaction.
According to Chris's post;
Chips of interest were ashed in a muffle furnace using a NEY Temperature Programmable furnace operated at 400oC for 1 hour. The gray layer remained intact and the red layer residue was prepared as described above and analyzed using a Philips CM120 TEM-SAED-EDS.
Was this a replication of the same experiment? Could someone explain this section please for people like myself? Thanks.
No, this wasn't a replication of the tests Farrer did for the Harrit team.
What Millette did is quite simply this: he carefully burned the organic matrix (epoxy), which largely turns into gas (CO2, H2O, ...) and disappears, to free the pigment from the matrix, so he can take a closer look at the pigments. Other methods to achieve the same result would have been: Dissolve paint in paint solvent (he tried that, but epoxy is notriously difficult to dissolve, so that didn't work), cut chips with a skalpell (that exposes only those pigments very close to the cut surface).
His furnace controls temperature carefully, but doesn't measure anything beyond that, as far as I understand.
The DSC (Differantiated Scanning Calorimeter) does something quite more complicated: You have to heating plates - one with your paint sample, the other without any sample as a control. You slowly feed energy (electric current I suppoose) to them, at the same rate, and you measure how the temperature of the device develops. You adjust energy input to both such that temperature rises constantly, for example at a rate of 10°C per minute. You really measure and plot how much energy you need to feed to your plates to achieve that heating rate.
No, if your sample contains some water, for example, that water will boil off around 100°C. This cools the plate, and you need to feed more energy to the plate with that sample than to the control plate, to still achieve the same heating rate. The same happens if an endotherm reaction occurs in your sample - that is a reaction that needs more energy input than comes out of it. If, on the other hand, your sample undergoes an exotherm chemical reaction (for example: it burns), then the sample heats up the plate, and you must reduce the heat you feed to your sample plate, or else your heating rate gets to high.
You do that process, starting perhaps at room temperature (20°C), and continue till you reach a desired maximum temperature (typically, DSCs will go to 600 or 700°C).
So what you basically measure at each temperature along the scale is the energy flow of the chemical and physical reactions that take place at that temperature. If you know the initial mass of your sample, you can compute from the curve a property called "energy density". But the shape of the curve can also be of interest, and of course the temperatures at which reactions peak.
What Farrer, Harrit and the others found was that thei chips react exothermally mostly around a peak at 430°, plus minus 50°. This means their reaction already started under 400°C. This is consistent with Millette's ashing of the sample at 400°C - a polymer like epoxy will degrade thermally at that temperature, no surprise. Also, it is no surprise that this reaction is exotherm. The energy densities that Farrer measured were unremarkable for organic substances: between 1.5 and 7.5 kJ/g. Paper and wood would have more than 15kJ/g, many plastics between 20 and 40 kJ/g, fossil fuels more than that. I don't have values for epoxy handy, but wouldn't be surprised at all if epoxy, too, had like 20kJ/g. Farrer's value were much lower than that, because (as Harrit e.al. concede themselves) their chips had this gray layer, which does not react, so it adds mass, but not energy, lowering the energy per mass. However, their values would be highly remarkable, if not downright impossible, for thermite: Ideal thermite has only 3.9kJ/g - two of Farrer's samples were above that value, so something that was not thermite must have burned, there is no way around that conclusion in this universe, and they even admit it.
Now, something that I am missing from Millettes ashed-in-a-furnace samples is - microspheres
And many, many thanks to Jim Millette as well!
