hi everyone, im new here. this thread seems to have been idle for quite some time. but it would be great if you guys could help me out somehow. I'm a summer student at McGill University in Canada and my summer research project primarily involves thermites. I place stoichiometric thermite powder mixtures in a Setaram DTA/TGA machine to plot a thermogram of heat flow of the mixture during heating and cooling. we are interested in the temperature where it starts reacting and if the reactions occur before or after the melting point of the metal. the heating is held in an inert argon atmosphere. i was told that most people beleive that the melting of aluminum coincides with the onset of a reaction (with Fe2O3 for example..). but our thermograms show that the bulk of the reactions (characterized by large exotherms on the thermogram) occur much later than 660C (m.p. Al). does this make sense? someone in our group suggested that this happens because the aluminum powder melts inside its own aluminum oxide shell (All aluminum quickly develops an oxide shell when in contact with atmosphere). so even though it is melted, it does not come in contact with the metal oxide powder because the shell (very high m.p.) prevents it.. maybe eventually, something happens and the aluminum leaks out and starts reacting at around 1200C (according to our graphs). which of these theories do you think is more valid? does the reaction genuinely occur much later than the melting point, or is it just because of the oxide shell which delays the reaction. also, i tried melting pure aluminum powder 99.9% in the DTA machine. it clearly melted (based on the thermogram) and then resolidified upon cooling. but it came back out in powder form, as if it never even melted in bulk form... very strange. i tried the same with pure aluminum wire.. and it came back out in wire form.. no bulk melting.. this is why the oxide shell theory was raised by one of my friends.. please give me some feedback into this.. thanks a lot.. I trust everyone's experience here...
