Finely powdered aluminium can make almost anything more pyrotechnically interesting, from fireworks to machine shop cleanups – even ceramics, as [Degree of Freedom] discovered. He was experimenting with mixing aluminium powder with various other substances to see whether they could make a thermite-like combination, and found that he could shape a paste of aluminium powder and clay into a form, dry it, and ignite it. After burning, it left behind a hard ceramic material.
[Degree of Freedom] was naturally interested in the possibilities of self-firing clay, so he ran a series of experiments to optimize the composition, and found that a mixture of three parts of aluminium to five parts clay by volume worked best. However, he noticed that bubbles of hydrogen were forming under the surface of the clay, which could cause cracks during the firing. The aluminium was reacting with water to form the bubbles, somewhat like a unwanted form of aerated concrete, and for some reason the kaolinite in clay seemed to accelerate the reaction. Trying to passivate the aluminium by heating it in air or water didn’t prevent the reaction, but [Degree of Freedom] did find that clay extracted from the dirt in his back yard didn’t accelerate it as kaolinite did, and the mixture could dry out without forming bubbles.
This mixture wasn’t totally reliable, so to make it a bit more consistent [Degree of Freedom] added some iron oxide to accelerate the burn through an actual thermite reaction – some mixtures burned hot enough to start to melt the clay. After many tests, he found that sixteen parts clay, seven parts aluminium, and five parts iron oxide gave the best results. He fired two cups made of the mixture, a thin rod, and a cube, with mixed results. The clay expanded a bit during firing, which sometimes produced a rough finish, cracking, and fragility, but in some cases it was surprisingly strong.
The actual chemistry at work in the clay-aluminium mixtures is a bit obscure, but not all thermite reactions need to involve iron oxide, so there might have been some thermite component even in the earlier mixtures. If you need heat rather than ceramic, we’ve also seen a moldable thermite paste extruded from a 3D printer.
Thanks to [kooshi] for the tip!
He should try dusting the outside with copper or silver powder before igniting. See what happens.
Put a tray of water next to the plastic bag, and invert a jar in the water. Run a tube from the ziplock bag to the jar so that the gas from the ziplock bag is captured, and now you have something that can be tested.
Put a smouldering match in the jar, and if it catches fire then the gas is probably oxygen. Put a lit match into the jar, and if it goes out it’s probably CO2. Bubble the gas through lime water, if it turns milky the gas is CO2. If the gas goes pop when a lit match is used, it’s probably Hydrogen. Ammonia and Chlorine can be detected by smell, neither is particularly dangerous in small diluted amounts (open the jar away from your nose and gently waft the contents towards you).
If you know what gas is being generated, it will help you pin down the reaction.
Additionally, I strongly suspect that inconsistencies of burning are due to inconsistencies of mixing. If you had some sort of machine that could mix the clay more thoroughly it would probably lead to better results. A simple meat grinder would probably work – hand mix the powders into the clay, then run everything through the meat grinder 5 or 10 times for good consistency.
‘Hella intreresting video.
ChatGPT does not know of any reaction that might happen between Aluminum powder and clay, so this video might be doing original science. A literature search would confirm this.
ChatGPT does point out that in certain circumstances the environment might catalyze the aluminum with water to release hydrogen, viz:
2Al+3H2O→3H2+Al2O3
So if anyone captures the gas and does the flame test, be a little careful – a pint of hydrogen can pop strongly enough to shatter a glass container.
Who cares what ChatGPT “knows”? Search the literature yourself.
I wonder if mixing in a small amount of some kind of oil (either mineral oil or sillicone oil) could act like flux to improve the smoothness of the resulting surface?