|more wootz experiments
||[Jan. 4th, 2011|03:07 pm]
When I last posted, I had completed a series of 7 experiments on making wootz using a thermite-based process, all of which failed to produce wootz (see my Oct 1 post). But after a month off they resumed, and here are the results.|
At the end of the last series, I was able to produce molten iron (and alumina) at >3000F. The basic methodology was to mix up some thermite (by weight 3 parts Al to 8 parts Fe2O3), add in a trace amount of vanadium pentoxide (VO5), and attempt to incorporate carbon via different methods, including incorporating charcoal in the thermite, charcoal in the crucible, and a lump of cast iron in the crucible. Tests showed that none of these successfully alloyed the carbon into the iron. For the next series, an additional set of clay flower pots, clay-graphite crucibles, and powdered cast iron of a known carbon content (2.85%) was acquired.
For the following tests, the thermite charge was 300g Al, 800g Fe2O3. Vanadium Pentoxide was not used. Test results show that this mix results in ~330g of iron along with a mass of greyish aluminum oxide (alumina). To achieve the desired final carbon content, an equal amount of powdered cast iron was used (330g). The target for a historically accurate wootz would be 1.2% to 1.8% carbon, so 1.4% should result from such a mix.
For this test, cast iron powder was added into the thermite mix. The reaction started, but failed to achieve a temperature high enough to melt the alumina, resulting in a flowerpot-shaped lump of iron in a semi-fused alumina matrix, currently sitting on the floor pending a decision as to what to do with it. In other words, fail. My best hypothesis for failure is that the cast iron was absorbing too much heat for the alumina to become fully molten. FAIL.
For this test, the cast iron powder was placed in the crucible and the molten iron/alumina allowed to fall and mix. Due to the crucible having picked up some moisture (the weather had been rainy for a couple of days previous), instead the crucible exploded, invalidating this test. FAIL.
Same as above, save except for exploding, the molten iron failed to mix with the cast iron powder, resulting in a layer of iron over a layer of cast iron powder semi-fused together, but not forming an alloy. FAIL.
Since it was clear that the thermite reaction was not sufficient to melt the cast iron, pre-heating the cast iron prior to the reaction was tried. The furnace was a propane-air fired casting furnace, with a #10 crucible inserted with the cast iron powder inside. It quickly became apparent that the air blast was causing cast iron powder to lift out and ignite in the air -- many bright sparks were observed exiting the furnace and bursting once they hit oxygen.
Above: This is the crucible in the furnace just after ignition.
Above: with the lights turned out, you can see some of the sparks. Note as well, the new air supply for the furnace is enough to cause the fire to exit furnace as a flat disc.
The furnace was shut down, the crucible removed, and the cast iron charge re-weighed and brought back up to 330g, with an addition of borax flux added to keep the cast iron powder from flying out. The furnace was reignited, and the crucible and its contents brought up to a yellow heat. The furnace was then shut down, and the thermite mix, contained in the usual flower pots and surrounded by firebrick to contain any sparks, set in place. The thermite was ignited, and drained into the crucible. The setup was allowed to cool overnight.
10 hours later, (this morning, in fact) the crucible was still hot to the touch, but below the boiling point, so the result was taken out of the crucible, the alumina removed, and the cake examined. While not all of the cast iron powder appeared to have been incorporated (there was a section of the bottom of the ingot that showed sintered powder, presumably the cast iron), there was evidence of reticulation (retained carbide crystals in a softer iron matrix) on both surfaces.
Pics from this morning will be up later, but right now I'm having problems getting them uploaded.
Given the partial success, it appears the next logical step is to pre-heat the crucible to the point that the cast iron becomes liquid, and instead of adding flux, a crucible cover be employed to prevent the cast iron powder from escaping. While the furnace is adequately hot enough to melt cast iron (a previous experiment currently lines the bottom of the furnace with a thin layer of iron), an additional layer of kaowool will be added to reduce the size of the burn chamber and increase the insulative efficiency. Should this prove successful, the next step after that is to start scaling up to 1-kg quantities and above, as well as trying different alloying agents to change the carbide patterns.