I've been experimenting with rocket fuel mixtures for "r-candy" (KNO3 + table sugar) rockets. The canonical mixture is 60/40 by weight (always oxidizer first), or 65/35 for less pure KNO3, though that is known to give far worse Isp.
While a 60/40 mixture is highly efficient, it can have problems with combustion stability at high thrust values. Since thrust is varied by changing the size of the bore and the ratio Ab/h (burn surface vs length), it can become impractical to have very high thrust engines since they would end up being very wide, and thus be inefficient anyway - so it can be worth adding catalyst or intensifier chemicals to those engines.
I tried two intensifiers: Fe(OH)3 (rust) and Al2O3 (aluminum oxide). Both provided between 10 and 20 percent additional thrust, but caused some combustion instability and a very marked decrease in efficiency.
The control engine burned for four seconds and produced a normal thrust curve: large ignition thrust, 4.5 seconds of sustained thrust, and a 0.5 second burnout taper.
My first mixtures were 55% oxidizer, 35% fuel, and 10% intensifier by weight (55/35/10).
The iron oxide mixture exploded the first time due to being improperly mixed. Its second test produced a huge initial thrust for about 0.22 seconds, about one second of sustained low thrust, and then exploded due to a crack in the grain. The third test produced a thrust profile very similar to the control engine with a much larger initial thrust, a shorter (only 3 second) sustained burn at about 18% more thrust, and then instead of a taper-off another spike which looked like the result of a small explosion.
My aluminum oxide mixture worked well all three times I fired it with none of the explosive properties of the iron oxide mixture. They produced a much more sedate curve; rather than an initial massive acceleration with a long sustained burn, they fired for a total of between 2 and 2.5 seconds at between 20% and 5% more thrust than the sustained thrust of the control engine. There were, however, occasional dips to below 50% nominal control thrust, accompanied by an expellation of sparks and very dark smoke. I don't know why this is; I'll be investigating further in a later post.
These engines did have a tendency to fail to ignite, but I think that can be solved with a better ignition system.
Overall, the 10% aluminum oxide fuel was much more successful; it was easier to use without risk of CATO and produced significantly more thrust; however, it had only 60% of the total impulse.