Last night, Protospace member Chris and I performed a test of how well PLA burns.
Before testing the rocket, I wanted to first test how well PLA would burn when its surface area-to-volume ratio (SA:V) was high. There have been videos showing how well PLA filament burns (drippingly), and MakerBot and others have warned against making 3D printed candle holders, but I had not seen any that showed it burning furiously as would be desired in a rocket.
So, I took one of the two rocket nozzles I'd printed (the one that didn't print well) and put some PLA bits (chopped-up skirt from printing the nozzles) into it, topped with some sawdust and butane to start it. Chris and I went out in the parking lot to test it.
The first few attempts didn't work—the igniter spark gap was too far from the sawdust.
The BBQ igniter ignited it successfully after several attempts, with Chris squirting butane as I squeezed the igniter repeatedly. It burned pretty well, but the sawdust and butane burned out before it got down the PLA.
We tried again, this time with forced air from below using the air gun (not actually the same air gun as used later in this experiment, not that it really matters). This resulted in the side of the nozzle catching fire, and it burned very well, with a blue flame like a Bunsen burner or blowtorch.
The actual PLA bits that I had intended to test didn't really catch fire much, but that's OK—the nozzle itself demonstrated that PLA burns strongly with forced air. This is quite promising for using PLA as rocket propellant.