Bad news/Good news on the 1g sample trials: bad first, getting 0.002 to 0.004g shortfalls on the std process yields using ingredient changes, which on a 1 gram sample means 0.2 to 0.4 points lower than the expected 1.1% yield.  Sources of these differences could include: scale precision, product losses due to std method transfers, poor choice of ingredient changes, operator error.

I hope to reduce the scale precision risk in the future by using an analytical balance to measure weight.  To reduce the other risks, brings me to the good news.

The best yielding run this time, yielding 1.1%, involved doubling the lime addition from the scaled quantity with preparation of paste and pulls in the same test tube.  Specifically: 0.5g lime and 3mL distilled water were mixed in the test tube, 3 Army Painter mixing balls and 1g cactus powder were added, then using a mini-vortex mixer, the paste was mixed for 3 minutes, then rested for 4 minutes.  Using fresh EA, 6mL EA was added to the test tube and vortex mixed for 1 minute, rested for 2 minutes, and poured through a coffee filter into another test tube; then, using same times and 4mL of EA, 5 more pulls were done.    The clear combined pulls were then salted with 0.3g citric acid, swirled, passively crystallized for 48 hrs, then filtered and washed.

I’m going to focus on this method for my next runs, trying to identify additional changes that will also yield at the expected %.