For one thing, you're making things tougher to measure by using a sample that small (2 sq.in.). 1 or 2 milliamps count when you are anodizing in "microscale".
If your sample is 2.0 sq.in. then (2/144) x 4.5 A/sq.ft. = 0.0625A = 62.5mA. Your anodizing current of 0.06A (60mA) is a current density of 4.32A/sq.ft. Your peak voltage calculation of 10.8 V is correct at 4.32A/sq.ft. At 4.5A/sq.ft. you would have expected 11.25V.
Notice that just 2.5mA made a difference of 0.45V.
Since I promised not to talk about the "C word", let it suffice to say that many have trouble with the "C word" until they understand what "tight" means in the DC electrical sense. Erratic voltage swings indicate this problem. 1 year old red dye would certainly be suspect. The pH of the dye changes with time, like a week or maybe less. A year? Who knows.
I don't know how effective the nitric acid / ammonium bifloride would be at removing the native oxide (essential step). I suggest you try using a sample maybe 5X larger to avoid metology issues, and some old fashioned sandpaper to strip the oxide until you get things sorted out.
|