W.T. Lagarde

I am currently performing genotyping in mouse tails using Proteinase K. In order to prolong the shelf life of our Proteinase K we wanted to create a storage buffer. One important part of this buffer is 50% (v/v) glycerol. According to all the literature I could find, the pH of glycerol should be about 7-7.5. However, after discovering our Proteinase K still would not keep, I decided to check the pH of all my ingredients. It turned out, our glycerol (99+%) was actually around 3.5. A different batch of glycerol, with a slightly lower concentration of glycerol (87%, diluted with Milli-Q) was even slightly lower at a pH of 3.2. Even stranger, diluting both batches of glycerol to the 50% required by our buffer brought the pH down even further. To me, and many of my colleagues, this made no sence at all. All possible explanations I am able to come up with seem very unlikely. Are both batches of glycerol somehow tainted? Is there some reaction going on? Can some of the trace elements (Heavy metals or magnesium <5ppm) have such an extreme effect on pH? Since I am not able to figure this thing out, I hope someone out there might have the answer for me. In any case, all help will be greatly appreciated.