A Matter of Scale

This year we tried to quantitate the recovery of caffeine isolated from beverages (an other caffeine sources) with HPLC. Normally, I suggest the mass difference method. Obtain the mass of the empty beaker, add the dried dichloromethane extract, evaporate the dichloromethane in a hot water bath, and reweigh the beaker with the caffeine residue. We have quite modern digital scales – 0.01 gram precision, but the students get some crazy readings sometimes. I’m not sure if the scales can’t be trusted or if there are other issues going on here. This year, the students submitted their caffeine in a 20 mL “scintillation” vial. They could use methanol to transfer the caffeine from their beaker to the vial. I diluted the contents of the vial to 20 mL and took a 10 microliter aliquot of that solution to put in an HPLC “uni-prep” vial. Interestingly, I found that caffeine is actually more soluble in water than it is in methanol. Even though caffeine loves chlorinated solvents like dichloromethane and chloroform, it is more soluble in water than it is in alcohols. So I used water as a diluent as I went along. I had made a standard curve for caffeine in the HPLC at 270nm detection. The students received the HPLC report with peak areas and calculated their original mass of caffeine. The results from the difference mass and the HPLC method rarely were close! In general, the mass determined by HPLC was much less than the difference mass method. I am not exactly sure what it going on. Maybe the residue is not really pure caffeine. Maybe the students tended to lose caffeine in transferring the sample from beaker to vial. Maybe the poor solubility in methanol was a factor as well.

Advertisements

One Comment

  1. Posted December 4, 2010 at 1:57 pm | Permalink

    Is it possible that how much methanol a student used before you dissolved it to 20 mL mattered before doing the HPLC? Could using more methanol make the caffeine more soluble in it and students using different amounts of methanol could somehow fudge the end results?

    I also noticed that sometimes, other people would spill stuff on the scale and another student would weigh their beaker+caffeine on the scale with the other powder on it and that can mess up their scale readings. So it can also be a really big student error.


Post a Comment

You must be logged in to post a comment.
%d bloggers like this: