June 14 to June 18

This past week, we ran into a little bit of a stall and had a "back to the drawing board" moment. We originally wanted to do high-performance liquid chromatography (HPLC) analysis, but we ran into a problem with that because our chocolate samples wouldn't dissolve in the solvents that the HPLC machine uses. The next idea was to use liquid chromatography-mass spectroscopy (LCMS) or to find another HPLC method, so we sat down with some literature to see what other people have done and find ourselves a starting point.

For more details:

We eventually narrowed down our search to two articles. The first was an article by Sinclair and Marc published in Journal of Chemical Education's Nov. 2009 issue. The article quantified methylxanthine, a purine base that compounds such as caffeine, theobromine, and theophylline derive from. Incidentally, those three compounds were the three that the article quantified and can also be found in cocoa products. The results of this article showed that theobromine and caffeine could be quantified in chocolate using HPLC and another method called thin-layer chromatography (TLC), both of which are available for use in our labs. The method of separation was also relatively simple (basically dissolving chocolate powder in water and then running the sample through a C18 column in order to separate the methylxanthines from the other compounds that make up chocolate), so we decided to give this analysis a shot. We're hoping that the result of this test will show some sort of correlation between the amount of methylxanthine and the percentage of cocoa found in our samples.

As of right now, we are testing this method with cocoa powder to make sure that our machines can detect methylxanthines and will update about our progress.

Another test that we are thinking of possibly trying is to use LCMS in order to detect the amount and type of free fatty acids (FFA) in our samples. This is because it is possible for LCMS to distinguish between different FFAs. The reason why we would want to do this is because luxury brands of chocolate use mostly cocoa butter (which is fatty acid extracted from cocoa upon roasting the cocoa beans), while cheaper brands of chocolate are more likely to use vegetable shortening in place of cocoa butter. We're still in the process of figuring out of the extraction for this method is possible with the resources that we have, so until we know definitively that we can do this test, FFA analysis is just another possible idea.

One more thing we did this past week was that we obtained some white chocolate for our inventory. The good thing about white chocolate is that it is mainly cocoa butter and sugar, so it should be a good control for our data. Andrew is still tweaking is spectra analysis program, and one of the things we're thinking of trying is to subtract the spectrum data of white chocolate from the other samples to see if it is possible to find a difference between our samples when the sugar and cocoa butter spectra are removed. We're definitely getting closer to obtaining results for this method, so stay tuned!