Isolation of Caffeine from Tea

This used to be one of a sequence of "natural product isolations" we would do in one lab. Now we do it in conjunction with the following exercise, in which students analyze their product by thin-layer chromatography and use it as a standard to identify caffeine in over-the-counter analgesics. These are both short exercises and fit easily into a single lab period.

The general approach is to make the tea alkaline, extract with methylene chloride, and recover the caffeine and the solvent by distillation. We used to use a cheap brand of cola for the isolation, which we allowed to degas completely before using, since carbonic acid interferes with the extraction. I have read that a cup of cola contains between 25 and 50 mg of caffeine, whereas coffee or tea has two or three times as much. Now I brew double-strength tea for students to use, and yields are better.

Students are told to add 10 mL of 10% sodium carbonate to 70 mL of tea in a 125-mL separatory funnel, then add 5 mL of methylene chloride. After stoppering, they swirl the mixture gently (to avoid emulsification), being careful to vent the funnel frequently. The methylene chloride layer is drained into a small beaker and the extraction is repeated. The beaker containing the combined solvent layers invariably contains some of the tea. If there isn't too much, the solvent can be transferred cleanly by careful pouring into a large (1-inch) test tube. If there is lot of tea, small drops, or an emulsion, two scoops of sodium chloride are added to the beaker and swirled a few times; the methylene chloride can then be easily poured into the large test tube. This test tube has been preweighed empty and dry. Students then set up a simple distillation arrangement whereby a bent glass tube carries vapors to a 18 x 150 mm test tube in an ice bath. Heat is supplied by a 600-mL beaker containing water at 75-80 degrees. When the solvent is gone, the cooled test tube is weighed again to determine the mass of (crude) caffeine recovered. If necessary due to traces of water, the tube is dried briefly in an oven, and students are cautioned not to weigh the tube until it is cool. Invariably, some students find that the mass of the tube after distillation is less than the mass before (!), so I simply have them weigh their tubes again later when they are clean. Reasonable recoveries range from 10 to 20 mg.

The exercise concludes with a series of questions leading them to an understanding of why sodium carbonate is used, incorporating the ideas of tertiary amines' acid-base chemistry and "like dissolves like."


Return to CHM 1102

Return to Barry Ganong's Homepage

Revised 9/1/06