THE CHEMISTRY OF COFFEE 175 



Collect the combined chloroform solutions in a tared flask ; immerse 

 in boiling water; distill. 



Evaporate off the solvent (chloroform), while hot, by an air 

 current. 



Dry to constant weight ; weigh residue as caffeine. 



The weight represents the amount of caffeine present in 10 grms. 

 of coffee.^ 



Crystalline Caffeine Properties 



Caffeine forms long, white, silky, flexible needles, which readily 

 adhere and mat together to form light fleecy masses. When deposited 

 slowly, crystals present a characteristic appearance under a magnify- 

 ing power of 100 to 300 diameters. Caffeine crystals heated to 

 iOO°C. become opaque and friable owing to the loss of water; and 

 the residue is anhydrous caffeine which dissolves without turbidity 

 in chloroform. Anhydrous crystals of caffeine can also be deposited 

 from ether or alcohol. 



Melting Point (Strecker) 233° to 234°C. 



(Allen) 23i.5°C. 



(German Pharmacopoeia) .... 230. 5°C. 



(U. S. Pharmacopoeia) 236.8°C. after drying. 



Sublimation Point (German Pharm.) i8o°C. 



(U. S. Pharm.) I78°C. 



Crystalline caffeine: water content (British Pharm.) 8.49%; 

 (Allen) 7.05% to 7.10%. Caffeine is odorless and gives a bitter 

 taste. It forms an acid solution in ether and alkaline in chloroform. 

 The presence of caffeine can be detected by evaporating a given 

 liquid with nitric acid. The residue becomes crimson or purplish- 

 red when ammonia is added. 



The nature of caffetannic acid has been a matter of dispute. 

 (For method of preparation, see Allen's Comm. Org. Anal. 6 (19 12) 

 645.) Gorter has shown it to be a mixture of chlorogenic and caffalic 



^ Paul and Cownly noted that the caffeine obtained by the evaporation of 

 chloroform is likely to contain small quantities of a brownish, waxy, or 

 resinous impurity, and therefore should be purified by resolution in boiling 

 water and recovered by evaporating the filtered solution and drying the 

 residual alkaloid at ioo°C. These investigators found that the proportion of 

 caffeine in coffee varied but slightly, and is not materially affected by roasting 

 excepting a loss in amot:nt up to 21%. This fact recommends the estimation 

 of tlie alkaloid in commercial coffee as a means of ascertaining the proportion 

 of chicory or other adulterant present. 



