2 DETECTION AND DETERMINATION OF ALCOHOLS, ETC. 



will probably vary from 0.1 to 1 per cent. An amount less than 0.1 

 per cent would, in the writer's opinion, call for further investigation 

 to determine its significance ; an amount greater than 1 per cent would 

 probably represent a degree of decomposition such as to render the 

 product of such* an appearance as to be unsalable. The method used 

 to estimate amounts of alcohol within these limits must be of such a 

 nature as to leave no possible doubt that the substance determined is 

 ethyl alcohol. Most of the present methods for detecting or deter- 

 mining small quantities of ethyl alcohol are weak at this point, being 

 based on the quantity of potassium bichromate or other oxidizing 

 agents which the alcohol will reduce. Such are the methods proposed 

 by Bourcart, 1 Nicloux and Baudreur, 2 Benedict and Norris, 3 and 

 others. It is evident from a careful study of these methods that 

 slight changes in temperature cr concentration materially affect the 

 accuracy of the results; indeed, Pozzi-Escot 4 claims that even when 

 uniform conditions are rigidly maintained it is impossible to obtain 

 accurate results for alcohol determinations by methods involving 

 oxidation with potassium bichromate and sulphuric acid. 



As very many substances reduce potassium bichromate, it is 

 evident that in a determination based on that process there is no 

 surety whatever that the substance so determined is ethyl alcohol. 

 The method of Verley and Bolsing, 5 in which the acetic ester is pre- 

 pared by the action on alcohols of acetic anhydrid in the presence of 

 pyridin, is not accurate enough to be applied to very small quantities 

 of alcohol, and it is evident that nearly any alcohol will give the same 

 reaction. 



As none of the proposed methods for determining alcohol when 

 present in small amounts answers the requirements for accuracy and 

 reliability of the work under the food and drugs act, studies to 

 improve them were begun. The possibility of concentration of the 

 alcohol from solutions containing only a small percentage of this 

 substance was first considered. Nicloux and Baudreur 6 found that, 

 in distilling 600 cc solutions containing 0.2, 0.1, and 0.033 per cent of 

 alcohol, respectively, from a liter distilling flask, there was obtained 

 in the first 30 cc of distillate 50, 52, and 55 per cent of the total 

 alcohol present. In repeating this work, using a Glinsky distilling 

 head, the writer found that by these distillations it is only possible to 

 obtain from 50 to 60 per cent of the total alcohol, when attempting to 

 concentrate 0.1 per cent solutions into such a volume as to obtain 

 solutions containing from 2 to 4 per cent of alcohol. 



It was also found that if a considerable quantity of calcium or 

 sodium eWorld was dissolved in the dilute alcohol solution before 



i Zts. anal. Chem., 1890, 29: 609. * Ann. chim. anal., appl., 1907: 112. 



Bull. soc. chim., 1891, (3) 17: 424. Ber. d. chem. Ges., 1901, 34: 3354. 



J. Amer. Chem. Soc., 1898, 20: 293 Loc. cit. 



