ESTIMATION 



identical with the nicotinic acid colour in a yield equivalent to 70 % 

 of the theoretical conversion. The solution was treated with Lloyd's 

 reagent as before and the colour evaluated. The value obtained in 

 the second estimation was equivalent to the amount of nicotinic acid 

 derivatives measured in the first estimation plus 70 % of the tri- 

 gonelline. 



Hogberg et al.^^ observed that the colour produced by nicotin- 

 amide was more intense and more stable in butanol solution than that 

 produced by nicotinic acid, so that if the coloured derivatives were 

 extracted into butanol and the extract left for ten minutes the resid- 

 ual colour was due solely to nicotinamide. To estimate bound nico- 

 tinic acid and nicotinamide, the material was heated with o-i N 

 hydrochloric acid in a sealed tube at 100° C. and the estimation 

 repeated. 



Other Chemical and Physical Methods of Assay 



Atkin et al^^ made use of the Hoffman reaction with bromine and 

 potassium hydroxide to destroy nicotinamide, in order to improve the 

 accuracy of the microbiological method of assay (page 227). The 

 same reaction was used by J. M. Goodyear and H. W. Murphy ** as a 

 method of estimation, the 3-aminopyTidine into which nicotinamide 

 was converted by this treatment being diazotised and coupled with 

 N-i-naphthylethylenediamine. The intensity of the red dye thus 

 produced was proportional to the amount of nicotinamide originally 

 present. 



J. J. Lingane and O. L. Davis ^^ developed a polarographic method 

 of estimating nicotinic acid, which gives a characteristic wave, the 

 position of which varies with the base solution used. The best results 

 were obtained with a tetramethylammonium borate buffer solution of 

 pK 9 ; nicotinic acid then had a half -wave potential of — i-6 volts, 

 but an unbuffered potassiimi chloride solution gave quite satisfactory 

 results. The method has the advantage that aneurine, riboflavine 

 and nicotinic acid can be estimated simultaneously in the same solu- 

 tion. 



Nicotinamide is also reduced at the dropping mercury electrode ^® 

 in alkaline solutions giving a well-defined wave at — i-6 to — 1*7 

 volts, the height of which is proportional to the concentration of 

 amide. The mechanism of the reaction is not known. N^-Methyl- 

 nicotinamide was reduced in two steps ; the first, at — 1-025 volts, 

 is possibly due to the formation of a dimer, and the second, at 

 — 1-6 to — 1-8 volts, to the formation of a dihydro-derivative. 

 Trigonelline gave a single step at — 1-35 volts due to the formation 

 of a dimer. 



15 225 



