BIXIN 



257 



c) From bixa-seeds*. The seeds are covered with water and allowed to stand 

 for several hours. They are then stirred mechanically and filtered through a sieve. 

 The turbid solution is allowed to stand overnight in large percolators and the lower 

 layer is separated and centrifuged. The residue is broken up and dried, first in air 

 and then in vacuum over calcium chloride, until it is not brittle but can be pressed. 

 After grinding in a mill, 100 kg of seeds yield about 5-6 kg of a material which 

 contains 15-30% of bixin. 



This material, in portions of 200 g, is immediately covered with portions of two 

 litres of ethanol, warmed to 60-65° on a waterbath, and ammonia is passed in until 

 the colour change is complete and the solution contains free ammonia. The reaction 

 mixture is allowed to stand for 20 minutes, filtered warm, and the residue is stirred 

 up with 1 litre of ethanol. Into this mixture, ammonia is again passed at 60°. It is 

 then allowed to stand for 1 hour and filtered. The combined filtrates precipitate 

 ammonium bixate on cooling. In order to complete the precipitation, 1 ml of acetic 

 acid is added to each litre of solution and the solution is vigorously stirred. After 

 a short time, a dark red resin separates on the stirrer and on the walls of the vessel. 

 The resin is separated from the mother liquor and treated with acetic acid, with 

 vigorous mechanical stirring. After some hours, free bixin separates and can be 

 filtered and dried in vacuum over sodium hydroxide and calcium chloride. It is 

 then crystallised from acetic acid. (About 18 g of boiling glacial acetic acid are 

 required to dissolve 1 g of crude bixin). About 120-160 g of pure bixin are obtained 

 from 100 kg of seeds. 



Chemical Constitution 



CH3 CHj CH3 CHj 



II II 



HaCOOC- CH=CH- C=CHCH=CH- C=CHCH=CHCH=C- CH=CHCH=C- CH=CH- COOH 



Bixin 



The elucidation of the constitution of bixin extended over several years. 

 Heiduschka and Panzer^ determined the correct molecular formula C25H30O4 

 An unsymmetrical structural formula was first proposed, but was abandoned 

 when KuHN and Winterstein^ put forward the now accepted formula of 

 bixin. This has been confirmed by the investigations of Karrer and co- 

 workers^", in the course of which perhydronorbixin was synthesised. 



Herzig and Faltis^^ recognised that bixin was the monomethyl ester of 

 an unsaturated dicarboxylic acid. Bixin absorbs 9 mols of hydrogen on cata- 

 lytic hydrogenation and is thus converted into the half-ester of a saturated 

 dicarboxylic acid. From the deep red colour of the pigment it can be concluded 

 that the 9 double bonds are conjugated. By the ozonisation of methyl bixin 

 RiNKES and van Hasselt^^ obtained the methyl ester of /3-acetylacrylic acid 

 and methylglyoxal. Methylglyoxal must be derived from the grouping 

 = CH.C(CH3) =, while the first product indicates the presence of the grouping: 



A detailed description is given by R. Kuhn and L. Ehmann, Helv. chim. Acta 12 

 (1929) 904, and E. ForcAt, Dissertation, Zurich, 1930. 



References p. 2^o-2g4. 

 Carotenoids 17 



