596 VI. CAROTENOIDS AND RELATED COMPOUNDS 



The correctness of the Kuhn-Winterstein formula was proved by Karrer 

 and associates/**'**^ who demonstrated the identity of the synthetically 

 prepared perhydrobixin ester with the compound formed on hydrogenation 

 of the natural bixin. 



The so-called "stable bixin" is a compoimd having a structural formula 

 identical with that of the natural (labile) bixin. The occurrence of the 

 isomer was first demonstrated by Herzig and Faltis**^ during the prepara- 

 tion of bixin from orlean. In a single, unreproducible experiment, these 

 workers failed to obtain the usual (labile) bixin, but olitained instead a sub- 

 stance which had a higher melting point, a greater stability, and a longer 

 wave-length spectrum. They called this isomer "jS-bixin"; the term 

 "stable bixin" is now applied to it. 



Labile and stable bixins are now known to be cis-trans isomers. Karrer, 

 Helfenstein, Widmer, and van Itallie**^ confirmed the occurrence of the 

 stable form and were the first to suggest that the two types of bixin are 

 geometrical isomers. The cis-trans relationship was indicated by the fact 

 that the transformation of ordinary bixin to stable bixin (/3-bixin) could be 

 effected by iodine''^-**^ even when this was used in catalytic amounts, or by 

 perbenzoic acid in chloroform.**^ The structural identity of the two bixins 

 was proved by the demonstration that when two hydrogens were added 

 either to labile bixin or to stable bixin by treatment with zinc and glacial 

 acetic acid or titanium chloride and ammonia, the same dihydrobixin re- 

 sulted in both cases. **^'**^ If the dihydrobixin so formed is oxidized in the 

 air in the presence of pyridine, stable bixin results. It is therefore possible 

 to change the labile into the stable isomer by this procedure. 



Zechmeister*^"'*^' has presented proof that labile bixin (also called 

 natural bixin, a-bixin, bixin II, and lower melting bixin) has the cis con- 

 figuration, while the stable bixin (/3-bixin, bixin I, and higher melting bixin) 

 has an oW-trans structure. The formulas for the two isomers are repre- 

 sented diagrammatically in Figure 3. 



b. Norbixins. The norbixins have the same structural formulas as the 

 bixins except that both carboxyl groups are free. The empirical formulas 

 are C24H28O4. The structural formula is represented here. Two forms of 



^*^ P. Karrer, F. Benz, R. Morf, H. Raudiiitz, M. Stoll, and T. Takahashi, Helv. 

 Chim. Acta, 15, 1218-1219 (1932). 



4« P. Karrer, F. Benz, R. Morf, H. Raudnitz, M. Stoll, and T. Takahashi, Helv. Chim. 

 Acta, 15, 1399-1419 (1932). 



"6 J. Herzig and F. Faltis, Ann., 431, 40-70 (1923). 



^« T. Takahashi, J. Pharm. Soc. Japan, 56, No. 1, 352-355 (1936); in German, pp. 

 48-50; Chem. Abst., 30, 6348 (1936). Cited bj^ L. Zeehmeister, Chem. Revs., 84, 324 

 (1944). 



«8 R. Kuhn and P. J. Drumm, Ber., 65, 1458-1460 (1932). 



*" R. Kuhn, P. J. Drumm, M. Hoffer, and E. F. Moller, Ber., 65, 1785-1788 (1932). 



«" L. Zeehmeister and R. B. Escue, /. Am. Chem. Soc, 66, 322-330 (1944). 



«i L. Zeehmeister, Chem. Revs., 34, 267-344 (1944). 



