The preparation of decahydroquinoxaline by four different methods, with no determination of geometrical configuration made, was reported in the literature. The purpose of this thesis was to establish the geometrical configurations of the decahydroquinoxalines prepared by these four methods and to determine if, as the work of previous investigators suggested, the two pure isomers of decahydroquinoxaline isomerized, each giving an equilibrium mixture of both isomers. Decahydroquinoxaline was prepared by three of the four previously reported methods of preparation. Comparison of the properties of the decahydroquinoxalines prepared and the preparation, comparison, and analyses of the diacetyl and the dinitroso derivatives of these decahydroquinoxalines showed that two of the methods, Shultz's, et. al., and Weston's, et. al., gave the same high melting decahydroquinoxaiine, while the decahydroquinoxaline prepared by the method of Broadbent and Whittle gave the low melting isomer. The fourth method of preparation, that of Mousseron and Combes, gave a product which by comparison with the N,N'-dinitroso derivative of 1,2,3,4-tetrahydroquinoxaline and in view of the ease of dehydrogenation of cis-decahydroquinoxaline appeared to be 1,2,3,4-tetrahydroquinoxaline instead of decahydroquinoxaline. It was found that the preparation of decahydroquinoxaline by the method of Shultz gave a mixture of cis- and trans-decahydroquinoxaline, with the trans isomer being isolated in suprisingly low yields (12.5%) by recrystallization from petroleum ether. In contrast, the decahydroquinoxaline prepared by the method of Broadbent and Whittle was obtained in nearly quantitative yields and had the cis configuration. A little trans-decahydroquinoxaline and traces of 1,2,3,4-tetrahydroquinoxaline were also obtained from the latter preparation. Previously possibilities of the isomerization of each isomer of decahydroquinoxaline, one into the other, were considered. These were found not to occur, but that the change in melting point of cis decahydroquinoxaline with standing was established as being due to absorption of carbon dioxide from the air forming the carbonate salt. The infrared spectra of aged cis-decahydroquinoxaline of changed melting point clearly indicated that carbonate salt formation had occurred. cis-Decahydroquinoxaline stored under carbon dioxide-free nitrogen did not change meIting point on standing. Chromatographic separations by previous workers of aged cis-decahydroquinoxaline suggested that isomerization occurred. However, fresh cis-decahydroquinoxaline was similarly found, by chromatographic separations, to contain both isomers. Chromatographically pure cis-decahydroquinoxaline on standing did not give any trans-decahydroquinoxaiiiie on being chromatographed. Neither isomer or decahydroquinoxaline was found to isomerize on standing and the melting point was established as being due to carbonate salt formation only. In addition to analytical data, confirmation of the structure of decahydroquinoxaline was further obtained by pK_a data, infrared spectral data, and the dehydrogenation of cis-decahydroquinoxaline to 1,2,3,4-tetrahydroquinoxaline. Ultraviolet spectra of cis-decahydroquinoxaline was finally obtained which did not indicate the presence of unsaturation. It was found that the traces of 1,2,3,4-tetrahydroquinoxaline contaminating cis-decahydroquinoxaline and responsible for the indications of unsaturation in the spectra could be removed by careful vacuum fractional sublimation or by chromatography. The spectra of cis-decahydroquinoxaline so purified did not show indications of unsaturation. Determination of the geometrical configuration of the isomers of decahydroquinoxaline was attempted by resolution of the racemic (D,L) trans isomer. Resolution was either not effected or not detectable. However, the stereospecific synthesis of decahydroquinoxaline from authenic trans- 1,2-diaminocyclohexane showed the high melting isomer, prepared by either the method of Shultz or the method of Weston, to be the trans isomer. The low melting isomer prepared by Broadbent and Whittle was, thus, established as the cis isomer. A number of new N-substituted and N,N'-disubstituted derivatives of decahydroquinoxaline was prepared. Table 1 lists these compounds. Very likely the predominant factor causing, in some instances, a lesser yield of the substituted cis compound compared with the trans compound was steric hindrance in the cis isomer.



College and Department

Physical and Mathematical Sciences; Chemistry and Biochemistry



Date Submitted


Document Type








Included in

Chemistry Commons