Infrared determination of organic structures

Abstract

vigorous mulling caused a clear-cut polymorphic transition without any detectable changes of the first kind. Pelleting brought about the same conversion, but produced some amorphous regions at the same time. Heating the pellets quickly eliminated the amorphous phases and then slowly aided change to the second structure. Since the original crystal structure was subsequently found to be unstable at normal conditions, the effectiveness of mulling in aiding transition is easily understood. Less readily explainable are the differences in the spectrum of stable 2-aminobenzoic acid given in Fig. 7. The first and second spectra are from a hand ground mull and a machine ground mull, respectively, and they show very marked differences in the bands indicated by arrows. Since the bands which show increased intensity after vigorous mulling have not broadened, the changes may be due to polymorphism. _ If so, then pelleting in KBr causes transition to still a different crystalline structure as shown in the fourth spectrum. The third spectrum is from a KI pellet ground for 10 sec. and is very similar to the first mull; the only changes are a slight broadening and decreased resolution of some bands, but the changes are much less than those produced by vigorous mulling. Similar changes have been observed sufficiently often in a number of other compounds such as 2,4-dinitropheny 1-hydrazine, salicylic acid, etc., that any comparison of solid-state spectra, whether mulls or pellets, should allow for possible crystal changes. Conclusion Although the pellet technique has some serious limitations, it is still eminently suited for obtaining spectra of microsamples and of amorphous polymer or resin samples. With some precautions, crystal distortions usually can be minimized in all but the most sensitive samples, and the pellets, while occasionally giving spectra not quite as good as mulls, will yield spectra which are good enough qualitatively for most problems. In particular, cautious grinding followed by appropriate heat treatment will give many excellent spectra from pellets which would otherwise give very poor spectra. On the other hand, rather than trying to avoid crystal changes, these have been induced purposely to solve problems encountered in polymorphism. This is usually easier and faster than recrystalliza-tion or cocrystallization. Rather than being relatively rare, polymorphism occurs very frequently in organic preparations, especially if stabilizing impurities are present or if different solvents are used in recrystallization. For the reasons outlined above, the pellet technique should have a place beside the mineral oil mull in every laboratory. In combination, these two methods of obtaining spectra will yield more information about a sample than will either one alone. The spectra of pyridine and 33 substituted pyridines are compared to establish correlations of the absorption bands with the molecular structure of the compounds. Five groups of compounds are considered, namely, 2-, 3-and 4-monosubstituted, and di-and tri-substituted types. Absorption bands that are generally characteristic of an alkylpyridine system are found near 1600, 1570 and 1000 cm.-1. For 3-alkylpyridines and 2,5-dialkylpyridines, the latter peak is removed to 1021-1034 cm.-1. Peaks in the regions 1280-1330 cm.-1 and 1222-1253 cm.-1 are strong confirmatory evidence for the alkylpyridine system. The separation of the absorption bands near 1600 and 1570 cm.-1 is approximately 40 cm.-1 for 4-monoalkyl-pyridines and 20 cm.-1 for the other two types of monosubstituted pyridines. The 4-monosubstituted pyridines also have a band in the region 1067-1072 cm.-1. The 2-monosubstituted pyridines have bands at 1050 cm.-1 and in the region 1146-1152 cm.-1. The 3-monosubstituted pyridines have bands at 1117-1131 cm.-1 and 1180-1196 cm.-1. Disubstituted pyridines all have a peak in the region 1099-1136 cm.-1. No correlations were found in this region for trisubstituted pyri-dines. Out-of-plane deformation vibrations for 2-monoalkylpyridines were found in the region 743-750 cm.-1; for 3-monoalkylpyridines at 789-810 cm.-1 and 712-715 cm.-1; for 4-monoalkylpyridines at 785-822 cm.-1; for disubstituted pyridines at 816-833 cm.-1 and 725-743 cm.-1; and for trisubstituted pyridines at 724-732 cm.-1. For monoalkylpyridines the relative intensity and location of absorption bands in the 1667-2080 cm.-1 region differ only if the position of substitution differs. Typical absorption patterns in the region are given for 2-, 3-and 4-monoalkylpyridines. Patterns are also suggested for 2,3-, 2,4-, 2,5-and 2,6-disubstituted pyridines. The patterns for the disubstituted compounds may need some modification as additional compounds become available. Correlation of the infrared absorption bands of simple compounds with their structures can provide a means for determining the structure of more complex compounds. This paper presents correlations based on the infrared spectra of pyridine and 33 substituted pyridines. The infrared absorption spectra are discussed in relation to each of five classes. These are 2-, 3-and 4-monosubstituted, disubstituted and trisubstituted classes. A sufficient number of compounds of each monosubsti-tuted type were examined to indicate the general applicability of the correlations. The disubsti-tuted and trisubstituted compounds were principally methyl-substituted. Hence the ranges given for the correlations of the spectra of the latter (1) Bureau of Minea, Regipn III, Laramie, "Wyo. compounds may have to be broadened as additional compounds become available. Most of the previous work has been based on the spectra of pyridine2-4 and the methylpyridines.6 Bellamy6 has reviewed work through 1953 concerning the assignment of vibrations for pyridine, the monomethylpyridines, 2,6-dimethylpyridine and various alkaloids containing the pyridine (2) C.