Synthesis and physical studies in heterocyclic systems.

Abstract

The research described in this thesis entitled "Synthesis and Physical Studies in Heterocyclic Systems" had its origin in an attempt to answer the question: "Does the placement of a substituent on a phenyl ring affect its magnetic anisotropic properties by way of modifying its 'ring current. PART I That specific groupings affect chemical shifts anisotropically was recognised early in the practice of NMR spectroscopy. Notwithstanding the imperfect development of an exact and quantitative theory of the chemical shift, the fact of existence of anisotropic effects has been usefully employed in numerous structure elucidations of molecules of both chemical and biological interest. These matters are touched on in Chapter I. For attempting to answer the question raised in the first paragraph, it was necessary to search for an appropriate system constituting a series down which a change in substitution induces a variation in the chemical shift of a strategically located probe group. It was necessary that any observed change in chemical shift of the probe group be largely attributable to a change in the anisotropic property of, say, a phenyl ring included in the structure, possibly as a result of a modification of its 'ring current'. The choice fell on systems 1 where CH?CN is a probe group and the location of the changing substituent could be considered as remote enough to exert little or no through-bond effects. The synthesis of systems 1 was based on a known procedure (Chart I) for substituted pyridines. It was adapted for the synthesis of cycloalkenopyridines such as 1 by an appropriate choice of the ?-methylene ketones 2. A strategy, consisting in an attempt to shorten the procedure somewhat by allowing the initial condensates 4 to react directly with ammonia for conversion into the amidines 1 and thence to the required products 6, in a one-pot reaction, proved successful even though it led to a certain loss of starting vinamidiniums 3 by way of formation of side-products, the diarylpyridines 9 and arylpyrimidines 10. The required separation proved to be quite facile. With the test systems 1 thus obtained, the substituent-induced change in the chemical shift of CH?CN was found, not unexpectedly, to be rather small but it distinctly fell outside the range of experimental error in chemical shift measurements. It was found that the change did not correlate with substituent properties even though the correlations were essayed versus different dual substituent parameter sets. It was then ensured that transmission of effects into the pyridine ring of 1, comparable to that in 4-substituted biphenyls, was indeed taking place by noting the similarities in the correlation of chemical shifts at C-3 and C-4 of 1, respectively, with C-1 and C-4 of the biphenyls. An attempt to explain the lack of correlation at CH?CN is the subject of Chapter III. PART II The diarylpyridines 9a could only have arisen from reactions involving two molecules of the vinamidinium salts. The detection of arylpyrimidines 10a as products that accompanied the formation of the pyridines 9a as well as the formation of isomeric diarylpyridines 9b and arylpyrimidines 10b from the similar reactions of vinamidiniums 2b allowed the setting down of general mechanisms for the formation of all these products. These mechanisms involved an apparently hitherto unreported dimerization of azadienes of type 11 derived from the vinamidiniums by replacement of at least one dimethylamino moiety by ammonia. The regioselectivity implied in the formation of pyridines 9ab and pyrimidines 10ab from systems 2ab indicated that only certain mutual orientations of the cyclodimerizing azadienes are conducive to the formation of precursor cyclates. It was of interest to test if these 'successful' orientations would be the ones predicted by frontier molecular orbital (FMO) perturbation theory. It was found that all the observed products, including those from reactions in which a 1:1 mixture of the vinamidiniums 2a and 2b had been taken for reaction, must have been derived from initial cyclates that are predicted by FMO theory. However, reactivity considerations, apparently extraneous to the theory, had to be invoked for attempting to account for changes in the relative proportions of the components of the same or similar product mixtures formed in the different reactions. Before describing the method of application of FMO and results thereof, a short piece of background information on the theory is provided (Chapter VI). Full experimental details, methods of spectroscopic identification of products, description of calculational procedures, etc., follow in Chapter VII. Finally, attempts to prepare certain substituted anthranils 12, the intended precursors of 7-substituted 4-acetylbenzofurazan N-oxides 13, are described in an Appendix. The purpose was to study the indicated equilibria. The various methods tried either did not lead to the required regioselectivities or did not yield expected products. The experiments and product identifications are described.