Challenging Problems Involving Benzenoid Polycyclics and Related Systems

Five theoretical and/or experimental problems involving benzenoid polycyclics and related eystems are discussed. The dualiet graph characterizes uniquely any polycyclic aromatic hydrocarbon (PAH or pol.yhex), and differentiates between catafusenes, perifusenes and coronafusenes (catahexes, perihexes, coronahexes). 1. The topology of catafusenes, expre8sed by their dualist graphs, can be correlated with the rates of Diels-.Alder addition to maleic anhyd-. ride (determined by Bierxnann and Schmidt, 1980). The most important topological parameter in this correlation is the number of rings in the longest acene portion of the catafu8ene. The topology at the ends of this longest acene portion constitutes the second parameter. 2. As known, the carcinogenic activity of PAH's is connected with the existence of bay—regions. Simple topological rules making use of dual.. ist graphs can be implemented by means of a computer program to explore all PAH structures with or without bay—regions. However, the problem is much more complex, because there exist many PAll's with bay—regions that are not carcinogenic, and a few carcinogenic PAH's without a bay—region. 3. Owing to its remarkable structure of coronafusene, kekulene may afford aza—analogues which should lead to highly interesting metallic complexes. The number of possibilities is much higher than in the case of porphyrins and related structures. These possibilities are reviewed and their potential applications are highlighted. As yet, no experimental data are available, but such polydentate ligands represent an exciting synthetic challenge for organic chemists. 4. So far, no experimental determinations of "local ring current" or "benzene character" in PAll's exist, though the theory has been much discussed (Clar, Polansky, Herndon, Randi6, Hosoya, Aihara, Gutman, Trinajsti, Bonchev). By means of proton NMR spectra of N—aryl—2,4,6— trimethylpyridinium salts (readily available from primary arylamines and 2,4,6—trimethylpyrylium salts), one can measure ring currents in the N—aryl group, and one should be able to calculate therefrom local ring currents. Experimental data with all isomeric naphthyl and anthryl groups show, however, that the theory of shielding is not precise enough Johnson—Bovey's data yield too high, Haigh—Mallion's data too low effects relatively to the measured ones. Even using corrections for dihedral angles and C—N bond distances, discrepancies are too large. 5. By means of dualist graphs, a simple rule is devised for establishing whether a given polyhex is a free radical (mono— or poly—radical), or a normal compound possessing a Kekulé structure. However, there exists a class of diradicals which are not encompassed by this rule, so that a universal procedure for establishing simply whether a given PAM has, or does not have, Kekulê structures, has now been found (Appendix).