Cyclic trimer versus head-tail dimer in metal-nucleobase complexes: importance of relative orientation (syn, anti) of the metal entities and relevance as a metallaazacrown compound.

The formation and crystal structure analysis of a cyclic trinuclear Pd complex with bridging 1-methylcytosinato model nucleobases is reported: [[(tmeda)Pd(1-MeC(-)-N3,N4)]3] (ClO4)3.5.5H2O (tmeda = N,N,N',N'-tetramethylethylenediamine; 1-MeC- = 1-methylcytosine deprotonated at exocyclic amino group) is obtained, among others, from the hydroxo-bridged dinuclear species [[(tmeda)Pd(OH)]2](ClO4)2, which likewise has been characterized by X-ray crystallography, and 1-MeC (1-MeC = neutral 1-methylcytosine) in aqueous solution. The usual head-tail dimer (HT1) appears not to be formed presumably because of the steric bulk of the tmeda ligand, which prevents a close approach of two tmeda ligands. There is also no evidence for formation of an alternative head-tail dimer structure (HT2) which, in principle, would not lead to any steric clash of ligands, but would require an orientation of the metal at N4 that is almost perpendicular to the nucleobase plane. In the Pd3 compound, the bridging metals are approximately in an anti arrangement, thereby leading to Pd...Pd separations within the Pd3 triangle close to 5.2 angstroms. This arrangement is reflected in the 1H NMR spectrum by a strongly deshielded H5 resonance of the nucleobase, occurring at 6.56 ppm (D2O). The overall structure of the Pd3 is that of a double cone, with ClO4- counterions approaching the cavities from either side. The trinuclear structure is also maintained in Me2SO-d6. In this solvent, Pd3 acts as a fluoride anion receptor, with F- binding to the N4H protons, as evident from large downfield shifts of these protons. The compound is compared with cyclic adeninato complexes of hexacoordinated metal ions, and a conceptional analogy with [12]metallacrown-3 species is outlined.