Density functional theory investigations of the homoleptic tris(dithiolene) complexes [M(dddt)(3)](-q) (q = 3, 2 ; M = Nd(3+) and U(3+/4+)) related to lanthanide(III)/actinide(III) differentiation.

The structures of the homoleptic lanthanide and actinide tris(dithiolene) complexes M(dddt)(3) (q = 3, M = Nd(3+) and q = 3 or 2, M = U(3+/4+)) have been investigated using relativistic Density Functional Theory (DFT) computations including spin-orbit corrections coupled with the COnductor-like Screening Model (COSMO) for a realistic solvation approach. The dithiolene ligands are known to be very efficient at stabilizing metal high oxidation states. The aim of the work is to explain the peculiar symmetric folding of the three Mdddt metallacycles in these complexes, some of them existing under a polymeric form, in relation with the Ln(III)/An(III) differentiation. In the M(dddt)(3)(py) species, where an additional pyridine ligand is linked to the metal center, the Mdddt moieties appear to be almost planar. The study brings to light the occurrence of a M...C=C interaction explaining the Mdddt folding of the U(dddt)(3) uranium species, the metal 5f electrons playing a driving role. No such interaction appears in the case of the Nd(III) complex, and the folding of the rather flexible dddt ligands in the polymeric structure of this species should be mainly due to steric effects. Moreover, the analysis of the normal modes of vibration shows that the U(III) complex U(dddt)(3), which has not yet been isolated, is thermodynamically stable. It appears that the X-ray characterized U(IV) complex U(dddt)(3) should be less stable than the calculated U(III) complex in a polar solvent.