Structural models for the substrate-catalyst adduct in hydrodenitrogenation catalysis: Oxygen vs sulfur ligation

A comparison between models for the substrate-catalyst adduct in hydrodenitrogenation (HDN) catalysis is made with respect to oxygen vs sulfur ancillary ligands. Reacting [eta(2)(N,C)(NC5Bu3H2]Ta)-Bu-t(OAr)(2)Cl (1, Ar = 2,6-(C6H3Pr2)-Pr-i) With (KOBu)-Bu-t affords orange crystals of the alkoxide [eta(2)(N, C)-(NC5Bu3H2)-Bu-t]Ta(OAr)(2)((OBu)-Bu-t) (2), while 1 and (LiSBu)-Bu-t react to form the red thiolate analogue [eta(2)(N, C)-(NC5Bu3H2)-Bu-t]Ta(OAr)(2)((SBu)-Bu-t) (3). Structural studies of both complexes 2 and 3 are reported and compared with other eta 62(N,C)-(NC5Bu3H2)-Bu-t derivatives. A trace of the bromide complex [eta(2)(N, C)-(NC5Bu3H2)-Bu-t]Ta(OAr)(2)Br (4) is isolated from reacting [eta(2)(N, C)-(NC5Bu3H2)-Bu-t]Ta(OAr)(2)Cl (1) with EtMgBr in THF/Et2O solution and is also structurally characterized for comparison. Complexes 2-4 reveal a severe interruption of aromaticity within the heterocycle, different rotational preferences of the pyridine NC5 plane with respect to the Ta(OAr)(2)X moiety, and various aryloxide ligand structural differences. From this comparison, arguments will be presented that support the ancillary ligand cc-donor ability decreasing as (OBu)-Bu-t > OAr > (SBu)-Bu-t > Cl approximate to Br > Et, although evidence suggests that the (SBu)-Bu-t ligand is a better sigma + pi donor overall than OAr or (OBu)-Bu-t.