Scission of small multiply bonded molecules using trasition metal complexes. A DFT study.

The analysis described herein applies density functional theory to the activation and scission of the small multiply bonded molecules dinitrogen, carbon monoxide and carbon dioxide using transition metal catalysts. Starting from the Laplaza-Cummins 3-coordinate molybdenum amide complex MoL3 (L = N(tBu)Ar ) we have applied electronic structure methods in combination with the ONIOM approach to complete a comprehensive study of the effect of ligand bulk on theactivation of dinitrogen. Our results show that not only is there expected destabilisation of the intermediate on the pathway, due to direct steric interactions of the bulky groups, but also there is significant electronic destabilisation as the size of the ligand increases. This latter destabilisation is due to the inability of the molecule to accommodate a rotated amide group bound to the molybdenum once the amide reaches a certain size. Interestingly the Laplaza-Cummins catalyst is experimentally inactive towards carbon dioxide despite binding and cleaving one C-S bond in the similar CS2 molecule. We have used density functional theory (DFT) to show that, at first glance, the reaction of 3 L3Mo + CO2 should proceed smoothly to give L3Mo-O + L3Mo-CO-MoL3. However, initial coordination of the CO2 molecule to L3Mo does not take place because of the bending of CO2, the energy required to cross from the doublet to the quartet state, and the lower metal-CO2 binding energy compared to metal-CS2. From this analysis we predicted that replacement of the central metal with a d2 transition metal would provide improved binding. Our calculations in this regard suggest that the tantalum analogue, TaL3, will successfully bind to CO2 in a mononuclear ˜í‚àë2 arrangement and, importantly, will strongly activate one C-O bond to a point where spontaneously C-O cleave occurs. This strongly exothermic reaction takes into consideration formation transition barriers, spin crossings, ligand bulk and even the DFT functional choice. The product from this reaction, CO, is known to react with a similar 3-coordinate Ta(silox)3 (silox = OSi(tBu)3) complex, initially forming a ketenylidene (silox)3Ta-CCO, followed by a dicarbide (silox)3Ta-CC-(silox)3 structure. We again applied DFT methods to this reaction revealing an intricate mechanism whereby the previously unknown intermediate [(silox)3Ta-CO]2 was identified. The mechanism has been extended to consider the effect of altering both the metal species and the ligand environment. Specifically we predict that introducing electron-rich metals to the left of Ta on the periodic table to create mixed metal dinuclear intermediates shows great promise, as does the ligand environment of the Cummins-style 3-coordinate amide structure. Finally our interest in CO2 reactions lead to the exciting oxygen-atom transfer from carbon dioxide to a Fischer Carbene at (PNP)Ir reaction by the Grubbs group. We have confirmed the mechanism for the important CO2 reaction and have successfully rationalised the selective cleavage of the CS and CN bonds in OCS and PhNCO. The formation of the iridium-supported carbene itself has also been investigated and a fascinating autocatalytic mechanism has been discovered which nicely fits the observed experimental behaviour. This formation analysis has also been extended to consider the reactions with other linear and cyclic ethers that are known to form either carbenes or vinyl ether adducts. We have successfully rationalised the factors dictating reaction direction where both ether structural arrangement and (PNP) ligand environment contribute to the formation reaction outcomes.