Ab Initio Studies on Ruthenium-Based Olefin Metathesis Catalysts and Insights into Design of Improved Catalysts

Presenter: Rick Muller1
Collaborators: Michael T. Feldmann1
Sheng Ding,1,2 Robert H. Grubbs,2
and William A. Goddard III1*

1Materials and Process Simulation Center, Beckman Institute (139-74),
2The Arnold and Mabel Beckman Laboratory of Chemical Synthesis,
1,2Division of Chemistry and Chemical Engineering,
California Institute of Technology, Pasadena, CA 91125

Abstract:

We report ab initio quantum mechanical (QM) calculations [using DFT at the B3LYP/LACVP** level] to investigate olefin metathesis by homogeneous ruthenium-based catalysts involving coordinated phosphines (Lp) and imidazole-2-ylidene (IMes)-like carbenes (Lc). We find that mechanism of the Ru(Cl)2(Lp)(Lc)(CH2) catalyst involves (1) activation by eliminating Lp, (2) binding of olefin substrate at the Lp site trans to Lc, and (3) formation of the metallacyclobutane. The activation step is endothermic by D H300K ~18 kcal/mol, while the subsequent catalytic steps of olefin binding and formation of the metallacyclobutane are exothermic by D H300K ~10 kcal/mol and 6 kcal/mol, respectively, with no energy barriers.

We find that the activation energy depends critically upon both the s basicity and p acidity properties of the Lc and Lp ligands. Based on these results we developed a rapid screening computational model to predict the activity for new choices of Lc and Lp (without the need for QM calculations on the catalyst). We tested this model against experimental activity for several ligands and then used it to predict the activities of several Lc and Lp not previously studied experimentally. This leads to new Lc and Lp predicted to be even more active than the current catalysts. Thus, this conputational model for catalyst design and modification allows rapid screening to find ligands likely to achieve improved catalyst performance.