Objective

In recent decades, much effort has been put forth to find methods to selectively oxidize hydrocarbons. Although it has been proved that synthetic metalloporphyrins can catalyze olefin epoxidation at mild conditions, simple metalloporphyrins are easily destroyed by autoxidation due to m-oxo dimer formation. In natural enzyme systems, the stability and selectivity of metalloporphyrins are improved by a surrounding protein framework that isolates the active site where the reactions are carried out and provides steric and electronic constraints on the reaction. Modifications of the porphyrin have been carried out in the literature to mimic this, but the chemical synthesis is difficult.

Recently, Nguyen, Hupp and coworkers have developed an alternate strategy using supramolecular chemistry to construct self-assembling systems that contain regular cavities for porphyrin encapsulation [1,2]. These structures, known as molecular squares (MS), have rhenium atoms at the corners and zinc porphyrin ligands on the sides forming cavities of ~18Å. By analogy to natural systems, MS were designed to improve the stability of porphyrin catalysts by immobilizing them in tailored environments. Porphyrins with nitrogen-containing ligands bind to the walls of the MS, as shown in Figure 1. In addition to enhancing the catalyst stability, MS can induce size-exclusion selectivity [3]. This work addresses two important concepts needed to understand these reaction systems and optimize their performance: the mechanism for the reaction and the overall rate law. To understand the kinetics of epoxidation reactions when manganese porphyrins are encapsulated within molecular squares, kinetic data were collected, and quantum mechanical calculations were used to investigate the individual elementary steps in the reaction mechanism.

----------------

Results and Discussion

The reaction studied was the epoxidation of styrene with iodosylbenzene as the oxidant. Before studying the encapsulated porphyrin, two different free porphyrin systems were analyzed: Mn tetraphenyl porphyrin (MnTPP) and Mn dipyridyl porphyrin (MnDPyP). Kinetic experiments were performed in batch reactors at constant temperature over wide concentration ranges for styrene (0.05 M-1 M) and catalyst (0.045 mM-1.1 mM). By monitoring reactant disappearance and product formation, initial reaction rates and catalyst deactivation were quantified. Results for free MnTPP indicate that the order of the reaction, assuming power law kinetics, for styrene increases from 0.4 to 1 as the concentration of the catalyst decreases. The order of the reaction for MnTPP when the concentration of the catalyst is higher than 0.2 mM is 0.8. However, at lower catalyst concentrations, the order of the reaction for MnTPP decreases from 3 to 1.5 as the concentration of styrene is increased. For MnDPyP, the orders of the reaction for the porphyrin and styrene were found to be 0.6 and 0.5, respectively.

The dependence of the reaction order for a given species on the concentrations of the other components suggests that a power law kinetics description is inadequate. A microkinetic description, in which elementary steps are explicitly considered, was therefore developed. Solution of the microkinetic model requires rate constants for each elementary step. Quantum mechanics calculations were carried out to obtain information about reaction energetics as a first step in quantifying rate constants. Density functional theory (DFT) and quantum/classical ONIOM calculations were performed in order to obtain optimized geometries for MnTPP and the intermediates along the proposed reaction path. For these calculations, a simplified porphyrin, in which the phenyl groups were replaced by methyl groups, was used. For the ONIOM calculations, two levels of theory were used. For the DFT calculations and for the high-level part of ONIOM, PW91 was used as the exchange-correlation functional, and the effective core potential, LANL2DZ, was employed as the basis set. The UFF force field was used for the low-level part of the ONIOM calculations. No symmetry constraints were applied in any of the calculations. In all results, the porphyrin had a saddle conformation, and bond lengths obtained were in very good agreement with the crystal structure of MnTPP [4]. Different spin states of the porphyrin and the intermediates were analyzed with both levels of theory. Figure 2 illustrates the energy changes for the different steps of the reaction pathway using the spin state for each configuration that had the lowest energy. State I is the reactants, II is the oxidized Mn-porphyrin, iodobenzene and the substrate, III is the substrate adsorbed to the oxidized Mn-porphyrin, and IV is the final products. The ONIOM and DFT results were in very good agreement.

References

1. R.V. Slone and J.T. Hupp, Inorg. Chem., 36 (1997) 5422.
2. P.H. Dinolfo and J.T. Hupp, Chem. Mater., 13 (2001) 3113.
3. M.L. Merlau, M.P. Mejia, S.T. Nguyen and J.T. Hupp, Angew. Chem. Int. Edit., 40 (2001) 4239.
4. A. Tulinsky and B.M.L Chen, J. Am. Chem. Soc., 99 (1977) 3647.

 

[Polymer Recovery] | [Catalysis] | [Mechanistic Modeling] | [Novel Biochemical Transformations]