(530b) Influence of Oxidant Chemical Functionality on Alkene Epoxidation over Lewis Acid Zeolites: Intermediate Stabilization through Inner- and Outer-Sphere Interactions

Atomically-disperse groups 4 and 5 catalysts activate H₂O₂ and alkyl hydroperoxides (i.e., t-butyl hydrogen peroxide (TBHP) and cumene hydroperoxide (CHP)) to form hydro- or alkyl-peroxide species (M-OOR; R = H, t-Bu, cumyl) that are active for alkene epoxidation. These M-OOR species interact with proximate solvent structures and bound alkenes at transition states which greatly influences catalysis; however, the molecular details that describe these interactions and the ensuing differences in catalytic phenomena are not discussed in the open literature. These M-OOR species and the corresponding epoxidation transition states may be stabilized through “inner-sphere” interactions, where dispersive forces between the -R group and the alkene lower the free energy of the transition state complex. Alternatively, the pendant moiety on the reactive intermediates (i.e., H, t-Bu, or cumyl) may (de)stabilize through participating in or disrupting molecular interactions (e.g., hydrogen bonding) with the solvent through “outer-sphere” interactions. Here, we attempt to unravel these inner- and outer-sphere interactions by studying cyclohexene epoxidation over Ti- and Nb-atoms substituted into zeolite BEA and grafted onto SBA-15 and utilize combination of kinetic, thermodynamic, and spectroscopic methods to understand how the structure of the hydroperoxide oxidant affects the stability of surface species present during the epoxidation catalytic cycle.

The rates of cyclohexene (C₆H₁₀) epoxidation are highly dependent on the combination of oxidant (i.e., H₂O₂, TBHP, CHP), active site identity (e.g., Ti, Nb), solvent (e.g., CH₃CN, CH₃OH), and silicate support (e.g., BEA, SBA-15). These differences, however, are not attributed to differences in the general mechanism for epoxidation, which appears to be indistinguishable among catalysts, solvents, supports, and oxidants. For example, within Ti-BEA with CH₃CN, rates of C₆H₁₀ epoxidation are proportional to the concentration of C₆H₁₀ and independent of oxidant concentration at low C₆H₁₀-to-oxidant molar ratios (<1 (mol C₆H₁₀)(mol oxidant)^-1), which suggests that active sites are saturated with Ti-OOR intermediates. Turnover rates become independent of C₆H₁₀ concentration and increase linearly with oxidant concentration when the ratio of C₆H₁₀-to-oxidant is large (>1 (mol C₆H₁₀)(mol oxidant)^-1), suggesting that the most abundant surface intermediates are derived from C₆H₁₀. These observations are consistent with a mechanism where the oxidant reversibly adsorbs to Ti active sites and is irreversibly activated to form Ti-OOR intermediates, which then react with fluid-phase C₆H₁₀ through to form the corresponding epoxide. Despite the similarity in the mechanism for epoxidation, turnover rates using TBHP and CHP as the oxidant are 10- and 20-fold lower than when H₂O₂ is used. Comparisons of apparent activation enthalpies (ΔH_E,App^‡) and entropies (ΔS_E,App^‡) show similar enthalpic stability (41 – 50 kJ mol^-1) for the formation of the epoxidation transition states by coordination of C₆H₁₀ with the vicinal O-atom of the Ti-OOR intermediate. The large disparity in rates result from differences in ΔS_E,App^‡ as epoxidations that proceed through Ti-OO-cumyl intermediates lead to the greatest entropic losses (-150 versus -50 J mol^-1K^-1 for CHP and H₂O₂, respectively). These inner-sphere interactions between the -R­ group and the cyclohexyl ring of the C₆H₁₀ epoxidation transition state results in the variable entropic losses as the identity of the oxidant is varied. On-going work seeks to further probe inner-sphere interactions between Ti-OOR intermediates and the alkyl group of the transition states while also investigating the role of outer-sphere interactions through changes in the solvent.