UC Berkeley

Low molecular weight alkenes comprise a significant fraction of the effluent from fluid catalytic cracking units and of the C2-C5 hydrocarbons produced from Fischer Tropsch synthesis. Yet these products are too volatile to be blended into transportation fuel. Therefore, the aim of this research was to investigate pathways by which light alkenes could be converted to molecules with favorable gasoline or diesel characteristics. This research is relevant because such pathways would allow for the production of fuels from non-petroleum feedstocks such as biomass, coal, and natural gas.

Previous work has identified butanol and 2-ethylhexanol as suitable blending components or replacements for gasoline and diesel, respectively. These molecules can be synthesized via propene hydroformylation to yield butanal which can then be hydrogenated to butanol or can undergo self-condensation followed by hydrogenation to form 2-ethylhexanol. Although many of these reactions are catalyzed effectively by homogeneous catalysts, our research has focused on the development, characterization, and kinetics of heterogeneous, industrially implantable catalysts for these reactions.

Hydroformylation processes are commonly plagued by the separation of the organic products from the homogeneous Rh complexes. Supported Ionic Liquid Phase (SILP) catalysts that immobilize a Rh complex on a support can address this concern; however, little was known about the effects of catalyst preparation on the activity and stability of such catalyst. Our studies have shown that the activity and stability are strongly influenced by the ligand-to-rhodium ratio and the surface density of silanol groups on the silica support. In situ spectroscopic studies suggest that HRh(CO)2SX (SX, sulphoxantphos) complexes are bound to the support by interactions of the sulfonate groups of SX with silanol groups from the silica support. The function of the ionic liquid is to prevent the formation of catalytically inactive [Rh(CO)(μ-CO)SX]2 or HRh(SX)2 species by enhancing the dispersion of the active monomers. The realization that the support interaction was necessary to maintain high dispersion led to the development of ionic liquid-free hydroformylation catalysts (X-Rh/SiO2, X = xantphos). Reaction kinetics for hydroformylation were investigated and found to be dependent on reaction conditions.

To achieve the tandem hydroformylation-hydrogenation of propene to butanol in a single reactor, a heterogeneous catalyst is needed for the selective hydrogenation of aldehydes in the presence of CO and alkene. An organometallic complex, Shvo's catalyst, was dispersed onto SiO2. This catalyst meets the necessary criteria because it operates through an outer sphere mechanism, thereby permitting the hydrogenation of butanal but not propene. Reaction kinetics of butanal hydrogenation were investigated and then used to predict the butanol yield in a tandem process that converts a mixture of propene and synthesis gas into butanol.

The aldol condensation of n-butanal was investigated over a solid-base organocatalyst and Ti-SBA-15 catalysts. The bifunctional nature of these catalysts is necessary to promote several of the elementary steps of the reaction mechanism. The catalyst was made by grafting site-isolated amines on tailored silica surfaces to allow for a co-operative interaction between basic amine and the weakly acidic silanols. Secondary amine functionalized silica was found to be nearly five times more active than primary amines, whereas grafted tertiary amines exhibited negligible catalytic activity. For secondary amines grafted on silica, it was found that the surface silanol groups serve as adsorption sites for the aldehydes and as a Brønsted acids to activate the carbonyl group of the aldehyde to form enamines. The bifunctional nature of this catalyst was confirmed by experimental evidence and theoretical calculations which demonstrate the need for spatial separation of the weakly acidic silanol and the basic amine groups in order to achieve high catalytic activity.

To understand how the coordination and connectivity of Ti affect the kinetics of aldol condensation, catalysts were made by either incorporation or impregnation of Ti precursors into or onto the SBA surface. Irrespective of the catalyst synthesis procedure, site isolated tetrahedral Ti sites were found to be more active than catalysts containing hexacoordinated Ti oligomers. Furthermore, it was observed that isolated tetrahedral titanols were more active than tetrahedral dimers. Theoretical calculations demonstrated that Ti dimers are less active due to steric constraints.