Committee Chair: Friederike Jenttoft
The use of biomass-derived lingo-cellulose as a possible alternative source of fuels and chemicals to fossil-based hydrocarbons has been an active area of research for both energy security and the search for more carbon neutral forms of energy. However, biomass offers many challenges based on processing and their high oxygen content. Various biomass upgrading routes, such as pyrolysis, fermentation, and hydrolysis, produce a variety of organic molecules with alcohol, carbonyl, and carboxylic functionality. The highly oxygenated compounds require catalytic upgrading before use as fuel or platform chemicals. One promising upgrading route is deoxydehydration, a reaction which combines a deoxygenation by a sacrificial reductant and dehydration in a single step to selectively convert vicinal diols into an olefin. This reaction has been demonstrated to be highly selective and effective at temperatures as low as 130 °C using homogeneous oxo-rhenium catalysts. The catalytic reaction involves the coordination of the diol with the metal center through a condensation reaction, reduction of the metal center by the sacrificial reductant, and finally the elimination of the alkene and re-oxidation of the catalytic metal center.
Oxo-rhenium complexes have proven to be very effective at deoxydehydration because they can easily undergo the necessary changes in coordination and oxidation state, however the high cost of rhenium and difficulty of homogeneous catalyst recovery makes these catalysts untenable for large scale biomass upgrading. This thesis details work in developing more robust and economical solid catalysts for deoxydehydration. A series of oxide-supported oxo-rhenium catalysts which demonstrate high activity for the deoxydehydration reaction, in some cases on par with benchmark homogeneous catalysts, and up to 95% selectivity. However, due to the liquid phase chemistry of the reaction, these catalysts demonstrated various propensities to deactivate over multiple runs due to the leaching of the active rhenium species. We undertook a study to determine which parameters control the leaching behavior. By understanding the effect of solvent, reactant, temperature, reductant, and conversion on leaching, the reaction conditions can be optimized to minimize leaching.
In an alternative strategy to mitigate the challenges associated with the high cost and rarity of rhenium, alternative redox active transition metals including vanadium, tungsten, molybdenum, and manganese were screened for deoxydehydration activity. The molybdenum catalysts were found to be the most promising, with the homogeneous ammonium heptamolybdate catalyst achieving 50% alkene yield and the oxide-supported molybdenum catalysts achieving 22% alkene yield using triphenylphosphine as the reductant. Interestingly, sodium molybdate was not found to be catalytically active under similar conditions. In the course of investigating the role of the ammonium and sodium counterion, it was discovered that ammonium-based reductants such as ammonium chloride, ammonium sulfate, and urea are active reductants for molybdenum-catalyzed deoxydehydration.