Date of Award

Summer 8-22-2019

Level of Access

Open-Access Thesis

Degree Name

Master of Science (MS)

Department

Chemical Engineering

Advisor

Thomas Schwartz

Second Committee Member

Brian Frederick

Third Committee Member

William DeSisto

Abstract

Converting biomass to alternative fuels has attracted significant interest in recent decades. Lignin, a principal component of biomass, is composed of phenolic monomers, which can be depolymerized using fast pyrolysis to yield a “bio-oil”. However, bio-oil is not immediately suitable as a biofuel because of its high oxygen content, and it is necessary to efficiently remove these oxygen atoms by hydrodeoxygenation (HDO). This project is focused on the elucidation of the reaction kinetics associated with carbon-oxygen hydrogenolysis in phenolic molecules, which is a significant reaction for the production of hydrocarbon fuels from biomass pyrolysis oils. The studied molecules are 5-hydroxymethylfurfural (HMF) and phenol, both used as model compounds. For phenol hydrodeoxygenation (HDO) the optimal pathway is direct deoxygenation (DDO), but at relevant temperatures, C-C double bond saturation is a significant side reaction, following the hydrogenation pathway (HYD). The importance of metal-TiO2 sites has been shown for a variety of reactions. Previous research in our group has shown that Ru/TiO2 is highly active for the conversion of phenol to benzene and that water could act as a co-catalyst. In this work, we clarify iv the role of water in C-O hydrogenolysis catalyzed by this material. Here, we designed and carried out a series of reaction kinetics experiments that illustrate the complex effect water has on the DDO mechanism. We measured reaction orders for phenol hydrogenolysis with respect to water and phenol over Ru supported on TiO2 rutile and anatase using a high-pressure liquid phase flow reactor operated in a kinetically-controlled regime. Our most interesting results show that the reaction is positive-order with respect to water for Ru/rutile and negative-order for Ru/anatase. These observations correlate with heats of water adsorption measurements on TiO2 indicating that anatase is more hydrophilic than rutile 1 and suggest that under particular circumstances, water molecules at the interfacial sites could become the most abundant surface intermediate (MASI). Also, the reaction is zero order with respect to phenol in the absence of water for Ru/anatase and it shifts from positive to negative order at higher phenol concentrations for Ru/rutile. Those differences with catalytic support identity suggest that the reaction mechanisms are different for each catalyst. For rutile, we believe that phenol is a MASI at two different sites: the interfacial site and the metal site that leads to the HYD pathway. For anatase, it is expected that phenol is only a MASI at one of those sites. HMF can also undergo hydrodeoxygenation in a series of intermediate reactions until it becomes 2,5-dimethylfuran. We perform a comparison between different Ni and Ru catalysts supported on Co3O4 with respect to literature performance. Our results were consistent with the literature, but we have obtained a different reaction intermediate, product distribution, and site time yields. We believe that those differences are due to difficulties in reproducing the catalysts by the co-precipitation method.

Share