Date of Award

Spring 1-1-2014

Document Type


Degree Name

Doctor of Philosophy (PhD)


Chemical & Biochemical Engineering

First Advisor

J. Will Medlin

Second Advisor

Daniel K. Schwartz

Third Advisor

Charles B. Musgrave

Fourth Advisor

Garry Rumbles

Fifth Advisor

Joel L. Kaar


Elucidating the site requirements for reactions over heterogeneous catalysts can be challenging since, unlike their homogenous counterparts, the surfaces of these materials are comprised of a variety of diverse features where reactions can occur. Because there is strong evidence suggesting that reactions will occur at a specific type (or group) of active site(s), significant effort has been put forth in the field to try to determine the site requirements for particular reactions. Here, we report a new approach for controlling the availability of specific active sites on supported Pd/Al2O3 catalysts using alkanethiolate self-assembled monolayers (SAMs). SAMs form consistent and ordered structures on metal surfaces; these features were exploited to form monolayers that preferentially block or expose specific types of sites on catalyst particles.

The structure of the tail moiety played an important role in the overall organization of the resulting monolayer and was tuned to control the types of active sites that were exposed on metal particle surfaces. Infrared spectroscopy using CO as a probe molecule was used to characterize site availability on the catalysts. A monolayer formed from a bulky caged molecule like 1-adamanetanethiol (AT) restricted the availability of contiguous active sites on particle terraces with respect to the uncoated catalyst. Increasing the density of the monolayer using a linear alkanethiol like 1-octadecanethiol (C18) further restricted adsorption on terraces. For both of these monolayers, adsorption at particle edges and steps was nearly unaffected but could be blocked by forming a high coverage monolayer from 1,2-benzene dithiol (BDT).

The relationship between monolayer spacing and reactivity dramatically depended on the structure of the reactant, even for small species. For example, employing a C18 monolayer reduced the rate of epoxybutane hydrogenation by nearly 3 orders of magnitude with respect to the unmodified catalyst, whereas hydrogenation of propionaldehyde was only reduced by a single order of magnitude. These results suggested 1) that site requirements for conversion of these two species were substantially different and 2) this modification approach could potentially be utilized to direct selectivity of more complex reactants where competing reaction pathways have different site requirements.

To demonstrate the feasibility of the latter hypothesis, this approach was utilized for the hydrogenation of furfural. On palladium catalysts, furfural tends to decarbonylate to produce furan and carbon monoxide. A competing process involves hydrogenation of the aldehyde to produce furfuryl alcohol and subsequent hydrodeoxygenation producing methylfuran. Site requirements for each pathway were investigated by comparing reactivity on uncoated, C18-coated and BDT-coated catalysts. Comparison between reaction rates and site availability suggested that decarbonylation occurred primarily on terrace sites while hydrodeoxygenation occurred on particle steps and edges. Aldehyde hydrogenation, and its reverse process of alcohol dehydrogenation, was found to occur on both terrace or edge sites, with the dominant pathway dependent on surface coverage as determined by reaction conditions.