Tailoring metal-carbon support interactions for the selective hydrogenation of multifunctional chemicals

No Thumbnail Available
Bateni, Hamed
Major Professor
Tessonnier, Jean-Philippe
Shanks, Brent H
Roling, Luke T
Rossini, Aaron J
Raman, Dave R
Committee Member
Journal Title
Journal ISSN
Volume Title
Research Projects
Organizational Units
Journal Issue
Is Version Of
Chemical and Biological Engineering
Biomass is an abundant, renewable, and inexpensive resource for the sustainable production of fuels and chemicals to replace those derived from non-renewable fossil fuels. In addition, biomass-derived feedstocks hold particular promise for producing novel compounds to overcome the petrochemical industry’s product stagnation and meet new market requirements. However, technological development has been identified as one of the critical obstacles to expanding the sustainable use of biomass resources while maximizing economic, social, and environmental outcomes. The highly functionalized nature of biobased species created an inevitable challenge for catalytic conversion of biomass-derived chemicals. In fact, conventional catalysts developed for the petrochemical industry are often inadequate for biomass processing due to the substantial differences in the nature of the feedstocks. Petroleum-derived species contain limited functionalities and are often catalytically processed at elevated temperatures in the gas phase. In contrast, chemical conversion of biobased chemicals is typically performed in the liquid phase under relatively mild operating conditions due to the low volatility and high reactivity of the multifunctional reactants. Therefore, it is crucial to develop finely tuned and robust catalysts tailored to these particular reaction conditions and the nature of the biobased molecules. Supported metal nanoparticle catalysts offer great flexibility. The typical practice to adjust the performance of these catalysts is often limited to engineering the composition, geometry, and crystallographic structure of the active sites of metals; however, the importance of support characteristics should not be overlooked. In fact, the support not only provides a scaffold for the dispersion of metal nanoparticles but can also alter the performance of metal nanoparticles through metal-support interactions (MSI). These interactions can induce geometry, chemistry, and electronic changes in the catalyst that consequently govern its behavior under reaction conditions. Significant efforts have been devoted to understanding and utilizing the MSI for the rational design of metal oxide-supported metal catalysts. However, the common metal oxide supports are often incompetent for liquid-phase reactions due to several limitations, including irreversible phase transition, leaching, and instability of the pore structure in the liquid phase. Carbon materials are promising supports for biomass conversion reactions in the condensed phase due to their excellent chemical and structural stability in a wide range of solvents, even in aggressive acidic or alkaline conditions. While carbon is generally regarded as an inert material, the differences in the activity per unit exposed area of the metal nanoparticles supported on carbon supports suggest that carbon is actually a non-innocent support. The carbonaceous materials usually consist of layers of fused aromatic rings (graphene layers) with varying degrees of imperfections, defects, and disorders within the structure. Oxygen-containing groups, e.g., carboxyl, hydroxyl, lactone, and phenyl groups, are the primary functionalities of the carbon materials, introduced either deliberately through an oxidation process or inadvertently through exposure to the ambient atmosphere. The change in the structural features and the surface chemistry of carbon materials results in a change in their physicochemical and electronic properties. Such a change in the electronic properties of semiconducting carbon materials can influence the charge transfer between the support and the overlaying metal nanoparticles through electronic metal support interactions (EMSI) that consequently impact the catalyst’s performance. The electronic-metal support interactions were originally discussed in Schwab’s works based on the solid-state physics of a metal-semiconductor junction. In fact, due to the difference in the Fermi level of the metal and the support, a contact potential is formed at their interface when the metal is deposited on the support. This contact potential serves as a driving force for the charge transfer at the interface until a uniform Fermi level is established across the junction. This interfacial charge transfer also leads to band bending in the semiconductor and the formation of Schottky barrier, representing the energy barrier for electron crossing through the interface. The large Schottky barrier for metal-insulator junction limits the charge transfer at their interface; however, the narrow bandgap of the semiconductor results in a smaller Schottky barrier where the “hot electrons” can transfer at the interface. Therefore, the amount of charge transfer not only depends on the properties of the metal particles (nature and size) but also is a strong function of the support’s nature and electronic properties, providing an additional degree of freedom for fine-tuning the performance of the catalysts. Carbon is uniquely placed to explore the potential of the EMSI in catalyst design, given the tunability of its electronic properties by adjusting its surface chemistry and structural features. Carbon materials were rationally synthesized to control their electronic properties to examine this hypothesis. Carbon-supported palladium catalysts were then carefully prepared with particular attention to minimizing the potential parasitic support effects enabling a better capture of the support’s direct electronic influence. Particular attention was devoted to understanding the electronic footprint of various oxygen functionalities, while nitrogen doping was also evaluated as a complementary approach to modify the carbon support. The performance of the catalysts was evaluated in liquid-phase hydrogenation of cinnamaldehyde, where a strong correlation between the electronic properties of the support and the catalyst selectivity was observed. As expected, the reduction in the work function of the support limited the formation of an electron-depleted region in the Pd particle (Pdδ+) due to the smaller energy gap between the Fermi-level of the metal and the support. Such a reduction in the interfacial charge transfer resulted in less electron perturbation of the Pd particles, restoring the intrinsic behavior of the Pd catalysts for C=C hydrogenation. This correlation was also examined in the case of the nitrogen-doped carbon support, where a linear correlation between the support’s work function and the catalyst’s performance was observed regardless of the dopant nature. These results take a step towards addressing the lack of knowledge about the rational strategy for designing carbon-supported metal catalysts. Additional studies are still needed to elucidate the interplay between the structure, electronic properties, and catalyst performance. Therefore, we also explored some potential candidate approaches for synthesizing carbon supports with tailored properties for next-generation studies. In particular, hydrothermally carbonized material and mesoporous carbons obtained from resorcinol-formaldehyde resin are identified as scalable versatile carbon supports enabling more rigorous evaluation of the carbon’s chemistry.
Subject Categories