Loading...
Towards understanding the catalysis of various complex systems---a theoretical study
Zhang, Kaihua
Zhang, Kaihua
Citations
Altmetric:
Abstract
Quantum mechanical simulations are powerful tools for studying chemical reactivity and modeling the properties of molecular and solid-state catalysts. Density functional
theory (DFT) is extensively used in the investigation of catalysis due to its good
balance between chemical accuracy and computational cost.
The work in this thesis consists of the theoretical investigation of mechanistic
details for different complex systems, including heterogeneous catalysis (metal-doped
silica), homogeneous reactions (metal-ligand molecular catalysts), and
molecular/heterogeneous catalyzed reactions. The present study also includes the
testing of a hybrid methodology to simulate large complex systems more effectively.
Metal-doped amorphous silicates are promising materials for heterogeneous catalysis due to their large tunable surface area and easy fabrication. However, the amorphous nature of the surface makes the characterization difficult. Thus, improvements rely on a trial-and-error approach. DFT simulations can, in principle, provide direct structure-property relations, but quantitative studies are challenging as brute force sampling would require simulations on tens of thousands of sites. We use the Nb-catalyzed epoxidation of ethylene as a test reaction to analyze various aspects of the modeling that need to be considered for simulations of effective reaction rates. We show that each site can host a variety of transition state structures that represent the same reaction event, but that can differ considerably in reaction barrier. We then use machine learning to identify the most important descriptors of the bare active site that correlate directly with the energy barrier. Although the present test set is too small for quantitative reaction rate predictions, we discuss the critical features of a very active site that can drive the kinetics in the actual material.
In terms of homogeneous catalysis, we have investigated the homogeneous catalysts Polypyridyl dicarboxylates with Ru(II)/uranyl(VI) ($UO_2^{2+}$) complexes as part of a collaboration with the Blakemore group. The redox properties of Ru(II) and uranyl(VI) complexes of $2,2'-bipyridyl-6,6'-dicarboxylate$ (\textbf{bdc}), $2,2':6',2"-terpyridyl-6,6"-dicarboxylate$ (\textbf{tdc}), $4'-phenyl-2,2':6',2"-terpyridyl-6,6"-dicarboxylate$ (\textbf{Phtdc}) have been investigated, revealing that these ligands can enable both ligand- and metal-centered electrochemical reduction.
Our simulations suggest the electron density mainly resides in the ligand for the reduced Ru(II)\textbf{pdc} complex. While for the Uranyl complex, the electron reduction is both ligand- and metal- centered. Therefore, \textbf{pdc} ligands are non-innocent and valuable for new schemes for reductive activation of challenging metal-containing species.
For molecular catalyst on the solid support, the $chloro-(2,2':6',2"-terpyridine-4'-carboxylic acid)$ palladium(II)chloride onto the amorphous silica catalyst is studied in collaboration with the VANNUCCI research group at the University of South Carolina. The heterogeneously catalyzed hydrodeoxygenation of oxygenated aromatics is highly active and selective $(>99\%)$. The reaction is under mild conditions and in non-polar solvent dodecane. The molecular/heterogeneous catalyst is formed by anchoring the molecular complex bound to the amorphous silica support. Moreover, separating the catalyst is much easier than in the corresponding homogeneous catalysis. However, experimental characterization of the binding mechanism is limited. Therefore, we studied the binding mechanism with density functional theory to provide atomistic and mechanistic insights into the binding and the orientation of the complex on the amorphous surface. Multiple binding sites that feature different silanol and siloxane configurations on the surface are considered to account for the disordered nature of the amorphous silica surface. Our simulations indicate hydrogen bonding is often favored over covalent bonding unless the silanol or siloxane groups have strained bonds. Additionally, the relative orientation of the complex on the surface is strongly influenced by the local hydrogen bonding between the carboxylate group and the surrounding local silanols of the binding site. These computational results may provide essential information for the design of catalysts with stronger binding on amorphous solid support.
The development of more efficient methods is of paramount importance in order to treat large molecular systems effectively. We tested a point charge embedding method for excited states that extrapolate entire excitation bands to simulate UV-Vis spectra. The embedding charges are computed to reproduce the electronic potential of the entire chromophore at the low level of theory, with proper constraints to avoid overpolarization issues at the boundary between layers. We proved the validity of the approach and its superiority to the simulation without embedding. This methodology can now be applied to the large cluster models necessary for heterogeneous catalysis simulations.
Metal-doped amorphous silicates are promising materials for heterogeneous catalysis due to their large tunable surface area and easy fabrication. However, the amorphous nature of the surface makes the characterization difficult. Thus, improvements rely on a trial-and-error approach. DFT simulations can, in principle, provide direct structure-property relations, but quantitative studies are challenging as brute force sampling would require simulations on tens of thousands of sites. We use the Nb-catalyzed epoxidation of ethylene as a test reaction to analyze various aspects of the modeling that need to be considered for simulations of effective reaction rates. We show that each site can host a variety of transition state structures that represent the same reaction event, but that can differ considerably in reaction barrier. We then use machine learning to identify the most important descriptors of the bare active site that correlate directly with the energy barrier. Although the present test set is too small for quantitative reaction rate predictions, we discuss the critical features of a very active site that can drive the kinetics in the actual material.
In terms of homogeneous catalysis, we have investigated the homogeneous catalysts Polypyridyl dicarboxylates with Ru(II)/uranyl(VI) ($UO_2^{2+}$) complexes as part of a collaboration with the Blakemore group. The redox properties of Ru(II) and uranyl(VI) complexes of $2,2'-bipyridyl-6,6'-dicarboxylate$ (\textbf{bdc}), $2,2':6',2"-terpyridyl-6,6"-dicarboxylate$ (\textbf{tdc}), $4'-phenyl-2,2':6',2"-terpyridyl-6,6"-dicarboxylate$ (\textbf{Phtdc}) have been investigated, revealing that these ligands can enable both ligand- and metal-centered electrochemical reduction.
Our simulations suggest the electron density mainly resides in the ligand for the reduced Ru(II)\textbf{pdc} complex. While for the Uranyl complex, the electron reduction is both ligand- and metal- centered. Therefore, \textbf{pdc} ligands are non-innocent and valuable for new schemes for reductive activation of challenging metal-containing species.
For molecular catalyst on the solid support, the $chloro-(2,2':6',2"-terpyridine-4'-carboxylic acid)$ palladium(II)chloride onto the amorphous silica catalyst is studied in collaboration with the VANNUCCI research group at the University of South Carolina. The heterogeneously catalyzed hydrodeoxygenation of oxygenated aromatics is highly active and selective $(>99\%)$. The reaction is under mild conditions and in non-polar solvent dodecane. The molecular/heterogeneous catalyst is formed by anchoring the molecular complex bound to the amorphous silica support. Moreover, separating the catalyst is much easier than in the corresponding homogeneous catalysis. However, experimental characterization of the binding mechanism is limited. Therefore, we studied the binding mechanism with density functional theory to provide atomistic and mechanistic insights into the binding and the orientation of the complex on the amorphous surface. Multiple binding sites that feature different silanol and siloxane configurations on the surface are considered to account for the disordered nature of the amorphous silica surface. Our simulations indicate hydrogen bonding is often favored over covalent bonding unless the silanol or siloxane groups have strained bonds. Additionally, the relative orientation of the complex on the surface is strongly influenced by the local hydrogen bonding between the carboxylate group and the surrounding local silanols of the binding site. These computational results may provide essential information for the design of catalysts with stronger binding on amorphous solid support.
The development of more efficient methods is of paramount importance in order to treat large molecular systems effectively. We tested a point charge embedding method for excited states that extrapolate entire excitation bands to simulate UV-Vis spectra. The embedding charges are computed to reproduce the electronic potential of the entire chromophore at the low level of theory, with proper constraints to avoid overpolarization issues at the boundary between layers. We proved the validity of the approach and its superiority to the simulation without embedding. This methodology can now be applied to the large cluster models necessary for heterogeneous catalysis simulations.
Description
Date
2022-01-01
Journal Title
Journal ISSN
Volume Title
Publisher
University of Kansas
Collections
Archive Status
This item contains archived web content.
Files
Zhang_ku_0099D_18733.pdf
Adobe PDF, 14.25 MB
- Embargoed until 2172-05-31
Research Projects
Organizational Units
Journal Issue
Keywords
Amorphous silica, catalysis, non-innocent ligand, ONIOM-EE