Computational Studies on Homogeneous Water Oxidation Catalysts
Abstract: To combat the increasing energy demand and climate change, artificial photosynthesis is a promising approach to producing renewables by storing energy into chemicals and fuels. Water oxidation, responsible for offering electrons and protons to the reduction reactions, suffers from slow kinetics. Consequently, developing highly efficient water oxidation catalysts (WOCs) reacting via desirable mechanisms plays an essential role in reaching the goal of artificial photosynthesis. Molecular WOCs in homogeneous catalysis could serve as models to understand the structure-property relationship and the reaction mechanisms owing to their well-defined structures and easily characterized properties. In Chapter 1, a variety of Ru- and Fe-based molecular WOCs are introduced. The two catalytic mechanisms for the O-O bond formation, water nucleophilic attack (WNA) and interaction of two metal-oxos (I2M), are described in Chapter 1.To understand the catalytic mechanisms for designing better WOCs, various theoretical techniques have been applied. Density functional theory (DFT) method was used to study the molecular properties of catalysts and the reaction energetics. Molecular dynamics (MD) and potential of mean force (PMF) were employed to investigate the behavior of catalysts and to calculate the free energy change along a specific coordinate in the explicit solvent. The theory for computing redox potentials, acid dissociation constants, reaction rate constants, and the descriptions of different solvation models are presented in Chapter 2.The family of Ru(bda)(py)2 (bda2- denotes 2,2'-bipyridine-6,6'-dicarboxylate, py denotes pyridine) complexes exhibits extremely high catalytic activity via the I2M mechanism under acidic conditions. Extensive studies have previously been conducted on the primary and secondary coordination environments. In Chapter 3, the first section focuses on three different modifications on the para-sites of the two pyridines. It is found that the complex with the longer hydrophobic group substituting the para-sites of two pyridines demonstrates the highest activity, which is attributed to the stronger binding energy between two RuV monomers. We conclude that the hydrophobic effect is dominating in enhancing the catalytic performance via the I2M mechanism. The second section of Chapter 3 studies the isolated crystal structure of the pseudo seven-coordinate RuIII-aqua intermediate obtained by connecting two meta-sites of pyridines with an ethylene glycol ether linker. DFT was used to study the formed H-bond network between the distal ligand and Ru(bda)(py)2. The influence of micro-solvation on the incoming aqua ligand was analyzed in the form of bonds and interactions. Incorporating a distal ligand could be an effective strategy to investigate the outer coordination environment effects. In the third section of Chapter 3, the linkers connecting two meta-sites of pyridines were changed and three more Ru(bda)(py)2-based catalysts with hydrophobic (aliphatic) and hydrophilic (ethylene glycol ether) linkers of different lengths were synthesized to study the outer coordination environment effects. The hydrophobic ligands lower the potentials slightly by stabilizing the key intermediates. The complex with the longer hydrophobic distal ligand demonstrates the highest TOF with the first-order kinetics. The I2M mechanism is suppressed owing to the limited flexibility of distal ligands validated by the nuclear magnetic resonance spectroscopy and the DFT-calculated energy differences between the conformations of distal ligands in the front and in the back of the bda ligand. The strategy of introducing hydrophobic outer coordination environment could be beneficial to design catalysts involving PCET reactions.Inspired by the Ru(bda)(py)2 catalyst and aiming to reduce the usage of Ru, a computational comparative study on Ru(bda)(py)2 and Fe(bda)(py)2 is presented in Chapter 4. Fe(bda)(py)2 was built by directly replacing Ru catalytic center of Ru(bda)(py)2 with Fe and maintaining the rest of the ligand system. The Fe-based complexes at different valence states prefer higher spin states while the Ru-based complexes are more stable at the lowest spin states. Unlike the Ru, the Fe center disfavors the 7-coordinate structure. Concerning the catalytic performance, the Fe(bda)(py)2 requires much higher potentials to reach the reactive FeV species than the Ru(bda)(py)2. The 6-coordinate [FeV(bda)(py)2=O]+ also has a higher energy barrier of the O-O bond formation via the I2M mechanism than the [RuV(bda)(py)2=O]+. We propose that directly substituting the Ru catalytic center with Fe fails to generate a viable catalyst and a significant ligand system modification is required.An alternative feasible catalytic mechanism is suggested for a highly active dinuclear Fe-based WOC in Chapter 5. The mechanism is proposed where two oxidation reactions are first required to reach the reactive state with calculated potentials matching the onset potential in the experiment. The reactive species is then decomposed by the nucleophilic chloride counter ion attack to produce two Fe-based monomers, which finally form the O-O bond by the radical coupling pathway. The calculated energy barriers and first-order kinetics match well with the experimental observations.
CLICK HERE TO DOWNLOAD THE WHOLE DISSERTATION. (in PDF format)