Structure-Function Relationships of Metal Coordination Complexes for Non-Aqueous Redox Flow Batteries

Abstract

Non-aqueous redox flow batteries could enable the large-scale deployment of variable energy generation technologies, such as wind and solar, as well as decrease the required capacity and increase the efficiency of conventional power plants. Redox flow batteries use liquid electrolytes containing soluble active materials, which allows for the decoupling of power and energy, yielding a battery with a theoretically infinite cycle life. Commercially available redox flow batteries utilize aqueous chemistries, which are limited by the voltage window of water. To overcome this limitation, recent studies have employed non-aqueous electrolytes using redox active polymers, redox active organics, or metal coordination complexes (MCCs) as the active material. The studies described in this dissertation seek to develop structure-function relationships for MCCs; in particular functional group modifications were used to better understand the impact of structure and composition on standard potentials, solubility, and stability. For metal acetylacetonates (acacs), a class of MCCs utilizing redox “innocent” ligands, functional groups consisting of long chain esters were found to significantly increase their solubilities, in some cases achieving up to 1.8M in acetonitrile. Computational analysis suggested that the increased solubilities were primarily due to decreased solvation energies. For the same ligand class, different metals were shown to have a significant effect on the experimentally determined standard potentials, and density functional theory (DFT) calculations were shown to accurately predict these potentials as being a function of the highest occupied or lowest unoccupied molecular orbital energies (HOMO/LUMO). Importantly, the cycle life of individual half-reactions for a number of MCCs with promising characteristics were quantified using cyclic bulk electrolysis, and the cycle life was shown to correlate to calculated HOMO or LUMO metal density values. For the metal acacs, X-ray absorption spectroscopy was used on two representative MCCs, tracking the charge distribution and structural changes that occur during reduction and oxidation. Significantly different side reactions were observed for each complex, with ligand shedding occurring for vanadium(III) acac during reduction and vanadyl(IV) acac conversion followed by a solvent stabilizing reaction occurring during oxidation. For ruthenium(III) acac, minimal structural changes occurred during reduction, leading to a robust cycle life, while the electrochemical signature dramatically changed during oxidation producing an unknown species with a structure identical to ruthernium(III) acac within 3Å from the Ru atom. Finally, an MCC class with non-innocent ligands, salicylaldimine, was investigated. Ultimately, solubility and stability relationships were used to generate a chromium complex with a solubility of 1.14M which can be reversibly oxidized for ~200 cycles. Combined, these studies provide a foundation for the rational design of next generation active species for non-aqueous RFBs.