| Summary: | Extending the classes of reactions that underlie electrochemical energy storage systems is of fundamental and practical importance to improving mobility, autonomy, medical devices and electronics. Most of the cathode materials developed so far are oxide-based: all commercial Li-ion cathodes utilize lithium transition metal oxides, while MnO₂, SOCl₂, and SO₂ are representative examples for Li primary cathodes. In contrast, fluoride-based cathodes are generally less investigated, with only several instances, but all with exceedingly high theoretical energy densities, such as carbon-monofluoride (CFₓ), transition metal fluorides, and the recently developed perfluorinated gas cathodes (SF₆ and NF₃). This indicates the strong potential for fluoride-based cathodes to surpass the current energy density limit. Therefore, to expand the landscape of fluoride redox to provide a new degree of freedom for the design of high-energy cathodes, this thesis examines the controlling parameters for fluoride bond redox activities, and their implications for Li and Li-ion batteries.
The first part of this thesis targets sulfur−fluorine (S−F) bonds. Using Li−SF₆ battery as a platform, the dominating effect of the electrolyte solvent properties on lithium fluoride (LiF, one of the discharge products) nucleation and growth was demonstrated. The electrode passivation induced by LiF is mitigated via increasing the fluoride solvation strength of the solvents, resulting in improved Li−SF₆ cell rate capabilities. Strategies to tune the S−F bond redox activity at molecular structure level was investigated next using liquid phase pentafluorosulfanyl arenes (RPh-SF₅), where one of the F-ligands in the SF₆ molecule is replaced by an aromatic group (R-Ph). The ring structure facilitates electron transfer by increasing molecular polarity, while R functionality alters the S−F bond reduction potential by changing the electron distribution around the –SF₅ group. As a new family of Li primary catholytes, the R-Ph-SF₅ reactants allow for full reactant defluorination and a total of up to 8 e− transfer per molecule, yielding capacities of 861 mAh/g subscript reactant and voltages up to ~2.9 V vs. Li/Li⁺. At a cell level, gravimetric energies of 1085 Wh/kg were attained at 50 ºC, exceeding all leading primary batteries based on electrode + 3 electrolyte (sub-stack) mass. Voltage compatibility of R-Ph-SF₅ and CFₓ solid cathodes further enabled design of a hybrid battery containing both fluorinated catholyte and cathode. The hybrid cells reach extraordinarily high cell active mass loading (~80%) and allow for significant boosting of sub-stack gravimetric energy of Li−CFₓ cells by at least 20%.
The carbon−fluorine (C−F) bonds in perfluoroalkyl group (R subscript F) were investigated next. The effect of extrinsic factors was examined using liquid perfluoroalkyl iodides (CFIs) as an example system, where the polarizable iodine supports electrochemical reduction with concerted F⁻ ligand expulsion. C−F bond redox activity was found to be influenced significantly by multiple parameters, including reactant concentration, discharge rate, temperature, and solvent properties (e.g. catholyte viscosity). A maximum of 8 e⁻/C₆F₁₃I, or 8/13 available F, is accessible, but only at ideal conditions (low reactant concentration and rate). Increasing concentration or rate exacerbates premature cell termination caused by deactivation of intermediates, resulting in <2 e⁻/C₆F₁₃I. This challenge was addressed via molecular design. By replacing the I-ligand with an alkene linker connected to a conjugated system, close-to-full defluorination of RF was achieved, yielding up to 15 e⁻ per molecule (or 15/17 available F), at voltages up to 2.6 V vs. Li/Li⁺. In addition to the ring structure and the R substitutional group, which facilitate charge transfer as that observed in R-Ph-SF₅, the alkene linker was found to be essential here for the reduction transformation propagating along the RF tail.
Lastly, Mn-F bond was studied in the context of electrochemical fluoridation of MnO, the product of which functions as rechargeable Li-ion cathodes. Previous studies showed that small MnO particle size (<10 nm) is necessary for MnO fluoridation via LiF splitting reaction. We demonstrated that such limitation is originated from LiF instead of MnO. With electrochemically formed LiF, which is nano-crystallized and in intimate contact with MnO, high MnO utilization (~0.9 e⁻/MnO) is achievable even with large MnO particle size (~400 nm).
Overall, the central advance of this thesis is the identification of multiple new electrochemical conversion motifs. In addition to the development of novel classes of Li primary cathodes with extraordinary electrochemical performances, this work also constructs a map of handles to tune the fluoride bond activities, providing a new platform for the design of battery materials for different applications. For instance, in rechargeable batteries, LiF is suggested to play an important role in stabilizing reactive interfaces and improving cycling stability.
|
|---|