Novel Catholyte and Electrolyte Optimisation for Aqueous Organic Redox Flow Batteries

Abstract

Redox flow batteries (RFBs) are a promising technology for grid-level energy storage. The ability to decouple energy and power, as well as the potential for low-cost and safe materials, make them particularly suited to this application. However, there is a lack of viable organic catholytes for RFBs and research thus far has primarily focussed on anolytes. Research in this thesis focuses on novel catholytes, degradation studies and electrolyte optimisation for aqueous organic redox flow batteries (AORFBs), the most challenging and yet the most promising area of this technology. In the first results chapter (Chapter 3), a series of triarylamines was synthesised. Initial electrochemical testing (using cyclic voltammetry) revealed one of these candidates, amino-functionalised 4-amino-trisphenyl amine, proved to be the most promising. However, battery cycling with this as the catholyte results in extensive polymerisation, leading to rapid capacity fade. This rapid capacity fade was improved by electrolyte optimisation, and utilising a mixed-salt system of 0.5 M HCl and 0.5 M H3PO4 it was possible to decrease capacity fade, increase coulombic efficiency and access more theoretical capacity. Chapter 4 explores commercially available phenothiazine dyes. Nicotinamide (NA) was used to increase solubility, specifically, the solubility of the most promising candidate explored, azure-a (AA), was doubled from 1 M to ca. 2 M. When cycled with NA in the supporting electrolyte, AA, had relatively stable cycling performance, though only half of the theoretical capacity was reached. Evidence suggests that this is the result of dimerisation of AA-based redox species. An extensive study using electrochemical impedance spectroscopy (EIS) showed that NA prevents a thick, charge-transfer blocking, film from forming on interphases in the cell (i.e. the electrode or membrane), thus improving the cycling performance. Chapter 5 further investigated both the dimerisation of AA-based species and their interaction with NA which leads to the observed improved performance. Through spectroscopic studies (NMR, EPR and UV/vis) it was found that there are at least four dimeric AA-based species in solution (most likely different dimer conformations). The role of pH on AA aggregation is also explored here for the first time. Finally, the origin of the improvement in battery performance using AA is shown to be preferential hydrogen-bonding with NA which intercepts AA aggregate, therefore reducing dimerisation and subsequent polymerisation. The final research chapter (Chapter 6) explores synthetic modification of phenothiazine, which is otherwise insoluble in aqueous conditions. A sulfonated propyl chain was found to improve solubility up to 1.15 M in 1 M HCl. However, upon cycling the sulfonate group was lost and an emulsion formed, leading to rapid capacity loss. This was improved by utilising NA as an additive (as shown previously in Chapters 4 and 5). Overall, this thesis has found that synthesising novel compounds presents many challenges, especially as the performance of candidate catholytes cannot be accurately predicted before experimental cycling. Ultimately the greatest improvements in cycling performance were achieved through electrolyte optimisation rather than synthetic changes in a particular catholyte family. It is therefore recommended to focus future research efforts on optimisation of the supporting electrolyte as the means for improving battery performance.