Identifying Performance-Defining Metrics of Ion Exchange Membranes for Non-Aqueous Redox Flow Batteries

Abstract

Redox flow batteries stand out as a promising candidate for large-scale and multi-hour energy storage due to their ability to independently scale power and energy, long cycle life, and facile manufacturing[1,2]. However, current deployment is hampered by elevated costs which motivate research into alternative electrolyte chemistries and advanced reactor concepts. As a result of the limited electrochemical stability window of water and the scarcity of inorganic redox active molecules, non-aqueous redox flow batteries (NAqRFBs) have emerged as a promising technology for low-cost energy storage if a number of technical challenges can be overcome. Leveraging the extended electrochemical window of organic electrolytes, NAqRFBs can enable high cell voltage and high energy density[3] and the use of abundant elements (e.g. H, C, N, O, S) coupled with molecular engineering enables almost unlimited tunability of the redox active molecules[4]. However, NAqRFBs suffer from large ohmic overpotentials due to the low ionic conductivity of the electrolytes[5] and the lack of tailored membrane materials for non-aqueous electrolytes. While existing knowledge on materials and characterization techniques in ion exchange membranes (IEMs) for NAqRFBs has been borrowed from their aqueous counterparts, there are some unique challenges (e.g. different ion size, solvent polarity[6]) that motivate research into new methodologies and polymer chemistries. In this work, we aim to elucidate material-property-performance relationships for IEMs relevant to NAqRFBs.Here, we employ an array of electrochemical and spectroscopic techniques to assess relevant membrane metrics for NAqRFBs. We deploy methods to determine membrane resistance, redox species crossover, ion exchange capacity, and electrochemical performance. To demonstrate this methodology we selected a series of commercial anion exchange membranes. We elect to study a model non-aqueous electrolyte system with redox active molecules based on ferrocene and phtalimide precursors chosen for their stability and facile kinetics, using TBAPF6 as supporting electrolyte in acetonitrile. In an effort to measure the real ion-exchange capacity using polyatomic fluorinated anions typically employed in NAqRFBs, we introduce quantitative fluorine nuclear magnetic resonance as analytical tool to track the loading of fluorinated anions in anion exchange membranes. Interestingly, we find that the loading of PF6 - is significantly lower than the ion exchange capacity measured in aqueous conditions, highlighting a rate-limited ion loading in non-aqueous electrolytes with bulky anions. Furthermore, we correlate these results with complementary metrics such as membrane resistance obtained with electrochemical impedance spectroscopy. Additionally, we incorporate a three-electrode setup equipped with a microelectrode within the electrolyte tanks to track in operando changes in concentrations. Linear sweep voltammetry reveals a reproducible and linear current-concentration response over a wide range of analyte concentrations, matching with the requirements of high-performance NAqRFBs. The integrated microelectrode sensor is suitable for an accurate measurement of state of charge or analyte crossover rate over a wide range of flow rates due to its positioning in the external tanks which display limited hydrodynamic turbulence. Our results demonstrate the need to develop novel methodologies to characterize membranes in non-aqueous electrolytes and informs the design of tailored membrane materials for next-generation NAqRFBs.