The zinc-bromine hybrid redox flow battery (RFB) is one of the few flow battery systems that have been commercially implemented for medium-to-large scale energy storage applications. One of the issues identified with this flow battery is bromine crossover from the positive electrolyte to the zinc side, leading to a loss of energy efficiency. This problem can be overcome through the use of inorganic-based complexing agents that bind the bromine in an immiscible liquid phase which is pumped from the electrode surface and stored in an external reservoir. This bromine-rich phase is recirculated back to the positive electrode during discharge of the battery.

Currently, N-methyl-N-ethylpyrrolidinium bromide (MEP) is the choice complexing agent for industrial batteries. However, interest lies in developing a complexing agent that will bind with the electrogenerated bromine but remain in the aqueous phase of the solution. This would improve the charged materials ability to disperse throughout the solution and reduce the complexity of the pumping requirements currently used. A number of novel additives (based on specific quaternary ammonium salts, QBr) have been synthesized and characterized by focusing on both their impact on the Br₂/Br⁻ electrochemical kinetics and on their physical properties.

Initial observations indicate that the amount of the immiscible liquid phase had been drastically reduced with the novel additives. Electrochemical techniques (including AC impedance spectroscopy and potentiodynamic polarisation) show that MEP and QBr have similar R_CT values with a depressed semi circular profile (Figure 1). However, cyclic voltammetry shows a reduced reversibility of the Br₂/Br⁻ reaction with QBr present compared to MEP. The data from Tafel extrapolation show a 10-fold reduction in the exchange current density (j₀) between QBr and MEP. However, for the various solutions containing the QBr, MEP and no additive, the same Tafel slope was evaluated (B_a ≈ 88 mV) as well as the same equilibrium potential (E₀ ≈ 0.88 V vs. SCE). This difference in the electrochemical kinetics between no additive/QBr and MEP from the CV is arguably due to the fact that the MEP can capture the electrogenerated bromine at the electrode surface better than QBr. These uncertainties led to the testing of the physical properties to determine how well the bromine is binding with the additives in question.

The physical properties were found to be quite similar for all the additives (including MEP). The Br₂ vapour pressure measured using an isoteniscope resulted in similar enthalpies of vaporization (ΔH ≈ 30 kJ mol⁻¹) and the quantity of Br₂ remaining within the aqueous phase from iodometric titrations. Therefore, it can be assumed that the complex additives used bind similar concentrations of Br₂ in the immiscible phase. This ability to capture (and release) larger amounts of bromine, was investigated using Raman spectroscopy and the presence of Br₃⁻ and Br₅⁻ species could be identified. By normalizing to the peak assigned to Br₃⁻ the effectiveness of the novel additives can be compared to MEP based on the Br₅⁻ content (Figure 2). The concentration ratios have an apparent effect on the intensity of the complexed peaks. However, this also shows that the new QBr salt has a similar effect on the physical properties of the electrolyte as MEP.

From this series of synthesised QBr compounds, the objective to reduce the immiscible liquid phase has been achieved. This is evident from both visual inspections and the effect that QBr had on the physical properties. MEP still proves to be the better additive in terms of electrochemical kinetics. This serves as a strong starting point for finding a potential candidate that is as effective as MEP with the absence of the immiscible liquid.