The redox flow battery (RFB) has seen an explosion in research activity in recent years due to its attractiveness for large-scale energy storage applications (e.g. grid level storage). RFBs^1,2 offer many advantages over conventional batteries including their relatively high reliability and long cycle life, as well as the ability to scale their energy and power outputs independently. The vanadium flow battery (VFB)^1,3-6 is currently the most promising of these technologies, with a number of large-scale applications and demonstration projects already underway.⁶ The power density of VFBs is limited primarily by the kinetics of the redox reactions at the carbon electrodes while the energy density is limited primarily by the concentration of vanadium in the electrolyte. In this talk we will review our work on two aspects of VFBs – kinetics on carbon electrodes and thermal stability of electrolytes – which impact these two limitations respectively.

We have investigated^3,7,8 the kinetics of the V^II-V^III, V^III-V^IV and V^IV-V^V redox couples for a range of different carbon materials using a variety of techniques, and it is clear that the kinetic rates depend strongly both on the type of carbon used and on the preparation of the electrode surface. We will discuss our investigations of the effects of anodic and cathodic pretreatments of carbon on electrode kinetics in the VFB. Using cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS), we showed for four different types of carbon that electrode treatments at negative potentials enhance the kinetics of V^IV-V^V and inhibit the kinetics of V^II-V^III while electrode treatments at positive potentials inhibit the kinetics of V^IV-V^V and enhance the kinetics of V^II-V^III. These observations may explain conflicting reports in the literature. We examined in detail the potentials required for activation and deactivation of electrodes. The results suggest that interchanging the positive and negative electrodes in a vanadium flow battery (VFB) would reduce the overpotential at the negative electrode and so improve the performance. This is supported by flow-cell experiments.³ Thus, periodic catholyte-anolyte interchange, or equivalent alternatives such as battery overdischarge, show promise of improving the voltage efficiency of VFBs.

We have developed^9-11 a standard methodology for measuring the thermal stability of vanadium catholytes based on their induction time for precipitation of V₂O₃. Using this, we have investigated the thermal stability of typical vanadium flow battery (VFB) catholytes at temperatures in the range 30–70°C for V^V concentrations of 1.4–2.2 mol dm^-3 and sulfate concentrations of 3.6–5.4 mol dm^-3. In all cases, V₂O₅ precipitates after an induction time, which decreases with increasing temperature. Plots of the logarithm of induction time versus the inverse of temperature (equivalent to Arrhenius plots) show excellent linearity and all have similar slopes, yielding a value of 1.791±0.020 eV for the activation energy. The logarithm of induction time also increases linearly with sulfate concentration and decreases linearly with V^V concentration. The slopes of these plots give values of concentration coefficients β_S and β_V5 which quantify the effects of concentration on induction times. Combining these with the Arrhenius slope, we have constructed a model¹¹ to predict the stability of sulfate-based vanadium catholytes. The model can also be used as a basis for accelerated testing¹² of electrolyte stability.

We have investigated a range of additives and shown that many suggested additives are ineffective. However, the Group V elements phosphorus and arsenic in the +5 oxidation state have a significant stabilizing effect.¹³ We have also used our methodology for measurement of thermal stability to examine the stability of VFB catholytes under actual operating conditions. Results on the effect of state of charge, and deterioration and recovery of stability will be discussed.

REFERENCES

1. M. Skyllas-Kazacos et al., J. Electrochem. Soc., 158, R55 (2011).
2. A. Weber et al., J. Appl. Electrochem., 41, 1137 (2011).
3. M. A. Miller et al., J. Electrochem. Soc., 163, A2095 (2016).
4. D. Oboroceanu et al., J. Electrochem. Soc., 163, A2919 (2016).
5. C. Petchsingh et al., J. Electrochem. Soc., 163, A5068 (2016).
6. Á. Cunha et al., Int. J. Energy Res., 39, 889 (2015).
7. A. Bourke et al., J. Electrochem. Soc., 163, A5097 (2016).
8. A. Bourke et al., J. Electrochem. Soc., 162, A1547 (2015).
9. D. Oboroceanu et al., J. Electrochem. Soc., 163, A2919 (2016).
10. D. Oboroceanu et al., J. Electrochem. Soc., 164, A2101 (2017).
11. D. N. Buckley et al., J. Electrochem. Soc., 165, A3263 (2018).
12. D. N. Buckley et al., J. Electrochem. Soc., 168, 030530 (2021).
13. D. Oboroceanu et al., J. Electrochem. Soc., 166, A2270 (2019).