Vanadium redox flow batteries (VRFBs) offer a scalable, long lasting, highly efficient means of energy storage and are a good option for large scale energy storage systems that can work on hand with renewable energy sources such as wind and solar farms. The proton exchange membrane that separates the negative and positive electrodes is a fundamental part of the system, since it allows the diffusion of H⁺ ions between electrodes to make possible the oxidation/reduction reactions in the half cells.¹ In VRFBs the membrane electrolyte is desired to have high proton conductivity, low vanadium permeability to avoid battery discharge caused by vanadium crossover and good durability. The most common membrane used in VRFBs has been Nafion due to its success in fuel cells. Previous work in our laboratory suggests that other membranes offer improved performance.^2,3 However, in a VRFB the working environment differs greatly from that of a fuel cell, for example, the membrane can adsorb electrolyte species other than protons and the high electrolyte concentration will tend to dehydrate the membrane. These conditions can substantially affect the membrane’s conductivity as well as other transport properties. In the flow battery, the trade-off between membrane conductivity and species transport across the membrane is key and is largely driven by the tendency of the membrane to imbibe different solution components as well as their intrinsic mobility.

Our group has developed a core set of measurements for membrane characterization that includes conductivity, porosity, acid, water and vanadium uptake, and vanadium permeability.^2,4 In this work we studied a series of 3M ionomer membranes, conditioned in solutions of different concentrations of sulfuric acid and vanadium. Membranes were tested as received and also after boiling them in water to expand the polymer channels by hydrating the ionic clusters.⁵ The effect of channel expansion in the polymer electrolyte was studied as function of the membranes equivalent weight. In addition, a set of membranes was subjected to heat treatments near the polymer glass transition temperature, to determine the temperature effect on the membranes properties.

Boiling the membranes dramatically increases their conductivity, porosity, water, acid and vanadium uptake, which is ideal for VRFBs, but also increases the vanadium permeability which is directly related to the undesired crossover. Figure 1, shows an example of how the membrane porosity, which is determined from density measurements obtained by gas pycnometry, is affected in a 3M825EW membrane. The boiled membrane shows higher porosity compared to that of the as received membrane. Higher porosity is an indication of the channels expansion in the membrane after boiling.

Acknowledgements

We would like to thank Dr. Greg Haugen and Dr. Tyler Matthews at 3M company for providing the membranes and funding for this study.

References

1. M. Sukkar, T., Skyllas-Kazacos, J. Memb. Sci., 222, 249–264 (2003).
2. Z. Tang et al., J. Electrochem. Soc., 161, A1860–A1868 (2014).
3. R. A. Elgammal, Z. Tang, C.-N. Sun, J. Lawton, and T. A. Zawodzinski, Electrochim. Acta, 237, 1–11 (2017)
4. J. S. Lawton et al., J. Electrochem. Soc., 163, A5229–A5235 (2016).
5. T. E. S. and S. G. Thomas A. Zawodzinski Jr., Charles Derouin, Susan Radzinski, Ruth J. Sherman, Van T. Smith, J. Electrochem. Soc., 140, 1041–1047 (1993).