The Effects of Sulfuric Acid and Vanadium Cations on the Morphology of Hydrated Nafion: MD Simulations

The vanadium redox flow battery (VRFB) is one of the promising electrical energy storage technologies in providing stable electricity supply from the primary solar and wind enery sources.1 For practical operation, it is critical to understand the electrolyte ion behavior within the VRFB membrane separator to gain insight into the ion permeation process. Here, we report our recent molecular dynamics (MD) investigations on the structure and interaction between hydrated Nafion membrane and the ionic species of the electrolyte solution: V2+, V3+, and HSO4 –. Classical MD simulations were undertaken on the following systems: Nafion/H2O/H3O+; Nafion/H2O/H3O+/HSO4 –; Nafion/H2O/H3O+/V2+; Nafion/H2O/H3O+/V3+; Nafion/H2O/H3O+/HSO4 –/V2+; and Nafion/H2O/H3O+/HSO4 –/V3+ at hydration levels (i.e., λ) of 6, 12, and 18 H2O/SO3H. The Nafion was modeled as a 15-mer ionomer with the system consisting of 64 macromolecules. A complete description of our methodology is detailed in recent manuscripts and is similar to our earlier work.2 The molecular potential models are described in prior publications.2-6 Figure 1(a) shows the spatial pair correlation function (PCF) between S-atom of the SO3 – group of Nafion with the H-atom of H3O+ and the cumulative H-atom number of H3O+ around each SO3 – (inset) for the Nafion/H2O/H3O+/HSO4 – system. The obtained first layer H3O+ numbers are: 1.18 (λ = 6); 0.72 (λ = 12); and 0.46 (λ = 18). The dashed line corresponds to the case of full association, i.e., each SO3 – has one H3O+ within the first hydration shell. Apparently, for λ = 6 the H-atom number is slightly larger than the full hydration, due to the correlation between sulfur atoms of Nafion and the H3O+ orientation. At high hydration level the H3O+in the first layer is significantly reduced due to the screening effect of water. Figure 1(b) displays PCFs between S-atom of the SO3 – group of Nafion with the H-atom of HSO4 – and the cumulative H-atom number of HSO4 - around the SO3 –. The dashed line at 0.052 indicates that all HSO4 - are in the first hydration shell of the sulfonate groups. Compare to H3O+ the H-atom number in the first layer at λ = 6 is below full hydration but still fairly high for an anion. We attribute this to the spatial limitation near the SO3 –sites due to the small size of the water cluster, among other factors. Figures 1(c) and (d) show the PCF of the sulfur atom of Nafion SO3 – with the V2+ and V3+, respectively, for the Nafion/H2O/H3O+/V2+ and Nafion/H2O/H3O+/V3+ systems. The PCFs exhibit a pronounced peak at about 2.70 and 2.40 Å for V2+ and V3+, respectively. For both ions, peak height significantly decreases with hydration level. The insets in Figures 1(c) and (d) show the cumulative number of vanadium cations around a SO3 –. At λ = 6, the number of V2+ ions is about 0.25. This number is significantly reduced at high hydration. Comparison of the two vanadium ions reveals that, the higher valence cations interact less directly with the sulfonate groups, with the number of V3+ 5 times smaller than V2+at λ = 6, and decreasing essentially to zero at λ = 18. This is probably due to the tight hydration shell of vanadium ions, which weakens the interaction between the sulfonate groups and the vanadium cations. Figures 1(e) and (f) plot the PCF between V2+, V3+ and the O2-atom (the three oxygen without bonded hydrogen) of HSO4 – in the Nafion/H2O/H3O+/HSO4 –/V2+, and Nafion/H2O/H3O+/HSO4 –/V3+ system, respectively. Comparing the PCFs for V2+ and V3+, it shows that the oxygen atom can make direct contact with the V2+ at λ = 6. The peak is missing at higher hydration levels λ = 12 and 18 due to the screening effect of water. The peak for the first layer is completely missing for V3+because of its tight water hydration layers. Acknowledgments The authors would like to thank the Navy for the support of this work through Grant N00014-12-1-0887 “Research Support for Naval Energy Needs.” 1. A. Z. Weber, M. M. Mench, J. P. Meyers, N. Philip, P. N. Ross, J. T. Gostick, and Q. Liu, Journal of Applied Electrochemistry 41, 1137 (2011). 2. S. T. Cui, J. Liu, M. E. Selvan, S. J. Paddison, and D. J. Keffer, Journal of Physical Chemistry C 112, 13273 (2008). 3. H. H. Loeffler, I. J. Yague, and B. M. Rode, Chemical Physics Letters 363, 367 (2002). 4. C. Kritayakornupong, Journal of Computational Chemistry 30, 2777 (2009). 5. C.-G. Ding, T. Taskila, K. Laasonen, A. Laaksonen, Chemical Physics 287, 7 (2003). 6. C. R. I. Chisholm, Y. H. Jang, S. M. Haile, and W. A. Goddard III, Physical Review B 72, 134103 (2005). Figure 1