Charge storage in Electrical Double-layer Capacitors (EDLCs) occurs through the accumulation of electrolyte ions in the electrode pores. Unlike batteries and fuel cells, energy is stored solely due to electrostatic attractions. The absence of electrochemical reactions results in high power outputs (> 10 KW/Kg) and offers EDLCs exciting applications in electric vehicles, high power tools, hybrid vehicle charging stations etc. A significant hindrance to the commercial applications of EDLCs is the self-discharge process. Self-discharge (SD) is the slow energy loss from a charged storage device. SD rates in EDLCs are usually two orders of magnitude larger than in batteries. Despite the limiting role of the SD in various applications, an understanding of the processes involved in SD and the options available to mitigate it are at a rudimentary stage. Conway et al. [1] reported that self-discharge could either be controlled by internal short-circuits (leakage) or by impurities in the cell that participate in electrochemical reactions (controlled by activation, diffusion, or shuttling between two different oxidation states). Kazaryan et al. [2] and Andreas et al. [3] reported SD measurements at different concentrations of Fe as an impurity, and suggested shuttle mechanism could be playing a role.

We designed a series of controlled experiments involving electrochemical isolation, monitoring of individual electrodes, and eliminating leakage-based SD altogether. Our experiments showed that shuttle-like steps play a role in the presence of Fe at concentrations higher than a critical value. These conclusions were reached without relying on model-fit based indirect methods. Shuttle mechanism is expected to discharge both the electrodes at an equal rate, similar to the leakage of charge from one electrode to the other. Our measurements clearly establish that this is not the case. The conclusion holds even when the electrode specific capacitance is considered to be a function of electrode potential as revealed by the measurements. The extent of self-discharge attributed to shuttle-like steps is however not the same on the two electrodes which raises a new set of questions. The study also confirms that Fe as an impurity plays a dramatically different role in the self-discharge of the two electrodes. We also studied the contributions of diffusion and activation-controlled processes at the individual electrodes at low concentrations of Fe, which, as seen in this work, does not involve self-discharge through shuttle-like steps. The careful observations made in our study to obtain model-independent inferences had excellent repeatability. Some of these show irreconcilable and challenging trends, and for the first time, raised questions hitherto unknown in the context of self-discharge of EDLCs. Some of the observations made to enlarge the earlier studies to obtain model-independent inferences have also pointed to repeatable but irreconcilable observations. These observations need the attention of the community at large. We believe that the efforts made to understand them qualitatively will lead to pathways for mitigation of self-discharge.

References:

1. Conway, B.E., Pell, W.G. and Liu, T.C., 1997. Diagnostic analyses for mechanisms of self-discharge of electrochemical capacitors and batteries. Journal of Power Sources, 65(1-2), pp.53-59.
2. Kazaryan, S.A., Kharisov, G.G., Litvinenko, S.V. and Kogan, V.I., 2007. Self-discharge related to iron ions and its effect on the parameters of HES PbO2∣ H2SO4∣ C systems. Journal of the Electrochemical Society, 154(8), p.A751.
3. Andreas, H.A., Lussier, K. and Oickle, A.M., 2009. Effect of Fe-contamination on rate of self-discharge in carbon-based aqueous electrochemical capacitors. Journal of Power Sources, 187(1), pp.275-283.