Impedance measurements of lithium-ion batteries are a powerful tool to investigate the electrolyte/electrode interface. To separate the contributions of anode and cathode to the full-cell impedance, a reference electrode is required. However, if the reference electrode is placed inappropriately, the impedance response can easily be biased and lead to erroneous conclusions. In this study, we present a novel micro-reference electrode for Swagelok-type T-cells which is suitable for long-term impedance and reference potential measurements. The reference electrode consists of a thin insulated gold wire, which is placed centrally between cathode and anode and is in-situ electrochemically alloyed with lithium. The resulting lithium-gold alloy reference electrode shows remarkable stability (>500 h) even during cycling or at elevated temperatures (40 • C). The accuracy of impedance measurements with this novel reference electrode is carefully validated. Further, we investigate the effect of different vinylene carbonate (VC) contents in the electrolyte on the charge transfer resistance of LFP/graphite full cells and demonstrate that the ratio of VC to active material, rather than the VC concentration, determines the impedance of the anode SEI.

The lifetime of lithium-ion batteries strongly depends on the properties of the interface between each electrode and the electrolyte. Electrochemical impedance spectroscopy is a simple and non-destructive method to investigate the electrode/electrolyte interfaces and the effect of electrolyte additives. As impedance measurements of full-cells always reflect the sum of both electrodes, it is difficult to deconvolute the contributions of the cathode and anode interface. Therefore, numerous reference electrode designs for impedance measurements have been suggested.1–6 However, if the reference electrode is not placed centrally between cathode and anode, which is the case for commonly used Swagelok T-cells, the measurement can easily be biased.2,7,8 Starting from the conventional T-cell design, we developed a novel micro-reference electrode which is placed centrally between both electrodes and two 200 µm thick glass fiber separators. The reference electrode consists of a thin gold wire. Due to the small wire diameter compared to the electrode separation, a distortion of the potential field between the electrodes is thus minimized. Similar as proposed by Abraham et al.6 for a tin-coated wire, we achieve a stable potential of the gold wire by electrochemical alloying with lithium. We show that the potential of this lithiated gold wire reference electrode (GWRE) is stable for several weeks, even under elevated temperatures. With this lithiated GWRE, we are able to record the potential of both electrodes in LFP/graphite full cells for more than 150 cycles. Further, we evaluate the capabilities of the lithiated GWRE to accurately measure the impedance of individual electrodes. We verify the impedance measurements of individual electrodes in full-cells with the lithiated GWRE by comparison with symmetrical cell measurements, which are commonly used for accurate electrode impedance analysis.9 Besides the electrolyte/electrode interface resistances, the lithiated GWRE is also suitable to probe electronic contact resistances of the electrodes. Using symmetrical cells, Burns et al.10 have shown that the charge transfer of a graphite anode depends almost linearly on the concentration of vinylene carbonate (VC) in the electrolyte. As a proof of concept, we conducted a similar study in LFP/graphite full cells with lithiated GWRE and different VC concentrations in the electrolyte. We can reproduce the findings by Burns et al.10 and further demonstrate that the total amount of VC per anode surface, rather than its concentration, is the key parameter for the electrolyte/anode interface resistance (s. Fig. 1). This result is important when electrolyte additives are tested in laboratory cells, as these cells typically have a higher electrolyte to active material ratio than commercial lithium-ion cells. Figure 1: Charge transfer resistance of graphite and LFP electrodes in full-cells with different amounts of VC added to the electrolyte (1M LiPF6 in EC:EMC 3:7). The impedance spectra were measured at 25 °C in LFP/graphite full-cells with a lithiated GWRE after 1 formation cycle at C/10 at 40 °C and a subsequent charge to 50% SOC. References: 1. T. S. Ong and H. Yang, J. Electrochem. Soc., 149, A1 (2002). 2. S. Klink et al., Electrochem. commun., 22, 120–123 (2012) 3. C. Bunzli, H. Kaiser, and P. Novák, J. Electrochem. Soc., 162, A218–A222 (2015) 4. C. Delacourt, P. Ridgway, V. Srinivasan, and V. S. Battaglia, J. Electrochem. Soc., 161, A1253–A1260 (2014) 5. J. Zhou and P. H. L. Notten, J. Electrochem. Soc., 151, A2173 (2004) 6. D. P. Abraham, S. D. Poppen, A. N. Jansen, J. Liu, and D. W. Dees, Electrochim. Acta, 49, 4763–4775 (2004) 7. M. Ender, A. Weber, and E. Ivers-Tiffée, J. Electrochem. Soc., 159, A128 (2012) 8. D. W. Dees, A. N. Jansen, and D. P. Abraham, J. Power Sources, 174, 1001–1006 (2007) 9. R. Petibon et al., J. Electrochem. Soc., 160, A117–A124 (2013) 10. J. C. Burns et al., J. Electrochem. Soc., 159, A85 (2012)   Acknowledgements: This work is financially supported by the BASF SE Battery Research Network. Funding for J. L. was provided by the Bavarian Ministry of Economic Affairs and Media, Energy, and Technology under the auspices of the EEBatt project. Figure 1