Use and Misuse of Electrochemical Impedance Spectroscopy (EIS) in Fuel Cell Research

Abstract

Electrochemical impedance spectroscopy (EIS) is a powerful method to analyze the sources of performance loss mechanisms in fuel cells and other electrochemical devices. Its major advantage is the reasonable experimental effort necessary to perform EIS measurements. However, the results of EIS measurements can easily be misinterpreted: in particular, two major mistakes can regularly be found in the fuel cell literature. The first of these mistakes consists in using a set of interpretation tools developed for electrochemical reactions near equilibrium. The charge transfer resistance (RCT) obtained from EIS is equivalent to a linearization of the Butler-Volmer equation around the operation point. While there is a linear relation between RCT and the corresponding charge transfer overpotential ηCT when the system is near equilibrium, these two values become fully uncorrelated for irreversible reactions such as the oxygen reduction reaction (ORR) of PEFCs (see Figure 1a). This is a direct and simply calculable consequence of being in the region where the Tafel approximation of the Butler-Volmer equation is valid. Thus, the value of RCT can by no means be considered a measure – or even an indication – of the magnitude of the activation losses related to the ORR. Despite this easily demonstrable fact (as developed for example in equation 10.37 of the reference book by Orazem and Tribollet (1)), values of RCT are often compared and discussed in terms of ORR activity. The second recurrent error is the comparison of measurements performed on highly inhomogeneous experimental cells with equivalent circuits assuming a homogeneous distribution of parameters, in particular along the flow channels. A common related misinterpretation is the attribution of the low frequency arc to only the oxygen diffusion across the gas diffusion layer (GDL). In their pioneering work, Brett et al. (2), Schneider et al. (3, 4) and Kramer et al. (5) identified the origin of this low frequency arc as being an oscillation of O2 concentration induced by the sinusoidal current, which propagates along the gas flow channels. A full model taking into account these oscillations was shown to reproduce the experimentally measured spectra. Moreover, measurements on segmented cells comparing the local impedance obtained by applying the sinusoidal perturbation either to the entire cell or to one specific segment confirmed without doubt the proposed explanation (4). This effect was later confirmed by other researchers (e.g. (6, 7)). The models presented in the literature take into account both the diffusion across the GDL and the oxygen concentration oscillation in the gas channels. However, it can easily be calculated (see result in Figure 1b) that a cell which with no diffusion limitations at all would also exhibit the low frequency loop characteristic of the concentration oscillation in the gas channels. Despite the strong evidence supporting the importance of this oscillation, it remains largely ignored when analyzing EIS data of fuel cells. In this critical review presentation, we will analyze the use of EIS in PEFC research and the frequency of the two interpretation mistakes described above – both in recent publications and on a historical perspective. We will present simple rules to assess EIS results interpretation critically, as well as our view on the correct procedures to extract useful information out of EIS results. References M. E. Orazem and B. Tribollet, Electrochemical Impedance Spectroscopy (2nd Edition), John Wiley & Sons, Inc., Hoboken, New Jersey (2017), ISBN 9781119341222.D. J. L. Brett, S. Atkins, N. P. Brandon, V. Vesovic, N. Vasileiadis and A. Kucernak, Electrochemical and Solid-State Letters, 6, A63 (2003).I. A. Schneider, S. A. Freunberger, D. Kramer, A. Wokaun and G. G. Scherer, Journal of The Electrochemical Society, 154, B383 (2007).I. A. Schneider, D. Kramer, A. Wokaun and G. G. Scherer, Journal of The Electrochemical Society, 154, B770 (2007).D. Kramer, I. A. Schneider, A. Wokaun and G. G. Scherer, ECS Transactions, 3, 1249 (2006).G. Maranzana, J. Mainka, O. Lottin, J. Dillet, A. Lamibrac, A. Thomas and S. Didierjean, Electrochimica Acta, 83, 13 (2012).A. A. Kulikovsky, Journal of The Electrochemical Society, 159, F294 (2012). Figure 1