Data Science Tools for Incorporating Physics-Based Models into Analysis of Impedance Spectra

Monday, 29 May 2017: 14:00
Prince of Wales (Hilton New Orleans Riverside)
M. D. Murbach and D. T. Schwartz (University of Washington)
Electrochemical impedance spectroscopy (EIS) is a widely used method for noninvasively characterizing many electrochemical systems – probing the physicochemical response at different timescales to provide insight into a wide range of kinetic, transport, and thermodynamic parameters. By fitting equivalent circuit analogs to experimental impedance spectra, lumped parameters (ohmic and charge transfer resistances, double layer capacitances, Warburg impedances, etc.) which contain information about the physicochemical properties of the system are extracted. While this usage of equivalent circuit models makes fitting the data straightforward, the lumped parameter nature of this process can lead to limitations on the information content and discriminatory power of these measurements.1,2 Moreover, while the extracted parameters are often used as inputs to complex physics-based models, the reverse flow of information – directly incorporating modeling insights into the analysis of experimental impedance spectra – remains challenging for many researchers using EIS.

Here, we present an open-source, web-based tool for easily integrating physics-based modeling into the analysis of experimental impedance data. Using a large dataset of simulated spectra across a well sampled physical range of model parameters, the best fit of an experimental dataset can be used to find parameter estimates for the model. This data driven approach enables a more robust method for finding parameter estimates than optimizing the model directly and visualizations of the resulting best estimates give researchers the tools to more deeply analyze their data. Interactive visualizations of the simulated spectra also make it possible to ask more insightful questions about the model itself. As an example, we present a dataset of impedance simulations from a pseudo 2-dimensional (P2D) lithium ion battery model3,4 and describe insights gained by incorporating this dataset into analysis of experimental battery impedance spectra. Finally, we discuss the tool as an extensible platform for adding additional modeling datasets and features from across the community and the role of open-source tools within electrochemistry.


1. Fletcher, S. Tables of Degenerate Electrical Networks for Use in the Equivalent-Circuit Analysis of Electrochemical Systems. J. Electrochem. Soc. 141, 1823–1826 (1994).

2. Harrington, D. A. & van den Driessche, P. Mechanism and equivalent circuits in electrochemical impedance spectroscopy. Electrochimica Acta 56, 8005–8013 (2011).

3. Doyle, M., Meyers, J. P. & Newman, J. Computer simulations of the impedance response of lithium rechargeable batteries. J. Electrochem. Soc. 147, 99–110 (2000).

4. Meyers, J. P., Doyle, M., Darling, R. M. & Newman, J. The impedance response of a porous electrode composed of intercalation particles. J. Electrochem. Soc. 147, 2930–2940 (2000).