EIS has been in use for material characterization since the 1950s with applications in corrosion, biosensors, battery development, fuel cell development, paint characterization, sensor development and physical electrochemistry. Battery testing is one application area among others where EIS is highly suitable.
EIS provides a non-destructive way of obtaining a snapshot of the cell’s current state, giving much insight into the properties of the battery cell. Due to this, EIS has become a standard tool for battery characterization, telling us for example about internal resistance, charge transfer kinetics, electrochemical reactions, capacity
and state of charge (SOC), electrolyte behaviour, aging and degradation.
In battery testing, EIS lets us evaluate state-of-charge, materials selection and electrode design. We can learn about various battery components by performing EIS at different frequencies. For example, charge transfer kinetics and the ohmic resistance of the electrolyte can be understood from the results of measuring at high frequencies, middle frequencies tell us about electrochemical reactions such as SEI capacitance and electron transfer rate and lower frequencies shed light on the diffusion processes of species in the insertion material. Broad frequency sweeps give us information on state of charge and capacity. More specific frequency-dependent features, indicative of electrolyte behaviour can be explored with EIS.
An EIS is performed by applying a test signal (most commonly a sinusoidal AC signal) to the battery. During the test, the voltage and current responses are recorded, tracking both amplitude and phase for the signals. From the measurement the impedance is calculated and the information is stored as a complex number, since it
conveniently contains both phase and magnitude.
The magnitude and phase of the impedance vary with the frequency and a complete EIS is composed of the results of several tests over a frequency range, often inside the limits of 1 mHz to 100 kHz. Most commonly the impedance results are visualised as a graph in the complex plane, with one point for the impedance for each frequency. The points are joined by lines and produce a curve trailing the frequencies. This is called a Nyqusit diagram or plot.
An analysis using EIS is usually performed in three steps
1. The EIS is measured to establish the impedance curve
2. A suitable model is fitted to the data representing the curve.
3. The elements of the fitted model are interpreted to provide understandable battery properties.
The predominant quantitative method for EIS data analysis is equivalent circuit modelling, which requires careful selection of equivalent circuit elements to represent physical processes in the system under study. It is not possible to reverse engineer the complete structure of the underlying physical network of components. Therefore, the analysis uses forward modelling to set model structures that match the physics of batteries and enables a user to interpret the result.
In single-frequency EIS, only one frequency is applied to the battery at a time. This is also the most common type of EIS measurement. A Nyquist plot is then assembled after multiple measurements. A more comprehensive characterization of a battery cell, including many frequencies can therefore take hours to complete.
An alternative to single-frequency EIS is to measure several frequencies simultaneously. This is what we call multifrequency EIS. It gives us a more comprehensive characterization of the battery cell in a shorter time, compared to performing multiple single-frequency EIS measurements. Giving higher frequency resolution,
multifrequency EIS is more suited to detecting more subtle changes in impedance behaviour across a broader frequency spectrum.
Current applications of simultaneous multifrequency EIS have also some known disadvantages. The set-up and instrumentation required is considered more complex and can require additional calibration steps or optimisation and control of experimental parameters to ensure accurate measurements. The increase in the volume of data that multifrequency EIS measurements give means that more storage space and computational resources are required. There are further challenges in analysing multifrequency EIS data as the decoding becomes more advanced. The more signals we squeeze into the testing time, the more information we will get out of the battery, and more care needs to be taken in the decoding.