General Issues with EIS Data Acquisition
EIS is a very sensitive technique. In many cases, a small change in the electrodeelectrolyte interface causes significant changes in the resulting spectrum, and this is very useful in detecting small changes in the system. The other side of the story is that it is not easy to obtain reproducible results. Small changes in electrode preparation, or electrolyte purity, or the reference electrode position in the cell can significantly impact the results (Ziino, Marnoto, and Halpern 2020). A long electrode stabilization time may be required to get reproducible results (Xu et al. 2019). If an oxide film grows on the electrode during the experiment, then as the film thickness changes, the impedance response will vary as a function of time. Minor variations in conducting the experiments, such as different wait times being used between measurements at each frequency, can result in a difference in the results. In other words, EIS being a very sensitive technique also makes it harder to obtain reproducible data in some cases. Techniques such as potentiodynamic polarization (often referred to as Tafel experiments) or cyclic voltammetry or chronoamperometry yield data with good repeatability in most of the cases.
For a rapidly changing system, it is very difficult to get reproducible results. However, a few attempts have been made in the literature to acquire the impedance spectra of such systems (Popkirov 1996; Koster et al. 2017). The utility of such data is further impaired by the fact that any analysis that accounts for the system changes is very complex (Stoynov 1993; Victoria and Ramanathan 2011; Stoynov 1992).
Signal-to-Noise Ratio, Linearity Requirements, and Experiment Duration
A small perturbation amplitude will lead to a poor signal-to-noise ratio, as seen in the example in Section 2.4.2. On the other hand, a large perturbation amplitude will lead to nonlinear effects being incorporated in the response. For a given system, a series of experiments with varying perturbation amplitudes need to be performed to identify the range of amplitudes where the linearity approximation is applicable, and the signal-to-noise ratio is good, and this is time consuming.
The duration of a single EIS run depends on the frequency range, in particular, the lowest frequency and the number of frequencies per decade. It takes a particularly long time to measure the impedance at very low frequencies. While the high-frequency data contain information about the double-layer structure, the data at mid and low frequencies are essential to characterize the reaction and mass transfer effects. Thus, it becomes imperative to acquire EIS at low frequencies, and the experiment duration will be long. The noise level tends to be higher at lower frequencies. Thus, several sinusoidal cycles have to be employed to obtain a good signal-to-noise ratio, and this increases the duration of the experiment even more.
Exercise – Experimental Aspects
Q2.1 Typically, in which systems are potentiostatic mode EIS employed, and in which systems are galvanostatic mode EIS employed? Why?
Q2.2 Select a few publications with EIS in the last 5 years and identify the commonly used supporting electrolytes. What is the range of concentration of the supporting electrolyte employed?
Q2.3 Select a few publications with EIS in the last 5years and summarize the amplitude and frequency range employed in EIS studies on (a) batteries or fuel cells (b) corrosion or electrodeposition and (c) sensors. What inferences can we draw from this summary?
Q2.4 (a) Create a data set consisting of two sine waves of frequencies Ю and 20Hz, with amplitudes of 10 and 5 mV respectively, for 0-0.1 seconds, at 1 ms time intervals, (b) Add them together to synthesize a multi-sine wave. Plot the multi-sine wave as a function of time.
Q2.5 In the above example, introduce a phase difference of 30° between the waves at 10 and 20 Hz and recalculate the multi-sine wave and plot.
The following exercises require access to a potentiostat with a FRA. Some of the experiments require access to suitable wet labs. Please follow the appropriate safety standards.
Q2.6 Connect the reference and counter electrode leads of a potentiostat to one end of a resistor of 20 Q resistance, and the working electrode lead to the other end of the resistor. Measure the impedance from 100 kHz to 1 mHz at 7 points per decade, logarithmically spaced. Use a single sine with a lOmV amplitude. Plot the results in a complex plane and Bode format.
Q2.7 Repeat the above with a 10 pF capacitor.
Q2.8 Create a circuit shown in Figure 1.18 and measure the impedance spectrum, as described in Q2.6. Record the time for acquiring data. If the software permits selecting the ‘quality’ of data acquisition, e.g., fast v.v. high quality, evaluate the results in both modes.
Q2.9 Repeat Q2.8 with an amplitude of 0.1, 1, 20, and 50 mV. What can we conclude from these results?
Q2.10 Repeat the experiment described in Q2.9 but with an applied dc bias of (1) 100 mV and (2) 200 mV. Compare the results with those of Q2.9.
Q2.ll Prepare a three-electrode system with an Au working electrode, an Ag/ AgCl (3.5 M KC1) (or any other suitable) reference electrode, and a Pt mesh counter electrode. Connect the appropriate leads from the potentiostat. Prepare an electrolyte with 5mM K,[Fe(CN)6] + 5mM K4[Fe(CN)6] and 0.1 M NaC104. Polish the electrode using 2000 grit paper, followed by alumina slurries of decreasing particle sizes. Measure the OCP vs. time.
Q2.12 In the above setup, run CV at 5 and 50mV s~' while keeping the electrode stationary. Conduct an EIS experiment with parameters (frequency and amplitude) described in Q2.6.
Q2.13 In the above setup, conduct the experiments in Q2.12 at various electrode rotational speeds (100, 400, 900rpm, and so on). Compare the results with those of Q2.12.
Q2.14 In the above system, at a fixed rpm, conduct an EIS experiment with a dc bias of +0.1 V and -0.1 V vs. the OCP, with parameters described in Q2.6.
Q2.15 Repeat the experiments described in Q2.11-2.14, after increasing the concentration of the supporting electrolyte from 0.1 to 1 M. Note: If 1 M NaC104 is used, then NaCl must be used in the reference electrode solution instead of KC1. Otherwise, KC104 will precipitate and may block the reference electrode. Alternatively, Na2S04 may be used as a supporting electrolyte, and Ag/AgCl (3.5 M KC1) can be used as a reference electrode.
Q2.16 What is the duration of the experiment when the impedance of a circuit is measured vs. when the impedance of an electrochemical cell with a nonzero dc bias is measured for the same set of frequencies and perturbation amplitudes? If there is a significant difference, what could be the reason?
Q2.17 If the instrument has the capability for multi-sine measurement, perform the experiments described in Q2.13 and Q2.14 in multi-sine mode and compare both the measured results and experimental duration with those of single-sine measurement mode.
Q2.18 Perform the experiments described in Sections 2.4.3-2.4.5 and compare the results with those shown.
Q2.19 Repeat Q2.12, but with an electrode polished using 600 grit paper, and compare the results. What is the effect of changing surface preparation?