2024 Volume 92 Issue 3 Pages 037007
Electrochemical impedance spectroscopy (EIS) of lithium insertion electrodes is a powerful technique to facilitate the further development of lithium-ion batteries (LIBs) because it provides useful information on electrochemical reactions. However, EIS methodology using a three-electrode cell is yet to be established owing to the difficulty in designing the reference electrode configuration. Therefore, a symmetric-cell method has been proposed to measure the EIS of a single electrode in a two-electrode cell, and it has been actively applied to EIS measurements in recent years. EIS measurements using symmetric cells have resulted in several new discoveries and deepened our understanding of the EIS behavior of lithium insertion electrodes. In this review, we outline the progress of EIS measurements using the symmetric-cell method for lithium insertion electrodes. First, we explain the principle, fabrication method, and EIS analysis of symmetric cells. Subsequently, an equivalent-circuit model representing lithium insertion electrodes is proposed based on the EIS measurement results. Furthermore, the factors affecting the various resistances comprising the equivalent circuit, such as charge transfer, contact, and ionic transport resistances, are discussed, including the dependence of the resistances on the electrode thickness, porosity, and measurement temperature on the resistances. Finally, applications of the EIS of a single electrode obtained using symmetric cells are described, including the prediction of the EIS of LIB and acquisition of the EIS of a single electrode from that of full cells.
Lithium-ion batteries (LIBs) are currently used as power sources in various applications ranging from small portable electronic devices to electric vehicles.1–3 Because the internal resistance of LIBs is attributed mainly to the resistances of the positive and negative electrodes, reducing the resistances of the electrodes is crucial for increasing the input and output powers of the battery. The lithium insertion electrodes used in batteries are porous electrodes wherein numerous active material particles are aggregated. Consequently, the resistance and capacitance components representing the electrodes, namely equivalent circuits, are considerably complex. Electrochemical impedance spectroscopy (EIS), which enables the distinction of each component with respect to the frequency response,4–6 is the most effective method for clarifying the resistance components of electrodes in LIBs.7–14
Electrochemical measurements using a two-electrode cell provide only the internal resistance of the battery, which is the sum of the resistances of the positive and negative electrodes and the ohmic loss of the electrolyte. A three-electrode cell equipped with a reference electrode is typically used to examine the resistance of a single electrode.15 However, the application of a three-electrode cell to the EIS measurements of the electrodes is challenging. This is because in LIBs, the positive and negative electrodes are placed facing each other with a very thin separator in between, which hinders the placement of the reference electrode in a position where it does not interfere with the current distribution. To date, various studies have focused on determining the appropriate location of the reference electrode and have reported errors depending on the location and shape of the reference electrode.16–23 In particular, EIS requires more attention because the measurements are performed over a wide frequency range (from more than 100 kHz to less than 10 mHz) and the voltage amplitude is extremely small at approximately 10 mV.
Instead of a three-electrode cell, a symmetric cell with the same configuration as that of a full cell has been proposed to measure the resistance of a single electrode without using a reference electrode.24 The symmetric cell comprises two equivalent electrodes, allowing the measurement of single electrode behavior in a two-electrode cell. Over the past two decades, several EIS studies on lithium insertion electrodes using symmetric cells have been conducted, and the EIS behavior of the electrodes has been gradually revealed.25–32
After introducing the principles of symmetric cells, their preparation methods, and EIS analysis examples, this review discusses the EIS measurements of lithium insertion electrodes for LIBs using the symmetric-cell method reported to date. Subsequently, we explain the equivalent circuit and factors affecting the charge transfer, contact, and ionic/electronic resistances. Finally, the applications of the EIS of a single electrode measured using symmetric cells for battery research are described.
A symmetric cell comprises two equivalent electrodes with the same material, weight, and thickness, as shown in Fig. 1a. Herein, the electrochemical response obtained using a symmetric cell is exactly twice that of a single electrode. Based on the equivalent circuit shown in Fig. 1b, the electrochemical impedance of the symmetric cell (Zsym) is defined as the sum of the impedances of the two single electrodes (Zsingle) and the resistance due to the ionic conduction of the electrolyte (Zsol), as expressed in Eq. 1. Therefore, when Zsingle is estimated by reducing Zsym by half, Zsingle includes Zsol, as expressed in Eq. 2.
| \begin{equation} Z_{\text{sym}} = 2 \times Z_{\text{single}} + Z_{\text{sol}}, \end{equation} | (1) |
| \begin{equation} \frac{Z_{\text{sym}}}{2} = Z_{\text{single}} + \frac{Z_{\text{sol}}}{2}. \end{equation} | (2) |
(a) Schematic of symmetric cell composed of two identical LTO electrodes. (b) Equivalent circuit of LTO/LTO symmetric cell. RLTO and CLTO represent the resistance and capacitance of the LTO electrode, respectively. Rsol denotes electrolyte resistance.
However, to analyze the impedance behavior of a single electrode, Zsol can often be ignored because Zsol exhibits resistance behavior and has a small value owing to the very thin separator used in the symmetric cell. This assumption is valid because in the impedance assignment of a Li-ion cell, the Zsol only affects Z of the cell in the high frequency limit and is not considered in the assignment of complex arcs. In this case, Zsingle can be obtained using the following equation:
| \begin{equation} Z_{\text{single}} = \frac{1}{2}Z_{\text{sym}}, \end{equation} | (3) |
| \begin{align} &| Z_{\text{single}} | = \frac{1}{2}| Z_{\text{sym}} |,\ \text{and}\ \theta_{\text{single}} = \theta_{\text{sym}},\ \notag\\ &Z_{\text{single}}' = \frac{1}{2}Z_{\text{sym}}',\ \text{and}\ Z_{\text{single}}'' = \frac{1}{2}Z_{\text{sym}}''. \end{align} | (4) |
where, θ, Z′, and Z′′ are the phase shift and real and imaginary parts of the impedance, respectively. Based on the above equations, the EIS of a single electrode can be obtained by applying a symmetric cell, although the symmetric cell is a two-electrode cell without a reference electrode.
The fabrication method for a symmetric cell comprises three steps:33,34 (1) preparation of two equivalent electrodes with the same weight and thickness; (2) fabrication of cells with a Li-metal electrode using each electrode and precycling to ensure that the electrochemical properties (capacity and polarization) of the two electrodes are equal; and (3) combination of the two electrodes to fabricate the symmetric cell. Figure 2 outlines the procedure for fabricating a symmetric cell with a Li[Li1/3Ti5/3]O4 (LTO) electrode as an example. The electrochemical behavior of the two identical LTO electrodes examined in the Li/LTO cells, which is shown in Fig. 2a, confirmed that the two LTO electrodes were equivalent in terms of reversible capacity and polarization. Both the LTO electrodes with the same composition of Li3/2[Li1/3Ti5/3]O4 exhibited exactly the same EIS behavior, as expected (Fig. 2b).
(a) Charge–discharge curves and (b) Bode plots of two identical LTO electrodes (32.5 mg with 113 µm and 33.1 mg with 115 µm) examined in Li/LTO cells for preparing the symmetric cell. (c) Voltage profiles of the LTO/LTO symmetric cell. The blue and red curves indicate the initial and final voltage profiles in the cycling test.
An LTO/LTO symmetric cell was fabricated by combining the two LTO electrodes obtained from the Li/LTO cells (Fig. 2c). Because the same electrode materials were used in the symmetric cell, the operating voltage was 0 V. Consequently, the same capacity as that of the Li/LTO cell was delivered, indicating that the symmetric cell worked well. In the symmetric cell, the chemical composition of the two LTO electrodes was the same (Li3/2[Li1/3Ti5/3]O4) at half capacity. Therefore, the impedance of the LTO/LTO symmetric cell was twice that of the LTO electrode.
Figure 3a shows the Z of the LTO/LTO symmetric cell and the LTO single electrode calculated using the Z of the symmetric cell using Eq. 4.34 For comparison, the impedance of the Li/LTO cell is shown. The impedance of the LTO single electrode was different from that of the Li/LTO cell, indicating that the impedance of the Li-metal electrode could not always be ignored unless the Z of the working electrode was sufficiently large.27,35–38 Figure 3b shows the Nyquist plots of Z for the LTO/LTO symmetric cell, LTO single electrode, and Li/LTO cell. The impedance of the single LTO electrode was calculated by reducing the real and imaginary parts of the impedance of the symmetric cell by half, as expressed in Eq. 4.
(a) Bode and (b) Nyquist plots of the LTO/LTO symmetric cell (blue) and its halves (red). The plots of the Li/LTO cells used to prepare the symmetric cell are also shown for comparison (black).
As described in the previous section, the fabrication of a symmetric cell includes several steps. A problem with the use of symmetric cells to measure the EIS of electrode materials is the change in the EIS of the electrode materials depending on the chemical composition.39–42 To examine the EIS of the electrode materials as a function of their chemical composition, many symmetric cells with electrodes of each composition are necessary, along with the number of compositions to be tested. To avoid this complexity, various types of symmetric cells have been reported to facilitate convenient measurements.43,44 The cell shown in Fig. 4a was designed to allow the composition of the electrode materials to vary, while maintaining a symmetric-cell configuration.43 In this cell, the Li-metal electrode was placed between two identical electrodes; consequently, the cell was transformed into a symmetric cell by removing the Li-metal electrode. Figure 4b illustrates the fabrication method for a symmetric cell comprising electrodes obtained from full cells.44 This symmetric cell is suitable for testing deteriorated electrodes after long-term cycling and storage tests wherein the deterioration mechanism differs between the symmetric and full cells owing to crosstalk reactions.45–48
Various types of symmetric cells were used for convenient EIS measurements. (a) A cell composed of a Li metal electrode placed between two identical electrodes, which enables the transformation of a symmetric cell by removing the Li metal electrode. Reprinted from Ref. 43, Copyright 2018 Elsevier. (b) Fabrication of symmetric cell consisting of electrodes obtained from full cells. Reprinted from Ref. 44, Copyright 2014 Elsevier.
In this section, the equivalent circuit of a lithium insertion electrode is discussed based on the EIS of a single electrode measured using symmetric cells. For the EIS of a flat-plate electrode, a Randles circuit wherein a resistor and a capacitor are connected in parallel, owing to charge-transfer resistance and electric double-layer capacitance, respectively, is used for equivalent-circuit analysis (Fig. 5a).49 The Randles circuit contains only one semicircle. However, the EIS of a lithium insertion electrode obtained using symmetric cells differs significantly from that of a Randles circuit. The EIS results for a symmetric cell with a graphite electrode at various temperatures are presented in Fig. 5b.50 Only one arc is observed in the EIS above room temperature; however, a new arc appears in the low-frequency region at low temperatures, resulting in two arcs, even for a single electrode.28,51–56 Thus, the Nyquist plot of a single electrode exhibits different characteristics from that of a Randles circuit.
(a) Typical Nyquist plot of a flat-plate electrode; the characteristics of a Randles circuit are shown in the figure. (b) Nyquist plots of symmetric cells with graphite electrodes at various temperatures. Reprinted from Ref. 50, Copyright 2017 IOP Publishing.
In the following section, we discuss two features of the EIS results of lithium insertion electrodes. As shown in Fig. 5b, the size of the high-frequency arc remains constant with change in the temperature, whereas that of the low-frequency arc changes significantly with temperature. Thus, the low-frequency arc is caused by the charge-transfer resistance, whereas the high-frequency arc is a resistance component caused by electronic and/or ionic conduction. The validity of this attribution is verified by the following results. Figure 6a shows the EIS results of a lithium insertion electrode at state of charge (SOC) values of 50 and 0 % (fully discharged states).51 A low-frequency arc is observed at 50 %, which disappears at 0 %. Thus, the low-frequency arc is caused by the lithium insertion reaction. In the EIS of the lithium insertion electrode using an electrolyte without Li ions (Fig. 6b), only a high-frequency arc is observed.57 Moreover, the EIS of a porous electrode that does not contain an active material exhibits only a high-frequency arc (Fig. 6c).58 As no lithium insertion reaction occurs in either case, the high-frequency arc is independent of the charge-transfer reaction and attributes to the contact resistance at the interface between the active materials and the current collector.18,56,58 Furthermore, when impedance measurements are performed using the diluted electrode method with small amounts of active materials, only the size of the low-frequency arc changes.58 Because the reciprocal of the resistance of the low-frequency arc is proportional to the active material content (Fig. 7), the low-frequency arc represents the charge-transfer resistance of the lithium insertion reaction.
Nyquist plots of (a) a lithium insertion material at SOC of 50 % (black) and 0 % (red) (reprinted from Ref. 51, Copyright 2019 IOP Publishing), (b) a lithium insertion electrode using an electrolyte without Li ions (reprinted from Ref. 57, Copyright 2020 Elsevier), and (c) a porous electrode without an active material (reprinted from Ref. 58, Copyright 2019 Elsevier).
Plots of the reciprocal of Rct calculated from the diameter of the high-frequency arc of the diluted LAMO electrodes against the loading weight. Reprinted from Ref. 58, Copyright 2019 Elsevier.
Another unique feature of the EIS of the lithium insertion electrode is the appearance of a linear region to the left of the low-frequency arc. This behavior is characteristic of porous electrodes wherein nano- or micron-sized active material particles are aggregated. The transmission-line model has been used to analyze the EIS of porous electrodes.7,55,59 In this model, in addition to the charge-transfer resistance (Rct) of each particle, the resistances related to electronic conduction (Re) and ionic conduction (Rion) should be considered for EIS. Figure 8 shows the simulated EIS results for porous electrodes with and without a charge-transfer reaction.60 In case of negligible resistance owing to electronic conduction (Re), the resistance related to ionic conduction (Rion) can be evaluated from the EIS in the straight-line region within the high-frequency region. Further, Rct can be estimated from the diameter of the semicircle.
Nyquist plots calculated from transmission-line models consisting of (a) Cdl and Rion, and (b and c) Rct in addition to Cdl and Rion. Reprinted from Ref. 60, Copyright 2012 IOP Publishing.
However, in actual EIS measurements of lithium insertion electrodes, a straight line is rarely observed clearly owing to the overlapping of distorted semicircles. Here, the Rion in the linear region can be analyzed by measuring the EIS of the electrode in its initial state (SOC = 0 %) (Fig. 6a).51,60,61 Conventionally, EIS measurements are performed for electrodes at the middle SOC where the lithium insertion reaction occurs; however, by intentionally measuring the EIS at 0 % SOC in the absence of a reaction, only the straight-line area related to Rion can be observed because the semicircle disappears owing to the infinite value of Rct. Accordingly, the EIS of the porous electrode contains information on the charge-transfer resistance and the resistance related to electronic/ionic conduction.
By considering all the resistances related to the lithium insertion electrodes, an equivalent-circuit model was obtained, as shown in Fig. 9a.52,62 The equivalent circuit comprised solution resistance Rsol, contact resistance Rcont, and a transmission-line model representing a porous electrode. In the transmission-line model, parallel circuits of charge-transfer resistance Rct and electric double-layer capacitance Cdl were infinitely lined up, and these RC parallel circuits were connected via resistances Re and Rion for electronic and ionic conduction, respectively. Figure 9b shows the Nyquist plot of the EIS results calculated based on the equivalent circuit. From the Nyquist plot, each resistance component was estimated as follows:
(a) Equivalent circuit of lithium insertion electrodes and (b) Nyquist plot calculated using the circuit.
Electrode thickness affects the Rct and Rion values because thicker electrodes contain many active material particles and have longer ionic conduction distances within the electrode.43,61,64 The thickness dependence of Rct (Fig. 10a) indicates that Rct decreases with increasing electrode thickness.64 Because Rct depends on the reactive surface area, the reciprocal of Rct is proportional to the electrode thickness. Further, Rion also increases in proportion to the electrode thickness (Fig. 10b) because thicker electrodes have longer ionic conduction distances.63 The thickness dependences of Rct and Rion have a tradeoff relationship; with increasing electrode thickness, Rct and Rion decreases and increases, respectively. Therefore, the total resistance of the electrode is dominated by Rct and Rion for thin and thick electrodes, respectively. Therefore, if the electrode is excessively thin or thick, the resistance of the electrode per unit area is high, and there is an optimal thickness at which this resistance is minimized.64
Electrode thickness dependence of (a) Rct and (b) Rion. The reciprocal of Rct is proportional to electrode thickness. Reprinted from Ref. 62, Copyright 2015 American Chemical Society.
Both Rct and Rion depend on the electrode density (Fig. 11).43 With an increase in electrode density, Rct decreases because a higher electrode density results in more active material particles. Increasing the electrode density decreases the porosity, resulting in fewer ion transport paths. Thus, Rion rapidly increases as the density approaches the theoretical value (no porosity).43,65,66 According to the results in Fig. 11, Rct exhibits reciprocal relationship with respect to the electrode density.
The resistance of an electrode varies depending on the type of conductive carbon.67–69 In a LiCoO2 electrode, Rcont changes significantly depending on the type of conductive carbon.50 In addition, the semicircle due to Rct appears at low temperatures and changes in size depending on the type of conductive carbon; however, the effect is smaller than that for Rcont. The amount of conductive carbon in an electrode affects Rion because the tortuosity increases with an increase in the amount of conductive carbon. Tortuosity is closely related to ion transport distance, despite having the same electrode thickness.57,59 Thus, Rion increases with increasing amount of conductive carbon (Fig. 12).61 In addition to the conductive carbon, the polymer binder also affects the tortuosity of an electrode; a larger amount of binder causes a larger Rion due to an increase in tortuosity.70–72
Plots of Rion as a function of loading mass for electrodes with various amounts of conductive additive. A larger amount of conductive additive increases tortuosity, resulting in an increase in Rion. Reprinted from Ref. 61, Copyright 2020 American Chemical Society.
Because the electrolyte significantly influences battery performance, such as cyclability and rate capability, many studies have been conducted on electrolytes in terms of solvents, Li salts, and concentration.73,74 The electrolyte affects the Rct of electrodes in addition to Rsol. EIS using symmetric cells is a powerful analysis method for separating the contributions of positive and negative electrodes with respect to the influence of the electrolyte. Previous studies reported that Rct changes depending on the type of solvent and Li salt.75 However, no comprehensive study has been conducted. Thus, the correlation with the physical properties of the electrolyte, such as conductivity and viscosity, is yet to be clarified. The concentration of Li salt also affects Rct; a higher concentration results in a smaller Rct.75 Additives to the electrolyte also change Rct;76,77 a highly effective additive reduces Rct by less than half by enhancing interface stability,78,79 although the details of its mechanism remain unknown.
5.4 Measurement temperatureThe temperature dependence of each resistance component of a lithium insertion electrode differs significantly (Fig. 13). For certain resistance components, the logarithm of the resistance is linearly related to the reciprocal of the absolute temperature,80,81 following the Arrhenius relationship. Among all resistances, Rct exhibits the greatest temperature dependence because of the high activation energy of the lithium insertion reaction.51,60,62 One reaction responsible for the high activation energy is the solvation/desolvation process.82–84 Further, Rion and Rsol, which are related to ionic conduction in the electrode pores and separator, respectively, exhibit temperature dependence; however, the degree of dependence is smaller than that of Rct. Conversely, the temperature dependence of Rcont is negligibly small because of electron transport. As the temperature dependences of the resistance components are significantly different, as discussed above, each resistance component can be separated and identified by examining the EIS at varying temperatures.
Finally, the applications of the EIS of a single electrode obtained from symmetric cells are described as follows: (1) identifying the origin of the impedance rise in deteriorated LIBs, (2) calculating the EIS of LIBs, and (3) estimating the EIS of a single electrode from a full cell.
6.1 EIS analysis of deteriorated LIBs using symmetric cellsOne useful application of the symmetric-cell method in actual LIBs is the analysis of degraded batteries to identify the causes of deterioration and consequently, investigate the contributions of the positive and negative electrodes to the increase in internal resistance.11,44,85–88 Determining the electrode that causes an increase in resistance from the EIS data of a full cell is challenging. The symmetric-cell method facilitates the evaluation of the resistance rise at the positive and negative electrodes, separately and independently. Figure 14 shows the results of the EIS of graphite/graphite and LiCoO2/LiCoO2 symmetric cells made from a graphite/LiCoO2 cell after a deterioration test.44 For both the graphite negative and LiCoO2 positive electrodes, the low-frequency arc increased after the degradation test, which increased the internal resistance of the battery. Furthermore, for the graphite electrode, Z increased in the frequency range of 100–1 Hz, whereas for the LiCoO2 electrode, Z increased in the frequency range below 1 Hz. Via the application of the symmetric-cell method, the cause of the increase in the internal resistance of the full cell was clarified. The resistance increase in the frequency range above 1 Hz was attributed to the graphite negative electrode, and that below 1 Hz was due to the LiCoO2 positive electrode. According to previous EIS analyses of LIBs, the two arcs at high and low frequencies correspond to the Rct of the negative and positive electrodes, respectively. In general, it is believed that the influence of the negative and positive electrodes is significant at high and low frequencies, respectively. However, more accurate predictions and analyses can be obtained by analyzing the results of symmetric cells.
Nyquist plots of symmetric cells with (a) graphite and (b) LiCoO2 electrodes after cycling the graphite/LiCoO2 cells. Reprinted from Ref. 44, Copyright 2014 Elsevier.
The EIS of a single electrode obtained using the symmetric-cell method enables more accurate prediction of the EIS of a full cell because the EIS of LIBs is the sum of the EIS of the positive and negative electrodes according to the following formulas:14,27,89
| \begin{equation} Z_{\text{cell}} = Z_{\text{PE}} + Z_{\text{NE}}, \end{equation} | (5) |
| \begin{equation} Z_{\text{cell}}' = Z_{\text{PE}}' + Z_{\text{NE}}',\ \text{and}\ Z_{\text{cell}}'' = Z_{\text{PE}}'' + Z_{\text{NE}}'', \end{equation} | (6) |
| \begin{equation} | Z_{\text{cell}} | = \sqrt{(Z_{\text{cell}}')^{2} + (Z_{\text{cell}}'')^{2}},\ \text{and}\ \theta_{\text{cell}} = \tan^{-1}(Z_{\text{cell}}''/Z_{\text{cell}}'). \end{equation} | (7) |
As shown in Fig. 15, the EIS of the full cell calculated from the EIS of a single electrode measured using symmetric cells closely matched the observed EIS of the full cell, which validated the accuracy of the EIS prediction by the symmetric-cell method.14 Furthermore, by comparing the EIS results of the positive and negative electrodes with those of the LIB, the following findings were clarified. (1) The EIS of the full cell was dominated by the graphite negative electrode only in the high-frequency region (>10 Hz), in contrast to the LiCoO2 positive electrode, which dominated over the entire frequency range. (2) Ensuring the assignment of high-frequency arcs to negative electrodes and low-frequency arcs to positive electrodes was not strictly possible.
The cell impedance (black) calculated from the single-electrode impedances obtained for the positive (red) and negative (blue) symmetric cells closely matches the actual cell impedance (green). Reprinted from Ref. 14, Copyright 2022 Authors CC BY 4.0.
The specific EIS of a single electrode provides an opportunity to calculate the EIS of another single electrode. If the EIS of either electrode is known using the symmetric-cell method, the EIS of the other electrode can be calculated from the EIS of the full cell, which is the sum of those of the two electrodes. For example, if the EIS of the negative electrode ZNE is known, that of the positive electrode ZPE can be calculated using the following equations:58,78,90
| \begin{equation} Z_{\text{PE}} = Z_{\text{cell}} - Z_{\text{NE}}, \end{equation} | (8) |
| \begin{equation} Z_{\text{PE}}' = Z_{\text{cell}}' - Z_{\text{NE}}'',\ \text{and}\ Z_{\text{PE}}'' = Z_{\text{cell}}'' - Z_{\text{NE}}''. \end{equation} | (9) |
To obtain EIS data for a single electrode using the above equations, an appropriate material must be selected for the counter (negative) electrode. Among the electrode materials reported thus far, an LTO electrode is ideal because its electrode potential is constant at 1.56 V,91–93 and its impedance undergoes minimal changes depending on the composition (Fig. 16a).34,89 For the LTO electrode, the impedance changes only at the end of charging and discharging. Thus, by fabricating a cell by adjusting the SOC of the LTO electrode, as shown in Fig. 16b, the impedance of the LTO electrode remains constant over the entire SOC range of the positive electrode.94
(a) Z of the LTO/LTO symmetric cell at different SOCs. Reprinted from Ref. 34, Copyright 2014 IOP Publishing. (b) Schematic of lithium-ion cell with NCM and extra-capacity LTO electrodes. Reprinted from Ref. 94, Copyright 2020 IOP Publishing. (c) Z of the NCM single-electrode calculated by subtracting the Z of the LTO single-electrode from the Z of the LTO/NCM cell.
The Z value of any positive electrode at any composition can be determined by subtracting the Z value of the LTO electrode from that of the full cell. Figure 16c shows an example of calculating the EIS of the LiNi1/3Co1/3Mn1/3O2 (NCM) electrode by subtracting the Z of the LTO electrode from the Z of the LTO/NCM cell. Consequently, the Z of the LTO/NCM cell changes depending on the SOC of the NCM electrode, which is consistent with previous reports.95,96 Thus, the Z of a single electrode can be indirectly calculated based on the Z value of the counter electrode obtained from the symmetric cell.
This review described the usefulness of the symmetric-cell method for analyzing the EIS of lithium insertion electrodes in LIBs. The significance of the symmetric cell is that, despite being a two-electrode cell, it can accurately and conveniently measure the EIS of a single electrode with the same electrode configuration in a battery. The symmetric cell is the only method used to measure the EIS of lithium insertion electrodes, except for an ideally designed three-electrode cell. This is because EIS using a cell with a Li-metal electrode cannot avoid contamination of the impedance of the Li-metal electrode. The importance of the EIS of a single electrode is that, from a scientific perspective, it provides a chance to identify the rate-determining step of the lithium insertion reaction based on an equivalent-circuit model. Further, from an engineering perspective, it is essential to predict and analyze battery performance.
An ideal three-electrode cell equipped with a reference electrode that does not interfere with the current distribution and can accurately measure the electrode potential would allow all electrochemical measurements for a single electrode, including EIS. Symmetric cells also play an important role in ensuring that three-electrode cells provide accurate information. By confirming that the impedance data measured with the symmetric cell are equivalent to those measured with the three-electrode cell, the reliability of the three-electrode cell, including the reference electrode, can be guaranteed.97–99
The following two points must be considered when analyzing the electrochemical measurements of symmetric cells. First, the EIS of a single electrode calculated from that of a symmetric cell inevitably includes Rsol, although Rsol is typically negligible because of its simple resistance behavior. Second, in symmetric cells composed of equivalent electrodes, the electrochemical reactions of the two electrodes occur in opposite directions (when one electrode is oxidized, the other electrode is reduced). Consequently, the symmetric-cell method cannot analyze the asymmetry of the electrochemical reaction, whereas the three-electrode cells can analyze the asymmetry using dynamic EIS measurements.100–103
Despite these drawbacks, the symmetric-cell method is a powerful technique for analyzing the EIS of lithium insertion electrodes. Thus, accumulating knowledge regarding the EIS of lithium insertion electrodes using the symmetric-cell method is expected to facilitate a deeper understanding of the kinetics of the lithium insertion reaction and further development of high-power LIBs.
The manuscript was written with contributions from the author. The author approved the final version of the manuscript.
Kingo Ariyoshi: Conceptualization (Lead), Visualization (Lead), Writing – original draft (Lead)
The author declares no competing financial interest.
This review was recommended for submission by the Editorial Board of Electrochemistry under the author’s proposal, and was accepted after peer review.
K. Ariyoshi: ECSJ Active Member
