Impedance Variation of All-Solid-State Battery Cell Using Li10GeP2S12; States-of-Charge Dependence Visualized by Distribution-of-Relaxation Time Analysis

Abstract

Electrochemical impedance spectroscopy (EIS) will support understanding of electrochemical phenomena in emerging energy devices, including all-solid-state Li ion batteries (ASSLiBs), only when the results are analyzed in a consistent way without arbitrariness. This study1 aimed to establish such a robust analysis for ASSLiB cells using a fast lithium ion conductor Li10 +x Ge1+x P2 −x S12 (LGPS)2, by applying distribution-of-relaxation time (DRT) analysis3. The DRT method distinctively visualized impedance changes depending on states of charge (SOCs). Pellet-type cells were prepared with an In–Li anode, LGPS separator, cathode composite comprising LGPS and LiNbO3-coated LiCoO2 powders4. Figure 1a shows the discharge curve that was recorded for the cell at the current density of 13.2 μA cm−2 (1/40 C-rate) and with the cut-off voltage of 1.9–3.6 V vs. Li+/In–Li (2.55–4.25 V vs. Li+/Li). The EIS measurements were performed during discharge process at various SOCs from 100% to 20%. Nyquist diagrams obtained from the EIS data are shown in Figure 1b, where two broad semicircles are observed at low and high frequency sides. This result indicates at least two electrochemical processes exist in the cell. Both semicircles become large in size upon decreasing SOCs from 100 % down to 20%, indicating that the cell has SOC-dependent electrochemical processes. Figure 1c shows DRT transformation of the EIS data, which helps to separate impedance components in detail. Here horizonal and vertical axes respectively represent the measurement frequency and polarization contribution at each frequency. The center frequency of each distinct peak corresponds to time constant [τ =1/(2πf)] for the respective electrochemical process, and thus the number of peaks matches that of electrochemical processes in the cell. Meanwhile, the peak area represents polarization contribution of each process. Consistently in the whole SOC range, the above DRT analysis indicates the semicircle at high frequency side is divided into two impedance contributions and that at low frequency side cannot be separated any more. This result is consistent with the DRT spectra from additionally prepared symmetric cells; in total three DRT peaks were observed with two peaks from interphase and interface between In–Li/LGPS, and one from LiNbO3-coated LiCoO2/LGPS interface. Accordingly, DRT diagram in Figure 1c was interpreted as indicated by arrows for each peak. This study demonstrated that DRT analysis supports a consistent interpretation of EIS data from ASSLiB cells at various SOCs and with different electrode configurations, and thus implied that the method will also help subsequent equivalent circuit model analysis for reproducible parametrization of electrochemical processes.1. S. Hori et al., “Understanding the impedance spectra of all-solid-state lithium battery cells with sulfide superionic conductors,” J. Power Sources 556, 232450 (2023).2. N. Kamaya et al., “A lithium superionic conductor,” Nat. Mater. 10, 682–686 (2011).3. S. Dierickx, A. Weber, E. Ivers-Tiffée, “How the distribution of relaxation times enhances complex equivalent circuit models for fuel cells,” Electrochim. Acta 355, 136764 (2020).4. N. Ohta et al., “LiNbO3-coated LiCoO2 as cathode material for all solid-state lithium secondary batteries,” Electrochem. Commun. 9, 1486-1490 (2007). Figure 1