In-situ Analysis of the Solid-Electrolyte Interphase Formed in Li[N(CF3SO2)2] and Li[N(FSO2)2] Tetraglyme Solvate Ionic Liquids

Abstract

Li metal has attracted attention as the negative electrode for secondary batteries due to its low density and negative redox potential. However, its negative operating potential also leads to the decomposition of electrolytes, resulting in the formation of some interfacial phase on the anode, i.e., solid-electrolyte interphase (SEI). The decomposition of the electrolyte causes the low coulombic efficiency of the Li anode. Therefore, the characterization of SEI is necessary in order to utilize the Li anode.Solvate ionic liquids (SILs) composed of 1:1 equimolar mixture of lithium bis(trifluoromethylsulfonyl)amide (LiTFSA) or bis(fluorosulfonyl)amide (LiFSA) and glyme have been investigated as the alternative electrolytes for lithium secondary batteries because of their high Li+ concentration [1, 2]. The deposition and dissolution of Li were reported to be possible in the SILs, although the cathodic decomposition of the electrolyte also proceeded at the potential more positive than that of Li(I)/Li. We have already reported the improvement of the morphology of Li deposits and the cyclability of the Li anode in LiTFSA-triglyme (G3) by the formation of SEI on a Cu electrode at 0 V vs. Li|Li(I) prior to the deposition of Li [3]. The cyclability of the Li anode in LiFSA-tetraglyme (G4) was found to be superior to that in LiTFSA-G4. However, the difference in the stability of Li deposits in these SILs has not been elucidated. In the present study, the formation of SEI on a Cu electrode and the interfacial reaction of the Li deposits were investigated in LiTFSA-G4 and LiFSA-G4 using electrochemical impedance spectroscopy (EIS) combined with electrochemical quartz crystal microbalance (EQCM).The solvate ionic liquids were prepared by mixing the equimolar quantities of G4 and LiTFSA or LiFSA. EIS and EQCM measurement were conducted on the same Cu-coated EQCM electrode (fundamental frequency: 9 MHz) using an impedance-type EQCM system (SEIKO EG&G, QCM922A) and a computerized electrochemical system (Hokuto Denko, HZ-7000). Li foil was used as a reference and counter electrode, respectively. SEI was formed at 0 V vs. Li|Li(I) on the Cu-coated EQCM electrode prior to the galvanostatic deposition of Li. EIS was performed at the open circuit potential (OCP) in the frequency range from 20 kHz to 1.0 Hz with an amplitude of 5 mV rms during SEI formation and after Li deposition.The resonance frequency (f) of a Cu-coated EQCM electrode gradually decreased with keeping the potential of the electrode at 0 V. A semicircle corresponding to the parallel circuit of the resistance and capacitance of SEI was observed in the Nyquist plots of the Cu electrode after keeping the potential at 0 V. The resistance of SEI, R SEI, increased with the lapse of time over 24 hours, indicating the formation and growth of SEI on the Cu electrode at 0 V in LiTFSA-G4 and LiFSA-G4. The changes in f and R SEI in LiFSA-G4 were smaller than those in LiTFSA-G4, suggesting the SEI formed in LiFSA-G4 was lighter and more conductive.The larger change in f of the Cu-EQCM electrode than that calculated from the passed electric charge implied that SEI was also formed on Li deposits during the galvanostatic cathodic reduction at 0.01 mA cm–2, as observed during cyclic voltammetry in LiTFSA-G3 [4]. Δf in LiFSA-G4 was smaller while the changes in the resonance resistance mainly reflecting the surface roughness of the electrode were close in the SILs, suggesting the reactivity against Li was low in LiFSA-G4. The electrode potential of the Cu-EQCM electrode changed over from that of Li(I)/Li after few tens of hours, suggesting the SEI formed in SILs did not suppress completely the reaction between Li and the electrolyte at the OCP. The transition time of the potential in LiFSA-G4 was longer than that in LiTFSA-G4, indicating Li deposits were more stable in LiFSA-G4.References H. Moon, et al., J. Phys. Chem. C, 118, 20246 (2014).S. Terada, et al., Aust. J. Chem., 72, 70 (2019).N. Serizawa, et al., J. Electrochem. Soc., 167, 110560 (2020).N. Serizawa, et al., J. Electrochem. Soc., 160, A1529 (2013). AcknowledgementThis study was supported in part by the JSPS KAKENHI (Grant No. 19K15682) and Advanced Low Carbon Technology Research and Development Program (ALCA, JPMJAL1301).