Switching Electrolyte Interfacial Model to Engineer Solid Electrolyte Interface for Fast Charging and Wide‐Temperature Lithium‐Ion Batteries

Abstract Engineering the solid electrolyte interphase (SEI) that forms on the electrode is crucial for achieving high performance in metal‐ion batteries. However, the mechanism of SEI formation resulting from electrolyte decomposition is not fully understood at the molecular scale. Herein, a new strategy of switching electrolyte to tune SEI properties is presented, by which a unique and thinner SEI can be pre‐formed on the graphite electrode first in an ether‐based electrolyte, and then the as‐designed graphite electrode can demonstrate extremely high‐rate capabilities in a carbonate‐based electrolyte, enabling the design of fast‐charging and wide‐temperature lithium‐ion batteries (e.g., graphite | LiNi0.6Co0.2Mn0.2O2 (NCM622)). A molecular interfacial model involving the conformations and electrochemical stabilities of the Li+‐solvent‐anion complex is presented to elucidate the differences in SEI formation between ether‐based and carbonate‐based electrolytes, then interpreting the reason for the obtained higher rate performances. This innovative concept combines the advantages of different electrolytes into one battery system. It is believed that the switching strategy and understanding of the SEI formation mechanism opens a new avenue to design SEI, which is universal for pursuing more versatile battery systems with greater stability.

Electrolyte preparation. All reagents were used directly without further purification. The stoichiometric ratio of lithium salt and solvent was calculated based on the molar concentration and then used to prepare the targeted electrolyte. Typically, the solvent was added to the brown bottle first, and then the lithium salt was then added under stirring to form a clear electrolyte solution. The entire process was handled carefully in the argon-filled glovebox, where the moisture and oxygen contents are controlled at about 0.5 ppm. The ionic conductivity was measured with a conductivity meter (Five Easy PlusTM-FE38, Mettler Toledo Co., Ltd) at room temperature.
Electrode preparation. The NCM powders, conductive materials (3.5 wt.% C45 and 1.5 wt.% KS-6), and poly (vinylidene fluoride) (PVDF) were mixed with the weight ratio of 92:5:3 in N-methyl pyrrolidinone (NMP). While for graphite anodes, the graphite powders, conductive carbon (SP), carboxymethyl cellulose (CMC), and styrene-butadiene rubber (SBR) were mixed with the weight ratio of 94.5:1.5:1.5:2.5 in water. The mixtures were milled using a Hasai planetary mixer for 5 min. Then, the uniform slurry was coated on the aluminum and copper foil, respectively. Finally, the NCM and graphite electrodes were dried at 120 °C and 80 °C in vacuum for 10 h, respectively. The mass loadings of the cathode and anode were about 8 mg cm -2 and 5.8 mg cm -2 , respectively.
Electrochemical measurements. All batteries were assembled using the 2032-type coin cell and disassembled in an argon-filled glovebox. The graphite electrode performance was tested in the Li | graphite half-cell, in which the ether-based electrolyte (i.e., 1.0 M LiTFSI, 0.4 M LiNO3 in DOL) or carbonate-based electrolyte (i.e., 1.0 M LiPF6 in EC/EMC (v/v, 3/7)) was used. For convenience, we label the carbonate-based electrolyte as EC/EMC and the etherbased electrolyte as DOL. The cut-off voltage was set at 0.01-3.0 V. The graphite@SEI electrode was obtained as below. The Li | graphite half-cell was cycled for three cycles, where the graphite@SEI was formed first and then taken out after disassembling the cell. The graphite@SEI was washed and dried carefully, which was then tested using different kinds of electrolytes (i.e., switching the electrolyte), for example, from the ether-based electrolyte to carbonate-based electrolyte, or from the carbonate-based electrolyte to ether-based electrolyte.
Besides, the graphite@SEI | NCM622 full battery was assembled and tested, in which the N/P ratio (i.e., the total capacity ratio of graphite@SEI/NCM622) was controlled at around 1.2 and the carbonate-based electrolyte was used. The cut-off voltage was set at 2.75-4.25 V. The high-temperature cycling test of the batteries was conducted in a constant temperature oven at 45 °C. The low-temperature cycling test of the batteries was performed in a low-temperature incubator at minus -10°C, which were pre-cycled three times at room temperature (25°C) before the low-temperature cycling. All galvanostatic charge/discharge curves were recorded by the Neware instrument. The electrochemical impedance spectroscopy (EIS) measurements were tested by the electrochemical station of Bio-Logic VMP3. The in-situ EIS test was performed on the entire discharge process, where the potential was decreased from the opencircuit voltage (OCV) to 0.01 V gradually by the interval of 5 mV. In each test, the sinusoidal AC perturbation of 5 mV over the frequency range from 100 kHz to 10 MHz was adopted.
The GITT method was used to measure DLi+ (testing at 0.1 A g −1 , pulse time 20 min, relaxation time 30 min).

Materials characterizations.
The morphology of electrodes was observed by scanning electron microscopy (SEM, Hitachi S-4800), while their structural characteristics were observed via transmission electron microscopy (TEM, FEI Tecnai G2 F20). The XPS spectra of the graphite and NCM electrodes were measured by X-ray photoelectron spectroscopy (XPS, ESCALABMKLL) with Al Kα radiation, which emits 1.4866 keV X-ray with the corresponding wavelength of 8.53 Å.
Theoretical simulation. The binding energy and molecular orbitals were studied based on the gas phase calculations and the implicit solvent models using the Gaussian09 package.