High‐Polarity Fluoroalkyl Ether Electrolyte Enables Solvation‐Free Li+ Transfer for High‐Rate Lithium Metal Batteries

Abstract Lithium metal batteries (LMBs) have aroused extensive interest in the field of energy storage owing to the ultrahigh anode capacity. However, strong solvation of Li+ and slow interfacial ion transfer associated with conventional electrolytes limit their long‐cycle and high‐rate capabilities. Herein an electrolyte system based on fluoroalkyl ether 2,2,2‐trifluoroethyl‐1,1,2,3,3,3‐hexafluoropropyl ether (THE) and ether electrolytes is designed to effectively upgrade the long‐cycle and high‐rate performances of LMBs. THE owns large adsorption energy with ether‐based solvents, thus reducing Li+ interaction and solvation in ether electrolytes. With THE rich in fluoroalkyl groups adjacent to oxygen atoms, the electrolyte owns ultrahigh polarity, enabling solvation‐free Li+ transfer with a substantially decreased energy barrier and ten times enhancement in Li+ transference at the electrolyte/anode interface. In addition, the uniform adsorption of fluorine‐rich THE on the anode and subsequent LiF formation suppress dendrite formation and stabilize the solid electrolyte interphase layer. With the electrolyte, the lithium metal battery with a LiFePO4 cathode delivers unprecedented cyclic performances with only 0.0012% capacity loss per cycle over 5000 cycles at 10 C. Such enhancement is consistently observed for LMBs with other mainstream electrodes including LiCoO2 and LiNi0.5Mn0.3Co0.2O2, suggesting the generality of the electrolyte design for battery applications.

exposure and oxidation in air. FTIR instrument (Nicolet, is50, U.S.A.) was conducted to study the Li anodes surface properties. Electrolyte viscosity was measured using a rheometer (Waters, VIS403039900, U.S.A.). The contact angle of the electrolytes on PP separator and various electrodes was tested by an automatic contact angle measuring instrument (ZhongChen, JC2000D3M, China). The thermal stability of various electrolytes was tested by DSC (NETZSCH, STA 449F3, Germany). OCV curves were tested with a heating rate of 5 °C/min. Electrolyte uptake was tested with the PP separator. Contact angle measurement for different electrolytes was based on the PP separator. Ignition tests for different electrolytes were based on glass-fiber films.
The Li + transference number was defined as the equation (Equation (1)).
where + is the mobility of Li + and − is the mobility of TFSI -.
The conductivity inside electrolyte was defined as the equation (Equation (2)).
where + and − are the numbers of anion and cation per unit volume, respectively, + and − are the mobilities of anion and cation, respectively, and is electronic charge.
The conductivity at electrolyte/anode interface in the 60%THE electrolyte was defined as the equation (Equation (3)).

= (1 − ) 1 + (3)
where 1 is the ion conductivity at electrolyte/anode interface in commercial electrolyte, is the ion conductivity of LiF and denotes as the increased ratio of F element after introducing THE.
The Li + transference numbers of various electrolytes were studied with AC impedance and DC polarization analysis. The polarization currents, referred to the initial current (I 0 ) and steady-state current (I S ) of the Li/Li cell, were obtained with a polarization potential (ΔV) at 10 mV. In addition, the initial and steady-state interfacial resistances (R O and R S ) of Li/electrolyte were determined through the impedance measurements before and after the potentiostatic polarization. The impedance measurements were performed at an open-circuit potential in a frequency range from 0.10 Hz to 1.0 MHz.
t Li + was calculated according to the Bruce-Vincent-Evans equation 1 (Equation (4)): The equation 1 (Equation (5)) of the conductivity inside electrolyte was given: where L is the thickness of the separator and S is the area of the electrode. Re is the Electrolyte uptake test: The electrolyte uptake was calculated through the following equation (6).
where and denote the weight of wet and dry separator, respectively.
ρ represents the electrolyte density. To eliminate the influence of different electrolyte densities, the mass of dry separator was multiplied with the electrolyte density to obtain the corresponding reference mass. In every 5 minutes, the mass of the wet separator with adsorbed electrolyte was weighed and divided by the reference mass to obtain the percentage of electrolyte uptake. The density calculation of various electrolytes was as follows: ρ 60%THE electrolyte = m 60%THE electrolyte / V 60%THE electrolyte = m LiTFSI + m THE + m DOL + m DME / V 60%THE electrolyte = m LiTFSI + ρ THE V THE + ρ DOL V DOL + ρ DME V DME / V 60%THE electrolyte = 0.2871 + 1.5398×0.6 + 1.0655×0.2 + 0.8665×0.2 / 1 = 1.5974 g/cm 3 ρ DOL+DME electrolyte = m DOL+DME electrolyte / V DOL+DME electrolyte = m LiTFSI + m DOL + m DME / V DOL+DME electrolyte = m LiTFSI + ρ DOL V DOL + ρ DME V DME / V DOL+DME electrolyte = 0.2871 + DOL-CH 3 CH 2 CH 3 , DME-CH 3 CH 2 CH 3 , THE-oxygen radical, DOL-oxygen radical, DME-oxygen radical, EC-oxygen radical, and DMC-oxygen radical were built and calculated. The adsorption energy was calculated with the equation: where E is the adsorption energy, and E a-b is the total energy of the relaxed a and b models at the equilibrium state. E a and E b are the self-consistent field (SCF) calculation energy values of geometry-optimized a and b models. Electron exchange correlation was constructed by Perdew-Burke-Ernzerhof (PBE) function with generalized gradient approximation (GGA). The calculation of binding energy between Al 3+ and different solvents adopted the same calculation methods and conditions as described above.
The breaking energy of C-F bond was calculated through a DMol 3 package with the , where E C-F is the breaking energy of C-F bond, and E electrolyte(e) is the electrolyte molecule system with a single electron. E electrolyte(-F) and E F are the energy values of geometry-optimized electrolyte molecule with C-F bond cleavage and F anion, respectively.

AIMD calculations.
Three electrolyte cells-a single LiPF 6 molecule dissolved in 7 EC and 7 DMC molecules, a single LiTFSI molecule dissolved in 5 DME and 7 DOL molecules, and a single LiTFSI molecule dissolved in 2 DME, 3 DOL and 4 THE molecules-were constructed in periodic boxes. We performed the AIMD calculation using the VASP package to understand Li + diffusion behavior. The ion-electron interaction was described with the projector augmented wave method, and the exchange-correlation energy was described by the functional of the Perdew-Burke-Ernzerhof form of the generalized gradient approximation. The plane wave energy cut-off of 400 eV was chosen and a minimal Г-centred 1 × 1 × 1 k-point grid was used.
All molecular dynamics simulations were performed in the NVT ensemble using a Nosé−Hoover thermostat. Each system was heated to 300 K and equilibrated for 10 ps and then simulated for 30 ps to obtain statistics.
Constrained AIMD simulations were also performed on the electrolyte/electrode systems to understand rate performance. The deposited Li (0 0 1)/electrolyte interfaces were modelled by packing the DOL, DME and THE in the deposited Li (0 0 1) box, respectively. In these simulations, a slow-growth method was used, in which a Li + within the electrolyte was chosen and constrained to a position from the deposited Li (0 0 1) surface. The force required to constrain the Li + at this particular position was monitored. The shifted distance was 9.92×10 -4 Å in every simulation step for approaching from the electrolyte to the surface, and the free energy of the system can be obtained by integrating the position dependent mean constraint force. ).
The inferior ionic conductivity (2.9 mS cm -1 for the 80%THE electrolyte) limits the rate capacity, when the volume fraction of THE exceeds 60%; while the low Li + transference numbers (0.48 and 0.53 for 20%THE and 40%THE electrolytes, respectively) become the limiting factor, when the volume fraction of THE falls below 60%. Therefore, the 60%THE electrolyte leads to the highest capacity compared with other concentrations.

Supplementary
where E a is the pseudo-activation energy, σ 0 is the pre-exponential factor, T 0 is the ideal glass transition temperature, and R is the gas constant.