Novel mixtures of ether-functionalized ionic liquids and non-flammable methylperfluorobutylether as safe electrolytes for lithium metal batteries

Abstract

Two mixtures of ether-functionalized quaternary ammonium ILs and non-flammable methylperfluorobutylether were used as new safe electrolytes for lithium metal batteries, and compared with the corresponding pure IL electrolytes. The mixed electrolytes had lower viscosity and higher conductivity than the pure IL electrolytes. The test results of flammability and flash point indicated that the addition of methylperfluorobutylether didn't reduce the safety of the pure IL electrolytes. The mixed electrolytes exhibited slightly narrower electrochemical windows due to the decrease of the cathodic stability. The effects of lithium redox on the Ni electrode and cycle performance of the symmetric lithium cell were investigated, and the addition of MFE could optimize the passivation layer on the Ni electrode and the SEI film on the lithium metal. The Li/LiFePO₄ cells using the mixed electrolytes had good cycle performance at 0.1 C, and showed better rate performance than the cells using the pure IL electrolytes.

---

Introduction

Ionic liquids (ILs) have attracted great interest of researchers due to their superior properties, including non-flammability, low volatility, high thermal stability, and good electrochemical stability.^1–4 Based on these properties, ILs have shown potential as safe electrolytes for lithium metal batteries, which use lithium metal anodes with high theoretical capacity (more than 3860 mA h g⁻¹).^5–7 Many kinds of ILs, such as quaternary ammonium, piperidinium, pyrrolidinium, quaternary phosphonium and guanidinium ILs, have been reported as electrolytes for lithium metal batteries during the past several years.^5,8–12

However, the electrochemical performances of the pure IL electrolytes, especially the rate performance, are nonideal owning to their high viscosity and low conductivity compared to those of conventional organic carbonates-based electrolytes. The addition of low-viscosity and flammable organic carbonates to the pure IL electrolytes is an approach to improve the rate performance, explored by some research groups, and the results of flammability test indicate that the mixed electrolytes composed of ILs and organic carbonates appear to be non-flammable if the content of ILs reaches a certain threshold.^13–18 Actually, the addition of flammable organic solvents still impairs the safety of the pure IL electrolytes. The flash points of the mixed electrolytes containing ILs and flammable organic solvents are obviously lower than those of the pure IL electrolytes because the vapors volatilized from organic solvents at elevated temperature are easy to be ignited.^15,19 C. Arbizzani et al. also think that the safety of the mixed electrolytes based on ILs and organic carbonates is underestimated, and the ignited vapors of organic carbonates can trigger the combustion of ILs at high temperature.²⁰ Therefore, in order to improve the rate performance as well as retain the high safety of the pure IL electrolytes, an optimal method is to add non-flammable and low-viscosity solvents to the pure IL electrolytes.

Hydrofluoroethers (HFEs) are one kind of non-flammable fluorinated solvents, and possess some attractive features, such as low viscosity, low melting point and low surface tension.^21,22 HFEs have been used as flame-retardant component in the mixed electrolytes based on HFEs and organic carbonates, and this type of mixed electrolytes with high content of HFEs can show good electrochemical performance.^21–27 Though HFEs are an ideal choice to be mixed with ILs as electrolytes with high safety, the issue that common ILs have bad miscibility with HFEs due to low dielectric constant of HFEs, must be faced firstly. For example, at room temperature the miscibility of 1-ethyl-3-methyl imidazolium bis(trifluoromethanesulfonyl)imide (EMITFSI) with methylperfluorobutylether (MFE) is only about 90 : 10 (mass ratio), and the miscibility of EMITFSI with HFEs reduces as the molecule size of HFEs increases.²⁸

Recently, we were inspired by “like dissolves like theory”, and found that ether-functionalization in the cation of ILs would increase the miscibility of ILs with MFE which also contained one methyl ether group. N-2-Ethoxyethyl-N,N,N-tri-(2-methoxyethyl)ammonium bis(trifluoromethanesulfonyl)imide (N(2o1)₃(2o2)TFSI) and N,N-di-(2-methoxyethyl)-N,N-di-(2-ethoxyethyl)ammonium bis(trifluoromethanesulfonyl)imide (N(2o1)₂(2o2)₂TFSI), which had four ether groups in the quaternary ammonium cation, could be miscible with MFE when the mass ratios were 75 : 25 and 65 : 35, respectively. The structures of the two ether-functionalized ILs and MFE were illustrated in Fig. 1. In this work, the mixtures of the two ether-functionalized ILs and MFE were applied in lithium metal battery as new electrolytes, and compared with the corresponding pure IL electrolytes. Viscosity, conductivity, safety of electrolyte, electrochemical stability, behavior of lithium redox on Ni electrode, cycle performance of symmetric lithium cell, and charge–discharge characteristics of Li/LiFePO₄ cell were studied. As expected, the safety of the mixed electrolytes was similar to that of the pure IL electrolytes, and the Li/LiFePO₄ cells using the mixed electrolytes had better rate performance.

Fig. 1 Structures of ILs and nonflammable hydrofluoroether.

Experimental

Two ether-functionalized quaternary ammonium ILs were synthesized according to the reported method.²⁹ The structure and purity of ILs were confirmed by ¹H NMR and ¹³C NMR (Avance III 400), and chloroform-d was used as the solvent. The NMR data were dealt with by MestReC soft (Fig. S1–S4†). The two ILs were dried under high vacuum for more than 48 hours at 110 °C before used. The water contents of the dried ILs were detected by a moisture titrator (Metrohm 73KF coulometer) based on Karl-Fischer method, and the values were less than 30 ppm. MFE was purchased from TCI Co., Ltd and used as received. 0.6 M lithium bis(trifluoromethanesulfonyl)imide (LiTFSI, Morita Chemical Industries Corporation Ltd) was added into the dried ILs to obtain the pure IL electrolytes, and the mixed electrolytes were prepared by dissolving 0.6 M LiTFSI in the mixtures of ILs and MFE. This procedure was carried out in an argon-filled glove box. In order to identify the difference of safety, the mixed electrolytes of ILs and diethyl carbonate (DEC, a flammable organic carbonate) were also prepared.

Viscosity was measured by using viscometer (DV-III ULTRA, Brookfield Engineering Laboratories, Inc.), and conductivity was measured by using DDS-11A conductivity meter in a dry chamber. Flammability test was carried out as the previous report.¹³ A glass filter was immersing in electrolyte, and then exposed to the flame of an alcohol lamp. The distance between the glass filter and the wick of alcohol lamp was kept at 10 cm. The lamp would be removed after 10 s. If the electrolyte didn't ignite during the test time or if the ignition of electrolyte ceased when the lamp was removed, the electrolyte was considered to be nonflammable. Each electrolyte was tested three times. The pictures of flammability test for two mixed electrolytes were shown as examples (Fig. S5†). Flash point was measured by using a flashpoint tester with the temperature range from room temperature to 200 °C (closed cup, Shanghai SUN Scientific Instrument Co., Ltd).

Electrochemical window was evaluated by linear sweep voltammogram in the glove box. The working electrode was a glassy carbon disk (3 mm diameter), and lithium metal was used as both counter and reference electrodes. The plating and stripping behaviors of lithium in different electrolytes were examined by cyclic voltammogram in the glove box. The working electrode was a nickel disk (2 mm diameter), and lithium metal was used as both counter and reference electrodes. Glassy carbon electrode or Ni electrode was polished with alumina paste (d = 0.1 μm). And the polished electrode was washed with deionized water and dried under vacuum. LSV and CV tests were performed by CHI 660D electrochemistry workstation.

Symmetric lithium coin cell (CR2016 stainless steel cell) was assembled in the glove box, and the borosilicate glass (GF/A, Whatman) was used as separator. After the cell retained at open circuit for 48 h at room temperature, charge–discharge cycling was conducted by CT2001A cell test instrument (LAND Electronic Co., Ltd) under a current density of 0.1 mA cm⁻² (charge for 16 min and discharge for 16 min). Impedance response of symmetric cell was measured before and after cycling by CHI660D electrochemistry workstation (100 kHz–100 mHz, applied voltage 5 mV).

Coin cell was used to evaluate the performance of electrolyte in lithium battery. Lithium metal was used as anode, and its area was 2.0 cm². Cathode was fabricated by spreading the mixture of 80 wt% carbon coated LiFePO₄ (Zhangjiagang Guotai-Huarong Co., Ltd), 10 wt% acetylene black and 10 wt% PVDF onto Al current collector (battery use). Loading of active material was about 2.0 mg cm⁻² and this electrode was directly used without pressing. The separator was the borosilicate glass (GF/A from Whatman). After all the components of cell were dried under vacuum, cell construction was carried out in the glove box. CT2001A cell test instrument was used to perform the galvanostatic charge–discharge cycling test at room temperature. Current rate was determined by using the nominal capacity of 170 mA h g⁻¹ for Li/LiFePO₄ cell. Charge had one process: constant current at a rate, cut-off voltage of 4.2 V, and discharge also had one process: constant current, cut-off voltage of 2.5 V.

Results and discussion

Viscosity and conductivity were two important properties of electrolyte applied in lithium metal battery. The two values of ILs, MFE and different electrolytes at 25 °C were summarized in Table 1. Like many pure IL electrolytes in the previous reports,^30–32 the viscosity increased and the conductivity decreased remarkably after dissolving LiTFSI in the two ether-functionalized ILs, which was attributed to the decreased mobility of all the ionic species caused by the increase of anion concentration and the strong coulomb interaction between lithium ion and anion.³³ When the low-viscosity MFE was added to the two pure IL electrolytes, both viscosity and conductivity were improved. And the mixed electrolyte composed of 65 wt% N(2o1)₂(2o2)₂TFSI and 35 wt% MFE had lower viscosity and higher conductivity than the mixed electrolyte containing 75 wt% N(2o1)₃(2o2)TFSI and 25 wt% MFE.

Table 1 Viscosity and conductivity of ILs, MFE and electrolytes at room temperature

Solvent or electrolyte	Viscosity/mPa s	Conductivity/mS cm^−1
MFE	0.6	—
N(2o1)3(2o2)TFSI	90	1.10
N(2o1)2(2o2)2TFSI	87	1.10
0.6 M LiTFSI/N(2o1)3(2o2)TFSI	280	0.43
0.6 M LiTFSI/N(2o1)2(2o2)2TFSI	228	0.48
0.6 M LiTFSI/75 wt% N(2o1)3(2o2)TFSI + 25 wt% MFE	114	0.64
0.6 M LiTFSI/65 wt% N(2o1)2(2o2)2TFSI + 35 wt% MFE	66	0.77

---

Unlike the addition of organic carbonates,^14–16 the addition of MFE didn't help to increase obviously the conductivity of electrolyte. For example, the presence of 25 wt% MFE in the mixed electrolyte based on N(2o1)₃(2o2)TFSI only increased the conductivity by 49%, and the conductivity of the mixed electrolyte containing 20 wt% propylene carbonate and 80 wt% N-butyl-N-methylpyrrolidinium bis(trifluoromethanesulfonyl)imide (P14TFSI) was about 130% higher than the conductivity of P14TFSI's pure IL electrolyte.¹⁵ It meant that the effect mechanism of MFE on the conductivity should be different from that of organic carbonates. The coordination between IL anion and lithium ion resulted in the dissolution of lithium salt in the pure IL. Because MFE had a weak ability to solvate ions,²⁴ the dissolution of lithium salt in the mixture of IL and MFE still relied on the solvation of IL anion. The reduced concentrations of ions after adding MFE would lead to the weaker Coulomb interaction between cation and anion, which could enhance the mobility of ions and the conductivity of electrolyte. For organic carbonates with a strong ability to solvate ions, they helped to form the solvated ions with better mobility in the mixed electrolyte,³⁴ and it contributed to the obvious increasing of conductivity.

Concerning the viscosity, a marked decrease was observed after adding MFE, which was similar to the effect of adding organic carbonates in pure IL electrolytes.^14–16 The reduced concentrations of ions after adding non-ionic solvent was directly associated with the decrease of viscosity. Moreover, for the configurations of ion pairs or complexes could impact the viscosity of pure IL electrolyte, it was speculated that the change of the configurations of ion pairs or complexes as a result of adding MFE or organic carbonates, might also be beneficial for the decrease of viscosity.

In order to clarify the safety of electrolyte, the tests of flammability and flash point were carried out. According to Table 2, not only the mixed electrolytes based on ether-functionalized ILs and MFE but also the mixed electrolytes based on the two ILs and DEC were observed to be non-flammable as well as the pure IL electrolytes. For the mixed electrolytes composed of ILs and flammable organic solvents, it had been reported that ILs could act as flame retardants, and the mixed electrolytes appeared to be non-flammable when the content of ILs reached a certain threshold.^13–16,18 The accepted mechanism of flame retardance was that the addition of non-flammable ILs could decrease the vapor pressure of flammable organic solvents, thereby preventing the mixed electrolytes from being ignited.¹³ However, the risk of electrolytes with flammable organic solvents, especially at high temperature, wouldn't be eliminated completely by the addition of ILs because the flammable vapors from organic solvents still existed. And this kind of risk could be verified by the test of flash point. For instance, the pure IL electrolyte of P14TFSI showed no flash point until 200 °C, but the mixed electrolyte of 80 wt% P14TFSI and 20 wt% 1,3-dioxolane possessed a very low flash point (19 °C).¹⁹ Here, the flash points of the mixed electrolytes containing 25 or 35 wt% DEC were only 46 and 43 °C, respectively. In contrast, both the pure IL electrolytes and the mixed electrolytes containing MFE exhibited no flash point until 200 °C, indicating that the addition of non-flammable MFE couldn't weaken the safety of the pure IL electrolytes. Hence, the significant difference of flash point proved that the mixed electrolytes based on ILs and MFE had better safety compared with the mixed electrolytes composed of ILs and flammable organic solvents.

Table 2 Flammability and flash point (T_f) of different electrolytes

Electrolyte	Occurrence of ignition	Tf/°C
0.6 M LiTFSI/N(2o1)3(2o2)TFSI	0/3	>200
0.6 M LiTFSI/N(2o1)2(2o2)2TFSI	0/3	>200
0.6 M LiTFSI/75 wt% N(2o1)3(2o2)TFSI + 25 wt% MFE	0/3	>200
0.6 M LiTFSI/65 wt% N(2o1)2(2o2)2TFSI + 35 wt% MFE	0/3	>200
0.6 M LiTFSI/75 wt% N(2o1)3(2o2)TFSI + 25 wt% DEC	0/3	46
0.6 M LiTFSI/65 wt% N(2o1)2(2o2)2TFSI + 35 wt% DEC	0/3	43

---

The electrochemical stability of electrolytes was investigated by linear sweep voltammogram (LSV), and the LSV curves were shown in Fig. 2. For the two pure IL electrolytes, the anodic limiting potentials were 5.3 V versus Li/Li⁺, and the cathodic limiting potentials were 0.6 V versus Li/Li⁺. So their electrochemical windows were 4.7 V. For the two mixed electrolytes containing MFE, the anodic limiting potentials were also 5.3 V versus Li/Li⁺, manifesting that the addition of MFE could not change the anodic stability of electrolyte. The cathodic limiting potentials of the two mixed electrolytes were 0.8 V versus Li/Li⁺, which were higher than those of the two pure IL electrolytes. It was apparent that the addition of MFE decreased slightly the cathodic stability of electrolyte.

Fig. 2 LSV curves of these electrolytes at room temperature. Working electrode: glassy carbon; counter electrode: Li; reference electrode: Li; scan rate: 10 mV s⁻¹.

Considering the cathodic limiting potential of these four electrolytes which were higher than 0 V versus Li/Li⁺, it was very possible that these electrolytes didn't support deposition of lithium due to continuous electrochemical reduction of electrolytes themselves before deposition of lithium. Nevertheless, according to the curves of cyclic voltammogram (CV) shown in Fig. 3, the plating and stripping of lithium on Ni electrode could be observed clearly. For the pure IL electrolyte of N(2o1)₃(2o2)TFSI (Fig. 3a), in the first cycle the plating of lithium was at about −0.14 V versus Li/Li⁺ and the anodic peak at about 0.36 V versus Li/Li⁺ in the returning scan corresponded to the stripping of lithium. The lithium redox in this electrolyte might be caused by the generation of a certain passivation layer on the Ni electrode. The cathodic peaks of lithium decreased slightly with the cycle number, and it suggested that the passivation layer changed so that the plating of lithium was restrained. The cathodic peak at about 0.29 V versus Li/Li⁺ could be found in the first cycle. This cathodic peak might be assigned to the electrochemical reduction of electrolyte, and at the same time it could be presumed that this reduction might generate the passivation layer on Ni electrode. In the second and third cycles the current of this peak decreased, so it meant that the passivation layer generating in the first cycle also inhibited the reduction of electrolyte. The lithium redox behaviors in the pure IL electrolyte of N(2o1)₂(2o2)₂TFSI (Fig. 3b) were similar to those in the pure IL electrolyte of N(2o1)₃(2o2)TFSI. Furthermore, a cathodic peak in the range from 2.0 V to 1.5 V versus Li/Li⁺ and a cathodic peak in the range from 1.3 V to 0.8 V versus Li/Li⁺ appeared in the first cycle for the two pure IL electrolytes, which might be caused by the electrochemical reactions of the trace water and oxygen on the Ni electrode. And these peaks disappeared in the second and third cycles due to the passivation layer forming in the first cycle. This kind of cathodic peaks caused by the trace water and oxygen, could also be found in the reported results of CV experiments for other pure IL electrolytes.^35–37

Fig. 3 CV curves of four electrolytes at room temperature (−0.5 V to 2.5 V vs. Li/Li⁺): (a) 0.6 M LiTFSI/N(2o1)₃(2o2)TFSI; (b) 0.6 M LiTFSI/N(2o1)₂(2o2)₂TFSI; (c) 0.6 M LiTFSI/75 wt% N(2o1)₃(2o2)TFSI + 25 wt% MFE; (d) 0.6 M LiTFSI/65 wt% N(2o1)₂(2o2)₂TFSI + 35 wt% MFE. Working electrode: Ni; counter electrode: Li; reference electrode: Li; scan rate: 10 mV s⁻¹.

After comparing the mixed electrolytes (Fig. 3c and d) with the pure IL electrolytes, two distinct differences of CV curves were observed. Firstly, the peak currents of lithium redox were enhanced by adding MFE, and the increase of the peak currents exceeded the increase of conductivity. For example, the mixed electrolyte containing 65 wt% N(2o1)₂(2o2)₂TFSI and 35 wt% MFE had 60% higher conductivity than the pure IL electrolyte of N(2o1)₂(2o2)₂TFSI, and the cathodic peak current of lithium in the first cycle for the former (1.38 mA cm⁻²) was 146% bigger than that for the latter (0.56 mA cm⁻²). It suggested that the addition of MFE could bring benefit to the lithium redox by adjusting the constitution of the passivation layer on Ni electrode. Secondly, the cathodic peaks in the first cycle caused by the trace water and oxygen, were inconspicuous after adding MFE. Besides the reduced concentrations of these trace impurities, it was speculated that MFE might help to inhibit the electrochemical reactions of the trace water and oxygen on Ni electrode.

Cycling test of symmetric Li/electrolyte/Li cell combined with electrochemical impedance spectra (EIS) were often used to investigate the interfacial characteristics of electrolyte/lithium metal.^7,38–41 Firstly, the time evolution of the impedance response of symmetric lithium cell at open circuit was investigated as the previous reports,^29,42 and the results were presented in Fig. S6.† The diameter of the semicircle was associated to the interfacial resistance of electrolyte/lithium metal. For the four electrolytes, the interfacial resistance increased gradually from 0 h to 12 h, and then kept dynamic stability after 48 h, which implied that a stable passivation layer could form on lithium metal after that time. Next, cycling test was performed after symmetric lithium cell had stayed at open circuit for 48 h. If the voltage profile of cycling test fluctuated remarkably, this kind of behavior indicated the changing of morphology on lithium metal. According to the results of cycling test in Fig. 4, the voltage profiles of the two pure IL electrolytes didn't show the obvious fluctuating behavior during the 100 cycles, which implied that the changing of morphology on lithium metal could be restrained efficiently due to the forming of stable solid electrolyte interphase (SEI). The addition of MFE couldn't have a negative effect on the cycling tests of symmetric lithium cells, and the voltage profiles of the two mixed electrolytes still kept stable.

Fig. 4 Symmetrical lithium cell cycling and EIS results: 0.1 mA cm⁻² constant current and 100 cycles at room temperature; (a) and (e) 0.6 M LiTFSI/N(2o1)₃(2o2)TFSI, (b) and (f) 0.6 M LiTFSI/N(2o1)₂(2o2)₂TFSI, (c) and (g) 0.6 M LiTFSI/75 wt% N(2o1)₃(2o2)TFSI + 25 wt% MFE, (d) and (h) 0.6 M LiTFSI/65 wt% N(2o1)₂(2o2)₂TFSI + 35 wt% MFE.

The results of EIS before and after cycle test were also presented in Fig. 4. The intercept with real axis of the response at high frequency was assigned to the bulk resistance of electrolyte, and the diameter of the semicircle was associated to the interfacial resistance of electrolyte/lithium metal. The interfacial resistances of electrolyte/lithium metal before and after cycle test reflected the passivation layer and the SEI film on lithium metal, respectively. No matter whether the electrolytes contained MFE or not, the interfacial resistances of electrolyte/lithium metal before cycle test were close to each other. The interfacial resistance of electrolyte/lithium metal changed after cycling test, indicating that the initial passivation layer on lithium metal might turn into the SEI film during the process of cycling test. After cycling test the interfacial resistances of the two mixed electrolytes were lower than those of the two pure IL electrolytes, and the mixed electrolyte based on 65 wt% N(2o1)₂(2o2)₂TFSI and 35 wt% MFE had the lowest interfacial resistance. It was presumed that the electrochemical reduction of MFE might adjust the construction of SEI film on lithium metal, and the lower interfacial resistance of the two mixed electrolytes after cycling test suggested that the addition of MFE would optimize the SEI film.

The charge–discharge (C–D) test of Li/LiFePO₄ cell was performed to evaluate these four electrolytes. Fig. 5a showed the C–D curves of Li/LiFePO₄ cell using the mixed electrolyte composed of 65 wt% N(2o1)₂(2o2)₂ TFSI and 35 wt% MFE at 0.1 C. The initial discharge capacity of this mixed electrolyte was 158 mA h g⁻¹. And the discharge capacity increased with the cycle number, which could result from the gradual penetration of electrolyte into the LiFePO₄ cathode during the C–D processes. The discharge capacity was stable after 10 cycles, and the value retained about 162 mA h g⁻¹ until the 50th cycle. Fig. 5b illustrated the discharge capacity and Coulombic efficiency during cycling of Li/LiFePO₄ cells using these four electrolytes at 0.1 C. The discharge capacities of the two mixed electrolytes were slightly higher than those of the two pure IL electrolytes. And the discharge capacities of all the four electrolytes increased with the cycle number and stabilized after some cycles. Whereas, the discharge capacities of the two pure IL electrolytes needed more cycle number (15 cycles) to reach the stable values, indicating that the addition of MFE could improve the wettability of electrolyte to the LiFePO₄ electrode. The Coulombic efficiencies of the four electrolytes were stable after the initial several cycles, and the Coulombic efficiencies of the two mixed electrolytes (about 99.2%) were higher than those of the two pure IL electrolytes (about 98.6%).

Fig. 5 (a) C–D curves of Li/LiFePO₄ cell using 0.6 M LiTFSI/65 wt% N(2o1)₂(2o2)₂ TFSI + 35 wt% MFE, (b) discharge capacity and Coulombic efficiency during cycling of Li/LiFePO₄ cells, and (c) EIS plots of Li/LiFePO₄ cells after 15 C–D cycles at 0.1 C.

EIS was utilized to analyze how the interfacial property of electrode/electrolyte affected the electrochemical performance of Li/LiFePO₄ cell. The EIS measurement was carried out after 15 C–D cycles at 0.1 C in order to ensure complete penetration of different electrolytes into the LiFePO₄ cathode, and the EIS plots were presented in Fig. 5c. The diameter of the semicircle represented the interfacial resistance of the SEI film.^43,44 Obviously, the two mixed electrolytes showed smaller interfacial resistances than the two pure IL electrolytes, suggesting that the addition of MFE could improve the SEI film and decrease the interfacial resistance. And it just explained the higher discharge capacities of the two mixed electrolytes at 0.1 C. Considering the better Coulombic efficiencies of the two mixed electrolytes at 0.1 C, it was inferred that the SEI film changed by adding MFE might restrain the side reactions more effectively during the C–D processes.

The discharge capacity during cycling of Li/LiFePO₄ cells at high rate (1.0 C or 2.0 C) was illustrated in Fig. 6, and the C–D test at high rate was also carried out after 15 C–D cycles at 0.1 C so as to avoid the effect of electrolyte's wettability on the rate performance. As expected, the two mixed electrolytes had better rate performance than the two pure IL electrolytes, because of lower viscosity, higher conductivity and more beneficial SEI film. For example, at 1.0 C, the initial discharge capacities of the mixed electrolyte of N(2o1)₂(2o2)₂TFSI were 142 mA h g⁻¹, which was about 20 mA h g⁻¹ higher than that of the pure IL electrolyte of N(2o1)₂(2o2)₂ TFSI. At 2.0 C, the Li/LiFePO₄ cell using the pure IL electrolyte of N(2o1)₂(2o2)₂TFSI wasn't able to discharge normally, but the mixed electrolyte of N(2o1)₂(2o2)₂TFSI still showed the stable discharge capacity (more than 110 mA h g⁻¹) during the 100 C–D cycles. Furthermore, the discharge capacities of the mixed electrolyte containing 65 wt% N(2o1)₂(2o2)₂TFSI and 35 wt% MFE were higher than those of the mixed electrolyte containing 75 wt% N(2o1)₃(2o2)TFSI and 25 wt% MFE at 1.0 C and 2.0 C. It implied that higher content of MFE in the mixed electrolyte was beneficial for the rate performance due to more remarkable improvement of viscosity and conductivity. Compared with the mixed electrolytes of organic carbonates and ILs,^14–16,18 the mixed electrolytes of MFE and ether-functionalized ILs in this work possessed the similar rate performance of Li/LiFePO₄ cell when the content of MFE was close to the content of organic carbonates.

Fig. 6 Discharge capacity during cycling of Li/LiFePO₄ cells using four electrolytes at (a) 1.0 C and (b) 2.0 C.

Conclusions

Two mixtures of ether-functionalized quaternary ammonium ILs and non-flammable MFE were used as safe electrolytes for lithium metal battery. In contrast to the corresponding pure IL electrolytes, the addition of low-viscosity MFE could improve the viscosity and conductivity of electrolyte. The test results of flammability and flash point indicated that the addition of non-flammable MFE didn't weaken the safety of the pure IL electrolytes. Owning to the decreasing of the cathodic stability, the two mixed electrolytes had slightly narrower electrochemical windows than the two pure IL electrolytes. The addition of MFE could be beneficial for lithium redox behavior on Ni electrode, and help to reduce the interfacial resistance of electrolyte/lithium metal after the cycling test of symmetric lithium cell. The Li/LiFePO₄ cells using the two mixed electrolytes showed good cycle performance at 0.1 C, and had better rate and performance than the cells using the two pure IL electrolytes.

Acknowledgements

This work is financially supported by the National Natural Science Foundation of China (Grants no. 21103108, 21173148 and 21373136).

References

1. T. Welton, Chem. Rev., 1999, 99, 2071–2083.
2. J. Dupont, R. F. De Souza and P. A. Z. Suarez, Chem. Rev., 2002, 102, 3667–3692.
3. S. A. Forsyth, J. M. Pringle and D. R. MacFarlane, Aust. J. Chem., 2004, 57, 113–119.
4. M. Armand, F. Endres, D. R. MacFarlane, H. Ohno and B. Scrosati, Nat. Mater., 2009, 8, 621–629.
5. H. Sakaebe and H. Matsumoto, Electrochem. Commun., 2003, 5, 594–598.
6. M. Egashira, H. Todo, N. Yoshimoto, M. Morita and J. I. Yamaki, J. Power Sources, 2007, 174, 560–564.
7. H. Yoon, G. H. Lane, Y. Shekibi, P. C. Howlett, M. Forsyth, A. S. Best and D. R. MacFarlane, Energy Environ. Sci., 2013, 6, 979–986.
8. H. Matsumoto, H. Sakaebe, K. Tatsumi, M. Kikuta, E. Ishiko and M. Kono, J. Power Sources, 2006, 160, 1308–1313.
9. H. Sakaebe, H. Matsumoto and K. Tatsumi, J. Power Sources, 2005, 146, 693–697.
10. S. Seki, Y. Ohno, H. Miyashiro, Y. Kobayashi, A. Usami, Y. Mita, N. Terada, K. Hayamizu, S. Tsuzuki and M. Watanabe, J. Electrochem. Soc., 2008, 155, A421–A427.
11. K. Tsunashima, F. Yonekawa and M. Sugiya, Chem. Lett., 2008, 37, 314–315.
12. S. Fang, L. Yang, J. Wang, H. Q. Zhang, K. Tachibana and K. Kamijima, J. Power Sources, 2009, 191, 619–622.
13. H. Nakagawa, Y. Fujino, S. Kozono, Y. Katayama, T. Nukuda, H. Sakaebe, H. Matsumoto and K. Tatsumi, J. Power Sources, 2007, 174, 1021–1026.
14. A. Guerfi, M. Dontigny, P. Charest, M. Petitclerc, M. Lagacé, A. Vijh and K. Zaghib, J. Power Sources, 2010, 195, 845–852.
15. R. S. Kühnel, N. Böckenfeld, S. Passerini, M. Winter and A. Balducci, Electrochim. Acta, 2011, 56, 4092–4099.
16. I. Quinzeni, S. Ferrari, E. Quartarone, C. Tomasi, M. Fagnoni and P. Mustarelli, J. Power Sources, 2013, 237, 204–209.
17. H. F. Xiang, B. Yin, H. Wang, H. W. Lin, X. W. Ge, S. Xie and C. H. Chen, Electrochim. Acta, 2010, 55, 5204–5209.
18. B. Yang, C. Li, J. Zhou, J. Liu and Q. Zhang, Electrochim. Acta, 2014, 148, 39–45.
19. S. Xiong, J. Scheers, L. Aguilera, D.-H. Lim, K. Xie, P. Jacobsson and A. Matic, RSC Adv., 2015, 5, 2122–2128.
20. C. Arbizzani, G. Gabrielli and M. Mastragostino, J. Power Sources, 2011, 196, 4801–4805.
21. J. Arai, J. Electrochem. Soc., 2003, 150, A219–A228.
22. K. Naoi, E. Iwama, N. Ogihara, Y. Nakamura, H. Segawa and Y. Ino, J. Electrochem. Soc., 2009, 156, A272–A276.
23. K. Naoi, E. Iwama, Y. Honda and F. Shimodate, J. Electrochem. Soc., 2010, 157, A190–A195.
24. J. Arai, J. Appl. Electrochem., 2002, 32, 1071–1079.
25. J. Arai, J. Power Sources, 2003, 119–121, 388–392.
26. M. Morita, T. Kawasaki, N. Yoshimoto and M. Ishikawa, Electrochemistry, 2003, 71, 1067–1069.
27. G. Nagasubramanian and K. Fenton, Electrochim. Acta, 2013, 101, 3–10.
28. M. B. Shiflett and A. Yokozeki, J. Chem. Eng. Data, 2007, 52, 2413–2418.
29. S. Fang, Y. Jin, L. Yang, S. I. Hirano, K. Tachibana and S. Katayama, Electrochim. Acta, 2011, 56, 4663–4671.
30. Y. Saito, T. Umecky, J. Niwa, T. Sakai and S. Maeda, J. Phys. Chem. B, 2007, 111, 11794–11802.
31. T. Frömling, M. Kunze, M. Schönhoff, J. Sundermeyer and B. Roling, J. Phys. Chem. B, 2008, 112, 12985–12990.
32. K. Hayamizu, S. Tsuzuki, S. Seki, Y. Ohno, H. Miyashiro and Y. Kobayashi, J. Phys. Chem. B, 2008, 112, 1189–1197.
33. S. Fang, Y. Tang, X. Tai, L. Yang, K. Tachibana and K. Kamijima, J. Power Sources, 2011, 196, 1433–1441.
34. S. Y. Lee, H. H. Yong, Y. J. Lee, S. K. Kim and S. Ahn, J. Phys. Chem. B, 2005, 109, 13663–13667.
35. H. Matsumoto, H. Sakaebe, K. Tatsumi, M. Kikuta, E. Ishiko and M. Kono, J. Power Sources, 2006, 160, 1308–1313.
36. J. Xu, J. Yang, Y. NuLi, J. Wang and Z. Zhang, J. Power Sources, 2006, 160, 621–626.
37. H. Matsumoto, H. Sakaebe and K. Tatsumi, J. Power Sources, 2005, 146, 45–50.
38. A. S. Best, A. I. Bhatt and A. F. Hollenkamp, J. Electrochem. Soc., 2010, 157, A903–A911.
39. G. H. Lane, A. S. Best, D. R. MacFarlane, A. F. Hollenkamp and M. Forsyth, J. Electrochem. Soc., 2010, 157, A876–A884.
40. A. Basile, A. F. Hollenkamp, A. I. Bhatt and A. P. O'Mullane, Electrochem. Commun., 2013, 27, 69–72.
41. A. I. Bhatt, A. S. Best, J. Huang and A. F. Hollenkamp, J. Electrochem. Soc., 2010, 157, A66–A74.
42. Y. Jin, J. Zhang, J. Song, Z. Zhang, S. Fang, L. Yang and S. I. Hirano, J. Power Sources, 2014, 254, 137–147.
43. N. Byrne, P. C. Howlett, D. R. MacFarlane and M. Forsyth, Adv. Mater., 2005, 17, 2497–2501.
44. J. Jin, H. H. Li, J. P. Wei, X. K. Bian, Z. Zhou and J. Yan, Electrochem. Commun., 2009, 11, 1500–1503.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra01713k

---