An Ionic Liquid Electrolyte Additive for High-Performance Lithium–Sulfur Batteries

With the development of mobile electronic devices, there are more and more requirements for high-energy storage equipment. Traditional lithium-ion batteries, like lithium–iron phosphate batteries, are limited by their theoretical specific capacities and might not meet the requirements for high energy density in the future. Lithium–sulfur batteries (LSBs) might be ideal next-generation energy storage devices because they have nearly 10 times the theoretical specific capacities of lithium-ion batteries. However, the severe capacity decay of LSBs limits their application, especially at high currents. In this study, an ionic liquid (IL) electrolyte additive, TDA+TFSI, was reported. When 5% of the TDA+TFSI additive was added to a traditional ether-based organic electrolyte, the cycling performance of the LSBs was significantly improved compared with that of the LSBs with the pure traditional organic electrolyte. At a rate of 0.5 C, the discharge specific capacity in the first cycle of the LSBs with the 5% TDA+TFSI electrolyte additive was 1167 mAh g−1; the residual specific capacities after 100 cycles and 300 cycles were 579 mAh g−1 and 523 mAh g−1, respectively; and the average capacity decay rate per cycle was only 0.18% in 300 cycles. Moreover, the electrolyte with the TDA+TFSI additive had more obvious advantages than the pure organic ether-based electrolyte at high charge and discharge currents of 1.0 C. The residual discharge specific capacities were 428 mAh g−1 after 100 cycles and 399 mAh g−1 after 250 cycles, which were 13% higher than those of the LSBs without the TDA+TFSI additive. At the same time, the Coulombic efficiencies of the LSBs using the TDA+TFSI electrolyte additive were more stable than those of the LSBs using the traditional organic ether-based electrolyte. The results showed that the LSBs with the TDA+TFSI electrolyte additive formed a denser and more uniform solid electrolyte interface (SEI) film during cycling, which improved the stability of the electrochemical reaction.


Introduction
With the development of mobile electronic equipment and clean energy vehicles, traditional lithium-ion batteries can no longer meet people's demand for future energy storage equipment because of their low theoretical specific capacity and energy density.Lithium-sulfur batteries (LSBs) have become next-generation energy storage equipment because of their extremely high theoretical specific capacity (1675 mAh g −1 ) and energy density (2600 wh kg −1 ) [1,2].At present, the electrolyte in LSBs generally uses organic ethers, such as 1,3-dioxolane (DOL), 1,2-dimethoxyethane (DME), or tetraethylene glycol dimethyl ether (TEGDME) [3,4].Organic ether-based electrolytes have the advantages of low prices, high ionic conductivities, and low viscosities, but they have many problems, such as a severe polysulfide shuttle effect [5,6], which will lead to the actual discharge specific capacity of the LSBs being lower than the theoretical discharge specific capacity and the charge-discharge not being stable enough during long-term cycling.Ionic liquids (ILs) have many advantages, like low polysulfide solubility [7,8], high ionic conductivity [9,10], and non-flammability [11][12][13].Therefore, many kinds of ILs have been reported as electrolytes [14][15][16].For example, Jun-Woo Park et al. reported a room-temperature IL N,Ndiethyl-N-methyl-N-(2-methoxyethyl) ammonium bis(trifluoromethanesulfonyl) amide ([DEME] [TFSI]) [17].When the [DEME] [TFSI] IL was used to replace the ether-based electrolyte TEGDME, the battery had a higher discharge specific capacity of 600 mAh g −1 after 100 cycles at 1/12 C and a more stable Coulombic efficiency, nearly 100%.This can be attributed to the weak Lewis alkalinity of TFSI − anions and the low donor capacity of the room-temperature IL, resulting in the poor solubility of polysulfides.Yang et al. reported an IL N-methoxyethyl-N-methylpyrrolidinium bis(trifluoromethanesulfonyl)imide (Pyr 1,2O1 TFSI)-based electrolyte and used it in LSBs [18].Pyr 1,2O1 TFSI and DOL were mixed at a ratio of 65:35 (w/w) to obtain a mixed electrolyte.The LSBs assembled with the mixed electrolyte had a good capacity performance of 650 mAh g −1 after 100 cycles at 0.2 C, and the Coulombic efficiency was higher than 99%, which was owing to the restrained polysulfide dissolution in the electrolyte.However, there are few reports on the use of ILs as electrolyte additives.Additives account for a small part of the electrolyte, but they can have a great impact on the performance of the battery [19,20].Lithium nitrate is a common additive [21,22], but it is hard to maintain the stability of the surface film during cycling owing to the lack of toughness of the film containing inorganic substances [23].So many salts and organic compounds have been used as additives in electrolytes [24][25][26][27][28][29].For instance, Vu et al. reported aluminum phosphate (AlPO 4 ) as an additive in a mixed liquid electrolyte; the specific capacity of LSBs containing 1.0 wt.% AlPO 4 remained at 453 mAh g −1 after 100 cycles at 1.0 C [30].The authors proved that the improved performance could be attributed to the adsorption of lithium polysulfide on the surface of the AlPO 4 .Lu et al. reported lithium bis(fluorosulfonyl)imide (LiFSI) as a functional additive in the fluorinated electrolyte for rechargeable LSBs.The electrolyte with 1.0 wt.% of the LiFSI additive delivered a high initial specific capacity of 1254 mAh g −1 and a capacity retention of up to 78% after 50 cycles at 0.2 C [31].The additive component improved the ion transport and slightly restrained the polysulfide dissolution of the electrolyte.The anions of lithium salts can form a stable SEI film, but few studies have been conducted on the capture of additives forming SEI films.
In this work, we report a kind of IL named TDA + TFSI as an electrolyte additive, which was synthesized by the quaternization reaction of tris(dioxa-3,6-heptyl) anime (TDA) and hydrochloric acid (HCl) followed by anion exchange with lithium bis(trifluoromethanesulfonyl)imide (LiTFSI), as shown in Scheme 1.The electrochemical performances of the LSBs using electrolytes with different TDA + TFSI additive contents were tested.When 5% of the TDA + TFSI additive was added to the ether-based organic electrolyte, the cycling performance of the LSBs was significantly improved.The LSBs with 5% of the TDA + TFSI electrolyte additive exhibited high specific capacities of 1167 and 924 mAh g −1 at 0.5 and 1.0 C, respectively, and good capacity retentions of 579 and 428 mAh g −1 after 100 cycles and 523 and 399 mAh g −1 after 300 cycles, respectively, at 25 • C. At the same time, the Coulombic efficiency of the LSBs with 5% of the TDA + TFSI electrolyte additive was more stable than that of the LSBs with the traditional organic ether-based electrolyte.The results of the SEM and electrochemical tests suggest that the LSBs with 5% of the TDA + TFSI electrolyte additive formed a denser and more uniform SEI film during cycling, which improved the stability of the electrochemical reaction.Scheme 1. Synthesis routine of TDA + TFSI.

Fabrication of IL Additive TDA + TFSI
TDA (17.97 g, 0.056 mol) and HCl (30.6 mL, 2 M) were mixed in a 100 mL round bottom flask and stirred at 40 °C for 24 h.Then, the solvent and redundant HCl were removed by rotary evaporation.Liquid products (19.75 g, 0.055 mol) were obtained and named as TDA + Cl.After that, 17.33 g (0.060 mol) of LiTFSI and 10 mL of deionized water were added to the TDA + Cl, and the mixture was stirred for 12 h and then layered.The lower liquid layer was collected as the final product, TDA + TFSI.The presence of Cl − ions in the TDA + TFSI was checked by adding a AgNO3 aqueous solution to a small portion of the TDA + TFSI.No precipitate of AgCl was observed, indicating that the TDA + TFSI did not contain any Cl − .The 4A molecular sieves were used to remove any trace of water so that the TDA + TFSI could be used as an electrolyte additive in LSBs.The TDA + TFSI was stored in a glovebox before use.To confirm that there was no trace of water in the TDA + TFSI, after the 4A molecular sieve was added, a piece of Li metal was directly placed in the TDA + TFSI to make sure that there was no reaction between the Li metal and possible trace of water.

Preparation of C/S Complex and Cathode
The C/S complex was prepared through the melt-infiltration method [32][33][34].Specifically, 210 mg of sublimed sulfur powder and 90 mg of Super P were mixed and ground in a mortar.Then, the mixture was put into a pipe furnace and heated to 150 °C at a heating rate of 5 °C min −1 .After being held at 150 °C for 6 h to ensure the sulfur powder had infiltrated the pores of the Super P, the obtained C/S complex was slowly cooled to room temperature.The cathode was made as follows: the C/S complex, Super P, and PVDF, in a weight ratio of 7:2:1, were dissolved in NMP to form a uniform slurry and then the slurry was coated on aluminum foil.The cathode was dried under vacuum at 70 °C for 12 h and then transferred to a glovebox before use.The diameter and area of the aluminum foil Scheme 1. Synthesis routine of TDA + TFSI.

Fabrication of IL Additive TDA + TFSI
TDA (17.97 g, 0.056 mol) and HCl (30.6 mL, 2 M) were mixed in a 100 mL round bottom flask and stirred at 40 • C for 24 h.Then, the solvent and redundant HCl were removed by rotary evaporation.Liquid products (19.75 g, 0.055 mol) were obtained and named as TDA + Cl.After that, 17.33 g (0.060 mol) of LiTFSI and 10 mL of deionized water were added to the TDA + Cl, and the mixture was stirred for 12 h and then layered.The lower liquid layer was collected as the final product, TDA + TFSI.The presence of Cl − ions in the TDA + TFSI was checked by adding a AgNO 3 aqueous solution to a small portion of the TDA + TFSI.No precipitate of AgCl was observed, indicating that the TDA + TFSI did not contain any Cl − .The 4A molecular sieves were used to remove any trace of water so that the TDA + TFSI could be used as an electrolyte additive in LSBs.The TDA + TFSI was stored in a glovebox before use.To confirm that there was no trace of water in the TDA + TFSI, after the 4A molecular sieve was added, a piece of Li metal was directly placed in the TDA + TFSI to make sure that there was no reaction between the Li metal and possible trace of water.

Preparation of C/S Complex and Cathode
The C/S complex was prepared through the melt-infiltration method [32][33][34].Specifically, 210 mg of sublimed sulfur powder and 90 mg of Super P were mixed and ground in a mortar.Then, the mixture was put into a pipe furnace and heated to 150 • C at a heating rate of 5 • C min −1 .After being held at 150 • C for 6 h to ensure the sulfur powder had infiltrated the pores of the Super P, the obtained C/S complex was slowly cooled to room temperature.The cathode was made as follows: the C/S complex, Super P, and PVDF, in a weight ratio of 7:2:1, were dissolved in NMP to form a uniform slurry and then the slurry was coated on aluminum foil.The cathode was dried under vacuum at 70 • C for 12 h and then transferred to a glovebox before use.The diameter and area of the aluminum foil were 12 mm and 1.13 cm 2 , respectively, and the sulfur content of the cathode was 0.90 ± 0.10 mg cm −2 .

Physical Characterization
1 H-NMR spectra were recorded on a Bruker 400 MHz spectrometer (Bruker Corp., Karlsruhe, Germany) with chloroform-d (CDCl 3 ) containing 3% tetramethylsilane (TMS) as the solvent.The sample concentration was approximately 15 mg mL −1 .Fourier-transform infrared (FTIR) spectra were recorded using a Bruker TENSOR II spectrometer (Bruker Corp., Karlsruhe, Germany).The mass spectrum of the TDA + TFSI was recorded on an Agilent quadrupole time-of-flight LC/MS G6545 mass spectrometer (Agilent, Santa Clara, CA, USA).The surface morphology of the lithium metal was observed using a Hitachi S-4800 scanning electron microscope (SEM, Hitachi, Ltd., Tokyo, Japan).The batteries were disassembled in a glovebox, and the lithium metal was cleaned with DOL/DME (volumetric ratio: 1/1) to remove the remaining electrolyte.

Electrochemical Characterization
CR2032-type coin cells were assembled in an Ar-filled glovebox for electrochemical characterizations and charge-discharge cycle tests.The separator used glass microfiber filters purchased from Whatman.The surface morphology and thickness of the glass fiber separator were characterized using SEM (Figure S1a,b, Supporting Information).The electrolyte/sulfur (E/S) ratio was 100 µL mg −1 .The charge-discharge cycling performance of the LSBs with different TDA + TFSI additive contents was characterized using the LAND battery testing system (CT2001A, Wuhan LAND Electronic Co., Ltd., Wuhan, China) at 25 • C, and the voltage range was 1.8-3.0V.The electrochemical stability window of the TDA + TFSI was measured using an electrochemical workstation (CHI660E, CH Instruments Inc., Shanghai, China).Linear sweep voltammetry measurements were performed from the open-circuit voltage to 5.0 V and 1.0 V vs. Li/Li + at a potential scan rate of 1 mV s −1 .The conductivities (σ) of the electrolytes with different TDA + TFSI additive contents were measured using the same instrument and a CR2032-type coin cell, of which both the anode and cathode were stainless-steel pistons.The impedances of the lithium symmetrical coil cells and full coil cells were measured using the same instrument in the frequency range from 100 mHz to 100 kHz at a perturbation amplitude of 5 mV.

Preparation of Electrolytes with Different TDA + TFSI Additive Contents
The electrolytes containing ionic liquids were prepared by adding TDA + TFSI to a mixed solution of DOL and DME.The formulae for the electrolytes with different TDA + TFSI additive contents are shown in Table 1.

Results and Discussion
The TDA + TFSI additive was synthesized by the quaternization reaction of TDA and HCl, followed by anion exchange with LiTFSI.Figure 1a shows the 1 H-NMR spectrum of TDA + TFSI, which confirms the chemical structure of TDA + TFSI.The chemical shifts of the peaks are assigned as follows: δ 3.94 (q, J = 5.3 Hz, 6H) (N + -CH 2 -CH 2 ), 3.64-3.58(m, 6H) (CH 2 -CH 2 -OCH 3 ), 3.54 (t, J = 4.8 Hz, 6H) (N + -CH 2 -CH 2 ), 3.51-3.46(m, 6H) (CH 2 -CH 2 -OCH 3 ), and 3.31 (s, 9H) (CH 2 -CH 2 -OCH 3 ).The 13 C-NMR spectral analysis of TDA + TFSI is also provided in Figure 1b, which confirmed the synthesis of the TDA + TFSI molecule.The chemical shifts of the peaks are assigned as follows: 13   FT−IR spectroscopy was used to characterize the structure of TDA + TFSI (Figure 2a).The result showed that the characteristic absorption peak at 2890 cm −1 is attributed to the C-H stretching vibration, and the peak at 1458 cm −1 corresponds to the C-H in-plane bending vibration.There are characteristic absorption peaks at 1353 cm −1 and 1136 cm −1 , which belong to the S=O symmetrical telescopic vibration and asymmetrical telescopic vibration of TDA + TFSI.Meanwhile, the characteristic absorption peaks at 1193 cm −1 , 1057 cm −1 , 790 cm −1 , and 612 cm −1 correspond to the C-O-C vibration, S-N-S stretching vibration, CF3 vibration, and O=S=O vibration, respectively [35]. Figure 2b,c show the mass spectra of TDA + TFSI.The positive and negative ions of TDA + TFSI show obvious peaks in the mass spectra.The strong ion peaks in the positive mode could be identified as N + H(CH2CH2OCH2CH2OCH3)3 (at m/z = 324.3),and other peaks correspond to fragments.The strong ion peaks in the negative mode could be identified as N − (SO2CF3)2 (m/z = 279.9),and the peak at m/z = 113.0represents an interior label.As shown in Figure 3a,b, TDA + TFSI was stable in the electrochemical window of 1.6 V-4.0 V at 25 °C, which is wider than the cutoff voltage of LSBs, i.e., from 1.8 V to 3.0 V.The cathodic current below 1.6 V vs. Li + /Li could be attributed to the reductive decomposition of the quaternary ammonium salt structure of TDA + TFSI, indicating that the additive forms an SEI film.The ionic conductivities of the electrolytes with different TDA + TFSI FT−IR spectroscopy was used to characterize the structure of TDA + TFSI (Figure 2a).The result showed that the characteristic absorption peak at 2890 cm −1 is attributed to the C-H stretching vibration, and the peak at 1458 cm −1 corresponds to the C-H in-plane bending vibration.There are characteristic absorption peaks at 1353 cm −1 and 1136 cm −1 , which belong to the S=O symmetrical telescopic vibration and asymmetrical telescopic vibration of TDA + TFSI.Meanwhile, the characteristic absorption peaks at 1193 cm −1 , 1057 cm −1 , 790 cm −1 , and 612 cm −1 correspond to the C-O-C vibration, S-N-S stretching vibration, CF 3 vibration, and O=S=O vibration, respectively [35]. Figure 2b,c show the mass spectra of TDA + TFSI.The positive and negative ions of TDA + TFSI show obvious peaks in the mass spectra.The strong ion peaks in the positive mode could be identified as N + H(CH 2 CH 2 OCH 2 CH 2 OCH 3 ) 3 (at m/z = 324.3),and other peaks correspond to fragments.The strong ion peaks in the negative mode could be identified as N − (SO 2 CF 3 ) 2 (m/z = 279.9),and the peak at m/z = 113.0represents an interior label.
OCH3), and 3.31 (s, 9H) (CH2-CH2-OCH3).The 13 C-NMR spectral analysis of TDA + TFSI is also provided in Figure 1b, which confirmed the synthesis of the TDA + TFSI molecule.The chemical shifts of the peaks are assigned as follows: 13   FT−IR spectroscopy was used to characterize the structure of TDA + TFSI (Figure 2a).The result showed that the characteristic absorption peak at 2890 cm −1 is attributed to the C-H stretching vibration, and the peak at 1458 cm −1 corresponds to the C-H in-plane bending vibration.There are characteristic absorption peaks at 1353 cm −1 and 1136 cm −1 , which belong to the S=O symmetrical telescopic vibration and asymmetrical telescopic vibration of TDA + TFSI.Meanwhile, the characteristic absorption peaks at 1193 cm −1 , 1057 cm −1 , 790 cm −1 , and 612 cm −1 correspond to the C-O-C vibration, S-N-S stretching vibration, CF3 vibration, and O=S=O vibration, respectively [35]. Figure 2b,c show the mass spectra of TDA + TFSI.The positive and negative ions of TDA + TFSI show obvious peaks in the mass spectra.The strong ion peaks in the positive mode could be identified as N + H(CH2CH2OCH2CH2OCH3)3 (at m/z = 324.3),and other peaks correspond to fragments.The strong ion peaks in the negative mode could be identified as N − (SO2CF3)2 (m/z = 279.9),and the peak at m/z = 113.0represents an interior label.As shown in Figure 3a,b, TDA + TFSI was stable in the electrochemical window of 1.6 V-4.0 V at 25 °C, which is wider than the cutoff voltage of LSBs, i.e., from 1.8 V to 3.0 V.The cathodic current below 1.6 V vs. Li + /Li could be attributed to the reductive decomposition of the quaternary ammonium salt structure of TDA + TFSI, indicating that the additive forms an SEI film.The ionic conductivities of the electrolytes with different TDA + TFSI As shown in Figure 3a,b, TDA + TFSI was stable in the electrochemical window of 1.6 V-4.0 V at 25 • C, which is wider than the cutoff voltage of LSBs, i.e., from 1.8 V to 3.0 V.The cathodic current below 1.6 V vs. Li + /Li could be attributed to the reductive decomposition of the quaternary ammonium salt structure of TDA + TFSI, indicating that the additive forms an SEI film.The ionic conductivities of the electrolytes with different TDA + TFSI additive contents were characterized, and the results are shown in Figure 3c.The ionic conductivity increased with increasing content of the TDA + TFSI additive.With increasing TDA + TFSI additive content, neutral organic molecules tend to substitute the TFSI − because of the stronger interaction between Li + and organic molecules [36].Therefore, the proportions of Li + -DME and Li + -DOL increased, while the coordination between lithium and TFSI − anions decreased.This yields a higher fraction of "free" TFSI − ions, leading to higher ionic conductivity [37].The high ionic conductivity is beneficial to the electrochemical reaction of the LSBs. Figure 3d is the photograph of TDA + TFSI, showing a dark brown liquid characteristic.
additive contents were characterized, and the results are shown in Figure 3c.The ionic conductivity increased with increasing content of the TDA + TFSI additive.With increasing TDA + TFSI additive content, neutral organic molecules tend to substitute the TFSI − because of the stronger interaction between Li + and organic molecules [36].Therefore, the proportions of Li + -DME and Li + -DOL increased, while the coordination between lithium and TFSI − anions decreased.This yields a higher fraction of "free" TFSI − ions, leading to higher ionic conductivity [37].The high ionic conductivity is beneficial to the electrochemical reaction of the LSBs. Figure 3d is the photograph of TDA + TFSI, showing a dark brown liquid characteristic.To evaluate the performances of different TDA + TFSI additive contents in LSBs, coin cells were assembled using carbon Super P/S composites with an ~71 wt.% sulfur content as the cathode (Figure S2).The galvanostat charge-discharge of LSBs with different TDA + TFSI additive contents were tested at 25 °C with a current density of 0.5 C. Figure 4 shows the charge-discharge curves in the first cycle of the LSBs assembled with 0%, 3%, 5%, and 10% TDA + TFSI electrolyte additives.The curves of all the LSBs using the electrolytes with the TDA + TFSI additive had two discharge plateaus.The first discharge plateau is located at 2.3 V-2.4 V, which corresponds to the reduction of the active substance, S8, to high-order polysulfides (S8 2− and S6 2− ).The second discharge plateau is located at 2.0 V-2.1 V, which corresponds to the higher-order polysulfides (S8 2− and S6 2− ) further obtaining electrons to become lower-order polysulfides (S4 2− and S2 2− ) [38,39].Similarly, there are two charge plateaus in the curves of these systems.The first charge plateau is located at 2.2 V-2.3 V.This plateau represents the reverse process of the corresponding reaction for the second discharge plateau, which corresponds to the oxidation of the final discharge product, Li2S, to form low-order polysulfides (S4 2− and S2 2− ).The second charge plateau is located at 2.3 V-2.4 V, this plateau represents the reverse process of the corresponding reaction for the first discharge plateau, which corresponds to low-order polysulfides further losing electrons to form high-order polysulfides (S8 2− and S6 2− ).These results indicate that the electrolytes with the TDA + TFSI additive had no effect on the electrochemical reaction of the LSBs.To evaluate the performances of different TDA + TFSI additive contents in LSBs, coin cells were assembled using carbon Super P/S composites with an ~71 wt.% sulfur content as the cathode (Figure S2).The galvanostat charge-discharge of LSBs with different TDA + TFSI additive contents were tested at 25 • C with a current density of 0.5 C. Figure 4 shows the charge-discharge curves in the first cycle of the LSBs assembled with 0%, 3%, 5%, and 10% TDA + TFSI electrolyte additives.The curves of all the LSBs using the electrolytes with the TDA + TFSI additive had two discharge plateaus.The first discharge plateau is located at 2.3 V-2.4 V, which corresponds to the reduction of the active substance, S 8 , to high-order polysulfides (S 8 2− and S 6 2− ).The second discharge plateau is located at 2.0 V-2.1 V, which corresponds to the higher-order polysulfides (S 8 2− and S 6 2− ) further obtaining electrons to become lower-order polysulfides (S 4 2− and S 2 2− ) [38,39].Similarly, there are two charge plateaus in the curves of these systems.The first charge plateau is located at 2.2 V-2.3 V.This plateau represents the reverse process of the corresponding reaction for the second discharge plateau, which corresponds to the oxidation of the final discharge product, Li 2 S, to form low-order polysulfides (S 4 2− and S 2 2− ).The second charge plateau is located at 2.3 V-2.4 V, this plateau represents the reverse process of the corresponding reaction for the first discharge plateau, which corresponds to low-order polysulfides further losing electrons to form high-order polysulfides (S 8 2− and S 6 2− ).These results indicate that the electrolytes with the TDA + TFSI additive had no effect on the electrochemical reaction of the LSBs. Figure 5a shows the long-term cycling performance of the LSBs using electrolytes with different TDA + TFSI additive contents at 25 °C and a current density of 0.5 C. The specific capacities of the LSBs with TDA + TFSI additive contents of 0%, 3%, 5%, and 10% were 971, 988, 1167, and 1079 mAh g −1 in the first cycle, respectively.After 100 cycles, the corresponding specific capacities decayed to 550, 461, 579, and 483 mAh g −1 , respectively.After 300 cycles, the corresponding specific capacities decayed to 478, 416, 523, and 400 mAh g −1 , respectively.Among the different LSBs, the LSBs using the electrolyte with 5% of the TDA + TFSI additive showed the excellent electrochemical performance during the whole cycle.The average capacity decay rate per cycle for the electrolyte with 5% of the TDA + TFSI additive was only 0.18% in 300 cycles, and the Coulombic efficiency of the LSBs using the electrolyte with 5% of the TDA + TFSI additive was 97-99%.However, the specific capacities of the electrolytes with 3% and 10% TDA + TFSI additive contents were no better than that of the electrolyte without the TDA + TFSI additive.This may be due to the different effects of the electrolytes with different TDA + TFSI additive contents on the formation of SEI films.
To verify the effects of different TDA + TFSI additive contents on SEI films, the impedances of LSBs using electrolytes with different TDA + TFSI additive contents were tested before and after 300 cycles at 0.5 C, and the results are shown in Figure 5b,c, respectively.The analog circuit shown in the inset of Figure 5b was used to simulate the experiment to obtain the impedance spectrum.The bulk resistance (Rb) (i.e., the intercept between the semicircle and real axis) values of the LSBs using electrolytes with 0%, 3%, 5%, and 10% of the TDA + TFSI additive were 3.1, 3.9, 4.4, and 3.5 Ω, respectively [40,41], while the interfacial resistance (Ri) (i.e., the diameter of the semicircle) values of the LSBs were 25.5, 39.3, 67.4, and 81.4 Ω, respectively.After 300 cycles were run at 0.5 C, the corresponding Rb values of the LSBs using electrolytes with 0%, 3%, 5%, and 10% of the TDA + TFSI additive became 4.2, 5.1, 3.6, and 3.1 Ω, respectively, and the Ri values of the LSBs became 7.6, 7.6, 5.4, and 15.9 Ω, respectively.The Rb values of the cells with 0% of the additive and 3% of the additive increased, while the Rb values of the cells with 5% of the additive and 10% of the additive decreased.This may be due to the reduction in the internal resistance of the LSBs during cycling with increasing TDA + TFSI additive content.Therefore, the battery can provide more stored electric energy to the external circuit.In addition, the change in Rb before and after cycling can be almost ignored, while the Ri value is substantially reduced.This can be attributed to the fact that the TDA + TFSI additives in the electrolyte participate in the reaction to form SEI films during cycling, which accelerates the charge transfer.The LSBs using the electrolyte with 5% of the TDA + TFSI additive had the lowest Ri value after cycling, which is consistent with the best performance of the 5% TDA + TFSI additive in the cycle test.These results might suggest that the SEI film formed with the 5% TDA + TFSI additive was denser and more uniform than the other SEI films.Figure 5a shows the long-term cycling performance of the LSBs using electrolytes with different TDA + TFSI additive contents at 25 • C and a current density of 0.5 C. The specific capacities of the LSBs with TDA + TFSI additive contents of 0%, 3%, 5%, and 10% were 971, 988, 1167, and 1079 mAh g −1 in the first cycle, respectively.After 100 cycles, the corresponding specific capacities decayed to 550, 461, 579, and 483 mAh g −1 , respectively.After 300 cycles, the corresponding specific capacities decayed to 478, 416, 523, and 400 mAh g −1 , respectively.Among the different LSBs, the LSBs using the electrolyte with 5% of the TDA + TFSI additive showed the excellent electrochemical performance during the whole cycle.The average capacity decay rate per cycle for the electrolyte with 5% of the TDA + TFSI additive was only 0.18% in 300 cycles, and the Coulombic efficiency of the LSBs using the electrolyte with 5% of the TDA + TFSI additive was 97-99%.However, the specific capacities of the electrolytes with 3% and 10% TDA + TFSI additive contents were no better than that of the electrolyte without the TDA + TFSI additive.This may be due to the different effects of the electrolytes with different TDA + TFSI additive contents on the formation of SEI films.
To verify the effects of different TDA + TFSI additive contents on SEI films, the impedances of LSBs using electrolytes with different TDA + TFSI additive contents were tested before and after 300 cycles at 0.5 C, and the results are shown in Figure 5b,c, respectively.The analog circuit shown in the inset of Figure 5b was used to simulate the experiment to obtain the impedance spectrum.The bulk resistance (R b ) (i.e., the intercept between the semicircle and real axis) values of the LSBs using electrolytes with 0%, 3%, 5%, and 10% of the TDA + TFSI additive were 3.1, 3.9, 4.4, and 3.5 Ω, respectively [40,41], while the interfacial resistance (R i ) (i.e., the diameter of the semicircle) values of the LSBs were 25.5, 39.3, 67.4, and 81.4 Ω, respectively.After 300 cycles were run at 0.5 C, the corresponding R b values of the LSBs using electrolytes with 0%, 3%, 5%, and 10% of the TDA + TFSI additive became 4.2, 5.1, 3.6, and 3.1 Ω, respectively, and the R i values of the LSBs became 7.6, 7.6, 5.4, and 15.9 Ω, respectively.The R b values of the cells with 0% of the additive and 3% of the additive increased, while the R b values of the cells with 5% of the additive and 10% of the additive decreased.This may be due to the reduction in the internal resistance of the LSBs during cycling with increasing TDA + TFSI additive content.Therefore, the battery can provide more stored electric energy to the external circuit.In addition, the change in R b before and after cycling can be almost ignored, while the R i value is substantially reduced.This can be attributed to the fact that the TDA + TFSI additives in the electrolyte participate in the reaction to form SEI films during cycling, which accelerates the charge transfer.The LSBs using the electrolyte with 5% of the TDA + TFSI additive had the lowest R i value after cycling, which is consistent with the best performance of the 5% TDA + TFSI additive in the cycle test.These results might suggest that the SEI film formed with the 5% TDA + TFSI additive was denser and more uniform than the other SEI films.To further examine the difference between the SEI films formed by different electrolytes, the LSBs using electrolytes with different TDA + TFSI additive contents were disassembled after 300 cycles at 0.5 C, and the surface morphologies of the lithium sheets were observed using SEM. Figure 6a-d show the surface morphologies of the lithium sheets after cycling using electrolytes with 0%, 3%, 5%, and 10% of the TDA + TFSI additive, respectively.For comparison, the surface morphology of a fresh lithium sheet is shown in Figure 6e.The surface of the fresh lithium sheet is smooth and flat, while the surfaces of the lithium sheets after charging and discharging appear to have a layered structure.This layered structure is the SEI film formed on the surface of the lithium sheet during the electrochemical reaction.Compared with the SEI films on the surfaces of the lithium sheets using electrolytes with 3%, 5%, and 10% of the TDA + TFSI additive, the SEI film on the surface of the lithium sheet using the electrolyte with 0% of the TDA + TFSI additive has more layered structures, indicating that the TDA + TFSI additive contributes to the formation of a uniformly dense SEI film.Among the different electrolyte additive contents, the 5% TDA + TFSI additive content produced a lithium sheet that had the flattest and most uniform surface without any obvious hollowing, although obvious hollowing is observed on the surface of the lithium sheet using the electrolyte with a TDA + TFSI additive content of 3%.This may be because the content of the TDA + TFSI additive in the electrolyte is too low, resulting in an SEI film that is insufficient to evenly cover the entire lithium sheet's surface.However, when the content of the TDA + TFSI electrolyte additive was increased to 10%, significant holes appeared on the surface of the lithium sheet, which is the result of the excessive growth of the SEI film caused by the excess additive in the electrolyte.The increase in the SEI film thickness forms an inert layer, leading to a reduced charge-transfer rate of the LSBs.This result is consistent with that of the cycling test: the smoother and denser the SEI film, the higher the specific capacity of the corresponding LSB.The cycling performances of lithium symmetrical cells using electrolytes with different TDA + TFSI additive contents and the impedances before and after 100 cycles were also studied.Figure 5d shows the cycling performances of lithium symmetrical cells at 0.1 mA cm −2 and a Li capacity of 1 mAh cm −2 .The electrolytes with 0%, 3%, 5%, and 10% of the TDA + TFSI additive had initial overpotentials of 19, 24, 20, and 22 mV, respectively.In addition, the overpotentials of the electrolytes with 0%, 3%, and 5% of the TDA + TFSI additive showed significant fluctuations after 50 h, which can be attributed to the destruction of SEI films and unstable interfacial channels between the metallic lithium and electrolyte.On the contrary, the electrolyte with 5% of the TDA + TFSI additive still maintains a relatively smooth voltage plateau and a low overpotential over 200 h, indicating that the electrolyte with 5% of the TDA + TFSI additive is conducive to stabilize Li anodes and lower the occurrence of polarization during continuous Li stripping/plating.These results suggest that the electrolyte with 5% of the additive forms the most stable and uniform SEI film on the Li metal anode.Figure 5e,f show the impedance spectra of the lithium symmetrical cells before and after 100 cycles.The R b values of the lithium symmetrical cells using electrolytes with 0%, 3%, 5%, and 10% of the TDA + TFSI additive were 3.9, 4.7, 3.5, and 3.2 Ω before cycling, respectively, and the R i values were 45.4,48.2, 72.5, and 95.2 Ω, respectively.After 100 cycles, the R b values became 3.6, 3.5, 3.1, and 3.5 Ω, respectively, and the R i values became 43.5, 21.6, 11.0, and 18.5 Ω, respectively.In addition, the relationship between the impedance and overpotential of the symmetrical batteries with or without the TDA + TFSI additive is as follows: On the one hand, owing to the higher viscosity of the electrolyte containing 5% of the TDA + TFSI additive compared with the electrolyte containing 0% of the TDA + TFSI additive, the permeability between the electrolyte with 5% of the additive and the electrode is relatively incomplete at the beginning of the cycle, resulting in a higher impedance value of the battery.This leads to the battery with 5% of the TDA + TFSI additive having a high initial overvoltage.The complete contact between the electrolyte and lithium metal electrode during continuous Li stripping/plating results in a decrease in the impedance value and a decrease in the overpotential.On the other hand, the electrolyte containing 5% of the TDA + TFSI additive is conducive to the formation of stable SEI films and can improve electrode interfaces, thereby reducing the interfacial impedance and charge-transfer resistance.The electrolyte with 0% of the additive has a low viscosity that enables the electrolyte to fully contact the electrode, resulting in minimal changes in the impedance value of the battery before and after cycling.However, owing to the instability of the SEI film formed by the electrolyte containing 0% of the TDA + TFSI additive, the SEI film continuously dissolves and reforms during the Li stripping/plating process, which leads to the continuous consumption of lithium nitrate in the electrolyte and causes overpotential fluctuations [22].This result is basically consistent with the impedance change before and after the cycling of the LSBs; that is, the R b values of the lithium symmetrical cells change slightly before and after cycling, while the R i values decrease significantly after cycling.This phenomenon is also due to the formation of an SEI film on the lithium sheet surface during cycling.It is worth noting that the R i value of the electrolyte without the additive changed little before and after cycling, which indicates that the TDA + TFSI additive in the electrolyte played a key role in the formation of the SEI film.
To further examine the difference between the SEI films formed by different electrolytes, the LSBs using electrolytes with different TDA + TFSI additive contents were disassembled after 300 cycles at 0.5 C, and the surface morphologies of the lithium sheets were observed using SEM. Figure 6a-d show the surface morphologies of the lithium sheets after cycling using electrolytes with 0%, 3%, 5%, and 10% of the TDA + TFSI additive, respectively.For comparison, the surface morphology of a fresh lithium sheet is shown in Figure 6e.The surface of the fresh lithium sheet is smooth and flat, while the surfaces of the lithium sheets after charging and discharging appear to have a layered structure.This layered structure is the SEI film formed on the surface of the lithium sheet during the electrochemical reaction.Compared with the SEI films on the surfaces of the lithium sheets using electrolytes with 3%, 5%, and 10% of the TDA + TFSI additive, the SEI film on the surface of the lithium sheet using the electrolyte with 0% of the TDA + TFSI additive has more layered structures, indicating that the TDA + TFSI additive contributes to the formation of a uniformly dense SEI film.Among the different electrolyte additive contents, the 5% TDA + TFSI additive content produced a lithium sheet that had the flattest and most uniform surface without any obvious hollowing, although obvious hollowing is observed on the surface of the lithium sheet using the electrolyte with a TDA + TFSI additive content of 3%.This may be because the content of the TDA + TFSI additive in the electrolyte is too low, resulting in an SEI film that is insufficient to evenly cover the entire lithium sheet's surface.However, when the content of the TDA + TFSI electrolyte additive was increased to 10%, significant holes appeared on the surface of the lithium sheet, which is the result of the excessive growth of the SEI film caused by the excess additive in the electrolyte.The increase in the SEI film thickness forms an inert layer, leading to a reduced charge-transfer rate of the LSBs.This result is consistent with that of the cycling test: the smoother and denser the SEI film, the higher the specific capacity of the corresponding LSB.
The charge-discharge curves of the electrolytes with 5% and 0% of the TDA+TFSI additive in different charge-discharge cycles are shown in Figures 7a and S3, respectively.The charge and discharge specific capacities of the two types of LSBs decreased with increasing number of cycles, but a clear charge-discharge plateau can still be observed, indicating that the addition of 5% TDA + TFSI to the electrolyte does not affect the electrochemical stability of the battery during the cycling process.The gradient rate capabilities of the LSBs using the electrolytes with 0% and 5% of the TDA + TFSI additive were investigated at 25 • C via the sequential variation in the current densities.As shown in Figure 7b, when the charge-discharge current was less than 0.5 C, there was little difference between the discharge specific capacities of the electrolytes with and without the additive.However, when the charge-discharge current was higher than 0.5 C, the electrolyte containing 5% of the TDA + TFSI additive obviously had a higher discharge specific capacity than the electrolyte without the additive.This result indicates that the 5% TDA + TFSI additive in the electrolyte forms a stable SEI film on the lithium sheet's surface, which makes the electrochemical reaction more stable, and this effect becomes more obvious with increasing charge-discharge current.In addition, when the charge-discharge current is recovered to 0.1 C, the specific discharge capacity of the LSB is basically the same as that at the beginning, indicating that the LSB with 5% of the TDA + TFSI additive in the electrolyte has an excellent rate capability.The charge-discharge curves of the electrolytes with 5% and 0% of the TDA+TFSI additive in different charge-discharge cycles are shown in Figures 7a and S3, respectively.The charge and discharge specific capacities of the two types of LSBs decreased with increasing number of cycles, but a clear charge-discharge plateau can still be observed, indicating that the addition of 5% TDA + TFSI to the electrolyte does not affect the electrochemical stability of the battery during the cycling process.The gradient rate capabilities of the LSBs using the electrolytes with 0% and 5% of the TDA + TFSI additive were investigated at 25 °C via the sequential variation in the current densities.As shown in Figure 7b, when the charge-discharge current was less than 0.5 C, there was little difference between the discharge specific capacities of the electrolytes with and without the additive.However, when the charge-discharge current was higher than 0.5 C, the electrolyte containing 5% of the TDA + TFSI additive obviously had a higher discharge specific capacity than the electrolyte without the additive.This result indicates that the 5% TDA + TFSI additive in the electrolyte forms a stable SEI film on the lithium sheet's surface, which makes the electrochemical reaction more stable, and this effect becomes more obvious with increasing charge-discharge current.In addition, when the charge-discharge current is recovered to 0.1 C, the specific discharge capacity of the LSB is basically the same as that at the beginning, indicating that the LSB with 5% of the TDA + TFSI additive in the electrolyte has an excellent rate capability.
The cycling performances of the LSBs using electrolytes with or without the TDA + TFSI additive were tested at a high charge-discharge current density (1.0 C), and the results are shown in Figure 7c.The discharge specific capacity of the LSB using the electrolyte with 5% of the TDA + TFSI additive decreases from 924 mAh g −1 in the first cycle to 428 mAh g −1 in the 100th cycle and 399 mAh g −1 in the 250th cycle.At the same time, the Coulombic efficiency of the LSB using the electrolyte with 5% of the TDA + TFSI additive is always maintained at 97-100%.Although the discharge specific capacity of the LSB using the electrolyte without the additive decreases from 817 mAh g −1 in the first cycle to 372 mAh g −1 in the 100th cycle and 339 mAh g −1 in the 250th cycle, the Coulombic efficiency of the LSBs using the electrolyte without the additive is 95-102%.The specific capacity of the LSBs with the electrolyte containing 5% of the TDA + TFSI additive is always higher than that of the LSBs with the electrolyte without the additive, which is consistent with our observational results of the rate performance.In addition, it is worth noting that the Coulombic efficiency fluctuation of the LSBs with the electrolyte containing 5% of the TDA + TFSI additive is much smaller than that of the LSBs without the electrolyte additive

Conclusions
In summary, a new type of ionic liquid, TDA + TFSI, was synthesized and used as an electrolyte additive for LSBs.The cycling performance of the LSBs using the electrolyte with 5% of the TDA + TFSI additive was significantly improved.The discharge specific capacity of the first cycle was 1167 mAh g −1 , and the residual specific capacities after 100 and 300 cycles were 579 mAh g −1 and 523 mAh g −1 , respectively, at 0.5 C. The average capacity decay rate per cycle was only 0.18% in 300 cycles, which is significantly higher than that of the LSB using the electrolyte without the TDA + TFSI additive.Moreover, the LSBs using the electrolyte with 5% of the TDA + TFSI additive had more obvious advantages at a high charge-discharge current (1.0 C).On the one hand, the discharge specific capacity of the LSB with 5% of the TDA + TFSI electrolyte additive was higher than that of the LSBs without the additive.On the other hand, the LSB with 5% of the TDA + TFSI electrolyte additive had The cycling performances of the LSBs using electrolytes with or without the TDA + TFSI additive were tested at a high charge-discharge current density (1.0 C), and the results are shown in Figure 7c.The discharge specific capacity of the LSB using the electrolyte with 5% of the TDA + TFSI additive decreases from 924 mAh g −1 in the first cycle to 428 mAh g −1 in the 100th cycle and 399 mAh g −1 in the 250th cycle.At the same time, the Coulombic efficiency of the LSB using the electrolyte with 5% of the TDA + TFSI additive is always maintained at 97-100%.Although the discharge specific capacity of the LSB using the electrolyte without the additive decreases from 817 mAh g −1 in the first cycle to 372 mAh g −1 in the 100th cycle and 339 mAh g −1 in the 250th cycle, the Coulombic efficiency of the LSBs using the electrolyte without the additive is 95-102%.The specific capacity of the LSBs with the electrolyte containing 5% of the TDA + TFSI additive is always higher than that of the LSBs with the electrolyte without the additive, which is consistent with our observational results of the rate performance.In addition, it is worth noting that the Coulombic efficiency fluctuation of the LSBs with the electrolyte containing 5% of the TDA + TFSI additive is much smaller than that of the LSBs without the electrolyte additive (Figure 7d) because the TDA + TFSI additive in the electrolyte formed a more stable SEI film during the electrochemical reaction, which improved the stability of the lithium-ion plating and stripping.In addition, the large fluctuation in the Coulombic efficiency of the LSB using the electrolyte without the additive represented the severe polysulfide shuttle during cycling [42].By contrast, the LSB with the TDA + TFSI additive formed an SEI film that could effectively prevent direct contact between the polysulfides and lithium sheet, thereby inhibiting the shuttle effect of the polysulfides.

Conclusions
In summary, a new type of ionic liquid, TDA + TFSI, was synthesized and used as an electrolyte additive for LSBs.The cycling performance of the LSBs using the electrolyte with 5% of the TDA + TFSI additive was significantly improved.The discharge specific capacity of the first cycle was 1167 mAh g −1 , and the residual specific capacities after 100 and 300 cycles were 579 mAh g −1 and 523 mAh g −1 , respectively, at 0.5 C. The average capacity decay rate per cycle was only 0.18% in 300 cycles, which is significantly higher than that of the LSB using the electrolyte without the TDA + TFSI additive.Moreover, the LSBs using the electrolyte with 5% of the TDA + TFSI additive had more obvious advantages at a high charge-discharge current (1.0 C).On the one hand, the discharge specific capacity of the LSB with 5% of the TDA + TFSI electrolyte additive was higher than that of the LSBs without the additive.On the other hand, the LSB with 5% of the TDA + TFSI electrolyte additive had a more stable Coulombic efficiency and a higher cycling stability than the LSBs without the TDA + TFSI additive.This is because the TDA + TFSI additive helps to form more dense and more uniform SEI films during cycling, thereby improving the stability of the electrochemical reactions.This work provides guidance for developing highly stable ionic liquid electrolytes for high-performance LSBs.

Figure 2 .
Figure 2. (a) The FT−IR spectrum of TDA + TFSI; (b) the mass spectrum of TDA + TFSI in the positive mode; (c) the mass spectrum of TDA + TFSI in the negative mode.

Figure 2 .
Figure 2. (a) The FT−IR spectrum of TDA + TFSI; (b) the mass spectrum of TDA + TFSI in the positive mode; (c) the mass spectrum of TDA + TFSI in the negative mode.

Figure 2 .
Figure 2. (a) The FT−IR spectrum of TDA + TFSI; (b) the mass spectrum of TDA + TFSI in the positive mode; (c) the mass spectrum of TDA + TFSI in the negative mode.

Figure 4 .
Figure 4. Charge-discharge curves in the first cycle of LSBs using electrolytes with different TDA + TFSI additive contents; the charge-discharge current density was 0.5 C.

Figure 4 .
Figure 4. Charge-discharge curves in the first cycle of LSBs using electrolytes with different TDA + TFSI additive contents; the charge-discharge current density was 0.5 C.

Figure 5 .
Figure 5. (a) Long-term cycling performances of LSBs using electrolytes with different TDA + TFSI additive contents at a charge-discharge current density of 0.5 C. (Inset: amplified cycling performance curves.)(b) EIS of LSBs using electrolytes with different TDA + TFSI additive contents before 300 cycles.The inset shows the analog circuit.(c) EIS of LSBs using electrolytes with different TDA + TFSI additive contents after 300 cycles.The inset shows the analog circuit.(d) The voltage profiles of the Li||Li symmetrical cells using electrolytes with different TDA + TFSI additive contents at 0.1 mA cm −2 and a stripping/plating capacity of 1 mAh cm −2 .(e) EIS of lithium symmetrical cells using electrolytes with different TDA + TFSI additive contents before the cycling test.The inset shows the analog circuit.(f) EIS of lithium symmetrical cells using electrolytes with different TDA + TFSI additive contents after the cycling test.The inset shows the analog circuit.

Figure 5 .
Figure 5. (a) Long-term cycling performances of LSBs using electrolytes with different TDA + TFSI additive contents at a charge-discharge current density of 0.5 C. (Inset: amplified cycling performance curves).(b) EIS of LSBs using electrolytes with different TDA + TFSI additive contents before 300 cycles.The inset shows the analog circuit.(c) EIS of LSBs using electrolytes with different TDA + TFSI additive contents after 300 cycles.The inset shows the analog circuit.(d) The voltage profiles of the Li||Li symmetrical cells using electrolytes with different TDA + TFSI additive contents at 0.1 mA cm −2 and a stripping/plating capacity of 1 mAh cm −2 .(e) EIS of lithium symmetrical cells using electrolytes with different TDA + TFSI additive contents before the cycling test.The inset shows the analog circuit.(f) EIS of lithium symmetrical cells using electrolytes with different TDA + TFSI additive contents after the cycling test.The inset shows the analog circuit.

Figure 7 .
Figure 7. (a) Charge-discharge curves of the LSB using the electrolyte with 5% of the TDA + TFSI additive in different cycles at a current density of 0.5 C. (b) Rate performances of LSBs using electrolytes with 5% of the TDA + TFSI additive and without the additive.(c) Cycling performances of LSBs using electrolytes with 5% of the TDA + TFSI additive and without the additive at charge-discharge current densities of 1.0 C. (d) The corresponding Coulombic efficiencies of the LSBs using electrolytes with 5% of the TDA + TFSI additive and without the additive at a current density of 1.0 C.

Figure 7 .
Figure 7. (a) Charge-discharge curves of the LSB using the electrolyte with 5% of the TDA + TFSI additive in different cycles at a current density of 0.5 C. (b) Rate performances of LSBs using electrolytes with 5% of the TDA + TFSI additive and without the additive.(c) Cycling performances of LSBs using electrolytes with 5% of the TDA + TFSI additive and without the additive at charge-discharge current densities of 1.0 C. (d) The corresponding Coulombic efficiencies of the LSBs using electrolytes with 5% of the TDA + TFSI additive and without the additive at a current density of 1.0 C.