High Conductivity Solvates with Unsymmetrical ... - ACS Publications

Article Cite This: Chem. Mater. 2018, 30, 246−251

pubs.acs.org/cm

High Conductivity Solvates with Unsymmetrical Glymes as New Electrolytes Devaraj Shanmukaraj,† Sandrine Lois,‡ Sebastien Fantini,‡ Francois Malbosc,‡ and Michel Armand*,† CIC EnergiGUNE, Parque Tecnológico de Á lava, 48, 01510 Miñano, Á lava, Spain Solvionic, Site Bioparc Sanofi, 195 Route d’Espagne, Toulouse 31036, France

† ‡

S Supporting Information *

ABSTRACT: A designed custom glyme ether with unsymmetrical (ethyl and butyl) end groups (UG) has been shown to give highly conducting liquid solvate electrolytes. Solvates prepared with UG glyme: salt ratios of 1:1, 2:1, and 3:1 using lithium bis(trifluoromethane sulfonyl) imide (LiTFSI) and lithium bis(fluoro sulfonyl) imide (LiFSI) lithium salts have been characterized with emphasis on the 2:1 composition. Conductivity in the order of 10−3 S/cm was observed at room temperature with an electrochemical stability window of 4.5 V. DSC studies indicate a low crystallization temperature between −60 °C and −75 °C for LiFSI based solvate electrolyte. Galvanostatic cycling studies (C/10) with LiFePO4 electrodes indicate a capacity deliverance of 145 mAh/g and good C-rate capability up to 2C at room temperature. Preliminary tests using unsymmetrical glyme-based solvate electrolytes with sulfur and LiNi1/3Mn1/3Co1/3O2 (NMC) electrodes reveal a low degree of polysulfide dissolution with sulfur electrodes and stable voltage profiles with NMC electrodes suggesting the use of these electrolytes for Li/S batteries and high-voltage Li-ion batteries.

1. INTRODUCTION Lithium batteries, whose success increases by the day, face the double challenge of increasing the energy density and improving safety.1−5 A lithium metal negative electrode is seen as one of the most effective strategies for the former.6 A change in electrolyte from the volatile/flammable mixtures of organic carbonate solvents is actively searched for, with ceramic, polymer, and ultraconcentrated salt solutions.7−12 To now, only PEO based polymer electrolytes have shown, in commercial cells, to be able to cycle well with a Li° electrode, but operate at 60−80 °C13 due to the low conductivity of the system at RT. Angell has studied extensively concentrated solution especially in water or in polymers (“polymer in salt”) and studied its fragility, which gives an indication of the decoupling of conductivity from viscosity.14−17 In the recent years, Watanabe and co-workers18−21 have shown that the so-called glymes, i.e., dimethyl ether of oligoethylene glycols, in this case, G3 and G4 having four respectively five coordinating oxygen, gave 1:1 complexes with Li salts of the imide family, namely, Li[(CF3SO2)2N] (LiTFSI), which were liquid at ambient temperature with conductivities in the range of 1.1 and 1.6 mS·cm−1 for G3 and G4. An added advantage was found in terms of the stability window, as all the ether oxygens are engaged in electron doublet donation with lithium, making the electrolyte much less prone to oxidization, and in practice, up to +4.5 V have been suggested from CV experiments. Though appealing, such materials, when used in an operating battery with lithiumexchanging electrodes, have problems of solvent balance at the © 2017 American Chemical Society

electrodes. At the anode (negative electrode on discharge, positive on charge) when lithium is injected in the electrolyte, there is no free glyme available to solvate the corresponding salt, and this will lead to its precipitation. Conversely, at the cathode, the desolvation process frees some glyme, not coordinated to Li+, and this lowers the potential at which the electrolytes get oxidized. Also, there is now strong indications that the glymes with their terminal O−CH3 ether group are mutagenic and should be phased out in the future (EPA norms22). We thus decided to address both concerns by redesigning the PEG ethers used for the preparation of solvates with the unsymmetrical end groups, ethyl and butyl in this study. Diethylene glycol was chosen as the PEG base. The lack of symmetry was expected to allow a wider liquid range, in an approach similar to that of ionic liquids, and lithium salt would be able to coordinate to form a monosolvate (3 oxygens) and a disolvate (6 oxygens), the latter satisfying all the coordination requirement of Li+. Further, the solvates would likely be miscible into excess solvent, due again to the low symmetry of the solvate.

2. EXPERIMENTAL SECTION Diethylene glycol ethyl-n-butyl ether (unsymmetric glyme-UG) was custom synthesized through a Williamson coupling from diethylene Received: October 10, 2017 Revised: December 18, 2017 Published: December 18, 2017 246

DOI: 10.1021/acs.chemmater.7b04270 Chem. Mater. 2018, 30, 246−251

Article

Chemistry of Materials glycol monoethyl ether and butyl chloride. The solvent was further dried by adding NaH followed by distillation in a bulb-to-bulb apparatus (Büchi-B585). All handling was carried out inside a drybox under Ar atmosphere (H2O/O2 < 0.1 ppm). LiTFSI (Li(CF3SO2)2N) came from Solvay, and LiFSI (Li(SO2F)2N) was received from Fluolyte (Suzhou, China). Solvates of UG/lithium salt (LiFSI/ LiTFSI) were prepared inside a glovebox under argon atmosphere (H2O and O2 < 1 ppm). Electrochemical Characterization. Ionic conductivity of the electrolyte solution in this study was measured using CDC749 (Radiometer analytical) cell connected to a VMP (Biologic, Claix, France) electrochemical workstation. Temperature dependent conductivity was measured with electrolytes in sealed vials, and temperature was maintained using a Thermo Fischer climatic chamber. The temperature was maintained for at least 30 min at each temperature before the measurements were taken. The reported values are an average of three measurements with a relative standard error of less than 5%. Cyclic voltammograms were taken using a platinum working/ counter electrode and Li foil as a reference electrode at a sweep rate of 10 mV/s. LiFePO4 (Aleees, Taiwan) laminates were cast on an Al current collector by preparing a slurry with 85% LiFePO4, 15% carbon black (Timcal), and 10% polyvinylidene fluoride (PVdF, Solvay) using N-methyl-2-pyrolidone (NMP, Sigma-Aldrich) as a solvent. The laminates were dried overnight at 120 °C under vacuum. Similarly, LiNi1/3Mn1/3Co1/3O2 (NMC, Targray) laminates were prepared by casting the slurry on a carbon paper current collector using the procedure as described above. Sulfur cathodes were prepared using the melt diffusion technique. 70% sulfur and 30% Csp were mixed in an agate mortar for 10 min and heated at 155 °C for 6 h in an oven. The composite was then casted on Al foil using NMP solvent with 90% sulfur/carbon composite and 10% PVdF binder followed by drying at 60 °C. CR2032 coin cells were assembled in an argon-filled dry glovebox. The cathodes and Li foil anode were separated using a Whatman GF/ D borosilicate glass fiber imbibed with UG/Lithium salt solvate electrolyte. Once assembled, the cells were cycled between 2 to 3.8 V, 3 to 4.2 V, and 1 to 3 V for LiFePO4, NMC, and sulfur cathodes, respectively, using a VMP3 (Biologic) system. Differential Scanning Calorimetry. DSC measurements were performed using DSC1, STARe, at a scan rate of 2 °C/min. The samples were prepared and sealed in aluminum pans inside a glovebox before being introduced in the DSC. NMR Measurements. NMR measurements were carried out using a 500 MHz Ascend Bruker spectrometer equipped with a prodigy multinuclear direct probe. The analysis was carried out at 298 K for 16 scans at a pulse of 30° and a relaxation time d1 of 3 s. Viscosity Measurements. The viscosity of the (2:1) UG:LiFSI glyme solvate was measured using a Brookfield cap 2000 viscometer in a dry room at 25 °C.

Ionic Conductivity Measurements. Temperature dependent conductivity studies of UG solvate electrolytes with LiTFSI and LiFSI are shown in Figure 2. Ionic conductivities from −20 to 45 °C reveal the overall high conductivity values achieved with the UG glyme solvates of the order of 10−5 to 10−3 S/cm between −20 and 45 °C. The conductivity increases exponentially with increase in temperature. The general trend of a lower conductivity for 1:1 glyme solvate when compared with 2:1 or 3:1 solvate electrolytes could be attributed to the higher viscosity of the 1:1 complex with each glyme molecule complexed to Li+ cation and strong ion pairing due to the inability of providing a complete solvation shell. Moreover, from the plot (Figure 2a,b), a dip in conductivity at low temperature between −20 °C to −15 °C was observed for solvates with LiTFSI and LiFSI salt. This could be as a result of solvate aggregates that can be formed at low temperatures due the presence of more than one free glyme molecule that can associate with a single Li+ cation.23 Comparing the overall conductivity of glyme solvate (3:1) and (2:1) with the two different salts LiTFSI and LiFSI, it can be seen that a higher conductivity is achieved for solvates with LiTFSI salt than with LiFSI salt. Although the crystallization kinetics is well-known for LiTFSI salt, that for LiFSI is still unknown. This behavior could be speculated as due to a higher degree of ionic association of Li+ cation with the anion in the case of LiFSI salt thereby leading to a higher degree of solvate aggregates. This decreases the overall ionic conductivity which is still however high when compared to other conventional ether glyme solvates.20 Cyclic Voltammetry. Our interest shifted toward the 2:1 and 3:1 solvates that displayed a better ionic conductivity than the 1:1 complex. Therefore, the cyclic voltammograms of UG/ lithium salt solvates with LiTFSI and LiFSI salts were carried out and are shown in Figure 3a−d. The oxidation curve from 0 to 5.5 V reveals that all electrolytes were stable up to 4.5 V. Watanabe et al. reported that equimolar complexes of glyme:Li salt displayed an electrochemical stability of close to 5 V owing to the oxidative stability enhancement due to the absence of free glyme molecules.21 However, it was also reported that glymes with slightly higher ratio of free solvent molecule, necessary to avoid salt precipitation under current, have an oxidative stability just above 4 V. As for the glymes with an unsymmetrical end group, an improvement in the electrochemical stability of 4.5 V is observed (i.e., oxidative stability) even though free ether groups were present for the ratios (3:1) and (2:1). DSC Studies. As confirmed from the 1H NMR spectra (Figure S1), pure UG glymes with a freezing point of −78.01 °C, melting point of −57.91 °C, and boiling point of 238.05 °C (Figure S2) were subjected to DSC analysis with LiTFSI and LiFSI salts (Figure 4). The 2:1 solvates show a melting point of −19.4 °C and crystallization temperature of −34.6 °C with LiTFSI salt whereas in the case of LiFSI solvate, a crystallization temperature was not observed. However, a transition between −60 and −75 °C was observed that could be attributed to slow crystallization with a melting point at −54.5 °C or to a glass transition. Similar behavior was observed for UG:LiTFSI and UG:LiFSI (3:1 and 1:1) as shown in Figure S3. C-Rate Capability Test. Figure 5 shows the C-rate capability test of the UG:Li salt solvates cycled with LiFePO4 electrodes. An optimal ratio of 2:1 UG glyme solvate (viscosity: 0.2 Pa.s at 30 rpm) was chosen and tested for its

3. RESULTS AND DISCUSSION Unsymmetric Glymes (UG) Electrolyte Composition. Glymes with unsymmetrical end groups were chosen (Figure 1) with R1 = ethyl and R2 = butyl groups. Solvates of UG glymes with LiTFSI and LiFSI lithium salts were prepared with UG:Li salt having 1:1, 2:1, and 3:1 molar ratios and molal concentrations of 2.095, 1.497, and 1.1655 mol/kg for 1:1, 2:1, and 3:1, respectively.

Figure 1. Glyme molecule under study, R1 = ethyl, R2 = butyl. 247


Article

Chemistry of Materials

Figure 2. Temperature dependent conductivity of UG/lithium salt solvates with (a) LiTFSI (●) 1:1, (■) 2:1, and (⧫) 3:1; (b) LiFSI (●) 1:1, (■) 2:1, and (⧫) 3:1.

Figure 3. Cyclic voltammetry of UG/lithium salt solvates at a mole ratio of 3:1 with (a) LiTFSI and (b) LiFSI and 2:1 with (c) LiTFSI and (d) LiFSI at a scan rate of 10 mV/s.

suggests a higher transference number for Li+ of the LiFSI salts in the solvates when compared to LiTFSI salt. This could also suggest that using LiFSI salts, practical C-rates (2C) could be achieved with the solvates using unsymmetrical glymes. Galvanostatic Charge/Discharge Studies. Galvanostatic charge/discharge studies using UG:LiFSI (2:1) solvates and commercial LP30 electrolytes are shown in Figure 6. The cells cycled at a C-rate of C/10 with LP30 electrolytes display a discharge capacity of 157 mAh/g, and with the UG solvates a discharge capacity of 145 mAh/g was achieved. Though a 10 mAh/g difference was observed in the discharge capacity, the UG:LiFSI (2:1) solvates showed a higher Coulombic efficiency

C-rate performance and compared with the commercial LP30 (EC/DMC (50:50 v/v) + 1 M LiPF6) electrolyte. As can be seen from the figure all three electrolytes delivered a 100% capacity at a C-rate of C/20. When the C-rate was increased, the solvates with LiFSI salt showed better performance when compared with solvates with LiTFSI salt up to 2C and was comparable with LP30 electrolytes. However, on still further increasing the C-rate to 5C, LP30 electrolyte was able to deliver nearly 50% of the capacity whereas negligible capacity percentage was observed with both solvates owing to their higher viscosity than LP30 electrolytes. The better performance of the UG: LiFSI solvates at 2C rate 248


Article


Figure 4. DSC thermogram of (a) UG:LiTFSI (2:1) and (b) UG:LiFSI (2:1).

especially with imide-type salts. In order to avoid this effect of corrosion, while using LP30, and compare the performance with the UG glyme electrolytes, carbon paper current collectors have been used. The voltage profiles as shown in Figure 7 show that no oxidative decomposition of the solvates was observed when cycled at a potential of 4.2 V. The use of such glymes in the Li/S battery system was also analyzed. Melt-diffused sulfur electrodes were used as cathode materials with metallic lithium as anode. DME:DIOX/1 M LiTFSI was taken as a reference electrolyte without any SEI layer forming electrolyte additives and compared with UG:LiTFSI (2:1) electrolyte. As shown in Figure 8 the cells with DME:DIOX/1 M LiTFSI electrolyte showed huge polysulfide shuttling effect in the second cycle whereas cells with UG:LiTFSI (2:1) electrolytes enabled the cycling of the cells. This shows that the electrolytes with UG glymes have reduced polysulfide solubility that is likely linked to a “saltingout” effect of the polysulfides. Due to the large concentration of imide salt in the electrolyte, competing for the solvating ethers from the corresponding Li+ results in Li2Sn precipitation, and further cycling showed an initial discharge capacity of 1200 mAh/g with the capacity stabilizing around 450 mAh/g after 38 cycles. When employing Li metal as anodes, the factors influencing the performance of the batteries are Li deposition (plating/ stripping), dendrite formation, and reaction of the deposited Li with electrolyte solvents. As observed by J. Qian et al.,24 when a higher concentration of the salt was used, 4 M LiFSI in DME, a stable Li plating/stripping with improved rate capability was observed. This shows the critical role of the presence of free ether solvent molecules that plays a major role in the performance of the battery. Although the electrolytes that have been employed in this work are more dilute UG glyme/ LiFSI (2:1), superior performance was observed that suggests the absence of any reaction of the free solvent molecules of the UG glyme with the deposited Li. These results suggest the practical use of these solvates with unsymmetrical glymes (Table 1), a first attempt to the best of our knowledge in lithium and post-lithium-ion batteries, and further improvement in the discharge capacity and wettability could be achieved by employing simple formation cycles and composite electrode optimization.

Figure 5. C rate vs capacity % graphs for LiFePO4 electrodes with (●) LP30, (●) UG:LiFSI (2:1), and (●) UG:LiTFSI (2:1) (all cells maintained at 40 °C and cycled twice at C/20 before tests).

Figure 6. Galvanostatic discharge capacity of LiFePO4 cathodes with (red ●) LP30 and (green ●) UG:LiFSI (2:1); Coulombic efficiency of (red x) LP30 and (green x) UG:LiFSI (2:1) RT.

(>98%) when compared to LP30 that showed an inferior Coulombic efficiency throughout the cycles. To test the performance of these solvates with a high voltage cathode (LiNi1/3Mn1/3Co1/3O2, NMC), NMC was chosen and cycled with the UG:glyme solvate (2:1) at a C rate of C/10 and compared with LP30 electrolytes. Commercially, when carbonate solvents are used in the electrolyte, carbon paper current collectors are employed to avoid Al corrosion 249


Article


Figure 7. Initial voltage profiles at C/10 of NMC cathode laminates on carbon paper current collectors with LP30 and UG:LiFSI (2:1) at RT.

Figure 8. Voltage profile of the first two cycles of sulfur electrodes with (a) DME:DIOX:1 M LiTFSI electrolyte and (b) UG:LiTFSI (2:1) electrolyte; (c) charge/discharge capacity of sulfur electrodes with UG:LiTFSI (2:1) electrolyte.

4. CONCLUSIONS

Table 1. Compatibility of Commonly Used Electrolyte Solvents with Cathode Materials

The results suggest that such custom synthesized PEG ether with unsymmetrical (ethyl and butyl) end groups is a potential candidate for high conducting electrolytes. Moreover the enhanced electrochemical stability and low crystallization temperature combined with good C-rate capability show the feasibility of such electrolyte compositions for use in lithium secondary batteries employing LiFePO4 or even NMC based electrodes. These electrolytes when employed with sulfur electrodes showed less polysulfide solubility ,enabling the

compatibility with cathode materials electrolyte solvents

LiFePO4

NMC

sulfur

unsymmetric glymes symmetric glymes carbonates dioxalane

√ √ √ √

√ √ √ X

√ X X X

250


Article


solid polymer electrolytes: advances and perspectives. Chem. Soc. Rev. 2017, 46, 797−815. (11) Wang, J.; Yamada, Y.; Sodeyama, K.; Chiang, C. H.; Tateyama, Y.; Yamada, A. Superconcentrated electrolytes for a high-voltage lithium-ion battery. Nat. Commun. 2016, 7, 12032. (12) Suo, L.; Hu, Y.-S.; Li, H.; Armand, M.; Chen, L. A new class of Solvent-in-Salt electrolyte for high-energy rechargeable metallic lithium batteries. Nat. Commun. 2013, 4, 1481. (13) Xue, Z.; He, D.; Xie, X. Poly (ethylene oxide)-based electrolytes for lithium-ion batteries. J. Mater. Chem. A 2015, 3, 19218−19253. (14) McLin, M.; Angell, C. Frequency-dependent conductivity, relaxation times, and the conductivity/viscosity coupling problem, in polymer-electrolyte solutions: LiClO4 and NaCF3SO3 in PPO 4000. Solid State Ionics 1992, 53, 1027−1036. (15) Xu, W.; Wang, L.-M.; Angell, C. A. PolyMOB”−lithium salt complexes: from salt-in-polymer to polymer-in-salt electrolytes. Electrochim. Acta 2003, 48, 2037−2045. (16) Angell, C.; Xu, K.; Zhang, S.; Videa, M. Variations on the saltpolymer electrolyte theme for flexible solid electrolytes. Solid State Ionics 1996, 86, 17−28. (17) Zhang, S.; Chang, Z.; Xu, K.; Angell, C. A. Molecular and anionic polymer and oligomer systems with microdecoupled conductivities. Electrochim. Acta 2000, 45, 1229−1236. (18) Tamura, T.; Hachida, T.; Yoshida, K.; Tachikawa, N.; Dokko, K.; Watanabe, M. New glyme−cyclic imide lithium salt complexes as thermally stable electrolytes for lithium batteries. J. Power Sources 2010, 195, 6095−6100. (19) Tamura, T.; Yoshida, K.; Hachida, T.; Tsuchiya, M.; Nakamura, M.; Kazue, Y.; Tachikawa, N.; Dokko, K.; Watanabe, M. Physicochemical properties of glyme−Li salt complexes as a new family of room-temperature ionic liquids. Chem. Lett. 2010, 39, 753− 755. (20) Ueno, K.; Yoshida, K.; Tsuchiya, M.; Tachikawa, N.; Dokko, K.; Watanabe, M. Glyme−lithium salt equimolar molten mixtures: concentrated solutions or solvate ionic liquids? J. Phys. Chem. B 2012, 116, 11323−11331. (21) Yoshida, K.; Nakamura, M.; Kazue, Y.; Tachikawa, N.; Tsuzuki, S.; Seki, S.; Dokko, K.; Watanabe, M. Oxidative-stability enhancement and charge transport mechanism in glyme−lithium salt equimolar complexes. J. Am. Chem. Soc. 2011, 133, 13121−13129. (22) US EPA (US Environmental Protection Agency). Ethylene glycol ether: Significant new use rule; 40 CFR Part 9 and 721, Federal Register, Vol. 79, No. 241, 2014. (23) Henderson, W. A. Glyme− lithium salt phase behavior. J. Phys. Chem. B 2006, 110, 13177−13183. (24) Qian, J.; Henderson, W. A.; Xu, W.; Bhattacharya, P.; Engelhard, M.; Borodin, O.; Zhang, J.-G. High rate and stable cycling of lithium metal anode. Nat. Commun. 2015, 6, 6362.

cycling of the sulfur cells thereby indicating the use of these electrolytes, with further optimization, in lithium sulfur batteries. An added advantage of using these glyme solvents is the stable and safe operation of lithium batteries due to their high thermal stability. As mentioned, the EPA norms of declaring glymes with methyl end groups as toxic substances urged us to find such alternative glymes for employing in lithium batteries along with their versatile use in lithium−sulfur batteries.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.chemmater.7b04270. Three figures showing the 1H NMR spectrum and DSC thermogram of unsymmetric glyme and glyme solvates (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*(M.A.) [email protected]. ORCID

Michel Armand: 0000-0002-1303-9233 Author Contributions

All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS The authors would like to acknowledge Dr. Laida Otaegui CIC Energi GUNE for viscosity measurements. REFERENCES

(1) Bouchet, R.; Maria, S.; Meziane, R.; Aboulaich, A.; Lienafa, L.; Bonnet, J.-P.; Phan, T. N.; Bertin, D.; Gigmes, D.; Devaux, D.; et al. Single-ion BAB triblock copolymers as highly efficient electrolytes for lithium-metal batteries. Nat. Mater. 2013, 12, 452−457. (2) Armand, M.; Grugeon, S.; Vezin, H.; Laruelle, S.; Ribière, P.; Poizot, P.; Tarascon, J.-M. Conjugated dicarboxylate anodes for Li-ion batteries. Nat. Mater. 2009, 8, 120−125. (3) Gachot, G.; Grugeon, S.; Armand, M.; Pilard, S.; Guenot, P.; Tarascon, J.-M.; Laruelle, S. Deciphering the multi-step degradation mechanisms of carbonate-based electrolyte in Li batteries. J. Power Sources 2008, 178, 409−421. (4) Scrosati, B.; Hassoun, J.; Sun, Y.-K. Lithium-ion batteries. A look into the future. Energy Environ. Sci. 2011, 4, 3287−3295. (5) Gao, X.-P.; Yang, H.-X. Multi-electron reaction materials for high energy density batteries. Energy Environ. Sci. 2010, 3, 174−189. (6) Xu, W.; Wang, J.; Ding, F.; Chen, X.; Nasybulin, E.; Zhang, Y.; Zhang, J.-G. Lithium metal anodes for rechargeable batteries. Energy Environ. Sci. 2014, 7, 513−537. (7) Tatsumisago, M.; Mizuno, F.; Hayashi, A. All-solid-state lithium secondary batteries using sulfide-based glass−ceramic electrolytes. J. Power Sources 2006, 159, 193−199. (8) Fergus, J. W. Ceramic and polymeric solid electrolytes for lithium-ion batteries. J. Power Sources 2010, 195, 4554−4569. (9) Thangadurai, V.; Kaack, H.; Weppner, W. J. Novel Fast Lithium Ion Conduction in Garnet-Type Li5La3M2O12 (M= Nb, Ta). J. Am. Ceram. Soc. 2003, 86, 437−440. (10) Zhang, H.; Li, C.; Piszcz, M.; Coya, E.; Rojo, T.; RodriguezMartinez, L. M.; Armand, M.; Zhou, Z. Single lithium-ion conducting 251


High Conductivity Solvates with Unsymmetrical ... - ACS Publications

Recommend Documents