Improving Electrochemical Stability and Low‐Temperature Performance with Water/Acetonitrile Hybrid Electrolytes

Abstract

headlines, [3][4][5] for which the highly flammable nonaqueous electrolytes used in LIBs are mainly responsible. It is under this context that aqueous LIBs (ALIBs) are revisited as a fundamental solution to safety, despite their low energy densities due to the narrow electrochemical stability window of water. [2,[6][7][8][9][10] Recently, a new class of high-voltage aqueous electrolyte was discovered by dissolving 21 molality (mol) lithium bis(trifluoromethane sulfonyl) imide (LiTFSI) in 1 kg of water. Such a "water-in-salt" electrolyte (WiSE) expands the electrochemical stability window from 1.23 to 3.0 V, which supports a 2.5 V chemistry using LiMn 2 O 4 (LMO) cathode and Mo 6 S 8 anode to stably deliver ≈100 Wh kg −1 for thousand cycles. The significantly improved electrochemical stability therein mainly comes from the depletion of free water molecules and the formation of an anion-derived solid-electrolyte interphase (SEI) on the anode surface. [9] However, because of the intrinsic repulsion of anions and adsorption of the Li + -4(H 2 O) solvates by the negatively polarized anode surface, [4] the formation of such an anion-derived SEI has been impossible below 1.9 V versus Li + /Li. [9] This "cathodic challenge" has essentially excluded many desired energy-dense anodes that operate at low potentials such as Li-metal, graphite, or silicon. Even Li 4 Ti 5 O 12 (LTO) that operates at mild potential (≈1.70 V Li + /Li in WiSE) suffers from irreversibility, because it sits right on the edge of the cathodic limit in WiSE. Efforts aiming to resolve the "cathodic challenge" with additional lithium salts such as lithium trifluoromethane sulfonate (LiOTf) [11] or lithium bis(pentafluoroethane sulfonyl) imide (LiBETI) [10] achieved limited success, because solubility limits of the salts impose restrictions on how high their concentration can go, while the effectiveness of added anions still faces intrinsic resistance from anode surface against their accumulation at inner-Helmholtz layer, not to mention that these additional salts further worsen the already problematic viscosity and ionic conductivity of WiSE. Introducing a nonaqueous solvent, dimethyl carbonate (DMC), [12] into WiSE expands the electro chemical window of the hybrid electrolyte to 4.1 V, because the neutral solvent is less sensitive to anode repulsion and hence participates in interphasial chemistry more easily than anions. The additional protection from an SEI consisting of both anion-and solvent-derived products enables LTO operation Although the "water-in-salt" electrolyte has significantly expanded the electrochemical stability window of aqueous electrolytes from 1.23 to 3 V, its inevitable hydrogen evolution under 1.9 V versus Li + /Li prevents the practical use of many energy-dense anodes. Meanwhile, its liquidus temperature at 17 °C restricts its application below ambient temperatures. An advanced hybrid electrolyte is proposed in this work by introducing acetonitrile (AN) as co-solvent, which minimizes the presence of interfacial water at the nega...

Aqueous Co‐Solvent in Zwitterionic‐based Protic Ionic Liquids as Electrolytes in 2.0 V Supercapacitors

“…Moreover, capacitance, coulombic efficiency, and IR drop versus current density of [EOAH] + [BET][MSA] − in 2 H 2 O/ACN/DMSO were in the range of (or even better than) those of LiTFSI not only in the same ternary liquid mixture but also in other solvent mixtures used previously (e. g., LiTFSI/H 2 O/ACN with molar ratios of 1 : 2.6 : 3.7 [5,55] and 1 : 1.1 : 1.1; [56] Figure 10d, Figures S17 and S23–S25, and Table S11 in the Supporting Information). Ragone plots corroborated that the performance of [EOAH] + [BET][MSA] − in 2 H 2 O/ACN/DMSO was similar to that of LiTFSI in the same solvent mixture as well as in the other solvent mixtures mentioned above [55,56] (Figure 10e, Figure S26, and Table S12 in the Supporting Information). Only LiTFSI/H 2 O/ACN with molar ratio of 1 : 2.6 : 3.7 performed better in terms of power density, but, as mentioned above, the flammability of this mixture might be a serious drawback for practical applications.…”

“…We described the preparation of novel [5,55,56] These results demonstrate how the judicious composition design of ZPILs helped overcoming the narrow ESWs typical of PILs. Thus, WcS-in-ZPIL electrolytes exhibited capacitances of up to approximately 83 F g À 1 at current densities of 1 A g À 1 and capacity retentions of approximately 90 % after 6000 cycles when operating at a voltage of 2.0 V and a current density of 4 A g À 1 .…”

“…In our case, . Nonetheless, it is worth noting that electrodes used in this work were different than those used in previous works, [55,56] thus making a straightforward comparison between data provided in this work and previously reported data difficult. Finally, we investigated the cycling performance of the SC using [EOAH] + [BET][MSA] À in H 2 O/ACN/DMSO as the electrolyte.…”

“…This ternary liquid mixture was of particular interest, mitigating the problem of flammability typically exhibited by ACN [54] and even by binary mixtures based on H 2 O/ACN when the H 2 O content is too low (Figure 9 and Figure S22 in the Supporting Information) [55,56] while somehow retaining a low viscosity (more so than 2 H 2 O/DMSO, see Table S7 in the Supporting Information) that eventually helped for a well‐balanced performance when used as electrolyte for EDLCs. Moreover, capacitance, coulombic efficiency, and IR drop versus current density of [EOAH] + [BET][MSA] − in 2 H 2 O/ACN/DMSO were in the range of (or even better than) those of LiTFSI not only in the same ternary liquid mixture but also in other solvent mixtures used previously (e. g., LiTFSI/H 2 O/ACN with molar ratios of 1 : 2.6 : 3.7 [5,55] and 1 : 1.1 : 1.1; [56] Figure 10d, Figures S17 and S23–S25, and Table S11 in the Supporting Information). Ragone plots corroborated that the performance of [EOAH] + [BET][MSA] − in 2 H 2 O/ACN/DMSO was similar to that of LiTFSI in the same solvent mixture as well as in the other solvent mixtures mentioned above [55,56] (Figure 10e, Figure S26, and Table S12 in the Supporting Information).…”

“…too low (Figure and Figure S22 in the Supporting Information) [55,56] while somehow retaining a low viscosity (more so than H 2 O/DMSO, see Table S7 in the Supporting Information) that eventually helped for a well-balanced performance when used as electrolyte for EDLCs. Moreover, capacitance, coulombic efficiency, and IR drop versus current density of [EOAH] + [BET][MSA] À in 2 H 2 O/ACN/DMSO were in the range of (or even better than) those of LiTFSI not only in the same ternary liquid mixture but also in other solvent mixtures used previously (e. g., LiTFSI/H 2 O/ACN with molar ratios of 1 : 2.6 : 3.7 [5,55] and 1 : 1.1 : 1.1; [56] Figure 10d, Figures S17 and S23-S25, and [55,56] (Figure 10e, Figure S26, and Table S12 in the Supporting Information). Only LiTFSI/H 2 O/ACN with molar ratio of 1 : 2.6 : 3.7 performed better in terms of power density, but, as mentioned above, the flammability of this mixture might be a serious drawback for practical applications.…”