Study on Fundamental Properties of Solvate Electrolytes and Their Application in Batteries

Abstract

Li salts and polar solvents form solvates, and certain solvates have low melting temperatures and remain in a liquid state at room temperature. Liquid-state solvates exhibit ionic conductivity and can be used as electrolytes in lithium batteries. The author and co-workers have systematically studied the interactions of Li⁺ ions with solvents and anions, Li⁺-coordination structures, thermal properties, transport properties, and electrochemical properties in molten-solvate electrolytes. In molten solvates, almost all solvent molecules are coordinated to Li⁺ ions, and uncoordinated (free) solvents are rare. Additionally, anions are involved in the coordination of the Li⁺ ions. The molten solvate electrolytes show non-flammability and negligible vapor pressure at room temperature because of the extremely low concentration (activity) of the free solvent, which can improve the thermal stability of Li batteries. The low activity of the free solvent results in a wide electrochemical window of the molten-solvate electrolytes, thereby suppressing undesired side reactions in Li batteries. The activity of the free solvent in the electrolytes significantly affects the electrochemical reaction processes, such as the reduction reaction of sulfur (S₈) in a Li–S battery and the oxygen reduction reaction (ORR) in a Li–air battery. The solubility of the reaction intermediates of the S₈ cathode and the ORR decreases with the decrease in solvent activity, which enables the highly efficient charge–discharge of Li–S and Li–air batteries. In molten solvates, Li⁺ ions diffuse and migrate by exchanging ligands (solvents and anions). Certain molten-solvate electrolytes show high Li⁺ ion transference numbers over 0.5, and these high transference numbers are useful in mitigating the concentration overpotential during the charging and discharging of Li batteries at high current densities.

1. Introduction

The electrolyte solutions for Li batteries are composed of Li salts and polar solvents. Li⁺ ions are solvated using the polar solvents to produce solvate cations [Li(solvent)_x]⁺, and the salts dissociate and exhibit ionic conductivity. Numerous combinations of Li salts and solvents are available for preparing electrolyte solutions (Fig. 1). Current Li-ion batteries (LIBs) use carbonate-based electrolytes, typically, LiPF₆ dissolved in a mixed solvent of ethylene carbonate (EC) and linear carbonates, such as dimethyl carbonate (DMC). The composition of carbonate-based electrolytes has been optimized to achieve high performance in LIBs.^1,2 LiPF₆ has been adopted as the electrolyte salt for LIBs because PF₆⁻ is oxidatively stable and a soft anion (weak Lewis base). PF₆⁻ weakly interacts with the solvated cations [Li(solvent)_x]⁺ in the solution, and a high degree of salt dissociation can be achieved. The concentration of LiPF₆ in the electrolyte solution is approximately 1 mol dm⁻³ because the ionic conductivity reached the maximum at this concentration. One of the issues of the conventional carbonate-based electrolyte solutions is thermal instability. Carbonate solvents are flammable, which increases the risk of the combustion of LIBs under severe abuse conditions.

Figure 1.

Chemical structures and abbreviations of the various solvents and anions of Li salts. Reproduced with permission from Ref. 18.

To improve the thermal stability of LIBs, room temperature molten salts, i.e., ionic liquids (ILs), have been investigated as non-flammable electrolytes.³ ILs composed of onium cations and soft anions have been widely investigated as electrolytes for LIBs. To provide the Li⁺ ion conductivity, Li salts are mixed with these ILs. In ILs, Li⁺ ions, onium cations, and anions contribute to ionic conduction, resulting in a low Li⁺ ion transference number. The addition of high concentrations of Li salts (∼1 mol dm⁻³) to ILs increases their viscosity and decreases their ionic conductivity, which is not favorable for producing LIBs with high-power density. Our group has investigated the quasi-ionic liquids, called “solvate ionic liquids (SILs),” as thermally stable electrolytes for LIBs.^4–6 When the author joined Yokohama National University in 2008, Prof. Watanabe’s group had already started the study on SILs: the molten-glyme solvates of Li salts. SILs contain a significantly high concentration of Li salt (∼3 mol dm⁻³) and show an ionic conductivity of ca. 1 mS cm⁻¹ at room temperature. SILs are mixtures of solvents and Li salts, which is common with other conventional electrolyte solutions. Initially, the author considered that the only difference between the SILs and conventional electrolyte solutions is the Li salt concentration. However, this makes the physicochemical properties of SILs significantly different from those of conventional electrolyte solutions. Subsequently, the author has studied the fundamental properties and battery applications of SILs. Glymes (CH₃O(CH₂CH₂O)_nCH₃) are oligoethers that form complexes with Li salts.^7–11 The ether oxygen atoms have lone pairs, and electrostatic and induction interactions between the oxygen atoms and Li⁺ result in the formation of a complex.^12,13 The coordination number of Li⁺ is typically 4–6 in solution.^14–16 The triglyme (G3) and tetraglyme (G4) molecules have four and five oxygen atoms, respectively (Fig. 1). G3 and G4 coordinate with Li⁺ ions in a molar ratio of 1 : 1. Figure 2 shows the optimized structure of a G3 solvate of Li bis(trifluoromethanesulfonyl)amide (LiTFSA).⁴ In the solvate structure, the G3 wraps around Li⁺ and forms a complex cation [Li(G3)]⁺. Further, G4 forms a complex cation, [Li(G4)]⁺, with a structure similar to that of [Li(G3)]⁺. Specific G3 and G4 solvates of Li salts have low melting temperatures and maintain a liquid state at room temperature. Spectroscopic analysis revealed that the complex structures of [Li(G3)]⁺ and [Li(G4)]⁺ are maintained in the molten solvates, and uncoordinated (free) glyme are rare.¹⁷ G3 and G4 molecules serve as multidentate ligands and form long-lived complexes with Li⁺ owing to the chelate effect. The molten G3 (or G4) solvates are ionic liquids composed of complex cations [Li(G3 or G4)]⁺ and counter anions. Similar to conventional ILs, SILs exhibit nonflammability and negligible vapor pressure.

Figure 2.

Most stable structure of [Li(G3)][TFSA] estimated by ab initio calculation (HF/6-311G** level). Reproduced with permission from Ref. 4. Copyright 2014 PCCP Owner Societies.

The concept of SILs can easily be extended to solvates other than the glyme solvates of Li salts. Additionally, various solvents, other than glymes, form low-melting solvates of Li salts. The author and co-workers have concurrently researched glyme-based SILs and other molten solvates.¹⁸ Further, other research groups have investigated highly concentrated electrolytes (HCEs) containing Li salts over 3 mol dm⁻³ and reported that specific HCEs can improve the performance of Li batteries.^19–28 The HCEs, in which all solvent molecules are coordinated to Li⁺, are regarded as molten solvates. In this paper, the author mainly focuses on the fundamental properties of the molten solvates of specific Li salts and the applications of molten solvates to next-generation batteries.

2. Solvent Activity in Molten-Solvate Electrolytes

The liquid structures of the electrolyte solutions were determined by Raman spectroscopy. The Raman bands of the ligands (solvents and anions) were sensitive to the formation of complexes with Li⁺ ions. The ligands bound to Li⁺ were distinguished from the free ones using Raman spectroscopy. Figure 3 shows the fractions of free and bound dimethyl sulfoxide (DMSO) in the LiTFSA/DMSO solution as a function of the molar ratio of LiTFSA to DMSO.²⁹ The fraction of DMSO bound to Li⁺ increased and free DMSO decreased with the increase in LiTFSA concentration. In the solution with the molar ratio [LiTFSA]/[DMSO] < 1/4, the solvation number of Li⁺ was estimated to be 3.7, which is consistent with the reported value.³⁰ Further, 9 % of free DMSO was observed in the 2.3 mol dm⁻³ solution ([LiTFSA]/[DMSO] = 1/4), and only trace amounts of free DMSO were present in the high-concentration LiTFSA/DMSO solutions, above 2.5 mol dm⁻³. Therefore, the liquids with a molar ratio of [LiTFSA]/[DMSO] > 1/4, which contain no free solvent, are molten solvates rather than solutions. Further, Raman spectroscopy revealed that TFSA anions were involved in Li⁺ coordination in the highly concentrated solution with a molar ratio of [LiTFSA]/[DMSO] > 1/4 and form contact-ion pairs (CIPs) and ionic aggregates (AGGs). Based on the Raman spectroscopic data, a schematic illustration of the liquid structure of the molten solvates has been drawn, as shown in Fig. 4.¹⁸

Figure 3.

Fractions of the integrated intensities of Raman peaks corresponding to free DMSO, I_f/(I_f + I_b), and bound DMSO, I_b/(I_f + I_b), in the Li[TFSA]/DMSO solutions. Reproduced with permission from Ref. 29. Copyright 2017 American Chemical Society.

Figure 4.

Schematic illustration of solvation structure in a molten Li salt solvate electrolyte (>3 mol dm⁻³); in a molten solvate electrolyte, Li⁺ ions form complexes with solvents and anions, and free solvent hardly exists. Reproduced with permission from Ref. 18.

The concentration (activity) of the free solvent in the electrolyte solution decreased with the increase in salt concentration and became negligible in the molten solvate. The attractive force between the Li⁺ cation and solvent was caused by the electrostatic and induction interactions between the ion and solvent.^12,13 This suppressed the evaporation of the solvent bound to Li⁺ in the solution, while the free solvent easily evaporated.^4,10 Therefore, the vapor pressure of an electrolyte solution decreases with the increase in salt concentration, and the molten solvates show non-volatility at ambient temperature. Additionally, the high concentration of Li salts and negligible activity of the solvent effectively improved the flame retardancy of the molten solvates. Consequently, the molten solvates can be used as thermally stable electrolytes in Li batteries.

3. Electrochemistry in Molten Solvate Electrolytes

3.1 Electrode potential of Li metal

The activity of the free solvent has a significant effect on the electrode potential of Li metal.¹⁷ At the interface between the Li metal electrode and an aprotic electrolyte solution containing Li salt, the solvated Li⁺, [Li(solvent)_n]⁺, is reduced to produce Li metal, and the desolvation of Li⁺ occurred as follows.   

\begin{equation} \text{[Li(solvent)$_{n}$]$^{+}$} + \text{e$^{-}$}\rightleftarrows \text{Li} + \text{$n$ solvent}. \end{equation} (1)

The equilibrium electrode potential, E, of this electrochemical reaction is described as follows according to the Nernst equation:   

\begin{equation} E = E^{\circ} + \frac{RT}{F}\ln \frac{a_{\text{[Li(solvent)${_{n}}$]${^{+}}$}}}{a_{\text{solvent}}^{n}}, \end{equation} (2)

where E° is the standard electrode potential, R is the gas constant, T is the absolute temperature, F is the Faraday constant, and $a_{[\text{Li}(\text{solvent})_{n}]^{ + }}$ and a_solvent are the activities of the solvated Li⁺ and free solvent, respectively. Figure 5 shows the plots of the electrode potential of the Li metal against log c_Li (where c_Li is the concentration of Li salt) in LiTFSA/ether solutions.¹⁸ The electrode potential linearly changes with log c_Li at concentrations lower than 1 mol dm⁻³, independent of the solvent species. The electrode potentials in the solutions with solvents, such as diglyme (G2), G3, and G4, which function as multidentate ligands, are lower than those with tetrahydrofuran (THF) and monoglyme (G1). This suggests that G2, G3, and G4 more stabilize Li⁺ than monodentate THF and bidentate G1 because the multidentate ligands (solvents) can form more stable complexes owing to the chelate effect.³¹ At concentrations higher than 1 mol dm⁻³, the electrode potential shifts nonlinearly in the positive direction with the increase in log c_Li.^17,32 This can be attributed to the decrease in the activity of the free solvent in the highly concentrated electrolytes. With the increase in the Li salt concentration, the fraction of free solvent in the electrolyte solution decreases and becomes negligible in molten solvates, which was revealed by Raman spectroscopy.¹⁷ As predicted from Eq. 2, $a_{[\text{Li}(\text{solvent})_{n}]^{ + }}$ and a_solvent affect the electrode potential. E becomes positive with the increase in $a_{[\text{Li}(\text{solvent})_{n}]^{ + }}$ and decrease in a_solvent. Therefore, the electrode potential is not simply linear to log c_Li at concentrations higher than 1 mol dm⁻³.

Figure 5.

Plots of the Li/Li⁺ electrode potential against a common logarithm of the Li salt concentration in G4 (tetraglyme), G3 (triglyme), G2 (diglyme), G1 (monoglyme), and THF–LiTFSA mixtures at 30 °C. Reference electrode was Li/Li⁺ in 1 mol dm⁻³ LiTFSA/G3. Reproduced with permission from Ref. 18.

3.2 Oxidative stability

The oxidative stability of the electrolyte solutions is improved by increasing the Li-salt concentration. For example, the linear sweep voltammograms of LiTFSA/G3 solutions ([Li(G3)_x][TFSA]; x is the molar ratio of G3 to LiTFSA) measured using two-electrode cells (Li/Pt) are shown in Fig. 6a.⁹ Ether solvents, such as G1 and G3, oxidatively decompose at electrode potentials higher than 4 V vs. Li/Li⁺; therefore, these solvents have not been used in 4 V-class Li batteries. The anodic current corresponding to the oxidation of G3 flows at 4 V in the solution containing excess G3 ([Li(G3)_x][TFSA] with x > 1), as shown in the inset of Fig. 6a. However, the anodic current decreases with decreasing x, and the onset of the anodic current shifts to 4.5 V for the SIL [Li(G3)₁][TFSA]. This is because the concentration of free G3 decreases with the decrease in x, and free G3 barely exists in the SIL. Additionally, the highest occupied molecular orbital (HOMO) energy level of G3 changes upon complexation with Li⁺. The HOMO energy levels of isolated G3 and [Li(G3)₁][TFSA] were estimated to be −11.45 and −12.10 eV, respectively, by ab initio calculations.⁹ The lone pairs of the ether oxygen of a G3 are attracted to Li⁺ because of the electrostatic and induction interactions, resulting in the decrease in the HOMO energy level of G3. The oxidation of G3 at the electrode/electrolyte interface occurs when electrons are extracted from the HOMO. Therefore, lowering the HOMO energy level of G3 through complexation with Li⁺ results in the enhancement of the oxidative stability of the electrolyte. Resultantly, the [Li(G3)₁][TFSA], in which almost all G3 molecules are involved in the Li⁺-coordination, becomes stable up to 4.5 V vs. Li/Li⁺. The solvate IL [Li(G3)₁][TFSA] was used as the electrolyte for a 4 V-class Li/LiCoO₂ battery (Fig. 6b). The battery can be operated for over 200 charge–discharge cycles regardless of the use of the ether-based electrolyte, suggesting that side reactions, such as decomposition of the electrolyte, do not significantly occur during battery operation. Moreover, the stable operation of the Li/LiCoO₂ cell indicates that the corrosion of Al current collector of the cathode is effectively suppressed, regardless of the use of LiTFSA salt. The oxidation of Al occurs at electrode potential higher than 3.8 V vs. Li in electrolytes containing LiTFSA,^33–35 and the dissolution of the Al oxidation products is considered to cause the Al corrosion. The solvation of the Al oxidation products may be inhibited in the [Li(G3)₁][TFSA] because the activity of free G3 solvent is negligible, resulting in the suppression of the Al corrosion.³⁶

Figure 6.

(a) Linear sweep voltammograms of [Li(glyme)_x][TFSA] (x = 1, 4, 8, and 20) measured using two-electrode cells (Li/Pt) at a scan rate of 1 mV s⁻¹ at 30 °C; inset depicts an enlarged view of current density; (b) charge and discharge curves of the [Li metal anode | [Li(G3)][TFSA] electrolyte | LiCoO₂ cathode] cell measured with a current density of 50 µA cm⁻² at 30 °C. Reproduced with permission from Ref. 9.

3.3 Li-ion intercalation into graphite

Graphite is the most popular anode material used in LIBs. In practical LIBs with carbonate-based electrolytes containing EC, the EC molecules reductively decompose on the graphite anode during the initial stage of the 1st charging process. Further, the decomposition products form a passivation layer, solid electrolyte interphase (SEI), on the graphite electrode.^1,2 The SEI is a Li⁺ ion conductive but an electronic insulator, and the reductive decomposition of the electrolyte is suppressed after the formation of the SEI. Additionally, the SEI inhibits the co-intercalation of the solvent with Li⁺ into the graphite electrode, and the desolvation of Li⁺ ions occurs at the SEI during charging. Subsequently, Li⁺ ions solely intercalate into graphite to form Li⁺–graphite intercalation compounds (Li⁺–GICs). After the formation of the SEI, reversible Li⁺ intercalation and deintercalation occur through the SEI during charging and discharging, respectively. In ether-based electrolyte solutions containing excess solvent, an effective SEI is not formed on the graphite electrode, and the co-intercalation of the ether solvent with Li⁺ occurs, destroying the crystal structure of graphite. Figure 7a shows the cyclic voltammograms of the graphite electrodes in LiTFSA/G3 solutions measured using two-electrode cells (Li/graphite).³² For electrolytes with excess G3 (x > 1 in [Li(G3)_x][TFSA]), the cathodic current caused by the reduction reaction of graphite (C_n) and the co-intercalation of G3 and Li⁺, i.e., the intercalation of solvated Li⁺ ion [Li(G3)₁]⁺, starts to flow at 1.2 V, corresponding to the following reaction.   

\begin{equation} \text{C$_{n}$} + \text{[Li(G3)$_{1}$]$^{+}$} + \text{e$^{-}$} \to \text{[Li(G3)$_{1}$]C$_{n}$}. \end{equation} (3)

The anodic current corresponding to the deintercalation was considerably smaller than the cathodic current, suggesting that the co-intercalation into the graphite electrode is irreversible (Fig. 7b). In contrast, for the SIL, [Li(G3)₁][TFSA], the cathodic current in the potential range of 1.2–0.2 V is significantly smaller than that for the electrolytes with excess G3. At potential lower than 0.2 V, the cathodic current for the intercalation of Li⁺ into graphite is observed, and the anodic current for Li⁺ deintercalation is observed. The co-intercalation of G3 and Li⁺ is suppressed, and the desolvation of Li⁺ occurs at the interface between graphite and [Li(G3)₁][TFSA] as follows.   

\begin{equation} \text{C$_{n}$} + \text{[Li(G3)$_{1}$]$^{+}$} + \text{e$^{-}$} \rightleftarrows \text{LiC$_{n}$} + \text{G3}. \end{equation} (4)

The electrode potential for the interaction of solvated Li⁺ ions (Eq. 3) depends on the activity of [Li(G3)₁]⁺ in the electrolyte according to the Nernst equation. Conversely, the electrode potential for Li⁺ intercalation, i.e., the desolvation of [Li(G3)₁]⁺ (Eq. 4) depends on the activity of [Li(G3)₁]⁺ and the activity of free G3, as for the Li metal electrode (Eq. 2). Therefore, the difference in the electrode potentials for the co-intercalation (Eq. 3) and Li⁺ intercalation (Eq. 4) decreases with the decrease in the activity of free G3. The activity of free G3 in the [Li(G3)₁][TFSA] is negligible, and the two reactions, the co-intercalation and Li⁺ intercalation (desolvation of Li⁺), compete at the graphite/[Li(G3)₁][TFSA] interface. Additionally, an SEI may be formed on the graphite electrode in the [Li(G3)₁][TFSA] electrolyte. In the [Li(G3)₁][TFSA] electrolyte, TFSA⁻ anions form CIPs with Li⁺ ions.³¹ According to the computational studies reported by Yamada et al., the complexation of the TFSA⁻ anion with Li⁺ ions results in the lowering of the energy level of the lowest unoccupied molecular orbital (LUMO) of TFSA.^37,38 This induces the reductive decomposition of TFSA⁻ anions on the graphite electrode, and the decomposition products form an SEI, which may block the co-intercalation of G3 into the graphite.

Figure 7.

(a) Cyclic voltammograms of the [Li metal | [Li(G3)_x][TFSA] electrolyte | graphite] cells at a scan rate of 0.01 mV s⁻¹ at 60 °C; (b) schematic illustration of the intercalations of [Li(G3)]⁺ and Li⁺ into graphite electrodes. Reproduced with permission from Ref. 32. Copyright 2014 American Chemical Society.

Regarding the reductive stability (cathodic limit) of the electrolytes, the complexation of ligands (solvents and anions) with Li⁺ gives significant effects. The complexation lowers the LUMO energy levels of the ligands bound to Li⁺, and the bound ligands reductively decompose at more positive electrode potentials compared with free ligands. Moreover, the activity of free ligands decreases with increasing the Li salt concentration, and this causes the positive shifts of the electrode potentials of the electrochemical reactions involving Li⁺ ions (vide supra). Furthermore, the SEI formation on the electrode affects the cathodic limit of the electrolytes. Therefore, we should consider several factors to understand the reductive decomposition and the stability of the electrolytes. In the highly concentrated electrolytes or molten solvate electrolytes, the anion-derived SEI may play a vital role in extending the cathodic limit of the electrolytes, as Yamada et al reported.^37,38

3.4 Reduction reactions of S₈ and O₂

The elemental S₈ can be electrochemically reduced to Li₂S in aprotic electrolytes, and this reaction is reversible.   

\begin{equation} \text{S$_{8}$} + \text{16Li$^{+}$} + \text{16e$^{-}$} \rightleftarrows \text{8Li$_{2}$S}. \end{equation} (5)

The theoretical capacity of S₈ is 1672 mAh g⁻¹, which is more than 10 times that of representative cathode material, LiCoO₂ (140 mAh g⁻¹), used in LIBs. Therefore, a high-energy-density Li–S rechargeable battery can be realized by combining S₈ as the cathode material with Li metal as the anode material. However, Li–S batteries with conventional electrolyte solutions have several limitations.^39,40 One of the serious issues is the dissolution of the reaction intermediates of the S₈ cathode. During the reduction reaction of S₈, Li polysulfides (LPSs), Li₂S_x, where 2 ≤ x ≤ 8, dissolve in the electrolyte solutions. This results in a low discharge capacity, short discharge–charge cycle life, and low Coulombic efficiency of the discharge–charge of the Li–S battery. Figure 8a shows the discharge–charge curves of a Li–S cell with a [Li(G4)₄][TFSA] electrolyte, which contains free G4 molecules.⁴¹ The cell exhibits a high capacity of over 900 mAh g⁻¹ in the initial cycle; however, the discharge capacity rapidly fades with the increase in the discharge–charge cycle number. Further, the Coulombic efficiency of discharge–charge decreases with the increase in the cycle number. The charging capacity increases after the 40th cycle, while the discharging capacity continuously decreases. Finally, the charging of the Li–S cell did not reach the cut-off voltage (3.3 V), and the charging capacity of the 55th cycle exceeded 2,000 mAh g⁻¹, which is greater than the theoretical capacity of the S₈ cathode. This phenomenon is caused by the redox shuttle effect of the LPSs. A dissolved LPS diffuses from the S₈ cathode to the Li metal anode and is reduced to produce another LPS with a relatively short chain. The relatively short LPS diffuses back to the S₈ cathode and is oxidized to produce a relatively long LPS. LPSs act as redox shuttles between the Li anode and S₈ cathode, causing a low Coulombic efficiency of discharge–charge. The LPSs should be solvated by G4 to dissolve in the electrolyte. The [Li(G4)₄][TFSA] electrolyte contains a large amount of free G4 molecules, and LPSs can easily dissolve in the electrolyte. Over 6 mol dm⁻³ of S (atomic concentration of S) can dissolve into the [Li(G4)₄][TFSA]. The solubility of LPSs can be decreased by increasing the Li salt concentration, viz., decreasing the activity of the free solvent in the electrolyte solution. In the SIL [Li(G4)₁][TFSA], almost all the G4 molecules are involved in the solvation of Li⁺ ions, and free G4 scarcely exists in the liquid. Therefore, the LPSs cannot receive solvation in the SIL, resulting in a considerably low LPS solubility of ca. 60 mmol dm⁻³, which is 1/100 compared with that in the [Li(G4)₄][TFSA].⁴² Figure 8b shows the discharge–charge curves of a Li–S cell with the [Li(G4)₁][TFSA] SIL. Although the discharge capacity gradually decreases with the increase in the cycle number, the Coulombic efficiency of discharge–charge is over 98 %, which is considerably higher than that of the cell with the [Li(G4)₄][TFSA] electrolyte. The redox shuttle effect can be effectively suppressed using the SIL electrolyte with the negligible activity of free G4, and this is effective in achieving a long cycle life of a Li–S battery. Similarly, high Coulombic efficiency of the discharge–charge of Li–S batteries can be achieved using other molten-solvate electrolytes, such as sulfolane solvate and glutaronitrile solvate of LiTFSA.^43,44

Figure 8.

Discharge–charge curves of Li–S cells with (a) [Li(G4)₄][TFSA] and (b) [Li(G4)][TFSA] measured at 30 °C; (a) is reproduced with permission from Ref. 41. Copyright 2013 American Chemical Society. (b) is reproduced with permission from Ref. 42. Copyright 2013 The Electrochemical Society.

Further, the solvent activity of the electrolyte solution affects the oxygen reduction reaction (ORR) process. The ORR in aprotic electrolyte solutions has been investigated as the cathode reaction of Li–air batteries.^45,46 Oxygen is electrochemically reduced to Li peroxide Li₂O₂ in aprotic solutions containing Li salts, and the Li₂O₂ (solid) is deposited on the cathode surface. The superoxide ion, O₂^−• (radical anion), is formed as a reaction intermediate during the ORR, and O₂^−• ions dissolve into the electrolyte solution. The superoxide ions are highly reactive and attack the species in the electrolyte solutions and the electrode material (carbon), and serious side reactions cause the short lifetime of Li–air batteries. We found that the dissolution of superoxide can be suppressed by increasing the Li salt concentration or decreasing the activity of the free solvent in the electrolyte solution.^29,47 It is postulated that the O₂^−• generated at the cathode surface forms an ion pair with Li⁺, and the produced LiO₂ cannot be solvated in a highly concentrated electrolyte with no free solvent, which is similar to the case of LPSs in the molten-solvate electrolytes. The suppressed LiO₂ dissolution effectively hinders undesirable side reactions in Li–air batteries.

4. Li-Ion-Hopping Conduction in Molten-Solvate Electrolytes

The theory of ion conduction in conventional electrolyte solutions has been considerably established.⁴⁸ In Li salt solutions containing excess solvent, the Li salt dissociates into ions through solvation, and a solvated Li⁺ ion, [Li(solvent)_x]⁺, diffuses/migrates, as shown in Fig. 9a. The hydrodynamic radius of the solvated Li⁺ ions becomes much larger than the ionic radius of Li⁺ in crystalline solids. Furthermore, the [Li(solvent)_x]⁺ is larger than the free solvent and anion, and the Li⁺ ion diffuses slowly, compared with the species in the solution, following Stokes’ law.⁴⁹ The ion conduction mechanism in highly concentrated electrolytes and molten solvates is different from that in conventional electrolyte solutions. We found that a unique Li⁺ conduction mechanism emerges in specific molten solvates of Li salts.⁵⁰ The self-diffusion coefficients of Li⁺ (D_Li), solvent (D_SL), and anion (D_BF4) in LiBF₄/sulfolane (SL) solutions were measured using pulsed-field gradient (PFG) nuclear magnetic resonance (NMR). Figure 9b shows the ratios of the self-diffusion coefficients, D_SL/D_Li and D_BF4/D_Li, in the LiBF₄/SL solutions as a function of LiBF₄ concentration. At concentrations lower than 1.5 mol dm⁻³, both D_SL/D_Li and D_BF4/D_Li are higher than 1, suggesting that the Li⁺ ion diffuses slower than the SL and BF₄ anions. However, at concentrations higher than 3 mol dm⁻³, the D_SL/D_Li and D_BF4/D_Li become lower than 1, suggesting that the Li⁺ ions diffuse faster than the SL and BF₄⁻ anions. However, this cannot be explained by the mechanism shown in Fig. 9a. In the liquids with LiBF₄ concentrations higher than 3 mol dm⁻³, the molar ratio of [LiBF₄]/[SL] is higher than 1/3, and the coordination number of Li⁺ ions (4–5) cannot be satisfied solely by SL molecules. In such electrolytes, almost all SL molecules are involved in the solvation of Li⁺, and the amount of free SL is negligible. Additionally, BF₄⁻ anions are coordinated to Li⁺ ions to form CIPs and AGGs, which are adjacent to each other. The D_Li value being larger than D_SL and D_BF4 suggests that Li⁺ ions diffuse by exchanging the ligands (solvents and anions) and move forward, leaving behind the ligands, i.e., Li⁺ hopping conduction occurs in the liquid electrolytes (Fig. 9c). The hopping conduction results in a relatively high transference number of Li⁺ in the electrolytes. The Li⁺ transference number in the electrolyte with a molar ratio of [LiBF₄]/[SL] = 1/2 is as high as 0.8, while that in a conventional carbonate-based electrolyte (1 mol dm⁻³ LiPF₆ dissolved in EC/DMC (50 : 50 wt%)) is lower than 0.3.¹⁸ Other molten solvates, such as the SL solvate of LiTFSA, γ-butyrolactone solvate of Li bis(fluorosulfonyl)amide (LiFSA), succinonitrile solvate of LiFSA, and glutaronitrile solvate of LiTFSA, show Li⁺ transference numbers over 0.5.^{43,44,51–53}

Figure 9.

(a) Schematic illustration of Li⁺ ion conduction in dilute electrolytes; (b) the ratios of self-diffusion coefficients, D_SL/D_Li and D_BF4/D_Li, of LiBF₄-SL electrolytes at 30 °C; (c) schematic illustration of Li⁺-ion-hopping conduction in molten-solvate electrolytes. (b) and (c) are reproduced with permission from Ref. 50. Copyright 2018 American Chemical Society.

A high Li⁺ transference number is effective in mitigating the concentration polarization in Li batteries during the charging and discharging at high current densities. The discharge curves of a Li/LiCoO₂ cell with an electrolyte of [LiFSA]/[SL]/[DMS] = 1/1.5/1.5 (where DMS is dimethyl sulfone) are shown in Fig. 10.⁵⁴ Further, this electrolyte shows Li⁺ ion-hopping conduction, and the Li⁺ transference number is 0.45. The ionic conductivity of this electrolyte is 2.28 mS cm⁻¹ (at 30 °C), which is considerably lower than that of a carbonate-based electrolyte, 1 mol dm⁻³ LiPF₆ in EC/DMC (1 : 1 v/v) (12.73 mS cm⁻¹ at 30 °C). However, the discharge rate capability of the cell with the [LiFSA]/[SL]/[DMS] = 1/1.5/1.5 electrolyte is comparable to that of one with the carbonate-based electrolyte. The decrease in the discharge capacity at high current densities is due to the concentration polarization. During the discharge at high current densities, the concentrations of Li salt decrease and increase in the vicinity of the LiCoO₂ cathode and Li metal anode, respectively. The operating voltage of the cell is lowered owing to the concentration overpotential, and the cell voltage reaches the cut-off voltage before achieving the full discharge capacity. A high Li⁺ transference number is effective in suppressing the development of the concentration gradient across the electrolyte.^55,56 Resultantly, the cell with the [LiFSA]/[SL]/[DMS] = 1/1.5/1.5 electrolyte exhibits a relatively high discharge capacity of over 110 mAh g⁻¹ even at a high current density of 10 mA cm⁻².

Figure 10.

Discharge curves of the Li/LiCoO₂ cells in the (a) ternary electrolyte with a composition of [LiFSA]/[SL]/[DMS] = 1/1.5/1.5 and (b) 1 M LiPF₆ in EC/DMC (1 : 1 v/v) solution recorded at various current densities at 30 °C. The current densities in the legends of the figures (a) and (b) range from 0.2 to 30 mA cm⁻² in order of capacity at the cutoff potential; (c) discharge capacities of the cells as a function of the current density; before each discharge measurement, the cell is fully charged at a low current density of 0.2 mA cm⁻². Reproduced with permission from Ref. 54. Copyright 2022 American Chemical Society.

5. Summary and Outlook

The unique physicochemical properties of the molten solvates of Li salts and their battery applications are reviewed herein. In molten-solvate electrolytes, almost all the solvent molecules are coordinated to Li⁺ ions, and the activity of the free solvent is negligible. This is the main reason for the high thermal stability, enhanced oxidative stability, unique electrochemical properties, and Li-ion-hopping conduction in molten-solvate electrolytes. In addition to the molten solvates of Li salts, the solvates of Na, K, and Mg salts can be prepared, and they possess similar properties.^6,57–65 The molecular structures of solvents and anions significantly affect the physicochemical properties of the molten solvates. The design and synthesis of novel solvents and anions may provide new functions for solvate electrolytes. Such research and development will be essential for realizing high-performance next-generation batteries.

Acknowledgments

The author sincerely thanks Prof. Masayoshi Watanabe, Dr. Seiji Tsuzuki, Prof. Yasuhiro Umebayashi, Prof. Kazuhide Ueno, Dr. Toshihiko Mandai, Dr. Naoki Tachikawa, Prof. Morgan L. Thomas, Prof. Ryoichi Tatara, Dr. Hisashi Kokubo, Prof. Shiro Seki, Prof. Kenta Fujii, Prof. Hikari Watanabe, Prof. Yang Shao-Horn, many students, postdoctoral researchers, and collaborators. Mr. Yosuke Ugata is acknowledged for his kind assistance in preparing the figures for this paper. The financial support from the Japan Society for the Promotion of Science (JSPS) (JSPS KAKENHI Nos. JP15H03874, JP18H03926, JP19H05813, and JP22H00340), Japan Science and Technology Agency (JST) (ALCA-SPRING, Grant No. JPMJAL1301), the Ministry of Education, Culture, Sports, Science, and Technology (MEXT) (Elements Strategy Initiative of MEXT, Grant No. JPMXP0112101003), and the New Energy and Industrial Technology Development Organization (NEDO) are gratefully acknowledged.

CRediT Authorship Contribution Statement

Kaoru Dokko: Writing – original draft (Lead)

Conflict of Interest

The authors declare no conflict of interest.

Funding

Japan Society for the Promotion of Science: JP15H03874

Japan Society for the Promotion of Science: JP18H03926

Japan Society for the Promotion of Science: JP19H05813

Japan Society for the Promotion of Science: JP22H00340

Advanced Low Carbon Technology Research and Development Program: JPMJAL1301

Ministry of Education, Culture, Sports, Science, and Technology: JPMXP0112101003

Footnotes

K. Dokko: ECSJ Active Member

References

• 1)  K. Xu, Chem. Rev., 104, 4303 (2004).
• 2)  K. Xu, Chem. Rev., 114, 11503 (2014).
• 3)  M. Watanabe, M. L. Thomas, S. Zhang, K. Ueno, T. Yasuda, and K. Dokko, Chem. Rev., 117, 7190 (2017).
• 4)  T. Mandai, K. Yoshida, K. Ueno, K. Dokko, and M. Watanabe, Phys. Chem. Chem. Phys., 16, 8761 (2014).
• 5)  M. Watanabe, K. Dokko, K. Ueno, and M. L. Thomas, Bull. Chem. Soc. Jpn., 91, 1660 (2018).
• 6)  T. Mandai, K. Dokko, and M. Watanabe, Chem. Rec., 19, 708 (2019).
• 7)  T. Tamura, T. Hachida, K. Yoshida, N. Tachikawa, K. Dokko, and M. Watanabe, J. Power Sources, 195, 6095 (2010).
• 8)  T. Tamura, K. Yoshida, T. Hachida, M. Tsuchiya, M. Nakamura, Y. Kazue, N. Tachikawa, K. Dokko, and M. Watanabe, Chem. Lett., 39, 753 (2010).
• 9)  K. Yoshida, M. Nakamura, Y. Kazue, N. Tachikawa, S. Tsuzuki, S. Seki, K. Dokko, and M. Watanabe, J. Am. Chem. Soc., 133, 13121 (2011).
• 10)  K. Yoshida, M. Tsuchiya, N. Tachikawa, K. Dokko, and M. Watanabe, J. Phys. Chem. C, 115, 18384 (2011).
• 11)  K. Ueno, K. Yoshida, M. Tsuchiya, N. Tachikawa, K. Dokko, and M. Watanabe, J. Phys. Chem. B, 116, 11323 (2012).
• 12)  S. Tsuzuki, W. Shinoda, S. Seki, Y. Umebayashi, K. Yoshida, K. Dokko, and M. Watanabe, ChemPhysChem, 14, 1993 (2013).
• 13)  S. Tsuzuki, T. Mandai, S. Suzuki, W. Shinoda, T. Nakamura, T. Morishita, K. Ueno, S. Seki, Y. Umebayashi, K. Dokko, and M. Watanabe, Phys. Chem. Chem. Phys., 19, 18262 (2017).
• 14)  S.-A. Hyodo and K. Okabayashi, Electrochim. Acta, 34, 1551 (1989).
• 15)  M. Morita, Y. Asai, N. Yoshimoto, and M. Ishikawa, J. Chem. Soc., Faraday Trans., 94, 3451 (1998).
• 16)  Y. Kameda, Y. Umebayashi, M. Takeuchi, M. A. Wahab, S. Fukuda, S. Ishiguro, M. Sasaki, Y. Amo, and T. Usuki, J. Phys. Chem. B, 111, 6104 (2007).
• 17)  K. Ueno, R. Tatara, S. Tsuzuki, S. Saito, H. Doi, K. Yoshida, T. Mandai, M. Matsugami, Y. Umebayashi, K. Dokko, and M. Watanabe, Phys. Chem. Chem. Phys., 17, 8248 (2015).
• 18)  Y. Ugata, K. Shigenobu, R. Tatara, K. Ueno, M. Watanabe, and K. Dokko, Phys. Chem. Chem. Phys., 23, 21419 (2021).
• 19)  S. K. Jeong, M. Inaba, Y. Iriyama, T. Abe, and Z. Ogumi, J. Power Sources, 175, 540 (2008).
• 20)  Y. Yamada and A. Yamada, J. Electrochem. Soc., 162, A2406 (2015).
• 21)  Y. Yamada, J. Wang, S. Ko, E. Watanabe, and A. Yamada, Nat. Energy, 4, 269 (2019).
• 22)  O. Borodin, J. Self, K. A. Persson, C. Wang, and K. Xu, Joule, 4, 69 (2020).
• 23)  M. Li, C. Wang, Z. Chen, K. Xu, and J. Lu, Chem. Rev., 120, 6783 (2020).
• 24)  L. Suo, Y. S. Hu, H. Li, M. Armand, and L. Chen, Nat. Commun., 4, 1481 (2013).
• 25)  J. Qian, W. A. Henderson, W. Xu, P. Bhattacharya, M. Engelhard, O. Borodin, and J. G. Zhang, Nat. Commun., 6, 6362 (2015).
• 26)  J. Alvarado, M. A. Schroeder, M. Zhang, O. Borodin, E. Gobrogge, M. Olguin, M. S. Ding, M. Gobet, S. Greenbaum, Y. S. Meng, and K. Xu, Mater. Today, 21, 341 (2018).
• 27)  X. Ren, S. Chen, H. Lee, D. Mei, M. H. Engelhard, S. D. Burton, W. Zhao, J. Zheng, Q. Li, M. S. Ding, M. Schroeder, J. Alvarado, K. Xu, Y. S. Meng, J. Liu, J. G. Zhang, and W. Xu, Chem, 4, 1877 (2018).
• 28)  S. Ko, Y. Yamada, and A. Yamada, Joule, 5, 998 (2021).
• 29)  R. Tatara, D. G. Kwabi, T. P. Batcho, M. Tulodziecki, K. Watanabe, H. M. Kwon, M. L. Thomas, K. Ueno, C. V. Thompson, K. Dokko, Y. Shao-Horn, and M. Watanabe, J. Phys. Chem. C, 121, 9162 (2017).
• 30)  Y. Yamada, Y. Takazawa, K. Miyazaki, and T. Abe, J. Phys. Chem. C, 114, 11680 (2010).
• 31)  C. Zhang, K. Ueno, A. Yamazaki, K. Yoshida, H. Moon, T. Mandai, Y. Umebayashi, K. Dokko, and M. Watanabe, J. Phys. Chem. B, 118, 5144 (2014).
• 32)  H. Moon, R. Tatara, T. Mandai, K. Ueno, K. Yoshida, N. Tachikawa, T. Yasuda, K. Dokko, and M. Watanabe, J. Phys. Chem. C, 118, 20246 (2014).
• 33)  L. J. Krause, W. Lamanna, J. Summerfield, M. Engle, G. Korba, R. Loch, and R. Atanasoski, J. Power Sources, 68, 320 (1997).
• 34)  H. Yang, K. Kwon, T. M. Devine, and J. W. Evans, J. Electrochem. Soc., 147, 4399 (2000).
• 35)  M. Morita, T. Shibata, N. Yoshimoto, and M. Ishikawa, J. Power Sources, 119–121, 784 (2003).
• 36)  C. Zhang, A. Yamazaki, J. Murai, J. W. Park, T. Mandai, K. Ueno, K. Dokko, and M. Watanabe, J. Phys. Chem. C, 118, 17362 (2014).
• 37)  Y. Yamada, K. Furukawa, K. Sodeyama, K. Kikuchi, M. Yaegashi, Y. Tateyama, and A. Yamada, J. Am. Chem. Soc., 136, 5039 (2014).
• 38)  K. Sodeyama, Y. Yamada, K. Aikawa, A. Yamada, and Y. Tateyama, J. Phys. Chem. C, 118, 14091 (2014).
• 39)  Q. Pang, X. Liang, C. Y. Kwok, and L. F. Nazar, Nat. Energy, 1, 16132 (2016).
• 40)  S. Zhang, K. Ueno, K. Dokko, and M. Watanabe, Adv. Energy Mater., 5, 1500117 (2015).
• 41)  J. W. Park, K. Yamauchi, E. Takashima, N. Tachikawa, K. Ueno, K. Dokko, and M. Watanabe, J. Phys. Chem. C, 117, 4431 (2013).
• 42)  K. Dokko, N. Tachikawa, K. Yamauchi, M. Tsuchiya, A. Yamazaki, E. Takashima, J.-W. Park, K. Ueno, S. Seki, N. Serizawa, and M. Watanabe, J. Electrochem. Soc., 160, A1304 (2013).
• 43)  A. Nakanishi, K. Ueno, D. Watanabe, Y. Ugata, Y. Matsumae, J. Liu, M. L. Thomas, K. Dokko, and M. Watanabe, J. Phys. Chem. C, 123, 14229 (2019).
• 44)  Y. Ugata, R. Tatara, K. Ueno, K. Dokko, and M. Watanabe, J. Chem. Phys., 152, 104502 (2020).
• 45)  A. C. Luntz and B. D. McCloskey, Chem. Rev., 114, 11721 (2014).
• 46)  D. Aurbach, B. D. McCloskey, L. F. Nazar, and P. G. Bruce, Nat. Energy, 1, 16128 (2016).
• 47)  H. M. Kwon, M. L. Thomas, R. Tatara, Y. Oda, Y. Kobayashi, A. Nakanishi, K. Ueno, K. Dokko, and M. Watanabe, ACS Appl. Mater. Interfaces, 9, 6014 (2017).
• 48)  K. Izutsu, Electrochemistry in Nonaqueous Solutions, Second, Revised and Enlarged Edition, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany (2009).
• 49)  K. Hayamizu, Y. Aihara, S. Arai, and C. G. Martinez, J. Phys. Chem. B, 103, 519 (1999).
• 50)  K. Dokko, D. Watanabe, Y. Ugata, M. L. Thomas, S. Tsuzuki, W. Shinoda, K. Hashimoto, K. Ueno, Y. Umebayashi, and M. Watanabe, J. Phys. Chem. B, 122, 10736 (2018).
• 51)  Y. Ugata, S. Sasagawa, R. Tatara, K. Ueno, M. Watanabe, and K. Dokko, J. Phys. Chem. B, 125, 6600 (2021).
• 52)  S. Kondou, K. Dokko, M. Watanabe, and K. Ueno, Electrochemistry, 89, 389 (2021).
• 53)  Y. Ugata, M. L. Thomas, T. Mandai, K. Ueno, K. Dokko, and M. Watanabe, Phys. Chem. Chem. Phys., 21, 9759 (2019).
• 54)  Y. Ugata, Y. Chen, S. Sasagawa, K. Ueno, M. Watanabe, H. Mita, J. Shimura, M. Nagamine, and K. Dokko, J. Phys. Chem. C, 126, 10024 (2022).
• 55)  M. Doyle, T. F. Fuller, and J. Newman, Electrochim. Acta, 39, 2073 (1994).
• 56)  K. M. Diederichsen, E. J. McShane, and B. D. McCloskey, ACS Energy Lett., 2, 2563 (2017).
• 57)  T. Mandai, R. Nozawa, S. Tsuzuki, K. Yoshida, K. Ueno, K. Dokko, and M. Watanabe, J. Phys. Chem. B, 117, 15072 (2013).
• 58)  T. Mandai, S. Tsuzuki, K. Ueno, K. Dokko, and M. Watanabe, Phys. Chem. Chem. Phys., 17, 2838 (2015).
• 59)  S. Terada, T. Mandai, R. Nozawa, K. Yoshida, K. Ueno, S. Tsuzuki, K. Dokko, and M. Watanabe, Phys. Chem. Chem. Phys., 16, 11737 (2014).
• 60)  S. Terada, H. Susa, S. Tsuzuki, T. Mandai, K. Ueno, Y. Umebayashi, K. Dokko, and M. Watanabe, J. Phys. Chem. C, 120, 23339 (2016).
• 61)  S. Terada, T. Mandai, S. Suzuki, S. Tsuzuki, K. Watanabe, Y. Kamei, K. Ueno, K. Dokko, and M. Watanabe, J. Phys. Chem. C, 120, 1353 (2016).
• 62)  S. Terada, H. Susa, S. Tsuzuki, T. Mandai, K. Ueno, K. Dokko, and M. Watanabe, J. Phys. Chem. C, 122, 16589 (2018).
• 63)  K. Hashimoto, S. Suzuki, M. L. Thomas, T. Mandai, S. Tsuzuki, K. Dokko, and M. Watanabe, Phys. Chem. Chem. Phys., 20, 7998 (2018).
• 64)  Y. Okamoto, S. Tsuzuki, R. Tatara, K. Ueno, K. Dokko, and M. Watanabe, J. Phys. Chem. C, 124, 4459 (2020).
• 65)  R. Tatara, S. Nishimura, Y. Okamoto, K. Ueno, M. Watanabe, and K. Dokko, J. Phys. Chem. C, 124, 15800 (2020).

Biographies

Kaoru Dokko (Professor, Faculty of Engineering, Yokohama National University)

Kaoru Dokko was born in 1973. He graduated from Graduate School of Engineering, Tohoku University in September 2001, and earned Doctor of Engineering, He was a postdoctoral researcher at Case Western Reserve University and Tokyo Metropolitan University. He joined Yokohama National University in 2008 as Associate Professor and was promoted to Professor in 2016. He received Young Researcher Award of The Electrochemical Society of Japan (Sano Award) from The Electrochemical Society of Japan in 2007. His current research interests are physicochemical properties of concentrated electrolytes, nanostructured materials for energy conversion and storage, and electrochemical reaction process in electrochemical devices.