Concentrated Electrolytes and Their Unique Interfacial Reactions in Rechargeable Batteries

Abstract

An electrolyte is an important component of rechargeable batteries. The Debye–Hückel theory is applicable to liquid electrolytes with salt concentrations up to approximately 0.01 mol/L. Thus, electrolyte solutions with approximately 1 mol/L salt concentration are considered “highly concentrated” from the perspective of the classical solution theory. However, electrochemical devices use them at these concentrations, because their ionic conductivity reaches the maximum level across the entire concentration range. Although a salt concentration of 1 mol/L is considered very high, recent studies have indicated that the use of further “super-concentrated” electrolytes, typically those with concentrations exceeding 3 mol/L and also known as “molten solvate” or “solvent-in-salt” can exhibit unique reaction behaviors in various electrochemical systems, including Li-ion batteries. This review examines the characteristics of such “super-concentrated” electrolytes.

1. Background of Classical Solution Chemistry

According to the classical solution theory, an electrolyte solution with a salt concentration of approximately 0.01 mol/L or less can be considered a “dilute solution.” In 1923, Debye and Hückel published a theory on ion–ion interactions in dilute electrolyte solutions¹ and defined the ionic strength of a solution, I, as a parameter that reflects inter-ionic interactions, as follows:

  

\begin{equation} I = \frac{1}{2}\sum\nolimits_{i}c_{i}z_{i}^{2} \end{equation} (1)

where c_i is the molar concentration of ion species and z_i is the net charge of the ion. Thus, they introduced a variable that adequately represents the Coulomb force between ions, which increases with increasing ionic valence and salt concentration in the solution. The Debye–Hückel theory assumes that (1) the electrolyte is completely dissociated, (2) only Coulombic interactions occur between ions, (3) the contribution of thermal disturbance is larger than the electrostatic potential energy (i.e., ions are not bound to each other by electrostatic forces in the solvent and are free to move under thermal stimulation), and (4) the dielectric constant of the solvent around an ion is equal to that in the bulk. The Debye–Hückel theory postulates that the counterion of a given ion exists on a spherical shell of radius r_D, which is termed Debye length and is given by

  

\begin{equation} r_{D} = \sqrt{\frac{\varepsilon_{0}\varepsilon_{r}RT}{2F^{2}I}} \end{equation} (2)

where ε₀ is the dielectric constant of vacuum, ε_r is the relative permittivity of the solvent, R is the gas constant, T is the absolute temperature, and F is Faraday constant. In an aqueous monovalent electrolyte solution, such as KCl aq., at 25 °C, the Debye length, r_D, is 96 Å at 1 mM concentration, 30.4 Å at 10 mM, and 9.6 Å at 100 mM (Fig. 1a). Given that the diameter of a hydrogen atom is approximately 1 Å, it is evident that a 100 mM (0.1 mol/L) solution is “highly concentrated.” Thus, the higher the ion concentration in a solution, the greater the number of counterions of opposite charge in the surrounding ionic shell, which restricts its mobility. The degree of “freedom” of an ion is expressed by an activity coefficient, γ (0–1), and the activity of an ion, a, in a solution of concentration c is defined as

  

\begin{equation} a = \gamma c \end{equation} (3)

The average activity coefficient γ_± is given by Eq. 4, known as the Debye–Hückel limiting law.²

  

\begin{equation} \log \gamma_{\pm} = Az_{+}z_{-}\sqrt{I} \end{equation} (4)

  

\begin{equation} A = \frac{F^{3}}{4\pi N_{A}\ln 10}\left(\frac{\rho}{2\varepsilon_{0}{}^{3}\varepsilon_{r}{}^{3}R^{3}T^{3}}\right)^{1/2} \end{equation} (5)

where A is a constant that depends on the dielectric constant, density (ρ), and temperature of the solvent. Dielectric constant- and concentration-dependent activity coefficients in A⁺B⁻-type monovalent electrolytes are shown in Figs. 1b–1c. Although the measured values agree with this law at very low salt concentrations (less than 0.01 mol/L), the ion–ion distance decreases and ionic size becomes a significant factor at high electrolyte concentrations. Therefore, by considering the average diameter of the ions, d, and using a constant, B, the following equation was derived for γ_±:

  

\begin{equation} \log \gamma_{\pm} = \frac{Az_{+}z_{-}\sqrt{I}}{1 + Bd\sqrt{I}} \end{equation} (6)

This equation accurately reproduces the experimental values at salt concentrations below 0.1 mol/L. However, at higher salt concentrations (>0.1 mol/L), the interaction between the ions and solvent molecules cannot be ignored. As this situation is difficult to treat theoretically, an empirical correction term was proposed, as shown below:

  

\begin{equation} \log \gamma_{\pm} = \frac{Az_{+}z_{-}\sqrt{I}}{1 + Bd\sqrt{I}} + bI \end{equation} (7)

Such formulas with various corrections are referred to as extended Debye–Hückel equations. However, in solutions with salt concentrations exceeding 1 mol/L, interionic distances become smaller, and electrostatic interactions become stronger. Consequently, the assumption that thermal disturbance predominates (|z_ieΦ| ≪ kT, where e is elementary charge, φ is Coulomb potential, and k is Boltzmann constant) does not hold. Ion pairs are formed in highly concentrated solutions, invalidating the assumption of complete ionic dissociation in a solution. Furthermore, it is necessary to consider the possibility of forming triple cations and anions, such as [ABA]⁺ and [BAB]⁻, in A⁺B⁻-type solutes. Consequently, we must acknowledge that handling these phenomena theoretically is currently exceedingly difficult. In the classical solution theory, even a solution with salt concentration of 1 mol/L is considered a “concentrated solution,” in which ionic interactions are significantly strong. In other words, the activity coefficient of the ion is considered to be much lower than 1.

Figure 1.

(a) Debye length, and (b) dielectric constant- and (c) concentration-dependent activity coefficient in A⁺B⁻-type monovalent electrolytes calculated from Debye–Hückel limiting law, assuming the solvent density is 1 g/cm³ and molality (mol/kg) is equal to molarity (mol/L). Note that Debye–Hückel limiting law is only applicable in dilute electrolytes (<0.01 mol/L).

The current lithium-based batteries typically use lithium salts dissolved in an organic solvent at a concentration of approximately 1 mol/L as electrolytes. This is because the ionic conductivity of the electrolyte, a crucial factor for electrolytes in electrochemical devices, reaches the maximum at approximately 1 mol/L.

The ionic conductivity, σ, of an electrolyte solution can be expressed as follows:^3,4

  

\begin{equation} \sigma = \sum\nolimits_{i}N_{i}Q_{i}u_{i} \end{equation} (8)

where N represents the number of ionic carriers per unit volume, Q is the charge per ionic carrier, and u is the mobility of the ionic carrier. As the concentration of the electrolyte increases, the number of carriers increases, but ionic mobility decreases due to the increased viscosity of the solution, resulting in a trade-off relationship between the electrolyte concentration and ionic mobility. Therefore, the ionic conductivity of an electrolyte solution reaches the maximum at a certain concentration (Fig. 2). Thus, a concentration of 1 mol/L is considered as the standard for practical electrolytes used in electrochemical devices. Electrolytes with higher ionic concentrations are generally considered to be less effective due to their higher viscosity, which leads to lower ionic conductivity. This situation highlights the lack of merit in using an excess of the expensive electrolyte salt.

Figure 2.

Concentration-dependent ionic conductivity and viscosity of LiPF₆ in a mixed ethylene carbonate (EC)/ethyl methyl carbonate (EMC) solvent (3 : 7 in vol/vol) at 25 °C.

Despite the aforementioned considerations, in recent years, there has been significant interest in so-called “super-concentrated” electrolytes with lithium salt concentrations in excess of approximately 3 mol/L (which usually have less than four solvent molecules per Li⁺) from the perspectives of both fundamental solution chemistry and device applications. As the typical solvation number of Li⁺ in a non-aqueous solution is 4–5, all the solvent molecules directly interact with Li⁺ and no “free solvent” is available in such super-concentrated electrolytes, as illustrated in Fig. 3.⁵ Many comprehensive reviews have discussed the applications of such concentrated electrolytes.^5–9 Therefore, this article mainly focuses on the historical background and fundamental aspects of concentrated electrolytes, and only highlights representative applications.

Figure 3.

Schematic illustration of the structures of (top) typical and (bottom) concentrated electrolyte solutions. Reprinted and partially modified from Ref. 5. © Y. Ugata et al. 2021 CC BY 3.0.

2. Parameters that Determine Solvent Properties

For describing the solution chemistry of a liquid electrolyte, it is essential to understand the parameters that characterize the properties of the solvents and salts used to prepare electrolytes. Some representative parameters are summarized here.

2.1 Dielectric constant

Dielectric constant is a measure of the ability of a dielectric material to polarize in response to an external electric field. In the context of solution chemistry, the term “dielectric constant” typically refers to relative permittivity ε_r of the solvent used, which is the ratio of the actual permittivity (ε) of the material to the permittivity of vacuum (ε₀).

  

\begin{equation} \varepsilon_{r} = \varepsilon/\varepsilon_{0} \end{equation} (9)

This parameter is commonly understood as a value proportional to the capacitance of a parallel plate capacitor. In electrolyte solutions, dissociated ions cause the solvent surrounding them to polarize. The greater the extent of this polarization is, the more significant the screening of the Coulomb forces between the ions is, which makes it harder for counterions to experience the electric field. Therefore, in solvents with a high relative permittivity, Coulombic interactions are weaker. For example, the relative permittivity of water is 78, and according to the Debye–Hückel limiting law (Eqs. 4–5), the mean activity coefficient of a monovalent electrolyte (e.g., KCl(aq.)) in a 1 mmol/L aqueous solution is approximately 0.96. In contrast, in an ether-based solvent with a relative permittivity of approximately 7, the mean activity coefficient for a 1 mmol/L electrolyte solution decreases to approximately 0.2 (Figs. 1b–1c). Thus, even in a relatively dilute 1 mmol/L electrolyte solution, a solvent with a low dielectric constant causes limited electrostatic shielding, resulting in stronger Coulombic interactions between ions of opposite charges. Thus, the dielectric constant of a solvent plays a critical role in governing the strength of ion–ion interactions and the degree of salt dissociation in a solution.

2.2 Donor and acceptor numbers

Lewis acids are molecules or ions that can accept an electron pair from other chemical species because of their vacant electron orbitals, resulting in chemical bonding or interactions. Conversely, Lewis bases are molecules or ions with an electron pair that can coordinate with a Lewis acid. The donor and acceptor numbers, proposed by Gutmann et al., represent the ability of a chemical species to donate or accept electron pairs, respectively.^10–12 As lithium ion acts as a Lewis acid and the solvents that solvate it act as Lewis bases, the donor number (DN) is a critical parameter that affects the properties of solvents in lithium-based batteries.

The DN indicates the ability of solvents or anions to coordinate with positively charged species. It is defined as the molar enthalpy of the reaction between SbCl₅ used as the reference acceptor and dichloroethane as the solvent at 1 mmol/L SbCl₅ concentration, which can be represented as

  

\begin{equation} \text{D} + \text{SbCl$_{5}$} \rightleftarrows \text{D} \cdot \text{SbCl$_{5}$}\ (-\Delta H_{\text{D${\cdot}$SbCl${_{5}}$}} \equiv \text{DN}) \end{equation} (10)

SbCl₅ was selected as the reference acceptor to evaluate the DN of the solvent, because it is a very strong acceptor and forms a 1 : 1 adduct with various donor molecules (i.e., solvents) and allows the evaluation of even very weak donors. However, as solvents that act as strong donors may react explosively, more convenient and indirect measurement methods, such as the evaluation of the NMR chemical shift of ²³Na in NaClO₄ dissolved in the solvent of interest or the shift in the maximum absorption wavelength of copper complexes during solvatochromism, have been employed. However, these estimations are merely based on the interaction of the relevant solvent with SbCl₅, Na⁺, or copper ions in complexes, and potential inconsistencies can arise between measurements, making it difficult to correlate the results with the fundamental properties of lithium electrolytes involving Li⁺ interactions. For instance, depending on the measurement method used, a wide range of DNs from 18 to 33 are obtained for water as the donor. Nevertheless, empirically, the DN serves as a good indicator of interactions between Li⁺ and solvents or anions in the electrolyte.

Further, the acceptor number (AN) of a solvent is empirically calculated from the ³¹P-NMR chemical shifts of triethylphosphine oxide (Et₃PO) in each pure solvent. Et₃PO is used as the reference because it acts as a strong base with a DN of approximately 40, allowing highly sensitive measurements. In this evaluation, the AN of hexane is set to zero, and the relative chemical shift of Et₃PO in a given solvent is used after corrections are made for differences in bulk susceptibility; the AN of the Et₃PO·SbCl₅ adduct in dichloroethane is set to 100.

Whereas the dielectric constant of a solvent is evaluated by treating it as a “continuum,” the DN or AN are evaluated by considering the explicit molecular properties of the solvent. Therefore, the DN and AN values correlate well with the coordination strength of solvate ions in the primary solvation shell. In contrast, the dielectric constant tends to be a better indicator of the electrostatic shielding effect of the medium in the outer shell, particularly in relatively dilute solutions, in which ion–ion interactions operate beyond the solvation shell. Table 1 summarizes certain parameters of representative solvents that could potentially be used in lithium-based secondary batteries.^3,10 The results of a preliminary investigation on the reactivity of each solvent with lithium metal and the solubility of lithium salts (e.g., lithium bis(trifluoromethanesulfonylamide); LiTFSA) in each solvent are also provided. Low-polarity solvents with extremely low DN values and dielectric constants tend not to dissolve lithium salts, whereas many high-polarity solvents react with Li metal, depending on their molecular structure.

Table 1. Donor numbers (DNs), acceptor numbers (ANs), and dielectric constants (ε_r) of solvents used to prepare electrolyte solutions.

Abbreviation	Name	DN	AN	εr	Stability against Li metal	LiTFSA Solubility
DCE	1,2-Dichloroethane	0	17	10
	Hexane	0	0	1.9	✓
	Toluene	2.4		0.1	✓	×
G4	Tetraglyme	17		7.8	✓	✓
G3	Triglyme	14		7.6	✓	✓
G2	Diglyme			7.3	✓	✓
G1	Monoglyme	24	10	7.2	✓	✓
THF	Tetrahydrofuran	20	8	7.6	✓	✓
PC	Propylene carbonate	15	18	64	✓	✓
EC	Ethylene carbonate	16		90	✓	✓
DMC	Dimethyl carbonate			3.1	✓	✓
DEC	Diethyl carbonate	16		2.8	×	✓
GBL	γ-Butyrolactone	18	17	39	✓	✓
SL	Sulfolane	14.8	19	43	△	✓
AN	Acetonitrile	14.1		36	×	✓
DMSO	Dimethyl sulfoxide	29.8	19	46	△	✓
Ac2O	Acetic anhydride	11		21	✓	✓
TMP	Trimethyl phosphate	23		21	✓	✓
DMI	1,3-Dimethyl-2-imidazolidinone	28		38	✓	✓
DOL	1,3-Dioxolane			7.1
DO	1,4-Dioxane	15	11	2.2
	Ethyl acetate	17.1		6
NM	Nitromethane	2.7	21	37	×
NB	Nitrobenzen	4.4		35	×
NMP	N-Methylpyrrolidone	27	13	32	×	✓
DMF	N,N-Dimethylformamide	27	16	37	×
DMAc	N,N-Dimethylacetoamide	28	14	38	×
TMU	Tetramethylurea	30		24	×
Py	Pyridine	33		13	×
HMPA	Hexamethylphosphoric triamide	39		30	×	✓
	Water	18–33	55	78	×	✓

3. Typical Electrolytes Used in Lithium-ion Batteries (LIBs)

The currently used LIBs employ organic solvents with dissolved lithium salts as electrolytes. Other types of rechargeable batteries, such as nickel-metal hydride and lead-acid batteries, use aqueous electrolytes because of their chemical stability and high ionic conductivity. However, aqueous electrolytes cannot be used in certain LIBs, such as those using LiCoO₂ with a cut-off voltage of ≥4.2 V. This is because the potential window of aqueous electrolytes is limited to 1.23 V, falling between the potentials of hydrogen evolution reaction (0 V vs. standard hydrogen electrode (SHE)) and oxygen evolution reaction (1.23 V vs. SHE).

Unlike aqueous electrolytes, organic electrolytes have wide potential windows.¹³ The desired properties of solvents for LIBs are a high ionic conductivity (that is, the ability to dissolve and dissociate lithium salts at high concentrations) combined with a low viscosity. However, there is a trade-off relationship between these two parameters. Solvents capable of dissolving and dissociating Li salts at high concentrations tend to have high dielectric constants and present high viscosities due to their high polarity. Conversely, low-viscosity solvents typically have low dielectric constants, owing to which they induce a weaker electrostatic shielding effect and thus have a lower capability to dissociate lithium salts. Consequently, current organic electrolytes use mixed solvents composed of a high-dielectric-constant solvent that effectively dissociates the Li salt and a low-viscosity solvent with low-dielectric-constant. Specifically, a mixture of a cyclic carbonate, such as ethylene carbonate (EC, ε_r = 90), and a linear carbonates (ε_r ∼ 3), such as dimethyl carbonate (DMC), diethyl carbonate (DEC), ethyl methyl carbonate (EMC), is the most commonly used solvent. Furthermore, the solvents used in batteries must meet stringent requirements in terms of compatibility with both the positive and negative electrodes. Positive electrodes, such as LiCoO₂¹⁴ with an average discharge voltage of 3.7 V (vs. Li/Li⁺) and a cutoff voltage of 4.3 V and the more recent high-voltage LiNi_1/3Mn_1/3Co_1/3O₂,¹⁵ require solvents with high oxidative stability.¹⁶ Using solvents with low oxidative stability, such as ethers, can result in oxidative decomposition of the electrolyte during charging. On the other hand, graphite used as the negative electrode operates at an electrode potential nearly equal to that of Li/Li⁺, subjecting it to extremely strong reductive conditions. Thus, solvents with high reductive stabilities are required for batteries using the graphite anode; solvents with low reductive stability, such as acetonitrile and sulfone derivatives, are not suitable for LIBs with a graphite anode. Furthermore, during the electrochemical intercalation of the lithium cation into the graphite anode, the solvated lithium cation [Li(solv)_n]⁺ must be desolvated, so that bare Li⁺ can be intercalated between the graphite layers (Fig. 4).

  

\begin{equation} \text{C$_{6}$} + \text{[Li(solv)$_{n}$]$^{+}$} + \text{e$^{-}$} = \text{C$_{6}$Li} + \text{$n$ solv} \end{equation} (11)

If desolvation does not occur and the solvated cations are inserted as they are (that is, solvent co-intercalation occurs), the larger solvated cations can cause the graphene layers to exfoliate, leading to irreversible side reactions.

  

\begin{equation} \text{C$_{6}$} + \text{[Li(solv)$_{n}$]$^{+}$} + \text{e$^{-}$} = \text{C$_{6}$[Li(solv)$_{n}$]} \end{equation} (12)

This is the reason why the graphite anode does not operate stably, even when solvents with high reductive stabilities, such as ethers, are used. However, when ethylene carbonate (EC) or vinylene carbonate (VC) is used as the solvent, the solvent molecule undergoes reductive decomposition at the graphite anode, forming a nanoscale film known as the solid–electrolyte interphase (SEI) at the graphite anode–bulk electrolyte interface. The so-formed SEI acts as a solid polymer electrolyte (it exhibits lithium-ion conductivity and no electronic conductivity) and promotes the desolvation of solvated lithium cations as they pass through it, allowing only the naked lithium cations to be intercalated into graphite. This prevents further reductive decomposition of the graphite anode, resulting in its stabilization. However, when propylene carbonate (PC) is used as the solvent, its co-intercalation proceeds preferentially over SEI formation via reductive decomposition at the graphite surface, inevitably resulting in graphite exfoliation. Due to these circumstances, the use of solvents or additives such as EC and VC, which generate a “good SEI,” is essential for the operation of the graphite anode. As carbonate-based solvents with relatively high oxidative stabilities are not problematic to be used with cathode materials such as LiCoO₂, current LIBs typically use an EC and linear carbonate (such as DMC, DEC, EMC) mixed solvent.

Figure 4.

Schematic illustration of the desolvation step at the graphite anode/electrolyte interface.

The choice of the lithium salt is also a critical factor for ensuring the stable operation of LIBs. The Li salt should be highly soluble in organic solvents, exhibit high dissociation, and be chemically stable. Because lithium cations are strong Lewis acids, relatively large anions that allow electron delocalization and thus have lower Lewis basicity are generally preferred as counterions. Common anions of Li salts include PF₆⁻, ClO₄⁻, BF₄⁻, [TFSA]⁻, and bis(fluorosulfonyl)amide ([FSA]⁻). However, ClO₄⁻ as a peroxide raises safety concerns, and [TFSA]⁻ and [FSA]⁻ are known to be incapable of preventing the corrosion of the aluminum current collectors on the cathode side. Therefore, LiPF₆ with a high oxidative stability and dissociation capability is commonly used as the Li salt. As previously mentioned, a lithium salt concentration of approximately 1 mol/L provides optimal ionic conductivity; therefore, 1 mol/L LiPF₆ in an EC-EMC solution is a standard electrolyte used in LIBs.

4. Ionic Liquids (ILs) as Electrolytes

As mentioned earlier, a solution with an electrolyte concentration of approximately 1 mol/L is generally considered ideal in terms of ionic conductivity. However, the unique properties of highly concentrated electrolytes, in which the salt is dissolved at much higher concentrations (>3 mol/L), are also well documented. Here, the historical relationship between highly concentrated electrolytes and ILs is discussed. ILs are defined as salts with a melting point near room temperature; they are solely composed of ions (zero-solvent liquid).¹⁷ The term “near room temperature” can be broadly interpreted as below 100 °C. In specific, substances that occur in liquid phase at room temperature (25 °C) are referred to as “room-temperature ILs.” [EtNH₃][NO₃] was reported as the first IL by Walden in 1914.¹⁸ Subsequently, from 1950s to 1980s, chloroaluminate salts (e.g., 1-ethyl-3-methylimidazolium tetrachloroaluminate; [C₂mim][AlCl₄]), which react vigorously with moisture and cannot be handled in ambient atmosphere, were discovered.^19,20 However, active research on ILs began after the discovery of [C₂mim][BF₄], which is a liquid at room temperature and remains stable in air, in 1992 by Wilkes et al.²¹ Thus, research on ILs began to gain momentum after nearly 80 years following the initial discovery.

Being solely composed of ions, ILs have an extremely low vapor pressure, low volatility, high thermal stability, and high ionic conductivity, and are nonflammable. Owing to these favorable properties, ILs are gaining attention as a third type of solvent, apart from water and organic solvents, and their application in various electrochemical devices, such as LIBs,^22,23 dye-sensitized solar cells,^24,25 fuel cells,^26–28 electric double-layer capacitors,^29,30 and actuators,³¹ is being widely explored. They are also used as solvents in polymer gels used as responsive^32,33 and self-healing materials,³⁴ in processes such as extraction and gas separation,^35,36 in organic synthesis and polymerization,^37,38 and in vacuum systems (e.g., electron microscopes),^39,40 among others.

Although ILs are classified in various ways,⁴¹ the most commonly known candidates are aprotic ILs (APILs) that lack active protons and are synthesized by the quaternization of a cation precursor and subsequent anion exchange. A typical APIL consists of an organic onium cation, a weak Lewis acid, and a relatively large anion, a weak Lewis base. This is because if both the Lewis acidity of the cation and Lewis basicity of the anion are high, the interaction between the cation and anion becomes strong, resulting in the formation of a solid. Therefore, cations and anions with extremely low Lewis acidity and basicity, respectively, are used to achieve good ion dissociation and maintain a liquid state even at room temperature, in the absence of a solvent.⁴²

Another class of ILs are “solvate ILs,” in which either the cation or anion, or both are solvated ions (Fig. 5).⁴³ A typical example is the glyme-lithium salt complex formed by mixing a lithium salt with a chelating solvent, such as glyme, in equimolar quantities. The formation of this complex in a liquid state was reported by Smyrl et al. in 2004.⁴⁴ Subsequently, systematic studies were conducted to explore its properties, discovering its IL characteristics.

Figure 5.

Preparation of a solvate ionic liquid, [Li(G3)₁][TFSA]. Reprinted and partially modified with permission from Ref. 43. Copyright 2013 Chemical Society of Japan.

As the glyme-lithium salt complex is an equimolar mixture of a salt and solvent, its ion concentration is approximately 3 mol/L or higher, and it can be categorized as a highly concentrated electrolyte. Note that anion BF₄⁻, which is widely used in ILs, is formed by the association of BF₃, a Lewis acid, with F⁻ as the Lewis base. Similarly, Li⁺, a Lewis acid, coordinates with glyme, a Lewis base, and the system behaves like a pseudo-cation.⁴⁵ Although a solvent (glyme) is used, all the added solvent molecules participate in the formation of solvated cations, leaving almost no glyme molecules that act as a “free solvent” in the system. Consequently, the glyme-lithium salt complex exhibits the properties of an IL. Using pulsed-field gradient (PFG) NMR, the self-diffusion coefficients of the components of this IL were measured, and the self-diffusion coefficients of the ¹H from glyme and the ⁷Li were found to be equal in the 1 : 1 mixture.^46,47 Furthermore, Raman spectroscopy confirmed the formation of stable solvated cations, as indicated by the characteristic spectral band (breathing mode at ∼870 cm⁻¹) that appears when oligoethers coordinate with metal cations.⁴⁵ Angell et al. categorized ILs consisting of solvated ions, such as the glyme-lithium salt complex, into a new class and designated them as “solvate ILs.”⁴¹

The following classification criteria have been proposed for solvate ILs:⁴⁵ (1) they form stoichiometric solvated ions composed of ions and ligands, (2) consist solely of solvated ions and their counterions, (3) do not exhibit the physicochemical properties of the pure solvent or precursor salt, (4) have a melting point below 100 °C, and (5) have a negligible vapor pressure.

For preparing lithium-based solvate ILs, various solvents and anions of lithium salts can be used; however, not all combinations present the properties of solvate ILs. For instance, when preparing 1 : 1 mixtures of triglyme (G3) or tetraglyme (G4) and lithium salts having anions of varying Lewis basicity (donor ability), stable solvate cations are formed upon using anions with relatively low Lewis basicity, such as [TFSA]⁻, bis(pentafluoroethylsulfonyl)amide ([BETA]⁻), or ClO₄⁻, which interact weakly with lithium ions and behave dissociatively.⁴⁸ Figure 6a plots D_Glyme/D_Li as a function of Λ_imp/Λ_NMR (sometimes referred to as “ionicity”) for each equimolar mixture of lithium salt and solvent. Λ_imp/Λ_NMR is the ratio of molar ionic conductivities; Λ_imp = σ/c (where σ is the ionic conductivity and c is the molar concentration of the IL) is determined using the electrochemical impedance method, and Λ_NMR is calculated from self-diffusion coefficients measured by PFG-NMR using the following Nernst-Einstein equation:

  

\begin{equation} \varLambda_{\text{NMR}} = F^{2}(D_{+}+D_{-})/RT \end{equation} (13)

where F is the Faraday constant, R is the gas constant, T is the absolute temperature, and D is the self-diffusion coefficient.

Figure 6.

(a) Relationship between ionicity scales and ratio of diffusion coefficients (D_Glyme/D_Li) for [Li(glyme)]X. Reprinted with permission from Ref. 48. Copyright 2012 American Chemical Society. (b) Schematic illustration for the effect of Lewis basicity of solvent and anion on the solvation structure in concentrated electrolytes.

Λ_imp/Λ_NMR indicates the ionic nature and self-dissociative properties of the system, while D_Glyme/D_Li obtained via PFG-NMR represents the ratio of the self-diffusion coefficient of the lithium cation to that of the solvent. If the solvated cations are stable, D_Glyme/D_Li should approach 1. For equimolar mixtures with a high Λ_imp/Λ_NMR value, the D_Glyme/D_Li value is nearly 1. In contrast, for equimolar mixtures containing anions with high Lewis basicity, such as [TfO]⁻, NO₃⁻, and trifluoroacetate ([TFA]⁻), the Λ_imp/Λ_NMR value is low, and the D_Glyme/D_Li value is greater than 1. Thus, when anions with high Lewis basicity are used, the interaction between the lithium cation and anion becomes dominant, leading to an unstable solvated cation structure (Fig. 6b). This trend was also supported by Raman spectroscopy results, which quantitatively revealed that for systems with an anion of low Lewis basicity, such as [TFSA]⁻, the content of free solvent molecules is less than 3 % (Fig. 7a).⁴⁹ In contrast, for anions with high Lewis basicity, more than 50 % of the solvent molecules remain free, indicating that the lithium cation and anion interact strongly. The absence of free solvent is also reflected in the electrode potential exhibited by lithium metal in each solution.^49–51 It is well known that the equilibrium potential of an electrochemical reaction is determined by the activity of the reactive species in the system, according to the Nernst equation.

  

\begin{equation} \text{Ox} + \text{$n$e$^{-}$} \rightleftarrows \text{Red} \end{equation} (14)

  

\begin{equation} E = E^{0} + \frac{RT}{nF}\ln \frac{a_{\text{Ox}}}{a_{\text{Red}}} \end{equation} (15)

where, a_ox and a_red are the activity of oxidant and reductant species, respectively, E is electrode potential, and E⁰ is standard electrode potential. In the case of the dissolution and deposition reactions of Li, the potential is determined as follows, by assuming the activity of solid lithium metal as 1:

  

\begin{equation} \text{Li$^{+}$} + \text{e$^{-}$} \rightleftarrows \text{Li} \end{equation} (16)

  

\begin{equation} E = E^{0} + \frac{RT}{F}\ln a_{\text{Li${^{+}}$}} \end{equation} (17)

However, this equation assumes that the activity of the solvent is 1, which holds true in a dilute electrolyte solution. When the solvent is also considered in the reaction system, the reaction can be represented as

  

\begin{equation} \text{[Li(solv)$x$]$^{+}$} + \text{e$^{-}$} \rightleftarrows \text{Li} + \text{$x$ solv} \end{equation} (18)

  

\begin{equation} E = E^{0} + \frac{RT}{F}\ln \frac{a_{\text{[Li(solv)]${^{+}}$}}}{a_{\textit{solv}}^{x}} \end{equation} (19)

If the activity of the solvent is assumed to be 1, then Eq. 19 becomes identical to Eq. 17. For dilute systems, which contain an excess of the solvent, the activity of the solvent can be assumed to be 1. However, in highly concentrated electrolytes, where the activity of the free solvent is extremely low, assuming the activity of the solvent to be 1 is inappropriate. In Eq. 19, the logarithmic term diverges to infinity as a_solv approaches zero. As shown in Figs. 7b–7c, in high-concentration regions, where free solvent is extremely scarce, the potential shifts significantly to the positive side. The magnitude of this shift has been reported to consistently correspond with the order of free-solvent concentration determined by Raman spectroscopy.

Figure 7.

(a) Estimated percentages of free glyme in equimolar molten mixtures [Li(glyme)₁]X at 30 °C. Plots of the Li/Li⁺ electrode potential against the common logarithm of the Li salt concentration for (b) [Li(G3)_n]X and (c) [Li(G4)_n]X at 30 °C. Li/Li⁺ in 1 mol dm⁻³ Li[TFSA]/G3 was used as the reference electrode. Open symbols represent the calculated electrode potential of Li/Li⁺, and a_solv was obtained from Raman spectral analysis. Reproduced with permission from Ref. 49. Copyright 2015 the Royal Society of Chemistry.

Furthermore, changing the chain length of the glyme solvent also affects the stability of the solvated cation structure. According to a previous study, Li[TFSA] with a sufficiently low Lewis basicity can maintain stable solvated cations in the G3 and G4 systems (as discussed in Fig. 6b). When the lithium salt was fixed to Li[TFSA], and the glyme chain length of the solvent was varied, the D_Glyme/D_Li value was observed to exceed 1 as the glyme chain length decreased, as shown in Fig. 8.⁵² This is attributed to the fact that even among the same ether-based solvents, multidentate glyme stabilizes the solvated cations more effectively due to its chelating effect. For ligands such as tetrahydrofuran or monoglyme (G1), which are monodentate or bidentate, the solvation of the lithium cation involves four coordinated ether oxygens at a molar ratio similar to that of the 1 : 1 mixture using G3 as the solvent. However, rapid ligand exchange leads to a shorter lifetime of the solvated cations, and they do not exhibit the properties characteristic of solvate ILs.⁵² Again, the stability of solvate cations is determined by the balance of interactions among the ligands (solvents), lithium cations, and anions. For forming stable solvated cations, the ligands should sufficiently stabilize the lithium cation, and the anion should have sufficiently low Lewis basicity (Fig. 6b).

Figure 8.

D_sol/D_Li ratio at 30 °C for [Li(glyme or THF)_x][TFSA] mixture. Reprinted with permission from Ref. 52. Copyright 2014 American Chemical Society.

As solvate ILs are formed using an equimolar mixture of a lithium salt and multidentate-solvent, they can be regarded as a type of highly concentrated electrolyte. The following section will focus on the broader concept of highly concentrated electrolytes and discuss their unique characteristics.

5. Concentrated Electrolytes

The previously mentioned solvate ILs are highly concentrated electrolytes composed of multidentate ligands and anions of low Lewis basicity. However, in highly concentrated solutions with bidentate or monodentate ligands, despite the lack of the chelate stabilization effect, which leads to volatility, various unique behaviors have been observed. Although these liquid systems cannot be referred to as solvate ILs because of their volatility, they still exhibit a range of unique behaviors that are not observed in solutions with salt concentrations of approximately 1 mol/L. Furthermore, the broader choice of available solvents allows for greater flexibility in electrolyte design compared to solvate ILs. This section discusses the unique interfacial reactions of highly concentrated electrolytes, including solvate ILs, with a focus on two main aspects: (1) expanding the potential window and (2) suppressing the solubility of reaction products.

5.1 Expanding the potential window

In 2003, Ogumi, Abe, and Inaba reported that despite using solvents that do not form an SEI, simply increasing the concentration of the lithium salt in PC resulted in stable charge–discharge behavior at the graphite anode.⁵³ To the best of the author’s knowledge, this is the first study that discusses the unique characteristics of using a highly concentrated electrolyte in an LIB. In relation to this, it has been reported that the reduction stability of the electrolyte is improved in highly concentrated electrolyte. For a highly concentrated electrolyte in acetonitrile, a solvent that has been considered to have poor reduction stability and unsuitable for lithium metal or graphite anodes, it has been demonstrated that reductive decomposition does not proceed even when the solvent contacts lithium metal, and the graphite anode operates stably.⁵⁴ It has been suggested that as the lithium salt concentration increases, the lowest unoccupied molecular orbital (LUMO) shifts from the solvent to anion, causing the anion to undergo reductive decomposition at the anode surface, thereby forming an SEI rich in anion-derived components (e.g., F-rich or S-rich), which prevent further reductive decomposition. This anion-derived SEI formation occurs effectively on the graphite anode in an EC-free electrolyte as well, enabling its stable operation. On the other hand, this trend can also be explained by the positive shift in the electrode potential of the desolvation reaction in the highly concentrated electrolytes mentioned earlier (as the solvent co-intercalation reaction does not involve free solvent, it does not cause a positive shift at high concentrations). However, the detailed underlying mechanisms remain unclear.

Furthermore, improvements in the oxidative stability of electrolytes have also been reported. For instance, in glyme-based electrolytes, the coordination of glyme to lithium cation lowers the highest occupied molecular orbital (HOMO) energy level of the ether oxygen in the glyme molecule. Moreover, the absence of an uncoordinated free solvent in the system further enhances the oxidative stability of the electrolyte. As shown in the linear sweep voltammogram in Fig. 9a, the oxidative stability of the solvent increases with the concentration of the lithium salt. It is generally known that ether-based solvents cannot be used in so-called 4 V-class cathodes such as LiCoO₂. Indeed, in an electrolyte with an excess glyme ([Li(G3)₄][TFSA]; 1 : 4 mixture), the Li/LiCoO₂ cell does not exhibit stable charge–discharge reactions owing to the oxidative decomposition of the solvent. However, in [Li(G3)₁][TFSA] (1 : 1 mixture), the increased oxidative stability allows for stable charge–discharge reactions over 200 cycles, despite the use of an ether-based solvent (Figs. 9b and 9c).⁵⁵ Additionally, leveraging the increased oxidative stability, it has been reported that LiNi_0.5Mn_1.5O₄, a 5 V-class cathode material, exhibits excellent cycling performance in highly concentrated electrolytes.^56–58 Moreover, not only is the oxidative stability improved but also the solvent molecules that are directly coordinated to Li⁺ in the first solvation shell find it difficult to access the catalytic sites in the LiNi_xMn_yCo_1−x−yO₂ electrode,⁵⁹ such as the NiO surface, that promote decomposition reactions. A highly concentrated electrolytes contains only a trace amount of the free solvent, which normally has easy access to the catalytic surface; consequently, electrolyte decomposition is suppressed.

Figure 9.

(a) Linear sweep voltammograms (left) of [Li(glyme)_x][TFSA] (x = 1, 4, 8, and 20) at a scan rate of 1 mV s⁻¹ at 30 °C. Charge and discharge curves of (b) [Li | [Li(G3)₁][TFSA] | LiCoO₂] and (c) [Li | [Li(G3)₄][TFSA] | LiCoO₂] cells recorded at a current density of 50 µA cm⁻² (1/8 C-rate) at 30 °C. Reprinted with permission from Ref. 55. Copyright 2011 American Chemical Society.

Some studies have also attempted to utilize the enhanced oxidative and reductive stability achieved by increasing the electrolyte concentration to operate 3 V-class LIBs using concentrated aqueous lithium salt electrolytes.^60,61 Although aqueous systems inherently have a potential window of only 1.23 V, recent studies have focused on expanding the potential window of aqueous LIBs by thermodynamically and kinetically suppressing decomposition reactions through SEI formation.

5.2 Reducing the solubility of reaction intermediates

The lack of free solvent in the electrolyte is reported to drastically decrease the solubility of the reaction intermediates of electrode reactions, simply because free solvent is not available to further dissolve the solute. Lithium–sulfur batteries, which have a theoretical capacity of 1672 mAh g⁻¹ (8Li₂S = S₈ + 16Li⁺ + 16e⁻), hold great promise as low-cost alternatives to LIBs;⁶² however, the dissolution of intermediates, such as Li₂S_x species, in the electrolyte is a major issue that hinders their application. When the dissolved intermediates reach the anode, they can cause a “redox shuttle” (Fig. 10), and the intermediates are repeatedly reduced at the anode surface during charging and then oxidized (charged) again at the cathode. However, in solvate ILs, only trace amounts of free solvents are available to solvate these intermediates. This limits the solubility of the Li₂S_x species, resulting in more stable charge–discharge reactions.^63–65

Figure 10.

Schematic illustration of redox shuttle in Li-S batteries.

The effect of the decreased solubility of reaction intermediates in metal–air (metal–O₂) batteries has also been reported. In lithium–air batteries, the oxygen reduction reaction in aprotic organic solvents containing lithium salts is widely believed to proceed as follows:^66,67

  

\begin{align} &\text{O$_{2}$} + \text{Li$^{+}$} + \text{e$^{-}$}\to \text{LiO$_{2}$}\notag\\ &\quad \text{(Electrochemical reduction of O$_{2}$)} \end{align} (20)

  

\begin{align} &\text{LiO$_{2}$} + \text{Li$^{+}$} + \text{e$^{-}$}\to \text{Li$_{2}$O$_{2}$}\notag\\ &\quad \text{(Electrochemical reduction of LiO$_{2}$)} \end{align} (21)

  

\begin{align} &\text{2LiO$_{2}$}\to \text{Li$_{2}$O$_{2}$} + \text{O$_{2}$}\notag\\ &\quad \text{(Chemical disproportionation of LiO$_{2}$)} \end{align} (22)

Using the rotating ring-disk electrode (RRDE) method, the solubility of the reaction intermediates can be evaluated using the ring current to reoxidize the oxygen reduction products formed at the disk electrode. In the RRDE measurements of systems using highly concentrated electrolytes with low free-solvent activity, no current derived from the reaction intermediates was detected at the ring electrode (Figs. 11a–11c).⁶⁸ In contrast, in typical (non-concentrated) glyme- or DMSO-based electrolytes, a current was detected at the ring electrode, confirming that the intermediates had certain solubility and lifetime, which allowed them to diffuse to the electrolyte. Additionally, as superoxide (O₂⁻·) is a radical species, it is known to decompose solvents like DMSO.⁶⁹ In conventional concentration regions, where intermediates dissolve, the LiOH byproduct was observed during the charge–discharge tests, whereas this side reaction was suppressed in highly concentrated solutions. This result suggests that the decreased solubility of LiO₂ leads to its precipitation on the electrode surface, where it is subsequently rapidly disproportionated to form Li₂O₂, as shown in Eqs. 21 and 22, resulting in shortened lifespan of the highly reactive intermediates. In contrast, in sodium–air batteries, unlike in lithium–air systems, NaO₂ is the final discharge product.⁷⁰ This is because the relatively larger ionic radius of Na⁺ (closer to the ionic radius of O₂⁻) allows NaO₂ to exist stably, unlike LiO₂. During the investigation of the precipitation morphology of NaO₂ using ether-based electrolytes, which are relatively stable against superoxide, an interesting behavior was observed: the particle size was the largest in the intermediate concentration region, but it decreased in high and low concentration regions (Figs. 11d–11h).⁷¹ The smaller particle size in the highly concentrated region could be due to the decreased solubility of NaO₂, similar to the case in lithium–air batteries, owing to the reduced amount of free solvent. In low concentration regions, the decreased presence of Na⁺ in the electrolyte leads to decreased solubility of NaO₂ owing to the entropy gain associated with the formation of Na₂O₂⁺ (for details, refer to the previous report⁷¹).

Figure 11.

RRDE responses on GC disk/GC ring electrode measured at 1000 rpm at 100 mV s⁻¹ in O₂ saturated (a) 0.1 mol dm⁻³ and (b) 3 mol dm⁻³ Li[TFSA]/DMSO at 30 °C. The potential of the GC disk electrode was swept from the open circuit potential of ca. 3 V to the negative direction, and the sweep direction was reversed at 1.5 V_Li. The potential of the GC ring electrode was set to be 3.3 V_Li. (c) Concentration dependency of peak ring/disk current ratio (N); data labels are the fraction of free DMSO determined by Raman analysis. Reprinted with permission from Ref. 68. Copyright 2017 American Chemical Society. SEM images of Na-O₂ cathode discharged at 100 mA g⁻¹ for 1000 mAh g⁻¹ in (d) 1 : 100, (e) 1 : 20, (f) 1 : 8, (g) 1 : 4 and (h) 1 : 2.5 mixtures of NaTFSA : dimethoxyethane. Reprinted with permission from Ref. 71. Copyright 2018 American Chemical Society.

Furthermore, when salts such as LiFSA and LiTFSA are used, corrosion reactions of aluminum, which is commonly used as a current collector on the cathode side, occur, preventing their practical application. However, it has been reported that even when LiTFSA or LiFSA is used, increasing the salt concentration leads to the absence of free solvent that could solvate the dissolved Al ions in the system.^72,73 Consequently, corrosion is halted at the first surface layer of the Al foil, which prevents further corrosion of the Al collector.

6. Conclusion

Traditionally, electrolyte solutions with a lithium salt concentration of approximately 1 mol/L have been used in LIBs. This is because of the trade-off relationship between the density of carrier ions and the viscosity of the solution, resulting in the maximum ionic conductivity of the electrolyte being observed near a lithium salt concentration of 1 mol/L. However, in recent years, the unique electrode reaction in highly concentrated electrolyte solutions with salt concentrations exceeding 3 mol/L have garnered attention. Generally, at ∼1 mol/L concentration, the molar ratio of the lithium salt to solvent is approximately 1 : 10, but the general solvation number of Li⁺ in an aprotic solution is considered to be 4–5. Therefore, at concentrations near 1 mol/L, the electrolyte contains “free solvent” that does not participate in the coordination of lithium cations. As the electrolyte concentration increases, the amount of “free solvent” decreases, and it becomes trace in highly concentrated electrolytes at more than 3 mol/L. The unique electrode reactions associated with the increase in electrolyte concentration are considerably influenced by the activity of the “free solvent.” This review discussed the activities of free solvents in various highly concentrated electrolytes and the corresponding electrode reactions. As highly concentrated electrolytes are in a largely unexplored concentration range, far from the Debye–Hückel limit, many aspects remain unclear, and further research is required to unravel their behaviors.

Acknowledgments

The author sincerely thanks Prof. Masayoshi Watanabe, Prof. Kaoru Dokko, Prof. Kazuhide Ueno, Prof. Hisashi Kokubo, Prof. Yang Shao-Horn, Prof. Shinichi Komaba, Prof. Kei Kubota, Prof. Tomooki Hosaka, many students, postdoctoral researchers, and collaborators. Financial support from JSPS KAKENHI (JP14J00165, JP21K14724, and JP23K13829), JST-PRESTO (JPMJPR2374), JST-GteX (JPMJGX23S4), NEDO Intensive Support Program for Young Promising Researchers (JPNP20004), TEPCO Memorial Foundation research grant (basic research), Takahashi Industrial and Economic Research Foundation, and ECSJ Kanto branch research grant is gratefully acknowledged.

The APC for the publication of this paper was supported by The Electrochemical Society of Japan.

CRediT Authorship Contribution Statement

Ryoichi Tatara: Writing – original draft (Lead)

Conflict of Interest

The author declares no conflicts of interest.

Funding

Japan Society for the Promotion of Science: JP14J00165

Japan Society for the Promotion of Science: JP21K14724

Japan Society for the Promotion of Science: JP23K13829

Japan Science and Technology Agency: JPMJPR2374

Japan Science and Technology Agency: JPMJGX23S4

New Energy and Industrial Technology Development Organization: JPNP20004

TEPCO Memorial Foundation

Takahashi Industrial and Economic Research Foundation

Electrochemical Society of Japan, Kanto Branch

Footnotes

R. Tatara: ECSJ Active Member

References

• 1)  P. Debye and E. Huckel, Z. Phys., 24, 185 (1923).
• 2)  P. Atkins, J. d. Paula, and J. Keeler, Atkins’ Physical Chemistry, Oxford University Press (1998).
• 3)  K. Izutsu, Electrochemistry in Nonaqueous Solutions (2nd ed.), Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim (2009).
• 4)  J. O. M. Bockris and A. K. N. Reddy, Modern Electrochemistry (2nd ed.), Springer US, New York (1998).
• 5)  Y. Ugata, K. Shigenobu, R. Tatara, K. Ueno, M. Watanabe, and K. Dokko, Phys. Chem. Chem. Phys., 23, 21419 (2021).
• 6)  Y. Yamada and A. Yamada, J. Electrochem. Soc., 162, A2406 (2015).
• 7)  Y. Yamada, J. Wang, S. Ko, E. Watanabe, and A. Yamada, Nat. Energy, 4, 269 (2019).
• 8)  M. Watanabe, M. L. Thomas, S. Zhang, K. Ueno, T. Yasuda, and K. Dokko, Chem. Rev., 117, 7190 (2017).
• 9)  K. Dokko, Electrochemistry, 90, 101003 (2022).
• 10)  V. Gutmann, The Donor-Acceptor Approach to Molecular Interactions, Plenum Press (1978).
• 11)  V. Gutmann and E. Wychera, Inorg. Nucl. Chem. Lett., 2, 257 (1966).
• 12)  U. Mayer, V. Gutmann, and W. Gerger, Monatsh. Chem., 106, 1235 (1975).
• 13)  K. Xu, Chem. Rev., 104, 4303 (2004).
• 14)  K. Mizushima, P. C. Jones, P. J. Wiseman, and J. B. Goodenough, Mater. Res. Bull., 15, 783 (1980).
• 15)  N. Yabuuchi and T. Ohzuku, J. Power Sources, 119–121, 171 (2003).
• 16)  M. Gauthier, P. Karayaylali, L. Giordano, S. Feng, S. F. Lux, F. Maglia, P. Lamp, and Y. Shao-Horn, J. Electrochem. Soc., 165, A1377 (2018).
• 17)  T. Welton, Chem. Rev., 99, 2071 (1999).
• 18)  P. Walden, Bull. Acad. Imp. Sci. St.-Petersbourg, 8, 405 (1914).
• 19)  F. H. Hurley and T. P. Wier, J. Electrochem. Soc., 98, 203 (1951).
• 20)  J. S. Wilkes, J. A. Levisky, R. A. Wilson, and C. L. Hussey, Inorg. Chem., 21, 1263 (1982).
• 21)  J. S. Wilkes and M. J. Zaworotko, J. Chem. Soc., Chem. Commun., 965 (1992).
• 22)  A. Lewandowski and A. Świderska-Mocek, J. Power Sources, 194, 601 (2009).
• 23)  M. Armand, F. Endres, D. R. MacFarlane, H. Ohno, and B. Scrosati, Nat. Mater., 8, 621 (2009).
• 24)  P. Wang, S. M. Zakeeruddin, P. Comte, I. Exnar, and M. Gratzel, J. Am. Chem. Soc., 125, 1166 (2003).
• 25)  R. Kawano, H. Matsui, C. Matsuyama, A. Sato, M. A. B. H. Susan, N. Tanabe, and M. Watanabe, J. Photochem. Photobiol. A, 164, 87 (2004).
• 26)  S. Y. Lee, A. Ogawa, M. Kanno, H. Nakamoto, T. Yasuda, and M. Watanabe, J. Am. Chem. Soc., 132, 9764 (2010).
• 27)  A. Noda, M. A. B. H. Susan, K. Kudo, S. Mitsushima, K. Hayamizu, and M. Watanabe, J. Phys. Chem. B, 107, 4024 (2003).
• 28)  T. Yasuda, S. Nakamura, Y. Honda, K. Kinugawa, S. Y. Lee, and M. Watanabe, ACS Appl. Mater. Interfaces, 4, 1783 (2012).
• 29)  M. Ue, M. Takeda, A. Toriumi, A. Kominato, R. Hagiwara, and Y. Ito, J. Electrochem. Soc., 150, A499 (2003).
• 30)  M. Galiński, A. Lewandowski, and I. Stępniak, Electrochim. Acta, 51, 5567 (2006).
• 31)  S. Imaizumi, H. Kokubo, and M. Watanabe, Macromolecules, 45, 401 (2012).
• 32)  Y. Kitazawa, T. Ueki, L. D. McIntosh, S. Tamura, K. Niitsuma, S. Imaizumi, T. P. Lodge, and M. Watanabe, Macromolecules, 49, 1414 (2016).
• 33)  Y. Kobayashi, Y. Kitazawa, T. Komori, K. Ueno, H. Kokubo, and M. Watanabe, Macromol. Rapid Commun., 37, 1207 (2016).
• 34)  T. Ueki, R. Usui, Y. Kitazawa, T. P. Lodge, and M. Watanabe, Macromolecules, 48, 5928 (2015).
• 35)  L. A. Blanchard, D. Hancu, E. J. Beckman, and J. F. Brennecke, Nature, 399, 28 (1999).
• 36)  E. D. Bates, R. D. Mayton, I. Ntai, and J. H. Davis, J. Am. Chem. Soc., 124, 926 (2002).
• 37)  A. J. Carmichael, M. J. Earle, J. D. Holbrey, P. B. McCormac, and K. R. Seddon, Org. Lett., 1, 997 (1999).
• 38)  H. Kokubo and M. Watanabe, Polym. Adv. Technol., 19, 1441 (2008).
• 39)  S. Kuwabata, A. Kongkanand, D. Oyamatsu, and T. Torimoto, Chem. Lett., 35, 600 (2006).
• 40)  T. Tsuda, N. Nemoto, K. Kawakami, E. Mochizuki, S. Kishida, T. Tajiri, T. Kushibiki, and S. Kuwabata, ChemBioChem, 12, 2547 (2011).
• 41)  C. Austen Angell, Y. Ansari, and Z. Zhao, Faraday Discuss., 154, 9 (2012).
• 42)  K. Ueno, H. Tokuda, and M. Watanabe, Phys. Chem. Chem. Phys., 12, 1649 (2010).
• 43)  R. Tatara, N. Tachikawa, H.-M. Kwon, K. Ueno, K. Dokko, and M. Watanabe, Chem. Lett., 42, 1053 (2013).
• 44)  T. M. Pappenfus, W. A. Henderson, B. B. Owens, K. R. Mann, and W. H. Smyrl, J. Electrochem. Soc., 151, A209 (2004).
• 45)  T. Mandai, K. Yoshida, K. Ueno, K. Dokko, and M. Watanabe, Phys. Chem. Chem. Phys., 16, 8761 (2014).
• 46)  T. Tamura, K. Yoshida, T. Hachida, M. Tsuchiya, M. Nakamura, Y. Kazue, N. Tachikawa, K. Dokko, and M. Watanabe, Chem. Lett., 39, 753 (2010).
• 47)  K. Yoshida, M. Tsuchiya, N. Tachikawa, K. Dokko, and M. Watanabe, J. Phys. Chem. C, 115, 18384 (2011).
• 48)  K. Ueno, K. Yoshida, M. Tsuchiya, N. Tachikawa, K. Dokko, and M. Watanabe, J. Phys. Chem. B, 116, 11323 (2012).
• 49)  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).
• 50)  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).
• 51)  R. Tatara, K. Ueno, K. Dokko, and M. Watanabe, ChemElectroChem, 6, 4444 (2019).
• 52)  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).
• 53)  S.-K. Jeong, M. Inaba, Y. Iriyama, T. Abe, and Z. Ogumi, Electrochem. Solid-State Lett., 6, A13 (2003).
• 54)  Y. Yamada, K. Furukawa, K. Sodeyama, K. Kikuchi, M. Yaegashi, Y. Tateyama, and A. Yamada, J. Am. Chem. Soc., 136, 5039 (2014).
• 55)  K. Yoshida, M. Nakamura, Y. Kazue, N. Tachikawa, S. Tsuzuki, S. Seki, K. Dokko, and M. Watanabe, J. Am. Chem. Soc., 133, 13121 (2011).
• 56)  T. Doi, Y. Shimizu, M. Hashinokuchi, and M. Inaba, J. Electrochem. Soc., 163, A2211 (2016).
• 57)  T. Doi, R. Masuhara, M. Hashinokuchi, Y. Shimizu, and M. Inaba, Electrochim. Acta, 209, 219 (2016).
• 58)  J. Wang, Y. Yamada, K. Sodeyama, C. H. Chiang, Y. Tateyama, and A. Yamada, Nat. Commun., 7, 12032 (2016).
• 59)  R. Tatara, Y. Yu, P. Karayaylali, A. K. Chan, Y. Zhang, R. Jung, F. Maglia, L. Giordano, and Y. Shao-Horn, ACS Appl. Mater. Interfaces, 11, 34973 (2019).
• 60)  L. Suo, O. Borodin, T. Gao, M. Olguin, J. Ho, X. Fan, C. Luo, C. Wang, and K. Xu, Science, 350, 938 (2015).
• 61)  Y. Yamada, K. Usui, K. Sodeyama, S. Ko, Y. Tateyama, and A. Yamada, Nat. Energy, 1, 16129 (2016).
• 62)  S. Zhang, K. Ueno, K. Dokko, and M. Watanabe, Adv. Energy Mater., 5, 1500117 (2015).
• 63)  N. Tachikawa, K. Yamauchi, E. Takashima, J. W. Park, K. Dokko, and M. Watanabe, Chem. Commun., 47, 8157 (2011).
• 64)  J.-W. Park, K. Yamauchi, E. Takashima, N. Tachikawa, K. Ueno, K. Dokko, and M. Watanabe, J. Phys. Chem. C, 117, 4431 (2013).
• 65)  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).
• 66)  K. M. Abraham and Z. Jiang, J. Electrochem. Soc., 143, 1 (1996).
• 67)  P. G. Bruce, S. A. Freunberger, L. J. Hardwick, and J. M. Tarascon, Nat. Mater., 11, 19 (2012).
• 68)  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. Watanabete, J. Phys. Chem. C, 121, 9162 (2017).
• 69)  D. G. Kwabi, T. P. Batcho, C. V. Amanchukwu, N. Ortiz-Vitoriano, P. Hammond, C. V. Thompson, and Y. Shao-Horn, J. Phys. Chem. Lett., 5, 2850 (2014).
• 70)  P. Hartmann, C. L. Bender, M. Vracar, A. K. Durr, A. Garsuch, J. Janek, and P. Adelhelm, Nat. Mater., 12, 228 (2013).
• 71)  R. Tatara, G. M. Leverick, S. Feng, S. Wan, S. Terada, K. Dokko, M. Watanabe, and Y. Shao-Horn, J. Phys. Chem. C, 122, 18316 (2018).
• 72)  Y. Yamada, C. H. Chiang, K. Sodeyama, J. Wang, Y. Tateyama, and A. Yamada, ChemElectroChem, 2, 1687 (2015).
• 73)  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).

Biographies

Ryoichi Tatara (Associate Professor, Department of Chemistry and Life Science, Yokohama National University)

Ryoichi Tatara was born in 1990. He graduated from Graduate School of Engineering, Yokohama National University in March 2017, earning a Doctor of Engineering. Following that, he served as a Postdoctoral Associate at the Massachusetts Institute of Technology. He then joined the Tokyo University of Science as an Assistant Professor. In 2024, he returned to Yokohama National University as an Associate Professor. He was awarded Young Researcher Award (Sano Award) from The Electrochemical Society of Japan in 2024. His current research interests include the physicochemical properties of electrolyte solutions, interfacial chemistry at the electrode/electrolyte interface, and electrochemical reactions in Li/Na/K batteries.