2025 年 93 巻 3 号 p. 037008
Surfactants possess unique properties in bulk solutions and at interfaces, naturally forming self-assembled structures. Herein, cetyltrimethylammonium trifluoroacetate (CTATFA) was incorporated into aqueous electrolytes as a cationic surfactant to enhance their ionic conductivity and electrochemical stability. The presence of CTATFA widened the electrochemical stability windows of both Li-based and Zn-based electrolytes. The Zn-based electrolyte exhibited high ionic conductivity and low viscosity in the bulk solution. In a Zn symmetric cell, the electrolyte containing 1 M Zn(TFA)2-0.5 M CTATFA demonstrated excellent Zn plating/stripping reversibility for over 800 h at 1 mAh cm−2 and 1 mA cm−2. A Zn-Cu cell with 1 M Zn(TFA)2-0.5 M CTATFA exhibited excellent reversibility, achieving over 300 plating/stripping cycles at 5 mA cm−2 and 5 mAh cm−2. The Zn/MnO2 cell using the Zn-based electrolyte also demonstrated a specific capacity of 105 mAh g−1 over 750 cycles at a current density of 0.5 A g−1. This study provides insight into the design of high-performance aqueous electrolytes based on the self-assembly and surface adsorption of cationic surfactants.
In modern society, an electricity-powered lifestyle has become a prevailing trend. Electrochemical energy storage systems (ESSs) play a pivotal role in both digital products and vehicle electrification.1 Electrolytes are indispensable components of these ESSs, serving as the medium for ion transport.2 Nonaqueous electrolytes, known for their wide electrochemical stability windows (ESWs) and relatively high ionic conductivities, are commonly used in lithium-ion batteries (LIBs).3 However, their flammability and toxicity pose significant safety concerns. As a result, aqueous electrolytes have garnered increasing interest owing to their nonflammable, nontoxic, cost-effective nature and high ionic conductivity.4 Despite these advantages, the narrow ESW (1.23 V) of water impedes the commercialization of aqueous LIBs.5 To address this challenge, various strategies, such as water-in-salt (WiSE),6 molecular crowding,7 and the introduction of organic solvent,8 have been explored to widen the ESW of aqueous electrolytes. However, these approaches often increase the production cost and compromise ionic conductivity, environmental compatibility, and nontoxicity. Therefore, an alternative strategy is required to overcome the narrow ESW of aqueous electrolytes while retaining their inherent advantages.
Surfactants are amphiphilic molecules consisting of a hydrophilic head and a hydrophobic tail.9 In addition to their widespread applications in colloid and interface sciences, ionic surfactants have also been used in aqueous battery electrolytes to widen their ESWs.10–12 In a previous study, LIBs using a concentrated aqueous solution of lithium dodecyl sulfate (an anionic surfactant) as the electrolyte exhibited a highly selective Li ion conduction with an ionic conductivity of 37 mS cm−1 and an ESW of 3.0 V.10 Various anionic11 and cationic12 surfactants have also been added to aqueous electrolytes to protect the electrode surface of sodium-ion and zinc metal batteries. However, the mechanisms underlying surfactant self-assembly in bulk solutions and at interfaces in aqueous batteries remain unclear. Therefore, understanding the relationship between the characteristics of surfactants in electrolytes and the electrochemical properties of aqueous batteries is crucial for their applications.
Polyelectrolytes are typically formed via the covalent bonding of ionic monomers, resulting in charged polyanions or polycations. Nonaqueous polyelectrolytes can enhance selective ion transport in bulk solutions and modify the electrode-electrolyte interface, thereby improving the performance of lithium- and sodium-ion batteries.13–17 Additionally, polyelectrolytes have been used in aqueous zinc metal batteries.18 Under appropriate conditions, ionic surfactants spontaneously form micelles with specific shapes and functions through noncovalent bonds.10,19 These ionic micelles can act as polyelectrolytes to provide favorable bulk and interfacial properties in aqueous batteries.
In this study, we used cationic surfactant-based aqueous electrolytes containing cetyltrimethylammonium trifluoroacetate (CTATFA), lithium trifluoroacetate (LiTFA), and zinc trifluoroacetate (Zn(TFA)2. These electrolytes can be easily prepared by dissolving CTATFA and salts (LiTFA or Zn(TFA)2) at different ratios in water. The addition of CTATFA enhanced the ESW of the aqueous electrolytes (1 M LiTFA-0.5 M CTATFA and 1 M Zn(TFA)2-0.5 M CTATFA). In a comparative study, the effects of different cations (Li+ and Zn2+) on the rheological behavior and ionic conductivity of the electrolytes were systematically investigated. The transport properties of bulk solutions were also studied using pulsed-field gradient nuclear magnetic resonance (PFG-NMR) measurements. The electrolyte containing 1 M Zn(TFA)2-0.5 M CTATFA exhibits low viscosity and high ionic conductivity. Its zinc stripping/plating reversibility, coulombic efficiency, and water corrosion resistance were significantly enhanced compared to the reference electrolyte without CTATFA. These results are highly beneficial for understanding the impact of self-assembled cationic surfactants on the transport and electrochemical properties of aqueous electrolytes, as well as for their introduction and application in aqueous batteries.
Cetyltrimethylammonium bromide (CTAB), hexadecyltrimethylammonium hydroxide, and deuterium oxide (D2O) were purchased from Tokyo Chemical Industry. LiTFA, Zn(TFA)2, lithium bis(trifluoromethanesulfonyl)imide (LiTFSI), lithium tetrafluoroborate (LiBF4), lithium trifluoromethanesulfonate (LiOTf), lithium acetate (LiOAc), lithium sulfate (Li2SO4), lithium bromide (LiBr), zinc powder, trifluoroacetic acid, and N-methyl-2-pyrrolidone (NMP) were purchased from Wako Pure Chemical. Trifluoroacetic anhydride (TFA) was purchased from Sigma Aldrich. Zinc sheets (200 µm thickness) were purchased from Nilaco Corporation. Prior to use, the zinc sheets were acid-washed with a 1 M HCl aqueous solution, followed by rinsing with deionized water, and then polished with sandpaper. Ultrapure water (18 MΩ cm) was obtained from a Milli-Q Integral 3 Pure water machine. Each salt was dissolved in ultrapure water in an appropriate ratio.
2.2 Preparation of CTATFA and Zn(TFA)2CTATFA was synthesized via a straightforward one-step neutralization reaction. First, trifluoroacetic acid, at a molar ratio of 1.2 times that of hexadecyltrimethylammonium hydroxide, was dissolved in an appropriate amount of water. Next, a methanolic solution of hexadecyltrimethylammonium hydroxide was added dropwise to the aqueous solution of trifluoroacetic acid. After stirring for one hour to ensure that the reaction was complete, the solvent was evaporated using a vacuum pump equipped with a liquid-nitrogen condenser until a white solid product was obtained. The white product was vacuum dried for 12 h.
Zn(TFA)2 was synthesized according to a previously reported method.20 Trifluoroacetic acid, at a molar ratio of 1.2 times that of Zn powder, was dissolved in an appropriate amount of water. Then, Zn powder was added to the trifluoroacetic acid solution and stirred until all the Zn powder had dissolved. The solution was filtered and evaporated using a vacuum pump equipped with a liquid nitrogen condenser until glassy substances were obtained. The glassy substances were then treated with trifluoroacetic anhydride, followed by evaporation using a vacuum pump equipped with two liquid nitrogen condensers. Zn(TFA)2 was obtained as a crystalline solid. Mn(TFA)2 was also synthesized using Mn powder in a similar manner to Zn(TFA)2.
2.3 Materials characterizationIonic conductivity (σ) was measured using an impedance analyzer (VMP-3, Biologic Science Instruments) over a frequency range of 500 kHz to 100 mHz, with a voltage amplitude of 10 mV. Before measurement, a platinized platinum electrode cell (CG-511B, DKK-TOA) was calibrated with a 0.01 M KCl aqueous solution. The viscosity was determined using a rheometer (Physica MCR102, Anton Paar) equipped with a cone plate geometry (diameter: 50 mm; angle: 1°). Morphological analysis of the electrodes was performed using a scanning electron microscope (SEM, Hitachi FE-SEM SU8000, Japan).
The self-diffusion coefficients of D2O, Li+, TFA−, and CTA+ were measured using PFG-NMR on a JEOL-ECX 400 spectrometer. A stimulated echo (STE) pulse sequence with a sinusoidal pulsed-field gradient was used for the measurements.21 1H and 19F NMR spectra were recorded using a JEOL ECX-500 spectrometer at frequencies of 500 MHz and 470 MHz, respectively.
The oxidative and reductive stabilities of the electrolytes were evaluated using linear sweep voltammetry (LSV) on a VMP-3 electrochemical workstation (Bio-Logic Science Instruments). The measurement was conducted at a scan rate of 1 mV s−1 in a three-electrode system, which included glassy carbon (GC) or aluminum as the working electrode, a platinum wire as the counter electrode, and an Ag/AgCl reference electrode in a saturated aqueous KCl solution. Electrochemical impedance spectroscopy (EIS) of Zn symmetric cells was performed using an impedance analyzer (VMP-3, Bio-Logic Science Instruments) in the frequency range of 500 kHz to 10 mHz at a voltage of 5 mV. Zn symmetric cells were assembled with the two zinc plates, glass fiber separator (GA-55, Advantec) and the aqueous electrolyte, and EIS data was obtained at open circuit voltage.
The cathode was prepared by mixing commercial MnO2, acetylene black (AB), and polyvinylidene fluoride (PVDF) in N-methyl-2-pyrrolidone (NMP) at a weight ratio of 80 : 10 : 10 using a mixing machine. The resulting slurry was then coated onto a carbon paper current collector (12 mm in diameter). The electrodes were dried at 80 °C under vacuum for 12 h, with a MnO2 loading of approximately 1–2 mg cm−2. To minimize the side reactions between the electrolyte and coin cell components, a Ti foil (18 mm in diameter) was placed between the positive electrode and the bottom of the casing. An Al foil (18 mm in diameter) was positioned between the negative electrode sheet and the spacer, and the spacer was also wrapped in Al foil.22
Glass filter paper (GA-55, Advantec) was used as the separator. CR2032-type coin cells were assembled for electrochemical testing in a controlled atmosphere. Galvanostatic discharge/charge measurements were performed within a voltage range of 0.8–1.9 V, with all tests conducted at a stable temperature of 30 °C. The specific capacity of Zn/MnO2 cells was calculated based on the mass of MnO2.
The contact angles were measured using a smart contact mobile device (A511). An automatic surface tension meter (CBVP-Z) was utilized to measure the surface tension. Attenuated total reflection (ATR)-Fourier transform infrared (FTIR) spectra were recorded using a JASCO FT/IR-6600 spectrometer equipped with a PIKE Technologies horizontal ATR accessory containing a ZnSe crystal. The spectra were collected over 128 scans within a wavenumber range of 1000–4600 cm−1 and an optical resolution of 4.0 cm−1. X-ray photoelectron spectroscopy (XPS) analysis was performed using a PHI Quantera SXM ULVACPHI spectrometer. After stripping/plating, the Zn and Cu electrodes were rinsed with 1,2-dimethoxyethane (DME, Kishida Chemical Co.) to remove the residual electrolyte and then dried under vacuum for 24 h. The XPS binding energy was calibrated using the C 1s peak of adventitious carbon at 284.8 eV.
Figure 1 shows the chemical structures of the metal salts and cationic surfactant (CTATFA) used in this study. LiTFA and Zn(TFA)2 were employed to investigate the effects of metal cations on the physicochemical and electrochemical properties of aqueous electrolytes in the presence of CTATFA. The 16-carbon long chain in CTATFA imparts strong hydrophobicity, while its quaternary ammonium group provides sufficient electrochemical stability.23 As shown in Fig. S1, the chemical structure was confirmed by the 1H and 19F NMR spectra, which are consistent with previous reports.24,25 An aqueous electrolyte of zinc sulfate (ZnSO4) was used as a reference sample.
Chemical structures of the salts and cationic surfactant used in this work.
Micelles are nanoscale aggregates formed by the spontaneous self-assembly of surfactant molecules. CTA+ is the cation of a cationic surfactant that can spontaneously form micellar structures when its concentration exceeds the critical micelle concentration (CMC ≈ 0.9 mM at 25 °C).26 In this study, a concentration range of CTA+ (0.1–0.5 M, M = mol dm−3), significantly exceeding its CMC, was employed to optimize the ionic conductivity and ESW. In a preliminary experiment, a suitable anion was selected to pair with CTA+ to serve as a highly water-soluble cationic surfactant. An ideal anion should contain fluorine-containing functional groups to promote the formation of a fluorinated solid electrolyte interphase (SEI)6 while also ensuring water solubility and electrochemical stability. Based on these considerations, simple solubility experiments were conducted to screen potential anions. In Fig. S2, 0.4 mL of a salt solution (10 mg L−1) was added to 2 mL of a CTAB solution (1 mg L−1) to investigate the solubility of hydrophobic CTA+ in the presence of the paired anion. As shown in Fig. S2, the solutions containing TFA− (trifluoroacetate), OAc− (acetate), SO42− (sulfate), and Br− (bromide) remained clear and transparent, whereas the solutions with TFSI− (bis(trifluoromethanesulfonyl)imide), BF4− (tetrafluoroborate), and OTf− (triflate) appeared turbid. Additionally, TFA− anions offer higher electrochemical oxidative stability than the other hydrophilic anions (OAc−, SO42−, and Br−)27 due to the electron-withdrawing -CF3 functional group. The presence of -CF3 also provides an opportunity for the formation of poorly water-soluble, fluorine-containing SEI via anion decomposition.28 Given these advantages, TFA was selected as a suitable anion for this study.
Contact angle measurements were performed to investigate the wettability of the Al substrates. The measurements were performed after contact, at which point the contact angle was observed to stabilize. This time point was consistent across all experiments, and the measurements were taken immediately after this stabilization period. As shown in Figs. 2a–2c and Figs. S3a–S3c, compared to surfactant-free solutions, such as 1 M ZnSO4, 1 M Zn(TFA)2, 1 M LiTFA, and 1 M Li2SO4, the contact angles for the 1 M Zn(TFA)2-0.5 M CTATFA and 1 M LiTFA-0.5 M CTATFA were significantly lower (approximately 25°). The surface tension of 1 M LiTFA (60.1 mN m−1) and 1 M Zn(TFA)2 (55.0 mN m−1) was reduced to 25.0 mN m−1 and 32.9 mN m−1 for 1 M Zn(TFA)2-0.5 M CTATFA and 1 M LiTFA-0.5 M CTATFA, respectively, suggesting the interfacial adsorption and micelle formation of CTA+.29 The strong electrostatic attraction between the cationic surfactant and the Zn electrode makes the CTA+ prefer to be adsorbed on the Zn anode, which makes the use of CTATFA as an additive in electrolyte possible. The results suggest that the adsorption of CTA+ at the interface significantly reduces the total free energy of the droplet on the solid surface,30 which offers the potential to improve the electrochemical performance in aqueous batteries.
Contact angle measurement of (a) 1 M ZnSO4, (b) 1 M Zn(TFA)2, and (c) 1 M Zn(TFA)2-0.5 M CTATFA aqueous solution droplets on Al foil.
In aqueous electrolytes, water electrolysis is suppressed when the hydrogen-bonding network of water is disrupted by the strong hydration of ions.31 ATR FT-IR spectroscopy was used to investigate the hydrogen-bonded network in bulk solutions. As shown in Figs. S4a and S4b, the peak related to the OH stretching mode of the Li-based electrolytes in the range from 2900 cm−1 to 3700 cm−1 slightly shifts to a higher frequency with the addition of the salt and CTATFA. This suggests that the interaction of the ions of the salt and/or surfactant with water reduces the number of hydrogen bonds between water molecules, which is consistent with the observations in aqueous electrolytes containing different concentrations of CTAB in a previous study.31 Meanwhile, the peak shift was marginal in the Zn-based electrolytes, indicating that neither Zn(TFA)2 nor CTAFTA strongly affected the hydrogen-bond network of water in the concentration range studied. Based on the FTIR results, the minor changes in the hydrogen bonding network in the Li- and Zn-based electrolytes are expected to have a negligible effect on ESW. Therefore, the improvement in electrochemical stability of Li-based and Zn-based electrolytes with the addition of CTATFA primarily arises from the hydrophobicity of CTA+ and the SEI created by the localized high-concentration salt regions at the interface.10,32,33 This conclusion is further validated by electrochemical measurements and surface analyses, which will be discussed in the latter sections.
The ionic conductivities and rheological behaviors were studied at 30 °C for Li- and Zn-based aqueous electrolytes with different concentrations of CTATFA, i.e., 1 M LiTFA-x M CTATFA and 1 M Zn(TFA)2-x M CTATFA, where x represents the CTATFA concentration. The self-diffusion coefficients of the components and the Li-ion transference numbers were also investigated to understand the ion transport behavior in Li- and Zn-based electrolytes in the presence of CTATFA.
As shown in Fig. 3a, the ionic conductivity of the 1 M LiTFA-x M CTATFA was the highest when x was 0.1 (43 mS cm−1) and decreased with increasing CTATFA concentration. The initial increase in conductivity is ascribed to the rise in the number of charge carriers derived from CTATFA, whereas the subsequent reduction in conductivity is caused by the decreased ion mobility in the presence of highly concentrated CTATFA. In contrast, 1 M Zn(TFA)2-x M CTATFA showed a monotonic decrease in ionic conductivity with increasing CTATFA concentration. The ionic conductivity of the Zn-based electrolyte is higher than that of the Li-based electrolyte when x is lower than 0.3. This was primarily attributed to the higher number of charge carriers associated with divalent Zn2+. When x is increased to above 0.3, the Li-based electrolyte exhibits a slightly higher ionic conductivity. Nevertheless, both electrolytes show high ionic conductivity exceeding 25 mS cm−1 even at the highest CTATFA concentration (0.5 M).
(a) Ionic conductivity (x = 0–0.5) and (b) Rheological behavior (x = 0.5) of 1 M LiTFA-x M CTATFA and 1 M Zn(TFA)2-x M CTATFA. Self-diffusion coefficients of the components in (c) 1 M LiTFA-x M CTATFA and (d) 1 M Zn(TFA)2-x M CTATFA (x = 0–0.5). (e) Li transference number of the 1 M LiTFA and 1 M LiTFA-0.5 M CTATFA at 30 °C.
As shown in Fig. 3b, the Li- and Zn-based electrolytes exhibit very different rheological behaviors. For example, 1 M LiTFA-0.5 M CTATFA demonstrates shear rate-dependent non-Newtonian fluid behavior (shear thinning) with a significantly high steady shear viscosity (3210 mPa s) at lower shear rates.34 In contrast, 1 M Zn(TFA)2-0.5 M CTATFA shows a lower steady shear viscosity (80 mPa s) and behaves as a Newtonian fluid in the measured range of shear rate.35 The differences in rheological behavior between the Li- and Zn-based electrolytes (non-Newtonian or Newtonian behavior) are independent of the concentration of CTATFA, as shown in Figs. S5a–S5b. The zero-shear-rate viscosity increases with increasing x (Fig. S5c). However, the viscosities of the Zn-based electrolytes are more than an order of magnitude lower than those of the Li-based electrolytes. The significantly higher viscosity and non-Newtonian behavior observed in the Li-based electrolytes indicate the formation of rod- or worm-like micelles.36,37 In contrast, the lower viscosity and Newtonian behavior of the Zn-based electrolyte suggest that the micelles remain spherical or small. The electrostatic shielding effect induced by Zn2+ can mitigate the electrostatic repulsion between the micellar aggregates in 1 M Zn(TFA)2-x M CTATFA, preventing the formation of larger aggregates of CTA+ (rod- or worm-shaped micelles) and resulting in lower solution viscosity.
PFG-NMR was used to measure the self-diffusion coefficients of the ions and D2O (Figs. 3c and 3d). By using the solution viscosity (η) obtained from either the literature on pure solvents or direct measurements of micellar solutions, the average radius of the micelle can be estimated through the well-known Stokes–Einstein Eq. 1:38,39
| \begin{equation} D = \frac{kT}{6\pi a\eta } \end{equation} | (1) |
where k is the Boltzmann constant, T is the temperature (303.15 K), η is the viscosity of water (0.7977 mPa s), D is the diffusion coefficient, and a is the Stokes radius. As shown in Table 1, the self-diffusion coefficient of CTA+ in the 1 M LiTFA-0.5 M CTATFA was approximately half that in the 1 M Zn(TFA)2-0.5 M CTATFA. From the Stokes–Einstein equation, the average Stokes radius of the micelles formed by CTA+ in 1 M Zn(TFA)2-0.5 M CTATFA was nearly half of that of the micelles in the 1 M LiTFA-0.5 M CTATFA. This indicates that, compared to Li+, the electrostatic repulsion was reduced in the Zn-based electrolytes, resulting in a smaller average Stokes radius of the micelles.
| Electrolytes (mol L−1, 30 °C) |
DCTA+ (×10−7 cm2 s−1) |
Stokes radius (nm) |
|---|---|---|
| 1 M LiTFA-0.5 M CTATFA | 0.17 | 3.1 |
| 1 M Zn(TFA)2-0.5 M CTATFA | 0.32 | 1.6 |
In Figs. 3c and 3d, the self-diffusion coefficients of CTA+ ($D_{\text{CTA}^{ + }}$) at different concentrations are significantly lower than those of D2O, Li+, and TFA−. The micelles formed by the self-assembly of the surfactant monomer CTA+, along with their entangled and/or densely packed structures, hinder the migration of the CTA+ cations. This low self-diffusion coefficient is similar to that of typical polymer solutions.40,41 Therefore, the electrolytes containing CTATFA are considered self-assembled noncovalent polycation-based electrolytes. $D_{\textit{TFA}^{ - }}$ and DD2O of Li- and Zn-based electrolytes were comparable despite the significant difference in the zero-shear-rate viscosity, indicating that the aggregated structure of CTA+ was responsible for the macroscopic viscosity in Fig. 3b. The translational motion of the anions in the solvents was not affected, even in the presence of different CTA+ aggregate structures. Given the higher $D_{\textit{CTA}^{ + }}$ in the Zn-based electrolyte and the comparable $D_{\textit{TFA}^{ - }}$ and DD2O in the Li- and Zn-based electrolytes, the less dissociation of Zn(TFA)2 can contribute to the lower conductivity of the Zn-based electrolytes at the higher CTATFA concentration in Fig. 3a.
The Li-ion transference number can be estimated from the diffusion coefficients of the ions using the following Eq. 2:32,33
| \begin{equation} t_{\textit{Li}^{+}}^{\textit{NMR}}= \frac{N_{\textit{Li}^{+}}D_{\textit{Li}^{+}}}{N_{\textit{Li}^{+}}D_{\textit{Li}^{+}} + N_{\textit{CTA}^{+}}D_{\textit{CTA}^{+}} + N_{\textit{TFA}^{-}}D_{\textit{TFA}^{-}}} \end{equation} | (2) |
where Ni is the individual number density of the ions. Notably, $t_{\textit{Li}^{ + }}^{\textit{NMR}}$ is calculated based on the assumption that the monovalent ions are fully dissociated and move independently without interacting with other ionic species, as predicted by the Nernst-Einstein relationship for ideal electrolyte solutions. The actual Li-ion transference number can be influenced by ion-ion interactions and the relative motions of ions in concentrated electrolyte solutions.42 As shown in Fig. 3e, despite the abundant CTA+ ions in the solution, only a slight decrease in $t_{\textit{Li}^{ + }}^{\textit{NMR}}$ is observed because the diffusion of CTA+ is significantly restricted by the formation of micellar structures. For Zn-based electrolytes, $t_{\text{Zn}^{2 + }}^{\text{NMR}}$ cannot be estimated because our PFG-NMR facility does not support the measurement of the self-diffusion coefficient of Zn2+.43
Figures 4a and 4b show the ESW of the Li- and Zn-based electrolytes on the Al and GC electrodes for reductive and oxidative stability, respectively. The electrolyte with 1 M LiTFA-0.5 M CTATFA showed onset potentials of −1.57 and 1.33 V vs. Ag/AgCl for reductive and oxidative decompositions, respectively. The ESW of 1 M LiTFA electrolyte was expanded from 2.3 V to 2.9 V with the addition of 0.5 M CTATFA (Fig. 4a). For Zn-based electrolytes, the addition of CTATFA also improved the reductive and oxidative stability (Fig. 4b). In the electrolytes with 0.5 M CTATFA, the adsorption and accumulation of hydrophobic CTA+ at the interface is expected to reduce interfacial water.44,45 This could be the primary reason for the enhanced ESW in the presence of the surfactant. CTA+ cannot directly adsorb onto the positively charged electrode, but can occupy regions near the electrode surface where anions are adsorbed. This occupation reduces the interfacial water on the positively charged electrode and improves the oxidative stability. In addition, the specific adsorption and accumulation of CTATFA with the metal ions can create a local high-concentration salt region at the interface.10,46 This region is highly favorable for the formation of an anion-derived fluorine-containing SEI on a negatively charged electrode. The combined effects of reduced interfacial water and the formation of SEI significantly suppress the kinetics of the hydrogen evolution reaction (HER), thus enhancing the ESW of aqueous electrolytes.44,45,47
Linear sweep voltammograms on Al and GC electrodes in (a) 1 M LiTFA and 1 M LiTFA-0.5 M CTATFA and (b) 1 M Zn(TFA)2 and 1 M Zn(TFA)2-0.5 M CTATFA.
To investigate the effects of CTATFA on SEI formation, we performed potentiostatic polarization on an Al current collector and studied the morphology of the formed SEI.48,49 As shown in Figs. S6a–S6b, the pristine Al foil exhibits a striated morphology containing small pits and microcracks owing to the manufacturing process. After the polarization at −1.2 V vs. Ag/AgCl for 20 h, a highly dense and smooth film was formed on the surface of the Al electrode in 1 M LiTFA-0.5 M CTATFA (Fig. S6c). In a magnified view (Fig. S6d), decomposition products with winkled structure are observed in the SEI layer. In contrast, the Al electrode in 1 M LiTFA exhibited numerous irregular depressions (Figs. S6e–S6f). This indicates that a localized high-concentration salt region formed at the interface via the surface adsorption of CTATFA, resulting in a denser and smoother SEI to widen the ESW.
Zn-symmetric cells were assembled, and EIS was performed to evaluate the interfacial effects of CTATFA in the Zn systems. As shown in Figs. 5a–5c, highly depressed semicircles were observed in all Zn-symmetric cells due to the overlap of the resistance associated with Zn2+ crossing the SEI and charge transfer resistance.50,51 Compared to the surfactant-free 1 M Zn(TFA)2 and 1 M ZnSO4, the electrolyte containing 1 M Zn(TFA)2-0.5 M CTATFA exhibited a much lower interfacial resistance (i.e., smaller diameter of the semicircles) of 179 Ω in the first cycle. Furthermore, the interfacial resistance did not show a significant increase during the second and third cycles. However, the Zn symmetric cells containing 1 M Zn(TFA)2 and 1 M ZnSO4 exhibited significantly higher interfacial resistance exceeding 1000 Ω, and the increase in the resistance during cycling was greater compared to that observed in the cell with 1 M Zn(TFA)2-0.5 M CTATFA. These EIS data further demonstrated that the addition of CTATFA could effectively alleviate the adverse chemical side reactions at the interface of the Zn anode and 1 M Zn(TFA)2-0.5 M CTATFA, thereby enhancing the interface stability.52
Nyquist plot of Zn symmetric cells with the aqueous electrolytes of (a) 1 M Zn(TFA)2-0.5 M CTATFA, (b) 1 M Zn(TFA)2, and (c) 1 M ZnSO4.
The morphology of Zn deposits on the Cu electrode was further studied using SEM. As shown in Figs. 6a–6c, the Cu foil after the initial Zn plating at 1 mA cm−2 and 1 mAh cm−2 in 1 M Zn(TFA)2 revealed dendritic growth of Zn deposits. Additionally, as shown in the photograph of the Cu electrode (Fig. 6d), the deposition of Zn was not uniform. After 20 cycles of plating and stripping at 1 mA cm−2 and 1 mAh cm−2, the dendrites remained prominent, and the digital images of the Cu electrode showed that the Zn deposition became more non-uniform (Figs. S7a–S7d). Similarly, as shown in Figs. S7e–S7h and S8, 1 M ZnSO4 electrolyte failed to produce a smooth Zn film, and the non-uniform deposition resulted in needle-like dendrites. In the absence of hydrophobic surfactants, a localized high-concentration region is difficult to form, leading to irregular zinc metal deposition.53 Furthermore, Zn2+ mass transport cannot be regulated without the adsorbed CTA+ layer at the interface, resulting in uneven distribution of current density.54 As a result, Zn cannot deposit in an orderly and controlled manner at the interface, and the rapid and chaotic Zn deposition eventually forms dendritic structures.
(a–c) SEM images and (d) photograph of the Cu electrode after the initial plating at 1 mA cm−2 and 1 mAh cm−2 in 1 M Zn(TFA)2. (e–g) SEM images and (h) photograph of the Cu electrode after the initial plating at 1 mA cm−2 and 1 mAh cm−2 in 1 M Zn(TFA)2-0.5 M CTATFA.
In contrast, in 1 M Zn(TFA)2-0.5 M CTATFA, the morphology of Zn deposits exhibits an ordered step-like morphology with a layer-by-layer flat platelet structure (Figs. 6e–6h). The Zn deposits remained uniform without significant dendrite formation, even after repeated plating and stripping over 20 cycles (Figs. S7i–S7l). The surface-accumulated CTA+ layers significantly reduced the amount of interfacial water. The highly concentrated CTATFA at the interface also facilitated the formation of an anion-derived SEI, greatly suppressing the HER.55 Moreover, the ordered aggregation of CTA+ ions at the interface, along with the formation of a structured ionic transport channel,56,57 can facilitate Zn2+ mass transfer and ensure an even current density distribution.58 This, in turn, promotes uniform Zn deposition on the electrode.
XPS measurements were performed on the same Cu electrodes after the initial deposition of Zn at 1 mA cm−2 and 1 mAh cm−2 to further investigate the composition of the SEI films formed in 1 M Zn(TFA)2 and 1 M Zn(TFA)2-0.5 M CTATFA. The Zn 2p and F 1 s spectra are shown in Fig. 7. The Zn 2p spectra were deconvoluted into 4 peaks at binding energies of 1021.35, 1022.32, 1023.30, and 1024.25 eV corresponding to Zn0, ZnO, ZnF, and Zn(OH)2, respectively (Figs. 7a and 7e).59–62 The F 1s spectra were also deconvoluted into peaks at 690.4, 687.5, and 684.35 eV corresponding to CF2CF2, CFx/CF3, and ZnF2, respectively (Figs. 7b and 7f).63–65 The presence of fluorinated components in the XPS spectra suggests the use of Zn(TFA)2 can produce anion-derived SEI with or without the addition of CTATFA.
XPS spectra of first plating onto Cu foil at 1 mA cm−2 and 1 mAh cm−2 in 1 M Zn(TFA)2-0.5 M CTATFA and 1 M Zn(TFA)2: (a) and (e) Zn 2p, (b) and (f) F 1s, (c) and (g) O 1s, (d) and (h) C 1s.
Zn(OH)2 is usually regarded as a by-product of Zn corrosion during the deposition process.66,67 The Zn 2p spectra indicate that the content of Zn(OH)2 formed in 1 M Zn(TFA)2 was higher than that in 1 M Zn(TFA)2-0.5 M CTATFA. Also, the amount of Zn(OH)2 was significantly greater than the amount of ZnF2 in the 1 M Zn(TFA)2. In contrast, the content of ZnF2 exceeded that of Zn(OH)2 in 1 M Zn(TFA)2-0.5 M CTATFA (Figs. 7a and 7e). The presence of ZnO62 and Zn(OH)268 was also verified by the peaks at 533.84 eV and 531.5 eV in the O 1s spectra (Figs. 7c and 7g). These results support that the SEI formed in 1 M Zn(TFA)2-0.5 M CTATFA contains a lower amount of Zn(OH)2. Hence, with the addition of CTATFA, the reduced interfacial water was likely to suppress the side reactions between Zn and H2O and reduce the formation of Zn(OH)2.
Furthermore, the SEI formed in 1 M Zn(TFA)2-0.5 M CTATFA consists of mixed components with a balanced organic to inorganic ratio (CF3/CFx : ZnF2), which is highly beneficial for Zn2+ transport and subsequent Zn deposition (Fig. 7f). In contrast, as shown in Fig. 7b, the SEI formed in 1 M Zn(TFA)2 displayed a significantly higher content of perfluorinated carbons compared to inorganic components. The deconvoluted C 1 s spectrum showed signals corresponding to C-O and C-C bonds at 286.4 and 284.8 eV,69 respectively. ZnCO3 (288.3 eV) was also observed in the C 1s spectra of 1 M Zn(TFA)2 and 1 M Zn(TFA)2-0.5 M CTATFA (Figs. 7d and 7h), consistent with a previous report.70,71 The C 1s spectra confirmed the presence of CF3/CFx (291.9 eV) in the 1 M Zn(TFA)2.72 In Fig. 7h, the most prominent peak of C-C (284.8 eV) clearly indicates that CTA+ was largely involved in the SEI formed in 1 M Zn(TFA)2-0.5 M CTATFA. The above XPS data suggested that the addition of CTATFA facilitated the formation of a localized high-concentration salt region at the interface. The resulting SEI contained ZnF2 and ZnCO3. The surface-adsorbed CTA+ layer was also responsible for the SEI. This hierarchical structure of the SEI effectively suppressed unfavorable side reactions (HER and Zn corrosion) and allowed for more uniform Zn deposition in 1 M Zn(TFA)2-0.5 M CTATFA.
The aqueous battery performance of the Zn-based electrolytes was further examined. In Figs. 8a–8d, the Zn symmetric cell with 1 M Zn(TFA)2-0.5 M CTATFA electrolyte exhibited Zn plating/stripping reversibility for over 800 h at 1 mA cm−2 and 1 mAh cm−2, and the Zn-Cu asymmetric cell with 1 M Zn(TFA)2-0.5 M CTATFA electrolyte exhibited Zn plating/stripping reversibility for 600 h at 5 mA cm−2 and 5 mAh cm−2. The increase in the overvoltage was also not significant during 300 cycles (Fig. 8e). Meanwhile, Zn-symmetric cells and Zn-Cu asymmetric cells using surfactant-free electrolytes exhibited short-circuiting and cell failure within 100 h (Figs. 8a–8d). In Zn-Cu cells (Fig. 8f) with 1 M Zn(TFA)2-0.5 M CTATFA, a Coulombic efficiency of approximately 99.9 % was achieved at a high specific capacity of 5 mAh cm−2 over 300 cycles, whereas the Coulombic efficiency in 1 M ZnSO4 was relatively low in the early stages with only about 30 stable cycles. It is often difficult to achieve stable cycling at a high capacity for (002)-textured zinc owing to its large lattice distortion.73,74 The XRD patterns obtained for the Zn deposits after 50 plating/stripping cycles (Fig. S9) verified that the preferred crystallization along the (1 0 1) lattice plane was more pronounced in 1 M Zn(TFA)2-0.5 M CTATFA, while crystal growth along the unfavorable (0 0 2) lattice plane was dominant in 1 M Zn(TFA)2.
(a) and (b) Galvanostatic cycling stability of Zn symmetric cells at 1 mA cm−2, 1 mAh cm−2 and (c) and (d) galvanostatic cycling stability of Zn-Cu asymmetric cells at 5 mA cm−2, 5 mAh cm−2. (e) Voltage profiles of a Zn-Cu cell with 1 M Zn(TFA)2-0.5 M CTATFA, and (f) Coulombic efficiency of Zn-Cu cells with 1 M Zn(TFA)2-0.5 M CTATFA and 1 M ZnSO4 at 5 mA cm−2 and 5 mAh cm−2.
The improved reversibility of Zn plating/stripping in 1 M Zn(TFA)2-0.5 M CTATFA is primary due to the uniform and less dendritic Zn deposition. The reduced side reactions such as HER and Zn corrosion and the smooth Zn deposits in the presence of CTATFA can also lead to more uniform Zn stripping. The improvements in both plating/stripping processes are therefore considered to be important factors contributing to the overall improved performance of the Zn-Cu cells.
Zn/MnO2 cells were also tested using a Zn-based electrolyte with or without CTATFA. Because Mn salts are usually utilized to inhibit side reactions at the MnO2 cathode during cycling,75 Mn(TFA)2 or Mn(SO4)2 was added to the Zn-based electrolytes in this study. As shown in Figs. 9 and S10, compared to 1 M Zn(TFA)2-0.2 M Mn(TFA)2 and 1 M ZnSO4-0.2 M MnSO4, incorporating 0.5 M CTATFA into the Zn/MnO2 cell enhanced the cycling performance, with a reversible capacity of 195 mAh g−1 and a stable coulombic efficiency of 99 % at a current density of 0.2 A g−1 after 250 cycles, and a reversible capacity of 105 mAh g−1 at a current density of 0.5 A g−1 and a stable coulombic efficiency of 99 % after 750 cycles. In previous studies on Zn/MnO2 battery, the dissolution of Mn2+ ions into the electrolyte was found to be the predominant cause of capacity fading.76,77 The presence of CTATFA likely contributes to the improved reversibility of MnO2 cathode: the surface-accumulated CTATFA can reduce interfacial water and provide more hydrophobic environment near the cathode, suppressing the dissolution of Mn2+ ions. The improvement also arises from the outstanding Zn deposition-stripping reversibility and good wettability of 1 M Zn(TFA)2-0.5 M CTATFA-0.2 M Mn(TFA)2.
Long cycling performance of Zn/MnO2 cells with 1 M ZnSO4-0.2 M MnSO4, 1 M Zn(TFA)2-0.2 M Mn(TFA)2, and 1 M Zn(TFA)2-0.5 M CTATFA-0.2 M Mn(TFA)2 at (a) 0.2 A g−1 and (b) 0.5 A g−1.
Based on the above battery performance and surface analyses, the possible mechanisms underlying the efficient Zn plating/stripping are summarized in Scheme 1. Compared to the interface without cationic surfactants, CTA+ can adsorb and accumulate at the interface, creating localized high-concentration salt regions. This can reduce the interfacial water, thereby inhibiting the HER.56,57 Additionally, the presence of these localized concentrated ion regions facilitates the formation of an anion-derived inorganic-rich SEI, further suppressing water decomposition and promoting the uniform deposition of metallic Zn. Moreover, the ordered surfactant layer promoted Zn2+ mass transfer and regulated current density distribution at the interface. These combined effects improved the reversibility of Zn stripping/plating and the cycling stability of the Zn/MnO2 full cell.
Schematic of efficient Zn plating/stripping via hierarchical organic–inorganic hybrid SEI in the Zn-based electrolyte with the addition of CTATFA and possible mechanisms for the formation of the SEI at the interface.
Our study investigated the electrochemical stability, solution properties, SEI formation, and aqueous Zn battery performance of cationic surfactant-based electrolytes. Significantly, 1 M Zn(TFA)2-0.5 M CTATFA exhibited a low viscosity of 80.2 mPa s and a higher conductivity of 27.5 mS cm−1. A wide ESW was achieved using CTATFA and alkali metal salts. In contrast to surfactant-free electrolytes, the greater hydrophobicity of CTA+ allows the creation of localized high-concentration salt regions at the interface, promoting the formation of a robust SEI that enhances the reversibility of Zn stripping/plating. The addition of CTATFA also promotes Zn2+ mass transport to the electrode and facilitates Zn deposition on the (1 0 1) plane, which is conducive to achieving stable cycling at a high capacity. For 1 M Zn(TFA)2-0.5 M CTATFA-0.2 Mn(TFA)2, the cycling stability of Zn/MnO2 full cells is enhanced at a high capacity, achieving over 250 cycles at a current density of 0.2 A g−1 and over 750 cycles at a current density of 0.5 A g−1. We believe that cationic surfactant-based electrolytes show great potential for practical applications in aqueous zinc metal batteries with long cycling and excellent stability.
This study was supported in part by JSPS KAKENHI (Grant Nos. 24K21803 and 23K17370) of the Japan Society for the Promotion of Science (JSPS). G. Shen acknowledges the China Scholarship Council (CSC) for its financial support and Dr. Frederik Philippi for the many helpful demonstrations.
The data that support the findings of this study are openly available under the terms of the designated Creative Commons License in J-STAGE Data at https://doi.org/10.50892/data.electrochemistry.28490156.
Guohong Shen: Data curation (Lead), Formal analysis (Lead), Investigation (Lead), Writing – original draft (Lead)
Shinji Kondou: Investigation (Supporting), Methodology (Lead), Validation (Lead), Writing – review & editing (Supporting)
Hiroki Nakagaki: Investigation (Supporting), Methodology (Lead)
Gakuto Wada: Data curation (Lead), Formal analysis (Lead), Methodology (Lead)
Masayoshi Watanabe: Resources (Lead), Supervision (Supporting), Validation (Lead), Writing – review & editing (Supporting)
Kaoru Dokko: Resources (Lead), Supervision (Lead), Validation (Equal), Writing – original draft (Supporting), Writing – review & editing (Lead)
Kazuhide Ueno: Conceptualization (Lead), Funding acquisition (Lead), Resources (Lead), Supervision (Lead), Writing – original draft (Lead), Writing – review & editing (Lead)
There are no conflicts to declare.
Japan Society for the Promotion of Science: 24K21803
Japan Society for the Promotion of Science: 23K17370
G. Wada: ECSJ Student Member
S. Kondou, K. Dokko, and K. Ueno: ECSJ Active Members
M. Watanabe: ECSJ Fellow
