High-Performance LiF@C-Coated FeF3·0.33H2O Lithium-Ion Batteries with an Ionic Liquid Electrolyte.

A new lithium-ion battery cathode material of LiF@C-coated FeF3·0.33H2O of 20 nm primary particles and 200-500 nm secondary particles is synthesized. The redox reaction mechanisms of the new cathode material and the influence of different electrolytes on the electrochemical performance of LiF@C-coated FeF3·0.33H2O are investigated. We show that LiF@C-coated FeF3·0.33H2O using a LiFSI/Pyr1,3 FSI ionic liquid electrolyte exhibits high reversible capacities of 330.2 and 147.6 mAh g-1 at 200 and 3600 mA g-1, respectively, as well as maintains high capacity over cycling. Electrochemical characterization shows that the high performance is attributed to higher electronic conductivity of the coating, continuous compensation of the loss of LiF product through the coating, higher ionic conductivity of both the coating and the electrolyte, and higher stability of the electrolyte.

Introduction

In recent years, lithium-ion batteries (LIBs) have been developed due to their wide applications in public transportation, grid energy storage, portable equipment, and other fields.[1] One of the top priorities is to develop high-capacity LIB cathode materials to meet the rapidly increasing demands from these energy storage devices.[2] Unfortunately, current commercial intercalation-type cathode materials (e.g., LiCoO2, LiFePO4, LiNiMnCoO2, etc.) have a specific capacity of 130–142 mAh g–1, limiting the energy density of LIBs to 250–300 Wh kg–1.[3] Conversion-type cathode materials are not limited by the number of void positions in the host lattice. On the contrary, energy storage is realized through a heterogeneous conversion reaction (MF + xLi+ + xe– ↔ M + xLiF).[4,5] Iron fluorides as a conversion-type cathode material has recently attracted attention due to its low cost and higher theoretical capacities. In particular, FeF3·0.33H2O is an iron fluoride polymorph with a specific hexagonal tungsten bronze structure that shows a high theoretical capacity of 712 mAh g–1 (discharge below 1.5 V, referring to 3e– transfer), leading to a potentially high theoretical energy density of 1951 Wh kg–1.[6,7]

However, FeF3·0.33H2O has its shortcomings; the Fe–F bond has a large band gap of 2.5 eV,[8] resulting in low electronic and ionic conductivity, which leads to rapid capacity fading and low actual capacity.[9] There are two main strategies to overcome the above shortcomings. One is to improve the electronic conductivity of FeF3·0.33H2O by coating graphene,[10] carbon nanotubes,[11] porous carbon,[12] acetylene black,[13] etc., and the other is to reduce the particle size of FeF3·0.33H2O to <20 nm to achieve a complete conversion reaction.[14] Therefore, the development of a simple and economical synthesis method for coating the conductive carbon layer and effectively reducing the particle size of FeF3·0.33H2O has become one of the necessary conditions for the wide application of FeF3·0.33H2O. We have previously reported a one-pot hydrothermal method of synthesizing a hierarchical nano/micro-structured FeF3·0.33H2O with nanoscale primary particles (1–2 nm) self-assembled into secondary particles (400 nm–1 μm) that reduced the ion diffusion distance in the particles and increased electrode packing density.[15]

Although coating with carbon can effectively improve the electrical conductivity of FeF3·0.33H2O, it does not fundamentally solve the rapid capacity degradation when FeF3·0.33H2O is discharged below 1.5 V over cycling. Previous research has shown that the reason for the rapid capacity decay of FeF3·0.33H2O is due to the gradual irreversible loss of LiF, the discharge product of FeF3·0.33H2O.[16] In addition, previous studies have shown that the choice of electrolyte affects the electrochemical performance of FeF3·0.33H2O, and a N-methyl-N-propyl pyrrolidinium bis(fluorosulfonyl)imide (Pyr1,3 FSI) ionic liquid electrolyte can improve the electrochemical performance of FeF3·0.33H2O.[17,18] However, the underlying mechanism of the ionic liquid electrolyte improving the electrochemical performance of FeF3·0.33H2O is currently unclear. Therefore, it is important to understand the influence of different electrolytes on the electrochemical performance of FeF3·0.33H2O.

Here, a new cathode material of LiF@C-coated FeF3·0.33H2O is synthesized. The introduction of a LiF@C layer to coat the FeF3·0.33H2O nanoparticles has three functions: (a) to improve electronic conductivity of FeF3·0.33H2O,[19] (b) to control the particle size of FeF3·0.33H2O in order to increase ionic conductivity, and (c) to continuously compensate the loss of LiF in order to overcome the shortcomings of poor cycling performance.[20] In addition, we prepare LiF@C-coated FeF3·0.33H2O batteries with two electrolytes, a standard LiPF6-based electrolyte and a Pyr1,3 FSI (LiFSI)-based ionic liquid electrolyte, to study the influence of different electrolytes on the electrochemical performance of the FeF3·0.33H2O material. The research results show that the LiFSI electrolyte improves the rate capability and cycling performance of the battery due to a higher Li+ migration number and formation of a stable cathode electrolyte interface (CEI) film. This provides a new reference direction for future research on new cathode materials and the complementary electrolyte on the LIB electrochemical performance.

Experimental Section

Synthesis of Pure FeF3·0.33H2O and LiF@C-Coated FeF3·0.33H2O

Pure FeF3·0.33H2O Nanosphere

Typically, iron(III) nitrate nonahydrate (Fe(NO3)3·9H2O, Adamas, China) and hydrofluoric acid HF (50%, Adamas, China) were utilized as raw materials, and cetyltrimethylammonium bromide (CTAB Adamas, China) was selected as a dispersant. First, CTAB (1.5 g) was dissolved in ethanol (60 mL) and stirred for 10 min. Fe(NO3)3·9H2O (2 g) was added in the mixture and stirred for 10 min to form a reddish-brown emulsion. Hydrofluoric acid (1 mL) was then added slowly into the mixed solution and stirred for 5 min. Deionized water (1 mL) was added and the mixture was continuously stirred for 30 min. The resulting mixture was transferred into a 100 mL Teflon reactor and kept at 150 °C for 10 h. The product was washed using ethanol by three times and then dried at 80 °C for 5 h in a vacuum atmosphere to remove the excess water, HF, and CTAB. Finally, the collected FeF3·3H2O powder was calcined at 220 °C for 3 h in Ar to obtain a tawny powder FeF3·0.33H2O marked as pure FF.

LiF@C-Coated FeF3·0.33H2O Nanocomposites

In a process of fabricating LiF@C-coated FeF3·0.33H2O nanocomposites, lithium fluoride (LiF, Adamas, China, 0.5 g) and graphite (Adamas, China, 0.1 g) were ball milled in a zirconium oxide (ZrO2) jar at 500 rpm for 5 h. The volume of the jar was 50 mL, and 20 zirconia balls (Φ = 5 mm) were used. The obtained LiF@C powder and FeF3·3H2O were further mixed in a ZrO2 jar at 500 rpm for 5 h. The product was washed using ethanol by three times and then dried at 80 °C for 5 h in a vacuum atmosphere. Finally, the collected FeF3·3H2O powder was calcined at 220 °C for 3 h under argon to obtain a black powder LiF@C-coated FeF3·0.33H2O marked as FF/LiF@C.

Material Characterization

The crystal structure of the as-synthesized materials was analyzed using a Bruker AXS D8 ADVANCE X-ray diffraction (XRD) machine with Cu Kα radiation operating at 40 kV and 40 mA between 10 and 90° 2θ at a scan rate of 2° min–1 with a step size of 0.02°. The morphology and particle size of the materials were investigated by scanning electron microscopy (SEM, Zeiss Sigma 500) and transmission electron microscopy (TEM, FEI Tecnai G2 F20). X-ray photoeletron spectroscopy (XPS, Thermo ESCALAB 250XI) was performed to investigate the oxidation state of the pristine and cycled electrodes.

Electrochemical Characterization

The working electrodes were made by mixing 280 mg of active materials (pure FF or FF/LiF@C), 80 mg of conductive carbon, and 40 mg of polyvinylidene fluoride (PVDF) in 4 mL of N-methyl pyrrolidone with a weight ratio of 70:20:10 according to ref (21). The slurry was cast uniformly on an aluminum foil using a doctor blade coater to prepare the electrode film, which was dried in a vacuum oven at 150 °C for 2 h to remove the solvent. The obtained electrodes were punched into circular discs of 1.2 cm in diameter. The electrode mass loading was 1–2 mg cm–2. The electrochemical tests were carried out using CR2032 coin cells with lithium metal foil as the counter electrode. The cells were assembled in an Ar-filled glove box with a H2O level <0.01 ppm and O2 level <0.01 ppm. Two types of electrolytes were used: one was LiPF6 based (marked as LP30) prepared from 1 M LiPF6 (Sigma-Aldrich) in ethyl carbonate (EC)/dimethyl carbonate (DMC) (1:1 vol %), and the other was lithium bis(fluorosulfonyl)imide (LiFSI) based (marked as LiFSI) prepared from 1 M LiFSI and dried Pyr1,3 FSI in a glove box for 2 h. Galvanostatic charge–discharge cycles were performed with a charge–discharge voltage range of 1.3–4.3 V vs Li/Li+ on a LAND CT2001A battery tester at different current densities at room temperature. The electrochemical impedance spectroscopy (EIS) was carried out in a frequency range from 100 kHz to 10 mHz on a Gamry interface 5000E electrochemical workstation at a potential perturbation of 5 mV at open-circuit voltage. Cyclic voltammetry(CV) was performed using an electrochemical workstation (Gamry interface 5000E) at a scan rate of 0.1–2 mV s–1 and a potential range of 1.3–4.3 V vs Li/Li+. Two LP30 and LiFSI FF/LiF@C cells were discharged to 1.3 V and charged to 4.3 V after 30 charge/discharge cycles, dissembled to retrieve the cathodes, and dried in the glove box. The dried electrodes were used for the XPS characterization.

Results and Discussion

We used XRD to investigate the crystal structures of FF and FF/LiF@C. Figure shows that all the diffraction peaks of FF are in good agreement with FeF3·0.33H2O (JCPDS card no. 76-1276), whereas the XRD diffraction peaks of FF/LiF@C are weak diffraction peaks at 26.3° (corresponding to the (001) characteristic peak of graphite) and at 38.7° (corresponding to the (111) characteristic peak of LiF). The above results show that both LiF and graphite may be coated on the surface of FF to form FF/LiF@C. In addition, the XRD pattern of FF/LiF@C shows wider diffraction peaks than FF, indicating that FF/LiF@C has smaller particle sizes.[22]

Figure 1

XRD spectra of FeF3·0.33H2 (pure FF) and LiF@C-coated FeF3·0.33H2O (FF/LiF@C).

We used SEM to investigate the morphologies of pure FF and FF/LiF@C. Figure a shows the SEM image of pure FF, and Figure b shows the SEM image of FF/LiF@C, showing that the particle size of FF/LiF@C was smaller than that of pure FF, in agreement with the XRD results, demonstrating that high-energy ball milling effectively reduced the particle size and the LiF@C coating prevented growth of the particles during the subsequent calcination.[23] More SEM images of pure FF and FF/LiF@C are in Figure S1. Figure c,d shows the TEM images of FF/LiF@C, showing more clearly that FF/LiF@C consists of ca. 20 nm primary particles and 200–500 nm secondary particles. This particle structure can effectively shorten paths of Li+ diffusion and electron transport and increase the contact area between the electrolyte and the active material. Figure e shows a high-resolution TEM (HRTEM) image of the edge of a FF/LiF@C primary particle, showing the LiF@C coating on the surface of the FF particle, and the lattice fringes with an average distance of 0.321 nm, which matches well with the (220) crystal planes of FeF3·0.33H2O. More TEM information about FF/LiF@C is in Figure S2. Figure f shows the energy-dispersive spectroscopy (EDS) mapping of FF/LiF@C, showing that the Fe, F, and C elements were distributed among the particles.

Figure 2

SEM images of the two types of active materials in a powder form: (a) pure FF and (b) FF/LiF@C. (c, d) TEM images of FF/LiF@C, (e) HRTEM image of FF/LiF@C, and (f) SEM EDS mapping of FF/LiF@C.

Figure a compares the cyclic voltammograms of pure FF and FF/LiF@C, both in the LP30 electrolyte. Figure b compares the cyclic voltammograms of FF/LiF@C in the LP30 and LiFSI electrolytes, all in the voltage range of 1.3–4.3 V. The voltammograms all show two pairs of oxidation–reduction peaks, one reduction peak at ca. 2.9 V and the other reduction peak at ca. 1.7 V. The peak at 2.9 V indicates that Fe3+ from FeF3·0.33H2O is gradually reduced to Fe2+ during Li insertion through an intercalation mechanism (Li+ + e– + FeF3·0.33H2O → LiFeF3·0.33H2O). The other peak at ca. 1.7 V corresponds to Li insertion through a conversion reaction mechanism (2Li+ + 2e– + LiFeF3·0.33H2O → 3LiF + Fe·0.33H2O).[24]Figure a shows that the potential difference (ΔE) between the oxidation peak (3.4 V) and the reduction peak (3.0 V) relating to the one-electron reaction of FeF3·0.33H2O for FF/LiF@C was 0.4 V, smaller than 0.6 V for pure FF, indicating lower internal resistance of LiF@C-coated FeF3·0.33H2O.[25]Figure a also shows that the oxidation peak of pure FF relating to the one-electron reaction of FeF3·0.33H2O was near 3.5 V, much higher than that of FF/LiF@C, which was caused by the formation of the solid electrolyte interface (SEI) film on the pure FF particles during the initial cycle due to the lack of LiF@C coating protection,[16] which is also shown by the EIS results (below). Figure b shows that the ΔE between the oxidation peak (3.4 V) and the reduction peak (3.1 V) relating to the one-electron reaction of FeF3·0.33H2O for FF/LiF@C in the LiFSI electrolyte was 0.3 V, smaller than 0.4 V for the same electrode material in the LP30 electrolyte, demonstrating small polarization and good reversibility of FF/LiF@C in the LiFSI electrolyte,[13] showing that the LiFSI electrolyte helps to reduce ΔE between the oxidation peak and the reduction peak of FeF3·0.33H2O. The CV curves of LP30 FF/LiF@C and LiFSI FF/LiF@C at various scan rates from 0.1 to 2 mV s–1 are shown in Figure S3.

Figure 3

Cyclic voltammograms at a 0.1 mV s–1 scan rate of (a) LP30 pure FF and LP30 FF/LiF@C and (b) LP30 FF/LiF@C and LiFSI FF/LiF@C.

In order to further explore the redox reaction mechanism, XPS was used to determine the oxidation state of the Fe element charged to 4.3 V in the two different electrolytes. Figure a shows the Fe2p spectrum of the pristine electrode, showing a distinct peak at a binding energy of 713.9 eV for Fe2p3/2, corresponding to the bond energy of Fe(III)–F,[6] indicating that a pure phase FeF3·0.33H2O material was synthesized. As shown in Figure b,c, when the LP30 electrolyte battery and the LiFSI electrolyte battery were charged to 4.3 V, two peaks of Fe3+-multiplet (713.0–717.0 eV) and Fe2+-multiplet (710.0–712.0 eV) were detected at the same time, which indicates that the reversible redox pair of Fe3+/Fe2+ participates in the charge and discharge process.[26] In addition, when the LiFSI FF/LiF@C battery was charged to 4.3 V, a weaker Fe2+ peak (peak intensity ratio of Fe2+:Fe3+ = 1.00) than the Fe2+ peak in LP30 (peak intensity ratio Fe2+:Fe3+ = 1.03) was detected, which indicates that the LiFSI electrolyte can improve the reversibility of the redox reaction of FeF3·0.33H2O. In addition, two iron elemental peaks were found between 700 and 710.0 eV, which indicates that a small amount of the Fe elemental phase will be generated when the iron fluoride material is charged to 4.3 V. This will cause the charging reaction product to be two phases of Fe simple substance and LiF, which is not conducive to the reversibility of iron fluoride. Therefore, it is necessary to coat an extra electrochemically active LiF@C layer to compensate for the loss caused by LiF deactivation.

Figure 4

XPS spectra of Fe 2p3/2 for (a) pristine FeF3·0.33H2O, (b) a cycled LP30 FF/LiF@C electrode, and (c) cycled LiFSI FF/LiF@C electrode.

Figure shows the galvanostatic charge and discharge curves of pure FF in the LP30 electrolyte, FF/LiF@C in the LP30 electrolyte, and FF/LiF@C in the LiFSI electrolyte at different charge and discharge currents. The discharge curves exhibit two plateaus; the region of the curves between 2–4.3 V corresponds to the Li+ + e– + FeF3·0.33H2O → LiFeF3·0.33H2O intercalation reaction, and the region of the curves between 1.3–2 V corresponds to the 2Li+ + 2e– + LiFeF3·0.33H2O → 3LiF + Fe·0.33H2O conversion reaction,[24] corroborating the CV results. Figure d compares the reversible capacities among the pure FF electrode in the LP30 electrolyte, the FF/LiF@C electrode in the LP30 electrolyte, and the FF/LiF@C electrode in the LiFSI electrolyte at different charge and discharge current densities. Using the same LP30 electrolyte, the reversible capacities of FF/LiF@C were 285, 223, 169, and 103 mAh g–1 at 200, 400, 800, and 1600 mA g–1 (approximately corresponding to 1, 2, 4, and 8 C), higher than 274, 189, 127, and 73 mAh g–1 for pure FF at the same current densities. The columbic efficiency for FF/LiF@C at the first cycle was 99%, significantly higher than 56% for pure FF. Figure d shows that FF/LiF@C in the LiFSI electrolyte exhibited the highest reversible capacities of 300, 283, 235, and 187 mAh g–1 at the same current densities. Especially at the high charge and discharge current density of 3200 mA g–1 (approximately corresponding to 16 C), FF/LiF@C in the LiFSI electrolyte exhibited a reversible capacity of 136 mAh g–1, while the capacities of the other two batteries became negligible. Figure e shows that FF/LiF@C in the LiFSI electrolyte also exhibited the highest cycling stability at 200 mA g–1. After 50 cycles at 200 mA g–1, the reversible capacity of LiFSI FF/LiF@C was 231 mAh g–1 and that of LP30 FF/LiF@C was only 138 mAh g–1. The discharge capacities and capacity retention of FF electrode materials are summarized in Table . Since the LiFSI/Pyr1,3 FSI ionic liquid electrolyte exhibits higher ionic conductivity than LiPF6 in organic carbonate solvents, it can effectively improve the rate capability of the battery.[18] The higher cycling performance for LiFSI/Pyr1,3 FSI than LP30 can be attributable to more stable properties for LiFSI, including higher thermal stability and less sensitivity to moisture.[18] Therefore, the high capacity, rate capability, and cycling performance of FF/LiF@C in LiFSI are mainly due to the higher electronic conductivity of the LiF@C coating, the higher ionic conductivity of the FF/LiF@C nanoparticles and LiFSI/ Pyr1,3 FSI ionic liquid electrolyte, the higher stability of LiFSI, and the compensation of Li+ through the LiF@C coating during the charge and discharge process. It is worth noting that the first Coulombic efficiency of FF/LiF@C in LP30 was greater than the first Coulombic efficiency in LiFSI mainly because FF/LiF@C reacted with LiFSI to form the cathode electrolyte interphase (CEI) film,[27] and the CEI film effectively improved the cycle performance of FeF3·0.33H2O for the following cycles. The constant current charge and discharge curves of the FF/LiF@C electrode in the LiFSI-based electrolyte at different rates are shown in Figure S4.

Figure 5

Galvanostatic charge/discharge curves of the electrodes of (a) pure FF in the LiPF6-based electrolyte, (b) LP30 FF/LiF@C in the LiPF6-based electrolyte, and (c) FF/LiF@C in the LiFSI-based electrolyte; (d) rate capability of the three types of batteries; and (e) the cycle performance of the three types of batteries.

Table 1

Summary of the Discharge Capacity for FF/LiF@C in the LiFSI-Based Electrolyte after 50 Cycles and a Comparison with the Literature

reference	current density (mAh g–1)	cycles	charge and discharge voltage range (V)	discharge capacity (mAh g–1)
our work	200	50	1.3–4.3	231
([11])	20	40	1.5–4.5	210
([7])	50	40	1.0–4.0	161
([28])	20	50	1.8–5.5	150
([29])	200	30	1.4–4.5	145

Electrochemical impedance spectroscopy (EIS) was used to further understand the influence of the LiF@C coating and the different electrolytes on the electrochemical behavior of pure FF and FF/LiF@C after the first cycle, the fifth cycle, and the 30th cycle, as shown in Figure a–c. Figure e shows the equivalent circuit used to fit all the EIS data; RΩ represents the ohmic resistance of the electrolyte and the electrode, Rct is attributed to the SEI film resistance formed after charging and discharging and the charge transfer resistance at the interface of the active material, Rcom is the resistance produced by Li+ through the LiF@C coating, CPEsei, CPEcom, and CPEct are interface double-layer capacitors, and Wc is the Warburg impedance caused by the semi-uniform diffusion of Li+ ions in the electrode, corresponding to the line at low frequency.[30]Figure a shows a semicircle at high frequency due to the formation of the SEI film that was detected after the first discharge cycle, which indicates that the LiFSI electrolyte can more easily form a stable SEI film on the surface of the FF/LiF@C material, which explains the better cycling performance of the LiFSI FF/LiF@C battery. In addition, the Rtotal value (35.0 Ω) of the LiFSI FF/LiF@C battery after initial charge and discharge was significantly lower than that of LP30 FF/LiF@C (80.6 Ω) and LP30 pure FF (96.7 Ω), showing that the LiFSI FF/LiF@C battery had a low degree of polarization. The Rtotal values of the LiFSI FF/LiF@C battery after the 5th cycle and the 30th cycle were 38.5 and 50.9 Ω, which were also the lowest among the three electrodes. Figure c illustrates the linear relationship between Z′ and the reciprocal square root of the lower angular frequencies ω–1/2. For each battery, the Warburg factor σ was obtained from the slope of the graphs following the equation below:[31]where Z′ is the real part of impedance and ω is the angular frequency in the low-frequency region. The Li+ ion diffusion coefficients (DLi) of the LP30 pure FF, LP30 FF/LiF@C, and LiFSI FF/LiF@C electrodes were calculated using the following equation:[31]where R is the gas constant, T is the absolute temperature, A is the surface area of electrodes, n is the number of electrons involved in the redox reaction, F is the Faraday constant, C is the molar concentration of Li+, and σ is the Warburg factor that is related to Z′. Table summarizes the individual resistance and DLi for pure FF in LP30, FF/LiF@C in LP30, and FF/LiF@C in LiFSI after the 1st, 5th, and 30th cycles, respectively, showing that FF/LiF@C in LiFSI consistently exhibited the highest Li+ ion diffusion coefficient among the three types of batteries. This also explains the better rate capability of the LiFSI FF/LiF@C battery.

Figure 6

Nyquist plot of the three types of batteries under the conditions of (a) after the first discharge cycle, (b) after the fifth discharge cycle, and (c) after the 30th discharge cycle; (d) after the first discharge cycle, Zre is plotted against ω–1/2 and (e) the equivalent circuit used for fitting the experimental EIS data.

Table 2

Summary of the Individual Resistance Obtained from the Equivalent Circuit Model and the Calculated Li+ Ion Diffusion Coefficients (DLi) for the Three Types of Batteries after the 1st, 5th, and 30th Charge and Discharge Cycles

after the first discharge cycle	RΩ (Ω)	Rsei (Ω)	Rct (Ω)	Rcom (Ω)	Rtotal (Ω)	DLi (cm2 s–1)
LP30 pure FF	2.8	44.4	49.5		96.7	4.78 × 10–17
LP30 FF/LiF@C	6.3		45.0	35.3	80.6	9.41 × 10–16
LiFSI FF/LiF@C	5.4	5.6	11.8	12.2	35.0	1.36 × 10–15

Conclusions

A new LiF@C-coated FeF3·0.33H2O cathode material of 20 nm primary particles and 200–500 nm secondary particles was synthesized. Electrochemical characterization shows that FeF3·0.33H2O undergoes two types of redox reaction mechanisms during the charge and discharge process: one is through an intercalation reaction between 2 and 4.3 V and the other is through a conversion reaction between 1.3 and 2 V to achieve an overall high capacity of 274 mAh g–1 at 200 mA g–1 (∼1 C). The LiF@C coating and the LiFSI-based electrolyte further increased capacity to 300 mAh g–1 at the same current density, and substantially increased rate capability to achieve a capacity of 136 mAh g–1 at 3200 mA g–1 (∼16 C). The LiF@C-coated FeF3·0.33H2O in the LiFSI-based electrolyte also maintained higher capacity over cycling. These are attributed to higher electronic conductivity of the coating, continuous compensation of the loss of the LiF product through the coating, higher ionic conductivity of both the coating and the electrolyte, and higher stability of the electrolyte.