Electrochemical Properties of 1,1-Dialkyl-2,5-bis(trimethylsilylethynyl)siloles as Anode Active Material and Solid-state Electrolyte for Lithium-ion Batteries

Abstract

1,1-Dialkyl-2,5-bis(trimethylsilylethynyl)-3,4-diphenylsiloles (R=Et, i-Pr, n-Hex; 3a-c) were prepared and utilized as anode active materials for lithium-ion batteries; 3a was also used as a filler for the solid-state electrolytes (SSE). Siloles 3a-c were prepared by substitution reactions in which the two bromine groups of 1,1-dialkyl-2,5-dibromo-3,4-diphe- nylsiloles, used as precursors, were substituted with trimethylsilylacetylene in the presence of palladium chloride, copper iodide, and triphenylphosphine in diisopropylamine. Among siloles 3a-c, 3a had the best electrochemical properties as an anode material for lithium-ion batteries, including an initial capacity of 758 mAhg^-1 (0.1 A/g), which was reduced to 547 mAhg^-1 and then increased to 1,225 mAhg^-1 at 500 cycles. A 3a-composite polymer electrolyte (3a-CPE) was prepared using silole 3a as an additive at concentrations of 1, 2, 3, and 4 wt.%. The 2 wt.% 3a-CPE composite afforded an excellent ionic conductivity of 1.09 × 10^-3 Scm^-1 at 60℃, indicating that silole 3a has potential applicability as an anode active material for lithium-ion batteries, and can also be used as an additive for the SSE of lithium-ion batteries.

INTRODUCTION

Silole (i.e., silacyclopenta-2,4-diene) derivatives have received much attention in both academic and applied fields, attributed to the σ* – π* conjugation in the silole ring, as well as the facile synthesis of 2,5-difunctional siloles through intramolecular cyclization reactions of diethynylsilanes.^1,2 The σ* – π* conjugation in the silole ring originates from the interaction of the σ* orbital of silylene with the π* orbital of butadiene, leading to the lowest unoccupied molecular orbital (LUMO) energy level compared to that of other similar heterocycles.^2,3 Therefore, the bandgap energy between the highest occupied molecular orbital (HOMO) and LUMO, for example, 4.90 eV for 1-silacyclopenta-2,4-diene, is much lower than that of other metallole heterocycles having similar structures.⁴

Graphite has been widely used as an anode material for lithium-ion batteries (LIBs), but its storage capacity per unit area is relatively low (372 mAhg^–1).⁵ As another anode active material, silicon has a very high specific capacity of 3,579 mAhg^–1. When silicon is fully lithiated to Li₂₂Si₅, the capacity increases significantly to 4200 mAhg^{–1. 5} Lithium metal also has a high specific capacity; however, lithium metal forms dendrites during the charging and discharging processes, which deteriorates the battery stability. Several alloys have also been developed by reacting lithium with species such as Si or Sn, but these alloys have strong drawbacks such as a volume expansion of approximately 300%.⁵ Enhanced performance of LIBs by using crystal-aligned LiNi_0.6Mn_0.2Co_0.2O₂ was reported.⁶ A multi-walled carbon nanotubes@poly(p-phenylenediamine)-Prussian blue composite was studied for the electrochemical sensing of hydrogen peroxide.⁷

Silole-containing and silole-bridged co-polymers have been synthesized, and their properties, for example, UV-vis absorption, fluorescence, and conductivity were examined for application as nonlinear optical (NLO) materials.⁸ Donor-acceptor π-conjugated silole chromophores containing alkyne-protecting groups were synthesized, and their electronic properties were examined for NLO optoelectronic materials.^9,10 Chromophores of co-polymers containing silole derivatives exhibited aggregation-induced emission (AIE), and were applied to fluorescent sensors for application in biological as well as optoelectronic fields.^11,12 Fluorene-substituted 2,5-silolene co-polymers were prepared, and their aggregation-induced emission enhancement (AIEE) properties were also investigated for application in chemosensors for the detection of explosives.¹³

Recently, we presented the first report of the electrochemical characteristics of silolene-containing co-oligomers as anode materials for lithium-ion batteries. These co-oligomers included oligo[(2,5-silolene)-co-(silylene)]s¹⁴ and co-oligomers bearing 1,1-dialkyl-3,4-diphenyl-2,5-silolene and aromatic diolene.¹⁵ We also reported the electrochemical characteristics of monomeric materials such as 2,5-bis(trimetylsilyl)silole derivatives as anode materials for lithium-ion batteries.^16,17 Various researches have focused on the advances in solid-state electrolyte materials such as composite polymers and polymer-based electrolytes for preventing Li dendrite formation.^18-20 Very recently, the development of all-solid-state batteries (ASSB) was reviewed due to the potential of these batteries for providing higher energy densities, improved safety, and a longer cycle life.²¹

Herein, several silole derivatives, including 1,1-dialkyl-2,5-bis(trimethylsilylethynyl)-3,4-diphenylsilole (R = ethyl, isopropyl, n-hexyl; 3a‒c), are synthesized and their electrochemical characteristics are evaluated for use as anode materials. 1,1-Diethyl-2,5-bis(trimethylsilylethynyl)-3,4-diphenylsilole 3a is analyzed as an all-solid-state electrolyte (ASSE) for lithium-ion batteries.

EXPERIMENTAL

General

The dichlorodisubstituted silanes, trimethylsilylacetylene, PdCl₂(PPh₃)₂, acetonitrile (99.8%), and polyvinylidene fluoride (PVDF, M_w ~ 5.34 × 10³) used in the experiments were purchased from Sigma Aldrich Company, Korea. Polyethylene oxide (PEO; M_w = 10⁶) and lithium bis(fluorosulfonyl)imide (LiTFSI) were purchased from Alfa Aesar. Super P, lithium iron phosphate (LiFePO₄, LFP), and lithium foil were purchased from MTI Corp. (USA). All glassware were collectively assembled and then flame-dried using a burner and flushed with argon prior to the experiments. All syntheses were performed under argon atmosphere using the standard dual manifold vacuum-line Schlenk technique. All solvents were distilled before use according to a previously reported method.²² For example, N-methyl-2-pyrrolidone (NMP) was distilled over calcium hydride while flushing with argon, and tetrahydrofuran (THF) was distilled over chopped sodium metal with benzophenone ketyl while flushing with argon.²²¹H and ¹³C NMR spectra were obtained using a JEOL JNM-ECZ 500R FT-NMR spectrometer (Japan) with CDCl₃ solvent. The chemical shifts in all NMR experiments were measured using the residual proton peaks of CDCl₃ or tetramethylsilane (TMS) as an internal standard. Infrared spectroscopy experiments were performed using a Thermo Fisher Scientific FT-IR iD50 or 630 spectrophotometer (USA). Fourier-transform infrared (FT-IR) spectra of the half-coin cells were obtained using a Thermo Scientific/Nicolet Continuum spectrophotometer (USA) in the attenuated total reflection (ATR) mode. UV-vis absorption experiments were performed using a Hewlett Packard 8453 spectrophotometer (USA). The thermal stability of the prepared materials was evaluated by thermogravimetric analysis (TGA) using a TA Instrument thermogravimetric analyzer (TGA Q500) (USA). TGA data were acquired from 25 ℃ to 900 ℃ at a heating rate of 10℃/min under nitrogen at a flow rate of 20 mL/min. To investigate the electrochemical properties of 3a‒c, a potentiostat (ZIVE MP1, WBCS3000, WonATech, Korea) was used at 30 ℃ in the voltage range of 0.01−3 V (vs. Li/Li⁺). The charge/discharge rate and long-term cycling performances were evaluated using a TOSCAT-3100 analyzer (TOYO System Co., Japan). The rate performance was measured between 0.1 and 2 C, and the long-cycling performance was measured at 1 C (1,000 mAhg^–1). The structural properties were investigated by X-ray diffraction (XRD, Empyrean-Malvern) using Cu-Kα radiation. The microscopic structures were identified, and elemental mapping was performed using a field-emission scanning electron microscope (FE-SEM, Hitachi, S-4800). The ionic conductivity of 3a-CPE was determined based on the resistance measured by electrochemical impedance spectroscopy (EIS) in the frequency range of 50 mHz–2 MHz at an amplitude of 5 mV. The ionic conductivity, σ (S/cm), was calculated using the equation σ = L/ (R × S), where L is the thickness (cm) of the solid electrolyte, R is the resistance (Ω) obtained through EIS measurements, and S is the cross-sectional area (cm²) of the solid electrolyte.²³ The solid electrolyte was placed between stainless steel (SS) and the electrode and then installed in a measuring device and pressure was applied to the electrode. The measurements were performed at intervals of 10 ℃ in the range of 20−70 ℃. Linear sweep voltammetry (LSV) was used to confirm the electrochemical stability. Coin cells were assembled in the form of [SSï3a-CPEïLi] and measurements were performed at a scan rate of 5 mVs^–1 in the voltage range of 3.0‒7.0 V (vs. Li/Li⁺). Li-plating/stripping cycling tests were conducted to determine the interfacial compatibility between lithium and the solid electrolytes. The coin cells used in the test were assembled in the form of [Liï3a-CPEïLi] and the current density was sequentially increased to 50, 100, 200, and 300 mAcm^–2.

Synthesis

1,1-Diethyl-2,5-dibromo-3,4-diphenylsilole (1a), 1,1-diisopropyl-2,5-dibromo-3,4-diphenylsilole (1b), and 1,1-dihexyl-2,5-dibromo-3,4-diphenylsilole (1c) were prepared by intramolecular reductive cyclization reactions of diethylbis(phenylethynyl)silane, diisopropylbis(phenylethylnyl)silane, and dihexylbis(phenylethynyl)silane with lithium naphthalenide, respectively, according to the procedure reported previously.^17,24

1,1-Diethyl-2,5-bis(trimethylsilylethynyl)-3,4-diphenylsilole (3a). A three-neck round-bottom flask (250 mL) equipped with magnetic stirrer bar, reflux condenser, and heating mantle was flame-dried utilizing a Bunsen burner and flushed with argon gas, then loaded with 1a (2.00 g, 4.46 mmol), trimethylsilylacetylene (2.16 mL, 15.4 mmol), triphenylphosphine (0.035 g, 0.14 mmol), and diisopropylamine (50 mL, as a solvent) under argon atmosphere. The mixture was warmed to 40 ℃ using a heating mantle, after which palladium(II) chloride (0.0056 g, 0.045 mmol), copper(I) iodide (0.043 g, 0.223 mmol), and diisopropylamine (20 mL) were added. The reaction mixture was heated to 70 ℃ and refluxed overnight under vigorous stirring. The reaction mixture was allowed to cool to room temperature with stirring. The diisopropylamine hydrobromide salt was removed by filtration and the solvent was removed under reduced pressure using a rotary evaporator. The residue was dissolved in dichloromethane, and the solution was washed successively with 5% aqueous hydrochloric acid and deionized water. The organic solution was dried overnight over anhydrous magnesium sulfate, filtered, and the solvent was removed under reduced pressure using a rotary evaporator. The obtained product was dissolved in a small amount of methanol and recrystallized. Compound 2a, 1.60 g, 74.4%) was obtained as a yellowish solid. ¹H-NMR (500 MHz, CDCl₃): δ 0.10 (s, 18H), 1.03 (q, J = 8.2 Hz, 4H), 1.15 (t, J = 7.7 Hz, 6H), 7.09‒7.12 (m, 4H), 7.13‒7.16 (m, 6H) ppm. ¹³C-NMR (125 MHz, CDCl₃): δ −0.039, 2.437, 6.862, 105.297, 105.412, 122.661, 127.019, 127.307, 129.255, 137.299, 162.908 ppm. ²⁹Si NMR (99 MHz, CDCl₃, d): −18.69, 17.12 ppm. FT-IR (neat) \(\begin{align}\tilde{v}_{\max}\end{align}\): 3050 (ν_arC-H), 2955, 2922, 2852 (ν_alC-H), 2118 (ν_C≡C), 1026, 1008 (ν_arC=C), 919 (ν_C-C), 871, 837, 742, 697 (ν_Si-C) cm⁻¹. UV-vis absorbance (THF) λ_max: (ε, cm^–1 M^–1): 393 (2.519 × 10⁶ ) nm.

1,1-Diisopropyl-2,5-bis(trimethylsilylethynyl)-3,4-diphe-nylsilole (3b) was prepared as a yellowish solid in 70.0% yield (1.50 g) by the reaction of 1b (2.00 g, 4.20 mmol) with trimethylsilylacetylene (1.47 mL, 10.5 mmol), analogous to the preparation of 2a. ¹H-NMR (500 MHz, CDCl₃): δ 0.09 (s, 18H), 1.23 (d, J = 7.45 Hz, 12H), 1.42‒1.51 (m, 2H), 7.08 (br, s, 4H), 7.14 (br, s, 6H) ppm. ¹³C-NMR (125 MHz, CDCl₃): δ 0.02, 11.26, 17.73, 106.06, 122.00, 127.11, 129.35, 137.49, 163.36 ppm. ²⁹Si NMR (99 MHz, CDCl₃, d): −18.76, 17.03 ppm. FT-IR (neat) \(\begin{align}\tilde{v}_{\max}\end{align}\): 3056 (ν_arC-H), 2955, 2923, 2864 (ν_alC-H), 2113 (ν_C≡C), 1086, 1074 (ν_C=C), 1027, 994, 917 (ν_C-C), 868, 836, 757, 745, 694, 645 (ν_Si-C) cm^-1. UV-vis absorbance (THF) λ_max: (ε, cm^–1 M^–1): 393 (1.662 × 10⁴) nm.

1,1-Dihexyl-2,5-bis(trimethylsilylethynyl)-3,4-diphenylsilole (3c) was prepared as a yellowish solid in 80.0% yield (1.70 g) by the reaction of 1c (2.00 g, 3.57 mmol) with trimethylsilylacetylene (1.27 mL, 9.06 mmol), analogous to the preparation of 2a. ¹H-NMR (500 MHz, CDCl₃): δ 0.09 (s, 18H), 0.88 (t, J = 6.59 Hz, 6H), 0.97‒1.02 (m, 4H), 1.30 (dd, J = 6.87, 3.44 Hz, 8H), 1.35‒1.43(m, 4H), 1.49‒1.59 (m, 4H), 7.06‒7.09 (m, 4H), 7.11‒7.14 (m, 6H) ppm. ¹³C-NMR (125 MHz, CDCl₃): δ 0.06, 10.81, 14.25, 22.65, 23.27, 31.58, 32.97, 105.36, 105.66, 127.11, 123.53, 129.37, 137.51, 162.75 ppm. ²⁹Si NMR (99 MHz, CDCl₃, d): −18.73, 12.99 ppm. FT-IR (neat) \(\begin{align}\tilde{v}_{\max}\end{align}\): 3056 (ν_arC-H), 2955, 2919, 2850 (ν_alC-H), 2116 (ν_C≡C), 1119, 1091,1072 (ν_arC=C), 1026, 997, 962 (ν_C-C), 869, 839, 748, 722, 695(ν_Si-C) cm⁻¹. UV-vis absorbance (THF) λ_max: (ε, cm^–1 M^–1): 393 (1.702 × 10⁶) nm.

Fabrication of 3a‒c coin cells for analysis of anode properties

The active anode materials were prepared with a composition of 30 wt.% 3a‒c siloles, 60 wt.% super P, and 10 wt.% PVDF, respectively. The active materials were dissolved in NMP and mixed by high-energy ball milling. The prepared anode slurry was coated on copper foil using a doctor blade, then dried in a vacuum oven at 110 ℃ for 12 h. A half-cell was fabricated using a CR2031 type coin cell. Polypropylene (PP) was used as the separator, Li metal was used as the anode and reference electrode, and LiPF₆ 1 M solution in a mixture of ethylene carbonate (EC) and diethyl carbonate (DEC) (1:1 vol %) containing 10% fluoroethylene carbonate (FEC)) was used as the electrolyte.

Fabrication of silole 3a solid composite polymer electrolyte (3a-CPE) film

The silole 3a solid composite polymer electrolyte (3a-CPE) film was fabricated as follows: PEO, LiTFSI, and silole 3a were combined in acetonitrile (ACN). The proportions of the materials used for the fabrication of 3a-CPE and the organic solid polymer electrolyte (SPE) without 3a are provided in Table 1. The solution was mixed using a magnetic stirrer at 400 rpm for 12 h. This blended solution was further stirred at 2,000 rpm for 15 min using a high-speed mixer. The mixed solution was poured into a Teflon Petri dish inside a glove-box under Ar atmosphere and dried for 24 h. The film was dried at 60 °C for 12 h in a vacuum oven.

Table 1. Proportions of PEO, LiTFSI and silole 3a dispersed in ACN for the fabrication of 3a-CPE

Fabrication of all-solid-state battery half battery (ASSB)

A LFP electrode was used as the cathode; the slurry was mixed at a ratio of 80 wt.% LFP, 10 wt.% super P, and 10 wt.% PVDF. The viscosity was controlled with NMP solvent and mixed using a high-energy ball mill. The prepared slurry was loaded onto aluminum foil and spread thinly using a doctor blade. The prepared electrodes were then dried in a vacuum oven at 110 ℃ for 12 h. The half-cell was fabricated using a CR2031 type coin cell; Li metal was used as the anode and reference electrode. The coin cells for charge/discharge, LSV, and Li plating/stripping measurements were manufactured using the [LFP|3a-CPE|Li], [SS|3a-CPE|Li], and [Li|3a-CPE|Li] structures, respectively. The prepared coin cell was activated at 70 ℃ for 2 h to improve the interfacial contact between lithium and the solid electrolyte. The electrochemical measurements were conducted at 60 ℃.

RESULTS AND DISCUSSION

Synthesis

1,1-Dialkyl-2,5-dibromo-3,4-diphenylsiloles (R = Et, i-Pr, n-Hex, 1a‒c), as starting materials in Scheme 1, were synthesized through the intramolecular cyclization reactions of di(phenylethynyl)(dialkyl)silanes (R = Et, i-Pr, n-Hex) using lithium naphthalenide, anhydrous zinc chloride, and N-bromosuccinimide (NBS), respectively, according to the procedure reported previously.^17,24 As shown in Scheme 1, 1,1-dialkyl-2,5-bis(trimethylsilylethynyl)-3,4-diphenyl siloles (R = Et, i-Pr, n-Hex, 3a‒c) were prepared in reasonably high yields (70‒80%) by Sonogashira cross-coupling of 1a‒c with trimethylsilylacetylene in the presence of a palladium catalyst (Ph₃P, PdCl₂) and CuI in diisopropylamine solvent.²⁵ The structures of the prepared compounds 1a‒c and 3a‒c were confirmed by ¹H, ¹³C, and ²⁹Si NMR spectroscopy and several other spectroscopic methods. The selected spectroscopic parameters of 3a‒c are listed in Table 2.^26-29 The ¹H NMR peaks of 3a (Table 2) at 0.10 ppm are due to 18H of the two trimethylsilyl groups. The peaks at 1.03‒1.15 ppm are due to 10H of the two ethyl groups, and those at 7.09‒7.16 ppm are due to 10H of the two phenyl groups. The ¹³C NMR spectrum of 3a showed carbon peaks at a chemical shift of −0.039 ppm due to the trimethylsilyl group, and at 2.437 and 6.862 ppm due to the ethyl groups. The ²⁹Si NMR spectrum of 3a showed a silicon peak at −18.69 ppm due to the trimethylsilyl group and at 17.12 ppm due to the silole ring. Furthermore, the FT-IR spectrum of 3a showed a stretching vibration of the ethynyl group at 2118 cm^–1 (Table 2).^26,29 All the obtained spectral data indicate that the structures of dialkylbis(trimethylsilylethynyl)siloles 3a‒c were consistent with the structures from the reaction shown in Scheme 1.

Table 2. Selected spectroscopic characteristics of 3a‒c

^aIn CDCl₃. ^bIn neat. ν_C≡C in the ethynyl group. ^cIn THF. ^d% weight at 200 ℃ under nitrogen atmosphere.

Scheme 1. Synthesis of 3a‒c.

The UV-vis absorption spectral data of 3a‒c in THF are summarized in Table 2.²⁸ The absorption maximum, λ_abs,max, of 3a‒c was 393 nm, with molar extinction coefficients of 1.662 × 10⁴ to 2.519 × 10⁶ cm⁻¹ M⁻¹, plausibly attributable to the silolene-ethynylene chromophores originating from siloles 3a‒c.

The thermal stability of 3a‒c was analyzed by TGA under nitrogen atmosphere. All the silole derivatives 3a‒c were stable up to 100 ℃ with less than 2% weight loss relative to the initial weight. Compounds 3a‒c rapidly lost weight above 200 ℃. For instance, 70‒90% of the initial weight of siloles 3a‒c was lost between 200 and 400 ℃. Notably, the percentage weight remaining at 200 ℃ under nitrogen was 90, 96, and 95% for siloles 3a‒c, respectively, as summarized in Table 2. 3a‒c lost 1‒16% of the initial weight at 600‒800 ℃. When compounds 3a‒c were heated to 900 ℃ under nitrogen, 91‒99% of the initial weight was lost and the char yields were only 1‒9%.

Cyclic voltammetry properties

Cyclic voltammograms were acquired using siloles 3a‒c as anodes in half cells at a scan rate of 10 mVs⁻¹ in the voltage range of 0.01‒3.00 V (vs. Li⁺/Li). All three silole derivatives 3a‒c exhibited similar cyclic voltammetry (CV) properties. As shown in Fig. 1, 3a,b exhibited similar cyclic voltammetry characteristics. The first charge/discharge cycles of 3a,b both involved an irreversible reaction due to the formation of a solid electrolyte interface (SEI) between the electrode material and electrolyte. Lithiation of the electrode material was observed near 0.01 V. The two anodic peaks at 0.3 V and 1.0 V are attributed to the delithiation of Li from the active material (siloles). The shapes of the cyclic voltammograms of 3a and 3b (Fig. 1(a, b)) after the 2nd cycle indicate that these were reversible processes.^14-17,30-32

Figure 1. Cyclic voltammograms of coin cells fabricated using (a) 3a and (b) 3b as active materials.

Anode properties of siloles 3a‒c

To confirm the capacity and cycle performance of siloles 3a‒c, measurements were carried out at current densities of 0.1‒2.0 A/g (Fig. 2). The 3a electrode afforded the highest capacity, with an initial capacity of 1,220 mAhg⁻¹ (0.1 A/g), which then decreased to 682 mAhg⁻¹, indicating that the initial irreversible reaction was large. The capacities were 613 (0.2 A/g), 568 (0.5 A/g), 530 (1 A/g), 497 (2 A/g) and 601 mAhg⁻¹ (0.2 A/g) depending on the current density. The rate characteristics of siloles 3a–c are listed in Table 3.

Figure 2. C-Rate performance of siloles 3a−c as anode materials.

Table 3. Specific capacities of anodes employing 3a‒c siloles at various current densities

To confirm the long-term lifetime performance and capacity of the silole 3a‒c anodes, measurements were performed over 500 charging and discharging cycles at a current density of 1 A/g, as shown in Fig. 3. The initial capacity of the silole 3a electrode was 758 mAhg^–1 and the initial coulomb efficiency (CE) was 53.0%. The capacity was initially reduced to 518 mAhg^–1, then increased as the cycle number increased, and finally reached 1,225 mAhg^–1 at 500 cycles. An increase in capacity was also observed for the other silole 3b and c electrodes, attributed to the increase in the delithiation reaction between lithium and the silole compounds as the cycles progressed, as confirmed by the CV data in Fig. 1. Galvanostatic charge and discharge curves showing the voltage according to the capacity per cycle are shown in Fig. 4. The initial coulombic efficiency, capacity per cycle, and capacity retention rate for materials 3a–c are listed in Table 4.

Figure 3. Long-cycle performance of coin cells fabricated using 3a‒c as anode materials.

Figure 4. Galvanostatic charge/discharge curves when (a) 3a, (b) 3b, and (c) 3c were used as anode materials.

Table 4. Specific capacities at various cycles, initial coulombic efficiencies (CE), and retention rates of silole 3a–c anodes

Properties of All-Solid-State Battery (ASSB)

Analysis of the siloles as anode materials confirmed that silole 3a had the best interaction with lithium. Based on these experimental results, the structural characteristics, ionic conductivity, and electrochemical stability of a silole 3a-composite polymer electrolyte (3a-CPE) were investigated using silole 3a as an additive in PEO-based solid electrolytes.

Structural Characteristics of 3a-CPE

Optical images of the 3a-CPE films are shown in Fig. 5. The surface of the film with 1% 3a content was not completely covered with the crystalline structure, as shown in Fig. 5(a), whereas the surfaces of the films with 3% and 4% 3a were covered with aggregated crystals, as shown in Fig. 5(c) and (d), respectively. The most uniform crystalline structure was achieved with 2% 3a, vide infra. Siloles emit light strongly in the solid phase via aggregation-induced emission, AIE.³³ Depending on the substituent of the silole, unique agglomeration patterns were formed.³⁴ The silole with the triphenylsilyethynyl group had higher thermal stability and lower mobility, and underwent molecular self-assembly.^33,34 The 3a molecules spontaneously assembled into uniformly organized structures by simple evaporation of the solvent.

Figure 5. Optical images of 3a-CPE films.

More details of the structure and elemental distribution were confirmed by FE-SEM and EDS elemental mapping, as shown in Fig. 6. When the 3a content was 1% (Fig. 6(a)), the CPE surface was not uniformly covered with 3a crystals (Fig. 6(a)). With 2% 3a, a uniform crystal structure was observed (Fig. 6(b)). When the 3a content was greater than 3%, the area covering the CPE surface decreased as the crystal structure became larger, as shown in Fig. 6(c‒d). EDS elemental mapping of the formed 3a crystals is shown in Fig. 6(e‒j). The crystals formed with silole 3a mainly comprised silicon and carbon atoms.

Figure 6. SEM images (a–e) of 3a-CPEs and EDS element mapping (f–j) of 3a crystals on the film surface.

XRD also confirmed that the crystalline phase of 3a-CPE depended on the 3a content, as shown in Fig. 7(a). The XRD pattern of the CPE without 3a (Fig. 7(a), SPE) showed peaks of the crystalline phase only at 2θ = 19.15° and 23.26°, which were mainly derived from PEO. In comparison, peaks of the crystalline phase were also observed at 2 = 15.08° and 17.08° (Fig. 7(a)) for the sample with 2% 3a; these peaks were clearly attributed to 3a. The most uniform crystal structure was formed with 2% 3a, as confirmed by the sharpest XRD peak, which is consistent with the uniformity of the crystals confirmed by the previous optical image and FE-SEM image.¹⁶ The FT-IR spectra of 3a-CPE with various 3a contents are presented in Fig. 7(b). Asymmetric stretching vibrations of the SO₂ and CF₃ groups of LiTFSI salt were observed as signals at 1340 and 1185 cm⁻¹, respectively. The signals of the CH₂ scissoring, CH₂ twist, and C–O–C stretching vibrations of PEO were observed at 1466, 1280, and 1095 cm⁻¹, respectively.¹⁹

Figure 7. (a) XRD patterns and (d) FT-IR spectra of 3a-CPEs according to the 3a content.

Electrochemical Properties of 3a-CPE

Electrochemical impedance spectroscopy (EIS) was used to determine the lithium-ion conductivity of 3a-CPF. Fig. 8(a‒d) shows the EIS data of the 3a-CPE samples with different 3a contents at various temperatures in the range of 20‒70 ℃. The ion conductivity (σ) was calculated from the charge transfer resistance (R_ct) determined through EIS analysis; the calculated values are summarized in Table 5.¹⁹ The variation of s depending on temperature is shown in Fig. 8(e). The values of 6.18 × 10⁻⁵ Scm^–1 at 30 ℃ and 1.26 × 10^–3 Scm^–1 at 70 ℃ were obtained with 2% 3a-CPE.

Figure 8. (a‒d) EIS curves for 3a-CPEs within temperature range of 20‒70 ℃ and (e) temperature-dependence of the ionic conductivity for 3a-CPE with 1, 2, 3, and 4% 3a.

Table 5. Ionic conductivities within temperature range of 20‒70 ℃

Linear sweep voltammetry (LSV) was used to evaluate the high-voltage stability of 3a-CPE depending on the 3a content (Fig. 9). The current increased both at 4.5 V and 5.5 V, attributed to decomposition of the TFSI-anions and PEO, respectively.³⁵ However, in the case of 2% 3a-CPE, the decomposition of PEO stopped at 5.8 V, which may be attributed to the improved electrochemical stability due to the uniformly formed 3a crystals on the CPE surface. The degradation of PEO restarted at 6.5 V, demonstrating very good high-voltage stability.

Figure 9. Linear sweep voltammogram using a [SS|ETS-CPE|Li] cell structure in the overpotential range of 3.0 to 7.0 V (vs. Li⁺/Li@60 ℃).

When lithium metal is used as the anode, a short-circuit can possibly occur owing to the contact between the anode and cathode due to dendrite growth, which is a serious problem. Therefore, a lithium stripping plating test was conducted to confirm the effectiveness of silole 3a for inhibiting the growth of lithium dendrites. The coin cell used in the test was fabricated in the form of [Li|3a-CPE|Li] and measurements were made at current densities of 50, 100, 200 and 300 mAcm^–2 at 60 ℃ (Fig. 10). When the electrodes were brought into contact with each other owing to the lithium dendrites, the voltage decreased rapidly and converged to zero.³⁶ The most stable cycle results were achieved with 2% 3a-CPE (Fig. 10(b)) and the overvoltage increase was also the smallest as the current density increased. This seemed to have the effect of forming a stable protective layer between lithium and the CPE and inhibiting the growth of lithium dendrites by uniformly distributing the 3a crystals. The overvoltage increased when 3a-CPE with more than 3% 3a was used owing to the increase in the interfacial resistance between lithium and the CPE due to the non-uniform deposition of the 3a crystals.

Figure 10. Galvanostatic lithium plating and stripping analysis of the [Li|3a-CPE|Li] cell structure at 60 ℃. Cycling performance of 3a-CPEs at current densities of 50, 100, 200, and 300 μAcm^–2.

To confirm the capacity and cycle performance of the 3a-CPE, measurements were carried out at current densities of 0.1‒2.0 C using a coin cell with an LFP anode, as shown in Fig. 11(a). The highest capacities of 140.8, 146.1, 132.4, 70.8, 37.2, and 144.7 mAhg^–1 at 0.1, 0.2, 0.5, 1, 2, and 0.2 C were achieved with 2% 3a-CPE. The effect of the 3a content of 3a-CPE on the rate characteristics is shown in Tables 6 and 7. The SPEs without 3a showed a rapid capacity decrease after 35 cycles, which converged to zero capacity after 75 cycles; all coin cells with the 3a-CPEs were successfully run for up to 100 cycles. The cell with 2% 3a-CPE afforded the highest capacity of 154.3 mAhg^–1, with an excellent capacity retention rate of 95% after 100 cycles (Table 7).

Figure 11. (a) Cycling performance for evaluating rate capability at C-rates of 0.1‒2.0 C at 60 ℃. (b) Long cycling performance for 100 cycles at 0.1 C at 60 ℃.

Table 6. Specific capacity of 3a-CPEs at various current densities

Table 7. Specific capacity of 3a-CPEs at various cycles and retention rates

CONCLUSION

Departing from the conventional use of polymeric materials as solid electrolytes for all-solid-solid batteries, monomeric materials, namely, siloles 3a–c, were evaluated, and their potential for application as additives to anode active materials for lithium-ion batteries and solid electrolytes for all-solid-state batteries was investigated. Siloles 3a–c were synthesized and fully characterized by ¹H, ¹³C, ²⁹Si NMR, and IR spectroscopy. Characteristic peaks were observed in the NMR spectra. The infrared absorption spectra also confirmed the synthesis of compounds 3a–c based on the characteristic peaks. Cyclic voltammetry (CV) analysis of siloles 3a–c as anode materials showed oxidation-reduction reactions between lithium and the siloles. Silole 3a afforded the best delithiation reaction and highest capacity at a current density of 1 A/g; the initial capacity was 758 mAhg^–1, and the initial coulombic efficiency (CE) was 53.0%. After an initial capacity reduction to 518 mAhg^–1, the capacity gradually increased to 1,225 mAhg^–1 at 500 cycles, because the delithiation reaction increased as the cycles progressed. The structural properties, ionic conductivity, and electrochemical stability of the PEO-based CPEs were investigated using silole 3a, which exhibited the best interaction with lithium. It was confirmed that 2% 3a-CPE formed the most uniform 3a crystal structure. The ionic conductivity (s) calculated by EIS measurements was highest for 2% 3a-CPE, with a value of 1.09 × 10^–3 Scm^–1 at 60 ℃. The LSV and lithium stripping plating tests indicated that 2% 3a-CPE exhibited the best electrochemical stability and inhibition of lithium dendrite growth. Charge and discharge tests demonstrated that 2% 3a-CPE showed the best capacity of 154.3 mAhg^–1 at a current density of 0.1 C and an excellent capacity retention rate of 95% after 100 cycles. This study confirms the feasibility of applying silole 3a compounds in secondary lithium batteries such as lithium-ion batteries and all-solid-state batteries.