Symmetric Vs Asymmetric Imide Anion Decomposition Pathways And Their Influence On Solid Electrolyte Interphase Stability For Si Anodes

Abstract

A sturdy solid electrolyte interphase (SEI) is imperative for extending the calendar‐life of Si anodes in lithium‐ion batteries (LIBs). However, carbonate electrolytes form unstable interphases, hindering their practical implementation. Alternatively, fluorinated sulfonylimide (FSI⁻/TFSI⁻) based ionic liquid (IL) electrolytes, coupled with Li‐imide salts, enable anion‐derived SEI formation, thereby enhancing capacity retention. However, the functional role of these anions in directing SEI formation and evolution within IL‐based systems is poorly understood. Moreover, the mechanistic interplay between the decomposition pathways of symmetric and asymmetric anions in governing interfacial chemistry remains elusive. Herein, we investigate the chemistry and morphology of SEIs formed using various imidazolium‐based ILs containing both symmetric and asymmetric anions. We reveal that the synergistic interactions between symmetrical bis(fluorinated sulfonyl)imide anions and imidazolium cations facilitate an inorganic‐rich (LiF/LiOH) inner and a Li₂SO₄/polymeric outer layer SEI, conformally coating the 3D Si interface. The complementary effects of inorganic‐rich and polymer components, featuring key Li₂SO₄ species, reinforce mechanical integrity and flexibility, suppressing pulverization and enabling reversible capacity of 2489 mAh/g at 1C over 250 cycles. Correlating electrochemical performance with surface analysis provides critical insights into the impact of fluorinated sulfonylimide on passivation behavior and battery performance, guiding future design of ionic liquid electrolytes for LIBs.

Symmetric bis(flourosulfonyl)imide anions in imidazolium‐based ionic liquids direct the formation of a conformal SEI on 3D a‐Si architecture, featuring an ionorganic‐rich (LiF/LiOH) inner and Li₂SO₄/polymeric outer layer. This anion‐derived, mechanically robust SEI suppresses pulverization and delivers a high reversible capacity of 2489 mAh g⁻¹ at 1C over 250 cycles.

Introduction

Next‐generation lithium‐ion batteries (LIBs) employing Si anodes paired with high‐voltage cathodes and electrolytes have the potential to achieve high energy densities, exceeding 400 Wh kg⁻¹, making them a promising sustainable candidate for long‐range electric vehicle applications.^{[ 1} , ² , ³ , ⁴ , ⁵ , ^{6 ]} However, the repeated volume change of the alloying anodes during prolonged cycling gives rise to two critical challenges.^{[ 7} , ⁸ , ^{9 ]} First, the enormous mechanical stress causes Si particles to pulverize and delaminate from the current collector (CC).^{[ 10} , ¹¹ , ¹² , ¹³ , ^{14 ]} The second major challenge arises from the unstable electrode/electrolyte interface.^{[ 15 ]} The fracture of the initially formed passivation layer repeatedly exposes the Si surface to the electrolyte, causing continuous SEI reformation, electrolyte depletion, and accelerates degradation of the anode architecture.^{[ 16} , ^{17 ]} Together, these issues lead to severe capacity fading, low coulombic efficiency (CE), increased internal resistance, and polarization, ultimately limiting the practical lifespan of Si anodes.^{[ 3} , ¹⁸ , ^{19 ]} Numerous strategies, such as surface coating and void space engineering, have demonstrated significant advancements in mitigating the structural deterioration of micron‐scale Si anodes.^{[ 18} , ²⁰ , ²¹ , ²² , ^{23 ]}

To enable durable operation of such high‐porosity anode architecture, parasitic interfacial reactions at the electrochemical boundary must be suppressed through the formation of a mechanically and electrochemically robust SEI.^{[ 18} , ²⁴ , ^{25 ]} The physicochemical characteristics of the SEI can be precisely tuned through strategic electrolyte engineering, enabling selective anion decomposition.^{[ 25} , ²⁶ , ²⁷ , ²⁸ , ^{29 ]} Since the solvation chemistry of the electrolyte strongly influences the SEI composition, implementing the correct anion will promote the formation of an inorganic‐rich SEI layer surrounded by stable organic compounds, thereby preventing additional electrolyte decomposition.^{[ 30} , ³¹ , ^{32 ]} For example, tailoring the standard carbonate electrolyte with a fluoroethylene carbonate (FEC) additive has been demonstrated to improve the interface stability by forming a surface film rich in LiF content, alongside Li _x SiO _y , Li₂CO₃ compounds, and insoluble organic poly VC species.^{[ 25} , ³³ , ³⁴ , ³⁵ , ³⁶ , ³⁷ , ³⁸ , ^{39 ]} Despite these advantages, conventional carbonate‐derived passivation layers are not stable enough to mitigate capacity loss over the long‐term cycling of Si anodes, and the electrolytes exhibit poor compatibility with high‐voltage cathodes.^{[ 40 ]}

Alternatively, ionic liquids (ILs) with imidazolium, pyrrolidinium, or pyridinium cations, coupled with fluorinated sulfonyl imide (TFSI⁻/FSI⁻) anions, are beneficial for forming stable, inorganic‐rich, anion‐derived SEIs with tunable compositions compared to carbonate electrolytes.^{[ 27} , ⁴⁰ , ⁴¹ , ⁴² , ⁴³ , ⁴⁴ , ⁴⁵ , ⁴⁶ , ⁴⁷ , ^{48 ]}In particular, imidazolium cation‐based ionic liquids (ILs), such as 1‐ethyl‐3‐methylimidazolium (EMIM⁺), are cost‐effective and exhibit a wide potential window stability, low viscosity, and high ionic conductivity, making them of growing interest.^{[ 49} , ⁵⁰ , ⁵¹ , ⁵² , ⁵³ , ^{54 ]} Shobukawa et al. demonstrated that an imidazolium IL electrolyte formulation exhibited superior compatibility with a Si anode compared to carbonate electrolytes.^{[ 55 ]} The critical challenge is that EMIM⁺ cations possess an acidic C‐2 proton, which decomposes rapidly below 1.0 V versus Li+/Li, leading to substantial electrolyte consumption, significantly greater than observed for other cations.^{[ 56} , ⁵⁷ , ^{58 ]} The electrochemical stability of the EMIM⁺ cation depends on the anions of the solvent and the salt employed. For example, Ishikawa et al. showed that when used in tandem with TFSI⁻ anion, the EMIM⁺ cation reduces and intercalates into the graphite anode, forming an unfavorable passivation layer that inhibits reversible Li⁺ intercalation/de‐intercalation.^{[ 59 ]} Although fluorinated sulfonylimide anions have emerged as key components in stabilizing SEI within ionic liquid electrolytes, a detailed investigation into the underlying functional role of these anions on passivation behavior will be instrumental in controlling interface reactions.^{[ 60 ]}

Investigations of IL systems with other cationic frameworks have focused on understanding SEI formation behavior. Piper et al. investigated pyrrolidinium‐based IL containing different symmetric imide anions (FSI⁻ and TFSI⁻) electrolyte formulations and showed that a symmetric FSI⁻ anion formulation promotes the formation of FSI⁻ anion‐derived sulfur species, stabilizing the interface.^{[ 27 ]} Recent work by Jafta et al. revealed that a symmetric FSI⁻ anion electrolyte decomposes at a higher potential, forming an inorganic‐rich LIF and other Li salts initial layer, which prevents EMIM⁺ co‐intercalation into the carbon electrode compared to a symmetric TFSI⁻ anion electrolyte.^{[ 57 ]} Furthermore, several critical studies have focused specifically on the kinetics of imide salt anion decomposition, especially for Li metal anodes. These studies have identified FSI⁻ anion‐derived passivation layers, which are critically influenced by the different cleavage pathways of S‐functional groups within the electrolyte structure.^{[ 61} , ^{62 ]}

While several studies have established the compatibility of imidazolium‐based ILs with Si anodes^{[ 40} , ⁶³ , ⁶⁴ , ^{65 ]} an in‐depth understanding of how molecular interactions of the coordinating anions with imidazolium cation influence the formation and evolution of complex anion‐derived interfacial chemistry remains poorly understood. In particular, a systematic comparison between symmetric and asymmetric imide anions (salt and solvent anion), as well as distinct decomposition pathways, especially the S‐functional cleavage routes, that affect the SEI composition, is still critically lacking. The mechanistic link between the electrolyte interfacial reactivity and the effect of the anion breakdown process on the resulting SEI's chemistry and morphology will be crucial to revealing the underlying cause of the improved cycling stability of Si anodes. Furthermore, most existing SEI studies focus on Si thin films or planar binder‐based anodes, while insights into the SEI morphology formed on 3D structured electrodes still need to be explored.

Here, we examine the SEI formation mechanism and anode structural evolutions for 3D a‐Si anodes within symmetric and asymmetric anion IL electrolytes. Three IL electrolyte formulations were explored: 0.2 M LiFSI in 0.8 M EMIFSI (IE1), 0.2 M LiTFSI in 0.8 M EMIFSI (IE2), and 0.2 M LiTFSI in 0.8 M EMITFSI (IE3), combining alkali metal salts (LiFSI or LiTFSI) with imidazolium‐based ILs containing FSI⁻ and TFSI⁻ anions. These electrolytes were also benchmarked against carbonate electrolytes with FEC additive (STD + FEC). X‐ray photoelectron spectroscopy (XPS) depth profiling revealed critical differences in the formation and evolution of anion‐derived SEI components and relative content as a function of the IL formulation. Crucially, the choice of anion plays a dominant role in dictating interfacial chemistry and stability and, consequently, battery performance. The reduction of the LiFSI–EMIFSI electrolyte (IE1) led to the formation of a SEI comprising Li₂SO₄ and polyether species, enabling chemical stability and flexibility, effectively suppressing further electrolyte decomposition and significantly extending the cycle life of the a‐Si anode. Complementary focused ion beam‐scanning electron microscopy (FIB‐SEM) analysis further confirmed that EMIFSI‐based ILs produced conformal and robust SEI layers across the 3D Si surface. These findings offer fundamental insights into anion‐mediated interphase chemistry and its impact on battery performance, providing guidance for design principles of next‐generation ionic liquid electrolytes.

Results and Discussion

Electrolyte Design and a‐Si Anode Architecture

To elucidate the decomposition mechanism of different ionic electrolyte formulations that stabilize the electrode–electrolyte interface and improve electrochemical performance, we examined the chemistry and morphology of the passivating layer formed on 3D structured alloy high‐capacity a‐Si NW anodes using various high‐voltage imidazolium ionic liquid formulations to establish the key factors enabling capacity retention. Figure 1a–e is a schematic illustrating the solvation structure of the electrolytes employed in the study. We prepared room temperature ionic electrolytes (IL) by combining bis(fluorosulfonyl)imide (FSI⁻) based imidazolium IL solvents with their analogous Li imide salts to form symmetric‐anion formulations, specifically LiFSI‐EMIFSI (IE1‐ Sym FSI⁻). A hybrid IL electrolyte was also formulated by combining an FSI⁻ anion‐based imidazolium IL solvent with different Li‐imide salts, LiTFSI‐EMIFSI (IE2‐ Asym FSI⁻/TFSI⁻), resulting in an asymmetric anion formulation. Additionally, a symmetric bis(trifluoromethanefluorosulfonyl)imide (TFSI⁻) anion‐based electrolyte was formed, LiTFSI‐EMITFSI (IE3‐ Sym TFSI⁻). For reference, a standard fluoroethylene carbonate (FEC), based carbonate electrolyte, 1 M LiPF₆ in EC/DMC + 3% FEC (STD + FEC), was also included in the study for comparison.

Figure 1.

a) Schematic of LIB with a‐Si NW anode; Structure and solvation chemistry of the electrolytes: b) LiFSI‐EMIFSI (IE1, Sym FSI⁻),c) LiTFSI‐EMIFSI (IE2, Asym FSI⁻/TFSI⁻,d) LiTFSI‐EMITFSI (IE3, Asym FSI⁻),e) STD + FEC; SEM images of a‐Si/Cu₁₅Si₄ NW electrode f) top view, and g) side view; h) STEM image of a‐Si/Cu₁₅Si₄ NW and corresponding EDX mapping of i) Si (red) and j) Cu (yellow); k) TEM image of a‐Si/Cu₁₅Si₄ NW; l) HRTEM image focusing the Cu₁₅Si₄ NW and a‐Si layer interface and corresponding FFT pattern.

Figure 1f shows the SEM image of the 3D binder‐free a‐Si/Cu₁₅Si₄ NW architecture synthesized by a two‐step process. In the first step, the metal‐rich copper silicide (Cu₁₅Si₄) NWs were formed directly on the surface of commercial Cu foil using a solvent vapor growth technique (SVG), following our previously reported procedure (Figure S1a–c).^{[ 66 ]} Consequently, the plasma‐enhanced chemical vapor deposition (PECVD) technique was used to deposit an electrochemically active a‐Si shell layer (Figure S1d,e),^{[ 2} , ^{66 ]} evident from the electrode's side‐view SEM image (Figure 1g). STEM‐ADF image of the FIB‐TEM lamella of individual NWs shows that the a‐Si shell conformally coats the Cu₁₅Si₄ NW core, with an overall diameter of ∼0.73 µm (Figure 1h). The EDS elemental maps (Figures 1i,j and S1f–h) show the Cu‐rich Cu₁₅Si₄ NW core with an a‐Si shell. Figure 1k shows the interface between the Cu₁₅Si₄ NW and a‐Si layer. HRTEM analysis of the core (Figure 1l) displays d‐spacing values of ∼3.1 and ∼1.9 Å, corresponding to (30$\overline{1}$) and (24$\overline{2}$) planes of cubic Cu₁₅Si₄. The XRD and Raman analysis of the a‐Si/Cu₁₅Si₄ NW electrode (Figure S1i,j) support the above observations.

Performance Evaluation of Imidazolium Electrolyte Formulations with Symmetric and Asymmetric Anions

The electrochemical performance of a‐Si/Cu₁₅Si₄ NW anodes was evaluated in a Li‐ion half‐cell battery configuration using selected room‐temperature ionic electrolytes with symmetric anion formulations (IE1‐Sym FSI⁻, IE3‐Sym TFSI⁻) and asymmetric anion formulations (IE2‐Asym FSI⁻/TFSI⁻). Figure S2a compares the 1st cycle CV profiles, investigating SEI formation on a‐Si/Cu₁₅Si₄ NW anodes with selected formulation, cycled in the potential range between 2–0.01 V at a scan rate of 0.1 mV s⁻¹. The IE1 and IE2 (using EMIFSI as a solvent) and STD + FEC each exhibited a reduction peak at ∼1.45 V, forming an anion‐derived initial SEI followed by characteristic two‐stage a‐Si alloying/dealloying behavior.^{[ 55 ]} In contrast, the IE3 (using EMITFSI as the solvent) did not exhibit a high‐voltage TFSI⁻ reduction peak or characteristics of a‐Si lithiation/de‐lithiation peaks, except for the electrolyte decomposition peak at ∼0.72 V. The initial charge‐discharge voltage profiles of the a‐Si/Cu₁₅Si₄ NW anode cycled galvanostatically using IE1, IE2, and STD + FEC exhibited reversible alloying charge/discharge plateaus (Figure S2b). Consistent with CV results, cycling using IE3 did not demonstrate reversible Li⁺ alloying/de‐alloying plateaus.^{[ 55} , ^{67 ]}

The cycling performance and corresponding capacity retention plot of Li/a‐Si/Cu₁₅Si₄ NW cells in different electrolytes are shown in Figure 2a,b. Comparatively, IE1 (Sym FSI⁻) demonstrated the stable cycling performance, retaining 98% of its charge capacity after 100 cycles, whereas IE2 (Asym FSI⁻/TFSI⁻) showed slightly lower stability with a capacity retention value of 90%. In contrast, STD + FEC exhibited steady capacity decay with 81% charge capacity retention after 100 cycles, respectively (Figures 2b and S3a). The corresponding voltage profiles exhibited typical charge/discharge characteristics of a‐Si (Figure S3b). Generally, a negative shift in the peak voltage due to overpotential is observed because of the SEI formation in the initial cycles. The lithiation onset potential (Figure S3c), derived from the dQ/dV plots (Figure S4a–d), indicates that the a‐Si/Cu₁₅Si₄ NW cycled using IE1 (Sym FSI⁻) exhibits the highest lithiation onset potential, characterized by a low energy barrier for Si alloying after the first cycle, compared to IE2 (Asym FSI⁻/TFSI⁻) and STD + FEC.

Figure 2.

a) Comparison cycling performance of a‐Si/Cu₁₅Si₄ NWs at C/2 rate using IE1 (Sym FSI⁻), IE2 (Asym FSI⁻/TFSI⁻), IE3 (Sym TFSI⁻), and STD + FEC electrolytes; b) corresponding capacity retention graph; c) corresponding coulombic efficiencies; interfacial impedance of a‐Si/Cu₁₅Si₄ NW anodes d) after 1st cycle, e) after 100th cycle; f) comparative long‐term cycling at 1C rate using IE1 and IE2 electrolytes; g) comparative voltage profiles at 1st, 150th, and 250th cycle; h) rate performance at C/10, C/5, C/2, C, 2C, 5C, and back at C/10; i) High loading a‐Si/Cu₁₅Si₄ NW cycling data using IE1; j) Voltage profiles of 1st, 25th, 50th, 75th, and 100th cycle.

The IE1 (Sym FSI⁻) also achieved a high initial coulombic efficiency (CE) of ∼66%, compared to IE2 (Asym FSI‐/TFSI⁻) (51%) and STD + FEC (∼45%) (Figure 2c). IE1 rapidly approached a CE of ∼99% within 20 cycles, whereas comparatively lower CE values were achieved with IE2 (∼97%) and STD + FEC (∼96%). IE1 also demonstrated a superior reversible discharge capacity of ∼1577 mAh g⁻¹ after 500 cycles at C/2, compared to IE2 with 1282 mAh g⁻¹ (Figure S5a–c), with an impressive CE of ∼99.3% after 500 cycles, confirming a stable electrode–electrolyte interface. Figure 2d,e shows the interfacial resistance (R _SEI & R _ct) of a‐Si/Cu₁₅Si₄ NW anode cycled using IE1 (Sym FSI⁻), IE2 (Asym FSI⁻/TFSI⁻), and STD + FEC electrolytes after the 1st and the 100th cycle. EIS spectra and fitting details of a‐Si NW anodes using different electrolytes are provided in Figure S6a–c. During the intial cycle, the anode cycled using IE1 showed the lowest total impedance (R _total = 48.56 Ω) compared to IE2 (82.05 Ω), and STD + FEC (672 Ω). Even after 100 cycles, IE1 retained the lowest total impedance R _total, following a sequence of STD + FEC (292.3 Ω) > IE2 (181.04 Ω) > IE1 (81.26 Ω).

At a higher C‐rate (1C) also, IE1 (Sym FSI⁻) electrolyte enabled superior cycling stability, with a high reversible discharge capacity of 2489 mAh g⁻¹ after 250 cycles compared to IE2 (Asym FSI⁻/TFSI⁻) with 1982 mAh g⁻¹ (Figure 2f). The corresponding comparative voltage profiles (Figure 2g) show an increase in voltage polarization over cycles using IE2 compared to IE1. Rate capability was assessed by increasing the C‐rate sequentially for five cycles each (Figures 2h and S7). The a‐Si anodes cycled using IE1 demonstrated superior capacity retention at every C‐rate compared to IE2. After cycling at a higher C‐rate, IE1 rebounded to a high reversible charge capacity of 2741 mAh g⁻¹ (C/10), 84% of the initial value, whereas IE2 retained only 58% (1853 mAh g⁻¹), showing the superior cycling stability of a‐Si anodes in IE1 (Sym FSI⁻) electrolyte at a faster charging rate compared to IE2 electrolyte (Asym FSI⁻/TFSI⁻). The superior performance of IE1 is linked to its symmetric FSI⁻ dominated solvation structure, which promotes stable SEI formation, enhances Li⁺ diffusion kinetics, and reduces overpotential at the Si interface, outperforming the mixed asymmetric FSI⁻/TFSI⁻ electrolyte.

Increasing the areal capacity of a‐Si anodes is a key requirement to meet practical energy targets; therefore, the active mass loading was increased (0.82 mg cm⁻²). This high areal loading (HL) using IE1 (Figure 2i) initially showed a high areal charge capacity of 2.5 mAh cm⁻² but suffered from capacity decay, retaining a discharge capacity of 1.1 mAh cm⁻² after 100 cycles. The corresponding voltage profiles (Figure 2j) show a retained discharge capacity of 1319 mAh g⁻¹ after 100 cycles. Post‐mortem analysis (Figure S8a–d) revealed that the HL anode cycled using IE1 forms a conformal passivating layer around the individual a‐Si/Cu₁₅Si₄ NWs, protecting them from irreversible volume change. However, stress‐induced NW delamination was identified as the primary cause of anode failure for high areal loading. Further, to demonstrate the compatibility of these anodes in a full‐cell, we paired the high areal loading a‐Si/Cu₁₅Si₄ NW anodes with a high‐voltage Li‐rich layered oxide (LRLO) cathode and performed a rate capability test by sequentially increasing the C‐rate for three cycles each (Figure S9). The full cell delivered an initial charge capacity of ∼300 mAh g⁻¹ at a C/10 rate and a reversible charge capacity of 214, 118, and 33 mAh g⁻¹ at C/5, C/2, and C rates, respectively. Further, upon cycling back to C/5, the cell exhibited superior capacity retention, with 183 mAh g⁻¹ after 50 cycles.

Post‐Cycling Morphological Analysis of a‐Si Anodes Under Selected Electrolyte Formulations

To elucidate the influence of distinct symmetric and asymmetric anion IL formulations on the structural evolution, we investigate the morphological changes of the a‐Si anode and the coating efficiency of the passivating layer. Top‐view and cross‐sectional analyses of a‐Si/Cu₁₅Si₄ NW anodes were performed using FIB‐SEM after 100 cycles under different electrolyte formulations, before and after SEI layer removal. The detailed assessment provides insights into how the electrolyte composition influences anode stability, the mechanical integrity of the SEI, and ultimately impacts electrochemical performance.

Figure 3a–h shows the top and cross‐sectional SEM images of the SEI formed on the a‐Si/Cu₁₅Si₄ NW interface after 100 cycles. The anode surfaces cycled using STD + FEC (Figures 3a and S10a) and IE3 (Sym TFSI‐) (Figures 3b and S10b) exhibited a thick, porous SEI film with multiple cracks. This unstable SEI formation on the Si anode cracks during cycling, allowing electrolyte infiltration, and promotes excessive thick SEI growth and increases interfacial resistance (Figure 2h). In contrast, IE1 (Sym FSI‐) & IE2 (Asym FSI‐/TFSI‐) produced a uniform, stable SEI (Figures 3c,d and S10c,d). The cross‐sectional analysis of the electrode cycled using STD + FEC (Figure 3e) reveals a non‐uniform SEI coverage around the core‐shell NW architecture, leading to high swelling. Cycling using IE3 (Figure 3f) presents a thick SEI layer with no visible core‐shell NW architecture. Although cycling using IE2 (Figure 3g) leads to uniform SEI coverage, the a‐Si/Cu₁₅Si₄ NW still undergoes moderate swelling and structural deterioration. The anodes cycled in IE1 (Sym‐FSI⁻) electrolyte exhibit a conformal SEI layer on the nanowire architecture, with minimal electrode swelling.

Figure 3.

a)–d) Post cycling SEM analysis of a‐Si/Cu₁₅Si₄ NW with SEI layer cycled using STD + FEC, IE3 (Sym TFSI⁻), IE1(Sym FSI⁻), and IE2 (Asym FSI⁻/TFSI⁻) electrolytes after 100 cycles. e)–h) Cross‐sectional analysis with SEI layer cycled using STD + FEC, IE3, IE1, and IE2 electrolytes after 100 cycles. i)–l) Electrodes cycled using STD + FEC, IE3, IE1, and IE2 electrolytes after 100 cycles without SEI layer.

Figure 3i–l shows the top view (with high magnification images inset) of the cycled anodes after SEI removal. Prominent cracks due to pulverization were observed across the electrode cycled using STD + FEC (Figures 3i and S11a), with the NWs undergoing huge volume change and losing structural integrity. The unstable SEI causes multiple deep cracks, explaining the capacity fade using STD + FEC. The anode cycled using IE3 undergoes delamination of a‐Si from the Cu₁₅Si₄ NWs, losing the core‐shell architecture (Figures 3j and S11b). In contrast, the anodes cycled using IE2 (Figures 3k and S11c) and IE1 (Figures 3l and S11d) retain their intact core‐shell architecture. Furthermore, the individual core‐shell NWs demonstrated only minor diameter increases, from ∼0.7 to 0.813 µm for IE2 (Figure 3k inset) and ∼0.81 µm (Figure 3l inset) for IE1. The difference in the SEI characteristics and related morphological evolution of the active material is summarized in Scheme 1. The SEI formed from IE1(Sym FSI⁻), after long‐term cycling (250 cycles), did not exhibit any prominent crack on the surface (Figure S11a), in contrast to IE2 (Asym FSI⁻/TFSI⁻), where noticeable cracks were observed (Figure S12b), explaining the capacity fading observed in IE2 (Figure S6a) over prolonged cycling. The symmetric FSI⁻ anion (IE1) dominated electrolyte facilitates a robust SEI, protecting the structural integrity of the a‐Si/Cu₁₅Si₄ NW (Figure S11c,d) and preventing pulverization.

Scheme 1.

Schematic showing a) pristine a‐Si/Cu₁₅Si₄ NW electrode; morphology of SEI layer upon cycled a‐Si/Cu₁₅Si₄ NW anode using, b) IE1 (Sym FSI⁻) and IE2 (Asym FSI⁻/TFSI⁻) (with SEI), c) IE1 and IE2 (after SEI removal), d) STD + FEC (with SEI), e) STD + FEC (after SEI removal).

SEI Formation and Interfacial Chemistry

The electrochemical analysis revealed that the symmetric FSI⁻ anion paired with imidazolium cation IL electrolyte exhibited superior capacity retention across all C‐rates compared to the analogous asymmetric FSI⁻/TFSI⁻ anion paired imidazolium electrolyte formulation. The chemical composition of the SEI layer governs its stability during the volume changes of a‐Si/Cu₁₅Si₄ NWs, which in turn critically affects cycling stability and interfacial resistance. To probe the SEI composition, we perform a detailed X‐ray photoelectron spectroscopy (XPS) depth profiling via Ar⁺ sputtering of the anode after 250 cycles to elucidate the relationship between SEI chemistry and electrochemical performance. The comparative XPS spectra of fluorine (F), sulfur (S), and oxygen (O) components of SEI layer formed on the a‐Si/Cu₁₅Si₄ NW surface using IE1 (Sym FSI⁻) and IE2 (Asym FSI⁻/TFSI⁻) electrolyte formulation are provided in Figure 4.

Figure 4.

a)–f) High‐resolution XPS depth profiles (0 to 1000 s) of SEI on a‐Si/Cu₁₅Si₄ NW using IE1 (Sym FSI⁻) and IE2 (Asym FSI⁻/TFSI⁻) after 250 cycles, (a) and (d) F1s spectra, (b) and (e) S2p spectra, (c) and (f) O 1s spectra. (g) Atomic percentage versus sputter time of components using IE1. h) Schematic illustration of the components' distribution in the IE1‐derived SEI layer. i) Bar graph representing compositional change. j) Atomic percentage versus sputter time of components. Using IE2, k) schematic illustration of components distribution in the IE2‐derived SEI layer. l) Bar graph representing compositional change.

The F 1s (Figure 4a) spectra of the SEI formed in IE1 (Sym FSI⁻) show the presence of LiF (∼684.8 eV) at low percentages (3%) on the surface.^{[ 68 ]} Upon Ar⁺ depth profiling, the percentage increases to ∼11% (1000s), confirming formation of LiF‐rich inorganic inner layer due to S─F bond cleavage in FSI⁻. The presence of LiF‐rich inner layer provides mechanical robustness to the inner SEI (Figure 4h,i).^{[ 62 ]} In S 2p (Figure 4b), the SO₄ ²⁻ peak (169.4 eV) confirms the presence of a stable Li₂SO₄ component,^{[ 62 ]} due to breakdown of the sulphonyl group.^{[ 69} , ^{70 ]} This portion was higher in the outer regions, 11% (0 to 300 s), than in the inner areas, 2.8% at 1000 s (Figure 4h,i). Li₂SO₄ has been reported to stabilize and encapsulate the Li₂CO₃ and LiOH growth in the inner region, preventing further electrolyte decomposition.^{[ 71 ]} Li₂SO₄ also imparts additional benefits of high mechanical and electrochemical stability and good Li ion conductivity.^{[ 69} , ⁷² , ^{73 ]} In the O 1s spectra, the dominant peak at ∼532.5 eV is assigned to SO₄, further confirming the presence of Li₂SO₄ (Figure 4c).^{[ 69} , ^{70 ]} After sputtering for 600 s, a new peak at 529 eV appears, characteristic of the oxide Li₂O.^{[ 22} , ^{74 ]} Simultaneously, an increase of the peak at ∼531 eV is observed, corresponding to LiOH and Li _x SiO _y .^{[ 36} , ⁵⁵ , ^{75 ]} The related Li 1s peak appears at 54.5 eV, and the presence of Li _x SiO _y is further confirmed in the Si 2p spectra with a peak at 100.5 eV (Figure S13a,b).^{[ 29 ]} Figure 4g summarizes the elemental composition in the SEI layers using IE1 (Sym FSI⁻). A reduced C % (33.01% (0 s) and 10.03% (1000 s)) upon sputtering and an increase in the Li ratio (15.6% (0 s) and 33.4% (1000s)), indicating an inorganic‐rich inner layer with the formation of LiF, LiOH, Li₂SiO₄, Li₂O, Li₃N, and outer SEI layer with Li₂SO₄ species and flexible polymeric layer which is confirmed by the formation of ether carbons C–O (286 eV) and C–O–C (286.5 eV) (Figure S15C1s).^{[ 57 ]} The segregated SO₂ groups interact with Li⁺ and electrons, forming Li₂S₂O₄, which, upon disproportionation (via reaction with free oxygen from the anode interface, cathode lattice oxygen, decomposed electrolyte products, impurities, and trace water in electrolyte), can form fully oxidized sulfate species Li₂SO₄.^{[ 27} , ⁷⁶ , ^{77 ]} The higher carbon and nitrogen ratio on the outer SEI is ascribed to the stable polymeric C–O, C–O–C, C–N, N–SO _{X ,} and EMIM⁺ cation decomposition due to the acidic proton in the C2 position.^{[ 56} , ^{57 ]} Therefore, IE1 (Sym FSI⁻) presents ideal SEI characteristics with a mechanically robust inorganic‐rich inner SEI (LiF and LiOH) surrounded by stable inorganic (Li₂SO₄) and polymer (C–O & C–O–C) components with elastic properties. The smaller FSI⁻ anion is positioned at the top of the EMIM⁺ cation ring, creating a favorable solvation structure with a weakly coordinating EMIM⁺‐FSI⁻ structure and weaker electrostatic interactions, enabling a larger diffusion coefficient of the ions. FSI⁻ anion‐rich solvation shells around Li^+, enables higher ion mobility.^{[ 78} , ^{79 ]} which in turn facilitates faster decomposition of the FSI⁻ anion and benefits the generation of anion‐derived SEI components with higher Li⁺ mobility and interphase stability.^{[ 80} , ⁸¹ , ^{82 ]}

In comparison, the F 1s spectra of IE2 (Asym FSI⁻/TFSI⁻) consistently show the presence of LiF (∼684.8 eV) at around 10.7% throughout the SEI layer (Figure 4d). In contrast to IE1 (Sym FSI⁻), a lower content (1.2%) of the SO₄ ²⁻ corresponding to Li₂SO₄ was observed in the outer region (Figure 4e).^{[ 69 ]} In O 1s spectra upon etching to 180 s (Figure 4f), a higher proportion of LiOH (∼531 eV) and Li₂CO₃ (∼531.8 eV), along with Li₂O (∼528.2 eV) and Li₂SiO₄ (∼530.1 eV) in lower amounts, was observed.^{[ 55} , ^{75 ]} The corresponding Li1s peak at ∼54.5 eV, and Si2p peak at 100.7 eV was observed (Figure S14a,b).^{[ 75} , ^{83 ]} Compared to IE1 (Sym FSI⁻), LIF, LiOH, and Li₂CO₃ are present throughout the SEI layer with increasing relative amounts upon depth profiling for IE2.

In IE2 (Asym FSI⁻/TFSI⁻), a Li‐rich SEI (∼48%) is present, based on inorganic components (LiF, LiOH, Li₂O, and Li₂CO₃) (Figure 4j). Furthermore, low sulfur (4%) indicates reduced TFSI⁻ anion decomposition products in the SEI. TFSI⁻ anions are larger compared to FSI⁻ anions and exhibit strong coordination with both EMIM⁺ cation and Li⁺. They are less reactive, causing steric hindrance and creating a more strongly coordinating, rigid solvation environment, thereby critically inducing a slower reduction of active anions, and decreasing the Li⁺ mobility.^{[ 78} , ^{79 ]} This stabilization could shift the anion decomposition pathway and modulate Li⁺ solvation, thereby reducing the likelihood of FSI⁻ reduction at the anode surface.^{[ 84 ]} The increased inorganic content throughout the SEI layer forms a rigid SEI layer, likely to rupture, causing additional electrolyte decomposition and Li inventory loss, leading to capacity fade (Figures 4k,l and S16).^{[ 3 ]} This clearly shows the significance of the Li₂SO₄ component as its presence in the outer region significantly benefited the IE1 electrolyte from further decomposition and consumption. In contrast, in IE2 (Asym FSI⁻/TFSI⁻), the absence of this component significantly contributed to the continuous growth of inorganic components within the inner and outer SEI, rendering it less flexible to act as a buffer for Si during volume expansion during long‐term cycling, which ultimately led to pulverization, as observed in the post‐mortem analysis.^{[ 3 ]} Li _x Si and SiO₂ were also observed after etching to 1000 s in IE1, indicating a compact SEI compared to IE2, as these peaks were absent after etching (Figure S13a).

In contrast to other ILs, IE3‐derived SEI shows two significant differences: the formation of M _x C _y on the inner SEI due to the interaction of acidic protons with Li⁺, forming imidazolium carbene (Figure S17a).^{[ 57 ]} We observe that the initial alloying of Li with Si forms Li _x Si _y (98.2 eV), which does not de‐alloy after imidazolium carbene formation (Figure S17b). Compared to IE1(Sym FSI⁻), the elemental composition in the SEI layers using IE3 (Sym TFSI⁻) shows very low S (1.21%), N (3%), and F (4%) at 1000 s, which indicates that TFSI⁻ reduction has not completely taken place (Figure S18). The atomic percentage of F increases in the outer layer due to the presence of organo‐fluoro complexes, indicating cleavage of S‐CF₃, with a resulting peak for CF₃ complexes at 293 eV (∼ 6%). However, the similarity was observed with respect to the IE2 electrolyte, with a low percentage of S and N species. However, the IE2 exhibited higher Li, C, and O content due to the formation of LiF, LiOH, and Li₂CO_3. However, the IE3 does not show any significant amount of Li‐rich inorganic compound in the inner region. Alternatively, a higher Si atomic percentage was observed (∼60%(1000 s), including elemental Li_xSi_y (∼26%), indicating that Si alloyed with Li did not undergo any dealloying process, 13% of elemental Si, indicating incomplete alloying of Si and formation of an irreversible Li₄SiO₄ compound with around 5% and also SiO₂ (10%). The absence of a significant atomic percentage of LiF (684.5 eV (1.3%)) (Figure S17c) and other inorganic compounds in IE1 & IE2 critically indicates the absence of an anion‐derived SEI in IE3 formulation.

To further confirm the presence of Li₂SO₄ in the SEI of IE1, postmortem morphological analysis of the a‐Si/Cu₁₅Si₄ nanowire (NW) anodes cycled in IE1 (Sym‐FSI⁻) was performed using cryo‐TEM (Figure S19a). Examination of the NW edges revealed clear evidence of Li₂SO₄ formation in the outermost SEI layer, while LiF was detected closer to the inner region. The selected‐area FFT of the HRTEM from the marked region (pink) showed d‐spacing values of 2.32 Å, corresponding to the (3) plane of monoclinic Li₂SO₄ (Figure S19b). Inner regions displayed d‐spacing values of 2.27 Å, calculated from the selected‐area FFT pattern associated with the {111} planes of cubic LiF (Figure S19c). Similar results were observed for another anode cycled in IE1, confirming the Li₂SO₄ formation. The fast FFT pattern from another sample showed a d‐spacing value of 1.97 Å, closely matching the (004) plane of monoclinic Li₂SO₄ (Figure S19d).

Conclusion

Overall, the a‐Si/Cu₁₅Si₄ NW cycled using a symmetric FSI⁻ anion dominated electrolyte formulation (LiFSI‐EMIFSI) displayed notably superior capacity retention compared to asymmetric anion formulations (LiTFSI‐EMIFSI) and carbonate electrolyte, as explained through comprehensive electrochemical testing and post‐mortem analysis. XPS revealed that EMIFSI facilitates a higher decomposition rate of FSI⁻ and TFSI⁻ anion, resulting in an anion‐derived inorganic‐rich SEI. However, the FSI⁻ anion‐dominated solvation structure facilitates higher S─F, N─S bond cleavage and S reduction, resulting in higher S, N, and F‐rich SEI with key Li₂SO₄ compound formation on the outer layer, inhibiting further electrolyte decomposition.^{[ 27} , ^{62 ]} The LiFSI‐EMIFSI SEI comprises an inorganic‐rich (LiF/LiOH) inner layer and a Li₂SO₄ outer layer surrounded by flexible polymer compounds. Additionally, the SEI facilitated by LiFSI‐EMIFSI is more uniform and conformal, protecting the 3D a‐Si NW structure. In contrast, the asymmetric FSI⁻/TFSI⁻ system‐derived SEI lacks the stable Li₂SO₄ component, resulting in the continuous growth of the inorganic species LiOH and Li₂CO₃, forming a rigid SEI prone to fracture, resulting in capacity fade. The study highlights that these decomposition pathways are highly dependent on the anions used in the electrolyte formulation, which play a crucial role in determining the passivating nature of the resulting interphase. In particular, the composition employing FSI⁻ anions demonstrate superior performance in forming a stable solid electrolyte interphase (SEI) compared to formulations with alternative anions. The insights on the interfacial chemistry and morphology of SEI using ionic liquids in this study will be crucial for further electrolyte engineering for Si anode development.

Supporting Information

Experimental methods, additional physical (XRD, SEM, TEM, FIB‐SEM, and XPS), and electrochemical data that support the study are provided in the Supporting Information.

Conflict of Interests

The authors declare no conflict of interest.

Supporting information

Supporting Information

Sankaran A., Laffir F., Maresca G., Kapuria N., Van der Velden M., Adegoke T. E., Israel S. S., Appetecchi G. B., Geaney H., Ryan K. M, Angew. Chem. Int. Ed. 2026, 65, e22709. 10.1002/anie.202522709

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.