Mitigating the volume expansion and enhancing the cycling stability of ferrous fluorosilicate-modified silicon-based composite anodes for lithium-ion batteries

Abstract

Silicon has emerged as a prominent candidate for anodes in advanced lithium-ion batteries due to its exceptional theoretical capacity and low operational potential. Despite its advantages, silicon-based anodes face significant challenges, including substantial volume changes, formation of an unstable solid-electrolyte interphase (SEI) film, and voltage hysteresis during lithium alloying/dealloying, which compromise their cycling stability. This study introduces a novel ferrous fluorosilicate (FeSiF₆)-modified silicon-based composite anode. FeSiF₆ is prepared via a simple reaction between Fe–Si alloys and hydrofluoric acid (HF). Various treatment methods are employed to create modified silicon-based composites with different compositions and morphologies. This innovative composite material prevents the formation of crystalline Li₁₅Si₄ and facilitates the formation of a stable SEI film, thereby markedly improving the cycling stability of the silicon-based anodes. Among these, the composite material Fe–Si@F@C (consisting of Fe–Si alloy@FeSiF₆@graphite) demonstrates a stable discharge capacity of 975 mA h g⁻¹ after 200 cycles at 1 A g⁻¹, with ∼94% capacity retention, and outstanding rate performance (664.4 mA h g⁻¹ at 4 A g⁻¹). In comparison, the Fe–Si alloy/graphite anode without FeSiF₆ shows a much lower discharge capacity of 458 mA h g⁻¹ at 1 A g⁻¹ after 200 cycles and 291.8 mA h g⁻¹ at 4 A g⁻¹. These findings underscore the critical role of FeSiF₆ in modifying silicon-based anodes and enhancing their cycling stability, significantly increasing their potential for commercial application in next-generation lithium-ion batteries.

---

1. Introduction

Silicon has gained widespread attention as a potential anode material for next-generation lithium-ion batteries (LIBs), owing to its superior theoretical capacity (3589 mA h g⁻¹ based on the alloying reaction 4Si + 15Li⁺ + 15e⁻ → Li₁₅Si₄ at room temperature), abundant availability (constituting about 28% of the Earth's crust), and cost-effectiveness (approximately 5 US$ per kg).^1,2 Despite these advantages, the extensive volume change (nearly 383% expansion/contraction) experienced by silicon during the charge/discharge cycles results in electrode cracking/fracturing, formation of an unstable solid electrolyte interphase (SEI) film, loss of electrical contact, and considerable capacity fade.^3–5 Consequently, the research community has dedicated efforts towards microstructural/compositional optimization, binder enhancement, and electrolyte engineering to address these issues.^4,6–8 Nanostructuring of silicon into forms such as nanoparticles, nanowires, and nanosheets has become a focal point of investigation,^9,10 aimed at reducing stress from the formation of crystalline Li₁₅Si₄ (c-Li₁₅Si₄) during full lithiation¹¹ and improving the diffusion rates of lithium ions within these nanoarchitectures.^12,13 In particular, silicon nanoparticles with a critical size of approximately 150 nm have shown resilience against cracking or fracturing in their initial lithiation cycle.¹⁴ However, the large specific surface area of these nanostructures predisposes them to aggregation, leading to excessive lithium/electrolyte consumption (which reduces the initial coulombic efficiency), and impedes electron transport at the electrode/electrolyte interface, thus adversely affecting battery performance.^15,16

To tackle these issues, one can employ encapsulation technology and surface modification, incorporate electrolyte additives, or combine silicon with other materials (i.e., compositing).^15,17,18 Encapsulation involves enveloping silicon in the highly conductive and buffering matrix (for example, carbon layer, carbon nanotubes, and graphene), mitigating the adverse effects of the electrode's expansion and improving cycling stability.¹⁹ Surface modification primarily entails functionalizing the surface of silicon particles through specific techniques (for instance, atomic/molecular layer deposition²⁰) to enhance their interface stability with electrolytes.²¹ Furthermore, the use of electrolyte additives like organic fluoroethylene carbonate (FEC) and vinylene carbonate (VC) has been proven to facilitate the formation of an SEI film, predominantly comprising LiF and cross-linked polyethylene oxide.^18,22 This enhancement in SEI film stability reduces electrolyte decomposition, boosts coulombic efficiency, and thereby aids in better capacity retention. However, it is important to highlight that the FEC additive is fully consumed when the delithiation cutoff voltage exceeds 0.7 V, utilizing lithium metal as the counter electrode, which prompts a rapid decline in capacity.²³ In addition, the presence of Lewis acid PF₅ in LiPF₆-based electrolytes triggers the defluorination of FEC at higher temperatures,²⁴ leading to the emission of gaseous species (such as CO₂ and H₂) and corrosive HF.²⁵ This corrosive HF can breach the SEI layer, reacting with the silicon and resulting in the degradation of the active silicon material.⁵

Additionally, the formation of composites through the combination of silicon with other materials, such as carbonaceous materials, metal oxides and intermetallic compounds, boosts both its structural stability and electrochemical performance.^26,27 Composite materials typically provide benefits such as straightforward preparation, increased mechanical stability,⁴ improved electron transport/Li-ion diffusion,²⁸ increased surface reaction activity,²⁹ and stabilized SEI film.^30,31 These benefits have turned silicon-based composite anode materials into a hotspot for research and development, offering new possibilities for the application of next-generation lithium-ion batteries. For example, silicon-based nanocomposites with FeSi₂ and carbon exhibit a reversible charge capacity of 1045 mA h g⁻¹ and an initial coulombic efficiency of 87%.³² This benefits from the buffer effect of the highly elastic/lithium-inactive FeSi₂ phase and conductive carbon layers. Furthermore, the study demonstrates that different carbon-based conductive modifiers are crucial for enhancing the conductivity and electrochemical performance of silicon-based anodes in lithium batteries.^33–35 Liu et al. found that fluorine-containing precursor carbon materials significantly improve the performance of pure-phase nano-silicon particles.³⁶ The porous nanocomposites they prepared exhibited a specific capacity of approximately 660 mA h g⁻¹ after 50 cycles, maintaining 75% of the capacity retention rate. Interestingly, compared to other commonly used carbon sources, such as polyvinyl chloride, pitch, and sucrose, the Si/C composites made with fluorine-containing precursors also showed improved capacity retention rates.

While conventional additives partially stabilize the SEI film and mitigate continuous side reactions between silicon and the electrolyte, their effectiveness is often compromised by their inability to uniformly withstand silicon's substantial volumetric changes, resulting in ongoing lithium consumption. Fluorine-containing compounds possess strong antioxidative properties, which help mitigate the issue of Li–Si alloying by enhancing the SEI film on the anode surface.³⁷ In addition to additives, composite materials also play a crucial role in suppressing silicon lithiation and enhancing its stability. Electrochemical performance analysis indicates that the presence of ferrous fluorosilicate (FeSiF₆) can prevent the irreversible formation of Li₂SiF₆ to some extent, especially after combining silicon with graphite, further stabilizing the anode. Moreover, these additives/composites play a crucial role in suppressing the lithiation of silicon, enhancing its stability. It is also hypothesized that the presence of FeSiF₆ can to some extent prevent the irreversible formation of Li₂SiF₆, especially following the integration of silicon with graphite, which further stabilizes the anode.³⁸

This study proposes a straightforward synthesis method for FeSiF₆ and investigates its effectiveness in improving the performance of silicon-based anodes. We characterized the electrochemical performance of the materials and explored the impact of varying FeSiF₆ concentrations on anode performance. Differential capacity analysis indicates that the FeSiF₆ modifier plays a critical role in inhibiting the formation of c-Li₁₅Si₄ during lithiation, thereby significantly reducing electrode expansion. The Fe–Si alloy@FeSiF₆@graphite (Fe–Si@F@C) composite anodes exhibited robust performance, maintaining a stable discharge capacity of 975 mA h g⁻¹ after 200 cycles at a current density of 1 A g⁻¹, which corresponds to approximately 94% capacity retention. The composite anode also demonstrated impressive rate performance (664.4 mA h g⁻¹ at 4 A g⁻¹). The findings from this study are anticipated to significantly advance research on silicon-based anodes and underscore their potential for commercial applications, marking a notable contribution toward the development of more resilient and efficient LIBs.

2. Experimental section

2.1 Material preparation

2.1.1 Preparation of pure-phase FeSiF₆. The schematic in Fig. S1† details the synthesis of FeSiF₆ as follows: (a) 1 g of Fe₄₅Si₅₅ alloy powder (500 mesh, 99% purity, Benyu Metal Materials Co., Ltd., China) is placed into a plastic beaker; (b) 30 mL of 1 M hydrofluoric acid is added, and the beaker is then sealed with cling film; (c) the reaction proceeds at 40 °C for 24 hours; (d) unreacted impurities are removed via vacuum filtration, followed by transfer of the solution to a Petri dish; (e) the sample is dried in a vacuum oven at 80 °C for 24 hours, resulting in the production of light green crystals.

Phase composition analysis by X-ray diffraction (XRD) revealed the presence of FeSi and FeSi₂ phases in the initial Fe–Si alloy powders, as shown in Fig. S2.† The XRD pattern of the synthesized FeSiF₆ crystals, presented in Fig. S3,† matches closely with the reference pattern for FeSiF₆·6H₂O (PDF#26-0799), demonstrating that the Fe–Si alloy (FeSi_x) reacted with HF to produce FeSiF₆, as per the reaction: FeSi_x + HF → FeSiF₆ + H₂ (gas). The morphology and structure of the FeSiF₆ crystals are depicted in Fig. S3a and b,† illustrating a hexagonal columnar formation with a width of around 2 µm and relatively smooth side surfaces. A high-resolution image in Fig. S3c† reveals stacked structures at the top, presumed to form during the crystallization phase, indicative of the FeSiF₆ crystals' continuous growth through evaporation and crystallization. In contrast, scanning electron microscopy (SEM) images (Fig. S4†) of the initial Fe₄₅Si₅₅ alloy show irregular particles with an uneven size distribution, highlighting the synthesis method's effectiveness in producing a uniform FeSiF₆ morphology.

Transmission electron microscopy (TEM) images (Fig. S3d†) depict the columnar structure of FeSiF₆, encased in a transparent thin outer layer. The high-magnification TEM image in Fig. S3e† reveals hollow structures within the crystals and their uneven distribution, suggesting that these voids may serve as insertion sites for Li⁺ ions in electrode materials, potentially enhancing stability. The high-resolution TEM (HRTEM) images in Fig. S3f† display distinct lattice fringes, with measurements at three different locations yielding lattice spacings of 0.42 nm, 0.38 nm, and 0.33 nm, corresponding to the (012), (021), and (202) planes of FeSiF₆ (PDF#26-0799), respectively. Furthermore, elemental mapping in Fig. S3g† showcases the even distribution of F, Fe, and Si elements within the FeSiF₆ structure, affirming its compositional uniformity.

2.1.2 Preparation of Fe–Si composite anode materials. Fe–Si composite alloy anodes were developed using three distinct methods to optimize performance (as listed in Table 1): employing an excess of FeSi alloy (incomplete reaction), utilizing an excess of HF (the Fe–Si alloy fully reacted), and applying an equal mixture of both Fe–Si alloy and FeSiF₆. Quantitative analysis of the phase compositions of the materials was conducted using the Jade software. The objective was to identify the most effective synthesis approach.

Table 1 Details of the preparation steps and compositions for the Fe–Si composites, including Fe–Si–T@C, Fe–Si@C–T, and Fe–Si@F@C

Abbreviations for materials	Differences	Preparation method	Composition (wt%)
Fe–Si–T@C	Employing an excess of FeSi alloy (incomplete reaction)	(a) Add 10 mL of 1 M hydrofluoric acid to a beaker; (b) mix in 1 g of Fe25Si75 alloy powder (500 mesh, 99% purity, Benyu Metal Materials) and seal the beaker; (c) heat at 40 °C for 12 hours to react; (d) dry the mixture at 80 °C under vacuum for 24 hours; (e) mix 0.6 g of the resultant powder with 0.4 g of graphite (from Dongguan KELIDA) in a ball mill; (f) mill with stainless steel balls for 30 minutes in a SPEX 8000M mill (with a fixed clamp speed of 875 cycles per minute); (g) rinse with ethanol, filter, and dry at 80 °C under vacuum for 6 hours	Graphite 40%, silicon 21.2%, FeSiF6 22.8%, and FeSi2 16%
Fe–Si@C–T	Utilizing an excess of HF (the Fe–Si alloy fully reacted)	(a) Combine 0.6 g of Fe25Si75 alloy powder with 0.4 g of graphite in a ball milling jar; (b) add 40 g of 3 mm and 10 g of 1.5 mm stainless steel balls to the jar and mill for 30 minutes using a SPEX 8000M mill; (c) wash the milled powder with ethanol, filter, and dry at 80 °C under vacuum for 6 hours; (d) soak the dry powder in 10 mL of 1 M hydrofluoric acid, seal with plastic wrap, heat to 40 °C on a heating plate for 12 hours to ensure complete reaction; (e) finally, transfer the reacted mixture to a Petri dish and dry under vacuum at 80 °C for 24 hours	Graphite 40%, silicon 12.2%, FeSiF6 41.3%, and FeSi2 6.5%
Fe–Si@F@C	Applying an equal mixture of both Fe–Si alloy and FeSiF6	(a) Measure 0.3 g each of Fe25Si75 alloy powder and FeSiF6 powder, along with 0.4 g of graphite, and place them into a ball milling jar; (b) add 40 g of 3 mm and 10 g of 1.5 mm stainless steel balls to the jar, then mill for 30 minutes using a high-energy ball mill (SPEX 8000M); (c) rinse the milled composite with ethanol, filter out the powder, and dry it under vacuum at 80 °C for 6 hours	Graphite 40%, silicon 21.8%, FeSiF6 29.1%, and FeSi2 9.1%

---

2.2 Morphology and composition characterization

The crystal structure of the resulting samples was characterized by X-ray diffraction (XRD) using a multifunctional powder X-ray diffractometer (X'Pert Powder, PANalytical, The Netherland) with a Cu K_α radiation source. The morphology and microstructure of the samples were characterized by field-emission scanning electron microscopy (FE-SEM, Hitachi SU8010, Japan) and high-resolution transmission electron microscopy (TEM, FEI Talos F200X, USA). X-ray photoelectron spectroscopy (XPS, Axis Supra+, Kratos Shimadzu, Japan) was used to identify the chemical states of Si, C, F, O, Li, and Fe in the samples.

2.3 Electrochemical measurements

The electrochemical performance was characterized by assembling a CR2032 button cell with lithium cells serving as the counter/reference electrode. The anode consists of 70 wt% active materials, 20 wt% acetylene black, and 10 wt% polyvinylidene fluoride (PVDF). These powders were mixed with N-methyl-2-pyrrolidone and stirred on a magnetic stirrer for 4 hours to obtain a homogeneous slurry. The slurry was evenly coated onto a copper foil and then dried in an oven at 60 °C for 8 h. The weight of the active material on the electrode is about 0.5–1.0 mg cm⁻², with a thickness of ∼15 µm, as measured using a micrometer. The weight of the active material is calculated based on the weight of the composite material. The electrolyte is LB-064 (Do Do Chem) comprising 1.0 M LiPF₆ in ethylene carbonate (EC) and dimethyl carbonate (DMC) (EC : DMC = 1 : 1 by volume). A microporous polypropylene film (Celgard 2400) was used as the separator for the button cell. Cyclic voltammetry (CV) was performed on an electrochemical workstation (model VSP-300, BioLogic, France) with a voltage range of 0.005–3 V and sweep rates of 0.1 and 0.5 mV s⁻¹. Electrochemical impedance spectroscopy (EIS) tests were conducted on the same electrochemical workstation at open circuit potentials with frequencies ranging from 100 kHz to 0.01 Hz and an AC amplitude of 5 mV. A multi-channel battery test system (LAND CT2001A, Wuhan LAND Electronics Co., Ltd., China) was used for galvanostatic charge–discharge (GCD) measurements in the range of 0.005–3 V (vs. Li⁺/Li).

3. Results and discussion

Pure-phase crystalline FeSiF₆ is efficiently synthesized from Fe–Si alloys (comprising FeSi and FeSi₂) using a straightforward reaction with HF, followed by vacuum filtration and drying. This method is simpler and more resource-efficient than traditional approaches, offering environmental benefits.^39,40 The material's transformation was evaluated through SEM images and the reaction process is illustrated in Fig. 1a–c. XRD analyses for Fe–Si@C–T, Fe–Si–T@C, and Fe–Si@F@C composites, presented in Fig. 1d, confirm the formation of FeSiF₆ in all samples post-HF treatment. The Fe–Si@C–T sample shows a complete disappearance of FeSi₂ peaks, leaving only the silicon phase. Meanwhile, Fe–Si–T@C reveals a mix of Si, FeSi₂, and FeSiF₆, with Si being the predominant phase due to the partial conversion of FeSi₂ into FeSiF₆, thereby reducing the presence of both FeSi and FeSi₂.

Fig. 1 (a–c) Schematic diagram for the synthesis of Fe–Si@C–T, Fe–Si–T@C, and Fe–Si@F@C. (d) XRD results of Fe–Si@C–T, Fe–Si–T@C, and Fe–Si@F@C samples.

SEM analyses were conducted on the three materials to examine the effects of various preparation methods on morphology, as shown in Fig. 2. The analysis highlights notable differences in morphology: Fe–Si@C–T prepared by ball milling and acid treatment (Fig. 2a) differs significantly from those prepared by other methods. Specifically, Fe–Si@C–T (Fig. 2b) and Fe–Si@F@C (Fig. 2c) show a layered graphite structure, offering a buffer for silicon anode volume expansion and improving stability.^15,41 Conversely, the Fe–Si@C–T processed through ball milling and acid treatment lacks this layered structure, presenting smoother surfaces with graphite well integrated with FeSiF₆ and Si. The TEM images (Fig. 2d) corroborate these observations, displaying a uniform graphite distribution in Fe–Si@C–T and distinct FeSiF₆ particles. Fig. 2e and f clearly depict the graphite layers enveloping other components in a sandwich-like manner. Elemental analysis of Fe–Si@C–T (Fig. 2g) indicates a core–shell structure, with Fe, F, and C forming an outer layer around Si, suggesting that this method effectively creates core–shell morphologies.

Fig. 2 SEM images of (a) Fe–Si@C–T, (b) Fe–Si–T@C, and (c) Fe–Si@F@C. TEM images of (d) Fe–Si@C–T, (e) Fe–Si–T@C, and (f) Fe–Si@F@C. (g) HAADF-STEM image and corresponding EDX mapping images of Fe–Si@C–T.

Research shows FeSi₂ enhances silicon anodes' cycling stability through a buffering effect,^15,42 while graphite creates a conductive network, improving electrical conductivity.^15,28,43 CV tests on Fe–Si@C–T, Fe–Si–T@C, and Fe–Si@F@C (Fig. S5†) revealed similar oxidation–reduction peak positions for the materials, with variations in intensity due to different compositions and morphologies. The initial scans show peaks below 0.1 V indicating Li⁺ insertion into crystalline silicon to form amorphous Li–Si (a-LiSi_x) alloys,^44,45 with further lithiation (at 30–60 mV) transforming these into crystalline phase c-Li₁₅Si₄.⁴⁶ Peaks at 0.34 V and 0.53 V are linked to the reconversion of a-LiSi_x and c-Li₁₅Si₄ to amorphous silicon.⁴⁷ Despite FeSi₂ being present, it shows no distinct redox activity with Li.⁴⁸ However, FeSiF₆'s CV curve (Fig. S6†) displays reversible peaks at 0.92 V and 1.4 V for oxidation, and 2.02 V and 1.62 V for reduction, similar to FeOF's Li⁺ storage reactions,⁴⁹ suggesting these peaks represent Fe(III) ↔ Fe(II) intercalation and conversion processes, respectively.

During the initial lithiation, Li-ions intercalate into FeSiF₆, causing a partial oxidation of Fe²⁺ to Fe³⁺ without altering the FeSiF₆ structure. In subsequent lithiation, the accumulation of Li-ions continues to reduce Fe³⁺ back to Fe²⁺ until all Fe³⁺ is fully converted. The oxidation states of other elements remain stable within the tested voltage range, indicating no significant changes. The CV curve for the Fe–Si@F@C material shows a slight shift in the Fe oxidation–reduction voltage in FeSiF₆, possibly due to energy absorption by other components, leading to a minor voltage increase.^50,51 A comparison of the CV curves of the three materials suggests that a higher presence of the FeSi₂ phase correlates with an elevated oxidation peak voltage. The similarity of curves between 1.0 V and 3.0 V in Fig. S5a and c,† along with XRD data, implies a relationship with FeSiF₆ content. Additionally, with more charging and discharging cycles, the redox peaks' intensity grows, indicating enhanced electrolyte penetration and gradual activation of the anode material.⁵²

The GCD curves for the three composite anodes at 0.1 A g⁻¹ reveal distinct characteristics in Fig. S7a and b,† and 3a. An extended voltage plateau below 0.1 V during the first discharge (Fig. 3a and S7b†) signifies the alloying of lithium with silicon, forming Li_xSi alloys, while a plateau from 0.3 V to 0.6 V during charging indicates de-alloying.⁵³ Comparing the GCD curves of the three composite materials reveals that all three exhibit a distinct lithiation reaction plateau around 1.5 V, which disappears in subsequent cycles. The initial discharge curves of Fe–Si@F@C and Fe–Si@C–T are more similar, likely due to the sufficient content of FeSiF₆ in both materials. During discharge, the presence of FeSiF₆ promotes the formation of a different SEI film, consuming some Li⁺ and increasing the Li storage capacity of the anode. Consequently, this reaction does not occur during subsequent charge–discharge cycles, leading to a capacity decrease and lower initial coulombic efficiency.^54–56 Despite similar electrochemical behavior across Fe–Si@C–T, Fe–Si–T@C, and Fe–Si@F@C samples, their specific capacities differ. Fe–Si@C–T displays a first-cycle discharge/charge capacity of 1291.5/907.5 mA h g⁻¹, achieving an initial coulombic efficiency (CE) of 70.2%, which improves to 92.7% in the second cycle. Fe–Si–T@C shows initial capacities of 1828.8/1414.8 mA h g⁻¹ with an initial CE of 77.36%. Fe–Si@F@C records initial capacities of 1924.9/1497.7 mA h g⁻¹, with the CE increasing to 94.6% in the second cycle and stabilizing at approximately 96.8% in later cycles. FeSiF₆ alone has an initial CE of 49.68%, suggesting FeSiF₆ and graphite significantly enhance the reversible lithium storage capacity of silicon-based anodes. Fe–Si@C–T demonstrates the most consistent rate performance, delivering capacities of 1187.5 to 534.4 mA h g⁻¹ across current densities from 0.1 to 4 A g⁻¹ (Fig. 3b). Fe–Si@F@C shows competitive rate performance at higher current densities, with capacities of 789 and 664.4 mA h g⁻¹ at 2 and 4 A g⁻¹, respectively. Fig. 3c showcases the cycling performance of the three anodes, with Fe–Si@C–T and Fe–Si@F@C demonstrating remarkable stability and minimal capacity loss after 200 cycles at 1 A g⁻¹, maintaining discharge capacities of 685.7 and 974.9 mA h g⁻¹, respectively, for a ∼94% capacity retention. By comparing the composition and electrochemical performance of the three composite materials, it is evident that the higher the FeSiF₆ content, the more stable the electrochemical performance of the material. Both Fe–Si–T@C and Fe–Si@F@C composites exhibit a slight increase in capacity, primarily due to structural defects introduced during ball milling.⁵⁷ The initial capacity decline in Fe–Si–T@C is attributed to the volumetric expansion of silicon. XRD analysis shows that the silicon and graphite content ratios are almost identical in both composites, with the difference being the varying amounts of FeSiF₆ and FeSi₂. It is apparent that FeSiF₆ stabilizes the silicon anode's performance better than FeSi₂.

Fig. 3 (a) GCD curves of Fe–Si@F@C in the 1st, 2nd and 5th cycles measured at 0.1 A g⁻¹. (b) Rate performance of Fe–Si@C–T, Fe–Si–T@C, and Fe–Si@F@C at different current densities. (c) Cycling performance of Fe–Si@C–T, Fe–Si–T@C, and Fe–Si@F@C at 1 A g⁻¹. (d) EIS spectra of Fe–Si@C–T, Fe–Si–T@C, and Fe–Si@F@C at a scan rate of 0.1 mV s⁻¹. The inset shows the equivalent circuit model used for EIS fitting. (e) Linear fitting of the Z′–ω^−1/2 (ω = 2πf) relationship for Fe–Si@C–T, Fe–Si–T@C, and Fe–Si@F@C.

To more intuitively analyze the material's performance and the reasonableness of its capacity, we compiled the data into Table 2 for comparison. The purchased high-energy-density graphite has a relatively high specific capacity. According to Wei et al.'s report,⁵⁸ the capacity of amorphous graphite can reach up to 810 mA h g⁻¹ in the voltage range of 0–1.2 V. Based on the study conducted by Zhao et al.,⁵⁹ which employed calculations to determine the theoretical capacity of materials, we can similarly use the data from the figure and the previously analyzed material composition to estimate the theoretical capacity of the composite material. The calculation formula is as follows:

Table 2 Initial discharge specific capacity (IDSC), initial charge specific capacity (ICSC), second cycle discharge specific capacity (SDSC), second cycle charge specific capacity (SDSC), theoretical specific capacity (TSC), discharge capacity after 200 cycles (C₂₀₀) for 1 A g⁻¹, and the retention percentage of specific capacity to IDSC (RPID) for electrodes Fe₂₅Si₇₅, graphite, FeSiF₆, Fe–Si–T@C, Fe–Si@C–T, and Fe–Si@F@C (the GCD curves of Si, graphite, and FeSiF₆ are shown in Fig. S8)

Electrode	Si	Graphite	FeSiF6	Fe–Si@C–T	Fe–Si–T@C	Fe–Si@F@C
IDSC (mA h g^−1)	3854.2	1313.4	1330.0	1291.5	1828.9	1924.9
ICSC (mA h g^−1)	2759.1	937.7	660.7	907.5	1414.9	1497.7
SDSC (mA h g^−1)	2718.3	896.7	653.0	1258.4	1439.6	1474.6
SCSC (mA h g^−1)	2565.3	827.3	602.5	1166.6	1521.2	1395.5
TSC (mA h g^−1)	—	—	—	1544.9	1645.5	1752.7
ICE (%)	71.58	71.32	49.68	70.2	77.36	78.84
C200 (mA h g^−1)	128.0	274.4	427.9	679.8	917.5	971.9
RPID (%)	23.75	135.5	93.6	96.45	104.17	104.25

---

Therefore, the theoretical capacity of the composite materials is:

Fe–Si@C–T: 3854.2 × 12.2% + 1330 × 41.3% + 1313.4 × 40% = 1544.9 mA h g⁻¹

Fe–Si–T@C: 3854.2 × 21.2% + 1330 × 22.8% + 1313.4 × 40% = 1645.5 mA h g⁻¹

Fe–Si@F@C: 3854.2 × 21.8% + 1330 × 29.1% + 1313.4 × 40% = 1752.7 mA h g⁻¹

From the calculation results, it can be seen that the test results of Fe–Si–T@C and Fe–Si@F@C are higher than the calculated values. This may be due to the modification effects of the ball-milled composite materials on the material interface, which significantly enhance the cycling stability of the materials. However, more Li⁺ is consumed during the SEI formation in the initial cycles.⁶⁰ As the composite materials are more stable, the capacity retention rate of the composite materials also improves significantly, resulting in an increase in capacity in subsequent cycles compared to the initial material capacity.⁵⁷

Furthermore, capacity calculations based on the material composition revealed that the initial cycle capacity is somewhat high. This is primarily due to two factors: first, the formation of the SEI film during the initial discharge consumes additional Li⁺ ions.⁶⁰ Second, the structure of the material itself contributes to this phenomenon. Characterization of the material revealed the formation of a unique sandwich structure, with FeSiF₆ containing some void structures. These structural features increase the material's defect density, allowing it to store more Li⁺.⁵⁷ By comparing the initial capacities of the composite material Fe₂₅Si₇₅@C and the pristine Fe₂₅Si₇₅ material (Fig. S9†), it is evident that the initial capacity of Fe₂₅Si₇₅@C is only ∼40% of the pristine material, whereas Fe₂₅Si₇₅–T exhibits an initial capacity of 93.6% of the pristine material. This indicates that although graphite improves the capacity stability of the material, it sacrifices a certain amount of capacity, possibly due to incomplete graphite encapsulation, leading to the reaction of silicon with Li₂PF₆ to form Li₂SiF₆, consuming lithium ions.³⁸ On the other hand, the addition of FeSiF₆ results in a smaller loss of capacity. The composite of the two materials achieves an improvement beyond the sum of their individual effects. Furthermore, from the analysis above, it can be inferred that the presence of FeSiF₆ can to some extent transform Li₂PF₆, rendering it a reversible process, thereby reducing Li⁺ consumption and increasing specific capacity.

EIS analysis of the three iron–silicon composite anodes before cycling provided insights into their electrochemical reaction processes, using equivalent circuit models to interpret Nyquist plots (Fig. 3d). The high-frequency semi-circle represents the SEI film resistance (R_f), and the medium-frequency one indicates charge-transfer impedance (R_ct), with a low-frequency straight line denoting Li⁺ diffusion within the material (Warburg impedance, Z_w). In the model, R_Ω is the cell's internal resistance, and CPE is the constant phase element. Fe–Si–T@C showed the lowest R_ct at approximately 40 Ω, suggesting excellent stability. Fe–Si@F@C and Fe–Si@C–T presented R_ct values of about 88 Ω and 245 Ω, respectively. Compared to the base material Fe₂₅Si₇₅ (Fig. S10†), which has an R_ct of 462 Ω, surface modifications like graphite coating improved conductivity, reducing R_ct to 273 Ω. With FeSiF₆, R_ct decreased further to 164 Ω, indicating FeSiF₆'s significant role in enhancing charge transfer kinetics, especially when combined with graphite. Analyzing the slope of the low-frequency Z′–ω^−1/2 curve (Fig. 3e) showed the Fe–Si@F@C anode with the lowest slope (306.3), implying a higher Li⁺ diffusion coefficient than the others.

Differential capacity (dQ/dV) curves after various cycles at 1 A g⁻¹ for the composite anodes (Fig. S11a–c†) highlight initial lithiation peaks for c-Li₁₅Si₄ (∼0.45–0.51 V) and a-LixSi (∼0.27–0.34 V), with shifts attributed to restricted Li-ion diffusion⁶¹ and polarization effects under high current density.⁶² As cycling progresses, the anodic peaks for Fe–Si@C–T and Fe–Si@F@C shift towards lower voltages, suggesting enhanced lithium insertion/extraction kinetics and reduced polarization.⁶³ In contrast, Fe–Si–T@C shows variable peak shifts, indicating a lower initial polarization that increases over cycles but improves versus the original material, possibly due to FeSiF₆'s effect on polarization. Notably, the sharp peaks at 0.15 V for Fe–Si–T@C and Fe–Si@F@C, versus a broader peak for Fe–Si@C–T, reflect different carbon structures from varied synthesis methods, linked to lithium deintercalation from lithiated graphite. Increased FeSiF₆ content in composite anodes correlates with higher anodic peak values after 100 cycles, suggesting FeSiF₆'s role in enhancing cycling stability. Higher FeSiF₆ content in silicon-based composite anodes correlates with increased anodic peak values after 100 cycles, suggesting FeSiF₆'s role in enhancing cycling stability. During delithiation, two cathodic peaks for amorphous silicon phases, L₁(a-Si) and L₂(a-Si), are noted, with L₂(a-Si) at 0.25 V becoming more pronounced and stable in capacity as FeSiF₆ content increases. Over cycles, both cathodic peaks diminish, with L₁(a-Si)'s decrease highlighting the challenge of lithiating amorphous silicon without affecting crystalline silicon, as per Obrovac's findings.⁶⁴ In contrast, L₂(a-Si) provides a transition region, stabilizing the cyclic performance of the material.

To elucidate FeSiF₆'s impact on composite material performance, dQ/dV plots for the initial and fifth cycles at 0.1 A g⁻¹ were analyzed, highlighting changes in c-Li₁₅Si₄ peak (at ∼0.45 V) intensities. Initially, Fe₂₅Si₇₅ showed a c-Li₁₅Si₄ peak intensity of ∼10.71 mA h g⁻¹ V⁻¹ (Fig. 4a), which reduced to ∼3.04 mA h g⁻¹ V⁻¹ after five cycles (Fig. 4b). In contrast, Fe–Si–T@C (Fig. S12a†), Fe–Si@C–T (Fig. S12b†), and Fe–Si@F@C (Fig. 4c) exhibited peak intensities of ∼7.09 mA h g⁻¹ V⁻¹, ∼1.84 mA h g⁻¹ V⁻¹, and ∼4.50 mA h g⁻¹ V⁻¹, respectively, indicating more stable lithium storage with higher FeSiF₆ content. This indicates that a higher FeSiF₆ content better inhibits the formation of c-Li₁₅Si₄. This inhibition can be attributed to two main factors: first, the addition of FeSiF₆ promotes the formation of a stable SEI film, which reduces the consumption of lithium ions through side reactions, thereby stabilizing the interface structure of the silicon anode and inhibiting the formation of c-Li₁₅Si₄;^18,65 second, the inclusion of FeSiF₆ can alter the lithium ion insertion pathways and diffusion rates, encouraging the formation of an amorphous phase and thus having a suppressive effect.^66–69 Additionally, SEI film modifications were noted in Fig. 4c, especially the emergence of a broad peak around 1.5 V linked to fluorine incorporation from FeSiF₆,¹⁸ enhancing SEI stability. To verify the reliability of the experimental results, the first cycle (Fig. 4e) and the second cycle (Fig. 4f) CV curves of the five materials were plotted at a scan rate of 0.1 mV s⁻¹, along with the CV curve of FeSiF₆ alone (Fig. S13†). The comparison reveals that after compounding the materials, the peak of Li₁₅Si₄ at around 0.5 V is smaller than that of the original Fe₂₅Si₇₅ material, indicating that the reactivity of the composite material at this potential has decreased.⁷⁰ By comparing the CV curves of different electrodes in the first cycle, it can be seen that the range of 0.15–0.75 V corresponds to the SEI formation process for the Fe–Si alloy anode material.^71,72 For the CV curve of the electrode with added FeSiF₆, a new broad peak appears between 1.33 V and 1.70 V. The position of this peak shows a certain deviation from the SEI formation position in the CV curve of pure FeSiF₆, indicating that this peak results from the combined effects of the composite material. Furthermore, this peak disappears in the second cycle, which may indicate that the composite material forms a relatively stable SEI film during the initial cycle and creates an effective passivation layer on the surface.⁷³ By comparing with the literature, this phenomenon may be due to the decomposition of FeSiF₆, where some substances (presumably F⁻) react with the electrolyte, resulting in the formation of a more stable and robust SEI film.¹⁸ The consistent performance and the presence of F-ions, as analyzed in CV curves and supported by the literature, imply that FeSiF₆ significantly contributes to SEI film stability.

Fig. 4 Differential capacity curves of Fe₂₅Si₇₅ at (a) the 1st cycle and (b) 5th cycle at 0.1 A g⁻¹. The differential capacity curves for Fe–Si@F@C anodes at (c) the 1st cycle and (d) 5th cycles at 0.1 A g⁻¹. CV profiles of (e) the 1st cycle and (f) 2nd cycle for different electrodes at a scan rate of 0.1 mV s⁻¹.

CV analysis of Fe₂₅Si₇₅, FeSiF₆, and Fe–Si@F@C electrodes after multiple cycles at 1 A g⁻¹ (Fig. 5a–c) revealed distinct behaviors. Fe₂₅Si₇₅ shows reduced peak intensities for a-Li_xSi and c-Li₁₅Si₄ transitions (∼0.3–0.5 V) with cycling (Fig. 5a), suggesting a decrease in reaction activity. In contrast, FeSiF₆ (Fig. S13†) exhibited increased peak intensities across cycles, indicating stable specific capacity due to the reversible Li⁺ insertion/extraction, enhanced by electrolyte permeation facilitating Li⁺ migration.⁷⁴ Fe–Si@F@C (Fig. 5b) displayed stabilized peaks related to a-Li_xSi and c-Li₁₅Si₄ transitions after ∼50 cycles, suggesting different protective mechanisms. Graphite improves mechanical stability and mitigates structural stress from volume expansion, while FeSiF₆'s protective effect, akin to FeSi₂'s role, warrants further investigation to elucidate its specific contributions to silicon-based anodes.

Fig. 5 Cyclic voltammetry profiles of (a) Fe₂₅Si₇₅ and (b) Fe–Si@F@C composite anode in the first five cycles at a sweep rate of 0.5 mV s⁻¹. High-resolution XPS spectra of F 1s of (c) Fe₂₅Si₇₅ and (d) Fe–Si@F@C under different cycle times.

The test results for the F element are shown in Fig. 5c and d below. It is evident from the graph that at 50 cycles, the Li–F (685 eV) peak intensity of Fe₂₅Si₇₅ is significantly lower,⁷⁵ and the SEI film contains a large amount of Si–F (687.4 eV).⁷⁶ In contrast, the composite material exhibits higher Li–F content with less Si–F. This further confirms that the presence of FeSiF₆ provides a substantial amount of F-containing bonds on the material's surface. During the formation of the SEI film, the Si–O–F and F–O bonds on the surface break, allowing more F ions to combine with Li and thereby generating more Li–F bonds within the SEI film. As the number of cycles increases, the binding energy of the F element in both cases becomes nearly identical. Moreover, both materials exhibit good stability in terms of capacity with increasing cycle numbers. This indicates a preference for forming LiF-rich SEI films on the anode surface, which enhances Li⁺ transport and minimizes side reactions and electrolyte consumption. The presence of LiF correlates with decreased impedance, highlighting its role in mitigating volume expansion and stabilizing electrode structures during cycling.⁷⁷ Nevertheless, Fe₂₅Si₇₅ exhibited a rapid decline in capacity during the first few cycles, likely due to its inability to fully counteract the micron-scale volumetric expansion of the silicon anode.⁷⁸ Conversely, the Fe–Si@F@C, with FeSiF₆, not only strengthens the SEI film but also mitigates volume expansion more effectively, leading to superior cycling stability.⁷⁹ This indicates the promoting effect of FeSiF₆ on the generation of LiF within the SEI film, thereby enhancing the electrochemical stability of the material.

XPS analysis of silicon-based anodes modified with FeSiF₆ revealed changes in the SEI film composition during cycling (Fig. S14a–c†). Initial C 1s spectra for all samples showed peaks typical of organic solvent molecules, including C–O, C=O, C–H, and C–C, mainly from graphite, carbon black, and PVDF binder, notably including C–F bonds. An increase in C–C/C–H bonds at 284.8 eV was observed across spectra, indicating electrolyte decomposition and SEI formation.⁸⁰ After 50 cycles, new peaks at 288.5 eV for ROCO₂Li and 289.6 eV for Li₂CO₃ appeared, representing major SEI components.¹⁸ Fe₂₅Si₇₅ showed increased ROCO₂Li and Li₂CO₃ (Fig. 5e), suggesting a thicker SEI film compared to the thinner films on the other materials, which correlated with their stable performance. After 200 cycles, the ROCO₂Li and Li₂CO₃ intensities decreased in Fe₂₅Si₇₅, remained stable in FeSiF₆, and increased in Fe–Si@F@C, likely due to SEI film thickening from cycling in Fe–Si@F@C. In contrast, Fe₂₅Si₇₅'s fluctuating peak intensities could be due to SEI film rupture and regeneration caused by volume expansion, whereas FeSiF₆ showed stability. The incorporation of FeSiF₆ into the Fe–Si alloy is inferred to strengthen the SEI film's toughness,⁸¹ with its thickness accumulating over cycles due to silicon's volume expansion. Fig. S15† presents the Si 1s spectra for the three electrode materials with varying cycle numbers, initially showing peaks for Si–Si (99.4 eV), Fe–Si (99.8 eV), and Si–C (102 eV) bonds.⁸² Notably, the composite electrode exhibits both Si–Si and Fe–Si bonds. Post-cycling, these peaks shift to SiO_x (∼101 eV) and Li_xSiO_y (102 eV), indicating Si–O bond involvement in reactions with the SEI film.^83,84 Fe₂₅Si₇₅ shows Si–Si/Fe–Si bond peaks at 150 and 200 cycles, contrasting with the more stable peak intensities of FeSiF₆ and Fe–Si@F@C. This stability discrepancy is attributed to Fe₂₅Si₇₅'s SEI film rupture from volume expansion, exposing internal materials, in line with existing research. Fig. S16† showcases the Li 1s spectra of the three materials at various cycles, revealing similar peak positions for them all, suggesting comparable Li-containing components in their SEI films, including Li₂O (54.3 eV), ROCO₂Li and LiCO₃ (54.8 eV), and LiF (55.7 eV). It also demonstrated the formation of an SEI film rich in LiF on the anode material surface, corroborated by the analysis of the F element.

The cycling stability and volume expansion mitigation of FeSiF₆ in silicon-based anodes were evaluated through cross-sectional morphology characterization before and after cycling (Fig. 6). The Fe₂₅Si₇₅ electrode expanded significantly from 6.15 µm to 21.89 µm, a ∼256% increase, after 200 cycles, exhibiting electrode fracture and detachment (Fig. 6a). In contrast, the FeSiF₆ electrode showed a minimal volume expansion of ∼34%, correlating with stable cycling performance (Fig. 6b). The Fe–Si@F@C composite electrode, incorporating FeSiF₆ and graphite, demonstrated controlled volume expansion of ∼84% after 200 cycles (Fig. 6c), indicating that FeSiF₆ effectively reduces volume expansion in silicon alloy anodes.

Fig. 6 Cross-sectional morphologies of the electrodes: (a) Fe₂₅Si₇₅, (b) FeSiF₆, and (c) Fe–Si@F@C after different number of cycles (initial state, 100 cycles and 200 cycles).

EIS measurements on Fe₂₅Si₇₅ and Fe–Si@F@C electrodes after various cycling stages (Fig. 7) showed a rapid reduction in charge transfer resistance (R_ct) for Fe₂₅Si₇₅, dropping from an initial 462 Ω to 93 Ω within fifty cycles and stabilizing at about 50 Ω after 100 cycles. This change is attributed to LiF accumulation in the SEI film, as identified by XPS, which improves Li⁺ migration and lowers impedance. Similarly, the Fe–Si@F@C electrode, benefiting from FeSiF₆ incorporation, shows a marked initial impedance decrease, with impedance fluctuating based on LiF content in the SEI film before stabilizing at approximately 39 Ω after 150 cycles. The integration of FeSiF₆ and graphite into silicon-based electrodes significantly improves interface contact stability and electronic transport to the copper foil. Our research identifies mechanisms for enhanced cycling stability of these electrodes (Fig. 7c),^18,28,74,81 focusing on three key findings: (1) dQ/dV and SEM analyses show FeSiF₆'s role in stabilizing the c-Li₁₅Si₄ alloy and mitigating volume expansion; (2) CV and dQ/dV investigations, supported by XPS, reveal that FeSiF₆ leads to a thinner, more stable, and flexible SEI film in early cycles, enhancing electrode–electrolyte interaction; and (3) the analysis suggests that FeSiF₆ may inhibit the reaction between the Si material and electrolyte to form Li₂SiF₆, thereby reducing Li⁺ consumption and volume changes. Comparing our results with the existing literature (Table S1 and Fig. S17†), our composite anodes demonstrate competitive capacities and stabilities, highlighting the effectiveness of adding FeSiF₆ and graphite in improving silicon-based anodes' performance.

Fig. 7 EIS spectra of (a) Fe₂₅Si₇₅ and (b) Fe–Si@F@C with a scan rate of 0.1 mV s⁻¹ under different cycle times. (c) Schematic diagram of the performance improvement mechanism in the Fe–Si@F@C anode.

4. Conclusions

This research presents a novel anode material modifier, FeSiF₆, through a straightforward chemical reaction, exploring its enhancement effects and lithium storage mechanisms on silicon-based composite anode materials. Three composites were synthesized, displaying stable performance, with Fe–Si–T@C achieving a high initial discharge capacity of 3250.8 mA h g⁻¹ at 0.1 A g⁻¹ and an initial CE of 86.88%. Despite its lower capacity retention of 52% after 200 cycles at 1 A g⁻¹, the other two materials maintained over 94% retention. All composites showed robust rate performance, with Fe–Si@C–T sustaining a capacity of 534.4 mA h g⁻¹ even at 4 A g⁻¹. The analysis identified FeSiF₆'s key roles: (1) suppression of Li₁₅Si₄ formation, offering a transitional L₂(a-Si) area to stabilize cycling performance; (2) formation of a LiF-rich SEI film, mitigating volume expansion and enhancing Li⁺ transport through LiF inclusion; and (3) potential reduction in irreversible Li₂SiF₆ formation, improving capacity and stability in graphite–silicon composites. This study introduces a viable, environmentally friendly silicon-based anode composite material, highlighting its unique lithium storage mechanism and potential for advancing silicon anode commercialization. Further investigation into this material could significantly benefit the development of next-generation high-performance LIBs.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Author contributions

Jichang Sun: writing – original draft, data curation, investigation, and visualization. Xiaoyi Liu: validation, data curation, and investigation. Penglun Zheng: conceptualization and writing – review & editing. Yang Zhao: writing – review & editing. Yun Zheng: funding acquisition, data curation, visualization, and writing – review & editing. Jingchao Chai: data curation and visualization. Zhihong Liu: funding acquisition and writing – review & editing.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors appreciate the funding support of the Excellent Discipline Cultivation Project by JHUN (No. 2023XKZ008), State Key Laboratory of Precision Blasting, Jianghan University (No. PBSKL2022102), and the Open Project of Civil Aircraft Fire Science and Safety Engineering Key Laboratory of Sichuan Province (No. MZ2023KF03).

References

1. J. R. Szczech and S. Jin, Energy Environ. Sci., 2011, 4, 56–72.
2. P. Guan, W. Zhang, C. Li, N. Han, X. Wang, Q. Li, G. Song, Z. Peng, J. Li, L. Zhang and X. Zhu, J. Colloid Interface Sci., 2020, 575, 150–157.
3. A. Mukhopadhyay and B. W. Sheldon, Prog. Mater. Sci., 2014, 63, 58–116.
4. X. Su, Q. Wu, J. Li, X. Xiao, A. Lott, W. Lu, B. W. Sheldon and J. Wu, Adv. Energy Mater., 2014, 4, 1300882.
5. N. Kim, Y. Kim, J. Sung and J. Cho, Nat. Energy, 2023, 8, 921–933.
6. W. Luo, X. Chen, Y. Xia, M. Chen, L. Wang, Q. Wang, W. Li and J. Yang, Adv. Energy Mater., 2017, 7, 1701083.
7. G. G. Eshetu and E. Figgemeier, ChemSusChem, 2019, 12, 2515–2539.
8. L. Deng, Y. Zheng, X. Zheng, T. Or, Q. Ma, L. Qian, Y. Deng, A. Yu, J. Li and Z. Chen, Adv. Energy Mater., 2022, 12, 2200850.
9. J. Sun, C. Liu, P. Zheng, A. Chaturvedi, K.-H. Nam, Z. Li, Q. Liang, S. Huang, D. Fang, J. Chai, Q. Liu, Z. Liu, E. H. T. Teo, Z. Jian, W. Shu, Y. Zheng and C.-M. Park, Next Mater., 2023, 1, 100053.
10. P. Zheng, J. Sun, H. Liu, R. Wang, C. Liu, Y. Zhao, J. Li, Y. Zheng and X. Rui, Batteries Supercaps, 2023, 6, e202200481.
11. D. S. M. Iaboni and M. N. Obrovac, J. Electrochem. Soc., 2016, 163, A255.
12. M. A. Rahman, G. Song, A. I. Bhatt, Y. C. Wong and C. Wen, Adv. Funct. Mater., 2016, 26, 647–678.
13. M. R. Zamfir, H. T. Nguyen, E. Moyen, Y. H. Lee and D. Pribat, J. Mater. Chem. A, 2013, 1, 9566–9586.
14. X. H. Liu, L. Zhong, S. Huang, S. X. Mao, T. Zhu and J. Y. Huang, ACS Nano, 2012, 6, 1522–1531.
15. P. Li, H. Kim, S.-T. Myung and Y.-K. Sun, Energy Storage Mater., 2021, 35, 550–576.
16. Y. Sun, N. Liu and Y. Cui, Nat. Energy, 2016, 1, 16071.
17. W. Luo, X. Chen, Y. Xia, M. Chen, L. Wang, Q. Wang, W. Li and J. Yang, Adv. Energy Mater., 2017, 7, 1701083.
18. C. Xu, F. Lindgren, B. Philippe, M. Gorgoi, F. Björefors, K. Edström and T. Gustafsson, Chem. Mater., 2015, 27, 2591–2599.
19. H.-J. Shin, J.-Y. Hwang, H. J. Kwon, W.-J. Kwak, S.-O. Kim, H.-S. Kim and H.-G. Jung, ACS Sustainable Chem. Eng., 2020, 8, 14150–14158.
20. Y. Zhao, L. Zhang, J. Liu, K. Adair, F. Zhao, Y. Sun, T. Wu, X. Bi, K. Amine, J. Lu and X. Sun, Chem. Soc. Rev., 2021, 50, 3889–3956.
21. J. Shi, X. Jiang, J. Sun, B. Ban, J. Li and J. Chen, J. Colloid Interface Sci., 2021, 588, 737–748.
22. Y. Jin, N.-J. H. Kneusels, L. E. Marbella, E. Castillo-Martínez, P. C. M. M. Magusin, R. S. Weatherup, E. Jónsson, T. Liu, S. Paul and C. P. Grey, J. Am. Chem. Soc., 2018, 140, 9854–9867.
23. S. Yamazaki, R. Tatara, H. Mizuta, K. Kawano, S. Yasuno and S. Komaba, J. Phys. Chem. C, 2023, 127, 14030–14040.
24. K. Kim, I. Park, S.-Y. Ha, Y. Kim, M.-H. Woo, M.-H. Jeong, W. C. Shin, M. Ue, S. Y. Hong and N.-S. Choi, Electrochim. Acta, 2017, 225, 358–368.
25. A. Schiele, B. Breitung, T. Hatsukade, B. B. Berkes, P. Hartmann, J. Janek and T. Brezesinski, ACS Energy Lett., 2017, 2, 2228–2233.
26. F. Dou, L. Shi, G. Chen and D. Zhang, Electrochem. Energy Rev., 2019, 2, 149–198.
27. Q. Liu, Y. Hu, X. Yu, Y. Qin, T. Meng and X. Hu, Nano Res. Energy, 2022, 1, e9120037.
28. A. Casimir, H. Zhang, O. Ogoke, J. C. Amine, J. Lu and G. Wu, Nano Energy, 2016, 27, 359–376.
29. J. D. McBrayer, M.-T. F. Rodrigues, M. C. Schulze, D. P. Abraham, C. A. Apblett, I. Bloom, G. M. Carroll, A. M. Colclasure, C. Fang and K. L. Harrison, Nat. Energy, 2021, 6, 866–872.
30. H. Jo, J. Kim, D.-T. Nguyen, K. K. Kang, D.-M. Jeon, A. R. Yang and S.-W. Song, J. Phys. Chem. C, 2016, 120, 22466–22475.
31. S. Fan, H. Wang, J. Qian, Y. Cao, H. Yang, X. Ai and F. Zhong, ACS Appl. Mater. Interfaces, 2020, 12, 16411–16416.
32. Y. M. Yang, C. Loka, D. P. Kim, S. Y. Joo, S. W. Moon, Y. S. Choi, J. H. Park and K.-S. Lee, Met. Mater. Int., 2017, 23, 610–617.
33. Z. Wen, J. Yang, B. Wang, K. Wang and Y. Liu, Electrochem. Commun., 2003, 5, 165–168.
34. F. Dou, L. Shi, G. Chen and D. Zhang, Electrochem. Energy Rev., 2019, 2, 149–198.
35. X. Li, K. Li, M. Yuan, J. Zhang, H. Liu, A. Li, X. Chen and H. Song, J. Colloid Interface Sci., 2024, 667, 470–477.
36. Y. Liu, Z. Wen, X. Wang, A. Hirano, N. Imanishi and Y. Takeda, J. Power Sources, 2009, 189, 733–737.
37. M. Han, D. Zheng, P. Song and Y. Ding, Chem. Phys. Lett., 2021, 771, 138538.
38. C. Yu, X. Chen, Z. Xiao, C. Lei, C. Zhang, X. Lin, B. Shen, R. Zhang and F. Wei, Nano Lett., 2019, 19, 5124–5132.
39. A. Pastushenko and V. Lysenko, RSC Adv., 2016, 6, 8093–8095.
40. Y. Zhang, Q. Zhang, X. He, L. Wang, J. Wang, L. Dong, Y. Xie and Y. Hao, Materials, 2023, 16, 1437.
41. J. Wu, Y. Cao, H. Zhao, J. Mao and Z. Guo, Carbon Energy, 2019, 1, 57–76.
42. M. Ruttert, V. Siozios, M. Winter and T. Placke, ACS Appl. Energy Mater., 2019, 3, 743–758.
43. S. Chen, Z. Song, L. Wang, H. Chen, S. Zhang, F. Pan and L. Yang, Acc. Chem. Res., 2022, 55, 2088–2102.
44. E. Radvanyi, E. De Vito, W. Porcher, J. Danet, P. Desbois, J.-F. Colin and S. J. S. Larbi, J. Mater. Chem. A, 2013, 1, 4956–4965.
45. H. Lin, K. Uosaki and H. Noguchi, Appl. Surf. Sci., 2021, 569, 151040.
46. H. Gao, L. Xiao, I. Plümel, G.-L. Xu, Y. Ren, X. Zuo, Y. Liu, C. Schulz, H. Wiggers, K. Amine and Z. Chen, Nano Lett., 2017, 17, 1512–1519.
47. M. N. Obrovac and L. J. Krause, J. Electrochem. Soc., 2007, 154, A103.
48. H.-T. Kwon, A.-R. Park, S.-S. Lee, H. Cho, H. Jung and C.-M. Park, J. Electrochem. Soc., 2019, 166, A2221.
49. V. L. Chevrier, G. Hautier, S. P. Ong, R. E. Doe and G. Ceder, Phys. Rev. B: Condens. Matter Mater. Phys., 2013, 87, 094118.
50. Y. Su, J. Chen, H. Li, H. Sun, T. Yang, Q. Liu, S. Ichikawa, X. Zhang, D. Zhu and J. Zhao, Adv. Sci., 2022, 9, 2201419.
51. Y. Su, S. Liu, D. Zhu, Y. Luo, X. Zhang, J. Yan, J. Chen, L. Geng, B. Guo and H. Li, Energy Storage Mater., 2023, 59, 102779.
52. H.-J. Zhang, Q.-C. Jia and L.-B. Kong, ACS Appl. Nano Mater., 2021, 4, 12514–12526.
53. B. Gattu, A Fundamental Study of Nanostructured Si Anodes for Lithium Ion Batteries, University of Pittsburgh, 2020.
54. R. Shao, J. Niu, F. Zhu, M. Dou, Z. Zhang and F. Wang, Nano Energy, 2019, 63, 103824.
55. S. Sekar, Y. Lee, D. Y. Kim and S. Lee, Nanomaterials, 2019, 9, 871.
56. Y. Zhang, L. Yue, H. Ding, Z. Xiong, L. Bai, M. Xu and Y. Qi, Nano Energy, 2024, 127, 109696.
57. S. Dong, Y. Song, M. Su, G. Wang, Y. Gao, K. Zhu and D. Cao, Chem. Eng. J., 2024, 481, 147988.
58. V. Subramanian, T. Karabacak, C. Masarapu, R. Teki, T.-M. Lu and B. Wei, J. Power Sources, 2010, 195, 2044–2049.
59. Y. Zhao, J. Wang, Q. He, J. Shi, Z. Zhang, X. Men, D. Yan and H. Wang, ACS Nano, 2019, 13, 5602–5610.
60. H. Kim, W. Choi, J. Yoon, J. H. Um, W. Lee, J. Kim, J. Cabana and W.-S. Yoon, Chem. Rev., 2020, 120, 6934–6976.
61. X. Gao, W. Lu and J. Xu, ACS Appl. Mater. Interfaces, 2021, 13, 21362–21370.
62. J. Tao, L. Liu, J. Han, J. Peng, Y. Chen, Y. Yang, H.-r. Yao, J. Li, Z. Huang and Y. Lin, Energy Storage Mater., 2023, 60, 102809.
63. D. Wang, M. Gao, H. Pan, J. Wang and Y. Liu, J. Power Sources, 2014, 256, 190–199.
64. M. Obrovac and L. Krause, J. Electrochem. Soc., 2006, 154, A103.
65. P. Bärmann, B. Krueger, S. Casino, M. Winter, T. Placke and G. Wittstock, ACS Appl. Mater. Interfaces, 2020, 12, 55903–55912.
66. Y. Liu, W. Sun, X. Lan, R. Hu, J. Cui, J. Liu, J. Liu, Y. Zhang and M. Zhu, ACS Appl. Mater. Interfaces, 2019, 11, 38727–38736.
67. L. Gu, J. Han, M. Chen, W. Zhou, X. Wang, M. Xu, H. Lin, H. Liu, H. Chen and J. Chen, Energy Storage Mater., 2022, 52, 547–561.
68. C. Gan, C. Zhang, W. Wen, Y. Liu, J. Chen, Q. Xie and X. Luo, ACS Appl. Mater. Interfaces, 2019, 11, 35809–35819.
69. J. Sung, N. Kim, J. Ma, J. H. Lee, S. H. Joo, T. Lee, S. Chae, M. Yoon, Y. Lee and J. Hwang, Nat. Energy, 2021, 6, 1164–1175.
70. G. Li, G. Fan, X. Zhang, J. Han, Y. Wang, Y. Liu, L. Jia, B. Guo, C. Zhu and M. He, eTransportation, 2024, 20, 100315.
71. M. Gu, Z. Wang, J. G. Connell, D. E. Perea, L. J. Lauhon, F. Gao and C. Wang, ACS Nano, 2013, 7, 6303–6309.
72. Q. Huang, K. Turcheniuk, X. Ren, A. Magasinski, D. Gordon, N. Bensalah and G. Yushin, Adv. Energy Mater., 2019, 9, 1803323.
73. J. Zhu, M. S. D. Darma, M. Knapp, D. R. Sørensen, M. Heere, Q. Fang, X. Wang, H. Dai, L. Mereacre and A. Senyshyn, J. Power Sources, 2020, 448, 227575.
74. Y. She, X. Li, Y. Zheng, D. Chen, X. Rui, X. Lin and Y. Qin, Energy Environ. Mater., 2023, e12538.
75. M. Nie, J. Demeaux, B. T. Young, D. R. Heskett, Y. Chen, A. Bose, J. C. Woicik and B. L. Lucht, J. Electrochem. Soc., 2015, 162, A7008.
76. E. Markevich, G. Salitra, M. Afri, J. Kulisch, P. Hartmann and D. Aurbach, J. Electrochem. Soc., 2019, 166, A1685.
77. L.-B. Huang, G. Li, Z.-Y. Lu, J.-Y. Li, L. Zhao, Y. Zhang, X.-D. Zhang, K.-C. Jiang, Q. Xu and Y.-G. Guo, ACS Appl. Mater. Interfaces, 2021, 13, 24916–24924.
78. N. M. Johnson, Z. Yang, Q. Liu and Z. Zhang, J. Electrochem. Soc., 2022, 169, 040527.
79. J. Lin, L. Wang, Q. Xie, Q. Luo, D. L. Peng, C. Buddie Mullins and A. Heller, Angew. Chem., Int. Ed., 2023, 135, e202216557.
80. M. Yu, E. Temeche, S. Indris, W. Lai and R. M. Laine, Green Chem., 2022, 24, 4061–4070.
81. K. Schroder, J. Alvarado, T. A. Yersak, J. Li, N. Dudney, L. J. Webb, Y. S. Meng and K. J. Stevenson, Chem. Mater., 2015, 27, 5531–5542.
82. Q. Ma, Y. Zhao, Z. Hu, J. Qu, Z. Zhao, H. Xie, P. Xing, D. Wang and H. Yin, Mater. Today Energy, 2021, 21, 100817.
83. C. Li, Z. Liang, Z. Li, D. Cao, D. Zuo, J. Chang, J. Wang, Y. Deng, K. Liu and X. Kong, Nano Lett., 2023, 23, 4014–4022.
84. M. Nie, D. P. Abraham, Y. Chen, A. Bose and B. L. Lucht, J. Phys. Chem. C, 2013, 117, 13403–13412.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ta02532f

---