Reinforcement effect of carbon nanofibers on silicon–carbon anode materials: a review

Abstract

This review examines the reinforcement effect of carbon nanofibers (CNFs) in enhancing the electrochemical performance of silicon–carbon anode materials for lithium-ion batteries. CNFs-reinforced silicon–carbon electrode materials exhibit the advantages of constructing conductive networks, buffering silicon volume expansion, facilitating ion transport and stabilizing electrode/electrolyte interface interactions. Various silicon–carbon nanofiber composites are generalized and listed as Si/CNF, Si–CNF@C, Si–o-CNFs, hierarchical carbon–Si/CNF, C–Si–CNF, and Si@porous carbon/porous CF. The key structural designs include sandwich structures to confine silicon expansion via carbon encapsulation, interwoven architectures to enhance particle dispersion through electrostatic self-assembly, honeycomb frameworks to optimize lithium-ion diffusion via ordered pores, and wrapping or core–shell structures to reserve expansion space and improve mechanical stability. Various preparation methods for forming silicon–carbon nanofiber composites are summarized, including electrostatic self-assembly, electrospinning, and electrophoretic deposition. Several challenges in active interfacial compatibility, new electrode materials and structures, comprehensive modeling and theoretical simulation calculations are proposed. The corresponding possible solution methods are suggested, which include surface functionalization, multiscale simulations, and machine learning-guided optimization. Perspective is also provided for the new design and further development of CNFs-reinforced silicon–carbon anode materials for electrochemical energy storage applications.

---

1. Introduction

Energy is the cornerstone of social development, but with the development of the social economy, various non-renewable energy sources are consumed, leading to global environmental pollution and energy shortages. Therefore, the development of various renewable and green sources of energy is imperative. This includes solar energy, nuclear energy, and wind energy. Renewable energy has advantages such as environmental friendliness, renewability, and abundant resources. However, these energy sources also have significant spatial limitations, intermittency, and volatility. These drawbacks have led to low utilization efficiency of renewable energy, high development costs, and difficulty in meeting supply needs. Therefore, it is necessary to develop a series of energy storage devices to match these demands.^1–3

In general, electrochemical energy storage devices involve various batteries and supercapacitors.^4–7 Lithium-ion batteries (LIBs) have attracted attention from the scientific and industrial communities due to their advantages, such as long cycle time, no memory effect, high energy density, and green and pollution-free properties.^8–12 Due to its excellent conductivity, good cycling stability, and low cost, graphite remains the most commonly used lithium-ion negative electrode material.^13,14 However, due to its low specific capacity (372 mAh g⁻¹),¹⁵ short cycle life, and poor rate performance, graphite is limited as a negative electrode material for high-performance LIBs. Among numerous anode materials, silicon is considered one of the powerful alternatives to graphite. Silicon, as an anode material, has advantages such as high theoretical specific capacity (4200 mAh g⁻¹),^16–18 wide availability, and environmental friendliness. However, the silicon anode undergoes significant expansion (∼400%) during repeated lithium removal and insertion processes, generating unstable solid electrolyte films and low electron and ion conductivity.^19–22 A common solution to the defects in silicon negative electrodes is to produce silicon–carbon composite materials. Carbon has excellent mechanical properties and can buffer the volume change of silicon. In recent years, significant progress has been made in the research of silicon–carbon composite materials, with various preparation methods available, such as chemical vapor deposition technology, high-energy mechanical ball milling technology, electrospinning technology,²³ and electrostatic self-assembly technology. However, carbon coatings prepared using these methods can only partially suppress the volume change effect of silicon, and the mechanical strength of the carbon layer is often insufficient to suppress the volume change after long-term cycling, thus only providing limited stability and cycling life. When lithium ions are embedded in the structure of silicon materials during charging and discharging, they are prone to structural fracture and powdering, resulting in poor cycling performance of the battery.

Carbon-derived materials can also improve battery performance in various ways, potentially optimizing electrode thermal conductivity, mechanical strength, stability, energy storage, and discharge rate.^24–27 In addition, they allow for the design of batteries that are both lightweight and increase the bulk density of the battery pack, which helps to improve performance. They can be used as an active material, conductive additive, current collector, and also as a structural component of batteries.^26–28 In the development of carbon materials for LIB, one-dimensional carbon materials have attracted people's interest due to their unique properties, such as excellent mechanical properties, considerable conductivity, large surface area, and good chemical stability.

Table 1 compares the performance of graphene-, carbon nanotube-, and carbon fiber-based electrode materials, clearly demonstrating the superior performance of carbon fiber-based materials.

Table 1 Performance comparison of graphene-, carbon nanotube-, and carbon fiber-based electrode materials

Type of carbon electrode material		Maximum capacity (mAh g^−1)	Capacity @high C rate (mAh g^−1)	Capacity retention	Ref.
Graphene-based materials	KOH etch graphite	363@0.1C	125@1C	74% after 100 cycles (6C)	29
	Porous graphite	193@0.1C	175@2.5C	96.7% after 100 cycles (2.5C)	30
	Multichannel graphite	365@0.1C	275@10C	85% after 3000 cycles (6C)	31
CNT-based materials	TiO2/CNT	190@1C	100@60C	76% after 1000 cycles (60C)	32
	Li5Cr7Ti6O25/MWCNT	137@0.3C	123.6@3.5C	95.3% after 400 cycles (6C)	33
	NiO@CNT	824@1C	460@5C	94% after 400 cycles (1C)	34
CFs-based materials	Hierarchical CFs	845@1C		77% after 100 cycles (1C)	35
	N-doped CFs	123.8@1C		98.8% after 500 cycles (1C)	36

---

The composite application of carbon fibers (CFs)^37–40 with silicon in LIBs primarily addresses critical limitations of silicon-based anodes, such as volume expansion and poor conductivity. The key roles of CFs in silicon–carbon composite systems are shown as follows.

1.1. Conductive network construction

1.1.1. Mechanism. The high aspect ratio and three-dimensional interconnected structure of CFs form a continuous conductive network, significantly enhancing electron transport efficiency between silicon particles. Studies show that adding 10% CFs increases the composite's electronic conductivity by 2–3 orders of magnitude.⁴¹

1.1.2. Experimental validation. In situ electrochemical impedance spectroscopy (EIS) reveals that silicon-based anodes with CFs exhibit approximately 60% reduction in charge transfer resistance (R_ct) during the cycling process and significantly mitigated polarization.⁴²

1.2. Volume expansion buffering

1.2.1. Structural design. CFs bond with silicon particles via C–O–Si chemical bonds or physical encapsulation, forming a rigid–flexible framework. The high mechanical strength with elastic modulus above 200 GPa accommodates silicon's extreme volume changes of ∼300% during the lithiation/delithiation process.⁴³

1.2.2. Performance improvement. Comparative tests demonstrate that CF-reinforced silicon composite electrodes retain 82% capacity after 100 cycles, whereas pure silicon electrodes degrade below 40% of their initial capacity within 20 cycles.

1.3. Ion transport optimization

1.3.1. Pore structure advantages. The random distribution of CFs creates hierarchical porosity (micro–meso–macro pores), shortening Li⁺ diffusion paths. The Li⁺ effective diffusion coefficient in nano Si reaches 1.2 × 10⁻¹⁰ cm⁻² s⁻¹, approaching graphite anode levels.^44,45

1.3.2. Kinetic advantages. Galvanostatic intermittent titration technique confirms that CF-containing silicon anodes retain 85% specific capacity at 2C rate, demonstrating their superior rate capability.⁴⁶

1.4. Interface stability enhancement

1.4.1. SEI regulation. CFs preferentially form a stable and uniform solid electrolyte interphase (SEI), reducing direct contact between silicon and electrolyte. XPS analysis shows the increased LiF content (>35%) and the reduced organic component in the composite effectively suppress side reactions.⁴⁷

1.4.2. Cycling durability. Transmission electron microscopy reveals a 90% reduction in the crack density of silicon particles and the minimized interfacial delamination in CF composite electrodes after 200 cycles.

Carbon nanofibers (CNFs) have high conductivity, which helps to increase electron transfer during electrochemical reactions.^24,25,40,48 Carbon nanofibers (CNFs), as nanoscale derivatives of traditional carbon fibers (CFs), inherit several beneficial properties, such as high electrical conductivity and chemical stability. Compared with CFs, CNFs offer a significantly larger specific surface area, higher surface reactivity, and improved flexibility. These features make CNFs particularly attractive for feasible application in electrochemical energy storage systems. CFs generally provide good mechanical strength and structural stability. CNFs, although mechanically weaker, offer higher electrical conductivity and porous networks that facilitate ion diffusion and improve interfacial contact between the electrode and electrolyte. Moreover, CNFs can partially fulfill the structural roles of CFs when used as a supporting framework in composite electrodes. In this sense, CNFs not only retain the key advantages of CFs but also introduce new benefits at the nanoscale, offering a promising and multifunctional platform for the development of advanced battery electrodes.

CNFs have advantages such as high specific strength, high specific modulus, low coefficient of linear expansion, and low density. CNFs are commonly used for the reinforcement of high-performance resin-based composites. CNFs play important roles in aerospace, transportation, energy, medical and other fields. Due to its graphitic structure, the surface of CNFs exhibits non-polarity and chemical inertness, making it difficult to bond well with the matrix resin. Therefore, the interface bonding between the CNFs and the matrix has become a difficulty and bottleneck in the development of CNFs. Therefore, scholars have conducted extensive research from the perspective of mechanisms. In terms of chemistry, active functional groups are introduced on the surface of fibers to improve the wettability and chemical adhesion of the resin on the surface of the fibers. Physically, etching or introducing other particles on the surface of CNFs can improve their roughness, increase their specific surface area, and thus strengthen the mechanical interlocking between fibers and resins.⁴⁹ The method of introducing functional nanomaterials onto the surface of CNFs has good designability. In addition, the introduction of nanomaterials is more suitable for large-scale preparation due to their ease of processing. Therefore, it is of great significance to study the introduction of nanomaterials into the surface of CNFs to enhance the interfacial bonding of CNFs.⁵⁰ Meanwhile, CNFs materials are helpful in improving the electrochemical performance of silicon-based materials. CNFs can effectively alleviate the volume expansion of silicon materials during charging and discharging processes. For example, Wang et al. used F127 as a pore-forming agent to prepare high-silicon-content silicon/porous carbon nanofiber (Si/PCNF) self-supporting electrodes through a simple electrospinning method.⁵¹ A continuous three-dimensional porous carbon nanofiber network can effectively alleviate the volume expansion/contraction of silicon during charging and discharging processes. Similarly, Qu and others did the same.⁵² A three-dimensional Si/C nanofiber composite film was prepared as a self-supporting electrode using a simple electrospinning technique using a mixture of nano silicon and polyaniline. The carbon nanofiber membrane generated by PAN can effectively disperse silicon nanoparticles and maintain their volume expansion during lithium/delithiation processes. Secondly, by crosslinking the wrapped CNFs with amorphous carbon, the intrinsic conductivity of SiNPs can be greatly improved. The uniform sealing of amorphous carbon and the strong mechanical elasticity of CNFs enable stable electrical contact of silicon nanoparticles during lithiation and dehydrogenation processes. In addition, CNFs can also serve as conductive additives. It is important to maintain a good electron conduction path for the nano silicon powder in the anode during the cycling process. For this reason, CNFs are often used as conductive additives to supplement carbon black. Carbon black is a spherical nano powder that provides point-to-point contact between silicon powders.⁵³ On the contrary, the CNFs can provide line-to-line contact between silicon particles, which increases the conductivity of the anode and improves electrochemical performance.⁵⁴ In addition, recent studies have shown that CNFs can improve the overall mechanical strength of silicon anodes through their three-dimensional network structure within the anode, thereby partially suppressing the adverse effects of silicon particle volume expansion.^55–57 In addition, due to the influence of CNFs on the size of the anode microstructure, CNFs may help reduce the tortuosity of electrolyte channels.⁵⁸ CNFs can not only be used as a conductive additive or as an active substance together with silicon materials, but also as a current collector to load silicon materials alone. 3D collectors with high specific surface area can be used to reduce local current density to achieve uniform lithium metal deposition. A large internal space is also beneficial for reducing the volume change of lithium metal during battery charging/discharging. As shown in Fig. 1, the advanced lithium metal anode current collector materials and their demonstrated electrochemical performance clearly validate these advantages. For different 3D collectors, a suitable material with a specific structure can directly determine the final performance of the battery. Therefore, in-depth research on different 3D collector materials is crucial for developing the next generation of high-energy-density LIBs.⁵⁹

Fig. 1 Advanced current collector materials of Li metal anodes and their electrochemical properties. Reproduced with permission from ref. 59. Copyright 2022 Elsevier.

CNFs in anode materials possess many advantages, including high axial strength and modulus, low density, superior specific performance, creep resistance under long-term stress, ultra-high-temperature resistance in non-oxidizing environments, excellent fatigue resistance, thermal and electrical conductivity intermediate between non-metals and metals, low thermal expansion coefficient and anisotropy, superior corrosion resistance, high electrical conductivity and electromagnetic shielding capability. However, they still present the following key challenges:

(1) Dispersion and reinforcement for special fiber electrode materials to produce homogeneous substrates with enhanced strength.

(2) Optimization of hot-pressing processes to achieve the required substrate compactness and surface flatness, while resolving issues of resin migration, thickness uniformity, and surface roughness.

(3) Development of high-temperature carbonization protocols to improve substrate strength and prevent brittleness/fracture, while maintaining target polar resistivity post-carbonization.

Notably, recent advances in diverse carbon fiber-based materials have significantly impacted electrochemical energy storage systems. Table 2 summarizes representative carbon fiber architectures in LIBs. This comparative analysis reveals critical performance metrics across various CFs, including: reversible capacity (mAh g⁻¹), current density (A g⁻¹), capacity retention (%), cycle number, and electrolyte compatibility.

Table 2 Summary of CFs materials reported in LIBs

CFs	Composite	Reversible capacity	Current density	Capacity retention	Cycle number	Electrolyte	Ref.
Interconnected hollow CFs	Si	903 mAh g^−1	0.2 A g^−1	89%	100	1 M LiPF6	60
CFs	Si/po-C	997 mAh g^−1	0.2 A g^−1	71%	150	1 M LiPF6	61
CFs	Si/po-C	603 mAh g^−1	0.5 A g^−1	—	300	1 M LiPF6	61
CFs	ZnCo2O4	1180 mAh g^−1	—	—	100	1 M LiPF6	62
CFs	FeS2	—	0.2 A g^−1	90%	100	1 M LiPF6	36
CFs	C/MnO2 NWs	710 mAh g^−1	0.2 A g^−1	90%	300	1 M LiPF6	63

---

Currently, the use of CNFs-loaded silicon materials to produce adhesive-free independent anodes has become an effective method. Therefore, we can use reinforced CNFs as the current collector to load silicon particles, and use CNFs to reinforce silicon electrodes. Alternatively, traditional metal collectors combined with reinforced CNFs and carbon materials can be used to further reinforce silicon as the anode material for LIBs.

However, the composite material of silicon and CNFs did not reach its performance potential due to some Si nanoparticles being exposed on the fibers’ surface. These exposed Si nanoparticles can directly react with the electrolyte, leading to the formation of heavy SEI. Without carbon protection, silicon nanoparticles will detach from the surface of the nanofibers after several cycles, causing severe fiber damage and capacity loss.⁶⁴ Carbon coating is an effective way to improve the electrochemical performance of silicon electrode materials, so we propose to add a carbon coating and CNFs to the composite material to provide dual insurance for the silicon material.

In this review, we summarized several common structures and material compositions of reinforced CNFs-loaded silicon materials, and demonstrated the role of these aspects in the electrochemical performance of silicon–carbon electrodes, aiming to provide a reference for the preparation and development of silicon–CNFs composite materials in the future.

2. Silicon–carbon nanofiber composites and structures

The two major challenges hindering the further application of silicon anodes are the inevitable volume change (400%), which leads to the continuous reconstruction of the SEI film, and the low conductivity of Si, resulting in poor anode performance. To address these two issues, significant efforts have been made on Si-based anodes for LIBs. Nanoscale is one of the common choices, such as nanospheres,⁶⁵ nanowires,⁶⁶ and nanotubes.⁶⁷ Nanostructured or nanoscale electrodes can be designed in the form of nanowire arrays or ordered mesoporous structures to provide the necessary free space volume to accommodate expansion during alloying processes, thereby minimizing material stress and electrode pulverization to achieve greater reversibility and cycling stability. The nanoscale size allows for rapid relaxation of stress, making nanoparticles more resistant to fracture than bulk particles. The use of nanoscale morphology can also enhance magnification performance and specific capacity.⁶⁸ However, the preparation of nano Si, such as chemical vapor deposition, pyrolysis, and high-energy ball milling, is costly and not suitable for industrialization. Si–C composite materials are another competitive anode candidate material because C is strong enough to limit the volume change of Si,⁶⁹ including core–shell structure,⁷⁰ yolk–shell structure,⁷¹ Si–C “sandwich” structure,⁷² and C-coated Si composite nanowire structure.

Carbon can alleviate the silicon volume expansion effect and improve the conductivity of anode materials. According to their structural characteristics, silicon–carbon composite materials can be divided into three categories: dispersed structure, load structure, and coating structure.⁷³ By simple mechanical grinding and heat treatment, silicon nanoparticles can be uniformly dispersed in the carbon material matrix to obtain composite materials with a dispersed structure.⁷⁴ In the loading structure, silicon nanoparticles are loaded onto carbon carriers such as CNFs, carbon nanotubes, or graphene.^75–77 In the above two structures, the nano silicon particles and carbon material are separated from each other, and the point contact between them makes it difficult to achieve optimal conductivity. The coating structure transforms point contact into surface contact, greatly improving conductivity, and this is gradually becoming a hot topic. For example, phenolic resin or dopamine can be directly polymerized on the surface of silicon particles to form a core–shell structure. Another interesting development is the deposition of carbon layers on three-dimensional porous silicon materials through CVD technology. The large specific surface area of the material ensures more contact. Lithium-ion diffusion sites and channels are formed between silicon and carbon, endowing the material with excellent cycling performance and high energy density.

The separate use of dispersed, load, and coating structures can improve the cycling performance and energy density of silicon CNFs negative electrode materials. Two of these structures can be combined or the same structure can be used repeatedly to form a new structure that can be combined with carbon materials on the basis of a Si/CNFs composite, for example, dispersing or coating nano silicon materials on carbon materials and then loading them on CNFs. Composite structures obtained in this way can further enhance the cyclic performance and energy density of Si/CNFs composite materials.

Honeycomb structure refers to a periodic, regularly arranged porous network structure formed by active materials (e.g., silicon nanoparticles, Si NPs) and carbon-based materials (e.g., CNFs) in electrode materials, resembling a hexagonal honeycomb-like pore array.⁷⁸ The honeycomb structure keeps a highly ordered pore distribution with uniform pore sizes, and the pore walls are composed of carbon-based materials (e.g., CNFs, graphene). Active materials (e.g., Si NPs) are typically embedded in the pore walls or uniformly distributed within the pores. The ordered pore structure optimizes lithium-ion transport pathways, mitigates volume expansion stress, and provides mechanical support. The high specific surface area facilitates electrolyte infiltration and enhances interfacial reaction kinetics. Typical preparation methods include templating (e.g., sacrificial templates, three-dimensional printing), self-assembly techniques, or chemical vapor deposition (CVD) combined with etching processes.

Wrapping structure refers to a composite system where carbon-based materials (e.g., CNFs, carbon layers) coat the surface of active materials (e.g., Si NPs) in continuous layers or shells, forming “core–shell” or “yolk–shell” morphologies.⁷⁹ Carbon-based materials (e.g., amorphous carbon, graphene) fully or partially cover the surface of the active material, creating a physical isolation layer. Controlled voids (e.g., yolk–shell structures) are formed between the Si NPs core and the carbon shell, reserving space for volume expansion. The carbon shell suppresses the fracture of Si NPs caused by volume expansion and reduces irreversible growth of the SEI layer. The high conductivity of the carbon layer improves the overall electron transport efficiency. Typical preparation methods include CVD, hydrothermal methods, sol–gel processes, or electrospinning combined with carbonization.

Interwoven structure refers to a three-dimensional continuous network formed by the physical or chemical entanglement of CNFs and active materials (e.g., Si NPs).^27,80 CNFs and Si NPs are randomly distributed, forming a non-directional entangled network, where interfacial bonding strength depends on interactions between fibers and particles (e.g., van der Waals forces, chemical bonds). It involves high porosity and disordered pore distribution. The three-dimensional conductive network provides rapid electron/ion transport channels while buffering volume changes of Si through the mechanical flexibility of the fibers. The preparation process is simple and suitable for large-scale production. Typical methods include electrospinning with in situ loading, mechanical mixing (e.g., ball milling), or solution blending.

In LIBs, the classification of key structural types such as honeycomb, wrapping, and interwoven structures is primarily based on the orderliness of their microstructures, interface interaction mechanisms, and functional design objectives. The honeycomb structure is characterized by periodic ordered pores, with active materials embedded in regularly arranged carbon-based pore walls, optimizing ion transport and stress relief through uniform porosity. The wrapping structure achieves physical isolation of active materials from the electrolyte via continuous layered carbon-based coatings (e.g., core–shell or yolk–shell configurations), balancing conductivity and volume buffering. The interwoven structure manifests as the disordered three-dimensional entanglement of CNFs and active materials, relying on mechanical interlocking to provide conductive pathways and expansion buffering. The fundamental differences among the three kinds of structures lie in the degree of morphological order (ordered vs. disordered) and interface interaction modes (isolative coating vs. physical mixing).

2.1. Sandwich structure

A sandwich structure is a commonly used structure made of composite materials. Various silicon negative electrode materials have been combined with other materials to form sandwich structures as examples of negative electrodes for LIBs. For example, Li et al. prepared sandwich-structured C/SiOx@rGO materials through an alcoholysis process.⁸¹ The 3D highly conductive network composed of graphene sheets and amorphous carbon layers not only improves the electrode reaction kinetics of SiOx, but also ensures uniform local current density and electrode reaction on the SiOx surface. In addition, the Si–O–C bonds between SiOx and graphene enhance the particle adhesion on graphene sheets, resulting in excellent structural integrity during charging/discharging processes. Benefiting from the aforementioned structural advantages, the C/SiOx@rGO material exhibits significant electrochemical performance in terms of excellent cycling performance and superior rate performance. The C/SiOx@rGO electrode provides a reversible specific capacity of 890 mAh g⁻¹ and a high capacity retention rate of 73.7% after 100 cycles at a current density of 0.1C.

The use of CNFs-loaded silicon particles or silicon derivatives combined with carbon layers to construct a sandwich-structured three-dimensional conductive network not only improves electrode reaction kinetics, but also homogenizes local current density to ensure that electrode reactions occur uniformly on silicon materials, thereby homogenizing the volume changes on the surface of silicon particles or silicon derivatives and effectively alleviating the volume changes of nano silicon. Meanwhile, the dense carbon film on the surface can effectively block the direct contact between nano silicon and the electrolyte, effectively preventing the repeated growth of SEI film,⁸² ensuring excellent structural integrity during the charging/discharging process.

Zeng et al. used BC aerogel as the starting template and carbon source to prepare silicon-coated carbon intermediates by the sol–gel method and the magnesium thermal reduction method.⁸³ Subsequently, the intermediate was coated with dopamine and converted to carbon through reduction, resulting in a C/Si/C three-dimensional sandwich-network-structured negative electrode material with excellent cycling stability and reversible specific capacity. After 200 cycles at a current density of 0.5 A g⁻¹, the capacity of the composite material can still be maintained at 889 mAh g⁻¹. Li et al. successfully synthesized Si/C@C composite nanofibers with a sandwich structure through electrospinning and subsequent hydrothermal treatment.⁷³ The electrospinning process is as follows: 0.75 g of polyacrylonitrile (PAN) (molecular weight: 86 000) was dissolved in 10 mL of N,N-dimethylformamide (DMF) to form a polymer precursor solution. Subsequently, nano-sized silicon particles (∼50 nm) were added to the polymer solution at different PAN/Si weight ratios (6 : 1, 4 : 1, and 2 : 1). The suspension was vigorously stirred for 1 hour, followed by ultrasonic dispersion for 30 minutes to obtain a homogeneous brown suspension. For the typical electrospinning process, the suspension was transferred to a 1 mL syringe and fed through a metallic needle using a syringe pump at a flow rate of 5 μL min⁻¹. The resulting PAN/Si composite nanofibers were collected. The composite nanofibers were pre-oxidized in air at 240 °C for 6 hours to preserve their morphology during subsequent carbonization. The pre-oxidized nanofibers were then transferred to a tube furnace and carbonized at 600 °C under an Ar atmosphere for 8 hours, yielding the desired C/Si composite nanofibers.⁸⁴ This C/Si composite nanofibers structure not only buffers the volume expansion of silicon, but also avoids the direct exposure of precipitated silicon on the surface of nanofibers to the electrolyte, enabling the formation of a stable SEI film. The results showed that the cycling stability of the electrode was significantly improved. After 100 cycles at a current density of 1 A g⁻¹, its reversible capacity was 601 mAh g⁻¹, and the capacity retention rate was 80%. This indicates that the electrode has potential application value in high-capacity silicon-based negative electrode materials.

The most important issue in producing silicon-based independent batteries containing carbon substrates is to maintain good contact between silicon and conductive substrates. Considering this issue, Shen et al. adopted low-cost waste silicon materials.⁸⁵ The presence of certain gaps between the carbon shell and silicon particles, and the flexibility of carbon, provides a buffer space for the volume change of the silicon particles. This ensures good contact between the silicon particles and the current collector. A thin carbon layer is formed around the silicon particles through ball milling process.

HCC–M–Si was synthesized using the obtained M–Si as the raw material. Specifically, M–Si is first coated with SiO₂ as a hard template for subsequent resin coating. Then the resin is carbonized after heat treatment. After etching off SiO₂, a void space is created between the Si particles and the carbon shell to accommodate Si volume expansion, ensuring a stable electrode structure and thus ensuring good contact between the Si active material and the collector.

In order to further improve the energy density of HCC–M–Si electrodes, a lightweight and low-cost CNFs network was used to replace the copper foil collector, polymer adhesive, and conductive additive used in traditional LIBs. Independent HCC–M–Si/CNF electrodes were prepared by vacuum filtration. Vacuum filtration is a simple (minimal equipment), fast (with only one program), low-cost, and “green” (involving benign solvents such as water and ethanol) method that has been widely used in the preparation of nanocomposite membranes and has made significant progress in improving mechanical and functional performance.⁸⁶ TEM images of the CNFs are shown in Fig. 2a–c. The low-magnification transmission electron microscopy images (Fig. 2a) and high-magnification transmission electron microscopy images (Fig. 2b) show the tubular structure of the CNF. As shown in Fig. 2c, CNF exhibits highly graphitized properties after high-temperature treatment. This graphite property ensures the good conductivity of CNFs. The top and bottom surfaces of the HCC–M–Si/CNF electrodes are covered by an additional CNFs network layer, with the aim of trapping HCC–M–Si within the CNFs network. The preparation process is shown in Fig. 3a. Firstly, a small amount of CNFs is filtered to obtain a thin layer of CNFs at the bottom, and then a mixture of HCC–M–Si and CNFs is filtered at the top as the main part of the electrode. Finally, another thin layer of CNFs is filtered at the top of the electrode to completely encapsulate the HCC–M–Si/CNF mixture, and the weight percentage of HCC–M–Si in the entire HCC–M–Si/CNF electrode is ∼60%. The SEM image of the electrode surface is shown in Fig. 2d, and we can observe that the surface is completely covered by CNFs. The length of the CNFs exceeds 100 μm. They entangle with each other to form a network, which helps to trap HCCM Si particles inside.

Fig. 2 Characterization and electrochemical performance of CNFs. (a) Low-magnification TEM image of CNFs. (b) TEM image of the open end of one CNF. (c) HRTEM image of the area marked by a red square in (b). It shows the graphitic nature of the CNF. (d) SEM image of the surface of the HCC–M–Si/CNF freestanding electrode. (e) Electrochemical performance of the freestanding CNFs electrode at a current density of 0.4 A g⁻¹. Reproduced with permission from ref. 85. Copyright 2018 Elsevier.

Fig. 3 (a) Schematic illustration of the fabrication process of a freestanding HCC–M–Si/CNF electrode. (b) Cycling performance comparison of freestanding HCC–M–Si/CNF and M–Si/CNF electrodes. Reproduced with permission from ref. 85. Copyright 2018 ACS Nano Article.

The cycling performance of a HCC–M–Si/CNF independent electrode at 0.4 A g⁻¹ is shown in Fig. 3b, and the capacity is calculated based on the total mass of the electrode. The first reversible capacity was 1595 mAh g⁻¹, which remained at 1015 mAh g⁻¹ after 100 cycles. Meanwhile, the coulombic efficiency of the HCC–M–Si/CNF self-supporting electrode increased from 62% in the first cycle to over 98% in the 20th cycle, and remained above 99% in the 100th cycle, demonstrating a stable reversible electrochemical reaction. At a current density of 0.4 A g⁻¹, based on the total mass of the electrode, the HCC–M–Si/CNF electrode reached over 1000 mAh g⁻¹ after 100 cycles. Compared with traditional electrodes, the lightweight and independent HCC–M–Si/CNF electrode significantly improves electrode energy density by 745%.

A M–Si/CNF self-supporting electrode was manufactured according to the same procedure as HCC–M–Si/CNF, and also provides cycling performance for comparison. It was found that the reversible capacity significantly decreased from 1078 mAh g⁻¹ in the first cycle to 349 mAh g⁻¹ in the 15th cycle. This is due to the pore evolution of Si particles during the cycling process, as well as the subsequent large volume changes of the particles, as shown in Fig. 4b–d, which cause the particles to lose contact with the CNFs network. Without the use of adhesive in traditional LIBs, the volume change of M–Si particles leads to faster capacity degradation of independent electrodes compared to traditional electrodes. On the contrary, the void space between the Si core and the carbon shell in HCC–M–Si can adapt well to the volume change of Si during the cycling process and help maintain the contact between the HCC–M–Si and the CNFs network, thereby maintaining the capacity of Si. The independent HCC–M–Si/CNF electrode not only provides a strategy to further improve the energy density of HCC–M–Si, but also confirms the function of the graded carbon coating structure in providing good contact between the Si and the collector.

Fig. 4 Post-cycling analysis of HCC–M–Si. (a) Schematic illustration showing the morphology evolution of M–Si and HCC–M–Si after cycling. (b–d) TEM images of M–Si after one cycle (b), 10 cycles (c), and 50 cycles (d). (e) STEM image of HCC–M–Si particle after 50 cycles. (f) EELS mapping of elements C and Si in the HCC–M–Si particle. (g) EELS spectrum of the HCC–M–Si particle.⁸⁵ Copyright 2018 ACS Nano Article.

2.2. Interwoven structure

An interwoven structure is another commonly used structure for designing negative electrode materials for LIBs. In a well-designed architecture, interwoven materials can not only construct highly conductive and flexible networks, but also serve as pillars, further increasing porosity and effectively promoting electron and ion transport. The excellent elasticity of CNFs can suppress the volume change during the charging and discharging processes of silicon materials, ensuring that lithium ions can easily penetrate the composite negative electrode and have good mechanical stability when reacting with silicon. Si/CNFs composite materials with self-assembling interwoven layered structure keep high conductivity, rich porosity, and excellent scalability. In addition, self-assembled interwoven structures can significantly improve the uneven distribution of silicon and carbon, thereby significantly enhancing the electrochemical performance of silicon anodes.

Bai et al. designed Si/CNFs composite materials through UV-ozone surface modification and electrostatic self-assembly.⁸⁰ The obtained Si–CNF composite material containing 70.0 wt% SiNP exhibits excellent cycling performance, with a coulombic efficiency of 74.8% in the first cycle and a discharge specific capacity of 1063 mAh g⁻¹ in the 400th cycle. On the basis of the above research, Bai et al. proposed a dual protection strategy,⁸⁷ which synthesized high-performance Si–CNF@C through two-step electrostatic self-assembly and subsequent pyrolysis. High CMC content acts more like a dispersant for modified SiNP rather than a binder, and is then used as a precursor for sintering amorphous carbon to achieve the encapsulation of the entire SiNP. Under the action of electrostatic attraction and chemical bonding, CNFs and amorphous carbon can effectively protect SiNPs from dispersion and contact problems. In the case of dual protection heterostructures, this new type of Si–CNF@C electrode exhibits highly reversible capacities of 1200 mAh g⁻¹, 982 mAh g⁻¹, and 849 mAh g⁻¹ after 100, 500, and 1000 cycles at 0.5 A g⁻¹, respectively. The long-term cycling stability is equivalent, with a capacity loss of 0.036% per cycle in 1000 cycles.

2.2.1. UV-ozone surface modification followed by electrostatic self-assembly⁸⁰. Electrostatic self-assembly is one of the effective techniques for preparing uniformly distributed silicon–carbon composite materials. Theoretically, it is believed that any molecule with opposite charges can be assembled into a film using this method without being limited by sample size and shape. Compared with other silicon–carbon composite material preparation technologies, electrostatic self-assembly has the advantages of a wide range of material sources, simple instruments and equipment, safety and environmental protection, and convenient and efficient large-scale production. The preparation of silicon–carbon composite materials using electrostatic self-assembly mainly involves the mutual attraction of silicon–carbon materials or polyelectrolytes with opposite charges in solution to form composite materials. The main driving force for electrostatic self-assembly into films is the electrostatic attraction between opposite charge components, while the repulsive force of the same charge can alleviate the aggregation phenomenon of materials.

A simple and rapid modification pathway (i.e., UV-ozone exposure) could primarily induce carboxyl groups on the outermost surface of CNFs within 20 minutes or less. The electrostatic self-assembly process is shown as follows. 200 mg of CNFs (150 nm in diameter) were subjected to UV-ozone exposure at 150 °C for 20 min with a wavelength of 253.7 nm by using a UVO-cleaner system. Thus, UV-ozone-modified CNFs (o-CNFs) were directly yielded in a simple and dry way. For comparison, conventional acid treatment was also conducted for CNFs. Pristine CNFs were soaked in a concentrated sulfuric and nitric acid solution (H₂SO₄ : HNO₃ 3 : 1 v/v), magnetically stirred for 12 h at 60 °C, and repeatedly rinsed with deionized (DI) water. Finally, acid-modified CNFs (a-CNFs) were obtained after filtering through 0.22-μm Teflon filter paper and vacuum drying at 50 °C for 12 h. For surface functionalization with amino groups, SiNPs with an average size of about 50 nm were firstly refluxed in a piranha solution (H₂SO₄ : H₂O₂ 3 : 1 v/v) at 80 °C for 1 h. After washing with DI water and vacuum drying at 50 °C for 12 h, 500 mg of the pretreated SiNPs was dispersed into 100 mL of anhydrous ethanol in an ultrasonic bath for 0.5 h. Subsequently, 0.5 mL of aminopropyltriethoxysilane (APTEs, 98%) was poured into the solution above and stirred at 70 °C for 12 h. Afterwards, APTEs-modified SiNPs (Si–APTEs) were produced by repeatedly rinsing with ethanol, filtering through 0.22 μm Teflon filter paper, and drying under vacuum at 50 °C for 12 h. The Si–APTEs were respectively dispersed with o-CNFs or a-CNFs using an atomic ratio of 1 : 1 in DI water under ultrasonication for 5 h. The resulting suspension solutions were dried to obtain Si–o-CNFs and Si–a-CNFs composites containing o-CNFs and a-CNFs, respectively.⁸⁰

In addition to significantly reducing processing time, it has been reported that dry UV-ozone treatment can maintain the integrity of the internal structure without being contaminated or structurally damaged. To our knowledge, this method of dry modification of CNFs has hardly been applied in the field of energy storage. The choice of CNFs is based on their superior mechanical performance and low cost. Therefore, negatively charged –COOH groups can be created on CNFs in a simple and effective manner. Under electrostatic self-assembly, the prepared silicon–carbon composite material containing 70.0 wt% SiNPs combines with the positively charged NH₂ group on the SiNPs. After 400 cycles, it exhibits a reversible capacity of 1063 mAh g⁻¹ at 0.5 A g⁻¹, demonstrating a significant cycling advantage. This improvement in performance demonstrates the enhanced interface connection between SiNPs and CNFs, as well as the mechanical strength provided by UV-ozone modified CNFs to maintain electrode integrity without the need for additional carbon coating or heat treatment.

Si–o-CNFs were prepared via UV-ozone treatment at 150 °C for 20 minutes, introducing carboxyl groups onto the CNF surface. Si–a-CNFs were obtained through a conventional acid oxidation using a H₂SO₄/HNO₃ (3 : 1, v/v) mixture under stirring at 60 °C, followed by water washing, filtration, and vacuum drying, resulting in carboxyl-functionalized CNFs.

The synthesis route for Si–o-CNFs composite materials is shown in Fig. 5. SiNPs are first functionalized by grafting amino groups (–NH₂) onto the surface of APTEs to form positively charged surfaces (–NH³⁺) in aqueous solution. Meanwhile, CNFs are exposed to ultraviolet ozone, and ozone (O₃) and its decomposition products (O₂ and O) can react quickly with carbon atoms through oxidation and ozone decomposition under dry conditions. Therefore, carboxyl groups (–COOH) appear on CNFs with negatively charged surfaces (–COO). Then, –NH₂ can easily combine with –COOH through electrostatic self-assembly to increase the dispersion and adhesion ability of SiNPs in the electronic conductive CNFs framework.

Fig. 5 Schematic preparation process for Si–o-CNFs composites. Reproduced with permission from ref. 80. Copyright 2021 Elsevier.

Fig. 6 shows SEM images of the original Si–CNFs, Si–o-CNFs, and Si–a-CNFs composite materials. For the original Si–CNFs, the smooth CNFs separate from the aggregated SiNPs, resulting in almost no interaction between SiNPs and CNFs. In the case of Si–a-CNFs composite materials, the surface of the CNFs is relatively rough, and some SiNPs are embedded on the CNFs in a porous structure. On the contrary, Si–o-CNFs composite materials exhibit rougher CNFs surfaces and more SiNPs are dispersed in CNFs clusters. Therefore, it proves that the contact area between SiNPs and CNFs in Si–o-CNFs samples is larger than that in Si–a-CNFs samples. This is consistent with the observed Raman spectroscopy results mentioned above. However, some SiNPs aggregate together, even with more connections with modified CNFs under UV-ozone exposure. This is due to the presence of more excess –NH₂ groups in Si–o-CNFs composites with a high content of SiNPs (70.0 wt%). From the perspective of matching the ratio of –NH₂ and –COOH groups, further research is being conducted to improve the uniform distribution of silicon carbide composite materials. In summary, the UV-ozone exposure of CNFs effectively promotes the contact and distribution of SiNPs and CNFs in porous and interconnected networks.

Fig. 6 SEM images of the (a) original Si–CNFs, (b) Si–a-CNFs, and (c) Si–o-CNFs. Reproduced with permission from ref. 80. Copyright 2021 Elsevier.

Fig. 7 shows the charge–discharge curves and electrochemical cycling results of pristine Si–CNFs, Si–o-CNFs, and Si–a-CNFs composite anodes at a current density of 0.5 A g⁻¹ in the voltage range of 0.01–1.5 V. The specific capacity of composite electrodes is determined by the mass of the active substance. From Fig. 7(a), it can be seen that for all composite negative electrodes, the first discharge curve shows a long plateau below 0.2 V, and an inclined plateau at 0.25–0.5 V during charging, corresponding to the lithiation potential of crystalline silicon and the lithiation process of lithium silicon alloy. Starting from the second cycle, the charge–discharge curve shows that the silicon–carbon composite negative electrode has the characteristics of amorphous silicon and carbon. The initial discharge specific capacity of the pristine Si–CNFs composite negative electrode is 1672 mAh g⁻¹, the charging specific capacity is 1177 mAh g⁻¹, and the initial coulombic efficiency is 70.4%.

Fig. 7 (a) The initial two charge–discharge curves and (b) cycling performance of electrochemical cells using pristine Si–CNFs, Si–o-CNFs, and Si–a-CNFs composite materials at 0.5 A g⁻¹. Reproduced with permission from ref. 80. Copyright 2021 Elsevier.

From Fig. 7(b), it can be seen that the coulombic efficiency of the pristine Si–CNFs composite negative electrode can be maintained between 97% and 100% during long-term cycling. However, the capacity decay of the pristine Si–CNFs composite negative electrodes is rapid. After 200 cycles, the discharge specific capacity decreased to 557 mAh g⁻¹, and compared with the second discharge specific capacity (1179 mAh g⁻¹), the capacity retention rate was only 47.2%. This result indicates that the SiNPs and CNFs in the directly mechanically mixed silicon–carbon composite material are unevenly dispersed, and the volume change of the silicon–carbon composite material is uncontrollable during the charging and discharging process, leading to a decrease in its cycling performance. Especially, after 400 cycles, the discharge capacity of the pristine Si–CNFs composite negative electrode is only 182 mAh g⁻¹, which means that SiNPs are basically ineffective in this battery.

In contrast, Si–a-CNFs and Si–o-CNFs composite anodes prepared by surface modification and electrostatic self-assembly techniques exhibit good cycling performance. The Si–a-CNFs composite electrode formed by electrostatic self-assembly of a-CNFs and Si–NH₂ has an initial discharge specific capacity of 2130 mAh g⁻¹, a reversible charging specific capacity of 1520 mAh g⁻¹, and an initial coulombic efficiency of 71.4%, which is higher than the reversible specific capacity of the pristine Si–CNFs composite electrode. Compared with the second cycle discharge specific capacity, after 200 cycles, the capacity retention rate can still reach 53.0%, while after 400 cycles, the discharge specific capacity decreases to 626 mAh g⁻¹. The initial discharge specific capacity (2218 mAh g⁻¹) and charging specific capacity (1658 mAh g⁻¹) of the Si–o-CNFs composite negative electrode were further improved, with an initial coulombic efficiency of 74.8%. After 400 cycles, the ultra-high discharge specific capacity of 1063 mAh g⁻¹ was still maintained. The stable electrochemical cycling performance indicates that UV-ozone surface-modified CNFs combined with electrostatic self-assembly technology mechanically confine Si–NH₂ in a continuous conductive carbon network, effectively suppressing and alleviating the volume change of the silicon material during charge and discharge processes, thereby improving the cycling stability of silicon–carbon composite electrodes.

Fig. 8 shows the rate plot of a Si–o-CNFs composite negative electrode at different current densities. It can be seen that at current densities of 0.1, 0.5, 1, and 2 A g⁻¹, the average discharge specific capacities of Si–o-CNFs composite anodes are 1794, 1281, 932, and 768 mAh g⁻¹, respectively. When the current density recovers from 2 A g⁻¹ to 0.1 A g⁻¹, the discharge specific capacity increases to 1698 mAh g⁻¹. Subsequently, as the current density further increased to 0.5 A g⁻¹, the specific capacity fluctuated slightly, but a high discharge specific capacity of 1123 mAh g⁻¹ could still be achieved after 250 cycles, with a small difference from the specific capacity before high-rate charging and discharging. Therefore, batteries using Si–o-CNFs composite materials as negative electrodes will not be damaged due to high rate operation, which means that the corresponding negative electrodes have good rate and structural stability. Overall, the UV-ozone surface modification method not only promotes the contact and distribution of more SiNPs in the conductive carbon network during the electrostatic self-assembly process of SiNPs and CNFs, but also improves the cycling stability of the silicon–carbon negative electrode.

Fig. 8 Rate capability, discharge capacity, and coulombic efficiency of a Si–o-CNFs electrode at current rates of 0.1, 0.5, 1, 2, 0.1, and 0.5 A g⁻¹ along with cycle number. Reproduced with permission from ref. 80. Copyright 2021 Elsevier.

2.2.2. Interwoven silicon–carbon negative electrode with secondary electrostatic self-assembly⁸⁷. One-step electrostatic self-assembly is not enough to improve the long-term cyclic performance of silicon–carbon composites. In order to further improve the distribution uniformity of silicon–carbon composites, a dual protection strategy is proposed in this part. Silicon–carbon composites with high specific capacity and high cycle life are prepared by secondary electrostatic self-assembly and heat treatment technology. On the basis of the first electrostatic self-assembly, the –COOH group in high-concentration CMC electrostatically interacts with the SiNPs at the –NH₂ sealing end, and the silicon is adhered to the CMC chain. In addition, the surface-modified CNFs with carboxyl groups are the same as the carboxyl groups of CMC and electrostatic repulsion occurs, thus promoting the distribution uniformity of carbon materials. Therefore, in the case of uniform distribution of carbon materials, –NH₂-functionalized SiNPs adhere to the CNFs–CMC frame on multiple sides to form a uniform interwoven structure, which will further enhance the structural stability of the silicon–carbon composite electrode. Due to the multi-faceted adhesion of SiNPs, the entire SiNPs are completely embedded in the crisscrossed CNFs and amorphous carbon conductive network, which can achieve full electrical contact of SiNPs on the basis of preventing SiNPs aggregation, thus effectively improving the stability of the SEI film and increasing the electrochemical cycle life.

Fig. 9 shows the preparation process for a Si–CNF@C composite. First, the surface of SiNPs was modified by APTEs to produce a positively charged amino group (Si–NH₂) on the surface of the silicon particles. At the same time, dry CNFs were exposed to UV-ozone irradiation for 20 min to produce negatively charged carboxyl groups (–COOH). Subsequently, the positively charged amino groups on the SiNPs were dispersed and adhered in the conducting CNFs–COOH network by the first electrostatic self-assembly technique to form the Si–CNFs composite material. The CMC polymer with carboxyl group and Si–NH₂ were electrostatically self-assembled in a pH 3 buffer solution for a second time, and CNFs–COOH was electrostatically excluded to form a Si–CNF–CMC composite with an interwoven structure. Therefore, the Si–CNF–CMC composite is obtained by the multi-sided adhesion of SiNPs on the CNFs or the CMC chain. Finally, the Si–CNF–CMC material is directly coated on copper foil and heat-treated in an argon atmosphere to convert the CMC polymer into amorphous carbon. SiNPs are completely encapsulated in a carbon nanofiber–amorphous carbon conducting network, resulting in Si–CNF@C composite electrodes. Notably, Si–CNF@C composite electrodes are used directly to assemble button batteries without the use of any additional additives or electrode preparation steps.

Fig. 9 Schematic illustration of the synthetic procedure for a Si–CNF@C electrode. Reproduced with permission from ref. 87. Copyright 2021 Elsevier.

The effect of the second step of electrostatic self-assembly and subsequent heat treatment on the surface morphology of Si–CNFs is shown in Fig. 10(a–c). Fig. 10(a) shows that some SiNPs are embedded on CNFs in Si–CNFs composite materials formed through the first self-assembly step. However, significant aggregation of SiNPs still exists. On the contrary, for the Si–CNF–CMC composite material obtained after the second electrostatic self-assembly step, SiNPs are uniformly dispersed and widely connected with CNFs and CMC, without obvious aggregation (Fig. 10(b)). Therefore, polymer CMC as a dispersant can effectively promote the multi-sided adhesion and uniform distribution of SiNPs in the entangled porous CNFs–CMC framework. Subsequently, after heat treatment of Si–CNF–CMC, the distribution of SiNPs in Si–CNF@C composite material seems to have increased (Fig. 10(c)). This growth can be explained by the complete or even coverage of amorphous carbon on SiNPs embedded in entangled CNFs. Therefore, both amorphous carbon and CNF covering different sides of SiNPs can embed the entire SiNPs into a continuous conductive network.

Fig. 10 (a) SEM images of Si–CNFs, (b) Si–CNF–CMC, and (c) Si–CNF@C composite materials. (d) TEM micrograph of Si–CNF@C materials and (e) HRTEM images obtained from rectangular regions on corresponding TEM images. Reproduced with permission from ref. 87. Copyright 2021 Elsevier.

The microstructure of Si–CNF@C is shown in Fig. 10(d and e), which shows the TEM and corresponding HRTEM images obtained from the rectangular region of the TEM image. It can be clearly seen that Si–CNF@C materials have heterogeneous nanostructures. The crystal plane spacing measured on the HRTEM image at 0.312 nm is very consistent with the d-spacing of the (111) plane of crystalline silicon. Combining high-angle annular dark field (HAADF) images and corresponding elemental spectra, these results reveal a heterogeneous nanostructure composed of CNFs and SiNPs coated with amorphous carbon. The formed conductive network can effectively provide chemical interactions and physical constraints on SiNPs, preventing direct contact between SiNPs and the electrolyte during cycling.

Fig. 11(a) shows the electrochemical cycle test diagram of the Si–CNFs, Si–CNF–CMC and Si–CNF@C composite negative electrodes at a current density of 0.5 A g⁻¹. The specific capacity of each silicon–carbon composite negative electrode is determined according to the total weight of the electrodes on the copper foil. The initial discharge capacity of the Si–CNFs composite negative electrode is 1774 mAh g⁻¹ and the coulombic efficiency is 74.8%. After 400 cycles, the specific discharge capacity of the Si–CNFs composite negative electrode decreased significantly. After 500 and 1000 cycles, the specific discharge capacity decreased to 621 and 141 mAh g⁻¹, corresponding to the second discharge specific capacity (1397 mAh g⁻¹), and the capacity retention rate of the silicon–carbon composite negative electrode decreased to 44.5% and 10.1%, respectively. Compared with the Si–CNFs composite negative electrode, the initial specific discharge capacity of the Si–CNF–CMC composite negative electrode is 1782 mAh g⁻¹. Although the specific discharge capacity of the Si–CNF–CMC composite electrode is slightly smaller than that of the Si–CNFs composite electrode during the first 400 cycles, the cycle stability of the Si–CNF–CMC composite electrode is significantly improved after 1000 cycles. After 500 and 1000 cycles, the specific discharge capacity can be maintained at 795 and 468 mAh g⁻¹. Compared with the specific capacity of the second discharge (1489 mAh g⁻¹), the capacity retention rate of the Si–CNF–CMC composite negative electrode can reach 53.4% and 31.4%, respectively. The electrochemical stability of the Si–CNF@C composite negative electrode is further improved. In the first cycle, the specific discharge capacity is 1799 mAh g⁻¹ and the coulombic efficiency is 71.6%. In the long cycle process, the coulombic efficiency can be maintained in the range of 97–100%, with high electrochemical reversibility. It is worth noting that at 500 and 1000 cycles, the discharge specific capacity of Si–CNF@C composite negative electrode is stable at 982 and 849 mAh g⁻¹, and the capacity retention rate is as high as 73.8% and 63.8%. After 1000 cycles, the Si–CNF@C composite negative electrode has a capacity loss of only 0.036% per cycle, demonstrating excellent long-term cycle durability.

Fig. 11 (a) Long-term cycling properties of Si–CNFs, Si–CNF–CMC, and Si–CNF@C electrodes at 0.5 A g⁻¹; (b) the rate capability of a Si–CNF@C electrode at different rates of 0.1–2 A⁻¹; (c) the 1st, 2nd, 20th, and 30th CV curves of a Si–CNF@C electrode at a scan rate of 0.1 mV s⁻¹ in the potential window of 0.01–1.5 V, and the detailed CV profile for the zoomed-in region of the first lithiation in the inset. Reproduced with permission from ref. 87. Copyright 2021 Elsevier.

Fig. 12 shows schematic illustration of structural evolution and Li-ion transport process of Si–CNF@C electrodes. The structural evolution and lithium-ion transport process of the composite negative electrode are presented to explain the excellent performance of the composite negative electrode in the lithiation and delithiation processes. Firstly, during the second electrostatic self-assembly process, SiNPs maintained good adhesion by electrostatic attraction and chemical connection with a large number of –COOH groups on the CMC polymer. According to reports, the chemical bonds between the surface of the SiNPs and CMC can withstand reversible lithiation and delithiation reactions. Secondly, a robust conductive network was constructed to maintain lithium-ion balance and ensure stable, fast ion transport rates in the Si-CNF@C electrode. The inherent conductivity of SiNPs is significantly improved by being encapsulated in a framework of cross-linked entangled CNFs and amorphous carbon. Especially, research has found that CMC polymers easily capture silicon particles in silicon–carbon negative electrode materials, forming a uniformly distributed interwoven structure. In this case, the amorphous carbon formed by carbonization of the CMC polymer uniformly covers the surface of the SiNPs. The uniform shell of amorphous carbon and the strong mechanical flexibility of CNFs enable SiNPs to maintain stable electrical contact during lithiation and delithiation processes. In addition, the pores in the composite negative electrode can effectively buffer the volume change of SiNPs during cycling. In the second electrostatic self-assembly, CMC connects to the surface of the SiNPs through electrostatic interactions and chemical bonds, forming polymer bridges between particles. Therefore, the intergranular gaps induced by carbonization of CMC will promote the separation of internal stresses related to the volume changes of the SiNPs. This makes it possible to from Si–CNF@C. The electrode has a complete structure and good reversibility during long-term cycling. Finally, the SiNPs are protected by both CNFs and amorphous carbon to avoid direct contact with the electrolyte, preventing the occurrence of side reactions and the repeated formation of SEI films. Therefore, this approach to maintaining structural integrity in Si–CNF@C composite materials can improve cycling performance and rate performance.

Fig. 12 Schematic illustration of structural evolution and Li-ion transport process of Si–CNF@C electrodes. Reproduced with permission from ref. 87. Copyright 2021 Elsevier.

2.3. Honeycomb structure

The honeycomb structure has a large specific surface area, abundant pores, and good stability.⁷⁸ Preparing Si–C composite materials with a honeycomb structure is an effective way to alleviate volume changes and improve electrochemical performance. Composite materials with good structural integrity can promote contact between active materials. In addition, the abundant pores in the honeycomb structure not only accommodate volume expansion,⁸⁸ but also increase the area between the active substance and the electrolyte.⁸⁹ In a honeycomb structure, when carbon nanofibers are combined with silicon material, the excellent conductivity of CNFs accelerates electron transport in electrochemical reactions and assists in buffering the volume change of Si in the carbon layer. Three-dimensional CNFs have a large specific surface area and good mechanical flexibility, which not only stabilizes the overall structure of silicon–carbon composite materials, but also effectively buffers the volume expansion of Si and rapidly transfers electrons. Meanwhile, the mesopores formed by the carbon layer facilitate the diffusion of Li⁺ ions.

Yao et al. successfully prepared CNFs/Si self-assembled composite films with a honeycomb structure using a one-step electrophoresis method.⁹⁰ This integrated anode electrode without adhesive or any other conductive additives greatly improves the electrochemical performance, providing a high reversible specific capacity of approximately 700 mAh g⁻¹. It has good cycling stability and excellent rate performance.

With the rapid development of Si–C fine structure design, many synthesis methods are frequently used, including high-energy ball milling (HEBM), thermal/chemical vapor deposition (TVD/CVD), atomic layer deposition (ALD), chemical/physical etching, RF magnetron sputtering, and electrospray. However, these methods often require high costs, complex methods, and harsh conditions, such as high processing temperatures, toxic atmospheres, or long production times, making the practical application of silicon-based anodes difficult.

Electrophoretic deposition (EPD) is commonly used to treat various coatings and independent objects, during which charged particles with small sizes (<30 μm) migrate to electrodes with opposite charges under an applied electric field in an appropriate suspension. Compared with the above-mentioned silicon-based anode method, EPD has the advantages of low cost, a simple and green process, uniform and controllable deposition, and diverse electrode shapes. In addition, the active material can be directly deposited on the collector (copper foil) through EPD technology, providing a large-scale manufacturing strategy and avoiding slurry preparation during subsequent coating processes. EPD has been proven to be an effective technique for preparing thin films, such as CNT/SiO₂ thin films coated on stainless steel substrates, titanium oxide/CNT composite coatings on metal substrates, manganese dioxide multi-walled carbon nanotube (MWCNTs) composite films with porous structures, and iron oxide films with spaced radial nanorods formed on carbon nanofiber scaffolds.

This section introduces the successful preparation of self-assembled CNFs/Si composite films with a honeycomb structure using a one-step EPD method. This adhesive-free integrated anode electrode without any other conductive additives greatly improves electrochemical performance, providing a high reversible specific capacity of about 700 milliampere hours, with good cycling stability and excellent rate performance.

To produce CNFs/Si composite films, surface modification of CNFs and silicon nanoparticles (SiNPs) is first carried out to achieve better particle dispersion and suspension stability. The main step is to reflux CNFs in a mixture of concentrated nitric acid and sulfuric acid (H₂SO₄ : H₂O₂, 1 : 2 v/v), and reflux SiNPs in a piranha solution (H₂SO₄ : H₂O₂, 1 : 2 v/v). Then, wash with deionized water and dry in a vacuum oven. Secondly, prepare suspensions of CNFs and SiNPs. Place two materials in ethanol and adjust the suspension pH using lithium hydroxide and hydrochloric acid. Finally, use the EPD method to produce CNFs/Si composite films. The process of preparing CNFs/Si composite films using the EPD method is shown in Fig. 13. Using a GEHN600e1.3 DC power supply, perform EPD under constant potential conditions with an applied voltage of up to 75 V cm⁻¹ and a deposition time of up to 10 minutes. Different applied voltages were also applied to study their effects on the morphology of the thin film. Connect copper foil (3 cm²) to the positive electrode as the working electrode and stainless steel (SS) to the negative electrode as the counter electrode. After deposition, the copper foil with a deposition layer is sequentially immersed in diluted HCl and DI water several times to remove residues, and then dried with Ar gas at 150 °C for 2 hours. As a comparison, pure CNFs and CNFs/Si were also dispersed in ethanol without Li₂B₄O₇ additive to form a stable suspension, and corresponding electrode films were prepared using the same EPD program.

Fig. 13 Schematic of the fabrication process for the CNFs/Si composite film by EPD method. Reproduced with permission from ref. 90. Copyright 2018 Elsevier.

Fig. 14 shows the SEM images of CNFs and SiNP surfaces before and after modification. As shown in Fig. 14a, pure CNFs have a relatively smooth surface with a diameter of approximately 200 nm. After acid treatment, the size is usually maintained, but the surface becomes relatively rough (Fig. 14b). Some CNFs with open ends can also be observed (illustrated in Fig. 14b), which is a typical feature of acid-treated CNFs. It is worth noting that pure CNFs only have a small amount of –COOH groups (2.20%), and after 2 hours of acid treatment, the carboxyl group content will increase to 6.55%. The average size of the SiNP is 50–100 nm (Fig. 14c), and there is no significant change in the size or morphology of the SiNP after surface treatment, as shown in Fig. 14d. XPS spectroscopy can further explain the surface state of the SiOH particles. As shown in Fig. 14e and f, before surface modification, the SiO₂ layer will be removed by HF etching method, so only one peak belonging to the Si group can be observed at ∼99 eV (Fig. 14e). After introducing the –OH group, a peak at 102.8 eV from the Si 2p spectrum (Fig. 14e) and a peak at 533.6 eV from the O 1s spectrum (Fig. 14f) can be observed, which belong to the Si–OH and –OH groups, respectively. The signal at 103.6 eV (as fitted by the brown curve in the figure) indicates the presence of residual oxide layers on the silicon nanoparticles, which should be attributed to the SiO Si groups generated by acid treatment. In addition, the peaks related to Si–O–H weakened or disappeared after sputtering, indicating that the surface of the particles had been successfully modified by hydroxyl groups. These functional groups are expected to help form a uniform suspension through repulsive forces.

Fig. 14 SEM images of CNFs before (a) and after (b) treatment with mixed acid, SEM images of (c) pure SiNPs, (d) Si–OH particles, high-resolution XPS spectra of Si–OH particles, (e) Si and (f) O SiOH. Reproduced with permission from ref. 90. Copyright 2018 Elsevier.

The corresponding cycling performance of the electrode is shown in Fig. 15. Pure CNFs membranes and CNFs/Si composites prepared without Li₂B₄O₇ only showed reversible capacities of 297 and 375 mAh g⁻¹, respectively. On the contrary, CNFs/Si membrane electrodes with honeycomb structure still maintain a reversible charging capacity of 670 mAh⁻¹ after 100 cycles, and the coulombic efficiency remains almost 100% after the first few cycles, far higher than pure CNFs prepared by EPD. The CNFs/Si composite electrode also exhibits good rate performance, as shown in Fig. 15b. As the current rate gradually increases from 400 mA to 2000 mA, the specific capacity decreases from 690 mA to 478 mA, 415 mA, 373 mA, and 358 mA, respectively. After cycling at different high current rates, when the current rate switches back to 400 mAh g⁻¹, the capacity can be restored to 660 mAh g⁻¹, indicating that the corresponding electrode has good structural stability. When calculating the mass-to-volume ratio, the effective active material is composed of CNFs and Si. If the mass of CNFs is excluded, the specific capacity of the electrode can be significantly increased, as based on TGA results, the actual mass ratio of the active material (Si) is only 22 wt%. The loading of Si during the EPD process is closely related to the condition of the suspension. If the concentration of Si is high, particles will deposit, resulting in uneven suspension and electrode film. However, during the EPD process, changing the corresponding parameters, such as the potential, pH, and the ratio of Si and CNFs, can clearly regulate the load. Therefore, in order to obtain a higher Si load on the electrode film, more experiments may be needed in the future and more parameters can be optimized.

Fig. 15 (a) Cycling performance of the CNFs/Si composite, CNFs/Si composite without the assistance of Li₂B₄O₇, and pure CNFs film at the current rate of 400 mA g⁻¹, (b) rate performance of the CNFs/Si composite film electrode. Reproduced with permission from ref. 90. Copyright 2018 Elsevier.

CNFs/Si composite film electrodes exhibit better electrochemical performance in terms of specific capacity, cycling stability, and rate capability, which can be attributed to their unique network structure and carbon conductive network. The nanoscale characteristics of Si particles in close contact with the CNFs network ensure complete lithiation reaction, thereby ensuring the high specific capacity of the electrode. Meanwhile, the porous structure formed during the self-assembly process can effectively buffer the volume change of Si caused by lithiation/delithiation, thereby producing excellent cycling stability and maintaining electrode integrity. In addition, carbon nanofiber networks can also prevent the aggregation of Si particles; compared with the control sample, the good adhesion between the porous matrix and Si and CNFs enhances lithium-ion and electron conduction, significantly promoting the electrode reaction kinetics and improving rate performance.

2.4. Wrapping structure

The wrapping structure is one of the most commonly used structures in LIBs, which transforms point contact into surface contact and greatly improves conductivity. Currently, most Si–C anodes only use carbon as the coating on Si particles. This carbon layer improves the charge transfer dynamics, but cannot improve the mechanical stability of the electrode structure.

According to research, porous carbon structures are more effective in adapting to changes in silicon volume than carbon-coated silicon anodes. In this study, a mixture of Si nanoparticles and carbon precursor polymers was incorporated into the CNFs felt collector fluid, and then carbonized to directly form a porous Si–C electrode. The entire electrode is synthesized using a one-pot carbonization process, thereby eliminating the polymer adhesive. The high Si load enhances the volume and weight energy density of the electrode. The obtained electrodes exhibit better capacity, recyclability, and rate performance, demonstrating the improved structural integrity.

This study reports a porous silicon–carbon anode used on a lightweight CNFs collector for LIBs.⁷⁹ This Si–C anode is synthesized through one-step carbonization of Si poly(acrylonitrile methyl acrylate) precursor directly deposited on carbon nanofiber felt. Carbon nanofiber collectors allow for higher loading of active materials, resulting in higher energy quality and area ratio. The obtained Si–C electrode exhibits excellent overall capacity, recyclability, and rate capacity.

The SEM image of the CNFs felt is shown in Fig. 16a, and the porous structure of the pad allows for high electrode material loading and better electrode current collection adhesion. According to observations, electrode materials coated thickly on CNFs can maintain good adhesion after carbonization. There are two reasonable explanations for achieving better adhesion: the CTE of CNFs mats may be similar to that of electrode materials, as they both contain carbon, and the porous structure of CNFs mats can alleviate some of the strain/stress generated by thermal expansion. Fig. 16b shows the surface of the Si–C electrode, which was intentionally damaged to reveal the CNFs beneath the electrode material. The cross-section shown in Fig. 16c shows the vertical structure of the Si–C electrode, partially penetrating into the CNFs felt. The active material is deposited into the CNFs felt, integrating the electrode and collector together to enhance interface integrity. The high-magnification image in Fig. 16d reveals the porous structure of the Si–C electrode.

Fig. 16 (a) SEM images of carbon nanofiber sheets, (b) intentionally ruptured Si–C electrodes to display the carbon nanofiber collector, (c) cross-section of Si–C electrodes, and (d) high-magnification image of porous electrode structures. Reproduced with permission from ref. 79. Copyright 2018 Elsevier.

Another advantage of this Si–C electrode is its high overall (including collector) capacity. Fig. 17a shows the voltage distribution for the 1st, 2nd, 20th, 60th, and 100th cycles. The first total discharge capacity was 728 mAh g⁻¹, and the first charging capacity was 519 mAh g⁻¹. Therefore, the irreversible capacity of the first cycle is 28.8%. It is worth noting that the proportion of CNFs felt (30 wt%) to the total capacity is less than 10% (CNFs mat capacity: 228 mAh g⁻¹). The total discharge and charging capacities in the second cycle are 781 mAh g⁻¹ and 848 mAh g⁻¹, respectively. After the second cycle, the coulombic efficiency rapidly increased to over 98%. Fig. 17b shows the total electrode capacity based on mass and area at a rate of 0.1C. After 100 cycles, the Si–C electrode can still maintain a total capacity of 650 mAh g⁻¹ or 3 mAh cm⁻².

Fig. 17 (a) Electrode capacity per mass of Si cycled at 0.1C and (b) Si charge rate performance of the Si–C porous electrode. Reproduced with permission from ref. 79. Copyright 2018 Elsevier.

To demonstrate the superiority of Si–C electrodes, cyclic testing was conducted in a glove box and the electrodes were extracted from the coin battery. The morphology of the surface and cross-section was evaluated using SEM and compared with the corresponding area of the uncarbonized cyclic precursor electrode. Fig. 18a and b show the surface morphology of the Si–C electrode and precursor, respectively, while Fig. 18c and d show the corresponding cross-sectional morphology. It is evident that the surface and entire structure (from a cross-sectional perspective) of the precursor electrode are covered by an SEI film. On the contrary, although the SEI film can also be seen in the Si–C electrode, its original structure can still be distinguished. This observation indicates that the carbon matrix of Si–C electrodes obtained through carbonization can promote the formation of stable SEI films, thereby improving recyclability.

Fig. 18 SEM images of the surface morphologies of (a) the porous Si–C electrode and (b) the precursor electrode and cross-section morphologies of (c) the porous Si–C electrode and (d) the precursor electrode. Reproduced with permission from ref. 79. Copyright 2018 Elsevier.

In this study, porous silicon–carbon anodes were synthesized on a lightweight CNFs current collector. The electrode exhibits good recyclability, rate capacity, and high total capacity. Our unique carbonization method directly synthesizes electrodes on the collector. The advantage of this Si–C electrode is its high structural integrity, as the entire electrode adopts an integrated carbon framework. The current results strongly indicate that the structural stability of the electrode is the key to achieving good electrochemical performance. Future work will focus on the uniform distribution of silicon particles in porous carbon frameworks. We will also study the pore size and pore size distribution of carbon frameworks. Finally, the carbonization technology will be optimized, including the type of polymer precursor, carbonization stage, and temperature.

Electrospinning has become a new method for designing silicon-based materials. The electrospinning-pyrolysis strategy enables precise control over the silicon distribution and carbon matrix crystallinity, which is critical for balancing conductivity and mechanical stability in Si/C composite anodes.²³ The porous CNFs network prepared by electrospinning can effectively alleviate the volume expansion and contraction of silicon during charge and discharge processes, and the porous structure can also provide a fast transport channel for lithium ions.

In this work, Li et al. designed a novel self-supporting electrode (represented as si@p-c/p-CF), composed of porous carbon fiber (p-CF) and Si nanoparticles coated with porous carbon (si@p-c).⁹¹ Prepared through a simple electrospinning technology, this carefully designed structure exhibits three main advantages: (1) the three-dimensional CFs network is a stable support skeleton for active materials and a good electron transfer channel; (2) porous CFs and porous carbon layers can effectively alleviate the volume expansion of Si, promote the formation of stable SEI membranes, and accelerate the transport of lithium ions; and (3) the prepared self-supporting electrode avoids the use of non-active components such as metal collectors and adhesives, which is beneficial for improving the weight energy density of the battery.

The preparation process for a p-CF electrode is shown in Fig. 19. The preparation is mainly divided into two steps: (1) Synthesis of porous carbon-coated silicon (denoted as si@p-c) nanoparticles using ZnO nanoparticles as hard templates and phenolic resin as carbon precursors; (2) Construction of self-supporting Si@/p-CF and (si@p-c)/P-CF electrodes through electrospinning of a homogeneous mixture containing the as-prepared si@p-c powder, polyacrylonitrile (PAN), and N,N-dimethylformamide (DMF).

Fig. 19 Schematic illustration of the preparation of Si_x/p-CF and (Si@p-C)_0.5/p-CF self-standing electrodes. Reproduced with permission from ref. 91. Copyright 2024 Elsevier.

The morphology and structure of Si/p-CF self-supporting electrodes with different silicon contents were studied using scanning electron microscopy and transmission electron microscopy. Fig. 20a shows the surface of the Si₁/p-CF electrode, which is the roughest due to the aggregation of high-concentration Si nanoparticles. As shown in Fig. 20b, a large number of Si aggregates are exposed on the surface of porous CFs, indicating insufficient encapsulation of Si nanoparticles within the fibers. As the content of Si nanoparticles decreases, the Si aggregates on the surface of the Si_0.5/p-CF fibers significantly decrease (Fig. 20c). The TEM image of Si_0.5/p-CF (Fig. 20d) shows good bonding between Si nanoparticles and CFs, with obvious pores inside the fibers. However, it can still be seen that some Si nanoparticles are directly exposed to the surface of Si_0.5/p-CF fibers, which may cause Si nanoparticles to detach from the fibers due to their volume expansion during cycling. As a comparison, the Si_0.25/p-CF electrode showed a smooth surface without Si particles (Fig. 20e). In addition, TEM images clearly show that Si nanoparticles are well coated within the fibers (Fig. 20f). The EDS elemental spectrum results of the Si_0.5/p-CF electrode in Fig. 20g further illustrate the uneven carbon layer on the Si surface. This may lead to the rupture of the thin carbon layer during lithiation/delithiation processes and cause damage to the electrode structure.

Fig. 20 SEM and TEM images of (a and b) Si₁/p-CF, (c and d) Si_0.5/p-CF, and (e and f) Si_0.25/p-CF. (g) EDS elemental spectrum of Si_0.5–p-CF. Reproduced with permission from ref. 91. Copyright 2024 Elsevier.

In order to determine the appropriate silicon content in the electrode, Si₁/p-CF, Si_0.5/p-CF, and Si_0.25/p-CF self-supporting electrodes were made separately. The cycling performance of Si₁/p-CF, Si_0.5/p-CF, and Si_0.25/p-CF self-supporting electrodes at 0.2 A g⁻¹ is shown in Fig. 21. The Si₁/p-CF self-supporting electrode has the highest silicon content and initial specific capacity, but it also has the worst cycling stability and cycling specific capacity. This is because most silicon nanoparticles are exposed to the surface of CFs, leading to severe capacity degradation. The cycling stability of Si_0.5/p-CF and Si_0.25/p-CF electrodes improved with the increase of carbon content because porous CFs reduce the volume expansion of silicon nanoparticles and increase the conductivity of electrodes. However, the capacity of Si_0.5/p-CF and Si_0.25/p-CF electrodes decreased after 35 cycles, which may be attributed to the limited ability of porous CNFs to further suppress Si volume expansion. This result indicates that simply increasing the carbon content cannot indefinitely improve the electrochemical performance of the electrode.

Fig. 21 Cycling performance of Si₁/p-CF, Si_0.5/p-CF and Si_0.25/p-CF self-supporting electrodes at 0.2 A g⁻¹. Reproduced with permission from ref. 91. Copyright 2024 Elsevier.

The cycling performance of the electrode was studied, and the results are shown in Fig. 22. After 100 cycles at 0.2 A g⁻¹, the Si_0.5/p-CF electrode exhibited a discharge specific capacity of 279.5 mAh g⁻¹. The significant decrease in capacity is mainly attributed to the volume expansion of silicon nanoparticles exposed to the outside of CFs during the cycling process, resulting in the crushing of silicon nanoparticles and subsequent structural damage. Compared to Si_0.5/p-CF electrodes (Si@p-c), the Si_0.5/p-CF electrode showed significantly improved cycling stability after 400 cycles, with a discharge specific capacity of 940 mAh g⁻¹. The porous carbon layer on the surface of the silicon nanoparticles can avoid direct contact between the silicon and the electrolyte, and alleviate the volume expansion of silicon during lithiation/delithiation processes. In addition, the pores within the Si-CNF@C composite anode effectively buffer volume changes of silicon nanoparticles (SiNPs) during cycling process. The strong adhesion between particles and porous CFs can prevent the active material from falling off and maintain the structural integrity of the self-supporting electrode even when bent or folded. However, in the first 10 cycles, the capacity of the Si_0.5/p-CF electrode significantly decreased, which may be related to the formation of the SEI film and electrolyte decomposition. Moreover, during the charging process, some lithium ions captured in the disordered structure of the hard carbon layer cannot escape, leading to an irreversible decrease in capacity. The Si_0.5/p-CF self-supporting electrode remained stable after 20 cycles, indicating the superiority of the designed composite material structure.

Fig. 22 (a) Cycling performance, (b) rate performance and (c) EIS curves of Si_0.5/p-CF and (Si@p-C)_0.5/p-CF self-supporting electrodes. Reproduced with permission from ref. 91. Copyright 2024 Elsevier.

Electrochemical impedance spectroscopy (EIS) measurements were conducted to further investigate the conductivity and electrochemical performance of the Si_0.5/p-CF and (Si@p-c)_0.5/p-CF self-supporting electrodes. As shown in Fig. 22b, compared with the Si_0.5/p-CF electrode, the Si_0.5/p-CF electrode has a smaller semi-circular diameter (related to charge transfer resistance) and lithium-ion diffusion resistance. Better conductivity may be attributed to the good cross-linked conductive network formed between the porous carbon coating structure and the CFs.

Compared to traditional adhesive-based electrode structures, this self-supporting silicon-based anode design provides improved safety and excellent mechanical strength/stability. At the same time, it minimizes the risk of electrode delamination, cracking, and potential short circuits, thereby reducing the possibility of thermal runaway and improving the safety of the battery. The implementation of dual protective layers is of great significance for suppressing the volume change of silicon-based anodes during cycling. The dual protective layer effectively acts as a barrier to prevent the expansion and contraction of silicon-based anode materials during electrochemical cycling. This stability helps to maintain the structural integrity of the anode and minimize physical and mechanical damage caused by volume changes. With a double-layer protective layer, the volume change of the silicon-based anode is greatly reduced. This leads to an improvement in cycling performance, as the anode can maintain a more stable electrode electrolyte interface and promote effective electron/ion transport during charging and discharging cycles. By reducing volume changes, the double protective layer helps to reduce stress and strain on the anode, which also extends the cycle life and improves the overall durability and lifespan of the battery.

In recent years, electrospinning technology has attracted widespread attention as an effective method for preparing Si/C composite nanofibers. Usually, Si/C nanofibers are prepared by electrospinning a polymer spinning solution containing highly dispersed Si nanoparticles (Si NPs) and then pyrolysis. Due to the excellent structural stability and high conductivity of carbon, this uniform composite structure can effectively buffer the volume expansion of silicon and provide excellent electron conduction during charging and discharging processes. Importantly, electrospinning makes the manufacturing of self-supporting nanofiber electrodes easy and provides satisfactory flexibility, thereby eliminating the need for adhesives and increasing the weight energy density of LIBs. On the other hand, this fiber electrode has excellent flexibility and is a promising candidate electrode for flexible batteries. CNFs can limit the volume change of electrode active materials within the fibers, avoiding the direct exposure of silicon deposited on the surface of the nanofibers to the electrolyte, reducing the occurrence of side reactions. The one-dimensional structure provides an effective channel for the rapid transmission of lithium ions and electrons.⁷³ However, simply combining silicon and carbon through electrospinning and pyrolysis cannot effectively limit the volume expansion of silicon particles. Further carbon coating provides an effective way to solve this problem. Chemical vapor deposition (CVD) is a commonly used method to coat the surface of silicon materials with a carbon layer. However, the CVD method always involves issues such as high-temperature equipment, hazardous gases, strict regulation, and uneven coating. In addition to the CVD method, the polymer deposition/pyrolysis strategy is another common way to coat carbon materials on Si/C nanofibers. Although impressive progress has been made in enhancing the cycling stability of Si/CNFs electrodes, most of these methods have only achieved limited results in the face of Si's volume expansion and unsatisfactory cycling life. In addition, creating a porous structure is one of the effective methods to regulate the volume expansion of Si/C composite materials.

This method is completed in two steps.⁹² Firstly, Si/C nanofibers (Si–CNFs) are fabricated using electrospinning technology. Then, a polydopamine layer is hydrothermally deposited on Si/C nanofibers for pyrolysis, resulting in the preparation of Si/C composite nanofibers with a carbon coating and hollow Si/SiOx (C–Si–CNF). The hydrothermal deposition/pyrolysis process not only achieves carbon coating on Si/C nanofibers, but also converts solid Si into hollow Si/SiOx. Due to the enhanced conductivity of the carbon coating, the carbon coating, hollow Si/SiOx, and SiOx shell reduce the volume expansion of Si, resulting in impressive structural stability and cycling life of the C–Si–CNF electrode as a LIB anode. In addition, with the optimization of silicon loading, the obtained C–Si–CNF electrode exhibits excellent flexibility and self-supporting characteristics, demonstrating its good application potential as a LIB binder-free self-supporting anode.

Generally speaking, the silicon content in composite materials should be proportional to the lithium storage capacity. However, the Si capacity of the Si/polymer solution is closely related to the spinnability and mechanical properties of Si/C nanofibers. Therefore, in order to obtain materials with good spinnability and strong mechanical properties, the Si content of Si/PAN was optimized. This experiment set the mass ratio of Si/PAN to 1 : 5 for further research.

The effect of hydrothermal deposition/pyrolysis on Si/C composite fibers is shown in Fig. 23. Pure CNFs have a smooth surface and uniform diameter. After the introduction of SiNPs, there is particle aggregation and protrusion on the surface of the CNFs, the surface becomes rough, and the diameter of the CNFs slightly increases. After hydrothermal deposition and pyrolysis, the surface becomes smooth and a carbon coating can be observed on the surface. It can be inferred that a carbon layer has been successfully coated on the surface of the Si/C composite fibers.

Fig. 23 SEM images of (a) CNF, (b) Si–CNF, and (c) C–Si–CNF, insets in (b) and (c) are corresponding optical images; schematic illustration of surface and cross sections for (a1 and a2) CNF, (b1 and b2) Si–CNF, and (c1 and c2) C–Si–CNF. Reproduced with permission from ref. 92. Copyright 2023 Elsevier.

For C–Si–CNF, this hierarchical porous structure and high specific surface area not only facilitate the diffusion of lithium ions, but also provide additional lithium storage sites. More importantly, the unique hollow porous structure is very beneficial for buffering the structural stress caused by the volume expansion of silicon.

Fig. 24a shows the cyclic performance of various fiber membranes. The specific capacity value is calculated based on the total weight of self-supporting electrodes. The discharge specific capacity calculated in the second cycle is 694.7 mAh g⁻¹ for the C–Si–CNF electrode and 799.5 mAh g⁻¹ for the Si–CNF electrode. The capacity retention rate of the C–Si–CNF electrode exceeded 100% (701.1 mAh g⁻¹) at the 100th cycle and reached 92.6% at the 300th cycle, which means that the decay rate of each cycle in the 300th cycle is 0.025%. However, the specific capacity of Si–CNF electrodes showed significant attenuation after 20 cycles. The capacity retention rate of the Si–CNF electrode is only 54.8% after 100 cycles and 46.9% after 300 cycles. It is worth noting that the cycling stability of C–Si–CNF electrodes is superior to most previously reported Si/C composite fiber electrodes.

Fig. 24 (a) Cyclic performance of Si, CNF, Si–CNF, and C–Si–CNF electrodes at 0.2 A g⁻¹; (b) rate performance of CNF, Si–CNF, and C–Si–CNF electrodes. Reproduced with permission from ref. 92. Copyright 2023 Elsevier.

Fig. 24b shows the rate performance of various fiber membranes. It can be seen that due to the highest Si content, Si–CNF electrodes have the highest capacity at low current densities. As the current density increases, the capacity of the Si–CNF electrode significantly decreases, while the C–Si–CNF electrode maintains high capacity, even comparable to that of the CNF electrode. When the current density is greater than 1.0 A g⁻¹, the specific capacity of the C–Si–CNF electrode is higher than that of the Si–CNF electrode. The C–Si–CNF electrode still exhibits a reversible capacity of about 210 mAh g⁻¹ at 5.0 A g⁻¹, with a capacity retention rate of 36.9%, while the Si–CNF electrode only maintains about 35.2 mAh g⁻¹. The significant difference in rate performance is closely related to the electronic and ionic conductivity of the fiber membrane.

The carbon coating enhances electronic conductivity and forms pores that provide rapid ion transfer. The C–Si–CNF electrode achieves high capacity under high current. When the current density recovers from 5.0 to 0.1 A g⁻¹, the reversible capacity of the C–Si–CNF electrode reaches nearly 100% of the initial value and remains stable. This also reflects that the C–Si–CNF electrode has good reversibility and stability.

In summary, a graded porous Si/C nanofibers membrane (C–SiCNF) with carbon coating and hollow Si/SiOx encapsulation was prepared by hydrothermal coating of polydopamine followed by pyrolysis. The hydrothermal treatment/pyrolysis process not only achieves carbon coating on Si/C nanofibers, but also converts solid Si into hollow Si/SiOx through hydrothermal etching of Si and conversion of Si into SiOx. The carbon coating significantly improves the conductivity of the fiber membrane and effectively enhances the mechanical strength of Si/C composite electrodes. The combination of carbon coating, hollow structure, and Si/SiOx hybrid texture greatly reduces the structural stress and deformation of composite fibers caused by Si volume expansion. As a self-supporting anode for LIB, the C–Si–CNF film exhibits high capacity (694.7 mAh g⁻¹ at 0.2 A g⁻¹), good rate performance (36.9% capacity retention when current density increases from 0.1 to 5 A g⁻¹), and excellent cycling stability (92.6% capacity retention after 300 cycles). The C–Si–CNF membrane has high capacity, excellent cycling stability, and good flexibility, demonstrating good application prospects in flexible LIBs.

2.5. Comparison of material properties

The processing and charge/discharge performance of the composites critically depend on the silicon (Si) content (e.g., low, medium, or high Si ratios), with significantly different requirements for carbon additives of CNFs. Table 3 shows the performance comparison of silicon/CNF composites fabricated through different methods. Varying the Si content exerts distinct impacts on the performance metrics of different materials. When the silicon content is low, CFs act as a highly conductive carbon matrix to compensate for the lack of active materials. When the silicon content is moderate, carbon fiber balances the conductive network and volume expansion buffering. When the silicon content is high, carbon fibers construct a three-dimensional mechanical constraint framework that suppresses severe volume expansion.

Table 3 Performance comparison of silicon/CNF composites fabricated through different methods

Structure	Si deposition method	Composite structure	Si mass ratio	Current density (mA g^−1)	Capacity@cycles (mAh g^−1)	Capacity retention	Structural features	Ref.
B and M in the last column indicate the use of binder and metal substrate, respectively. The capacities are normalized to the mass of active material.
HCC–M–Si/CNF	Vacuum filtration method	Sandwich structure	57.12 wt%	400	1015@100	63.64%	No B & M	85
Si–o-CNFs	Electrostatic self-assembly	Interwoven structure	70 wt%	500	1063@400	47.93%	B & M	80
Si–CNF@C	Two-step electrostatic self-assembly followed by sintering	Interwoven Structure	58.3 wt%	500	849@1000	47.19%	B & M	87
Si/CNFs	Electro-deposition	Honeycomb structure	—	—	730@50	83.9%	No B & M	90
Si–C	Coating	Wrapping structure	31 wt%	0.1C	650@100	89.29%	No B & M	79
Si@p-c/p-CF	Hydrothermal deposition/pyrolysis	Wrapping structure	4.9 wt%	200	643@300	92.6%	No B & M	91
C–Si–CNF	Electrospinning	Wrapping structure	—	200	940@400	51.79%	No B & M	92

---

This review focuses on Si–C LIB anode systems composed of random networks of CNFs blended with silicon nanoparticles (Si NPs) or carbon-coated Si NPs. In addition to the structures summarized above, several unique architectures have been developed. Advanced architectures like core–shell fibers⁹³ and fiber-in-fiber⁹⁴ composites demonstrate that hierarchical structural engineering is pivotal for synchronously addressing silicon's volume expansion and sluggish kinetics. It was reported that the dual-nozzle electrospinning technique was employed to produce fibers with a core–shell structure.⁹³ The silicon nanoparticle (Si NP)-filled core is encapsulated by a CFs shell, denoted as SiNP@C fibers. This unique structure addresses all the aforementioned challenges, enabling stable silicon battery operation and demonstrating exceptional electrochemical performance. (1) High specific capacity: 1384 mAh g⁻¹ at a 10C rate (based on the total weight of both Si and C). (2) Outstanding rate capability. (3) Discharge completed within 5 minutes (12C rate) while retaining 721 mAh g⁻¹. (4) Long cycle life: 300 cycles with 99% capacity retention. It was reported that a novel nano-in-micro fiber, silicon-rich composite anode could be fabricated using coaxial electrospinning technology with a polyvinyl alcohol (PVA)/silicon suspension as the core fluid and a polyacrylonitrile (PAN)/dimethylformamide (DMF) solution as the sheath fluid.⁹⁴ The high-quality nonwoven carbon microfibers derived from the PAN sheath provide structural protection for the internal short silicon-rich nanofibers and prevent further oxidation of silicon during fiber processing. Together with the thin carbon skin on the internal silicon–carbon fiber bundles, the microfiber forms a conductive network that serves as a three-dimensional current collector to facilitate lithium-ion transport and charge transfer during charge/discharge processes. Simultaneously, the short silicon nanofibers and their mesoporous buffer spaces help to effectively mitigate the stress induced by the volume changes of silicon nanoparticles during lithium insertion and extraction. All these features of the nano-in-microfiber composite anode collectively address the trade-off between electrode pulverization and high reversible capacity, resulting in promising electrochemical performance and capacity retention for silicon-rich anodes. The unique processing strategy presented in this work also proposes a new pathway for manufacturing high-solid-content composite nanomaterials with applications extending beyond LIBs.

In addition to the systems summarized above, other well-defined architectures are also involved, such as vertically aligned carbon nanofiber/nanotube (VACNF/VACNT) frameworks as three-dimensional current collectors. For example, Fan et al. reported a high-performance Si anode involving a well-optimized CNT–Si coaxial cone array, highlighting the superiority of the core–shell nanowire structure.⁹⁵

(1) Vertically aligned CNT cores with large diameters and wide interwire spacing effectively accommodate strain during long-term cycling and facilitate rapid charge transport at high rates.

(2) The tapered Si coating with a unique vertically open-pore structure eliminates excessive strain accumulation at nanowire roots and enhances structural stability.

(3) Electrochemical performance achieved >90% capacity retention over 100 cycles at 0.2C, delivered high delithiation capacities of 2515–718 mAh g⁻¹ at a charge/discharge rate of 0.515C.

(4) Key insight demonstrates that morphological control of CNT cores and Si shells critically determines electrochemical performance.

Klankowski et al. reported a hybrid Li-ion anode material featuring coaxial Si shells on VACNF cores. The VACNF arrays served as both a mechanically stable nanostructured template and an efficient electrical interface due to their cup-stacked graphitic structure.⁹⁶

(1) Structural advantages involved open vertical core–shell nanowire structures even with thick Si layers (nominal thickness of 1.5 μm) deposited on 10 μm-long VACNFs, enabling radial expansion/contraction of Si shells during Li⁺ insertion/extraction.

(2) Electrochemical performance achieved high-rate Li storage at 1C, high capacity, high rate capability, and exceptional cyclability, efficient electrical connectivity with bulk Si materials, and short Li⁺ transport paths enabled by the three-dimensional nanostructure.

Vertically structured Si–C composites are promising candidates for next-generation high-energy LIBs. This review summarized the different electrode systems. The random networks of CNFs blended with Si NPs or carbon-coated Si NPs exhibited inferior performance when compared to VACNF/VACNT architectures. Subsequent research should prioritize more structurally well-defined systems to advance this field.

3. Conclusions and prospective

This review mainly includes three themes: (1) composites of Si/CNFs; (2) the structure of Si/CNFs; (3) enhancement of CNFs. Given the current situation, an increasing number of academics are working hard to overcome obstacles in finding new and acceptable materials. Developing a unique combination of pure CNFs and other materials seems to be a convincing and effective choice. This strategy not only overcomes the limitations of pure CNFs, but also combines the advantages of other materials into the compound. Fortunately, significant progress has been made in the development of these compounds in recent years, particularly in the coupling of CNF with TMO, metal sulfides, metal selenides, MXene, and other materials. According to recent research, TMOs, sulfides, MXenes, and related materials have demonstrated excellent performance. Combining these materials eliminates many of the drawbacks of pure CNFs. For example, thorough research has been conducted on rate performance, cycling stability, and electronic conductivity to ensure that composite materials exhibit different characteristics in various applications. Research has shown that significant progress has been made in the manufacturing process of CNFs-based materials, providing new opportunities to improve their performance in many applications.

The preparation of silicon–carbon composite materials has become relatively mature and has been widely applied. Conventional silicon–carbon composite materials are usually prepared through methods such as pyrolysis, mechanical mixing and high-energy ball milling, or a combination of the two. However, after multiple cycles, the silicon material still experiences expansion, leading to the crushing of the polarizers. The cycle performance and lifespan are limited. Using carbon nanofibers to reinforce silicon materials to produce anode plates has become a new solution.

The silicon oxide formed by the distribution of O and Si atoms exhibits more stable electrochemical performance, such as higher ICE and better cycling ability.⁹⁷ Compared with Si, silicon oxide (SiOx) undergoes less volume change during cycling (∼200%). Due to its high theoretical specific capacity (SiO is 2680 mAh g⁻¹, SiO₂ is 1965 mAh g⁻¹), suitable working voltage (∼0.3 V), low cost, and easy synthesis, SiOx is currently one of the most promising silicon materials. The purpose of this review is to summarize the use of different CNFs reinforcement methods for silicon composite materials, or to further use CNFs to reinforce and produce anode plates on the basis of silicon–carbon composite materials. It provides a solution for the subsequent production of silicon anode plates. Similar methods can also be used to solve the expansion problem of silicon-related materials.

This review classifies Si/CNFs composite materials and systematically summarizes the latest developments in this field. Based on different carbon sources, various silicon–carbon nanofiber composite materials have been developed, which significantly improve lithium storage performance, including capacity, rate capability, coulombic efficiency, and cycle life. Although these well-designed composite materials and electrode systems have great potential in the next generation of LiBS, there are still some key issues that need to be addressed to accelerate the commercialization of silicon-based anodes. Several prospects are presented below.

(1) At present, the understanding of the reaction mechanism of lithium in Si/CNFs composite materials, especially the lithium/disulfide reaction in various microstructures, is not sufficient. It is crucial to quantitatively explore the lithium and dehydrogenation processes to provide valuable data for the design and optimization of Si/CNFs anode composite materials. Further research is needed on the transport kinetics of lithium ions in Si/CNFs electrodes, particularly the interfacial reactions between Si and CNFs, as well as the reactions between electrodes and electrolytes.

(2) Considering the practical application of LiBs, in addition to some carefully designed silicon anodes (such as pomegranate-inspired silicon design, microporous silicon, and graphene support), the weight and volume capacity of Si/CNFs electrodes should be considered simultaneously in the design and manufacturing process.

(3) In order to enable these batteries to be applied in high-power applications, further research is needed to improve the rate performance of Si/CNFs electrodes.

(4) Manufacturing costs will inevitably become a thorny issue affecting the practical application of Si/CNFs-based anodes.

(5) Composition regulation of Si/CNFs composite materials. Although the basic components of Si/CNFs composites include Si and CNFs, the source, price, and synthesis conditions of precursors are key factors affecting their practical applications. In addition, the content of CNFs and silicon, as well as the microstructure (disorder, specific surface area, etc.), have a significant impact on the energy storage performance of Si/CNFs composites. An appropriate carbon content is essential for buffering volume expansion and ensuring sufficient conductivity. On the other hand, excessive carbon content can affect the reversible capacity of Si-based anodes. Therefore, when designing and manufacturing feasible Si/CNFs anodes, the weight ratio of Si/CNFs must be strictly specified.

(6) Layered structure design of Si/CNFs composite materials. Although the composition of Si/CNFs composite materials has been determined, different structures provide different advantages. For example, one-dimensional structures can alleviate radial mechanical stress caused by volume expansion and prevent electrode material pulverization. In addition, one-dimensional structures facilitate effective electron transfer and rapid charge transfer along the axis. On the contrary, the 2D structure allows for easy acquisition of Li⁺, which not only improves ion adsorption but also accelerates ion diffusion. In addition, due to its unique sheet-like structure, the two-dimensional structure exhibits good compatibility with other materials with minimal volume changes. On the other hand, the characteristics of three-dimensional structures are three-dimensional interconnected structures and strong mechanical strength, which can withstand stress changes. Due to the functionality of 3D structures, they are also crucial for practical battery applications. These characteristics include high conductivity, interconnected porous channels, and excellent mechanical stability. Especially in three-dimensional microstructures, efficient conductive networks can promote rapid diffusion of electrons/ions, significantly improving the performance of batteries. By optimizing structural design, electrochemical performance can be improved while maintaining structural stability, thereby better accommodating Li⁺.

(7) In addition to optimizing the design of Si/CNFs composite materials themselves, the selection of binders and the optimization of electrode preparation processes are also crucial. Developing new electrolyte additives and polymer binders becomes effective methods.  Adding appropriate electrolyte additives helps to form a thin and stable SEI layer, which improves the cycling stability and charging efficiency of LiBs full batteries. The addition of binders can affect the adhesion between the collector and the active substance, as well as the integrity of the electrode during cycling. In the future, further improvements may be achieved by introducing new concepts of material synthesis and hierarchical structure construction simultaneously. We believe that the improvement of battery performance and the true application of LiBs based on Si/CNFs are imminent.

Several challenges need to be addressed for CNFs-containing anode materials.

(1) New electrode materials and structures with high efficiency for LIBs should be designed. New materials, such as high entropy and zero strain materials, are currently being improved and adjusted for their application as battery anodes.

(2) Active interfacial compatibility of electrode components should be designed to promote ongoing electrochemical reactions. Conductive materials are used to deposit with active materials where strong interfacial strength plays a crucial role in enhancing electrochemical and mechanical properties.

(3) Flexible CNFs and foam are applied for developing self-supporting and adhesive-free multifunctional electrodes. The porosity of nanofibers and foam-based electrodes is controlled to improve the overall battery capacity.

(4) Comprehensive modeling and theoretical simulation calculations are required for better understanding of the working process and the decay mechanisms of capacity.

The interfacial incompatibility between CNFs and silicon (Si) in LIBs stems from inherent differences in their chemical, thermodynamic, and electrochemical properties: insufficient chemical bonding (e.g., inert sp²-hybridized carbon surfaces and lack of covalent bonds with Si, resulting in interfacial bonding energy <0.5 eV), mismatched thermal expansion coefficients (Si CTE: ∼2.6 × 10⁻⁶ K⁻¹ vs. carbon: ∼0.8 × 10⁻⁶ K⁻¹), and charge transport barriers (Schottky barriers at interfaces increase resistance, reducing the Li⁺ diffusion coefficient to ∼10⁻¹⁴ cm² s⁻¹). These defects synergistically induce stress concentration, crack propagation, and polarization during cycling, leading to rapid capacity decay (e.g., <50% retention after 100 cycles). Strategies such as surface functionalization (e.g., plasma-modified CNFs with –NH₂ groups forming Si–N bonds), gradient interfacial layer design (amorphous/graphitic carbon transitions to buffer stress), and 3D interpenetrating architectures (core–shell/yolk–shell confinement to limit volume expansion) have been proposed. For instance, yolk–shell structures achieve 80% silicon expansion space utilization and extend the cycle life beyond 1000 cycles. Future efforts require integrating in situ characterization (operando TEM, synchrotron XANES) and multiscale simulations (molecular dynamics + finite element analysis), alongside machine learning for interface optimization, to advance practical CNFs/Si composite anodes.

In LIBs, agglomeration of silicon nanoparticles (Si NPs) severely compromises electrochemical performance. Driven by high surface energy, Si NPs aggregate via van der Waals forces into micron-scale clusters, reducing active surface area, obstructing Li⁺ diffusion, and exacerbating localized volume expansion (up to 300%). Internal stress heterogeneity within agglomerates triggers crack propagation, destabilizing electrode integrity, while poor electrolyte infiltration causes uneven SEI growth and persistent side reactions (e.g., electrolyte decomposition), ultimately degrading capacity (<50% retention after 100 cycles) and coulombic efficiency. Current mitigation strategies include surface modification (carbon/polymer coatings to lower surface energy), spatial confinement (encapsulating Si NPs in CNF channels or 3D porous frameworks), and chemical anchoring (covalent bonding via silane coupling agents). For example, Si@CNFs core–shell structures fabricated by electrospinning and in situ carbonization achieve >90% particle dispersion and 500-cycle stability. However, scalable, cost-effective dispersion methods remain challenging, necessitating optimized interfacial engineering and structural designs to balance dispersion stability and electrochemical performance.

Scalability challenges in electrospinning for CNF-based electrode production include low throughput (traditional single-needle systems: ∼0.1 g h⁻¹, insufficient for industrial-scale demand), poor fiber diameter uniformity (50–500 nm variations due to process fluctuations, degrading electrode homogeneity), and process instability (sensitivity to ambient humidity/temperature, multi-needle electric field interference). Additional hurdles in continuous manufacturing involve fiber alignment control in roll-to-roll systems, high-energy carbonization (∼1000 °C), and toxic solvent disposal (e.g., PAN/DMF precursors). While multi-needle arrays and bubble electrospinning boost yields to ∼10 g h⁻¹, trade-offs between fiber quality and process complexity persist. Future breakthroughs demand process–equipment–material co-optimization (e.g., low-cost biobased precursors, smart humidity/temperature control) and green manufacturing (low-toxicity solvents, waste recycling) to transition electrospinning from lab-scale to industrial applications.

The enhanced lithium storage capacity and electrochemical reversibility during charge/discharge processes arise from multiscale structural and interfacial engineering in composite electrodes. Specifically, CNFs/Si interfacial interactions (e.g., Si–C covalent bonds, Si–O–C coordination bonds) regulate Li⁺ diffusion kinetics.

3.1. Chemical bonding and charge transfer

Covalent bonds (Si–C) reduce interfacial electron transfer barriers, homogenizing charge distribution and minimizing polarization.

Polar functional groups (such as –Si–O–C–) accelerate Li⁺ desolvation (adsorption energy reduced by ∼0.15 eV at oxygen-rich interfaces), increasing diffusion rates (Li⁺ diffusion coefficient up to 10⁻¹² cm² s⁻¹).

3.2. Interfacial defects and diffusion pathways

Topological defects (such as pentagon/heptagon rings) and dopants (such as N, S) on CNFs create low-energy-barrier Li⁺ adsorption sites in the respect of molecular dynamics with the diffusion energy below 0.3 eV.

Gradient porous carbon coatings on Si with pore size of 1–5 nm improve electrolyte wettability via capillary effects, shortening ion transport distances.

Quantitative correlations between mechanical stability and cycling performance include the following factors.

3.2.1. Elastic modulus (E) and stress dissipation. High-modulus CNF networks (E > 50 GPa) absorb Si expansion stress via elastic deformation (finite element analysis: peak stress reduced by ∼30%), suppressing crack initiation.

3.2.2. Interfacial adhesion and capacity fade. Adhesion strength (e.g., nano-scratch critical load Lc >10 mN) correlates positively with capacity retention (experimental data: +5% retention per 1 mN increase in Lc after 100 cycles).

3.2.3. Porosity and strain tolerance. 30–50% porosity provides buffering space, limiting local Si strain to <10% (in situ SEM observations).

Universal theoretical models for these mechanisms remain elusive. Systematic multiscale experimental-simulation studies are also needed.

3.2.4. In situ interfacial characterization. Operando XPS to track Si–C bond dynamics; synchrotron X-ray imaging to resolve Li⁺ concentration gradients.

3.2.5. Cross-scale simulations. Molecular dynamics (atomic diffusion pathways) and phase-field modeling (mesoscale crack propagation) are used to quantify structure–property relationships.

3.2.6. Machine learning optimization. Neural networks trained on high-throughput data are used to predict optimal interfacial compositional parameters (C/Si ratios) and structural parameters (porosity, layer thickness).

3.2.7. Comparison of Silicon/carbon nanofibers-based batteries and Solid-state batteries. CNFs enhance battery performance through two key mechanisms of volume expansion buffering and conductivity enhancement. Concerning volume buffering, CNFs mitigate silicon's ∼300% expansion via structural constraints (e.g., yolk–shell designs achieving 80% expansion space utilization) and stress redistribution,⁹⁸ with finite element analysis showing ∼30% peak stress reduction. Flexible interfacial layers (e.g., N,P-doped porous CNFs) dynamically adapt to volume changes,⁹⁹ enabling stable Li metal anodes (1400 h cycling at 1 mA cm⁻², 99.89% coulombic efficiency). Mechano-electrochemical studies reveal interfacial adhesion strength (nano-scratch critical load directly correlates with capacity retention (+5% per 1 mN increase)). Concerning conductivity enhancement, CNFs’ three-dimensional conductive networks (10³–10⁴ S m⁻¹) and defect engineering (e.g., pentagon/heptagon defects, N/S doping) lower Li⁺ diffusion barriers (<0.3 eV via MD simulations), while Si–C covalent bonds (>2 eV bond energy) boost Li⁺ diffusion coefficients to 10⁻¹² cm² s⁻¹. Quantitative analyses include stress–strain simulations (e.g., NPCNFs films with 2.9 MPa tensile strength), in situ characterization (XPS tracking Si–C bond dynamics, Raman spectrum monitoring interface evolution), and machine learning models linking porosity (30–50%) to capacity retention (+8–12% per 10% porosity increase). Challenges remain in scaling electrospinning (yield ∼10 g h⁻¹, diameter uniformity 50–500 nm) and elucidating interfacial failure thresholds via multiscale simulations (phase-field modeling). Accelerated aging tests are needed to validate long-term stability beyond 1000 cycles for practical applications.

Compared to the theoretical capacity, the specific capacity is still relatively low. In addition, after cyclic deformation, the capacity rapidly decreases. The quality of the active material loaded on the carbon-based electrode is limited, which adversely affects the volume and area energy density. In addition, there is a lack of strategies and devices for low-cost, large-scale, and reproducible synthesis of carbon materials with specific structures and morphologies. Therefore, the development of flexible carbon-based anode materials characterized by higher-level flexibility, high energy capacity, scalability, and cost-effectiveness will be the main focus of future research. Finally, CNFs have significant potential for future use in energy storage devices such as LIBs. In order to improve the electrochemical performance, it is necessary to develop more effective methods. Although there are many challenges that need to be addressed, composite materials made from CNFs materials show great potential for expanding the use of CNFs-based anode materials in LIBs.

Solid-state batteries (SSBs) have emerged as a critical technology for powering future electric vehicles and other applications due to their enhanced safety and higher energy density compared to conventional LIBs.¹⁰⁰ Among next-generation energy storage systems, SSBs (semi-solid or all-solid-state) stand out as the most promising candidates in terms of safety, cost, performance, and compactness. The integration of silicon (Si) into SSBs has attracted significant efforts owing to its exceptionally high specific capacity (3590 mAh g⁻¹), low cost, and terrestrial abundance. Silicon-anode SSBs demonstrate compelling application potential. Compared to lithium metal counterparts, Si-based SSBs exhibit the following advantages. (1) Tolerance to high stacking pressures, improving anode–electrolyte interfacial contact, (2) operational capability at room temperature, (3) enhanced safety by suppressing short-circuit risks, (4) superior electrochemical performance with accelerated reaction kinetics. Furthermore, higher ionic conductivity can be achieved in Si-based composite anodes, as they demonstrate improved thermodynamic stability with most solid electrolytes (e.g., sulfide-based solid electrolytes). Future applications of related materials in other energy storage systems are anticipated.

Conflicts of interest

There are no conflicts to declare.

Data availability

All relevant data are within the paper.

Acknowledgements

This research work is supported by the Science and Technology Program of Suzhou City, China (SYG202342) and the Big Data Computing Center of the Southeast University, China.

References

1. H. Yu, J. Duan, W. Du, S. Xue and J. Sun, China's energy storage industry: Develop status, existing problems and countermeasures, Renewable Sustainable Energy Rev., 2017, 71, 767–784.
2. B. Dunn, H. Kamath and J. M. Tarascon, Electrical energy storage for the grid: a battery of choices, Science, 2011, 334, 928–935.
3. M. Srivastava, J. Singh, T. Kuila, R. K. Layek, N. H. Kim and J. H. Lee, Recent advances in graphene and its metal-oxide hybrid nanostructures for lithium-ion batteries, Nanoscale, 2015, 7, 4820–4868.
4. H. Yin, F. Rong and Y. Xie, A review of typical transition metal phosphides electrocatalysts for hydrogen evolution reaction, Int. J. Hydrogen Energy, 2024, 52, 350–375.
5. Q. Wu and Y. Xie, Hydrogel bonding interface-induced all-in-one supercapacitors: A review, J. Energy Storage, 2024, 104, 114480.
6. Y. Wang and Y. Xie, Self-healing and deformation responsive supercapacitor using smart gel polymer electrolyte and redox-activated electrode, J. Mater. Chem. C, 2024, 12, 13388–13410.
7. R. Hang and Y. Xie, Matching Faradaic reaction of multi-transition metal compounds as supercapacitor electrode materials: A reviewelectrode materials: A review, J. Alloys Compd., 2024, 1002, 175367.
8. M. Ko, P. Oh, S. Chae, W. Cho and J. Cho, Considering Critical Factors of Li-rich Cathode and Si Anode Materials for Practical Li-ion Cell Applications, Small, 2015, 11, 4058–4073.
9. S. H. Yu, S. H. Lee, D. J. Lee, Y. E. Sung and T. Hyeon, Conversion Reaction-Based Oxide Nanomaterials for Lithium Ion Battery Anodes, Small, 2016, 12, 2146–2172.
10. D. Deng, Li–ion batteries: basics, progress, and challenges, Energy Sci. Eng., 2015, 3, 385–418.
11. J. An, L. Shi, G. Chen, M. Li, H. Liu, S. Yuan, S. Chen and D. Zhang, Insights into the stable layered structure of a Li-rich cathode material for lithium-ion batteries, J. Mater. Chem. A, 2017, 5, 19738–19744.
12. Z. Wang, L. Zhou and X. W. Lou, Metal oxide hollow nanostructures for lithium-ion batteries, Adv. Mater., 2012, 24, 1903–1911.
13. Y. Xie and C. Yao, Interfacial effect investigation of lithium perchlorate-interacted oxygen-containing carbon paper, Surf. Sci., 2025, 751, 122615.
14. Y. Xie, J. Xu, L. Lu and C. Xia, Electrochemical Investigation of Lithium Perchlorate-Doped Polypyrrole Growing on Titanium Substrate, Inorganics, 2024, 12, 125.
15. J. R. Dahn, T. Zheng, Y. Liu and J. S. Xue, Mechanisms for Lithium Insertion in Carbonaceous Materials, Science, 1995, 270, 590–593.
16. M. T. McDowell, S. W. Lee, W. D. Nix and Y. Cui, 25th anniversary article: Understanding the lithiation of silicon and other alloying anodes for lithium-ion batteries, Adv. Mater., 2013, 25, 4966–4985.
17. J. Deng, H. Ji, C. Yan, J. Zhang, W. Si, S. Baunack, S. Oswald, Y. Mei and O. G. Schmidt, Naturally rolled-up C/Si/C trilayer nanomembranes as stable anodes for lithium-ion batteries with remarkable cycling performance, Angew. Chem., Int. Ed., 2013, 52, 2326–2330.
18. B. Key, R. Bhattacharyya, M. Morcrette, V. Seznec, J. M. Tarascon and C. P. Grey, Real-time NMR investigations of structural changes in silicon electrodes for lithium-ion batteries, J. Am. Chem. Soc., 2009, 131, 9239–9249.
19. Z. Xiao, C. Wang, L. Song, Y. Zheng and T. Long, Research progress of nano-silicon-based materials and silicon-carbon composite anode materials for lithium-ion batteries, J. Solid State Electrochem., 2022, 26, 1125–1136.
20. C. K. Chan, H. Peng, G. Liu, K. McIlwrath, X. F. Zhang, R. A. Huggins and Y. Cui, High-performance lithium battery anodes using silicon nanowires, Nat. Nanotechnol., 2008, 3, 31–35.
21. X. Zhou, L. J. Wan and Y. G. Guo, Electrospun silicon nanoparticle/porous carbon hybrid nanofibers for lithium-ion batteries, Small, 2013, 9, 2684–2688.
22. X. H. Liu, F. Fan, H. Yang, S. Zhang, J. Y. Huang and T. Zhu, Self-limiting lithiation in silicon nanowires, ACS Nano, 2013, 7, 1495–1503.
23. Y. Wang, X. Wen, J. Chen and S. Wang, Foamed mesoporous carbon/silicon composite nanofiber anode for lithium ion batteries, J. Power Sources, 2015, 281, 285–292.
24. X. Feng, F. Rong and Y. Xie, Electrochemical cycling stability of electrospun silicon–carbon nanofibers anode materials: A review, Phys. Chem. Chem. Phys., 2025, 27, 1720–1751.
25. Y. Xie, Electrochemical and hydrothermal activation of carbon fiber supercapacitor electrode, Fibers Polym., 2022, 23, 10–17.
26. Q. Jiang, A. Beutl, H. Kühnelt and A. Bismarck, Structural composite batteries made from carbon fibre reinforced electrodes/polymer gel electrolyte prepregs, Compos. Sci. Technol., 2023, 244, 110312–110320.
27. Y. Anil Kumar, G. Koyyada, T. Ramachandran, J. H. Kim, S. Sajid, M. Moniruzzaman, S. Alzahmi and I. M. Obaidat, Carbon Materials as a Conductive Skeleton for Supercapacitor Electrode Applications: A Review, Nanomaterials, 2023, 13, 1049.
28. T. Zhang, C. H. Kim, Y. Cheng, Y. Ma, H. Zhang and J. Liu, Making a commercial carbon fiber cloth having comparable capacitances to carbon nanotubes and graphene in supercapacitors through a “top-down” approach, Nanoscale, 2015, 7, 3285–3291.
29. Q. Cheng, R. Yuge, K. Nakahara, N. Tamura and S. Miyamoto, KOH etched graphite for fast chargeable lithium-ion batteries, J. Power Sources, 2015, 284, 258–263.
30. J.-H. Shim and S. Lee, Characterization of graphite etched with potassium hydroxide and its application in fast-rechargeable lithium ion batteries, J. Power Sources, 2016, 324, 475–483.
31. Q. Cheng and Y. Zhang, Multi-Channel Graphite for High-Rate Lithium Ion Battery, J. Electrochem. Soc., 2018, 165, A1104–A1109.
32. K. Zhu, Y. Luo, F. Zhao, J. Hou, X. Wang, H. Ma, H. Wu, Y. Zhang, K. Jiang, S. Fan, J. Wang and K. Liu, Free-Standing, Binder-Free Titania/Super-Aligned Carbon Nanotube Anodes for Flexible and Fast-Charging Li-Ion Batteries, ACS Sustainable Chem. Eng., 2018, 6, 3426–3433.
33. M.-C. Han, J.-H. Zhang, Y.-M. Li, Y.-R. Zhu and T.-F. Yi, Li5Cr7Ti6O25/Multiwalled Carbon Nanotubes Composites with Fast Charge-Discharge Performance as Negative Electrode Materials for Lithium-Ion Batteries, J. Electrochem. Soc., 2019, 166, A626–A634.
34. B. Ng, X. Peng, E. Faegh and W. E. Mustain, Using nanoconfinement to inhibit the degradation pathways of conversion-metal oxide anodes for highly stable fast-charging Li-ion batteries, J. Mater. Chem. A, 2020, 8, 2712–2727.
35. F. Wu, L. Shi, D. Mu, H. Xu and B. Wu, A hierarchical carbon fiber/sulfur composite as cathode material for Li–S batteries, Carbon, 2015, 86, 146–155.
36. Y. Zhu, X. Fan, L. Suo, C. Luo, T. Gao and C. Wang, Electrospun FeS2@Carbon Fiber Electrode as a High Energy Density Cathode for Rechargeable Lithium Batteries, ACS Nano, 2015, 10, 1529–1538.
37. J. Chen, F. Rong and Y. Xie, Electrospun polyvinyl alcohol/chitosan nanofibers modified with carbon nanotubes and sliver particles for electrochemical sensor application, React. Funct. Polym., 2025, 206, 106110.
38. H. Zhu and Y. Xie, Electrochemical performance of bridge molecule-reinforced activated carbon fiber-m-aminobenzenesulfonic acid-polyaniline for braidable-supercapacitor application, Chem. Eng. J., 2023, 478, 147416.
39. H. Wang, J. Xu and Y. Xie, Bridge molecule thiourea enforced polyaniline growing on activated carbon fiber for wearable braidable-supercapacitor, Appl. Surf. Sci., 2023, 637, 157980.
40. Y. Xie, Electrochemical Performance of Polyaniline Support on Electrochemical Activated Carbon Fiber, J. Mater. Eng. Perform., 2022, 31, 1949–1955.
41. M. Shi, Z. Liu, S. Zhang, S. Liang, Y. Jiang, H. Bai, Z. Jiang, J. Chang, J. Feng, W. Chen, H. Yu, S. Liu, T. Wei and Z. Fan, A Mott–Schottky Heterogeneous Layer for Li–S Batteries: Enabling Both High Stability and Commercial–Sulfur Utilization, Adv. Energy Mater., 2022, 12, 2103657–2103669.
42. L. Wang, P. Adiga, J. Zhao, W. S. Samarakoon, K. A. Stoerzinger, S. R. Spurgeon, B. E. Matthews, M. E. Bowden, P. V. Sushko, T. C. Kaspar, G. E. Sterbinsky, S. M. Heald, H. Wang, L. W. Wangoh, J. Wu, E. J. Guo, H. Qian, J. Wang, T. Varga, S. Thevuthasan, Z. Feng, W. Yang, Y. Du and S. A. Chambers, Understanding the Electronic Structure Evolution of Epitaxial LaNi(1-x)Fe(x)O(3) Thin Films for Water Oxidation, Nano Lett., 2021, 21, 8324–8331.
43. F. Hagemans, F. Camerin, N. Hazra, J. Lammertz, F. Dux, G. Del Monte, O. V. Laukkanen, J. J. Crassous, E. Zaccarelli and W. Richtering, Buckling and Interfacial Deformation of Fluorescent Poly(N-isopropylacrylamide) Microgel Capsules, ACS Nano, 2023, 17, 7257–7271.
44. N. Ding, J. Xu, Y. X. Yao, G. Wegner, X. Fang, C. H. Chen and I. Lieberwirth, Determination of the diffusion coefficient of lithium ions in nano-Si, Solid State Ionics, 2009, 180, 222–225.
45. N. Swaminathan, S. Natarajan and E. T. Ooi, Communication—A Fast and Accurate Numerical Technique for Impedance Spectroscopy of Microstructures, J. Electrochem. Soc., 2022, 169, 020543–020548.
46. S. Yuan, X. Duan, J. Liu, Y. Ye, F. Lv, T. Liu, Q. Wang and X. Zhang, Recent progress on transition metal oxides as advanced materials for energy conversion and storage, Energy Storage Mater., 2021, 42, 317–369.
47. J. Rebmann, H. Werners, F. Johst, M. Dohrmann, Y. U. Staechelin, C. Strelow, A. Mews and T. Kipp, Cation Exchange during the Synthesis of Colloidal Type-II ZnSe-Dot/CdS-Rod Nanocrystals, Chem. Mater., 2023, 35, 1238–1248.
48. A. Belgibayeva, S. Berikbaikyzy, Y. Sagynbay, G. Turarova, I. Taniguchi and Z. Bakenov, A review on electrospun polyvinylpyrrolidone-derived carbon composite nanofibers as advanced functional materials for energy storage applications and beyond, J. Mater. Chem. A, 2023, 11, 11964–11986.
49. S. Li, C. Zhang, J. Fu, Y. Zhou, J. Sun, Y. He, F. Nan and Z. Yu, Interfacial modification of carbon fiber by carbon nanotube gas-phase dispersion, Compos. Sci. Technol., 2020, 195, 108196–108200.
50. C. Zhang, S. Shi, C. Gao, D. Zhang, L. Bi, L. Wang and Y. He, Three-stage channel sprays carbon nanotubes onto carbon fibers surface, preparing high performance carbon fibers composites, J. Compos. Mater., 2023, 57, 2287–2295.
51. M.-S. Wang, W.-L. Song, J. Wang and L.-Z. Fan, Highly uniform silicon nanoparticle/porous carbon nanofiber hybrids towards free-standing high-performance anodes for lithium-ion batteries, Carbon, 2015, 82, 337–345.
52. E. Qu, T. Chen, Q. Xiao, G. Lei and Z. Li, Flexible freestanding 3D Si/C composite nanofiber film fabricated using the electrospinning technique for lithium-ion batteries anode, Solid State Ionics, 2019, 337, 70–75.
53. Q. Lin and J. N. Harb, Implementation of a Thick-Film Composite Li-Ion Microcathode Using Carbon Nanotubes as the Conductive Filler, J. Electrochem. Soc., 2004, 151, A1115–A1121.
54. J. W. Yap, T. Wang, H. Cho and J.-H. Kim, Comparison of carbon-nanofiber and carbon-nanotube as conductive additives in Si anodes for high-energy lithium-ion batteries, Electrochim. Acta, 2023, 446, 142108–142118.
55. J. Guo, A. Sun, X. Chen, C. Wang and A. Manivannan, Cyclability study of silicon–carbon composite anodes for lithium-ion batteries using electrochemical impedance spectroscopy, Electrochim. Acta, 2011, 56, 3981–3987.
56. D. Mazouzi, D. Reyter, M. Gauthier, P. Moreau, D. Guyomard, L. Roué and B. Lestriez, Very High Surface Capacity Observed Using Si Negative Electrodes Embedded in Copper Foam as 3D Current Collectors, Adv. Energy Mater., 2014, 4, 130178–130191.
57. L. Y. Lim, S. Fan, H. H. Hng and M. F. Toney, Operando X-ray Studies of Crystalline Ge Anodes with Different Conductive Additives, J. Phys. Chem. C, 2015, 119, 22772–22777.
58. Y. Kuang, C. Chen, D. Kirsch and L. Hu, Thick Electrode Batteries: Principles, Opportunities, and Challenges, Adv. Energy Mater., 2019, 9, 1901457–1901476.
59. D. Li, H. Hu, B. Chen and W. Y. Lai, Advanced Current Collector Materials for High-Performance Lithium Metal Anodes, Small, 2022, 18, e2200010.
60. J. Wu, X. Qin, C. Miao, Y.-B. He, G. Liang, D. Zhou, M. Liu, C. Han, B. Li and F. Kang, A honeycomb-cobweb inspired hierarchical core–shell structure design for electrospun silicon–carbon fibers as lithium-ion battery anodes, Carbon, 2016, 98, 582–591.
61. H. Zhang, X. Qin, J. Wu, Y.-B. He, H. Du, B. Li and F. Kang, Electrospun core–shell silicon–carbon fibers with an internal honeycomb-like conductive carbon framework as an anode for lithium ion batteries, J. Mater. Chem. A, 2015, 3, 7112–7120.
62. B. Liu, X. Wang, B. Liu, Q. Wang, D. Tan, W. Song, X. Hou, D. Chen and G. Shen, Advanced rechargeable lithium-ion batteries based on bendable ZnCo2O4-urchins-on-carbon-fibers electrodes, Nano Res., 2013, 6, 525–534.
63. J. Jiang, J. Zhu, W. Ai, Z. Fan, X. Shen, C. Zou, J. Liu, H. Zhang and T. Yu, Evolution of disposable bamboo chopsticks into uniform carbon fibers: a smart strategy to fabricate sustainable anodes for Li-ion batteries, Energy Environ. Sci., 2014, 7, 2670–2679.
64. K. Fu, L. Xue, O. Yildiz, S. Li, H. Lee, Y. Li, G. Xu, L. Zhou, P. D. Bradford and X. Zhang, Effect of CVD carbon coatings on Si@CNF composite as anode for lithium-ion batteries, Nano Energy, 2013, 2, 976–986.
65. J. E. Entwistle and S. V. Patwardhan, Enabling scale-up of mesoporous silicon for lithium-ion batteries: a systematic study of a thermal moderator, RSC Adv., 2021, 11, 3801–3807.
66. H. Chen, Z. Dong, Y. Fu and Y. Yang, Silicon nanowires with and without carbon coating as anode materials for lithium-ion batteries, J. Solid State Electrochem., 2010, 14, 1829–1834.
67. T. Song, J. Xia, J.-H. Lee, D. H. Lee, M.-S. Kwon, J.-M. Choi, J. Wu, S. K. Doo, H. Chang, W. I. Park, D. S. Zang, H. Kim, Y. Huang, K.-C. Hwang, J. A. Rogers and U. Paik, Arrays of Sealed Silicon Nanotubes As Anodes for Lithium Ion Batteries, Nano Lett., 2010, 10, 1710–1716.
68. J. R. Szczech and S. Jin, Nanostructured silicon for high capacity lithium battery anodes, Energy Environ. Sci., 2011, 4, 56–72.
69. J. Liang, Z. Fan, S. Chen, S. Zheng and Z. Wang, A novel three-dimensional cross-linked net structure of submicron Si as high-performance anode for LIBs, J. Alloys Compd., 2021, 860, 158433–158442.
70. Y. Xing, T. Shen, T. Guo, X. Wang, X. Xia, C. Gu and J. Tu, A novel durable double-conductive core-shell structure applying to the synthesis of silicon anode for lithium ion batteries, J. Power Sources, 2018, 384, 207–213.
71. D.-C. Jia, Y.-Y. Feng, C.-L. Zhang, J.-J. Li, B.-W. Zhang, Y.-H. Dou, J. C. Roy, X.-Y. Zhu and L. Zhang, Freestanding carbon fiber-confined yolk-shelled silicon-based anode for promoted lithium storage applications, Rare Met., 2023, 42, 3718–3728.
72. Y. Chen, Y. Hu, Z. Shen, R. Chen, X. He, X. Zhang, Y. Zhang and K. Wu, Sandwich structure of graphene-protected silicon–carbon nanofibers for lithium-ion battery anodes, Electrochim. Acta, 2016, 210, 53–60.
73. Y. Li, R. Wang, J. Zhang, J. Chen, C. Du, T. Sun, J. Liu, C. Gong, J. Guo, L. Yu and J. Zhang, Sandwich structure of carbon-coated silicon–carbon nanofiber anodes for lithium-ion batteries, Ceram. Int., 2019, 45, 16195–16201.
74. F. Zhao, M. Zhao, Y. Dong, L. Ma, Y. Zhang, S. Niu and L. Wei, Facile preparation of micron-sized silicon-graphite-carbon composite as anode material for high-performance lithium-ion batteries, Powder Technol., 2022, 404, 117455–117465.
75. Y. Liu, H. Liu, W. Huang, Y. Yu, X. Dai and Z. Shan, Self-assembly of hierarchical microsized hard carbon-supported Si encapsulated in nitrogen-doped carbon as anode for lithium-ion batteries, J. Mater. Sci., 2020, 55, 12373–12384.
76. A. Zhang, X. Fang, C. Shen, Y. Liu and C. Zhou, A carbon nanofiber network for stable lithium metal anodes with high Coulombic efficiency and long cycle life, Nano Res., 2016, 9, 3428–3436.
77. F. Wang, M. Gao, R. Hong and X. Lu, Magnetoelectric Plasma Preparation of Silicon-Carbon Nanocomposite as Anode Material for Lithium Ion Batteries, Appl. Sci., 2020, 10, 2672–2683.
78. Q. Zhang, Y. Yang, D. Wang, R. Zhang, H. Fan, L. Feng, G. Wen and L.-C. Qin, A silicon–carbon/reduced-graphene composite of honeycomb structure for high-performance lithium-ion batteries, J. Alloys Compd., 2023, 944, 169185–169195.
79. J. Guo, A. Sun and C. Wang, A porous silicon–carbon anode with high overall capacity on carbon fiber current collector, Electrochem. Commun., 2010, 12, 981–984.
80. X. Bai, H. Zhang, J. Lin and G. Zhang, UV-ozone contributions towards facile self-assembly and high performance of silicon-carbon fiber materials as lithium-ion battery anodes, J. Colloid Interface Sci., 2021, 598, 339–347.
81. Z. Li, Y. Yang, J. Wang, Z. Yang and H. Zhao, Sandwich-like structure C/SiOx@graphene anode material with high electrochemical performance for lithium ion batteries, Int. J. Miner., Metall. Mater., 2022, 29, 1947–1953.
82. Y. Huang, J. Luo, J. Peng, M. Shi, X. Li, X. Wang and B. Chang, Porous silicon–graphene–carbon composite as high performance anode material for lithium ion batteries, J. Energy Storage, 2020, 27, 101075–101083.
83. Q. Zeng, Q. Huang, Z. Luo, A. Liu, H. Li, M. Zhong and Z. Zhang, Sandwich network structure silicon–carbon anode material for lithium-ion batteries based on bacterial cellulose, Ionics, 2023, 30, 135–144.
84. L. Wang, C. X. Ding, L. C. Zhang, H. W. Xu, D. W. Zhang, T. Cheng and C. H. Chen, A novel carbon–silicon composite nanofiber prepared via electrospinning as anode material for high energy-density lithium ion batteries, J. Power Sources, 2010, 195, 5052–5056.
85. C. Shen, X. Fang, M. Ge, A. Zhang, Y. Liu, Y. Ma, M. Mecklenburg, X. Nie and C. Zhou, Hierarchical Carbon-Coated Ball-Milled Silicon: Synthesis and Applications in Free-Standing Electrodes and High-Voltage Full Lithium-Ion Batteries, ACS Nano, 2018, 12, 6280–6291.
86. Q. Wu, H. Bai, X. Yang and J. Zhu, Significantly increasing the interfacial adhesion of carbon fiber composites via constructing a synergistic hydrogen bonding network by vacuum filtration, Composites, Part B, 2021, 225, 109300–109309.
87. X. Bai, H. Zhang, J. Lin and G. Zhang, Durable silicon-carbon composites self-assembled from double-protected heterostructure for lithium-ion batteries, J. Colloid Interface Sci., 2022, 615, 375–385.
88. P. Zheng, J. Su, Y. Wang, W. Zhou, J. Song, Q. Su, N. Reeves-McLaren and S. Guo, A High–Performance Primary Nanosheet Heterojunction Cathode Composed of Na0.44MnO₂ Tunnels and Layered Na2Mn3O7 for Na–Ion Batteries, ChemSusChem, 2020, 13, 1793–1799.
89. J. Song, S. Guo, L. Kou, K. Kajiyoshi, J. Su, W. Huang, Y. Li and P. Zheng, Controllable synthesis Honeycomb–like structure SiOx/C composites as anode for high-performance lithium-ion batteries, Vacuum, 2021, 186, 110044–110051.
90. M. Yao, Z. Zeng, H. Zhang, J. Yan and X. Liu, Electrophoretic deposition of carbon nanofibers/silicon film with honeycomb structure as integrated anode electrode for lithium-ion batteries, Electrochim. Acta, 2018, 281, 312–322.
91. X. Li, K. Xu, Y. Yu, C. Chen, H. Zhang and W. Lei, Self-standing high-performance Si nanoparticle anchored porous carbon fiber composites for Li-ion batteries, Electrochim. Acta, 2024, 475, 143658–143667.
92. Y. Jin, Z. Ai, Y. Song, X. Zhang, J. Shi and C. Ma, Enhanced lithium storage performance of Si/C composite nanofiber membrane with carbon coating as binder-free and self-supporting anode for lithium-ion battery, Mater. Res. Bull., 2023, 167, 112429–112439.
93. T. H. Hwang, Y. M. Lee, B. S. Kong, J. S. Seo and J. W. Choi, Electrospun core-shell fibers for robust silicon nanoparticle-based lithium ion battery anodes, Nano Lett., 2012, 12, 802–807.
94. Y. Pei, Y. Wang, A.-Y. Chang, Y. Liao, S. Zhang, X. Wen and S. Wang, Nanofiber-in-microfiber carbon/silicon composite anode with high silicon content for lithium-ion batteries, Carbon, 2023, 203, 436–444.
95. Y. Fan, Q. Zhang, Q. Xiao, X. Wang and K. Huang, High performance lithium ion battery anodes based on carbon nanotube–silicon core–shell nanowires with controlled morphology, Carbon, 2013, 59, 264–269.
96. S. A. Klankowski, R. A. Rojeski, B. A. Cruden, J. Liu, J. Wu and J. Li, A high-performance lithium-ion battery anode based on the core–shell heterostructure of silicon-coated vertically aligned carbon nanofibers, J. Mater. Chem. A, 2013, 1, 1055–1064.
97. H. Li, H. Li, Z. Yang, L. Yang, J. Gong, Y. Liu, G. Wang, Z. Zheng, B. Zhong, Y. Song, Y. Zhong, Z. Wu and X. Guo, SiO(x) Anode: From Fundamental Mechanism toward Industrial Application, Small, 2021, 17, e2102641.
98. W. Li, Q. Wang, K. Cao, J. Tang, H. Wang, L. Zhou and H. Yao, Mechanics-based optimization of yolk-shell carbon-coated silicon nanoparticle as electrode materials for high-capacity lithium ion battery, Compos. Commun., 2016, 1, 1–5.
99. M. Gao, Z. Tang, M. Wu, J. Chen, Y. Xue, X. Guo, Y. Liu, Q. Kong and J. Zhang, Self-supporting N, P doped Si/CNTs/CNFs composites with fiber network for high-performance lithium-ion batteries, J. Alloys Compd., 2021, 857, 157554–157563.
100. P. Molaiyan, B. Boz, G. S. dos Reis, R. Sliz, S. Wang, M. Borsari, U. Lassi and A. Paolella, Paving the path toward silicon as anode material for future solid-state batteries, eTransportation, 2025, 23, 100391–100404.

---