Silicon has become the preferred anode for high-energy-density lithium-ion batteries due to its highest theoretical lithium storage capacity; however, its low cycle life severely hinders its commercial application. The low cycle life of silicon anodes stems from their significant volume expansion during charge and discharge. This volume expansion degrades the mechanical stability of silicon particles and electrodes, the electrical contact between active particles, and the stability of the SEI film. To improve cycle stability and mitigate these adverse factors, researchers have made considerable efforts in electrode fabrication processes, electrolyte selection, and battery design, achieving some success. This review summarizes the aforementioned research work, aiming to provide feasible technical approaches for the practical application of silicon anodes.
Significantly increasing the energy density of lithium-ion batteries is an urgent requirement for the development of a range of new technologies, including portable electronics, electric vehicles, and energy storage power stations. The theoretical lithium intercalation capacity of existing graphite anodes is only 372 mAh/g, severely limiting further improvements in lithium-ion battery energy density. Silicon, with its ten times the lithium storage capacity of graphite anodes, is considered the ideal anode for next-generation lithium-ion batteries. However, the significant volume changes involved in the lithiation/delithiation process of silicon anodes lead to poor cycle stability, hindering the practical application of silicon-based anodes.
To address this issue, researchers have conducted extensive studies to improve its cycling performance. Some of these studies focus on designing the composition and structure of silicon materials, such as composite materials that provide buffer space, or utilizing nanotechnology to reduce the volume expansion effect of active particles. Magasinski et al. prepared nano-Si/C composite materials using a two-step vapor deposition method; after 100 cycles, the specific capacity of the material remained above 1500 mAh/g, demonstrating excellent cycling performance. Cui Yi et al. prepared hollow silicon nanotubes using a template method, which exhibited a cycle life of up to 6000 cycles.
The structural design of silicon materials enables individual active particles to remain intact during repeated volume changes, thereby improving the material's cycle performance. However, the volume expansion of silicon also affects the mechanical stability of the electrode, the electrical contact of active particles within the electrode, and the stability of the SEI film on the electrode surface. Therefore, the performance of silicon anodes is not only related to the composition and structure of the material itself, but also closely related to the electrode fabrication process, the choice of electrolyte, and the design of the battery. To advance the application of silicon-based anodes, researchers have conducted a series of studies on electrode composition, binder selection, and electrolyte optimization. The main progress is as follows:
1. Electrode composition optimization
The fabrication process of silicon-based anodes typically involves coating a slurry made from a mixture of active materials, conductive agents, and binders onto a current collector to form the electrode. The conductive agent ensures good electrical contact between the active powder particles, thereby guaranteeing sufficient electrochemical reactions within the electrode's internal active particles. Therefore, the physicochemical properties of the conductive agent, such as conductivity, dispersibility, particle size, and specific surface area, as well as its dosage and dispersion method, significantly affect the electrochemical performance of the silicon-based anode. Commonly used conductive agents include acetylene black, SuperP, Ketjen black, and vapor-deposited carbon fibers, which possess different morphologies, conductivity, and liquid absorption capabilities. Therefore, different conductive agents are suitable for silicon-based materials with different structures and morphologies.
Xiaojie et al. compared the electrochemical performance of micron-sized porous silicon using SuperP and Ketjenblack as conductive agents. They found that when Ketjenblack was used as a conductive agent, the specific capacity of the electrode was slightly lower, but the cycling performance was more stable.
Bernard Lestriez et al. investigated the electrochemical performance of silicon electrodes using acetylene black, graphene oxide, and reduced graphene as conductive agents in 100 nm silicon powder. The results showed that silicon electrodes prepared using reduced graphene as a conductive agent exhibited better cycle life and coulombic efficiency at high loadings (Si 2.5 mg/cm²).
Lestriez B et al. also investigated the effects of the dispersion mode and morphology of the conductive agent on the electrochemical performance of silicon powder. The results showed that the conductive agent not only affected the electrode's conductivity but also the distribution of mechanical stress in the electrode's micro-regions. Compared to SuperP and VGCF, silicon electrodes prepared using expanded graphite as a conductive agent exhibited higher internal resistance but more stable cycling performance. They attributed this to the fact that expanded graphite improved the mechanical stability of the electrode's micro-regions and dispersed the internal mechanical stress, thereby enhancing the material's cycling stability.
Guyomard D et al. compared the electrochemical performance of micron-sized silicon powder electrodes using multi-walled carbon nanotubes, VGCF, and conductive carbon as conductive agents. The results showed that the electrode exhibited the best cycling performance and rate performance when MWNTs and VGCF were used in combination.
On the other hand, in addition to their conductive and adhesive properties, conductive agents and binders can also act as buffering media to mitigate the volume expansion of active silicon. Therefore, the electrochemical performance of silicon-based anodes is also closely related to their electrode composition.
Beattie S.D. et al. optimized the electrode composition through modeling, concluding that the electrode exhibits better cycle stability when the silicon mass ratio is less than 20%, and verified this result experimentally. Further experimental results show that due to the presence of a certain porosity in the electrode, the silicon (100nm) content can be increased to 33% (mass fraction) in practical applications.
2. Adhesive selection
Bridel J.S. et al. believe that high electrode porosity and the presence of interactions between the binder and the silicon surface are prerequisites for ensuring electrode cycling performance. Given the inverse correlation between electrode porosity and its volumetric energy density, researchers are focusing more on screening suitable binders.
As early as 2003, Dahn JR et al. used surfactants and silane coupling agents to prepare silicon electrodes to enhance the interaction between silicon particles and binders. Correspondingly, the prepared silicon electrodes exhibited better cycling performance. Other researchers prefer to directly utilize the interaction between the binder's own functional groups and the silicon material surface to improve electrode adhesion. For example, binders containing alcoholic hydroxyl groups, phenolic hydroxyl groups, carboxyl groups, or amino groups can form bonds with the abundant silanol groups or Si-O-Si bonds on the silicon-based material surface. These binders include sodium carboxymethyl cellulose, polyacrylates, sodium alginate, polyacrylic acid, polyvinyl alcohol, polysaccharides, and polycyclodextrins.
Wu Nae Lih et al. were the first to use a CMC-SBR mixture as a binder to prepare silicon electrodes, and found that the prepared electrodes exhibited a cycle life of more than 50 cycles, which was far superior to silicon anodes prepared using PVDF. Dahn JR et al. prepared silicon electrodes using PVDF, CMC, and CMC-SBR binders, and found that the electrode prepared with CMC binder exhibited the best cycle performance.
Kovalenko Igor et al. used sodium alginate as a binder to prepare silicon electrodes that maintained a specific capacity of over 1700 mAh/g after 100 cycles at a current density of 4.2 A/g.
Liugao et al. synthesized a conductive PPFOMB polymer and used it as both a binder and a conductive agent, mixing it with 50 nm silicon powder to prepare an electrode. This electrode maintained a specific capacity of over 2000 mAh/g after 650 cycles at 0.1C. Building on this, the research group introduced groups that improve electrolyte adsorption and polymer mechanical strength into PPFOMB, synthesizing a polymer PEFM. The electrode using this polymer as a binder exhibited a specific capacity close to the theoretical lithium storage capacity of silicon; after 50 cycles, the electrode still retained a specific capacity of over 3000 mAh/g.
Yandonglin et al. compared the electrochemical performance of graphite-Si-SiOx-C composite anodes using polyimide resin and CMC-SBR as binders, and found that the PI-based electrode exhibited a longer cycle life. They confirmed that this was because the PI binder had a strong adhesive force, which prevented the anode powder shedding due to volume expansion and improved the material's cycle performance.
Some researchers also mix, modify, and crosslink commonly used binders to develop inexpensive and efficient binder systems. RyouMyung-Hyun et al. grafted dopamine onto sodium alginate and polyacrylic acid to obtain novel binders Alg-C and PAA-C. Silicon-based anodes based on these two binders exhibited higher lithium storage capacity and longer cycle life.
KooBonjae et al. obtained the crosslinking product c-PAA-CMC by vacuum treatment of two polymers, polyacrylic acid and sodium carboxymethyl cellulose, at 150 °C. The silicon anode based on this binder maintained a specific capacity of over 2000 mAh/g after 100 cycles.
Komaba S. et al. generated a crosslinked polymer by amidation reaction of polyacrylic acid and polycarbodiimide, and studied the electrochemical performance of silicon-graphite composites when PAA-PCD was used as a binder. The results showed that PAA-PCD had better dispersion and adhesion, and the corresponding silicon anode exhibited better cycling performance.
Wang Donghai et al. prepared a PAA-PVA/Si network structure electrode using vacuum thermal polymerization. The hydrogen bonds and covalent bonds between the silicon active particles and the polymer provided sufficient adhesion for the silicon volume expansion. Therefore, the electrode exhibited cycling performance far superior to PVDF-based and Na-CMC-based electrodes. After 300 cycles, the capacity retention of the electrode was 69%.
Furthermore, the literature results revealed inconsistencies in comparative experiments using different binders. Christoph Erk et al. found that nano-silicon electrodes prepared using PAA binder exhibited the best cycling performance, followed by PVA, AA, PVDF, and CMC. This result contradicts most other studies. This discrepancy may be due to the use of different silicon materials by different researchers, resulting in variations in the interaction between the binder and silicon, thus leading to differing research findings.
3. Electrolyte composition optimization
Suitable electrode composition and binders can improve the mechanical stability of silicon electrodes during cycling, thereby improving the cycling performance of silicon anodes. However, the significant volume expansion of silicon anodes during lithium storage can cause repeated rupture and reconstruction of the surface SEI film, leading to an increase in SEI thickness, increased interfacial impedance, and decreased battery capacity. Therefore, the electrochemical performance of silicon-based electrodes is closely related to the composition, structure, and properties of their surface solid electrolyte film. A dense and stable SEI film is beneficial for ensuring the cycling stability of silicon-based anodes. The SEI film is formed by the reduction and decomposition of the electrolyte on the anode surface, and its structure and properties are greatly affected by the electrolyte composition. Therefore, optimizing the electrolyte composition plays a crucial role in improving the application performance of silicon-based anodes.
Researchers have found that adding film-forming additives to the electrolyte is beneficial for forming a stable SEI film, thereby improving the electrochemical performance of the material. According to literature reports, the main film-forming agents used for silicon-based anodes include: vinylene carbonate, fluoroethylene carbonate, succinic anhydride, tris(pentafluorophenyl)borane, and maleic anhydride.
ParkJung-Ki et al. compared the electrochemical performance of silicon thin films with SA, VC, FEC and TFPB as film-forming additives, and confirmed that silicon electrodes exhibited more stable cycle performance in electrolytes containing VC and FEC additives, and that the coulombic efficiency of the battery was higher when FEC was used as an additive.
BordesArnaud et al. studied a full-cell system composed of LiO2 and Si/graphene, finding that an electrolyte containing 5% FEC additive could effectively improve its cycle stability, increasing the 50-cycle stability from 81% to almost no capacity decay. With further research, FEC has been considered a suitable silicon anode additive, and its mechanism of action has been extensively studied.
Chen Xilin et al., through analysis of 7Li solid-state NMR spectra, found that after 35 cycles in an electrolyte containing 10% FEC, the thickness of the SEI film on the surface of the silicon anode was only 2.1 times that of the second cycle; while in a blank electrolyte without FEC additives, the SEI film thickened to 3.4 times that of the second cycle after the same number of cycles. The increase in SEI film thickness resulted in a capacity retention of only 41.3% for the silicon anode after 100 cycles, compared to 97.3% in an electrolyte containing 10% FEC.
Building upon this foundation, researchers have further investigated the impact of FEC as a cosolvent on the performance of silicon-based materials. Lin Yongmao et al. studied the electrochemical performance of silicon-based anodes in electrolyte systems with FEC contents of 0%, 20%, and 50%, finding that as the FEC content increased, the cycle stability and coulombic efficiency of the electrode improved, but the first-cycle efficiency decreased to some extent.
Besides additives, the decomposition of electrolyte salts and solvents in the electrolyte also significantly affects the SEI film. YuZ et al. compared the electrochemical performance of silicon electrodes in four electrolyte salts: LiPF6, LiBOB, LiClO4, and LiBF4. They found that when LiBF4 was used as the electrolyte, the battery impedance increased significantly after cycling, and the battery capacity decayed rapidly. When LiBOB and LiPF4 were mixed and used as the electrolyte, the silicon electrode exhibited stable cycling performance.
Philippe Bertrand et al. found that silicon anodes exhibited more stable cycle performance in electrolytes using lithium difluorooxalate as the electrolyte. This is because LiFSI can form a more stable SEI film on the surface of the silicon-based anode and does not generate hydrofluoric acid that corrodes the SEI film and silicon material due to the presence of trace amounts of water in the electrolyte. Lithium difluorooxalate borate has a high reduction potential, which is beneficial for forming a stable SEI film and has also been used as an electrolyte salt to improve the cycle performance of silicon-based anodes. However, due to the limited solubility of LiBOB in common organic carbonate solvents, the ionic conductivity of the electrolyte is low. Therefore, LiBOB is often used as an electrolyte additive or as a coelectrolyte salt in conjunction with LiPF6. Other borates, such as lithium difluorooxalate borate, have also been shown to have good film-forming properties, effectively improving the SEI film of both silicon anodes and cathodes and enhancing battery cycle stability.
Besides commonly used carbonate solvents, researchers have also focused on the electrochemical performance of silicon anodes when using cyclic ethers and novel liquid-ionic liquids as solvents. Etacheri Vinodkunar et al. compared the electrochemical performance of silicon nanowires in LiPF6/EC-DMC and LiTFSI/DOL electrolyte systems, finding that silicon nanowires exhibited superior cycling performance in the DOL electrolyte. They suggested that the presence of polydioxolane and its oligomers in the electrolyte effectively suppressed the volume expansion of the silicon nanowires, allowing them to maintain a specific capacity of over 1250 mAh/g after 1000 cycles at 60°C.
Ionic liquids possess advantages such as high conductivity and a wide electrochemical window, leading to increasing research focus on their application in silicon anodes. ChoiJi-Ae et al. investigated the electrochemical behavior of a silicon-based alloy thin-film electrode in the ionic liquid 1-butyl-1-methylpyrrolidine-diimide, finding that the silicon thin-film electrode exhibited significantly better cycling stability than its counterpart in conventional electrolyte systems. After 100 cycles, the impedance spectrum of the SEI film on the silicon anode surface remained almost unchanged, indicating that the SEI film formed in the ionic liquid electrolyte is more stable.
Song Jin-Woo et al. investigated the electrochemical behavior of SiO1.3 in the pyrrolidine-onium type ionic liquid lithium-1-butyl-1-methylpyrrolidine-onium diimine salt, and found that silicon oxide exhibited better cycling performance in the ionic liquid electrolyte. However, due to the generally high viscosity of ionic liquids, the initial charge-discharge capacity of silicon-based anodes is often low.
4. Outlook
Silicon, with its specific capacity of 3580 mAh/g, has become the preferred anode for next-generation lithium-ion batteries. However, the low cycle life caused by volume expansion is a significant obstacle to its commercial application. Reducing the size of silicon particles can prevent individual particles from pulverizing due to stress; however, this alone is insufficient to address the poor electrical contact of particles, decreased electrode mechanical stability, and the effects of SEI film rupture and repeated rebuilding caused by volume expansion. Comprehensive consideration and improvement of electrode composition, binders, and electrolyte systems hold promise for providing a feasible solution to the adverse effects of silicon particle volume expansion, and ultimately for the commercial application of silicon anodes.
