Metallurgical refractory lining-guided inorganic binder for stable lithium storage in silicon microparticle anodes

Abstract

Silicon is a promising anode material for next-generation lithium-ion batteries (LIBs), offering a significantly higher theoretical capacity than graphite. Although nano-silicon excels, its high cost limits its practicality, making silicon microparticles (μSi) a more economical and scalable alternative. However, μSi anodes are hindered by substantial capacity degradation, resulting from the over 300% volume expansion upon cycling. This study introduces an aluminum dihydrogen phosphate (Al(H₂PO₄)₃, AHP) binder system, designed based on the principles of refractory chemistry, which effectively mitigates interfacial instability and mechanical failures in μSi anodes. The water-soluble AHP binder forms a uniform electrode through in situ dehydration condensation, creating a covalently cross-linked inorganic network. This high-modulus cross-linked network restricts the expansion of μSi particles during cycling, thereby preserving electrode integrity. As a result, μSi anodes incorporating the AHP binder exhibit exceptional cyclability, retaining a capacity of 1300.4 mA h g⁻¹ after 200 cycles at 0.5 A g⁻¹, alongside impressive rate capabilities of 936.4 mA h g⁻¹ and 769.1 mA h g⁻¹ at 4 A g⁻¹ and 5 A g⁻¹, respectively. Additionally, the AHP binder demonstrates superior compatibility with lithium iron phosphate (LiFePO₄, LFP) cathodes. This work establishes inorganic binders as a practical and economical solution for high-performance μSi anodes, enhancing energy density and lifespan of LIBs.

---

Broader context

The rapid growth of the electric vehicle market, driven by the global transition to sustainable energy, has increased the demand for lithium-ion batteries with higher energy densities. However, conventional graphite anodes are reaching their theoretical capacity limits, necessitating the exploration of alternative materials. Silicon microparticle (μSi) anodes offer significant promise, with a theoretical capacity ten times higher than that of graphite anodes. However, their 300% volume expansion during lithiation leads to particle breakage and degradation at the electrode/electrolyte interface, compromising cycling stability. While organic polymer binders can mitigate this issue, they are costly and time-consuming to produce, hindering commercial viability. In contrast, in situ cross-linking technology enables precise control over material dispersion and reaction kinetics during electrode fabrication, providing a scalable solution. Drawing inspiration from metallurgical refractory linings and utilizing eco-friendly, cost-effective inorganic materials, this study introduces an aluminum dihydrogen phosphate (AHP) binder. The binder's exceptional rigidity effectively limits μSi expansion, preventing particle fragmentation, and ensures stable electrode integrity over extended cycles. When applied to lithium iron phosphate (LFP) cathodes, the corresponding battery achieves over 1000 stable cycles. Full cells using AHP binders for both the anode and cathode show excellent cycle life. This innovation establishes a foundation for cost-effective and environmentally friendly energy storage solutions.

Introduction

As global demand for sustainable energy intensifies, lithium-ion batteries (LIBs) have become essential for energy storage technologies, powering a wide range of devices, from electric vehicles to portable electronics.^1,2 However, the drive for higher performance, particularly in terms of energy density, has exposed limitations in conventional LIBs that utilize graphite anodes, which have a theoretical capacity of only 372 mA h g⁻¹.^3,4 To meet the growing demand for higher energy densities, considerable research has focused on the development of silicon-based anodes.^5–8 Silicon, with a theoretical capacity of approximately 4200 mA h g⁻¹, more than ten times that of graphite, holds the potential to significantly enhance battery energy density and address the increasing need for advanced energy storage solutions.^9,10 Despite this promise, silicon's substantial volume expansion and contraction (≈300%) during lithiation and delithiation trigger a triple failure mechanism: at the microscopic scale, silicon particles fracture and pulverize; at the macroscopic level, the electrode structure collapses and electrical contact is lost; and at the interfacial level, the solid electrolyte interphase (SEI) repeatedly ruptures and reforms.^11–15 These issues lead to the loss of active material and a surge in interfacial impedance, ultimately causing rapid battery capacity degradation, which hinders the large-scale application of silicon-based anodes. Therefore, innovative material designs and structural strategies are essential to mitigate expansion-induced damage and fully exploit silicon's high-capacity potential.

In response to the challenges posed by the volumetric expansion of silicon anodes during cycling, numerous strategies have been proposed. One approach involves developing silicon–carbon composite anodes, while another focuses on optimizing the microstructure and surface coatings of silicon materials, often through pore design or micro/nano-structuring, to manage silicon's expansion.^16–18 Nanosizing silicon particles has also been demonstrated to reduce expansion, but nano-silicon faces significant barriers, such as high synthesis costs and susceptibility to agglomeration, limiting its practical application.¹⁹ In contrast, silicon microparticles (μSi) are more economically viable due to their commercial availability, low cost, and high volumetric density.^20,21 However, μSi encounter even more pronounced expansion issues, as larger particles fragment into smaller ones upon expansion, further exacerbating degradation at the SEI and promoting the formation of inactive Li/Si alloys.^22,23 As a result, there has been increasing attention on the development of functional binders designed to mitigate the expansion of μSi anodes, thereby enhancing battery stability and cycle life.^23–28 Considerable research has focused on organic polymer binders, which have been shown to effectively alleviate the expansion of μSi. For instance, a composite binder incorporating 5 wt% polyrotaxane into conventional polyacrylic acid has been demonstrated to impart excellent elasticity to the polymer network, enabling stable cycling for μSi anodes.²⁴ Similarly, self-healing polymer coatings on silicon electrodes have significantly alleviated expansion issues and improved electrochemical stability.²⁵ Although polymer binders can reduce expansion and improve electrochemical performance, intramolecular and intermolecular interactions within these polymers can lead to severe agglomeration, resulting in poor dispersion of electrode components.^29,30 Poor dispersion compromises the contact between the binder and the active material, preventing the polymer network from uniformly covering the active material and leading to uneven stress distribution during volume changes. Furthermore, many polymers are complex to synthesize, require long preparation processes, and incur high production costs, limiting their commercialization potential. In contrast, achieving bonding through in situ cross-linking during electrode preparation offers a more feasible approach, balancing the advantages of active material dispersion with the controllability of the reaction.⁶ Compared to the high cost and environmental concerns associated with many organic materials, inorganic alternatives are of low cost and eco-friendly. However, achieving in situ cross-linking polymerization of inorganic materials presents challenges that warrant further exploration to realize effective cross-linking during the preparation of silicon anodes.

Aluminum dihydrogen phosphate (Al(H₂PO₄)₃, AHP) is a highly water-soluble inorganic compound known for its excellent thermal stability and is commonly used as a high-temperature binder in metallurgical refractory materials (Fig. 1(a)).^31–35 In refractory material preparation, an AHP binder reacts with fillers such as Al₂O₃ or SiO₂ through dehydration and condensation during high-temperature sintering, effectively binding the components together.^36,37 Inspired by refractory chemistry, AHP has been innovatively applied as a binder for μSi anodes, yielding promising results (Fig. 1(b)). AHP contains multiple hydroxyl groups (–OH), which facilitate the formation of strong covalent bonds with –OH groups on the μSi surface through condensation reactions. Its excellent water solubility ensures uniform dispersion of AHP, contributing to the formation of stable and homogeneous electrodes. Regarding the challenge of μSi expansion, the high Young's modulus of the inorganic cross-linked network formed by AHP dehydration and condensation imparts significant rigidity, effectively mitigating the significant volumetric expansion of μSi anodes. The AHP-bound μSi (μSi-AHP) anodes demonstrate exceptional electrochemical performance, retaining a capacity of 1300.4 mA h g⁻¹ after 200 cycles at a current density of 0.5 A g⁻¹, compared to a much lower capacity of 161.1 mA h g⁻¹ for μSi anodes bound with carboxymethyl cellulose (CMC). AHP has also proved to be compatible with lithium iron phosphate (LiFePO₄, LFP) cathodes, enabling stable cycling for over 1000 cycles. A full cell, assembled with AHP-bound μSi anodes and AHP-bound LFP cathodes, demonstrates remarkable electrochemical performance, maintaining 79.6% capacity retention over 100 stable cycles. This study introduces a novel approach to stabilizing μSi anodes using low-cost inorganic binders, presenting new opportunities for the development of next-generation high-capacity LIBs.

Fig. 1 Inorganic AHP as a binder for silicon microparticle anodes. (a) Inspiration: AHP is commonly employed as a binder in high-temperature furnace refractories within the metallurgical industry. (b) Application to μSi anodes: AHP serves as an aqueous binder for μSi anodes, with μSi-AHP electrodes constructed by combining μSi, AHP binder, and the conductive agent acetylene black.

Results and discussion

Physicochemical characterization of the AHP binder

This study investigates the potential of the inorganic binder AHP for stabilizing μSi anodes. As demonstrated in Fig. 2(a), AHP exists as a white crystalline powder (PDF#14-0546) with excellent water solubility, dissolving to form a homogeneous white sol. Scanning electron microscopy (SEM) images of the AHP powder show an irregular morphology, with elemental mapping confirming a uniform distribution of Al, P, and O (Fig. S1a). Particle size analysis reveals a median particle diameter (D50) of around 24.4 µm (Fig. S1b). Upon thermal treatment at 60 °C for 2 h, the AHP sol undergoes a sol–gel transition, resulting in the formation of a plastic gel with notable adhesion strength, capable of withstanding weights up to 100 g. This thermally induced sol–gel transition arises from the distinctive molecular architecture of AHP. Fourier transform infrared spectroscopy (FTIR) analysis was employed to investigate the molecular structure of AHP. As indicated in Fig. 2(b), characteristic peaks are observed at 1633, 885, and 519 cm⁻¹, corresponding to the –OH, HPO₄²⁻, and P–O–Al groups, respectively, confirming the identity of AHP as an inorganic compound of Al(H₂PO₄)₃.³⁷ The underlying mechanism involves temperature-driven dehydration condensation between the hydroxyl groups of adjacent AHP molecules, resulting in the formation of an interconnected polymeric network, as illustrated in Fig. 2(c). In contrast, carboxymethyl cellulose (CMC), an amorphous yellowish powder, forms a transparent sol upon aqueous dissolution (Fig. S2). The FTIR spectra of CMC display peaks at 3277, 1592, and 1019 cm⁻¹, corresponding to the –OH, –C=O, and C–O–C groups, typical of organic polymers (Fig. 2(b)).³⁸ In comparison to organic polymer binders such as CMC, which tends to aggregate during slurry preparation due to intramolecular and intermolecular interactions, the inorganic AHP demonstrates exceptional dispersibility during slurry preparation. Binder films prepared with a 1 : 1 binder–conductive agent (acetylene black) mass ratio reveal significant agglomeration in the CMC binder, whereas AHP is uniformly distributed in the films, as shown by SEM imaging (Fig. S3). This indicates AHP's capability to fabricate well-dispersed electrodes. The thermal stability of the binder is critical for ensuring the operational reliability of LIBs. AHP, a refractory binder, exhibits outstanding thermal resilience, maintaining its structural integrity up to 226 °C, significantly higher than the thermal tolerance of CMC (Fig. 2(d)). This suggests that the AHP binder is thermally stable within the operating temperature range of LIBs, which typically remains below 100 °C.

Fig. 2 Physicochemical characterization of the AHP binder. (a) Optical image and XRD analysis of the AHP powder, along with optical images of the variations leading to the bonding effect. (b) FTIR spectra of CMC and AHP powders. (c) Schematic representation of AHP dehydration condensation at high temperature to induce polymerization. (d) Thermogravimetric analysis of CMC and AHP powders. (e) Overview of the μSi-AHP electrode preparation process. FTIR spectra of (f) AHP binder and AHP + μSi, and (g) μSi and μSi + AHP binder, highlighting the hydrogen bonding interactions between AHP and μSi. (h) Schematic illustration of the bonding between the μSi and the AHP binder.

The fabrication of μSi anodes was subsequently carried out. The μSi material was characterized through SEM, XRD, and particle size distribution analysis (Fig. S4), confirming its identity as a crystalline micron-scale powder (PDF#27-1402) with a D50 of 1.66 µm. As illustrated in Fig. 2(e), electrode slurries were prepared by homogenizing μSi, the AHP binder, and acetylene black in an 8 : 1 : 1 mass ratio with deionized water, followed by controlled electrode coating. Organic polymer binders typically contain functional groups (e.g., –NH₂, –OH, and –COOH) that can form hydrogen bonds with the –OH groups on the surface of μSi, thereby improving adhesion.^39–43 Similarly, the multiple –OH groups in AHP facilitate the formation of hydrogen bonds with the –OH groups on the μSi surface during wet electrode preparation, promoting strong interfacial interactions. During the heating and drying process of the wet electrode, AHP induces the formation of Si–O–P covalent bonds through dehydration condensation with the –OH groups on the surface of μSi (Fig. 2(h)). This process strengthens the adhesion between μSi and AHP, thereby stabilizing the μSi particles. Concurrently, dehydration condensation occurs between AHP molecules, resulting in the formation of network-like polymer macromolecules. The interaction between AHP and μSi particles anchors μSi within the three-dimensional network formed by AHP polymerization, leading to the fabrication of uniform and robust μSi anodes. FTIR spectra (Fig. 2(f)) demonstrate a shift in the P–O–Al peak from 519 to 535 cm⁻¹ in AHP following μSi addition, suggesting the presence of covalent bonding. A similar shift is observed in the O–Si–O peak (1042 cm⁻¹ to 1024 cm⁻¹) in the FTIR analysis of μSi before and after incorporating AHP (Fig. 2(g)), which further confirms the formation of covalent bonding between AHP and μSi. Moreover, as shown in Fig. S5, AHP is not only highly affordable but also offers distinct advantages over common binders. Linear sweep voltammetry (LSV) test results of the binder film further demonstrate that AHP exhibits excellent electrochemical stability at low potentials (Fig. S6), further confirming its suitability as an efficient anode binder. These findings suggest that the low-cost AHP binder has the potential to prepare μSi anodes with excellent dispersion and strong bonding, ensuring enhanced electrode stability and performance.

Physicochemical characterization of μSi-AHP anodes

The choice of binder significantly influences the microstructure and mechanical properties of electrodes. To investigate these effects, viscosity and modulus tests were first performed on the μSi-CMC and μSi-AHP anode slurries, revealing that the μSi-AHP slurry exhibits exceptional stability (Fig. S7). Moreover, the AHP slurry demonstrates superior dispersion stability; after standing for 12 h, the CMC slurry showed significant sedimentation, while the AHP slurry exhibited only minimal sedimentation (Fig. S8). To further explore the influence of different binders, a series of characterization studies were conducted on μSi-AHP and μSi-CMC anodes. SEM analysis was used to compare the micromorphology of the two electrodes, revealing significantly different morphological characteristics. In the μSi-CMC anodes, the μSi particles exhibit uneven distribution, with CMC showing noticeable agglomeration. Correspondingly, the elemental mapping for Si/C/O/Na also demonstrates a heterogeneous distribution (Fig. 3(a)). This is attributed to the tendency of CMC, an organic polymer binder, to agglomerate due to intramolecular and intermolecular interactions. In contrast, the SEM images and elemental mapping of the μSi-AHP anodes (Fig. 3(b)) show a smooth, uniform surface with a homogeneous distribution of μSi particles and Si/C/O/Al elements, indicating superior structural consistency. The improved dispersion in the μSi-AHP anodes is attributed to the favorable water solubility of the inorganic AHP binder and its bonding mechanism, which involves in situ dehydration condensation. X-ray photoelectron spectroscopy (XPS) analysis of the Si 2p region for both μSi-CMC and μSi-AHP anodes (Fig. S9) reveals nearly identical spectra, suggesting that AHP does not alter the structure of the μSi. To assess the mechanical stability of the μSi-CMC and μSi-AHP anodes, folding tests were conducted, and the results are shown in Fig. 3(c) and (d). After folding the electrodes twice at the same angle, cracks appear on the surface of the μSi-CMC anodes, while the μSi-AHP anodes remain almost unchanged, indicating better bond strength and elasticity of the AHP binder compared to CMC. To conduct a more comprehensive investigation into the mechanical properties of the two electrodes, atomic force microscopy (AFM) analyses were performed. As shown in Fig. 3(e) and (f), the AHP film exhibits a significantly higher average Young's modulus than the CMC film (3 GPa for the CMC film vs. 14 GPa for the AHP film), with the corresponding force curves confirming that the AHP film possesses greater rigidity than the CMC film. This higher rigidity suggests that the inorganic AHP binder contributes to mitigating the volume expansion of μSi particles during cycling, thereby preserving the electrode's structural integrity. Furthermore, the proportion of irreversible plastic network in the AHP binder is found to be lower than that in the CMC binder, as demonstrated by the energy dissipation profiles [CMC film (51 keV) vs. AHP film (17 keV)] (Fig. 3(g)). This result indicates that the energy resulting from Si deformation is more effectively dissipated over time through the synergistic motion of the AHP binder. The SEM images of the μSi-CMC and μSi-AHP anodes after initial complete lithiation (Fig. S10) further confirm that the rigidity of AHP helps suppress electrode degradation caused by the significant volume expansion of μSi during lithiation. The adhesion strength to the copper current collector was evaluated through a stripping test (Fig. S11). The μSi-AHP anodes exhibit a stripping force of 6.36 N, significantly higher than the 3.02 N observed for the μSi-CMC anodes (Fig. 3(h)). This enhanced adhesion reduces the risk of μSi detachment due to volume expansion, ensuring a higher proportion of electrochemically active μSi and improving the overall performance of the electrode. In summary, the μSi-AHP anodes demonstrate excellent mechanical stability, effectively mitigating the adverse effects of substantial volume changes in μSi particles. This results in improved structural integrity, suggesting that the μSi-AHP anodes hold promise for achieving superior electrochemical performance in LIBs.

Fig. 3 Physicochemical characterization and mechanical property evaluation of the μSi-AHP anode. SEM images and corresponding elemental mapping of (a) μSi-CMC and (b) μSi-AHP anodes. Mechanical strength and adhesion characteristics assessed through folding tests for (c) μSi-CMC and (d) μSi-AHP anodes. Comparison of Young's modulus and force–displacement curves for (e) CMC and (f) AHP films. (g) Energy dissipation mapping comparison of CMC and AHP films. (h) Schematic of the peeling test and comparison of peeling forces for μSi-CMC and μSi-AHP anodes.

Stable lithium storage capacity of the μSi-AHP anodes

To evaluate the lithium storage capacity of the μSi-AHP anodes, they were paired with lithium metal anodes to assemble half-cells for electrochemical testing, with μSi-CMC cells used as a comparison. Fig. 4(a) shows the initial charge/discharge curves for both electrodes at a current density of 0.1 A g⁻¹. The μSi-AHP anodes display an initial charge capacity of 2815.4 mA h g⁻¹, with an initial coulombic efficiency (ICE) of 88.25%. In comparison, the μSi-CMC anodes exhibit a lower initial charge capacity of 2117.3 mA h g⁻¹ and an ICE of 86.35%, indicating that the μSi-AHP anodes offer superior lithium storage capacity and reversibility. This improvement can be attributed to the rigid AHP binder, which effectively mitigates electrode structural degradation caused by the pronounced volume expansion of μSi during lithiation, thereby enabling both high capacity and excellent reversibility. Subsequent rate capability tests were conducted at varying current densities. As shown in Fig. 4(b), the μSi-AHP anodes demonstrate superior rate performance compared to the μSi-CMC anodes. Notably, the capacity of the μSi-CMC anodes decreases dramatically at current densities of 0.1, 0.2, and 0.5 A g⁻¹, which can be attributed to the deactivation of the active material and intensified interfacial side reactions resulting from the large volume expansion of the μSi particles. In contrast, the μSi-AHP anodes maintain a stable and reversible lithium storage capacity at each current density. The average rate capacities of both electrodes at various current densities are presented in Fig. 4(c), where the μSi-AHP anodes retain average capacities of 1868.2, 1499.9, 1186.4, 936.4, and 769.1 mA h g⁻¹ at current densities of 1, 2, 3, 4, and 5 A g⁻¹, respectively. In comparison, the μSi-CMC anodes exhibit much lower capacities of 716.6, 388.2, 177.3, 58.0, and 20.2 mA h g⁻¹ at the same current densities. Furthermore, when comparing capacity retention at various current densities to that at 0.1 A g⁻¹, the μSi-AHP anode retains 28% of its capacity at 5 A g⁻¹, while μSi-CMC anode retains only 0.8% at 5 A g⁻¹. The charging/discharging curves of the rate test more intuitively reflect the superior rate performance of μSi-AHP over μSi-CMC (Fig. S12). The cycling performance of both electrodes was evaluated at 0.5 A g⁻¹ (Fig. 4(d)). The μSi-AHP anodes show a discharge capacity decay from 2268.8 mA h g⁻¹ to 1659.7 mA h g⁻¹ after 100 cycles (73.2% retention), maintaining 1300.4 mA h g⁻¹ after 200 cycles. In contrast, the μSi-CMC anodes experience severe capacity degradation, declining from 1628.7 mA h g⁻¹ to 339.8 mA h g⁻¹ (20.9% retention) after 100 cycles, and further deteriorating to 161.1 mA h g⁻¹ after 200 cycles. This difference in cycling stability is confirmed by the charge/discharge voltage profiles (Fig. 4(e) and (f)), with the μSi-AHP anodes showing superior electrochemical reversibility. The midpoint voltage during cycling, as shown in Fig. 4(g), provides further insights into the performance differences. The polarization of the μSi-CMC anodes increases significantly, reaching 0.49 V after 200 cycles, while the μSi-AHP anodes maintain a stable polarization voltage of only 0.25 V. The reduced polarization in the μSi-AHP anodes suggests that the AHP binder effectively mitigates the adverse effects of volume expansion in μSi particles, which would typically lead to pulverization, increased SEI formation, and more severe interfacial side reactions in conventional binders like CMC. The dQ/dV curves during cycling are compared for both electrodes to assess the impact of the binder on μSi alloying and dealloying (Fig. 4(h)). The μSi-AHP anodes exhibit stable peaks for silicon alloying and dealloying, indicating excellent reversibility of lithium storage. In contrast, the μSi-CMC anodes show a gradual decrease in their alloying and dealloying peaks, with redox peaks nearly disappearing after 50 cycles, suggesting a loss of lithium storage capability. Even when the binder content is reduced to 2.5 wt% or 5 wt%, the μSi-AHP anodes retain superior cycling performance relative to the CMC-based system under the same conditions (Fig. S13–S14). Moreover, at high areal loadings (about 1.5 mg cm⁻²), the μSi-AHP anodes still deliver stable cycling behavior, significantly outperforming their CMC counterparts (Fig. S15). To further understand the electrochemical performance, electrochemical impedance spectroscopy (EIS) was employed to compare the kinetic behavior of the two electrodes. As shown in Fig. 4(i), both electrodes exhibit similar resistances before cycling. However, post-cycling, the high-frequency region exhibits distinct semicircles, attributed to the resistance (R_SEI) and charge transfer (R_ct) of the SEI layer. Notably, the impedance value of the μSi-CMC anodes is significantly higher than that of the μSi-AHP anodes. The EIS fitting results in Fig. 4(j) further highlight this discrepancy, with μSi-CMC exhibiting R_SEI and R_ct values of 233.4 Ω and 356.9 Ω, respectively, compared to significantly lower values of 102.4 Ω and 245 Ω for the μSi-AHP anodes. The higher R_SEI value for the μSi-CMC anodes is attributed to the volume expansion of μSi particles during cycling, leading to cracking and pulverization, and the constant formation of new SEI layers. The larger R_ct value is due to the deactivation of μSi particles bound with CMC, which results in the loss of lithium storage capability. In contrast, the smaller impedance values observed for the μSi-AHP anodes after cycling indicate that the AHP binder mitigates the extensive expansion of μSi anodes, ensuring a stable and reversible lithium storage capacity. Additionally, galvanostatic intermittent titration technique (GITT) results (Fig. S16) reveal that the Li⁺ diffusion coefficients in the μSi-AHP anodes remain more stable throughout cycling compared to those in the μSi-CMC anodes, further indicating better structural and kinetic stability. In summary, the AHP binder effectively prevents the large volume expansion typically associated with the μSi anodes, preserving the structural integrity of the electrode during cycling and thereby demonstrating superior electrochemical performance compared to the CMC binder.

Fig. 4 Stable lithium storage capacity of the μSi-AHP anodes. (a) Initial charge/discharge curves of μSi-CMC and μSi-AHP electrodes. (b) Rate performance comparison and (c) rate average capacity between μSi-CMC and μSi-AHP electrodes. (d) Cycling performance at 0.5 A g⁻¹ of both μSi-CMC and μSi-AHP electrodes. Corresponding charge–discharge profiles of (e) μSi-CMC and (f) μSi-AHP extracted from (d). (g) Midpoint voltage during cycling at 0.5 A g⁻¹ of μSi-CMC and μSi-AHP electrodes. (h) The dQ/dV plots of μSi-CMC and μSi-AHP electrodes over 200 cycles. (i) EIS of fresh and cycled cells corresponding to μSi-CMC and μSi-AHP electrodes. (j) EIS fitting results for both electrodes after cycling.

Stress distribution and evolution of μSi anodes revealed by finite element simulations

To better elucidate the role of AHP in stabilizing μSi anodes, finite element simulations were conducted using COMSOL Multiphysics 6.2 software to illustrate the expansion and stress changes of silicon during the lithiation process. In this simulation, spherical silicon particles were initially modeled as randomly distributed in CMC and AHP media, respectively. The resulting expansion and stress behaviors of the silicon particles during lithiation were then analyzed for both binder systems. Upon Li⁺ insertion, adjacent silicon particles undergo volumetric expansion, causing an extrusion effect that generates localized stress concentrations. Notably, the expansion of silicon particles in the CMC medium (Fig. 5(a)) is found to be more pronounced than that in the AHP medium (Fig. 5(c)). This heightened expansion in the CMC medium leads to a significant increase in the stress generated during lithiation, as shown in Fig. 5(b). In contrast, the stress generated in the AHP medium (Fig. 5(d)) remains substantially lower, owing to the inherent rigidity and cross-linked structure of the AHP binder. These results indicate that the AHP binder effectively limits the expansion of the silicon anode during lithiation, reducing stress generation and preserving the mechanical stability of the electrode. This enhanced mechanical stability ultimately translates into improved electrochemical performance and stability of the μSi anodes, highlighting the potential of AHP as a superior binder for high-capacity LIBs.

Fig. 5 Expansion and stress distribution of the μSi anodes during the lithiation process, simulated using finite element analysis. The expansion and stress distribution of μSi particles with (a), (b) CMC and (c), (d) AHP binders at different lithiation stages, where (a), (c) depict expansion and (b), (d) illustrate stress distribution.

Mechanistic analysis of μSi stabilization by the AHP binder

Post-cycling analysis of μSi anodes was conducted to gain deeper insights into the impact of the binder on electrode performance. After cycling with μSi-CMC anodes, noticeable detachment of the active material occurred, exposing the copper collector at the electrode edges. The central region of the electrode appears rough and uneven, while the separator shows significant staining due to the detached active material. SEM images reveal a highly undulating surface with numerous cracks, and 3D profilometry confirms the extreme surface roughness (Fig. 6(a) and Fig. S16). This result is attributed to the extensive volume expansion of the μSi particles during cycling, which the CMC binder was unable to constrain, leading to electrode degradation, cracking, and severe shedding of the active material. In contrast, as shown in Fig. 6(b), the μSi-AHP anodes maintain their structural integrity after cycling, with minimal staining on the separator. SEM images of the surface indicate a smooth and even electrode, further supported by 3D profilometry (Fig. 6(b) and Fig. S17), which reveals minimal surface deformation. These results suggest that the AHP binder effectively mitigates μSi particle expansion, preserving the mechanical integrity of the anode and enabling stable, reversible cycling behavior. To quantitatively assess the effect of the binder on μSi expansion, electrode cross-sections were analyzed before and after cycling. The thickness of the μSi-CMC anodes expands from 11.8 µm to 34.2 µm after 10 cycles, accompanied by significant detachment of the active material from the copper collector (Fig. 6(c)). In comparison, the thickness of μSi-AHP anodes increases from 15.4 µm to 22.2 µm after 10 cycles, with much less active material loss (Fig. 6(d)). The expansion of the μSi-CMC anodes is calculated to be 189.8%, significantly higher than the 44.2% expansion observed for the μSi-AHP anodes (Fig. 6(e)), indicating that AHP substantially reduces the expansion of μSi during cycling. Furthermore, analysis of the porosity of the electrode in its initial state, after lithiation, and after cycling indicates that the μSi-AHP anodes exhibit significantly lower porosity compared to the μSi-CMC anodes following lithiation and cycling (Fig. S18–S20). This demonstrates that AHP effectively maintains electrical contact within the electrode during the cycling process, thereby enabling stable lithiation and delithiation. Additionally, as shown in Fig. S21, analysis of post-cycle SEM images of electrodes with high loading further highlights the positive role of the AHP binder in maintaining electrode integrity and ensuring electrical contact, whereas the CMC binder exhibits more severe structural failure than in electrodes with low loading. Transmission electron microscopy (TEM) images of the μSi particles after cycling reveal stark differences in particle integrity. The μSi particles in the μSi-CMC anodes exhibit pronounced cracking, with a thick and unevenly distributed SEI (Fig. 6(f)), while the μSi-AHP anode particles remain intact, with a thin and uniform SEI layer (Fig. 6(g)). This suggests that excessive expansion of μSi particles in μSi-CMC anodes leads to cracking and fragmentation, exacerbating interfacial side reactions and promoting uncontrolled SEI growth. In contrast, the AHP binder effectively preserves μSi particle integrity, mitigating these detrimental effects. XPS was utilized to characterize the interfacial properties of the μSi anodes after cycling. The F 1s spectra from the post-cycling electrodes show that the μSi-AHP anodes exhibit a greater percentage of LiF (at a peak of 684.8 eV) compared to the μSi-CMC anodes (Fig. 6(h)). This suggests that the μSi-CMC anodes undergo a greater number of side reactions due to the expansion and pulverization of μSi particles during cycling. Additionally, analysis of the Li 1s spectra (Fig. S22) reveals the presence of a lithium metal (52.4 eV) signal for μSi-CMC anodes, alongside a Li₂CO₃ signal (55.3 eV), indicating that the lithium storage capacity of the μSi-CMC anodes is reduced, by depositing Li⁺ directly on the electrodes instead of alloying with silicon. In contrast, the μSi-AHP anodes show only the Li₂CO₃ signal, indicating that the μSi-AHP anodes retain their high lithium storage capacity even after cycling. Based on these findings, the evolution of μSi-CMC and μSi-AHP anodes during cycling is schematized in Fig. 6(i) and (j). After cycling, the μSi-CMC anodes undergo dramatic expansion and pulverization, with some active silicon particles losing electrical contact (Fig. 6(i)). In contrast, the μSi-AHP anodes effectively suppress silicon particle expansion and maintain electrode integrity due to the higher Young's modulus of AHP and the stronger covalent bonding between AHP and μSi (Fig. 6(j)).

Fig. 6 AHP binder effectively stabilizes the μSi anode by mitigating expansion. Optical, SEM and 3D optical profiler images of (a) μSi-CMC and (b) μSi-AHP electrodes after cycling. Cross-sectional SEM images of (c) μSi-CMC and (d) μSi-AHP electrodes before and after cycling. (e) Comparison of the expansion rate of the two electrodes after cycling. TEM images of μSi particles in (f) μSi-CMC and (g) μSi-AHP electrodes after cycling. (h) F 1s XPS spectrum of both electrodes after cycling. Schematic diagrams illustrating the mechanism of (i) CMC and (j) AHP binders during the operation of μSi anode electrodes.

Electrochemical performance of the μSi-AHP‖LFP-AHP full cell

The potential of AHP as a binder for cathodes was also evaluated, with promising results indicating its compatibility with lithium iron phosphate (LFP) cathodes and good cycling stability. As demonstrated in Fig. 7(a) and (b), spherical LFP particles with well-preserved crystal structures (PDF#40-1499) were utilized to evaluate the suitability of AHP as a binder for the cathode. The LFP electrode prepared with the AHP binder is referred to as LFP-AHP, while the electrode prepared using a conventional polyvinylidene fluoride (PVDF) binder (LFP-PVDF) is used for comparison. The elemental mapping of the LFP-AHP cathodes reveals a uniform distribution of Fe/P/O/Al (Fig. 7(d)), whereas the LFP-PVDF cathodes show a relatively inhomogeneous distribution of Fe/P/O/F (Fig. 7(c)). This comparison indicates that the distribution of LFP particles is more uniform within the LFP-AHP cathodes, which can be attributed to the excellent water solubility of the AHP binder, thereby facilitating the even dispersion of components during electrode preparation. To further investigate the interactions between AHP and LFP, FTIR spectroscopy was conducted on LFP, AHP, and the AHP/LFP composite. The FTIR spectra reveal new peaks corresponding to P–O–P and P–O–H bonds in the AHP/LFP composite (Fig. S23), which are absent in the individual LFP and AHP spectra. This suggests that the –OH groups in AHP may form covalent bonds with the PO₄³⁻ in LFP, indicating a chemical interaction between AHP and LFP that enhances electrode stability. Cycling performance comparisons at a 0.5C (1C = 170 mA g⁻¹) current density (Fig. 7(e)) further highlight the advantages of the LFP-AHP cathodes. After 1000 cycles, the LFP-AHP cathodes retain a capacity of 133.9 mA h g⁻¹, with a high-capacity retention of 96.5%. In contrast, the capacity of the LFP-PVDF cathodes gradually decreases after 500 cycles, and the capacity is only 112.2 mA h g⁻¹ after 711 cycles. The dQ/dV curves reveal a gradual increase in polarization for the LFP-PVDF cathodes with continuous cycling, whereas polarization in the LFP-AHP cathodes remains nearly unchanged (Fig. S24), indicating that the AHP binder enhances the operational stability of the LFP cathodes. The AHP binder can be used for both μSi anodes and LFP cathodes, showing its compatibility with both electrodes. Next, full cells were assembled using the μSi-AHP anodes and LFP-AHP cathodes to evaluate the practical value of the AHP binder, and a schematic of the full cell is shown in Fig. 7(f). The full cell assembled using the μSi-CMC anodes matched with the LFP-PVDF cathodes is used as a comparison. It can be seen that the μSi-AHP‖LFP-AHP full cell exhibits significantly better cycling performance than the μSi-CMC‖LFP-PVDF full cell (Fig. 7(g)). The capacity retention rate of 79.6% after 100 cycles for the μSi-AHP‖LFP-AHP full cell is much higher than the 11.7% for the μSi-CMC‖LFP-PVDF full cell. The charge/discharge curves of the μSi-AHP‖LFP-AHP full cell (Fig. 7(h)) clearly show its considerable cycling performance. The AHP binder can be adapted to both μSi anodes and LFP cathodes, demonstrating universality and advancement in the field. This work contributes to the development of new ideas for the preparation of high-performance electrodes.

Fig. 7 AHP binder enables stable operation of the LFP-AHP electrode. (a) XRD and (b) SEM images along with elemental mapping plots of the LFP cathode materials employed. SEM images of the (c) LFP-PVDF and (d) LFP-AHP electrodes and the corresponding elemental mapping. (e) Cycling performance of the LFP-PVDF and LFP-AHP electrodes at 0.5C. (f) Schematic diagram of LFP-AHP‖μSi-AHP full cell. (g) Full-cell cycling performance of LFP-PVDF‖μSi-CMC and LFP-AHP‖μSi-AHP. (h) Charge–discharge curves of LFP-AHP‖μSi-AHP full-cell cycling performance.

Conclusions

In summary, the inorganic AHP binder, commonly used in metallurgical refractory materials, has been successfully applied to address the volume expansion challenges of μSi anodes. During electrode fabrication, AHP undergoes dehydration and condensation at elevated temperatures, leading to the formation of an inorganic cross-linked network. Covalent bonding between μSi and AHP ensures that μSi are securely embedded within this network, leading to homogeneous and stable μSi-AHP anodes. The rigidity of AHP, with its high Young's modulus, effectively suppresses the expansion of μSi during cycling, thereby preserving electrode integrity and ensuring stable electrical contact. Consequently, μSi anodes bound with AHP exhibit exceptional cycling stability, retaining 1300.4 mA h g⁻¹ after 200 cycles at 0.5 A g⁻¹, along with impressive rate capabilities of 936.4 mA h g⁻¹ and 769.1 mA h g⁻¹ at 4 A g⁻¹ and 5 A g⁻¹, respectively. Notably, AHP also exhibits compatibility with LFP cathodes, and LFP cathodes using the AHP binder remain stable for more than 1000 cycles. Full cells constructed with μSi anodes and LFP cathodes, both utilizing AHP as the binder, show promising cycling stability. By expanding the range of potential binders for μSi anodes, the unique bonding mechanism and rigidity of AHP substantially enhance the electrochemical stability of μSi anodes, offering a promising pathway toward the development of high-performance, cost-effective LIBs.

Author contributions

F. W. conceived ideas, designed experiments and edited article drafts. J. Z. executed and analyzed all the main electrochemical experiments and wrote the original draft. S. W. reviewed and edited article drafts. Y. L. and Q. L. helped with the electrochemical tests and edited article drafts.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information is available. Supplementary information: Experimental Section, Fig. S1–S24. See DOI: https://doi.org/10.1039/d5ee03194j.

Acknowledgements

The authors gratefully acknowledge the National Natural Science Foundation of China (no. 22479162, 51904344 and 52172264) and the Science and Technology Innovation Program of Hunan Province (no. 2023RC1016).

References

1. N. Nitta, F. Wu, J. T. Lee and G. Yushin, Mater. Today, 2015, 18, 252–264.
2. F. Wu, J. Maier and Y. Yu, Chem. Soc. Rev., 2020, 49, 1569–1614.
3. J. W. Choi and D. Aurbach, Nat. Rev. Mater., 2016, 1, 16013.
4. D. Li, F. Chu, Z. He, Y. Cheng and F. Wu, Mater. Today, 2022, 58, 80–90.
5. Y. Yu, C. Yang, J. Zhu, B. Xue, J. Zhang and M. Jiang, Angew. Chem., Int. Ed., 2025, 64, e202418794.
6. L. Wang, Z. Song, Y. Li, Y. Huang, H. Zhang, Z. Yin, J. Xiao, C. Zhu, Y. Zhao, M. Zhang, T. Liu, F. Pan and L. Yang, Matter, 2025, 8, 101952.
7. T. Zhu, H. Sternlicht, Y. Ha, C. Fang, D. Liu, B. H. H. Savitzky, X. Zhao, Y. Lu, Y. Fu, C. Ophus, C. Zhu, W. Yang, A. M. M. Minor and G. Liu, Nat. Energy, 2023, 8, 129–137.
8. P. Zhou, P. Xiao, L. Pang, Z. Jiang, M. Hao, Y. Li and F. Wu, Adv. Funct. Mater., 2025, 35, 2406579.
9. X. Wan, C. Kang, T. Mu, J. Zhu, P. Zuo, C. Du and G. Yin, ACS Energy Lett., 2022, 7, 3572–3580.
10. C. Sun, H. Zhang, P. Mu, G. Wang, C. Luo, X. Zhang, C. Gao, X. Zhou and G. Cui, ACS Nano, 2024, 18, 2475–2484.
11. Z. Li, Y. Zhang, T. Liu, X. Gao, S. Li, M. Ling, C. Liang, J. Zheng and Z. Lin, Adv. Energy Mater., 2020, 10, 1903110.
12. T. Liu, Q. Chu, C. Yan, S. Zhang, Z. Lin and J. Lu, Adv. Energy Mater., 2019, 9, 1802645.
13. B. Chen, D. Xu, S. Chai, Z. Chang and A. Pan, Adv. Funct. Mater., 2024, 34, 2401794.
14. Y. Liu, Z. Tai, T. Zhou, V. Sencadas, J. Zhang, L. Zhang, K. Konstantinov, Z. Guo and H. K. Liu, Adv. Mater., 2017, 29, 1703028.
15. P. Mu, S. Zhang, H. Zhang, J. Li, Z. Liu, S. Dong and G. Cui, Adv. Mater., 2023, 35, 2303312.
16. Z. Li, M. Han, P. Yu, J. Lin and J. Yu, Nano-Micro Lett., 2024, 16, 98.
17. S. Ou, C. Liu, R. Yang, W. Fan, Z. Xie, B. Li, T. Meng, C. Zou, D. Shu and Y. Tong, Energy Storage Mater., 2024, 73, 103814.
18. C. Xu, P. Jing, Z. Deng, Q. Liu, Y. Jia, X. Zhang, Y. Deng, Y. Zhang and W. Cai, Energy Storage Mater., 2025, 74, 103911.
19. Z. Zhao, F. Chen, J. Han, D. Kong, S. Pan, J. Xiao, S. Wu and Q.-H. Yang, Adv. Energy Mater., 2023, 13, 2300367.
20. A.-M. Li, Z. Wang, T. Lee, N. Zhang, T. Li, W. Zhang, C. Jayawardana, M. Yeddala, B. L. Lucht and C. Wang, Nat. Energy, 2024, 9, 1551–1560.
21. T. Liu, T. Dong, M. Wang, X. Du, Y. Sun, G. Xu, H. Zhang, S. Dong and G. Cui, Nat. Sustain., 2024, 7, 1057–1066.
22. Z. Li, Z. Wan, Z. Lin, M. Zheng, J. Zheng, S. Qian, Y. Wang, T. Song, Z. Lin and J. Lu, Energy Environ. Sci., 2025, 18, 2365–2380.
23. Z. Liu, Y. Wang, G. Liu, X. Yue, Z. Shi, Y. Tan, J. Zhao, Y. Lei, X. Yan and Z. Liang, J. Am. Chem. Soc., 2024, 146, 34491–34500.
24. S. Choi, T.-W. Kwon, A. Coskun and J. W. Choi, Science, 2017, 357, 279–283.
25. C. Wang, H. Wu, Z. Chen, M. T. McDowell, Y. Cui and Z. Bao, Nat. Chem., 2013, 5, 1042–1048.
26. Z. Xu, J. Yang, T. Zhang, Y. Nuli, J. Wang and S.-I. Hirano, Joule, 2018, 2, 950–961.
27. B. Zhang, Y. Dong, J. Han, Y. Zhen, C. Hu and D. Liu, Adv. Mater., 2023, 35, 2301320.
28. W. Yan, S. Ma, Y. Dong, M. Cao, S. Chen, Y. Li, Y. Lu, L. Chen, Q. Huang, Y. Su, F. Wu and N. Li, Energy Storage Mater., 2024, 73, 103852.
29. J. Feng, D. Wang, Q. Zhang, J. Liu, Y. Wu and L. Wang, ACS Appl. Mater. Interfaces, 2021, 13, 44312–44320.
30. J. Feng, M. Hou, Q. Zhang, D. Wang, Z. Li, J. Liu, Y. Wu and L. Wang, J. Colloid Interface Sci., 2023, 634, 621–629.
31. Z. Dong, L. Zhang, K. Yang, Z. Fang, Y. Ni, Y. Li and C. Lu, Materials, 2024, 17, 4542.
32. A. Mocciaro, M. S. Conconi, N. M. Rendtorff and A. N. Scian, J. Therm. Anal. Calorim., 2021, 144, 1083–1093.
33. S. Tang, Y. Yang, Z. Fan, L. Yang, Z. Yang, Q. Ling and P. Wang, Ceram. Int., 2022, 48, 7963–7974.
34. X. Xu, J. Zhang, P. Jiang, D. Liu, X. Jia, X. Wang and F. Zhou, Ceram. Int., 2022, 48, 864–871.
35. Y. Wu, M. Wang, Z. Sun and D. Zhang, Case Stud. Constr. Mater., 2024, 20, e03275.
36. H. Wei, T. Wang, Q. Zhang, Y. Jiang and C. Mo, J. Chin. Chem. Soc., 2020, 67, 116–124.
37. N. Somers, A. Monton, E. Ozmen and M. D. Losego, J. Am. Ceram. Soc., 2024, 107, 36–46.
38. S. Chen, H. Zhu, J. Li, Z.-W. Yin, T. Chen, X. Yao, W. Zhao, H. Xue, X. Jiang, Y. Li, H. Ren, J. Chen, J.-T. Li, L. Yang and F. Pan, Angew. Chem., Int. Ed., 2025, 64, e202423796.
39. Z. Wu, Y. Ma, S. Li, L. Que, H. Chen, F. Hao, X. Tao, H. Xing, J. Ye, D. Qian, M. Ling, W. Zhu and C. Liang, Small, 2024, 20, 2401345.
40. B. Sun, X. Jiao, J. Liu, R. Qiao, C. Mao, T. Zhao, S. Zhou, K. Shi, M. Ravivarma, J. Shi, H. Fan and J. Song, Nano Lett., 2024, 24, 7662–7671.
41. Z. Li, Z. Wan, X. Zeng, S. Zhang, L. Yan, J. Ji, H. Wang, Q. Ma, T. Liu, Z. Lin, M. Ling and C. Liang, Nano Energy, 2021, 79, 105430.
42. L. Hu, X. Zhang, P. Zhao, H. Fan, Z. Zhang, J. Deng, G. Ungar and J. Song, Adv. Mater., 2021, 33, 2104416.
43. J. Kim, J. Choi, K. Park, S. Kim, K. W. Nam, K. Char and J. W. Choi, Adv. Energy Mater., 2022, 12, 2103718.

---