Phosphorus-based anodes for fast-charging lithium-ion batteries: advances, challenges and prospects

RP

Structural characteristics and classification of RP

RP is commonly regarded as an amorphous allotrope of P, characterized by structurally complex, disordered polymeric networks with limited long-range order. However, RP is not a single-phase material but rather a family of thermodynamically distinct allotropes that differ in structural order and thermal stability. Based on these differences, early studies classified RP into five distinct types (Types I-V)^[62]. These phases are predominantly determined by synthesis parameters, polymer chain dimensions, and the chemical composition of terminal functional groups^[63]. RP is typically synthesized from WP using high-temperature thermal treatment or electromagnetic radiation. The thermal method involves heating WP at temperatures ranging from 150 °C to 530 °C, whereas the radiation-induced approach employs milder synthesis conditions. From the perspective of practical production, the scalability and economic viability of RP synthesis methods merit further consideration. The thermal transformation route from WP is relatively straightforward and low-cost, making it the most commonly adopted approach for bulk RP production; however, the use of WP (highly toxic and pyrophoric) introduces severe safety and environmental concerns that restrict its large-scale processing. Radiation-induced conversion provides a milder and potentially safer route, yet the requirement of specialized irradiation facilities and higher energy input raises its cost and scalability limitations.

The structure of RP has long been the subject of debate, with multiple allotropes encompassing both amorphous and crystalline forms. Among these allotropes, Type I is the only form unambiguously identified as amorphous and is the most commonly used in electrochemical energy storage due to its high reactivity and isotropic structure^[64]. Despite its widespread application, the atomic structure of amorphous RP remains controversial. Two prevailing structural models have been proposed^[65]. Model I suggests the presence of polymeric phosphorus nanotubes with pentagonal cross-sections, structurally analogous to Hittorf’s P (violet P, monoclinic system)^[66,67]. On the contrary, Model II describes a disordered layered architecture resembling orthorhombic BP^[67-70]. Shanabrook et al.^[68] provided strong support for Model II by identifying intermediate-range order in amorphous RP through systematic X-ray diffraction (XRD) and Raman spectroscopy. However, Olego et al.^[67] and Goodman et al.^[71] observed features consistent with Model I, pentagonal nanotubes, via Raman and X-ray photoelectron spectroscopy (XPS), indicating that the local structure of Type I RP may be heterogeneous and sensitive to synthesis conditions.

The crystalline RP phases, particularly Types II, IV, and V, exhibit well-defined atomic arrangements. In 1947, Roth et al.^[72] first identified Types II and III RP through differential thermal analysis, reporting three reproducible exothermic peaks corresponding to discrete phase transitions upon heating to 550 °C. More recently, Sun et al.^[73] proposed a solution-phase solvothermal strategy for synthesizing Type II RP using 1,2-diaminopropane as a nucleophilic reagent [Figure 2A]. XRD patterns in Figure 2B showed clear diffraction peaks, including a dominant diffraction peak at 15.66° (corresponding to a d-spacing of 5.66 Å), consistent with the crystallographic parameters reported by Roth et al.^[72] Meanwhile, the selected area electron diffraction (SAED) pattern confirmed the RP’s monoclinic symmetry [Figure 2C]. By combining state-of-the-art electron diffraction tomography, atomic-resolution scanning transmission electron microscopy (STEM), and powder XRD, Yoon et al.^[74] identified the triclinic unit cell of Type II RP and confirmed that the basic building blocks of Type II RP are twisted pentagonal tubes packed in a wavy arrangement. In contrast, Type III RP remains poorly characterized due to limited synthetic accessibility and insufficient structural data. Type IV and V RP, known as fibrous RP and Hittorf’s P, respectively, both exhibit structural homology, consisting of repeating polymeric tubular motifs with pentagonal cross-sections. These double tubes comprise sequential [P9]P2[P8]P2 units, as evidenced by crystallographic analysis^[75] [Figure 2D]. The principal distinction between the two lies in the packing orientation of the tubes: Type IV exhibits parallel alignment along the a-axis, whereas Type V displays perpendicular stacking along the c-axis, stabilized by van der Waals interactions^[76,77] [Figure 2E and F].

Figure 2. (A) Schematic diagram of the synthesis route of Type II RP through a solvothermal method; (B) The XRD pattern and (C) SAED pattern of the prepared Type II RP. Reproduced with permission^[73]. Copyright 2022, Royal Society of Chemistry; (D) Crystallographic representation of the repeating [P9]P2[P8]P2 tubular unit in Type IV RP. Reproduced with permission^[77]. Copyright 2017, American Chemical Society; Linkage of the P strands and stacking of the layers in (E) fibrous RP and (F) Hittorf’s P. The blue tubes are added to emphasize the fibrous nature. Reproduced with permission^[75]. Copyright 2020, Royal Society of Chemistry; Band structure, total density of states, and partial density of states of (G) monolayer hittorfene and (H) bulk Hittorf’s P based on first-principles calculations. Reproduced with permission^[79]. Copyright 2016, American Chemical Society. XRD: X-ray diffraction; SAED: selected area electron diffraction; RP: red phosphorus; P: phosphorus.

BP and phosphorene

Structural characteristics and electronic properties of BP and phosphorene

BP exhibits the highest thermodynamic stability and mass density among all known P allotropes under standard ambient conditions^[62]. Traditional methods for synthesizing bulk BP crystals typically rely on high-pressure conditions (often exceeding 1 GPa) to induce phase transitions from other P allotropes. In a pioneering study, Bridgman et al.^[83] successfully prepared bulk BP by heating molten purified WP sealed within a steel enclosure. This assembly was subjected to controlled pressure conditions of 1.2-1.3 GPa in a specialized apparatus, accompanied by thermal activation at 200 °C. The reaction system was maintained under these conditions for 5-30 min to complete the phase transformation. Despite its historical significance, this method requires expensive equipment and high energy consumption, restricting its economic viability for mass production.

Phosphorene, a single or few-layer derivative of BP, can be synthesized by using two principal strategies: top-down exfoliation and bottom-up growth. The top-down approach adapts techniques originally developed for graphene. In 2014, Liu et al.^[84] first isolated monolayer and few-layer phosphorene from bulk BP through mechanical exfoliation. The fabrication process involved iterative delamination of bulk crystals using adhesive tape, followed by precise transfer onto SiO₂/Si substrates (300 nm oxide thickness), yielding atomically thin phosphorene sheets with characteristic thicknesses approximating 0.85 nm. While this method provides high-quality flakes, it suffers from low throughput and poor thickness control. In comparison, bottom-up methods such as chemical vapor deposition (CVD) offer greater potential for scalable production^[85]. In a representative study, Smith et al.^[86] demonstrated the controlled growth of two-dimensional phosphorene on Si substrates by employing a Sn/SnI₄-mediated CVD process. The growth was performed under the thermogradient annealing from 950 °C to 600 °C at an argon pressurization of 27.2 atm. This synergistic process achieved directional laminar growth through coordinated optimization of mineralizer-assisted vapor transport and thermal gradient-controlled crystallization dynamics. In general, top-down exfoliation techniques yield high-quality phosphorene flakes but suffer from low throughput and poor reproducibility, limiting scalability. Bottom-up strategies such as CVD offer greater promise for large-scale synthesis with structural tunability, yet they often involve complex setups, high temperatures, and toxic precursors, which raise environmental and safety concerns.

Bulk BP exists in three distinct crystalline phases: orthorhombic, simple cubic, and rhombohedral structures^[65]. Through systematic XRD analysis, Jamieson revealed that BP undergoes a reversible two-step phase transformation under elevated pressure conditions^[87]. Specifically, the orthorhombic structure transforms into a rhombohedral phase at 8.3-10.2 GPa, followed by a transition to a simple cubic phase above 12.4 GPa. Upon decompression, the reverse sequence is observed: the simple cubic phase reverts to rhombohedral at 9.3-11.1 GPa, and full release of pressure restores the orthorhombic structure. These findings indicate that both simple cubic and rhombohedral phases are thermodynamically unstable under ambient conditions, thereby limiting their practical applicability. As a result, the orthorhombic phase remains the most extensively studied and technologically relevant form of BP, particularly for applications in electronics and electrochemical energy storage.

As displayed in Figure 3A and B, bulk BP exhibits a layered crystal structure analogous to graphene^[88]. Each P atom is covalently bonded to three adjacent P atoms via sp3 hybridized orbitals, forming stable, interconnected hexagonal rings^[89]. These rings collectively constitute an orthorhombic honeycomb lattice with a folded structure^[65]. Through time-of-flight neutron powder diffraction analysis, Cartz et al.^[90] found that BP crystallizes in the orthorhombic system (space group Cmca) with lattice parameters determined as a = 3.3133(9) Å, b = 10.473(5) Å, and c = 4.374(2) Å. The distinctive folded architecture of bulk BP arises from the non-coplanar arrangement of P atoms^[91] [Figure 3C]. This feature was further elucidated by Hultgren et al.^[92] via XRD coupled with Fourier analysis. The six-membered ring structure in monolayer BP exhibits evident non-planarity, with three P atoms residing within the upper lattice layer and the remaining three occupying a sublayer arrangement. This spatial organization generates a bilayer-mimetic folded structure featuring an internal corrugated structure with a 2.1 Å interlayer spacing, contrasting with the 5.25 Å spacing observed in van der Waals-stacked multilayered BP. Figure 3D demonstrates that covalent bonding within the phosphene unit cell exhibits a bond length of 2.18 Å and an average bond angle of 102°, forming folded chain structures along distinct crystallographic directions^[93]. This anisotropic bonding arrangement defines two main in-plane crystallographic directions: the armchair direction (parallel to folded chain alignment) and the zigzag orientation (perpendicular to the folding axis). Owing to its distinctive lattice symmetry, BP is an intrinsic direct bandgap semiconductor, with highly anisotropic and layer-dependent electronic characteristics^[88] [Figure 3E]. Combined with optical absorption spectroscopy and photoluminescence experiments, Li et al.^[94] resolved the layer-number dependence of phosphorene’s bandgap while quantitatively correlating this structural parameter with electronic properties through first-principles calculations. The layer-dependent electronic structure evolution arises from quantum confinement effects, which cause the conduction and valence bands to split into distinct two-dimensional sub-bands with increasing layer number. Theoretical simulations demonstrated that the direct bandgap of encapsulated phosphorene undergoes progressive reduction from 1.56 eV to 0.53 eV as layer count increases from monolayer to tetralayer structure, exhibiting quantitative agreement with experimental observations. The layer-number-dependent reduction in bandgap of BP directly correlates with increased electronic conductivity^[95,96]. This tunable electronic structure is critical for enhancing electron transfer within P-based anodes, a key requirement for achieving high-rate capability in LIBs.

Figure 3. Crystal structures of (A) graphene (planar) and (B) BP (puckered orthorhombic). Reproduced with permission^[88]. Copyright 2022, Elsevier; (C) Modeled mono-, bi-, and tri-layer BP structures, illustrating the stacking of puckered layers. Reproduced with permission^[91]. Copyright 2015, Springer Nature; (D) Top view of single-layer BP (phosphorene), showing bond angles (96.3° and 102.1°) and the anisotropic lattice directions (zigzag vs. armchair). Reproduced with permission^[93]. Copyright 2015, National Academy of Sciences; (E) Evolution of BP bandgap with sample thickness, showing the thickness-dependent bandgap change. Reproduced with permission^[88]. Copyright 2022, Elsevier. BP: Black phosphorus.

TMPs

Besides elemental P, TMPs have garnered significant attention as promising alternatives to conventional graphite anodes for fast-charging LIBs due to their moderate lithiation potentials and relatively high theoretical capacities. Based on the electrochemical reactivity of the metallic constituents with lithium, TMPs can be divided into two categories: inactive and active^[9,58].

Inactive TMPs

In this class, the metal component is electrochemically inert toward lithium and mainly functions as a conductive matrix, while the P element actively participates in lithium storage. Representative materials include FeP, CoP, Ni_xP, etc. The general lithium storage mechanism can be represented by the reversible conversion reaction^[32]:

(1) $$ \mathrm{MP}_{\mathrm{x}}+3 \mathrm{xLi}^{+}+3 \mathrm{xe}^{-} \leftrightarrow \mathrm{M}+\mathrm{xLi}_{3} \mathrm{P} $$

During lithiation, Li⁺ reacts with P to form lithium-rich phosphides (e.g., Li₃P), accompanied by the reduction of the transition metal to its metallic state. The resultant metal nanoparticles, though electrochemically inactive, construct a percolating conductive framework that improves electronic conductivity and buffers structural stress from volume fluctuations. Upon charging, partial reconversion of Li₃P and redistribution of the metal can occur, depending on particle size, crystallinity, and cycling conditions. Due to their chemical stability, metallic backbone, and structural robustness, inactive TMPs often exhibit enhanced rate capability and cycling durability, making them attractive for fast-charging scenarios^[32,99].

FeP is a typical inactive TMP, possessing a high theoretical capacity (924 mAh g^-1), moderate lithiation potential, abundant natural reserves, and low toxicity^[40,41]. The lithium storage mechanism of FeP involves a two-step reversible lithiation/delithiation process^[100-102]: FeP + Li ↔ LiFeP and LiFeP + 2Li ↔ Li₃P + Fe. Generally, FeP is synthesized via a two-step approach. Initially, a Fe-based precursor with a tailored nanostructure is prepared, which is followed by phosphorization in an inert atmosphere using a phosphorus source, resulting in the formation of FeP. The morphology and crystallinity of the final product are predominantly dictated by the precursor architecture, phosphorization temperature, and reaction atmosphere. For instance, Veluri et al.^[103] prepared FeP via a polyol-mediated route to fabricate porous Fe₂O₃ precursors, which were then mechanically mixed with NH₄H₂PO₄ and polyethylene glycol-400, followed by thermal conversion at 800 °C under an Ar/H₂ atmosphere. At a current density of 100 mA g^-1, a discharge capacity of 618 mAh g^-1 was delivered. Even at a high current density of 1 A g^-1, the discharge capacity of FeP can reach 314 mAh g^-1. Regarding the poor intrinsic electronic conductivity Li₃P product (1 × 10^-2 S m^-1)^[104], which slows reaction kinetics and causes rapid capacity degradation under high-rate cycling conditions, various nanostructured FeP-based composites have been developed. Li et al.^[105] developed a one-step phosphorization strategy that integrated vapor deposition and self-assembly processes. This approach enabled the uniform dispersion of ultrafine FeP nanoparticles (10-15 nm) within a P-doped carbon matrix anchored on MXene (Ti₃C₂) conductive networks (FeP@C/Ti₃C₂) [Figure 4A]. The Ti₃C₂ MXene enhances conductivity through its integration with the P-doped carbon network and simultaneously provides mechanical flexibility to buffer volume change during cycling. This architecture significantly enhances structural integrity and accelerates reaction kinetics. Based on these structural merits, the FeP@C/Ti₃C₂ electrode exhibited excellent cycling performance, delivering a high reversible capacity of 809.6 mAh g^-1 with 93.1% capacity retention over 1,500 cycles at a current density of 2 A g^-1 [Figure 4B]. Subsequently, Yu et al.^[106] employed a metal-organic framework (MOF) derivation method to construct biconical-nanorods-like Co-doped FeP confined by P-doped carbon composites (Co-doped FeP@P-C). Co-doping modulates the electronic structure of FeP, enhancing charge transfer kinetics and creating additional Li⁺ adsorption sites. Meanwhile, the unique bicone nanorod morphology increases contact interfaces with the electrolyte for accelerated ion/electron transfer, while the conformal P-doped carbon coating simultaneously suppresses volume change and improves conductivity and structural stability. As a result, the Co-doped FeP@P-C anode exhibited a high lithium storage capacity (784 mAh g^-1 at 0.2 A g^-1) and long-term cycling stability (555.6 mAh g^-1 at 1 A g^-1 after 700 cycles).

Figure 4. (A) Schematic illustration of the microstructure of FeP@C/Ti₃C₂ nanocomposite; (B) The cycling performance of the FeP@C/Ti₃C₂ electrode at a current density of 2 A g^-1, maintaining a capacity retention of 93.1% over 1,500 cycles. Reproduced with permission^[105]. Copyright 2022, Elsevier; (C) Schematic illustration of the fabrication process of CoP@CC; (D) The first-four charge/discharge profiles of the CoP@CC electrode at 0.1 A g^-1. Reproduced with permission^[107]. Copyright 2025, John Wiley and Sons; (E) The structure and the charge/discharge profiles of the GeP/SWCNTs electrode in the first two cycles at 0.1 A g^-1 with a potential range of 0.005-3.0 V. Reproduced with permission^[112]. Copyright 2025, Elsevier; (F) The cycling performance of the a-ZnP₂/Zn₃(PO₄)₂/P/C electrodes at a current density of 20 A g^-1. Reproduced with permission^[117]. Copyright 2024, Elsevier. CoP@CC: CoP deposited on carbon cloth electrode; SWCNTs: single-walled carbon nanotubes.

CoP is another common inactive TMP anode for LIBs due to its high theoretical capacity (894 mAh g^-1) and relatively low redox potential (~0.6 V vs. Li⁺/Li)^[42,43]. During lithiation, CoP reacts with Li⁺ to form Li₃P and metallic Co nanoparticles according to CoP + 3Li⁺ + 3e^- → Co + Li₃P. Upon delithiation, the reverse transformation restores the initial CoP phase. The in-situ generated metallic Co nanoparticles contribute to improved electrode conductivity and provide a percolated electron transport network, while the Li₃P phase serves as the active lithium host. In terms of synthesis, CoP is generally prepared via phosphorization of Co-based precursors, following strategies analogous to FeP fabrication. Yang et al.^[107] developed a self-standing CoP deposited on carbon cloth electrode (CoP@CC) through an integrated fabrication strategy involving direct electrodeposition followed by in-situ phosphorization, enabling the construction of vertically-aligned CoP nanosheet arrays on CC substrates [Figure 4C]. The CoP@CC architecture demonstrates robust interfacial adhesion between active materials and the current collector, thereby enhancing the overall electrical conductivity. The carbon cloth acts as both a mechanical support to maintain structural integrity and a spatial framework to prevent CoP nanosheet aggregation through the geometric constraint effect. Moreover, the three-dimensional hierarchical architecture composed of vertically-oriented CoP nanosheets and interconnected CC networks creates abundant void spaces that effectively accommodate volumetric fluctuations during electrochemical cycling, while simultaneously establishing shortened pathways for both electron transport and ion diffusion, which is crucial for high-rate performance. The CoP@CC electrode delivered an impressive initial discharge capacity of 1,866.9 mAh g^-1 (3.8 mAh cm^-2) with a high initial CE of 80.3% at 0.1 A g^-1 [Figure 4D]. It also exhibited excellent rate capability, maintaining 635.3 mAh g^-1 (1.3 mAh cm^-2) at 4 A g^-1, along with outstanding cycling stability, retaining 793.4 mAh g^-1 (1.6 mAh cm^-2) after 420 cycles at 2 A g^-1.

In addition, Ni_xP is a class of inactive TMPs that has garnered considerable interest as anode materials for LIBs due to their relatively high theoretical specific capacities compared to graphite (542 mAh g^-1 for Ni₂P and 468 mAh g^-1 for Ni₁₂P₅), low polarization characteristics, and good structural robustness^[44,45]. Their lithium storage mechanism primarily involves a conversion reaction, synergistically coupled with intercalation processes. Turarova et al.^[108] developed a self-supporting Ni_xP/C composite electrode through a one-pot electrospinning strategy combined with in-situ steam oxidation during carbothermal reduction. Precise control of precursor drying temperatures enabled the rational design of crystalline hydrate/nitrate composite oxidants, achieving customized surface oxide layers, optimized electronic structures, and refined microstructural regulation. This multiscale synergistic mechanism substantially enhanced the intrinsic reaction kinetics while effectively mitigating volume expansion effects during electrochemical cycling. When tested at a current density of 0.1 A g^-1, the Ni_xP/C achieved the highest initial charge capacity of 719.5 mAh g^-1 and maintains a high capacity of 590.7 mAh g^-1 after 100 cycles. To further enhance the electrochemical performance of Ni_xP, Co-doped bimetallic Ni_2-XCo_xP materials have been developed. Li et al.^[109] prepared a series of three-dimensional Ni_2-XCo_xP flower-like electrode materials by using a self-induction and self-assembly method, which showed improved electronic conductivity, ion transport kinetics, and electrochemical activity. In particular, the optimal Ni_1.2Co_0.8P anode showed excellent rate capability (808 mAh g^-1 at 0.1 A g^-1 and 137 mAh g^-1 at 5 A g^-1) and long-term cycling stability (3,000 cycles at 1 A g^-1), proving inspiration for TMPs as feasible anodes for applications in fast-charging LIBs.

Active TMPs

Unlike inactive TMPs, both the metal and P participate in reversible lithiation reactions for active TMPs. Representative materials include GeP, Sn₄P₃, ZnP₂, etc. Their lithium storage mechanism can be expressed as:

(2) $$ \mathrm{MP}_{\mathrm{x}}+(3 \mathrm{x}+\mathrm{y}) \mathrm{Li}^{+}+(3 \mathrm{x}+\mathrm{y}) \mathrm{e}^{-} \leftrightarrow \mathrm{Li}_{\mathrm{y}} \mathrm{M}+\mathrm{xLi}_{3} \mathrm{P} $$

Li⁺ ions insert into both the metal and P sites, forming lithium-rich alloys (e.g., Li_yM and Li₃P)^[32]. The dual participation not only enhances theoretical capacity but also improves electronic conductivity compared to pure P-based systems.

GeP features a high theoretical specific capacity of 1,900 mAh g^-1 and excellent electronic conductivity of ~5 × 10⁴ S cm^-1 attributed to the metallic character of Ge^[110,111]. During lithiation, GeP undergoes a multi-phase conversion reaction, forming Li₃P and Li_xGe. However, in-situ transmission electron microscopy (TEM) studies by Shen et al.^[112] revealed that GeP suffers from significant structural fracture during initial lithiation, primarily due to its large volume expansion and limited mechanical resilience. Thus, they fabricated a GeP/single-walled carbon nanotubes (SWCNTs) composite by combining few-layer GeP nanosheets with SWCNTs via ball milling and ultrasonication. The resulting hybrid architecture offers a mechanically flexible and conductive framework that accommodates volume changes and mitigates stress accumulation, enhancing lithiation reversibility and prolonging cycling life. An initial specific capacity of 778/911 mAh g^-1 with a CE of ~85% was obtained at 0.1 A g^-1 [Figure 4E]. After 200 cycles, the capacity of 456 mAh g^-1 was maintained for the GeP/SWCNTs anode in LIBs.

As an emerging anode candidate in the TMP family, Sn₄P₃ offers a synergistic combination of the high specific capacity of P with the superior electronic conductivity of Sn (1,255 mAh g^-1 and 30.7 S cm^-1)^[113]. Its layered crystal structure supports dual lithium storage mechanisms, involving both the alloying of Sn (Sn ↔ Li_xSn) and the conversion of P to Li₃P. Besides, the moderate lithiation potential (0.4 V vs. Li⁺/Li) helps reduce the risk of lithium plating during fast charging. However, the associated volume change (~200%) poses a critical challenge to structural integrity and cycling stability^[114,115]. He et al.^[116] designed a Sn₄P₃ encapsulated in P-doped porous carbon nanosheets (Sn₄P₃/P-C@CNs) composite anode through a one-step thermal treatment approach, producing ultrafine Sn₄P₃ nanoparticles uniformly embedded within the conductive P-doped carbon framework. The resulting architecture provides a mechanically robust and electrically continuous matrix that effectively buffers volumetric fluctuations, preserves electrode integrity, and accelerates charge transport, thereby enabling enhanced cycling stability and rate capability. Consequently, the Sn₄P₃/P-C@CNs composite electrode delivered a high reversible capacity of 770 mAh g^-1 at 0.1 A g^-1 upon 100 cycles and exhibited 717 mAh g^-1 at 1 A g^-1 after 1,000 cycles.

Recently, ZnP₂ has emerged as a promising active TMP anode for fast-charging LIBs due to its high theoretical specific capacity (1,580 mAh g^-1), superior electrical conductivity, and cost efficiency. The lithium storage mechanism of ZnP₂ follows a conversion reaction of ZnP₂ + 7Li⁺ + 7e^- → LiZn + 2Li₃P, wherein both Zn and P participate in lithiation via alloying and phosphide formation, respectively. Liu et al.^[117] developed a multicomponent amorphous ZnP₂-based hybrid [a-ZnP₂/Zn₃(PO₄)₂/P/C] via a one-step high-energy ball milling route, employing ZnO, RP, and carbon black as precursors. This strategy enabled the formation of a disordered composite matrix featuring homogeneously distributed Zn, P, and conductive carbon domains. The amorphous nature of the material not only facilitates isotropic Li⁺ diffusion and stress relaxation but also suppresses phase separation and mechanical failure during cycling. Benefiting from structural and interfacial merits, the optimized a-ZnP₂/Zn₃(PO₄)₂/P/C electrode delivered an impressive reversible capacity of 985 mAh g^-1 at 5 A g^-1 with 92.3% retention over 2,200 cycles. Even under ultrafast charging conditions (20 A g^-1, 12C), it retained 734 mAh g^-1 after 2,000 cycles and 592 mAh g^-1 after 2,700 cycles [Figure 4F], highlighting its remarkable rate capability and structural robustness.

With regard to scalability and sustainability, TMP synthesis faces additional trade-offs. Conventional phosphorization approaches (using RP, PH₃, or NH₄H₂PO₄ as P sources) are effective in structural control but suffer from hazardous precursors, high temperatures, and complex reaction atmospheres, which increase both cost and environmental burden. MOF-derived routes and other nanostructuring strategies can deliver precise nanostructures with enhanced electrochemical performance, yet their reliance on costly precursors and multistep treatments challenges industrial feasibility. Ball-milling or one-step solid-state phosphorization methods provide more economical and scalable alternatives, though they often yield products with structural inhomogeneity. Moving forward, establishing low-cost, environmentally friendly, and scalable synthetic frameworks for TMPs will be vital to bridge the gap between laboratory-scale demonstrations and industrial adoption in fast-charging LIBs.

To sum up, the structural diversity and lithium storage mechanisms of P-based materials offer a rich foundation for designing high-performance anodes tailored for fast-charging LIBs. A simple comparison of these materials in terms of their structural characteristics and lithium storage properties has been summarized in Table 2. Amorphous RP features isotropic ionic pathways and structural adaptability, while crystalline RP variants provide insights into structure-dependent reactivity. BP and its few-layer derivative phosphorene exhibit anisotropic charge transport and tunable band structures governed by quantum confinement, enabling rapid kinetics and enhanced rate performance. Inactive TMPs primarily utilize the P component for reversible lithiation via conversion reactions, with the transition metal matrix acting as a conductive scaffold to buffer volume changes and maintain electrode integrity. In contrast, active TMPs leverage synergistic lithiation of both metal and phosphorus components, offering higher theoretical capacities and improved electrical conductivity. Detailed lithiation/delithiation mechanisms of various P-based anodes are illustrated in Figure 5. These insights provide a fundamental framework for further optimization of P-based anode materials, paving the way for their practical implementation in next-generation fast-charging LIBs.

Figure 5. Schematic illustrations of lithiation/delithiation mechanisms for P-based anodes: (A) RP, (B) BP, (C) inactive TMPs, and (D) active TMPs. RP: Red phosphorus; BP: black phosphorus; TMPs: transition metal phosphides; P: phosphorus.

Table 2

Comparison of RP, BP, and TMPs in terms of their structural characteristics and lithium storage properties in LIBs

Material	Sub-type	Crystal structure	Thermodynamic stability	Electronic conductivity	Volume expansion	Lithium storage mechanism
RP	Type I	Amorphous	Moderate	Extremely low (~10-14 S cm-1)	~300%	Alloying reaction
BP	Bulk	Orthorhombic	Very High	Moderate (~102 S m-1)	~200%-300%	Intercalation-alloying reaction
	Phosphorene		Low (air-unstable)	High
TMPs	Inactive	Variable (depending on material)	High	Moderate-high	＜ 100%	Conversion reaction
	Active				~150%-200%	Conversion + alloying reaction

LIBs: Lithium-ion batteries; BP: black phosphorus; RP: red phosphorus; TMPs: transition metal phosphides.

Severe volume variation

P-based anodes can undergo extreme volumetric expansion (up to ~300%) during lithiation. This pronounced volume fluctuation imposes substantial mechanical strain on the electrode architecture, leading to structural disruption and destabilization of the solid electrolyte interphase (SEI) layer. In essence, SEI is electronically insulating yet Li⁺-conductive. The mechanical integrity and continuity of SEI layers are critical for sustaining rapid Li⁺ transport and preventing continuous electrolyte decomposition, especially under fast-charging conditions. Operando electrochemical atomic force microscopy (EC-AFM), combined with ex-situ spectroscopic analyses, has revealed the dynamic fracture and reconstruction of the SEI layer during repeated lithiation/delithiation^[32,118]. Such continuous rupture-reformation not only accelerates electrolyte decomposition but also causes persistent loss of active P species and progressive thickening of the interfacial layer, ultimately resulting in increased impedance, compromised rate capability, and shortened cycling life. Additionally, repeated expansion-contraction induces delamination between active materials and current collectors, causing electrical isolation and capacity decay^[9,47]. In amorphous RP, the isotropic but substantial expansion (~300%) leads to structural pulverization and continuous disintegration due to the lack of long-range structural support^[76,77]. In BP, the anisotropic lattice results in directional expansion-significantly larger along the armchair direction than along the zigzag-aggravating interlayer slippage and local stress accumulation, which accelerates fracture propagation^[89]. Monolayer or few-layer phosphorene, though sharing the same lattice symmetry with BP, demonstrates enhanced mechanical compliance due to its reduced dimensionality. This allows for more efficient strain relaxation during lithiation, but the ultrathin nature also increases surface area exposure to electrolyte, making it vulnerable to fracture and interfacial instability when not properly protected^[91]. For TMPs, especially active types such as Sn₄P₃ and GeP, dual lithiation mechanisms involving both Li₃P formation and metal alloying contribute to combined volume expansion of ~150%-200%. While the presence of metallic frameworks can alleviate some stress and maintain partial conductivity, repeated cycling still leads to nanoparticle pulverization and electrode integrity loss^[112,116]. Collectively, these expansion-related issues represent a key bottleneck for the long-term stability of P-based anodes, particularly under fast-charging conditions.

Intrinsically low electronic conductivity

Elemental P, particularly RP, suffers from extremely low electronic conductivity (~10^-14 S cm^-1 at room temperature), which severely limits electron transfer and active material utilization. This poor conductivity leads to high charge-transfer resistance, sluggish reaction kinetics, and large polarization under high current densities, ultimately degrading rate capability and power performance. While BP and some TMPs offer better conductivity, their performance still requires significant enhancement, especially under fast-charging conditions^[48,49].

Surface passivation

Both few-layer BP flakes and phosphorene are highly susceptible to surface passivation arising from rapid chemical degradation upon exposure to ambient oxygen and moisture. Upon exposure to air, P atoms at the surface undergo spontaneous oxidation, forming P_xO_y species and phosphoric acid derivatives, which disrupt the lattice continuity and alter the local electronic environment. This degradation significantly elevates contact resistance, reduces carrier mobility, and impairs the overall charge transport capability of anode materials^[50,51].

Dissolution of polyphosphide intermediates

During the lithiation and delithiation processes of P-based anodes, a series of lithium polyphosphide species is generated as metastable intermediates. These species exhibit high solubility in conventional carbonate-based electrolytes, which leads to the continuous dissolution and migration of active P-containing moieties away from the electrode. This phenomenon results in irreversible loss of active material and severely diminishes CE and reversible capacity over extended cycles. Moreover, the dissolved polyphosphide anions can diffuse across the electrolyte and undergo redox reactions at the counter electrode, inducing a shuttle effect similar to that observed in Li-S batteries. This parasitic phenomenon not only consumes cyclable lithium but also promotes electrode corrosion and uncontrolled SEI growth, thereby accelerating capacity fading and performance degradation^[47,52].

Constructing C-P composites

As a primary strategy targeting both severe volume variation and intrinsically low electronic conductivity of P-based anodes, the integration of P with highly conductive carbon matrices has become widely adopted. The conductive carbon frameworks [e.g., graphite, graphene, carbon nanotubes (CNTs), fullerenes, etc.]^[53] not only provide electron transport pathways to address low conductivity but also act as mechanical buffers to accommodate volumetric expansion during lithiation/delithiation. For instance, graphite, a conventional anode material for LIBs, features a layered crystal structure that facilitates reversible Li⁺ intercalation/deintercalation while offering structural support to mitigate the volume changes of P. Graphene is a type of two-dimensional, monoatomic-thick carbon material with excellent conductivity and ultrahigh specific surface area.

Although structurally analogous to graphite, graphene possesses superior electronic conductivity and lithium storage capacity due to its quantum confinement effects. CNTs are one-dimensional nanocarbon materials with high conductivity and mechanical strength, which synergistically enhance lithium storage performance when hybridized with graphene through the formation of three-dimensional conductive networks. Fullerenes, another class of nanocarbons distinguished by unique cage-like structures and strong electron affinity, can be incorporated into graphene composites to expand interlayer spacing, thereby optimizing Li⁺ storage capacity. These carbon matrices demonstrate critical applications in LIBs owing to their notable electron transport capability, structural stability, and chemical inertness^[54].

P-C composites exhibit three key advantages as advanced anode materials: (1) The conductive carbon matrices form percolating networks that enhance bulk electronic conductivity, facilitating efficient electron transport to the active P-based material, which is essential for achieving high rate capability; (2) The strategic integration of carbon structures, particularly porous carbons (PC), CNTs, and graphene frameworks, enables effective buffering of volumetric variation during cycling, preserving electrode integrity and promoting the formation of stable SEI layer; (3) The hierarchical carbon matrix with optimized porosity and surface characteristics offers abundant electrolyte-permeable channels, significantly shortening ion diffusion pathways and enhancing ion transport kinetics through improved electrode-electrolyte interfacial contact improved electrode-electrolyte interfacial contact^[55].

To improve the electrochemical performance of P-based anodes, Zhang et al.^[48] adopted a vapor condensation strategy to infiltrate PC with RP (RP-PC). The abundant edge sites in the carbon matrix facilitated the physisorption and polymerization of P₄ molecules, along with the formation of stable P-C bonds [Figure 7A]. This structural design enabled a high phosphorus loading capacity (approaching 50 wt%) and enhanced electrochemical safety. Uniformly distributed RP nanoparticles with an average particle size of 15.68 nm were anchored on the PC surface, effectively mitigating stress induced by volumetric expansion during cycling. The RP-PC composite anode demonstrated notable electrochemical stability, retaining a high specific capacity of 965.2 mAh g^-1 after 1,100 cycles at a current density of 1,000 mA g^-1 [Figure 7B]. Notably, even under an ultrahigh current density of 8,320 mA g^-1 (~16.7 C), the composite maintained a capacity of 496.8 mAh g^-1 after 1,000 cycles, exhibiting superior cycling stability under extreme operational conditions. As observed by scanning electron microscopy (SEM), the RP-PC electrode maintained a smooth and compact surface after 120 cycles at 260 mA g^-1 and no obvious cracks or pulverization even after 1,100 cycles at 1,000 mA g^-1, in sharp contrast to the complete pulverization of the pristine RP electrode and severe cracking of the RP-multi-walled CNTs (RP-MWCNT) control electrode after 100 and 120 cycles at 260 mA g^-1, respectively. Meanwhile, the fresh RP-PC electrode exhibited a smaller charge-transfer resistance of ~618 Ω compared to that of ~1,037 Ω for the RP-MWCNT electrode. Additionally, both the RP-PC and RP-MWCNT electrodes demonstrated reduced charge-transfer resistances after cycling, indicating the establishment of the stable electrode-electrolyte interface. Besides, Zhang et al.^[119] synthesized a three-dimensional conductive architecture by chemically crosslinking BP with N-doped CNTs (BP@CNTs). This synthesis involved a process comprising ultrasonic dispersion of BP, urea addition for crosslinking, and incorporation of N-doped CNTs. During the reaction, the decomposition of urea releases NH⁴⁺ ions, facilitating the uniform attachment of BP nanosheets onto CNT surfaces. As shown in Figure 7C, the N-doped CNTs serve as a structural framework, providing abundant O- and N-containing functional groups that enhance interfacial charge transfer and stabilize BP anchoring. As a result, the charge-transfer resistance of the BP@CNTs composite (88.9 Ω) is much lower than that of BP (415.0 Ω). Benefiting from the supporting and network-skeleton role of CNTs, the BP@CNTs composite electrode maintained good morphology and structure after repeated lithiation/delithiation processes. Electrochemical tests further verified that the BP@CNTs composite anode demonstrated excellent rate capability and cycling stability. Specific capacities of 420 and 380 mAh g^-1 were delivered even at high current densities of 1,500 and 2,500 mA g^-1 [Figure 7D]. Meanwhile, it maintained a specific capacity of approximately 750 mAh g^-1 after 100 cycles at 100 mA g^-1, significantly outperforming both pure CNT electrodes and pristine BP electrodes. Notably, the composite exhibited nearly 100% CE over cycling. In addition, Liu et al.^[120] developed a low-temperature molten salt system (NaCl-KCl-AlCl₃) employing a controlled phosphatization strategy to synthesize peapod-like CoP@C nanostructures at 250°C. This method utilized industrial-grade PCl₃ as the P precursor and Mg as the reducing agent. High-resolution TEM (HRTEM) images revealed that the CoP@C nanoparticles with diameters below 50 nm were uniformly encapsulated within a one-dimensional CNT shell, forming a distinctive zero-dimensional-in-one-dimensional structure [Figure 7E]. The highly reversible structural evolution of the CoP@C peapod nanostructure during charge/discharge cycles results in long cycling life and excellent rate performance. As shown in Figure 7F, the resultant nanopeapod-like CoP@C composite exhibited excellent lithium storage capabilities, demonstrating a specific capacity of 500 mAh g^-1 after 800 cycles at a high current density of 1.0 A g^-1, highlighting remarkable long-term stability. Furthermore, electrochemical impedance spectroscopy demonstrated that there is almost no deterioration of charge-transfer and Li⁺ diffusion kinetics even after long-term cycling.

Figure 7. (A) Schematic illustration of RP-PC with stable P-C bonds prepared by the vaporization-condensation process and corresponding transmission paths of electron and Li⁺; (B) The cycling performance of the RP-PC and RP-MWCNT electrodes for 1,000 cycles at a current density of 1,000 mA g^-1. Reproduced with permission^[48]. Copyright 2021, American Chemical Society; (C) Schematic illustration of the charge transfer process for the multilayered BP (left) and BP@CNTs hybrid (right); (D) The rate capability of the BP@CNTs, BP, and CNT electrodes at different current densities ranging from 100 mA g^-1 to 2,500 mA g^-1. Reproduced with permission^[119]. Copyright 2020, John Wiley and Sons; (E) The TEM image of the peapod-like CoP@C; (F) The cycling performance of the peapod-like CoP@C at 1.0 A g^-1 for 800 cycles. Reproduced with permission^[120]. Copyright 2018, John Wiley and Sons. RP-PC: Red phosphorus-porous carbons; RP-MWCNT: RP-multi-walled CNTs; CNTs: carbon nanotubes; BP: black phosphorus; TEM: transmission electron microscopy.

These studies confirm the effectiveness of C-P composite strategies in simultaneously addressing the inherent low electrical conductivity and severe volume expansion of P-based anodes. Key design principles include maximizing interfacial contact (e.g., via P-C bonding), tailoring pore structure for P confinement and strain buffering, and constructing three-dimensional conductive networks to facilitate electron/ion transport. Furthermore, future efforts should focus on enhancing P loading capacity, exploring sustainable phosphorus sources, and integrating carbon compositing with complementary strategies to further promote the electrochemical performance of P-based anodes under fast-charging conditions.

Surface modification

Surface modification is mainly employed to suppress surface passivation and improve electrical conductivity and interfacial stability against the detrimental effects of volume variation and polyphosphide dissolution. In recent years, diverse surface modification strategies, such as coating with conductive polymers, surface functionalization, core-shell structure construction, etc., have been developed. These strategies not only effectively mitigate electrode pulverization caused by volume changes but also enhance the interfacial stability and inhibit the shuttle effect of polyphosphide intermediates, thereby substantially improving the electrochemical performance of P-based anodes^{[56,57,121,122]}.

Elastomeric conductive polymer coatings can effectively accommodate the volumetric expansion of active materials, enhance electrical conductivity, and prevent direct contact with electrolyte. This synergistic mechanism preserves the structural integrity of both the active material and the SEI layer, thereby markedly improving the cycling stability and rate performance of LIBs^[121,122]. As a representative study, Sun et al.^[121] fabricated a BP/CNT@polypyrrole (PPy) composite with a robust core-shell structure through a two-step synthesis strategy involving high-energy ball milling and subsequent liquid-phase [Figure 8A]. The core component consists of RP-derived BP, while the incorporated CNT network effectively mitigates volume expansion of BP during cycling. The conformal PPy shell exhibits multifunctional roles: its inherent conductivity accelerates charge transfer kinetics; the flexible polymeric matrix accommodates volume changes while physically isolating the active materials from direct electrolyte contact; and, importantly, the nitrogen functionalities in PPy strongly chemisorb polyphosphide intermediates, thereby effectively suppressing their dissolution and shuttle effect. As shown in Figure 8B, electrochemical testing demonstrated that the BP/CNT@PPy anode maintains a specific capacity of 1,376.3 mAh g^-1 after 200 cycles at 260 mA g^-1, representing a 2.6-fold enhancement compared to the BP/CNT control (530.1 mAh g^-1). Similarly, Wang et al.^[56] synthesized few-layer BP nanosheets through liquid-phase exfoliation, subsequently fabricating BP-graphene oxide-polyaniline (BP-GO-PANI) ternary hybrid anodes via a straightforward two-step chemical process. As illustrated in Figure 8C, the formation of P-C and P-O-C bonds ensures intimate and robust interfacial contact between BP and carbon materials, thereby enhancing the structural integrity of the hybrid anode. The PANI coating effectively maintains unobstructed Li⁺ diffusion pathways, elevating the initial CE from 18% to 40%. Furthermore, the synergistic interaction among few-layer BP, GO, and PANI suppresses volume expansion during charge/discharge cycles from 381.1% to 13.4%, while alleviating active-material cracking and even partial dissolution, thereby markedly improving cycling stability compared to bulk BP. As demonstrated in Figure 8D, the hybrid anode delivered a high reversible specific capacity of 652 mAh g^-1 at 0.1 A g^-1 after 50 cycles, representing a 6.2-fold enhancement over the pristine BP electrode. Furthermore, the hybrid system exhibited excellent long-term cycling performance, retaining 958 mAh g^-1 at 5 A g^-1 after 2,000 cycles [Figure 8E].

Figure 8. (A) Schematic illustration of the synthesis process of the BP/CNT@PPy composite; (B) Comparison of the cycle performance between BP/CNT@PPy and BP/CNT composites at a current density of 260 mA g^-1 within 0.01-3.0 V. Reproduced with permission^[121]. Copyright 2021, American Chemical Society; (C) Sketch map of the whole compounding process of BP-GO-PANI via a solvothermal reaction between BP and GO and then coated in-situ by PANI; Cycling performance of the BP-GO-PANI, BP, and BP-GO electrodes at (D) 0.1 A g^-1 and (E) 5 A g^-1. Reproduced with permission^[56]. Copyright 2022, Elsevier; (F) Schematic illustration of self-healing and delithiation accelerant effects of the LM-BP/G anode; (G) The cycling performance of LM-BP/G and BP/G anodes at 5 A g^-1; (H) The rate performance of LM-BP/G and BP/G anodes at varied rates from 0.5 to 15 A g^-1. Reproduced with permission^[123]. Copyright 2025, Elsevier. BP: Black phosphorus; CNTs: carbon nanotubes; PPy: polypyrrole; BP-GO-PANI: BP-graphene oxide-polyaniline; LM-BP/G: liquid metal-BP-graphite; RP: red phosphorus.

Although the polymer coating strategy enhances the structural stability of P-based anodes by buffering mechanical stress, it may induce significant polarization under high-rate cycling. Additionally, conventional solid-state polymer coatings lack intrinsic self-healing capabilities. To address these limitations, Zhang et al.^[123] fabricated a liquid metal-BP-graphite (LM-BP/G) composite anode by homogeneously dispersing liquid metal, BP, and graphite through ultrasonic treatment, followed by ball-milling. The incorporation of LM, with high conductivity and fluidity, serves as a dynamic self-healing agent that effectively preserves electrochemical connectivity between pulverized P and Li_xP particles during prolonged lithiation/delithiation cycles. Furthermore, LM synergistically facilitates Li⁺ transport and enhances the delithiation kinetics of Li₃P [Figure 8F]. In addition, the strong affinity of LM toward soluble polyphosphide intermediates suppresses their dissolution and shuttle effect, thereby mitigating active-material loss and further stabilizing the electrode-electrolyte interface. Consequently, the LM-BP/G anode demonstrated superior rate capability and cyclability across a broad temperature range (20-50 °C). Specifically, it achieved a specific capacity of 1,123 mAh g^-1 at 5 A g^-1 (~4 C, 15 min charging) with 80% capacity retention after 200 cycles [Figure 8G]. Even under ultrahigh current density (15 A g^-1, ~7 min charging), the anode maintained 1,194.6 mAh g^-1 with comparable cycling performance [Figure 8H]. What is more, the practicability of the LM-BP/G composite was further confirmed in a full-cell configuration with an LiNi_0.5Co_0.2Mn_0.3O₂ (NCM523) cathode, where the NCM523||LM-BP/G cell delivered stable capacities across a wide range of current densities. A specific capacity of 106.0 mAh g^-1 was achieved even at the high rate of 400 mA g^-1, revealing that the LM-BP/G composite exhibits excellent structural stability and electrochemical reversibility under practical full-cell conditions.

Although the above studies demonstrate that surface modification can effectively enhance the conductivity, buffer the volume variation, and suppress polyphosphide dissolution in P-based anodes, their practical application still faces notable challenges. In particular, many reported strategies rely on complex synthesis and costly functional components, which hinder scalability and cost-effectiveness. Future research should therefore prioritize the development of low-cost, multifunctional, and self-healing surface modifiers, together with simplified and scalable fabrication processes, to accelerate the practical deployment of P-based anodes in fast-charging LIBs.

Nanostructuring

Nanostructuring is a versatile strategy to simultaneously alleviate volume variation and enhance Li⁺ transport kinetics in P-based anodes. Downsizing to the nanoscale shortens Li⁺ diffusion pathways, increases electrode-electrolyte contact area, and creates sufficient void space to buffer volumetric expansion, thereby suppressing mechanical degradation during repeated lithiation/delithiation cycles. These structural and kinetic advantages represent the dominant benefits of nanostructuring in most P-based systems. Notably, when the particle size approaches the quantum confinement, additional modulation of the electronic band structure may occur, further facilitating charge transfer and partially overcoming the intrinsically low conductivity^[55,58,59]. Thus, nanostructuring provides both structural/kinetic improvements and, in certain cases, electronic structure tuning, establishing it as a fundamental strategy for enabling fast-charging P-based anodes.

Meng et al.^[124] initially fabricated Ti₃C₂ nanosheets (TNS) through a two-step approach using Ti₃AlC₂ as the precursor. Subsequently, BP quantum dots (BPQDs) anchored on TNS (BPQD/TNS) were synthesized via mechanical grinding of bulk BP, followed by ultrasonic processing in 1-methyl-2-pyrrolidone (NMP). The assembly of BPQDs onto TNS was then achieved through chemical bonding interactions under prolonged ultrasonication and stirring in NMP. The resultant BPQD/TNS composite anode features quantum-confined BPQDs with dimensions below 10 nm. This nanostructure provides abundant lithiation active sites, shortens Li⁺ ion diffusion pathways, and effectively mitigates electrode volume expansion during cycling. Furthermore, the integration of BPQDs and TNS via P-O-Ti covalent bonds not only enhances the interfacial charge transfer rate and stability but also further polarizes atomic charges, strengthening the pseudocapacitive storage effect. This thus improves the battery capacity, long-term cycling performance, and rate capability. As shown in Figure 9A, the BPQD/TNS anode delivered reversible capacities of 1,124 mAh g^-1 at 50 mA g^-1 and 828 mAh g^-1 at 100 mA g^-1. Most notably, it maintained 520 mAh g^-1 after 2,400 cycles at 1,000 mA g^-1, corresponding to a capacity retention exceeding 100% [Figure 9B]. Similarly, Zheng et al.^[125] established a scalable pulsed laser-assisted strategy for high-yield synthesis of two-dimensional few-layered BP nanosheets (L-BP-S) in three low-boiling-point solvents of hexane, 2-propanol, and methanol (L-BP-S_hexane, L-BP-S_2-propanol, and L-BP-S_methanol) [Figure 9C]. The systematic investigation of their electrochemical performance revealed that precise morphological control of BP nanosheets directly determines electrochemical behaviors. Specifically, reduced layer thickness enhances charge transfer kinetics and facilitates Li⁺ ion diffusion, whereas increased aspect ratio improves charge transport efficiency and introduces extended Li⁺ diffusion pathways. Notably, the optimized L-BP-S_hexane anode demonstrated notable cycling durability, maintaining a specific capacity of 404.4 mAh g^-1 after 200 cycles at a current density of 1.0 A g^-1 [Figure 9D], indicating its superior lithium storage capability. Furthermore, Liu et al.^[126] constructed a series of Co/Fe-based Prussian blue analogue@BP quantum dots (PBA@xBQ, where x represents the content of BQ) composites by encapsulating BQ within the size-matched nanopores of cobalt/iron-based PBA MOF via coprecipitation [Figure 9E]. The atomic-level contact between BQ and Co/Fe catalytic sites effectively shortens Li⁺ diffusion pathways and significantly accelerates Li⁺ transport kinetics. Meanwhile, the porous structure of Co/Fe-based PBA enables enhanced Li⁺ storage capacity through synergistic interaction with BP. Surface functionalization of MOF with BQ via a grafting technique effectively suppresses parasitic reactions at the electrode-electrolyte interface. The synergistic interplay of these components substantially enhances lithium storage capabilities. As shown in Figure 9F, the Fe-Co PBA@10BQ composites demonstrated notable electrochemical performance. The fabricated Li||Fe-Co PBA@10BQ cell exhibited high cycling stability, maintaining 70% capacity retention after 500 cycles at a current density of 100 mA g^-1.

Figure 9. (A) Rate performance of the BPQD/TNS composite electrode at varied from 50 mA g^-1 to 2,000 mA g^-1 followed with a cycling performance at 100 mA g^-1; (B) The cycling performance of the BPQD/TNS composite electrode at a high current rate of 1,000 mA g^-1 between 0.005-3.0 V. Reproduced with permission^[124]. Copyright 2018, John Wiley and Sons; (C) Schematic illustration of the preparation procedure of L-BP-S by a liquid-phase laser-assisted exfoliation method with 8 ns pulse duration and a repetition rate of 10 Hz; (D) Cycling performance of the L-BP-S_hexane, L-BP-S_2-propanol, and L-BP-S_methanol electrodes at 1.0 A g^-1 between 0.001-3.0 V. Reproduced with permission^[125]. Copyright 2020, John Wiley and Sons; (E) Schematic illustrations of the preparation process of Co/Fe-based PBA@xBQ (x = 8, 10 and 12) through the coprecipitation method; (F) The cycling performance of the Fe-Co PBA@10BQ electrode at 0.5C, maintaining 70% capacity retention after 500 cycles. Reproduced with permission^[126]. Copyright 2023, Elsevier. BP: Black phosphorus; BPQDs: BP quantum dots; TNS: Ti₃C₂ nanosheets; PBA@xBQ: prussian blue analogue@BP quantum dots; ns: nanosecond.

Recently, Cui et al.^[127] developed a universal “assembly-phosphorization” strategy to encapsulate monodispersed TMPs (MP_x: Ni₂P, FeP, CoP, and CuP₂) within flexible, N-doped carbon multi-chambers with pre-reserved voids (MP_x@NC). This architecture simultaneously enhances electronic conductivity, shortens Li⁺ diffusion pathways, and buffers conversion-induced volume change. As a representative, the Ni₂P@NC anode delivered 613 mAh g^-1 at 0.2 A g^-1 after 200 cycles and retained 475 mAh g^-1 even at a high rate of 2 A g^-1 after 800 cycles. Furthermore, a full-cell based on the Ni₂P@NC anode and the commercial LiFePO₄ cathode achieved a high capacity of 150.1 mAh g^-1 with 87.4% capacity retention after 100 cycles at 0.1 A g^-1. In-situ XRD confirmed a reversible conversion reaction, while in-situ/ex-situ TEM revealed effective volume accommodation within the carbon chambers and preserved structural integrity without excessive SEI growth. Complementary DFT calculations further indicated strong MP_x-C coupling, improved electronic density of states, and reduced Li⁺ diffusion barriers (0.16 eV vs. 0.44 eV for bare Ni₂P), which collectively rationalize the superior rate capability and long-term cycling stability. Overall, this work highlights a synergistic C-P composite, nanostructuring, and interfacial engineering strategy that integrates mechanistic reversibility with device-level feasibility, providing valuable guidance for the multidimensional design of high-performance P-based anodes.

Electrolyte optimization

Electrolyte optimization is primarily focused on addressing the dissolution of polyphosphide intermediates and stabilizing the SEI layer, which are critical for mitigating capacity degradation caused by active material loss and interfacial instability. While P-C composites mitigate volumetric variation, the inherent instability of SEI (exacerbated by volume changes) and the dissolution of polyphosphide intermediates in conventional electrolytes remain unresolved. These issues trigger electrolyte consumption, capacity degradation, and decreased CE with excessive SEI accumulation, further elevating interfacial impedance and impairing fast-charging performance^[9,60]. Thus, electrolyte optimization strategies target these specific challenges to enhance stability. Since P and S are both molecular crystals with similar lithium storage mechanisms, the cleavage of the P-P bonds generates low-molecular-weight polyphosphate species. These species interact with solvent molecules, inducing solvation and triggering polysulfide shuttling. This phenomenon ultimately results in active phosphorus material loss, electrode corrosion, and capacity degradation^[47]. To address these issues, the optimization of electrolyte primarily focuses on constructing a uniform and mechanically robust SEI while effectively mitigating the dissolution of reactive intermediates and suppressing detrimental shuttle phenomena^[61].

Lei et al.^[60] developed a straightforward approach for synthesizing highly reversible P-based hybridized anode materials (P-SP) through high-energy ball-milling of RP and Super P carbon black (SP). This method yields nanoscale P-SP particles, which effectively shorten the electron/ion diffusion pathways and alleviate volume expansion during repeated lithiation/delithiation cycles. To further improve interfacial stability, 10 vol% fluoroethylene carbonate (FEC) was introduced as a film-forming additive to facilitate the formation of a LiF-rich SEI layer. As shown in Figure 10A-C, the resulting SEI exhibits dual advantages: (1) a reduced thickness of 3-5 nm compared to 30-50 nm without FEC; (2) significantly lower SEI resistance (51.7 Ω after the first cycle for FEC-containing cells versus 143.5 Ω for FEC-free cells). Undoubtedly, this ultra-thin, LiF-rich, and stable SEI layer should effectively suppress the polyphosphide shuttle, thereby minimizing active-material loss and ensuring interfacial stability during long-term cycling. With these modifications, the optimized P-SP anode demonstrated notable rate and cycling performance, maintaining a reversible capacity of 1,087.4 mAh g^-1 at 0.5C over 300 cycles [Figure 10D]. Furthermore, Han et al.^[128] developed a ternary lithium bis(fluorosulfonyl)imide in triethyl phosphate/FEC (LiFSI-TEP/FEC) electrolyte system by replacing conventional LiPF₆ with thermally stable LiFSI salt and optimizing the ratio of the flame retardant TEP to the film-forming agent FEC [Figure 10E]. This system substantially enhanced the fast-charging performance of the RP anode. The core mechanism involves Li⁺ preferentially forming a unique low-energy solvation structure {[Li-FEC-3TEP]⁺} with TEP/FEC. This facilitates the preferential reduction of both FEC solvent and FSI^- anions, forming a robust, LiF-rich SEI layer (~10 nm) on the RP surface that is free from corrosive Li_xPF_y compounds. This SEI effectively buffers the significant volume expansion of RP, suppresses electrode/electrolyte side reactions, stabilizes the RP/electrolyte interface for reduced active P loss, and promotes Li⁺ interfacial diffusion during cycling. Consequently, the RP anode delivered a capacity retention of 482.8 mAh g^-1 after 100 cycles at a high rate of 3 A g^-1, approximately 7.3 times higher than that achieved with conventional LiPF₆-based electrolytes, while simultaneously improving battery safety [Figure 10F].

Figure 10. HRTEM images of P-SP hybrids without additive (A) and with 10 vol% FEC (B) after 100 cycles; (C) The corresponding interfacial resistance of P-SP hybrids without additive and with 10 vol% FEC after 100 cycles. Inset: equivalent circuit model; (D) Comparison of cycling performance of the P-SP hybrid anode with 10 vol% FEC to the P-SP hybrid anode and the P-SP mixture anode without additive at 0.5C. Reproduced with permission^[60]. Copyright 2019, American Chemical Society; (E) Schematic illustration of the functions for each component and the synergy in the designed ternary LiFSI-TEP/FEC electrolyte for RP anode; (F) Cycling performance of RP/CNT electrodes at 3 A g^-1 in LiPF₆-based and LiFSI-TEP (7):FEC(3) electrolytes, respectively. Reproduced with permission^[128]. Copyright 2020, Elsevier. HRTEM: High-resolution TEM; TEM: transmission electron microscopy; FEC: fluoroethylene carbonate; P: phosphorus; LiFSI-TEP/FEC: lithium bis(fluorosulfonyl)imide in triethyl phosphate/FEC; RP: red phosphorus; CNTs: carbon nanotubes; P-SP: red phosphorus-super P carbon black hybrid anode.

Overall, electrolyte optimization has proven to be an effective strategy for stabilizing the SEI and suppressing polyphosphide dissolution, thereby mitigating active-material loss, reducing interfacial impedance, and enabling improved rate capability and cycling stability of P-based anodes. Nevertheless, the persistent instability of the SEI-exacerbated by repeated volume changes and uncontrolled interfacial reactions-remains a critical bottleneck. Future efforts should focus on developing multifunctional electrolyte systems that simultaneously regulate solvation structures, construct robust and self-healing SEI layers, and suppress the shuttle effect, which are essential for advancing P-based anodes toward practical fast-charging applications. Moreover, to enable a direct comparison across material classes and architectures, Table 3 summarizes representative electrochemical performance for various P-based anodes discussed above.

In-situ multiscale characterization of Li⁺ transport and interfacial kinetics

The comprehensive elucidation of Li⁺ diffusion pathways and interfacial reaction mechanisms in P-based anodes under fast-charging conditions constitutes a pivotal challenge for optimizing their electrochemical performance. Conventional characterization methodologies (e.g., ex-situ XRD, SEM) encounter inherent limitations in dynamically tracking intricate phase transition behaviors and interfacial evolution during high-rate cycling. Recent breakthroughs in in-situ characterization techniques have enabled significant progress^[9,32]. Atomic-scale visualization of structural deformations and dynamic SEI reconstruction during lithiation/delithiation can now be achieved through in-situ TEM coupled with electron energy loss spectroscopy. The spatially and temporally resolved analytical capabilities of synchrotron-based X-ray absorption spectroscopy (XAS) combined with in-situ Raman spectroscopy prove particularly effective in elucidating transient changes in local bonding structures and intermediate species within P-based materials. Furthermore, in-situ EC-AFM provides quantitative assessments of mechanical-electrochemical coupling behaviors, especially regarding the correlation between interfacial mechanics and volumetric expansion. And in-situ cryogenic electron microscopy (cryo-EM) can be applied to directly visualize SEI formation and evolution in working cells, revealing unprecedented nanoscale details of interfacial chemistry and decomposition pathways. However, fundamental gaps remain in correlating structural models with electrochemical behavior. For example, the local structure of amorphous RP remains debated (polymeric nanotube vs. layered model), and its correlation with Li⁺ transport has not been conclusively established. Similarly, the phase evolution of crystalline RP and TMPs during high-rate cycling is poorly understood. Addressing these controversies will require integrated in-situ multimodal probes coupled with atomistic simulations to provide unified mechanistic insights. Future research should prioritize the advancement of high-resolution, cross-scale, multimodal characterization platforms that integrate cryo-EM, XAS, EC-AFM, neutron imaging, and ab initio molecular dynamics simulations, etc., to establish quantitative structure-property correlations across multiple length scales, thereby offering theoretical foundations for the rational design of kinetically stable electrode interfaces.

Precise design and dynamic regulation of SEI

The unstable SEI layer represents one of the primary bottlenecks limiting the fast-charging performance of P-based anodes. Current modification strategies primarily encompass two approaches: electrolyte optimization and surface engineering. The introduction of film-forming additives (e.g., FEC) facilitates the formation of inorganic-organic hybrid SEI layers (thickness < 5 nm) enriched with LiF, which notably reduce interfacial resistance. Pre-constructing artificial SEI layers through advanced deposition techniques (e.g., atomic layer deposition, CVD) on P surfaces lowers the diffusion energy barrier of Li⁺ ions, while simultaneously mitigating polyphosphide dissolution and enhancing mechanical resilience. Nevertheless, significant knowledge gaps remain. The chemical heterogeneity of SEI layers, their dynamic reconstruction under fast-charging, and the causal relationship between polyphosphide dissolution, SEI thickening, and electrolyte decomposition are not fully understood. To address these issues, future research should integrate advanced characterization methodologies such as cryogenic electron microscopy and time-of-flight secondary ion mass spectrometry to enable high-resolution mapping of SEI composition and spatial architecture. Concurrently, machine learning-driven predictions could streamline the development of optimal electrolyte formulations and interfacial modification protocols.

Constructing highly conductive channels for fast Li⁺ transport

Establishing continuous and stable rapid Li⁺ transport channels within solid-state electrode systems is critical for achieving high-rate charging/discharging performance, yet their practical implementation still faces multiple challenges. RP suffers from sluggish reaction kinetics due to its extremely low electronic conductivity and relatively high solid-state diffusion barriers. In BP and phosphorene, the strongly anisotropic lattice leads to directional dependence of Li⁺ transport, while surface passivation further compromises interfacial charge transfer. For TMPs, although the presence of metallic species provides enhanced conductivity, repeated cycling still induces severe volume expansion, structural pulverization, and redistribution of metallic phases, which together disrupt long-term ion/electron conduction pathways. To address these issues, researchers primarily employ three optimization strategies: (1) construction of carbon-based composite architectures to establish conductive networks for minimizing interfacial resistance; (2) functional interface engineering to mitigate parasitic reactions while enhancing structural integrity; and (3) nanoscale structural engineering to create three-dimensional interconnected ion transport pathways. Recently, the integrated application of DFT-based multiscale simulations, artificial intelligence (AI)-driven materials-genome approaches, and high-throughput experimentation has established a powerful paradigm for the rational design of P-based anode materials, accelerating the optimization of ion-transport mechanisms. By curating standardized, multi-source descriptors (composition, microstructure/morphology, interfacial chemistry, and operating conditions), physics-informed predictive models can rapidly down-select candidates to meet target constraints or, conversely, define multi-objective design spaces to identify promising formulations and architectures. Coupled with precise microstructural characterization, machine learning enables interpretable surrogate models that link microstructure to Li⁺/electron transport and rate performance. When integrated in a closed loop with high-throughput synthesis and parallel electrochemical testing, active-learning/Bayesian optimization can autonomously propose the next experiments, quantify uncertainty, and accelerate inverse design of P-based anodes and compatible electrolytes for practical fast-charging systems.

Full-cell integration and collaborative enhancement of safety and scalability

To translate the promising electrochemical performance of P-based anodes into practical fast-charging LIBs, it is imperative to address the compatibility of P-based anodes with commercially viable cathodes and electrolytes. Full-cell assembly poses critical challenges in balancing the capacity ratio, operating voltage window, and dynamic Li⁺ diffusion kinetics across electrodes. For instance, the large specific capacity and moderate potential of P-based anodes must be carefully matched with cathodes such as Ni-rich layered oxides, LiFePO₄, or high-voltage spinel oxides to ensure optimal energy density, voltage stability, and safety. Mismatches in electrode kinetics and Li⁺ transport rates can lead to over-lithiation or Li plating, especially under high-rate cycling. Meanwhile, the selection of compatible electrolytes is critical. Fluorinated weakly solvating electrolytes, localized high-concentration formulations, ionic liquids, and emerging solid-state electrolytes have shown promise in forming robust SEI layers, suppressing polyphosphide dissolution, and maintaining interfacial stability. Recent full-cell demonstrations further highlight both the feasibility and the limitations of P-based anodes in practical fast-charging LIBs. For example, Sun et al.^[59] reported nanoscale RP confined in conductive carbon frameworks, which in full-cells delivered much better fast-charge capability than graphite or Li₄Ti₅O₁₂ at industrially relevant areal capacities, though with notable first-cycle irreversible loss due to SEI formation. Jin et al.^[13] designed a BP-graphite/polyaniline composite that, when paired with LiCoO₂ cathodes, enabled ~80% state-of-charge recovery within ~9 min and maintained 90% capacity retention over 2000 cycles, far outperforming graphite under identical conditions. TMPs such as AlP have also shown promise: AlP/CNT||LiFePO₄ full-cells retained ~80% capacity after 85 cycles, though limited by moderate initial CE^[129]. More recently, Xie et al.^[130] demonstrated that fluorinated weakly solvating electrolytes combined with RP@carbon anodes achieved ~2,600 mAh g^-1 in half-cells and 87.6% retention in LiFePO₄ full-cells after 300 cycles at 4C, simultaneously improving safety and fast-charging tolerance. Collectively, these studies demonstrate that P-based anodes can surpass conventional graphite in terms of theoretical capacity, fast-charge capability, and long-term cycling stability. Nevertheless, unresolved issues, including large initial irreversible capacity, SEI instability under repeated expansion, and the need for scalable, low-cost synthesis routes, still hinder their competitiveness at the full-cell level. In terms of cost, the relatively high price of high-purity phosphorus precursors and the additional expenses associated with structural engineering (e.g., nanostructuring, carbon coating, and heteroatom doping) raise concerns regarding economic feasibility. From the perspective of scalability, most of the current fabrication strategies, such as high-energy ball-milling, vapor deposition, or solution-based processes, are difficult to translate into large-scale, continuous production, necessitating the development of more industrially compatible routes. Furthermore, the practical implementation of P-based anodes remains challenged by critical safety and environmental concerns: (1) Synthetic safety: the generation of WP (highly toxic and pyrophoric) must be rigorously suppressed during electrode fabrication, which requires the establishment of environmentally benign synthetic pathways; (2) Thermal stability: localized heat accumulation during high-rate cycling can trigger SEI decomposition and potential thermal runaway, highlighting the need for flame-retardant electrolytes and thermoresponsive protective coatings; and (3) Environmental impact: the energy-intensive synthesis and potential release of phosphorus-containing byproducts may raise sustainability concerns, underlining the importance of green chemistry approaches and recycling strategies. In addition, the recyclability of phosphorus-containing electrodes remains underexplored, and dedicated protocols for material recovery and reuse will be essential to minimize waste and close the material loop. From a supply-chain perspective, the availability and cost of high-purity RP and BP precursors pose further bottlenecks, given their limited global production capacity and regional concentration. This chain reaction could be mitigated through strategies such as developing flame-retardant electrolytes, fabricating thermoresponsive polymeric coatings, and establishing electro-thermal-mechanical coupled simulation platforms to predict critical failure thresholds under extreme conditions while optimizing Battery Management Systems. Additionally, cost-effective large-scale manufacturing technologies for P-based materials must be systematically investigated to advance industrial scalability. From an environmental perspective, the design of green and energy-efficient synthesis pathways, the minimization of toxic intermediates such as WP, and the establishment of recycling protocols for phosphorus-containing electrodes are also essential. Such efforts would not only reduce the environmental footprint of large-scale production but also enhance sustainability, thereby paving the way for the safe, economical, and environmentally responsible commercialization of P-based anodes in next-generation LIBs.

In conclusion, owing to their notable theoretical capacity and tunable electronic architecture, P-based anodes demonstrate high potential for fast-charging LIBs. Future research should synergistically integrate multiscale characterization, computational materials science, and AI to elucidate Li⁺ transport kinetics and interfacial degradation mechanisms at the atomic scale. Such multidisciplinary approaches will facilitate the rational design of integrated “structure-function” electrodes with simultaneously improved dynamic performance and long-term cyclability. Addressing these interdisciplinary challenges-spanning atomic-scale characterization, interface engineering, ion transport optimization, and scalable manufacturing-will be crucial for translating the high theoretical promise of P-based materials into practical, high-performance fast-charging LIBs. Concurrent systematic investigations into safety evaluation protocols and engineering-oriented application frameworks will expedite their transition from laboratory prototypes to commercial viability, thereby establishing new paradigms for developing next-generation high-energy-density batteries with ultrafast charging capabilities.