Rapid Synthesis of Honeycomb-Structured FeP₂@NHC for High-Rate and Durable Lithium Storage

Abstract

The concurrent preservation of structural integrity and improvement of electrical conductivity in FeP₂ anodes presents a persistent challenge. Herein, FeP₂ nanoparticles embedded within a 3D N-doped honeycomb-like carbon framework composite (FeP₂@NHC) are synthesized through a phosphorization process with a honeycomb-like Fe₃C@NHC as a precursor. The in situ incorporation of FeP₂ nanoparticles into the 3D carbon matrix effectively restrains the aggregation, pulverization, and stripping of material during cycling, and significantly enhances reaction kinetics and structural stability, achieving a superior electrochemical performance. Specifically, FeP₂@NHC electrodes demonstrate remarkable reversible capacity (1433.9 mA h g⁻¹ at 0.1 A g⁻¹), excellent rate-capability (399.9 mA h g⁻¹ at 10 A g⁻¹), and ultra-long cycle life (631.5 mA h g⁻¹ after 1000 cycles at 2 A g⁻¹). Moreover, XRD analysis reveals that iron-rich Fe₃C and Fe₃O₄ precursors can react with NaH₂PO₂ to form FeP₂ and FeP, respectively. This study offers a rational and practical strategy for designing other phosphorus-rich metal phosphide anode materials.

1. Introduction

Over the past three decades, to address the escalating energy demands of electric vehicles and large-scale energy storage systems, lithium-ion batteries (LIBs) have garnered significant attention [1,2,3]. The anode material plays a crucial role in achieving a fast charging capability for LIBs, meanwhile maintaining high energy density and cycling stability [4]. Recent research has focused on conversion reaction-based transition metal binary compounds (M_aX_b) resting with their high theoretical capacity and versatile compositional options. However, previous studies predominantly focused on metal oxides [5,6], sulfides [7,8], and selenides [9,10]. Though these materials exhibit impressive reversible capacities, their main capacity-contributing redox reactions typically occur at relatively high oxidation potentials (above 1.5 V vs. Li⁺/Li), leading to lower voltage outputs and energy densities [11]. In contrast, transition metal phosphides (TMPs), such as CuP₂ [12], Cu₃P [13], CoP [14], CoP₂ [15], Ni₂P [16], FeP [17], and FeP₂ [18], demonstrate higher specific capacities and lower redox potentials during extended charge–discharge cycles, making them more suitable as anode materials.

Notably, iron is the most abundant transition metal element, and phosphorus reserves are plentiful and readily accessible. Consequently, Fe-P based materials are highly promising candidates for sustainable LIB anodes, with their environmental friendliness, abundant natural resources, and low cost [19]. Iron phosphides encompass various compositions, including Fe₄P, Fe₃P, Fe₂P, FeP, FeP₂, and FeP₄. The theoretical specific capacity increases with a higher phosphorus content, albeit at the expense of reduced electrical conductivity [20,21]. Although phosphorus-rich FeP₂ boasts a notably high theoretical capacity (1365 mA h g⁻¹) and a redox potential of approximately 1 V, its practical implementation is hindered by significant challenges, including irreversible volume changes exceeding 300% and sluggish reaction kinetics, leading to rapid capacity fading [22]. To address these issues, researchers commonly employ a strategy combining nanostructuring with highly conductive carbonaceous materials [23,24,25,26,27]. Nanostructuring enhances complete utilization of active materials in electrochemical reactions and shortens the diffusion distance of Li⁺ from the surface to the core, thereby improving reaction efficiency [28]. Carbonaceous materials such as carbon nanotubes [29], carbon nanofibers [19], graphene [30], porous carbon [20], and hollow carbon shells [18] can mitigate excessive volume expansion during lithiation and enhance the conductivity of the anode material, and their porous structures facilitate increased Li⁺ diffusion rates.

The synthesis techniques for TMP anodes primarily encompass high-energy ball milling, temperature-programmed reduction, organometallic synthesis, and solvothermal/hydrothermal methods [31,32,33]. However, TMPs synthesized via ball milling exhibit challenges in morphology control, leading to suboptimal rate performance and long-term stability. Though organometallic synthesis offers precise control over the phase, size, and morphology of TMPs, its multi-step reaction process presents scalability challenges for industrial production. Moreover, nanoparticles produced through solvothermal/hydrothermal routes often suffer from issues of insolubility and aggregation [32]. Notably, the NaH₂PO₂ reduction method enables the rapid chemical conversion of precursors into TMPs within 300–400 °C while maintaining the structural integrity of the original nanostructures. Although this phosphorization approach has been extensively employed for FeP synthesis, its application in fabricating FeP₂ compounds remains unreported in the existing literature [23,34]. Consequently, the rational design of high-performance FeP₂ hybrid materials remains a significant challenge.

Herein, we report for the first time in situ embedding of FeP₂ nanoparticles within a 3D N-doped honeycomb-like carbon matrix (FeP₂@NHC) using sol–gel pyrolysis blowing and the NaH₂PO₂ phosphorization method. This unique structure offers several advantages. (a) The well-dispersed FeP₂ nanoparticles can substantially mitigate the risk of fracture during repeated lithiation/de-lithiation cycles by isotropically distributing stress [35]. (b) The 3D porous carbon matrix provides ample space to accommodate volume expansion and ensures a stable pathway for rapid ion/electron transport. (c) The in situ incorporation of FeP₂ nanoparticles within the porous carbon framework prevents the agglomeration and pulverization of active particles, achieving efficient lithium storage. As expected, FeP₂@NHC demonstrates outstanding cycling stability and excellent rate capability. Furthermore, the formation mechanisms of FeP and FeP₂ are revealed by XRD and XPS analysis.

2. Materials and Methods

2.1. Synthesis of FeP₂@NHC Nanocomposite

First, 0.5 g Fe (NO₃)₃·9H₂O (Sinopharm Chemical Reagent Co., Ltd., Shanghai, China) and 0.5 g polyvinylpyrrolidone (PVP K130, Aladdin, Shanghai, China) were mixed with 50 mL of deionized water to form a homogeneous solution [20]. Then, continuous heating and stirring were carried out until a gel state was achieved. Subsequently, the gel was subjected to heat treatment in Ar at 700 °C for a duration of 3 h to yield the Fe₃C@NHC precursor. Finally, the Fe₃C@NHC precursor and Na₂H₂PO₂·H₂O (Macklin, Shanghai, China) were placed in two porcelain boats with a mass ratio of 1:30 and positioned separately at downstream and upstream locations within a tube furnace. This phosphorization process was carried out at 360 °C for 0.5 h with a heating rate of 4 °C min⁻¹ under Ar atmosphere. The cooled product was named FeP₂@NHC.

2.2. Synthesis of FeP-FeP₂@C and Pure FeP

The synthesis of FeP-FeP₂@C was carried out by replacing PVP K130 with glucose, while all other procedures remained identical to those for the synthesis of FeP₂@NHC. The synthesis of pure FeP followed similar procedures to those of FeP-FeP₂@C, with the exception that no glucose was added.

2.3. Materials Characterization

The morphology and architecture of the samples were investigated through scanning electron microscopy (SEM, TESCAN) and transmission electron microscopy (TEM, JEM F200, JEOL, Tokyo, Japan). The crystal structures and phases were collected by X-ray powder diffraction (XRD, Smartlab SE, Rigaku, Tokyo, Japan). The chemical states were performed by thermogravimetric analysis (TGA, STA 300, Netzsch, Selb, Germany) and Raman spectroscopy (DXR3, Thermo Fisher Scientific, Waltham, MA, USA). The electronic states of elements in the surface were analyzed by X-ray photoelectron spectroscopy (XPS, K-Alpha, Thermo Fisher Scientific, Waltham, MA, USA). The N₂ adsorption/desorption isotherm was obtained at 77.3 K with a specific surface area analyzer from a Brunauer–Emmett–Teller test (BET, Quadrasorb SI-3MP, Quantachrome, Boynton, FL, USA), while the Barrett–Joyner–Halenda (BJH) method was employed to determine the distributions of pore sizes.

2.4. Electrochemical Measurements

The working electrodes were made by coating a slurry consisting of as-prepared materials, polyvinylidene fluoride (PVDF), acetylene black (mass ratio: 8:1:1), and N-methyl-2-pyrrolidone (NMP) on copper foils and then dried at 80 °C overnight. The composite material loading in each electrode was around 1.0 ± 0.2 mg cm⁻². Li-storage measurements were carried out in stainless-steel coin cells (CR2032) assembled in an argon-filled glovebox, with Li foils as counter electrodes, polypropylene membrane (Celgard 2500) as a separator, and 1.0 M LiPF₆ dissolved in DMC, EC, and EMC (1:1:1 vol.%), with the addition of 5 vol.% FEC as an electrolyte. The as-prepared materials were characterized using the galvanostatic charge/discharge (GCD) method within 0.01–3 V (vs Li/Li⁺) on the Land CT2001A. In our work, the total mass of the composite material was utilized in the calculation of specific capacity. The cyclic voltammetry (CV) and electrochemical impedance spectrum (EIS) in the frequency of 100 kHz to 0.01 Hz were measured on an electrochemical workstation (CHI 760E).

3. Results

The synthesis strategy of FeP₂@NHC, as illustrated in Figure 1, involved three sequential steps. Firstly, the aqueous solution of Fe (NO₃)₃·9H₂O and PVP K130 was continuously heated and stirred to form a gel. Secondly, the gel underwent a high-temperature heat treatment process to obtain the honeycomb-like 3D porous Fe₃C@NHC precursor. PVP played an essential role in the formation of this hierarchical structure. This was due to a substantial amount of gas generated during the decomposition of the colloidal mixture, leading to the formation of bubbles. Once the pressure within the bubble cavities reached a critical threshold, the bubbles burst, and their walls interconnected to form a honeycomb-like structure. Meanwhile, Fe₃O₄ generated from the decomposition of Fe (NO₃)₃ was incorporated into the framework. As the temperature increased, the framework underwent gradual carbonization, and Fe₃O₄ was reduced to form Fe₃C. Finally, Na₂H₂PO₂·H₂O was used as the phosphorus source, and Fe₃C@NHC was rapidly reacted with PH₃ under Ar at 360 °C to produce the FeP₂@NHC nanocomposite. FeP₂@NHC retained the morphology and structure of the Fe₃C@NHC precursor, attributing to the resistance of the carbon framework to the phosphorization process, thereby exhibiting a distinctive multi-scale hierarchical structure.

As shown in Figure 2, the microstructural characteristics and morphological evolution of the synthesized materials were systematically investigated through SEM and TEM analyses. Figure S1a presents the SEM image of the rod-shaped Fe₃O₄. Pure FeP was observed as micron-sized agglomerates, as shown in Figure S1b. Figure S2a,b depict the aggregates of Fe₃C@C and FeP-FeP₂@C nano-spherical particles, respectively. As shown in Figures S3 and S4, Fe₃C@C and FeP-FeP₂@C are encapsulated by a carbon layer, obtained through the high-temperature carbonization of glucose. Notably, FeP₂@NHC displayed a unique honeycomb-like 3D porous architecture. In this specific structure, FeP₂ was encapsulated within a thin carbon layer and subsequently embedded in two-dimensional N-doped carbon nanosheets. These nanosheets were interconnected, forming a hierarchical three-dimensional framework. The low-magnification TEM image (Figure 2c) confirmed that FeP₂ particles, ranging from 20 to 200 nm in size, were embedded within the carbon nanosheets, consistent with the SEM results. This advantageous combination of nanosheets and nanoparticles effectively prevented the aggregation and pulverization of active materials during the cycling process, enhanced the transmission efficiency of electrons and ions, and mitigated volume changes [18,20]. The 2.49 Å lattice fringes observed in the high-resolution TEM (HRTEM) image (Figure 2d) corresponded to the (200) plane of FeP₂, indicating excellent crystallinity. Furthermore, the HRTEM image confirmed that FeP₂ was uniformly coated with a thin carbon layer. The selected area electron diffraction (SAED) pattern (Figure 2e) showed (200) and (010) crystal planes along the [001] zone axis, corresponding to the FeP₂ phase of the orthorhombic crystal system. The EDX analysis (Figure 2f) confirmed the presence of elements such as C, N, Fe, and P within the FeP₂@NHC sample. Additionally, the atomic ratio of Fe to P was approximately 1:2, indicating a successful transformation of Fe₃C and PH₃ into FeP₂ following calcination. The TEM mapping images (Figure 2g–k) indicated that Fe and P elements were predominantly localized within the FeP₂ nanoparticle regions, whereas C and N elements exhibited a uniform distribution throughout the composite material, suggesting that the FeP₂ nanoparticles were effectively encapsulated within the N-doped honeycomb-like carbon matrix.

The crystal structures of the materials and formation mechanisms of FeP and FeP₂ were systematically investigated through XRD. As shown in Figure 3a,b, in the presence of a carbon source, Fe (NO₃)₃ was transformed into Fe₃C (ICDD#01-077-0255) at high temperatures. For FeP₂@NHC, multiple distinct peaks were observed, indicating that FeP₂ (ICDD#01-089-2261) nanoparticles were well-crystallized. The reactions for synthesizing FeP₂@NHC can be expressed by Equations (1)–(4). However, when glucose served as the carbon source, Fe₃C was encapsulated within the carbon sheets, leading to the formation of aggregates. This structure restricted the diffusion rate of the PH₃ gas and reduced the contact area with Fe₃C. Consequently, the phosphorization process predominantly proceeded via Equation (5) to form FeP (ICDD#01-078-1443), with only a minor contribution from Equation (4) for the formation of FeP₂. Figure 3c illustrates that, in the absence of any assisting polymer, Fe (NO₃)₃ underwent thermal decomposition to form Fe₃O₄ (ICDD#01-071-6766), which subsequently reacted with PH₃ via Equation (6) to produce pure FeP.

$$\mathrm{F}\mathrm{e}{\left(\mathrm{N}{\mathrm{O}}_3\right)}_3=\mathrm{F}{\mathrm{e}}_3{\mathrm{O}}_4+2\mathrm{N}{\mathrm{O}}_2\left(\mathrm{g}\right)+\mathrm{N}\mathrm{O}\left(\mathrm{g}\right)$$

(1)

$$\mathrm{P}\mathrm{V}\mathrm{P}{\left({\mathrm{C}}_6{\mathrm{H}}_9\mathrm{N}\mathrm{O}\right)}_{\mathrm{n}}\to \mathrm{C}+\mathrm{C}{\mathrm{O}}_{\mathrm{x}}\left(\mathrm{g}\right)+\mathrm{N}{\mathrm{O}}_{\mathrm{y}}\left(\mathrm{g}\right)+{\mathrm{H}}_2\mathrm{O}\left(\mathrm{g}\right)$$

(2)

$${\mathrm{F}\mathrm{e}}_3{\mathrm{O}}_4+5\mathrm{C}={\mathrm{F}\mathrm{e}}_3\mathrm{C}+4\mathrm{C}\mathrm{O}\left(\mathrm{g}\right)$$

(3)

$${\mathrm{F}\mathrm{e}}_3\mathrm{C}+6\mathrm{P}{\mathrm{H}}_3\left(\mathrm{g}\right)+3.5\mathrm{C}=3\mathrm{F}\mathrm{e}{\mathrm{P}}_2+4.5\mathrm{C}{\mathrm{H}}_4\left(\mathrm{g}\right)$$

(4)

$${\mathrm{F}\mathrm{e}}_3\mathrm{C}+{3\mathrm{P}\mathrm{H}}_3\left(\mathrm{g}\right)+1.25\mathrm{C}=3\mathrm{F}\mathrm{e}\mathrm{P}+2.25\mathrm{C}{\mathrm{H}}_4\left(\mathrm{g}\right)$$

(5)

$${\mathrm{F}\mathrm{e}}_3{\mathrm{O}}_4+{3\mathrm{P}\mathrm{H}}_3=3\mathrm{F}\mathrm{e}\mathrm{P}+{4\mathrm{H}}_2\mathrm{O}\left(\mathrm{g}\right)+{\mathrm{H}}_2\left(\mathrm{g}\right)$$

(6)

Figure 3d illustrates the Raman spectrum of the materials. FeP₂@NHC exhibited two prominent peaks at 1340 cm⁻¹ and 1580 cm⁻¹, corresponding to the D-band and G-band of the carbon sheet, respectively [17,36]. To be specific, the G-band indicated the graphitization level of the carbon atoms, and the D-band reflectred the extent of structural defects in the carbon lattice [37,38]. The intensity ratio of the D-band to the G-band for FeP₂@NHC and FeP-FeP₂@C was approximately 0.902 and 0.83, respectively, suggesting that the honeycomb-like carbon framework possessed a higher density of defects, enhanced electronic conductivity, and facilitated efficient electron transport. The surface area and porosity characteristics of FeP₂@NHC can be characterized using adsorption–desorption isotherms and the pore size distribution curve (Figure 3e). The adsorption–desorption isotherm exhibiteda type IV pattern with a pronounced hysteresis loop, characteristic of mesoporous materials [30,39]. Based on the BET method, the specific surface area of FeP₂@NHC was approximately 942.2 m² g⁻¹. In contrast, the specific surface areas of FeP-FeP₂@C and pure FeP were notably lower (Figure S5), measuring 12.23 and 10.54 m² g⁻¹, respectively. The average pore diameter was estimated to be around 5 nm using the Barrett–Joyner–Halenda (BJH) model. This unique honeycomb-like structure with a large surface area and suitable pore size was anticipated to enhance the contact between the electrolyte and the active materials, provide ample space for accommodating significant volume changes, and facilitate the rapid transport of ions and charges during the electrochemical reaction [35,40]. As illustrated in Figure 3f, TGA was employed to quantify the carbon content in FeP₂@NHC. The mass loss below 200 °C can be set with the volatilization of adsorbed water. The substantial mass loss observed from 200 to 800 °C primarily remained with the oxidation of FeP₂ into Fe₂O₃ and P₂O₅, along with the oxidation of the carbon [41]. The contents of FeP₂ and carbon were determined to be 36.4% and 63.6%, respectively. XPS analysis was employed to further characterize the chemical composition and surface state of materials. The XPS survey spectrum in Figure 3g confirmed the presence of C, O, N, Fe, and P elements in FeP₂@NHC, with the oxygen signal attributed to partial surface oxidation. The high-resolution Fe 2p spectrum (Figure 3h) indicated that peaks at 711.8 and 725.7 eV were attributed to Fe-O bond, and peaks at 707.6 and 720.4 eV were assigned to Fe-P 2p_3/2 and Fe-P 2p_1/2, respectively [19,21]. Figure 3i shows the P_2p spectrum, in which three peaks observed at 129.7, 130.5, and 134.3 eV corresponded to P 2p_3/2, P 2p_1/2, and P-O bond, respectively [20,22]. In the high-resolution C 1s spectrum (Figure S6a), the peak at 284.8 eV corresponded to the C-C bond, and peaks at 285.6 and 289 eV were attributed to C-P and C-O bonds, respectively [17,18,29]. Furthermore, the N 1s spectrum (Figure S6b) exhibited three distinct peaks at 400.0, 400.7, and 402 eV, corresponding to pyridinic N, pyrrolic N, and graphitic N, respectively [24]. Abundant nitrogen atoms within the carbon matrix can furnish an abundance of active sites and defects, thus facilitating the diffusion and transport of Li⁺ ions and enhancing reaction kinetics [42,43]. Additionally, given that the electronegativity of nitrogen exceeds that of carbon, this can boost the conductivity of material. All the above results collectively validated the successful synthesis of FeP₂ nanoparticles embedded within a honeycomb-like N-doped carbon matrix. The XPS spectrum of FeP-FeP₂@C and pure FeP displayed comparable fitting peaks. The fitted Fe 2p peak position shifted in FeP-FeP₂@C to a higher binding energy, attributed to lattice differences in the mixed phosphides. Notably, the oxygen peak in pure FeP was markedly stronger, likely due to the absence of carbon protection, leading to a more pronounced surface oxidation.

The lithium-storage performance of materials was evaluated within a voltage range of 0.01–3.0 V (Figure 4). The first five cyclic voltammogram (CV) curves, recorded at a scan rate of 0.2 mV s⁻¹, are presented in Figure 4a. The peak observed at approximately 1.17 V in the initial cathodic scan suggested the formation of a solid electrolyte interlayer (SEI) film, and this phenomenon disappeared in subsequent cycles. The cathodic peak at 0.73 V and the anodic peak at 1.17 V corresponded to the conversion reaction process of FeP₂, respectively, as described by Equations (7) and (8) [20]. In the second and subsequent cycles, the cathodic peak shifted to approximately 0.6 V, while the anodic peak moved to a higher potential, and a broad peak emerged around 1.5 V. These shifts in peak positions can be attributed to the transformation of FeP₂ crystals into an amorphous state and the development of strain during the initial lithiation/de-lithiation processes [18,21]. Figure S8a presents the CV curves for the initial five cycles of FeP-FeP₂@C. For FeP-FeP₂@C, a series of cathodic peaks observed between 0.1 and 1.8 V during the first cycle can be attributed to the formation of an SEI film and the varying voltages associated with the mixed phosphide conversion reactions. Additionally, anodic peaks consistently appearing at 0.16 V across all cycles were indicative of multiple conversion reactions. Figure S8b presents the CV curves of pure FeP, aligning with previous reports [24].

$$\mathrm{F}\mathrm{e}{\mathrm{P}}_2+6{\mathrm{L}\mathrm{i}}^{+}+6{\mathrm{e}}^{\_}\to \mathrm{F}{\mathrm{e}}_{\left(\mathrm{a}\mathrm{m}\mathrm{o}\mathrm{r}\mathrm{p}\mathrm{h}\mathrm{o}\mathrm{u}\mathrm{s}\right)}+2\mathrm{L}{\mathrm{i}}_3\mathrm{P}$$

(7)

$$\mathrm{F}{\mathrm{e}}_{\left(\mathrm{a}\mathrm{m}\mathrm{o}\mathrm{r}\mathrm{p}\mathrm{h}\mathrm{o}\mathrm{u}\mathrm{s}\right)}+2{\mathrm{L}\mathrm{i}}_3\mathrm{P}\to \mathrm{F}\mathrm{e}{\mathrm{P}}_{2\left(\mathrm{a}\mathrm{m}\mathrm{o}\mathrm{r}\mathrm{p}\mathrm{h}\mathrm{o}\mathrm{u}\mathrm{s}\right)}+6{\mathrm{L}\mathrm{i}}^{+}+6{\mathrm{e}}^{\_}$$

(8)

Figure 4b shows the charge–discharge profiles of the FeP₂@NHC electrode at 0.1 A g⁻¹. FeP₂@NHC exhibited initial lithiation/de-lithiation specific capacities of 1870 mA h g⁻¹ and 1183.3 mA h g⁻¹, respectively, with an initial Coulombic efficiency (CE) of 63.3%. The capacity loss during the first cycle primarily lay with the formation of the SEI film and the irreversible trapping of Li⁺ within the porous structure [19]. Furthermore, observed charge–discharge voltage plateaus aligned with the findings from the CV curves. As shown in Figure 4c, the capacity of the FeP₂@NHC electrode initially decreased and subsequently increased at a current density of 0.1 A g⁻¹, with a reversible capacity of 1433.9 mA h g⁻¹ after 200 cycles. The capacity loss during the initial 20 cycles was attributed to an incomplete redox reaction of FeP₂, the continuous formation of the SEI film, and the decomposition of the electrolyte [19]. As the electrode material gradually activated, the capacity began to rise slowly after several dozen cycles. This phenomenon is commonly observed in Fe-based materials [18]. In contrast, the reversible capacities of FeP-FeP₂@C and pure FeP after 200 cycles were 859.3 and 168.1 mA h g⁻¹, respectively. These results confirm the superior intrinsic theoretical specific capacity of phosphorus-rich FeP₂ and the unique hierarchical structure in the FeP₂@NHC composite. The rate performance of FeP-FeP₂@C, and FeP was inferior to those of FeP₂@NHC under all working conditions. For FeP₂@NHC, reversible capacities of 1210.7, 1136.1, 1072.8, 979.9, 934.3, 851.9, 665.2, and 399.9 mA h g⁻¹ were achieved at 0.05, 0.1, 0.2, 0.5, 1, 2, 5, and 10 A g⁻¹, respectively (Figure 4d). Notably, when the current density was reset to 0.1 A g⁻¹, the capacity recovered to a significantly higher value of 1283.8 mA h g⁻¹. To further characterize the stable cycling performance of FeP₂@NHC, as illustrated in Figure 4e, all electrodes underwent 1000 cycles at 2 A g⁻¹. FeP₂@NHC initially showed a decrease in cycling performance, followed by an increase and eventual stabilization, achieving a reversible capacity of 631.5 mA h g⁻¹. Moreover, compared with previously reported Fe-based metal phosphide anodes, FeP₂@NHC exhibited a superior rate capability (

Table 1. Comparison of rate performance of recently reported Fe-based phosphides in LIBs.

Materials	Electric Current Density/A g−1	Capacity/mA h g−1	Ref.
FeP2@NHC	0.1 2 10	1136.1 851.9 399.9	Our work
FeP@C/Ti3C2	0.1 2 5	816 563.5 466.7	[17]
FeP2@C	0.1 3	932 339	[18]
FeP2@CNs	0.1 2 5	1132.2 695.9 580.7	[19]
FeP2 NPs@CK	0.1 2	1023 724	[20]
FeP2/C/CNTs@GA	0.1 2 5	886 691 685	[21]
FeP@NC	0.1 2 10	704.6 434.3 305.9	[24]
FeP/C	0.2 2 5	990 490 314	[25]
FeP2/CNT	0.137 1.37 6.85	590 380 240	[29]
Fe2P/C	0.1 2 5	727 500 365	[37]
FeP@CNBs	0.1 2	608 380	[41]

). The SEM and TEM images (Figure S9) of all electrode materials after 200 cycles at 0.1 A g⁻¹ revealed that the distinctive porous honeycomb structure of FeP₂@NHC remained intact, indicating the robust structural stability of FeP₂@NHC. The superior electrochemical performance can be attributed to the inherent redox reactivity of FeP₂ and the innovative structural design that minimized volume changes while facilitating rapid electron–ion transport.

The charge transfer kinetics of electrodes were examined through EIS. As shown in Figure S10, each Nyquist plot comprised a high-frequency semicircle and a low-frequency inclined straight line, corresponding to the charge transfer process at the electrode–electrolyte interface and the Li⁺ diffusion process, respectively. The charge transfer resistance (R_ct) can be estimated by analyzing the diameter of the semicircle observed in the high-frequency region of the impedance plot. The Warburg resistance (W) can be derived from the slope of the straight line [44]. The R_ct values of FeP₂@NHC before and after cycling were lower than those of FeP-FeP₂@C and FeP, indicating that the honeycomb-like N-doped carbon framework enhanced the charge transfer and preserved the structural stability of active materials. The W value of FeP₂@NHC indicated a higher rate of Li⁺ diffusion in this anode material. To further investigate the reaction kinetics of Li⁺, CV was conducted at various scan rates ranging from 0.2 to 2 mV s⁻¹. As depicted in Figure 5a,d,g, all electrodes exhibited an increase in peak current and a broader potential range as scan rates were increased. The storage mechanism can be examined using the following equations [28,44]:

$$i=a{v}^b$$

(9)

$$logi=blogv+loga$$

(10)

where i represents the peak current, v denotes the scan rate, and the remaining symbols (a, b) represent related parameters. The b-value can be determined by fitting the slope of the line in logarithmic plots of v versus i, reflecting the type of charge storage mechanism. A b value of 0.5 signifies the dominance of only redox reactions, whereas a value of 1 demonstrates solely capacitive effects. The calculated results of all electrodes, as shown in Figure S11, revealed that b values fell within the range of 0.5 to 1, indicating that the combined action of both mechanisms contributed to the electrochemical performance. The b value of redox peak for FeP₂@NHC was lower than that of FeP-FeP₂@C and pure FeP, which further corroborated the ultrafast charge transfer rate in FeP₂@NHC, along with the enhanced diffusion and reaction rates of lithium ions. Consequently, fewer electrochemical reactions took place at the surface.

Moreover, the contribution ratios of diffusion and capacitive control behaviors throughout the process can be quantified by Equations (11) and (12) [45]:

$$i_v=k_{1}v+k_2{v}^{1/2}$$

(11)

$$i_v/v^{1/2}=k_1{v}^{1/2}+k_2$$

(12)

where k₁v is associated with the pseudocapacitive contribution, and k₂v^1/2 is linked to the diffusion contribution. The constants k₁ and k₂ are determined by the linear plots of i_v/v^1/2 and ν^1/2 at a given potential. The pseudocapacitive contribution at 0.8 mV s⁻¹ is shown in Figure 5b,e,h, respectively, with the green region accounting for approximately 56%, 65%, and 68% of the total area. The contribution percentages presented in Figure 5c,f,i reveal a consistent increase in the capacitive contribution ratio as the scan rate increased. The N-doped honeycomb structure enhanced the charge storage and accelerated the reaction kinetics, which aligned with the fitted b value results.

Additionally, GITT was employed to determine the diffusion coefficients of Li⁺ in the electrodes, as depicted in Figure 6 and Figure S12. The diffusion coefficients were determined by the subsequent equation [46]:

$$D={\displaystyle \frac{4}{\pi \tau }}{\left({\displaystyle \frac{n_B{V}_m}{S}}\right)}^2{\left({\displaystyle \frac{∆E_S}{∆E_t}}\right)}^2$$

(13)

where τ represents the time of pulse, n_B is the number of moles (mol) of the electrode, and V_m is the molar volume of the electrode. Moreover, S denotes the total contacting area between the electrode and the electrolyte. ΔE_S refers to variations in the steady-state potential during the relevant steps, while ΔE_t represents the changes in a single-step experiment by eliminating the IR drop. The calculation results for the Li⁺ diffusion coefficients are presented in Figure 6b,c. Notably, compared with the FeP-FeP₂@C and FeP electrodes, the FeP₂@NHC electrode exhibited significantly higher ion diffusion coefficients during both insertion and extraction processes. This finding provides strong evidence explaining the superior rate performance and the excellent cycling stability of FeP₂@NHC.

The exceptional electrochemical performance of the FeP₂@NHC electrode lay within the intrinsically high specific capacity of phosphorus-rich FeP₂ and its distinctive 3D conductive networks, conferring multiple advantages for a superior performance. As illustrated in Figure 7, in situ generated FeP₂ nanoparticles embedded in carbon nanosheets effectively suppressed the grain growth of FeP₂. This nanostructure engineering not only mitigated the pulverization of the FeP₂ nanoparticles but also shortened the electron diffusion path. Furthermore, FeP₂ was encapsulated within a honeycomb-like carbon matrix, which shielded the active material from electrolyte degradation and provided ample space for volume expansion during repeated charge–discharge cycles, thus preserving the structural integrity and electrochemical stability. Furthermore, the 3D interconnected carbon network functioned as a high-speed conduit for both electron and ion transport, while also preventing the aggregation of FeP₂ nanoparticles. Benefiting from the optimized structural design, the FeP₂@NHC electrode exhibited exceptional rate performance and stable long-term cycling stability.

4. Conclusions

In conclusion, we successfully synthesized FeP₂ nanoparticles embedded in an N-doped honeycomb-like carbon skeleton via a combination of sol–gel pyrolysis blowing and sodium hypophosphite phosphorization methods. FeP₂ nanoparticles were embedded into a porous carbon matrix, which effectively prevented the pulverization and aggregation of the nanoparticles, thereby maintaining structural integrity. The 3D honeycomb-like carbon framework provided ample space to accommodate volume expansion during repeated charge–discharge cycles, and offered conductive pathways, along with rapid diffusion channels, for both electrons and ions. Moreover, N doping introduced additional active sites and enhanced electrical conductivity. Benefiting from this innovative structure, the FeP₂@NHC electrode demonstrated exceptional reversible capacity (1433.9 mA h g⁻¹ at 0.1 A g⁻¹), outstanding rate capability (399.9 mA h g⁻¹ at 10 A g⁻¹), and excellent cycling stability (631.5 mA h g⁻¹ at 2 A g⁻¹). Furthermore, XRD analysis confirmed that the iron-rich Fe₃C and Fe₃O₄ precursors reacted with NaH₂PO₂ to form FeP₂ and FeP, respectively. This work presents a straightforward approach for the synthesis of phosphorus-rich phosphide composites, which can provide valuable insights for advancements in the energy storage field.