Low-Temperature Solution-Based Phosphorization Reaction Route to Sn₄P₃/Reduced Graphene Oxide Nanohybrids as Anodes for Sodium Ion Batteries

1 Introduction

Considering the rare lithium resources and expensive cost of lithium element, the present technology for lithium ion battery (LIBs) cannot meet the demands for power sources in future energy storage applications, especially for large-scale energy storage systems.[1, 2] Recently, sodium ion batteries (SIBs) have attracted increased attention as a new rechargeable battery system as an alternative to LIBs because of the abundant sodium resources in the earth and low cost of sodium compounds.[3-6] However, it should be stressed that the lower energy capacity and poor cycling stability of SIBs than that of LIBs are the major obstacles for their potential large scale energy storage applications.

In order to improve the energy density of SIBs, it is of great importance to rationally design electrode materials with high specific capacities and appropriate redox potentials. Up to date, various interlayered transition metal cathode materials for SIBs have been developed successfully, including NaCoO₂,[7] NaFePO₄,[8] NaMnO₂,[9] and Na₃V₂(PO₄)₃[10-12] by replacing Li ions with Na ions. However, high performance SIBs anodes are still challenging. Previous study shows that pure graphite SIBs anode does not demonstrate an excellent performance.[4] Ge and Fouletier[13] reported the electrochemical behavior of graphite SIBs anode only shows a capacity of only about 35 mA h g⁻¹. Kim et al.[14] improved the Na ion storage performance of SIBs anode based on natural graphite by using an ether-based electrolyte system. However, the reversible capacity is still lower, only about 150 mA h g⁻¹. Therefore, in order to obtain SIBs with high specific capacity, new anode materials with high capacity and low redox potential should be introduced.[4]

Na alloy-based anodes have attracted much attention owing to the appropriate redox potentials and higher gravimetric and volumetric specific capacities, such as Sb,[15-19] Ge,[20] Sn,[21-23] and P[24-28] based materials. Sb shows excellent stable capacity retention performance because of its smaller volume expansion and reduced anisotropic mechanical stress.[19] However, it is hindered by its low theoretical specific capacity of 660 mA h g⁻¹ and high price.[4, 15-18] Ge also cannot meet the requirement of SIBs due to its lower theoretical capacity of only 396 mA h g⁻¹.[6, 20] Sn has the high electrical conductivity, low cost, environmental friendliness, and appropriate low charge–discharge potentials versus Na⁺/Na.[21-23] Sn can theoretically deliver a high capacity of 847 mA h g⁻¹, which corresponds to the formation of Na₁₅Sn₄ alloy.[6] Nevertheless, a 525% volume change from Sn to Na₁₅Sn₄ could cause the poor cycling performance.[21] Phosphorus is very promising as an anode for SIBs because of its high theoretical specific capacity (2596 mA h g⁻¹) and appropriate redox potentials (0.4 V vs Na⁺/Na). The large volume expansion (420%) and poor electrical conductivity (10⁻¹⁴ S cm⁻¹) are two critical issues that need to be urgently solved.[24-28] In addition, both the binder and electrolyte additive play an important role in the performance of SIBs anodes.[29-32] Fluoroethylene carbonate (FEC) additive in propylene carbonate (PC) electrolyte can suppress the unfavorable side reactions for SIBs electrode.[29, 30] Recently, it is reported that carboxymethyl cellulose (CMC) and polyacrylic acid (PAA) as the binders for SIBs can form a cross-linked structure to tolerate volume expansion.[31, 32] However, they are not widely used in batteries because of their higher cost. Compared with CMC and PAA binders, polyvinylidene fluoride (PVDF) is the most popular binder in rechargeable batteries because of its low cost, higher stability, and easy processing performance, although it demonstrates a relatively low mechanical strength for electrode material.

Recently, Sn₄P₃, combining with the advantages of Sn (high conductivity) and P (high capacity), has attracted much attention as an anode for SIBs.[33-38] Sn₄P₃ anode materials can display a theoretical volumetric capacity (6650 mA h cm⁻³), a good electrical conductivity (30.7 S cm⁻¹), and a gravimetric capacity of 1132 mA h g⁻¹. Sn_xP₃ is mostly reported to be synthesized via the mechanical alloying process. For example, Sn_4+xP₃@Sn-P nanohybrids synthesized by Li et al. using high-energy mechanical ball milling route can deliver a capacity of 465 mA h g⁻¹ at a current density of 100 mA g⁻¹ after 100 cycles.[34] Qian et al. reported Sn₄P₃/C SIBs anode by mechanical ball milling method, demonstrating a capacity of about 500 mA h g⁻¹ at a current density of 100 mA h g⁻¹ after 150 cycles.[36] However, the particle diameter of the anode material obtained by mechanical ball milling can reach about several micrometers, which is worse for electrode cycling performance, especially for the longer cycling life. Although Sn₄P₃ displays verity of advantages for SIBs anodes, unfortunately the progress for rational design and synthesis of Sn₄P₃-based SIBs anode with an optimized microstructure and electrochemical performance is still less significant. The rational design of controllable microstructures and chemical composition, thereby improving the electrochemical performance of SIBs anode materials with high specific capacity and longer cycling life, is of great challenge and fundamental importance.

Herein, Sn₄P₃/reduced graphene oxide (RGO) hybrids are synthesized for the first time through an in situ low-temperature solution-based phosphorization chemical transformation route from Sn/RGO. The Sn₄P₃ nanoparticles with an average size of 6 nm are homogenously loaded on the RGO nanosheets, interconnecting to form 3D mesoporous architecture structures. RGO nanosheet not only enhances the conductivity of the nanohybrids but also buffers the volume expansion during the Na⁺ insertion–extraction process. The Sn₄P₃/RGO hybrid architecture SIBs anodes exhibit significantly improved electrochemical performance of high reversible capacity, high-rate capability, and excellent cycling performance as anode materials for SIBs, exhibiting a superior long cycling life, delivering a high capacity of 362 mA h g⁻¹ after 1500 cycles at a high current density of 1.0 A g⁻¹. The outstanding cycling performance and rate capability of these porous hierarchical Sn₄P₃/RGO hybrid anodes can be attributed to the advantage of porous structure, and the synergistic effect between Sn₄P₃ nanoparticles and graphene nanosheets. This novel Sn₄P₃/RGO nanohybrid with good electrochemical performance can find potential applications as anode material for SIBs.

2 Results and Discussion

The synthesis process of Sn₄P₃/RGO hybrids is schematically illustrated in Figure 1. First, Sn/RGO nanohybrids are synthesized by reducing SnCl₂ to Sn, using NaBH₄ as a reducing agent. A polyvinylpyrrolidone (PVP) assisted reduced approach can assure uniform distribution of smaller Sn nanoparticles on the RGO matrix.[39, 40] Then Sn/RGO in-situ transforms to Sn₄P₃/RGO via a low-temperature solution-based phosphorization chemical reaction process.[37, 41] The corresponding powder X-ray diffraction (XRD) patterns of the Sn/RGO and Sn₄P₃/RGO hybrids can provide information on phase components of the synthesized products. Figure S1 (Supporting Information) shows the XRD pattern of the synthesized Sn/RGO samples. As expected, the sharp diffraction peaks at 30.6°, 32.0°, 43.8°, 44.9°, 55.3°, 62.5°, 63.7°, and 64.5° can be assigned to the (200), (101), (220), (211), (301), (112), (400), and (321) planes of a tetragonal structured Sn ( space group: I41/amd(141)) (JCPDS No.: 65-2631), respectively.[39, 40] Figure 2 depicts the XRD pattern of the synthesized Sn₄P₃/RGO samples. The sharp diffraction peaks at 28.8°, 30.3°, 31.5°, 44.5°, and 45.7° can be assigned to the (015), (0012), (107), (0114), and (110) planes of a rhombohedral Sn₄P₃ with a space group of R-3m(166) (JCPDS No.: 73-1820), respectively.[41, 42] In the XRD pattern of Sn₄P₃/RGO-1 sample, there are two smaller peaks at about 20.1° and 22.1°, which may be ascribed to P₄ obtained from the solvothermal reaction. In the XRD pattern of Sn₄P₃/RGO sample, the broad RGO (002) peak degree at about 22°–26° increased with increasing the RGO concentration. The RGO peak of Sn₄P₃/RGO-2 sample is very weak, which can be attributed to the lower RGO concentration and the lower graphitization degree, indicating the disordered stacking nature of graphene sheets.[43-45] In the Sn₄P₃/C sample (30 wt% carbon)[36] and Sn₄P₃/C nanosphere sample,[37] the carbon peaks are also indistinct. The Raman spectra in Figure S2 (Supporting Information) further testified the presence of RGO nanosheets in synthesized samples.

The graphitization degree of the RGO in the synthesized samples can be well characterized by Raman spectra. Figure S2 (Supporting Information) shows the Raman spectra and thermogravimetric analysis (TGA) curves of the synthesized Sn₄P₃/RGO samples. As shown in Figure S2a (Supporting Information), two peaks at 1322 and 1587 cm⁻¹ are observed, which are attributed to the D (disordered) band and the G (graphite) band of carbon, respectively. For the graphitization degree of the carbon, it is generally evaluated according to the ratio of the intensity of the D band to G band (I_D/I_G). The ratio of intensity between I_D and I_G band is 1.50, 1.46, and 1.42, indicating a low graphitization degree for the synthesized Sn₄P₃/RGO samples.[46] Figure S2b (Supporting Information) shows the TGA curves of the bare Sn₄P₃ sample and Sn₄P₃/RGO samples. The pure Sn₄P₃ sample exhibits a good thermal stability. It almost keeps completely structural stable in the temperature range of 0–750 °C. Only a few weight losses were observed in higher temperature. According to the TGA results in Figure S2b (Supporting Information), the content of carbon in the Sn₄P₃/RGO-1, Sn₄P₃/RGO-2, Sn₄P₃/RGO-3 hybrid can be estimated to be 6.3, 10.4, 15.5 wt%, respectively.

Transmission electron microscopy (TEM) is applied to further investigate the detailed microstructures of the synthesized Sn₄P₃/RGO-2 materials, as shown in Figure 3. From low-magnification TEM images in Figure 3a,b, it is revealed that the monodisperse Sn₄P₃ fine nanoparticles are loaded on the RGO nanosheets uniformly. The 3D porous interconnected architectures for the synthesized Sn₄P₃/RGO products can be clearly revealed. The different color contrast in Figure 3a suggests the presence of different layers of the RGO nanosheets. The diffraction rings in the upper inset of Figure 3b are an electron diffraction pattern, corresponding well with that of the (015), (110), (107), and (027) planes of Sn₄P₃, in good agreement with the XRD results.[37] The electron diffraction pattern suggests a polycrystalline nature for the rhombohedral Sn₄P₃ nanoparticles. The high-magnification TEM images in Figure 3c and Figure S3 (Supporting Information) indicate that the Sn₄P₃ nanoparticles with an average size of 6 nm are homogenously loaded on the RGO nanosheet matrix. The HRTEM lattice image of the edge for the RGO matrix shows two to three atomic carbon layers for the RGO nanosheets. The marked d-spacings of 0.33 and 0.31 nm in the high resolution transmission electron microscopy (HRTEM) lattice image in Figure 3d are in good agreement with that of the (012) and (015) planes of rhombohedral Sn₄P₃.[37] Figure 3e–h shows the scanning transmission electron microscopy (STEM) image and the energy dispersive X-ray (EDX) elemental mapping images of Sn, P, and C of the Sn₄P₃/RGO-2 sample, indicating the synthesized products are composed of Sn, P, and C elements, and Sn, P, and C elements are homogeneously distributed among the whole region of Figure 3e for the Sn₄P₃/RGO-2 sample.

The nitrogen adsorption–desorption isotherms and the corresponding pore size distribution curves of the synthesized Sn₄P₃/RGO samples obtained by Barrett–Joyner–Halenda method are shown in Figure S4 (Supporting Information). The Sn₄P₃/RGO-2 sample shows a characteristic of IV type with an obvious capillary condensation step, indicating a mesoporous nature (Figure S4b, Supporting Information).[46, 47] The specific surface area and average pore diameter of Sn₄P₃/RGO-2 are about 231 m² g⁻¹ and 8.6 nm, respectively. The higher surface area and appropriate pore diameter of Sn₄P₃/RGO-2 sample can be useful to enhance the sodium storage performance. The hysteresis loop shown in isotherms of Sn₄P₃/RGO-1 sample is very narrow and small, suggesting the presence of macroporous in the sample (Figure S4a, Supporting Information).[47] The specific surface area and average pore diameter of Sn₄P₃/RGO-1 sample are about 137 m² g⁻¹ and 38.4 nm, respectively. The surface area of Sn₄P₃/RGO-3 samples can reach a value of about 337 m² g⁻¹, which can be attributed to the high content of RGO nanosheets.

To understand the electrochemical reaction of the synthesized Sn₄P₃/RGO samples, the cyclic voltammetry (CV) curves for the first five cycles of the Sn₄P₃/RGO-2 sample are investigated at a scan rate of 0.1 mV s⁻¹ in a voltage range of 0.01–3.00 V versus Na⁺/Na (Figure 4). In the case of the first cathodic half-cycle, three reduction peaks located at 2.0, 1.2, and 0.03 V are observed, respectively. The broad cathodic peak at around 2.0 and 1.2 V could be attributed to the formation of an solid electrolyte interface (SEI) film.[37] These peaks disappear in the subsequent cycles. Another cathodic peak at 0.03 V can be ascribed to the Na insertion reaction in Sn₄P₃ to form Na₁₅Sn₄ and Na₃P alloy (Equation (1)).[36] In the reversed scan and subsequent cycles, the Sn₄P₃/RGO-2 electrode displays two pairs of symmetric peaks at 0.3 and 0.65 V, indicative of a two-stepped redox reaction. In the case of the first anodic process, the anodic peak at about 0.3 V corresponds to the dealloying process of the Na₁₅Sn₄ alloys (Equation (2)).[36, 37] The peaks at a higher potential of 0.65 V could be attributed to the reversible conversion reaction of P element (Equation (3)) together with further Na-Sn reaction (Equation (2)). In the second cathodic cycle, the large reduction peak at 0.1 V observed in the first cathodic process splits into two peaks at about 0.3 and 0.1 V, though the peak at 0.3 V is indistinct. The splitting process can be attributed to the electrode activation.[35] The peak at 0.3 V corresponds to the formation of Sn and Na₃P. The other peak at 0.1 V is believed to be associated with the formation of Na₁₅Sn₄ alloy.[35, 36] After the first cycle, the CV curves of the Sn₄P₃/RGO-2 electrode in the second cycle remain similar to the first cycle, and the curves after the second cycle are nearly overlapped, suggesting an excellent cycle performance of the electrodes.

Sn4P3+24Na++24e−→Na15Sn4+3Na3P(1)

Na15Sn4↔4Sn+15Na++15e−(2)

Na3P↔P+3Na++3e−(3)

To confirm above the reaction mechanism, we disassemble coin cells at different discharge/charge stages and track the phase component evolution of electrode materials by ex situ XRD. Figure 5 shows the XRD patterns of the Sn₄P₃/RGO-2 electrode discharged/charged at different states in the first cycle. As shown in Figure 5, the anode before discharge shows a clear XRD pattern corresponding to that of Sn₄P₃. When the Sn₄P₃/RGO-2 electrode is discharged at 0.25 V, the characteristic peaks of Sn and Na₃P are observed in the XRD pattern. The primary diffraction peak of Sn₄P₃ at 28.8° and 31.5° gradually diminishes during the discharge process, corresponding to Equation (1). It indicates that the Sn₄P₃ lattice has transformed to metallic Sn and Na₃P phases at an initial stage of Na insertion reaction before the discharge potential of about 0.25 V.[36] Besides Na₃P, Na₁₅Sn₄ peak at 21.3° is also detected in the XRD pattern, when the Sn₄P₃/RGO-2 electrode completely charges to a terminate potential of 0.01 V, the peak of Sn₄P₃ completely vanishes, corresponding to Equation (1). However, the Sn peak is still present, though slightly weakens, indicating that Sn extracted from Sn₄P₃ cannot react with Na completely to form the Sn-Na alloy. This conclusion can also be confirmed by Qian et al.[36] When the Sn₄P₃/RGO-2 electrode is charged to 2.0 V, a stronger Sn peak and a weaker P peak can be observed in the XRD, corresponding to the reaction of Equations (2) and (3).

To reveal the sodium storage mechanism of Sn₄P₃/RGO composite, ex situ TEM is used to investigate the structure and phase changes of the Sn₄P₃/RGO-2 electrodes at different discharge–charge states. Figure S5 (Supporting Information) shows the TEM images of the Sn₄P₃/RGO-2 electrode after first discharge to 0.01 V and full charge to 2.0 V. From low-magnification TEM images of these two states in Figure S5a,c (Supporting Information), it is suggested that the monodisperse nanoparticles are loaded on the RGO nanosheets uniformly after cycling. The particle diameter at fully discharged state is larger than that of charged state due to the Na allaying process. Figure S5b (Supporting Information) shows the HRTEM image and the selected area electron diffraction pattern of Sn₄P₃/RGO-2 sample which was discharged to 0.01 V.Two phases of Na₃P and Na₁₅Sn₄ can be identified. The marked d-spacing of 0.281 and 0.249 nm in the HRTEM lattice image corresponds well to d-spacing of the (332) plane of Na₁₅Sn₄ and the (110) plane of Na₃P.[37] The diffraction rings correspond to the (110) and (207) planes of Na₃P, and the (108) plane of Na₁₅Sn₄. When the electrode is charged to 2.0 V, metallic Sn phase is detected in the sample. The marked d-spacing of 0.206 nm corresponds well to d-spacing of the (220) plane of Sn, as shown in Figure S5d (Supporting Information). The diffraction rings are in good agreement with the (501), (211), (200), and (101) planes of metallic Sn. No crystalline P is observed in the HRTEM images, which may be attributed to an amorphous state for P.[48-50] The ex situ TEM characterization further reveals the electrochemical reaction mechanism.

Successful sodiation of Sn₄P₃ is also manifested by the elemental mapping images of the fully discharged electrode, as shown in Figure S6 (Supporting Information). A stronger Na peak can be observed in the energy dispersive spectroscopy (EDS) spectrum. Na, Sn, P, and C elements are homogeneously distributed among the whole electrodes. Ex situ XRD, SEM, and TEM analysis suggests successful sodiation and desodiation of Sn₄P₃/RGO hybrid electrodes.[48-50]

The electrochemical impedance spectroscopy (EIS) is used to investigate the charge transfer kinetics of the Sn₄P₃/RGO SIBs electrodes. Figure 6 shows the Nyquist profiles of the AC impedance measured at an open circuit voltage state of fresh cells and cells fully charged state after several cycles for the Sn₄P₃/RGO electrodes. As shown in Figure 6a, Sn₄P₃/RGO samples all show one depressed semicircle in high frequency and a straight line portion in low frequency range. The semicircle in the high and middle frequency regions is attributed to contact resistance (R_f) and charge-transfer impedance on electrode–electrolyte interface (R_ct), whereas the inclined line in low frequency region of the spectrum is ascribed to the Warburg impedance (R_w), which is related to the diffusion coefficient of Na⁺ at the interface between the electrode and the electrolyte.[43, 46, 49] Sn₄P₃/RGO-2 and Sn₄P₃/RGO-3 samples show a smaller semicircle than that of Sn₄P₃/RGO-1 sample, indicating that these two samples have the improved charge transfer ability at the interface between the electrolyte and electrode. Figure 6b shows the Nyquist plots of the Sn₄P₃/RGO-2 hybrid electrode after the 1st, 10th, 50th, and 100th cycles at the full charge state. The semicircles in the high and middle frequency regions become smaller with cycle times increasing, indicating a smaller charge-transfer impedance on electrode–electrolyte interface after cycling. The decrease in the charge transfer resistance of the electrode is due to the surface activation and lattice expansion, which triggers sodium ion transport.[50] The Nyquist profiles after the 1st, 10th, 50th, and 100th cycles exhibit a smaller charge transfer impedance, indicating no significant growth of the SEI layer and a stable structure of the anode materials.

Galvanostatic technique is utilized to investigate the charge–discharge cycling performance of Sn₄P₃/RGO electrodes in the potential window of 0.01–3.00 V at a current density of 100 mA g⁻¹. Figure 7 shows the capacity performance, Coulombic efficiency, rate capacity curves, and long cycling life of the Sn₄P₃/RGO electrodes. It is clearly indicated that the Sn₄P₃/RGO-2 sample displays an outstanding electrochemical performance. The capacity value is calculated based on the whole Sn₄P₃/RGO weight. The Sn₄P₃/RGO-2 hybrid anode displays an initial discharge and charge capacity of 1663 mA h g⁻¹ and 775 mA h g⁻¹, respectively, showing a Coulombic efficiency of about 46.6%. The Sn₄P₃/RGO-2 anode demonstrates the better cycling performance, showing a superior reversible capacity of 656 mA h g⁻¹ at a current density of 100 mA g⁻¹ after 100 cycles. While Sn₄P₃/RGO-1 and Sn₄P₃/RGO-3 samples show a lower specific capacity relatively. The specific capacity of Sn₄P₃/RGO-1 and Sn₄P₃/RGO-3 samples is only 275 mA h g⁻¹ and 435 mA h g⁻¹, respectively. The first Coulombic efficiency of Sn₄P₃/RGO-1 and Sn₄P₃/RGO-3 samples is 42.8% and 37.8%, respectively. The initial capacity loss of the synthesized samples is primarily due to the consumption of Na-ions for the formation of an SEI layer during the first discharge step.[37] It is believed that binders, electrolyte additive, and microstructure are the main factors to influence the Coulombic efficiency of the electrode. Komaba et al.[29] reported that FEC is an effective additive in PC electrolyte to help the formation of a contact SEI film on the surface of the electrode. Certain concentration FEC additive in PC electrolyte can suppress the unfavorable side reactions to improve the first Coulombic efficiency for sodium ion battery. Komaba et al.[31] also investigated the binder effect on the performance of Sn SIBs anodes. It is shown that PAA binder can improve the specific capacity and the first Coulombic efficiency instead of PVDF binder. The first Coulombic efficiency of Sn SIBs electrode can reach 86% for PAA binder from 61% for PVDF binder. The volume expansion will cause the fresh anode surface re-expose to the electrolyte and to form the SEI again, resulting in further electrolyte decomposition and low Coulombic efficiency. In addition, the surface area of the electrode material is also an important factor to influence the Coulombic efficiency. Carbon materials with large surface area usually demonstrate low Coulombic efficiency and large irreversible capacity due to the electrolyte decomposition and undesirable irreversible reaction. For example, Wang et al.[51] reported the RGO material as SIBs anode, showing a low first Coulombic efficiency of about 20% due to the irreversible reaction. In Sn₄P₃/RGO sample, RGO with large surface area will adsorb more electrolyte and lead to undesirable irreversible reactions and electrolyte decomposition. All these factors result in the lower Coulombic efficiency. So appropriate binder, electrolyte additive, and rational microstructure can enhance the Coulombic efficiency of the Sn₄P₃/RGO electrode. Compared with the Sn₄P₃/RGO anode, the specific capacity of pure Sn₄P₃ nanoparticle electrode is also tested, as shown in Figure S7a (Supporting Information), showing only 38 mA h g⁻¹ after 100 cycles at a current density of 100 mA g⁻¹. The enhanced electrochemical performance of Sn₄P₃/RGO-2 sample can be attributed to the synergistic interaction between Sn₄P₃ nanoparticles and the graphene matrix.

It should be noted that the synthetic route and microstructures of the Sn₄P₃/RGO architecture nanohybrids display advantages compared with the previously high-energy balling or mechanical alloying route to synthesize Sn_xP₃. The in situ low-temperature solution-based phosphorization chemical transformation route for fabricating Sn₄P₃/RGO is more effective, and that the microstructure and chemical components of the Sn₄P₃/RGO nanohybrids are more easier to be controlled. More importantly, the specific capacity and rate capability of the Sn₄P₃/RGO electrodes is much superior to that of previously reported results. For example, Li et al.[34] prepared Sn_4+xP₃@Sn-P nanohybrid by direct low-speed ball milling of the P and Sn powders and studied its electrochemical performance as an anode for SIBs, displaying a capacity of 465 mA h g⁻¹ at the current density of 100 mA g⁻¹ over 100 cycles. Qian et al.[36] synthesized Sn₄P₃/C nanohybrid by a mechanical chemical reaction of metallic Sn, elemental P, and carbon, exhibiting a capacity of 500 mA h g⁻¹ at the current density of 100 mA g⁻¹ over 150 cycles. Yu and co-workers[37] synthesized yolk–shell Sn₄P₃@C nanospheres via a top-down phosphorized route, showing a capacity of 515 mA h g⁻¹ after 50 cycles at a current density of 100 mA h g⁻¹. Table S1 (Supporting Information) shows the capacity comparison of the present work with that of the reported Sn₄P₃ materials. It should be noted that the cutoff voltage window in precious work is at about 0.01–2.00 V. In the present work, the voltage range is larger than that of the previous report. It is shown that the capacity change is very small from the voltage of 2.00–3.00 V, as depicted in the discharge–charge curves of Figure S8 (Supporting Information). Compared with previous reports, the present Sn₄P₃/RGO anode displays the competitive retained capability.

To investigate the rate capability of the Sn₄P₃/RGO hybrid anodes, the Sn₄P₃/RGO hybrid anodes are cycled for each ten cycles under an increasing current density from 0.1 A g⁻¹ to 2.0 A g⁻¹ (Figure 7b). It is revealed that Sn₄P₃/RGO-2 sample shows the best rate capability performance. In the first ten cycles at a current density of 0.1 A g⁻¹, the reversible capacity of the Sn₄P₃/RGO-2 hybrid anode decreases with the cycle time increasing, displaying a reversible capacity of 670 mA h g⁻¹ at the 10th cycle. Subsequently, the reversible capacity of the Sn₄P₃/RGO-2 hybrid anode shows a reversible capacity of 522 mA h g⁻¹ at a current density of 0.5 A g⁻¹ after 30 cycles, and 391 mA h g⁻¹ at a current density of 2.0 A g⁻¹ after 50 cycles. As the current density is set back to 0.1 A g⁻¹, the capacity is able to return to more than 628 mA h g⁻¹. Figure S8 (Supporting Information) shows the charge–discharge curves of Sn₄P₃/RGO-2 sample at different current densities from 0.1 to 2.0 A g⁻¹. All the capacities are calculated based on the total mass of Sn₄P₃/RGO composite, not just on the Sn₄P₃. The first discharge and charge capacity are 1640 and 758 mA h g⁻¹, respectively, corresponding to a Coulombic efficiency of about 46.2%. The charge plateaus can be observed clearly at the lower current density. However, increased polarization is observed at the higher current density.

The rate capacity of Sn₄P₃/RGO-1 and Sn₄P₃/RGO-3 samples is only 44 mA h g⁻¹ and 159 mA h g⁻¹ at a current density of 2.0 A g⁻¹, respectively. The bare Sn₄P₃ sample shows a specific capacity of only 9.8 mA h g⁻¹ at a current density of 2.0 A g⁻¹ after 50 cycles (Figure S7b, Supporting Information), demonstrating that the electrochemical performance of the Sn₄P₃ can be enhanced remarkably by coupling with graphene. It should be stressed that the Sn₄P₃/RGO-2 sample shows the superior long cycling life and still retains a high capacity of 362 mA h g⁻¹ over 1500 cycles at a high current density of 1.0 A g⁻¹. The superior rate capacity and longer cycling stability can be potential for its large scale application.

The remarkable electrochemical performance especially the high specific capacity and superior cycling stability of the Sn₄P₃/RGO hybrids could be attributed to the synergistic effect and interaction between Sn₄P₃ and RGO. (1) The monodisperse Sn₄P₃ nanoparticles with an average diameter of 6 nm are uniformly loaded on the RGO nanosheets, interconnecting with RGO matrix to form 3D porous architectures structures. The mesoporous structure not only provides an elastic buffer space to accommodate the volume expansion/contraction of Sn₄P₃ during Na⁺ ions insertion/extraction process but also shortens the ion diffusion path, maintaining large specific capacity, high rate capability, and longer cycling stability. In order to investigate the effect of graphene in alleviating and accommodating volume expansion/contraction of Sn₄P₃/RGO SIBs anode, the changes of electrode thickness of Sn₄P₃ and Sn₄P₃/RGO SIBs anode during sodiation and desodiation of Sn₄P₃/RGO and Sn₄P₃ electrodes are observed, as indicated in Figure S9 (Supporting Information). The electrode thickness of bare Sn₄P₃ demonstrates about 94% larger change from 17.3 to 33.6 μm compared with that of Sn₄P₃/RGO electrode. While the Sn₄P₃/RGO-2 electrode shows smaller change of only about 60% in thickness, from about 16.5 to 26.4 μm. (2) Sn₄P₃ combines with the advantages of high conductivity of Sn and high capacity of P and displays the high theoretical volumetric capacity of 6650 mA h cm⁻³ and high gravimetric capacity of 1132 mA h g⁻¹. (3) Flexible RGO sheet not only alleviates the large volume expansion of Sn₄P₃ but also provides a conductive path to enhance capacity retention ability during repeated cycles. (4) EIS spectra indicate that RGO nanosheets can greatly improve the charge transfer kinetics of Na⁺ ions at the interface between the electrolyte and electrode, enhancing the rate capacity of the Sn₄P₃ anode.

It should be noted that the Sn₄P₃/RGO-3 with higher RGO concentration does not demonstrate better cycling performance and higher rate performance, which can be attributed to the structure characteristics and the synergistic effect between RGO and Sn₄P₃ active material. The rate capability and cyclability of the Sn₄P₃/RGO composite depend on many factors such as charge transfer kinetics, microstructure such as surface area and porous structure, and the amount of the active Sn₄P₃ in the Sn₄P₃/RGO hybrid. First, carbon materials with large surface area, such as graphene, mesoporous carbon, carbon aerogel, etc.,[52, 53] demonstrate low Coulombic efficiency and large irreversible capacity due to the electrolyte decomposition and undesirable irreversible reactions.[46] Second, the charge transfer ability of Sn₄P₃/RGO-3 sample (493 Ω cm²) is only slightly improved compared with that of Sn₄P₃/RGO-2 sample (515 Ω cm²) (Figure 6a). Last, the theoretical capacity of the whole Sn₄P₃/RGO composite decreases with increase in the concentration of RGO. In the present work, Sn₄P₃/RGO-2 composite with a rational concentration of RGO exhibits optimized synergistic effects between RGO and Sn₄P₃ nanoparticles.[54, 55] Similar result is reported for the electrochemical performance of the SnO₂/RGO composite by Wang et al.[55]

3 Conclusion

A facile low-temperature solution-based phosphorization reaction strategy has been developed for the fabrication of Sn₄P₃/RGO nanohybrids with the adjustable graphene content. The monodisperse Sn₄P₃ nanoparticles with a smaller diameter of 6 nm can be loaded on the graphene nanosheets uniformly, interconnecting to form 3D mesoporous architecture structures. The Sn₄P₃/RGO-2 nanohybrid with a graphene concentration of about 10.4 wt% exhibits a higher sodium storage capacity of 656 mA h g⁻¹ at a current density of 100 mA g⁻¹ over 100 cycles, and a remarkable rate capability of 391 mA h g⁻¹ even at a higher current density of 2.0 A g⁻¹. In addition, the longer cycling performance with a capacity of 362 mA h g⁻¹ at a current density of 1.0 A g⁻¹ can be obtained after 1500 cycles. The outstanding cycling performance and rate capability of these porous hierarchical Sn₄P₃/RGO hybrid anodes can be attributed to the advantage of porous structure, and the synergistic action between Sn₄P₃ nanoparticles and graphene nanosheets. The Sn₄P₃/RGO nanohybrids with high rate capacity and long cycling life can find potential applications as anode materials in SIBs.

4 Experimental Section

Preparation of Sn₄P₃/RGO Sample: Sn₄P₃/RGO nanohybrids were synthesized by a facile solvothermal route using Sn/RGO nanohybrid and red phosphorus as precursors. The graphene oxide nanosheets used in this experiment were synthesized by a modified Hummer's method.[56] The graphene oxide was filtrated and washed with de-ionized water until pH to about 7 at the last stage of preparation, and then graphene oxide was preserved in de-ionized water to form the graphene oxide suspension solution.

First, Sn/RGO nanohybrids were synthesized by reducing SnCl₂ using NaBH₄ as a reducing agent. In a typical synthesis of Sn/RGO nanohybrid, PVP (0.01 g) and citric acid (0.02 g) were dissolved in 20 mL de-ionized water under ultrasonication. 5 mL graphene oxide suspension (about 2 mg mL⁻¹) was added into above solution to form solution A. 0.2 g SnCl₂·2H₂O was dispersed in 20 mL de-ionized water under ultrasonication to form solution B. Solution B was added into solution A drop by drop under stirring for 1 h. Then, the mixture was transferred into a 250 mL round-bottom flask. The flask was purged with high purity Ar as a protecting gas. 20 mL NaBH₄ (90 mg) aqueous solution was gradually added to the mixture to reduce Sn²⁺ to Sn, and the oxidized graphene (GO) could also be reduced to reduced oxidized graphene (RGO). The mixture was stirred for 1 h. The resultant black products were separated from the reaction mixture by centrifugation and washed several times with ultrapure water and ethanol to remove impurities. Finally, Sn/RGO nanohybrids were dried in a vacuum oven at 60 °C.[39, 57]

In a typical synthesis of Sn₄P₃/RGO nanohybrids, 0.6 g Sn/RGO nanohybrid and 0.232 g red phosphorus were added into 30 mL ethylenediamine solution and kept stirring for 12 h. Then, the resultant mixture was transferred into a 40 mL teflon-lined stainless-steel autoclave. Teflon-lined autoclave was put into an oven at 200 °C for 40 h. The system was then cooled to ambient temperature naturally. To remove the redundant tin from the as-produced samples, the products were washed under vigorous stirring in dilute HCl aqueous (0.1 mol L⁻¹) for 12 h, then washed with water and ethanol, and then dried in a vacuum oven at 60 °C.[37, 41, 42, 54] The prepared samples were denoted as Sn₄P₃/RGO-1, Sn₄P₃/RGO-2, and Sn₄P₃/RGO-3 for 5, 10, and 15 mL graphene oxide suspension.

Structure and Electrochemical Characterization: The crystal structure of the synthesized samples was determined by powder XRD (Rigaku D/Max-KA diffractometer with Cu Ka radiation). Raman spectroscopy was carried out using a JY HR800 micro Raman spectrometer. SU-70 field emission scanning electron microscopy and transmission electron microscopy (TEM JEM-2010) were employed to analyze the microstructures of the synthesized products. The specific Brunauer–Emmett–Teller surface area was determined by N₂ adsorption/desorption on a V-Sorb 2800 series analyzer (Gold APP Co. Ltd.). TGA was carried out with the temperature ramp of 10 °C min⁻¹ from 40 to 750 °C using thermogravimetric analyzer (SDTQ600).

The working electrode slurry was prepared by mixing 80 wt% active materials, 10 wt% acetylene carbon black, 10 wt% PVDF binder, and an adequate amount of N-methyl-2-pyrrolidone. The performance of the SIBs was tested using standard 2025 type coin cells with copper foil as the current collectors, Na foil as counter electrode and reference electrodes, glass fiber (GF/D) from Whatman as a separator, and 1.0 m NaClO₄ in PC as the electrolyte, plus 5 wt% FEC. The cutoff voltage window is 0.01–3.00 V. Galvanostatically cycled test was carried out on a LAND CT2001A instrument (Wuhan, China) at room temperature. Electrochemical workstation (PARSTAT2273) was used to study CV behavior in a potential window of 0.01–3.00 V at a scan rate of 0.1 mV s⁻¹. EIS measurements were performed in the frequency between 100 kHz and 10 mHz and the amplitude is 5 mV.[46, 47]

Acknowledgements

The authors acknowledge support from the project supported by the Sate Key Program of National Natural Science of China (Grant No.: 51532005), the National Nature Science Foundation of China (Grant Nos.: 51472148 and 51272137), and the Tai Shan Scholar Foundation of Shandong Province.