High performance carbon-coated hollow Ni₁₂P₅ nanocrystals decorated on GNS as advanced anodes for lithium and sodium storage†

Huinan Guo ^a, Chengcheng Chen ^ad, Kai Chen ^a, Haichao Cai ^a, Xiaoya Chang ^a, Song Liu ^a, Weiqin Li ^a, Yijing Wang *^ab and Caiyun Wang *^c
aKey Laboratory of Advanced Energy Materials Chemistry (Ministry of Education), College of Chemistry, Nankai University, Tianjin 300071, China
^bCollaborative Innovation Center of Chemical Science and Engineering, College of Chemistry, Nankai University, Tianjin 300071, China. E-mail: wangyj@nankai.edu.cn
^cIntelligent Polymer Research Institute, ARC Centre of Excellence for Electromaterials Science, University of Wollongong, Wollongong, NSW 2500, Australia. E-mail: caiyun@uow.edu.au
^dChina Electronic Product Reliability and Environmental Testing Research Institute (CEPREI), Guangzhou 510610, China

First published on 2nd October 2017

---

Transition-metal phosphides have been considered as promising anode materials for rechargeable secondary batteries owing to their low cost and high capacity. However, low electronic conductivity and poor stability limit their further development. Herein, we have designed a template-free refluxing method for synthesizing tailored carbon-coated hollow Ni₁₂P₅ nanocrystals in situ grown on reduced graphene oxide nanosheets (denoted as Ni₁₂P₅@C/GNS). The hollow structure can accommodate volume expansion and shorten the ion transfer path. The GNS loading and carbon shell can efficiently prevent Ni₁₂P₅ from aggregating and improve the electronic conductivity. As an anode of Li-ion batteries (LIBs), the hollow Ni₁₂P₅@C/GNS composite displays an excellent discharge specific capacity of 900 mA h g⁻¹ at a current density of 100 mA g⁻¹ after 100 cycles and outstanding rate capability. Furthermore, it also shows a good Na storage capability with a reversible capacity of 235 mA h g⁻¹ at 100 mA g⁻¹. Therefore, our work demonstrates that this hollow Ni₁₂P₅@C/GNS composite has great potential for Li/Na storage.

---

1. Introduction

Rechargeable secondary batteries have been one of the most important energy storage technologies recently.^1–3 LIBs have occupied a dominant position among portable electronic devices due to their high operating voltage, high energy density and limited self-discharging.^4–9 Besides, sodium-ion batteries (SIBs) are expected to be broadly applied to large energy storage equipment,^10–14 owing to the low cost and natural abundance of sodium resources.^15,16 However, commercial anode materials hardly show satisfactory performances in both Li/Na storage systems, due to the more sluggish kinetics of sodium than that of lithium.^17,18 Therefore, it is urgent to search for advanced anode materials in terms of high capacity, long cycle life and superior rate capability for both LIBs and SIBs.^19,20

Transition-metal phosphides (TMPs) have been applied to several fields, especially in electrocatalysis and photocatalysis.^21–23 Nevertheless, their applications as anode materials are not fully developed. Conversion type TMPs have been rising anode materials for rechargeable secondary batteries in recent years.^24–28 As anode materials for lithium/sodium ion batteries, based on the reversible reaction between phosphorus and Li/Na (Li₃P/Na₃P), TMPs display high specific capacity (up to ∼1800 mA h g⁻¹),^29,30 low operating potential, and metallic features as well as good thermal stability. However, the large volume expansion and poor electronic conductivity of TMPs seriously limit their kinetic properties and cycling stability. Facing similar challenges, Lou's group³¹ prepared a tailored hollow structure aiming to buffer the substantial volume change and boost the electrochemical performances. The study demonstrates that the hollow structure can significantly accommodate pulverization by restricting volume expansion. Additionally, Mai and co-workers³² synthesized novel layered Li₃V₂(PO₄)₃, in which LVP layers were uniformly alternated with reduced graphene oxide nanosheets to provide effective electron transport. The rate capacities of the prepared Li₃V₂(PO₄)₃/rGO&C are better than most of the state-of-the-art reported results. Thus, it is interesting and worth designing a hollow nanocrystal, and combining it with carbon materials to improve the electrochemical performances of Ni₁₂P₅.

On the other hand, the synthesis methods of TMPs are usually defective. Generally speaking, there are two primary routes to prepare TMPs. High energy mechanical milling is simple; however, the products of this method are always without a hollow morphology.^33–35 Another way is a solid reaction using hypophosphites (NH₄H₂PO₂ and NaH₂PO₂) as the P source. This process releases very flammable and poisonous PH₃ gas.^36,37 Hence, designing a facile and low toxicity method for the synthesis of TMPs is not only urgent, but also still remains challenging.

Herein, we first report a novel hollow structure of carbon-coated Ni₁₂P₅ nanocrystals supported on reduced graphene oxide nanosheets via a simple refluxing method without any templates and surfactants, as high performance anodes for LIBs and SIBs. This pathway is facile and effective. During the synthesis, the phosphorus source is consumed in situ without hazardous gas escape. Meanwhile, the Ni₁₂P₅@C/GNS is endowed with three merits: (i) the hollow Ni₁₂P₅@C with a small average size of 35 nm effectively shortens ion/electron transport paths and facilitates the reaction kinetics; (ii) the hollow structure and carbon shell can buffer volume expansion wonderfully; and (iii) the GNSs prevent aggregation and provide an electric net between grains. As expected, when evaluated as an anode material for LIBs, the tailored Ni₁₂P₅@C/GNS electrode shows a high reversible capacity of 900 mA h g⁻¹ at a current density of 100 mA g⁻¹ after 100 cycles, and an excellent rate capability with discharge specific capacities of 702.2, 644.1, 566.5, 482.8, 423.2 and 905.9 mA h g⁻¹ at 100, 200, 500, 1000, 2000 and 100 mA g⁻¹, respectively. We also adopt this material as an anode of SIBs for the first time, which delivers excellent performance with a reversible capacity of 235 mA h g⁻¹ at 100 mA g⁻¹. Therefore, the Ni₁₂P₅@C/GNS composite is a promising anode material with high discharge specific capacities and good rate capabilities for LIBs and SIBs.

2. Experimental section

2.1 Chemicals

Nickel(II) acetate tetrahydrate (AR) was purchased from Tianjin Guangfu Fine Chemical Research Institute and triphenylphosphide (TPP) was purchased from Alfa Aesar. Oleylamine (OLA) was purchased from J&K. All reagents were used as received without further purification.

2.2 Synthesis of reduced graphene oxide

Graphene oxide was prepared following a modified Hummers' approach,³⁸ and then reduced by using H₂ (5%)/Ar gas.

2.3 Synthesis of Ni₁₂P₅/GNS

Ni₁₂P₅ was synthesized using a facile refluxing condensation method. Briefly, 1 mmol Ni(Ac)₂·4H₂O (0.25 g) was dispersed in 30 mmol OLA (8 g) in a 100 mL three-necked flask. The flask was heated using a heating panel while being magnetically stirred under an Ar flow. The solution mixture was heated to 120 °C until all Ni(Ac)₂·4H₂O was dissolved and the solution turned green. Then 10 mmol TPP (2.6 g) and a certain amount of GNS were added into the solution. The growth solution was heated to 270 °C and kept at this temperature for 30 min. Then the solution was naturally cooled to ambient temperature. The black precipitate was isolated and washed three times with a mixture of hexane and ethanol by centrifugation. Finally, the products were dried at 80 °C under vacuum for further characterization.

2.4 Synthesis of Ni₁₂P₅@C/GNS

In a typical transformation process, the as-prepared Ni₁₂P₅/GNS was annealed at 450 °C in a H₂ (5%)/Ar atmosphere for 30 min.

2.5 Characterization

The products were characterized by X-ray diffractometry using a Rigaku MiniFlex II diffractometer equipped with a Cu Kα radiation source (λ = 1.54178 Å) at a scan rate of 4° min⁻¹. Raman measurement was performed on an inVia/Reflex Laser Micro-Raman spectrometer (Horiba Jobin Yvon, France). XPS spectra were collected by using a VG Scientific ESCALAB 2201XL System equipped with a monochromatic Al Kα source. The weight percentage of carbon was determined by elemental analysis (EA, Thermal Conductivity Detector, Elemental vario E cube). Inductively coupled plasma-atomic emission spectroscopy (ICP-AES) was performed on a PerkinElmer Optima 8300 instrument. Field-emission scanning electron microscopy (SEM) measurements were performed on a JEOL JSM-7500F (5 kV). Transmission electron microscopy (TEM) studies were conducted on an FEI Philips Tecnai electron microscope at an operating voltage of 200 kV.

2.6 Electrochemical measurements

Half cells were assembled in standard CR2032 coin-type model cells, with Li/Na foil as the counter and reference electrodes. The working electrode was prepared by mixing Ni₁₂P₅@C/GNS powder, super P and polyvinylidene fluoride (PVDF) in a weight ratio of 7:2:1. Then the mixture was ground in a mortar with N-methyl-2-pyrrolidone (NMP) as a solvent for making a slurry and pasted on pure copper foil (99%). The average mass loading of the active materials within the film was about 0.98 mg cm⁻². For lithium ion batteries, the electrolyte was 1 M LiPF₆ in ethylene carbonate (EC)/diethyl carbonate (DEC) (1:1 by volume ratio). A polypropylene film (Celgard 2400) was used as the separator. For sodium ion batteries, the electrolyte was 0.1 M NaClO₄ in diethylene glycol dimethyl ether (DEG-DME) with the addition of 5% fluoroethylene carbonate (FEC). Glass fiber was used as the separator. The battery tests were carried out in an Ar-filled glovebox (H₂O, O₂ < 1 ppm, Mikrouna). Cycling performances were recorded on a CHI660E electrochemical workstation at a scan rate of 0.2 mV s⁻¹ from 0.1 to 3 V. Galvanostatic charging and discharging tests were performed using a LAND CT2001A instrument.

The electrochemical performances of full cells were also measured using a two electrode coin-type cell (CR2032). To prepare the positive electrode, a mixture of LiCoO₂, Super P and PVDF in a weight ratio of 7:2:1 was dispersed in NMP solvent and ground thoroughly into a slurry, and then pasted on aluminum foil. After being dried at 80 °C, the aluminum foil with active materials was cut into small rounds for the battery cathode. The negative electrode of Ni₁₂P₅@C/GNS was prepared by the same method. For assembling full cells, the mass ratio of cathode to anode was 16:1. The excess cathode was used to ensure that the Ni₁₂P₅@C/GNS anode reacts thoroughly.

3. Results and discussion

3.1 Structural and morphological characterization

We employ GNS as an assistant agent to build hollow Ni₁₂P₅ nanocrystals through a facile template-free refluxing method (Scheme 1). The XRD patterns confirm that the Ni₁₂P₅@C/GNS, pure Ni₁₂P₅@C and GNS are pure phases, as displayed in Fig. 1a and S1.† The products fully correspond to the tetragonal unit cell (a = 8.464 Å, c = 5.070 Å, JCPDS card no. 22-1190) with the I4/m space group. In this structure (Fig. 1b), Ni atoms occupy 16i and 8h sites connected with two P atoms, while there are two kinds of P atoms, which reside in the 2a sites connected with eight Ni and 8h sites connected with four Ni, respectively. The SEM images (Fig. S2†) clearly show the sheet-like structure of GNSs and nanocrystal structure of pure Ni₁₂P₅@C. Fig. S2b† displays the serious aggregation of Ni₁₂P₅@C without GNS. Fig. 1c exhibits the SEM image of the Ni₁₂P₅@C/GNS composite. Notably, Ni₁₂P₅@C nanocrystals with a uniform size are homogeneously dispersed on the surface of GNS. In addition, compositional information is determined by using the energy-dispersive spectrum (EDS, Fig. S3†). The atomic ratio of Ni to P is measured to be 2.32:1, in accordance with that of stoichiometric Ni₁₂P₅. According to N₂ adsorption/desorption measurement and Brunauer–Emmett–Teller (BET) analysis, the existence of mesopores in the Ni₁₂P₅@C/GNS is observed and the specific surface area is estimated to be about 78.7 m² g⁻¹ (Fig. 1d), which is higher than that of pure Ni₁₂P₅@C (41.8 m² g⁻¹, Fig. S4†). The elemental analysis result shows that the carbon content of Ni₁₂P₅@C/GNS is 18.15%. The content of Ni in the Ni₁₂P₅@C/GNS composite is estimated to be 66.3 wt% by using the ICP instrument. Thus, the weight percentage of Ni₁₂P₅ in the hybrid is about 80.9%. Raman spectroscopy is employed to investigate the conductivity of carbon of Ni₁₂P₅@C/GNS. As shown in Fig. S5,† two strong carbon peaks located at 1346 cm⁻¹ (D-band, sp³-coordinated behavior) and 1580 cm⁻¹ (G-band, sp²-hybridized carbon) can apparently be observed.^39,40 The Raman peak intensity ratio (I_D/I_G = 0.78) is inversely proportional to the graphitization degree of materials, indicating a higher electronic conductivity.³⁹

The detailed morphology and microstructure of pristine Ni₁₂P₅@C nanocrystals and prepared Ni₁₂P₅@C/GNS are further characterized by TEM. The low-resolution TEM images show that the pure Ni₁₂P₅@C nanocrystals are aggregated with an average diameter of ∼19 nm (Fig. 2a and b and S6a†). The carbon layer thickness is ∼3.14 nm. The high-resolution TEM (HRTEM) image displays the characteristic lattice fringe spaces of 0.185 and 0.301 nm, which are consistent with the (312) and (220) planes of Ni₁₂P₅ nanocrystals, respectively, demonstrating a high degree of crystallinity (Fig. 2c). The low-magnification TEM images show the hollow structure of the carbon-coated product and the particle size distribution histogram displays the ∼35 nm size of Ni₁₂P₅@C/GNS (Fig. 2d, e and S6b†). A typical HRTEM image (Fig. 2f) reveals that the Ni₁₂P₅@C anchored on the GNS surface is fully hollow and the carbon shell is clear. The carbon layer thickness is ∼2.64 nm and the thickness of Ni₁₂P₅ is ∼6.34 nm. The characteristic lattice fringe spacings of 0.193, 0.235 and 0.217 nm are consistent with the (312), (330) and (321) planes of Ni₁₂P₅@C nanocrystals, respectively. Meanwhile, the angle between the (312) and (330) lattice planes is 53.6° consistent with the calculation value. Additionally, the crystal type of both pure Ni₁₂P₅@C and Ni₁₂P₅@C/GNS is polycrystalline (Fig. S7†). Notably, GNS effectively suppress the aggregation of Ni₁₂P₅@C and adjust the growth direction of crystals to construct a hollow architecture. The formation mechanism of the hollow structure can be ascribed to the Kirkendall effect. It's known that GNS are negatively charged. When dispersed in a suspension, the surface functional groups of GNS strongly attract Ni²⁺, which leads to the faster out-diffusion tendency of inner Ni²⁺ and formation of hollows.^41–43 The carbon shell is derived from the carbonation of OLA. Moreover, the elemental mapping analysis (Fig. 2g) further clearly confirms the uniform distribution of C, Ni and P elements in the Ni₁₂P₅@C/GNS.

3.2 Electrochemical performances of LIBs

The electrochemical Li-storage performances of hollow Ni₁₂P₅@C/GNS are given in Fig. 3. As shown in Fig. 3a, cyclic voltammetry (CV) is adopted to verify the reversible Li⁺ storage behavior of hollow Ni₁₂P₅@C/GNS. The CV curves display weak redox peaks and a box-like shape. This is due to the sorption of lithium ions on the pores and functional groups of the GNS surface, as well as the sorption on the edge of defective sites. Thus, the electrochemical behaviors are surface capacitive lithium storage behaviors.^44–47 In the first cycle, the reduction peaks are located at around 1.5 and 0.7 V, which represent the irreversible structural change and lithiation and formation of a solid electrolyte interphase (SEI), respectively. Besides, the oxidation peaks are observed at around 1.0, 1.5 and 2.4 V, which can be primarily attributed to the gradual reduction of Li₃P. Subsequently, the curves remain stable in the following cycles implying its high reversibility. According to this mechanism, the reaction process could be summed up as

Ni12P5 + xLi+ + xe− ↔ mLi3P + nNi	(1)

Fig. 3b presents the representative discharge–charge curves of Ni₁₂P₅@C/GNS at a current density of 100 mA g⁻¹, which are consistent with CVs. The initial discharge and charge specific capacities of Ni₁₂P₅@C/GNS could reach 1543.1 and 702.1 mA h g⁻¹, respectively. This indicates first cycle irreversible losses of about 54.5%, resulting from the irreversible formation of the SEI and structural change. However, the coulombic efficiency rapidly increases to 86.9% in the second cycle with a discharge capacity of 756.6 mA h g⁻¹ and a charge capacity of 657.5 mA h g⁻¹. Remarkably, the discharge capacity remains at about 668.4 mA h g⁻¹ after 20 cycles at a current rate of 100 mA g⁻¹ and its coulombic efficiency remains consistently at 97.14%. To further confirm the structural change, ex situ XRD, XPS and selected area electron diffraction (SAED) are employed. Eight different key potentials (Fig. 3c) are selected in the initial cycle to analyze the phase change. Ex situ XRD (Fig. 3d) shows two obvious transformations. Firstly, the peaks of Ni₁₂P₅ gradually disappear and the peaks of Li₃P and Ni appear following the lithiation process. In contrast, the peaks of Ni₁₂P₅ arise while the peaks of Li₃P and Ni fade away during delithiation. This indicates that the lithiation/delithiation of Ni₁₂P₅@C/GNS is a conversion reaction. During the discharge process, the Li⁺ insets the structure of Ni₁₂P₅, then forms Li₃P and Ni. Subsequently, the Li₃P and Ni reversibly generate Ni₁₂P₅ in the charge process. In addition, the XPS spectra reveal the valence change of Ni, as shown in Fig. 3e. The bond energy of Ni 2p_2/3 decreases at 0.005 V and then increases at 3.00 V, corresponding to the descent and then rise of Ni valence. Furthermore, the SAED patterns (Fig. 3f and g) also verify the presence of Ni and Li₃P phases in the discharged state and the Ni₁₂P₅ phase in the charged state, which agree well with XRD analysis.

When considering the performance of a whole electrode, the proportion between Ni₁₂P₅@C and GNS can seriously influence the performances. Broadly speaking, with the increase of GNS, the electroconductivity of Ni₁₂P₅@C/GNS gradually increases. However, the capacity will reach a saturation point and not increase any more with the content of GNS increasing. Therefore, in order to explore the best ratio between the active material Ni₁₂P₅@C and GNS, we further prepare four Ni₁₂P₅@C/GNS composites with different carbon contents (7.68, 11.13, 18.15 and 34.19 wt%, respectively). The carbon contents are characterized by EA. Cycling performances and rate capacities are displayed in Fig. 4. It could be observed that the electrochemical performances of Ni₁₂P₅@C/GNS electrodes with carbon content above 18.15 wt% are superior to those with 11.13 and 7.68 wt% carbon. The cycling performances of the Ni₁₂P₅@C/GNS with 7.68, 11.13, 18.15 and 34.19 wt% carbon at 100 mA g⁻¹ are 517.7, 564.1, 900 and 923 mA h g⁻¹ after 100 cycles, respectively (Fig. 4a). Fig. 4b shows the rate capacities of Ni₁₂P₅@C/GNS with different carbon contents. As expected, with the GNS content increasing, the capacities of Ni₁₂P₅@C/GNS tend to improve, because more GNS offer more ion and electron transport paths to achieve good performances. However, the capacity of Ni₁₂P₅@C/GNS no longer linearly grows with the further increase of the GNS amount. Considering that the preparation of GNS is costly and time-consuming, the Ni₁₂P₅@C/GNS of 18.15 wt% carbon is optimum.

The electrochemical lithium storage performances of Ni₁₂P₅@C/GNS are given in Fig. 5. Fig. 5a displays the cycling performances of the pure Ni₁₂P₅@C and Ni₁₂P₅@C/GNS at a current density of 100 mA g⁻¹. The discharge specific capacity of Ni₁₂P₅@C/GNS remains at 900 mA h g⁻¹ after 100 cycles, which is much better than that of Ni₁₂P₅@C (420 mA h g⁻¹). The TEM images of the Ni₁₂P₅@C/GNS composite after 100 cycles are shown in Fig. S8,† which show that Ni₁₂P₅@C/GNS can maintain its original appearance and no significant aggregation is observed. The batteries show a gradual gain in capacity. According to previous reports, the major reasons can be summarized as follows: (1) The slowly increased capacities are driven by the activation effects. (2) The stabilization process of the SEI layer will result in gradually increasing capacities. (3) Decreasing polarization and resistances are directly connected to the increase of capacities.^48,49 Then rate capabilities are examined by discharging/charging at various current densities (Fig. 5b). Ni₁₂P₅@C/GNS delivers discharge specific capacities of 702.2, 644.1, 566.5, 482.8, 423.2 and 905.9 mA h g⁻¹ at 100, 200, 500, 1000, 2000 and 100 mA g⁻¹, respectively, which are much higher than those of pure Ni₁₂P₅@C. Moreover, Ni₁₂P₅@C/GNS can be cycled with high stability at a higher current density of 500 mA g⁻¹ (Fig. S9†) with discharge capacity maintaining 308.9 mA h g⁻¹ after 250 cycles and at 2 A g⁻¹ (Fig. 5c) with the discharge capacity reaching up to 237.3 mA h g⁻¹ after 200 cycles. This firmly verifies its excellent cycling stability and long service life. Table 1 shows that the electrochemical performances of the materials prepared in the present work for LIB anodes are higher than those of most of the related materials reported in seven years.^50–59

---

Electrochemical impedance spectroscopy (Fig. 6a) demonstrates that Ni₁₂P₅@C/GNS shows a lower charge transfer resistance (279.3 Ω, fitted by Zview) than pure Ni₁₂P₅@C (1021.0 Ω).The ion diffusion capabilities of the electrodes are estimated according to the slope of the lines (σ, the value of which represents the Warburg factor, Fig. 6b) between Z′ and ω^−1/2. Meanwhile, the value of σ² is in inverse ratio to the diffusion coefficient of Na⁺ ions (D_Na).⁶⁰ It can be seen that the slope (σ) of Ni₁₂P₅@C/GNS is smaller than pure Ni₁₂P₅@C. This indicates the D_Na of Ni₁₂P₅@C/GNS is higher than that of pure Ni₁₂P₅@C. These results declare the Ni₁₂P₅@C/GNS electrode exhibits superior electronic/ionic conductivity, which could be reasonably attributed to two merits (Fig. 6c). (1) The hollow nanocrystal structure shortens the ion transport path and provides ample room to accommodate the volume expansion, so that the cycling performance is improved drastically. (2) The carbon shell and adding of GNS effectively prevent the agglomeration between nanocrystals and improve the electronic conductivity of Ni₁₂P₅@C accelerating the transmission of electrons, which greatly elevate the rate capacities. Thus, the designed Ni₁₂P₅@C/GNS exhibits excellent cycling performance and outstanding rate capability, indicating its overwhelming superiority in applications.

We also further test the full battery cycling performances with the prepared Ni₁₂P₅@C/GNS as the anode and commercial LiCoO₂ as the cathode. Fig. 7a displays the galvanostatic charge/discharge curves of the full LIBs at 100 mA g⁻¹. The reversible specific capacity (based on the anode material) can reach 350 mA h g⁻¹, and the average working voltage is about 2.3 V. What's more, the curves overlap well in the course of 50 to 100 cycles, which reveals the high reversibility and stability of the LiCoO₂//Ni₁₂P₅@C/GNS full cell. Fig. 7b shows that a stable capacity of 315.1 mA h g⁻¹ can be achieved at a current density of 100 mA g⁻¹ after 50 cycles. To demonstrate its potential application, the full cell is used to light up red LEDs.

3.3 Electrochemical performances of SIBs

As mentioned before, the Ni₁₂P₅@C/GNS electrode will also show promising electrochemical performances for sodium-ion storage in half cells. Fig. 8a shows that the charge and discharge capacities are 212.7 and 234.9 mA h g⁻¹ of the 5th cycle at 100 mA g⁻¹, respectively. Remarkably, the Ni₁₂P₅@C/GNS electrode shows a stable cycling performance (Fig. 8b). After 500 cycles, the discharge capacity remains at 164.8 mA h g⁻¹. Fig. 8c exhibits the rate performances of the Ni₁₂P₅@C/GNS electrode. The reversible capacities are 171.6, 162.4, 144.4, 126.8, 110 and 105.6 mA h g⁻¹ for current densities of 50, 100, 200, 500, 1000 and 2000 mA g⁻¹, respectively. More importantly, the capacity can be restored to 199.2 mA h g⁻¹ when the current density returns to 100 mA g⁻¹, implying its high current adaptation. Na⁺ (1.02 Å) is heavier and larger than Li⁺ (0.76 Å), leading to the inferior reaction dynamics of Na⁺. This significantly influences the electrochemical performances of the Ni₁₂P₅@C/GNS electrode in terms of Na⁺ storage resulting in lower capacity and poor stability. Therefore, further studies should be carried out to improve its sodium storage properties with some novel strategies.

4. Conclusions

In summary, a hollow Ni₁₂P₅@C/GNS composite is first prepared by a simple refluxing method. The Ni₁₂P₅@C nanocrystals with a particularly hollow structure and thin carbon shell are anchored on the GNS. This architecture combines the advantages of the hollow structure and high electron conductive GNS microstructure revealing excellent electrochemical performances in LIBs and SIBs in terms of cycle stability and rate capability. As an anode for LIBs, it displays an excellent specific capacity of 900 mA h g⁻¹ at a current density of 100 mA g⁻¹ after 100 cycles, and a high rate capability with specific capacities of 702.2, 644.1, 566.5, 482.8, 423.2 and 905.9 mA h g⁻¹ at 100, 200, 500, 1000, 2000 and 100 mA g⁻¹, respectively. As for Na-ion storage, this hollow Ni₁₂P₅@C/GNS composite also maintains a reversible capacity of approximately 235 mA h g⁻¹ at 100 mA g⁻¹. These promising features make hollow Ni₁₂P₅@C/GNS an ideal candidate for the next generation of battery materials which can be extended to electrochemical applications in various fields, such as fuel cells, supercapacitors, electrochemical sensors, and so forth.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was financially supported by the MOST (2016YFA0202500), NSFC (51471089 and 51501072), MOE (IRT13R30), NSFT (17JCYBJC17900), and 111 Project (B12015).

Notes and references

1. L. Shi and T. S. Zhao, J. Mater. Chem. A, 2017, 5, 3735 CAS .
2. A. Devaraj, M. Gu, R. Colby, P. Yan, C. M. Wang, J. M. Zheng, J. Xiao, A. Genc, J. G. Zhang, I. Belharouak, D. Wang, K. Amine and S. Thevuthasan, Nat. Commun., 2015, 6, 8014 CrossRef CAS PubMed .
3. W. H. Zuo, W. H. Zhu, D. F. Zhao, Y. F. Sun, Y. Y. Li, J. P. Liu and X. W. Lou, Energy Environ. Sci., 2016, 9, 2881 CAS .
4. G. Q. Xu, P. Liu, Y. R. Ren, X. B. Huang, Z. G. Peng, Y. G. Tang and H. Y. Wang, J. Power Sources, 2017, 361, 1 CrossRef CAS .
5. X. X. Zuo, J. Zhu, P. M. Buschbaum and Y. J. Cheng, Nano Energy, 2017, 31, 113 CrossRef CAS .
6. Y. Zhao, Y. Zhang, H. Sun, X. L. Dong, J. Y. Cao, L. Wang, Y. F. Xu, J. Ren, Y. HWang, I. H. Son, X. L. Huang, Y. G. Wang and H. S. Peng, Angew. Chem., Int. Ed., 2016, 55, 14384 CrossRef CAS PubMed .
7. C. C. Chen, J. B. Wang, Q. Zhao, Y. J. Wang and J. Chen, ACS Energy Lett., 2016, 1, 1165 CrossRef CAS .
8. D. Sun, Y. G. Tang, D. L. Ye, J. Yan, H. S. Zhou and H. Y. Wang, ACS Appl. Mater. Interfaces, 2017, 9, 5254 CAS .
9. F. Mueller, N. Loeffler, G. T. Kim, T. Diemant, R. J. Behm and S. Passerini, ChemSusChem, 2016, 9, 1290 CrossRef CAS PubMed .
10. H. N. He, D. Sun, Q. Zhang, F. Fu, Y. G. Tang, J. Guo, M. H. Shao and H. Y. Wang, ACS Appl. Mater. Interfaces, 2017, 9, 6093 CAS .
11. Y. Liu, J. Y. Liu, M. Y. Hou, L. Fan, Y. G. Wang and Y. Y. Xia, J. Mater. Chem. A, 2017, 5, 10902 CAS .
12. H. C. Gao, W. D. Zhou, K. Park and J. B. Goodenough, Adv. Energy Mater., 2016, 6, 1600467 CrossRef .
13. H. N. He, H. Y. Wang, D. Sun, M. H. Shao, X. B. Huang and Y. G. Tang, Electrochim. Acta, 2017, 236, 43 CrossRef CAS .
14. H. N. He, Q. Zhang, H. Y. Wang, H. H. Zhang, J. D. Li, Z. G. Peng, Y. G. Tang and M. H. Shao, J. Power Sources, 2017, 354, 179 CrossRef CAS .
15. J. Y. Li, X. L. Wu, X. H. Zhang, H. Y. Lü, G. Wang, J. Z. Guo, F. Wan and R. S. Wang, Chem. Commun., 2015, 51, 14848 RSC .
16. X. P. Han, F. Y. Cheng, C. C. Chen, F. J. Li and J. Chen, Inorg. Chem. Front., 2016, 3, 866–871 RSC .
17. F. Niu, J. Yang, N. N. Wang, D. P. Zhang, W. L. Fan, J. Yang and Y. T. Qian, Adv. Funct. Mater., 2017, 27, 1700522 CrossRef .
18. J. Liu, P. Kopold, C. Wu, P. A. V. Aken, J. Maier and Y. Yu, Energy Environ. Sci., 2015, 8, 3531 CAS .
19. C. C. Chen, Y. Y. Dong, S. Y. Li, Z. H. Jiang, Y. J. Wang, L. F. Jiao and H. T. Yuan, J. Power Sources, 2016, 320, 20 CrossRef CAS .
20. J. Peters, D. Buchholz, S. Passerini and M. Weil, Energy Environ. Sci., 2016, 9, 1744 CAS .
21. Y. W. Tan, H. Wang, P. Liu, C. Cheng, F. Zhu, A. Hirata and M. W. Chen, Adv. Mater., 2016, 28, 2951–2955 CrossRef CAS PubMed .
22. Y. Feng, X. Y. Yu and U. Paik, Chem. Commun., 2016, 52, 1633–1636 RSC .
23. M. Sun, H. J. Liu, J. H. Qu and J. H. Li, Adv. Energy Mater., 2016, 6, 1600087 CrossRef .
24. J. W. Zhou, J. Qin, L. C. Guo, N. Q. Zhao, C. S. Shi, E. Z. Liu, F. He, L. Y. Ma, J. J. Li and C. N. He, J. Mater. Chem. A, 2016, 4, 17370–17380 CAS .
25. G. L. Guo, Y. Y. Guo, H. T. Tan, H. Yu, W. H. Chen, E. Fong and Q. Y. Yan, J. Mater. Chem. A, 2016, 4, 10893–10899 CAS .
26. E. J. Sheets, R. B. Balow, W. C. Yang, E. A. Stach and R. Agrawal, Nanoscale, 2015, 7, 19317–19323 RSC .
27. F. P. Zhao, N. Han, W. J. Huang, J. J. Li, H. L. Ye, F. J. Chen and Y. G. Li, J. Mater. Chem. A, 2015, 3, 21754–21759 CAS .
28. J. F. Qian, Y. Xiong, Y. L. Cao, X. P. Ai and H. X. Yang, Nano Lett., 2014, 14, 1865–1869 CrossRef CAS PubMed .
29. Q. S. Xie, D. Q. Zeng, P. Y. Gong, J. Huang, Y. T. Ma, L. S. Wang and D. L. Peng, Electrochim. Acta, 2017, 232, 465–473 CrossRef CAS .
30. Y. H. Cui, M. Z. Xue, Z. W. Fu, X. L. Wang and X. J. Liu, J. Alloys Compd., 2013, 555, 283–290 CrossRef CAS .
31. X. S. Zhou, L. Yu and X. W. Lou, Adv. Energy Mater., 2016, 6, 1600451 CrossRef .
32. Q. L. Wei, Y. N. Xu, Q. D. Li, S. S. Tan, W. H. Ren, Q. Y. An and L. Q. Mai, Chem. Commun., 2016, 52, 8730–8732 RSC .
33. Y. Y. Lu, P. F. Zhou, K. X. Lei, Q. Zhao, Z. L. Tao and J. Chen, Adv. Energy Mater., 2017, 7, 1601973 CrossRef .
34. Y. Kim, Y. Kim, A. Choi, S. Woo, D. Mok, N. S. Choi, Y. S. Jung, J. H. Ryu, S. M. Oh and K. T. Lee, Adv. Mater., 2014, 26, 4139–4144 CrossRef CAS PubMed .
35. A. Ueda, M. Nagao, A. Inoue, A. Hayashi, Y. Seino, T. Ota and M. Tatsumisago, J. Power Sources, 2013, 244, 597–600 CrossRef CAS .
36. Z. Q. Li, L. Y. Zhang, X. L. Ge, C. X. Li, S. H. Dong, C. X. Wang and L. W. Yin, Nano Energy, 2017, 32, 495–502 CrossRef .
37. C. Y. Peng, L. Kang, S. Cao, Y. Chen, Z. S. Lin and W. F. Fu, Angew. Chem., Int. Ed., 2015, 54, 15725–15729 CrossRef CAS PubMed .
38. A. Alazmi, O. E. Tall, S. Rasul, M. N. Hedhili, S. P. Patole and P. M. J. Costa, Nanoscale, 2016, 8, 17783–17787 RSC .
39. Y. J. Fang, L. F. Xiao, X. P. Ai and H. X. Yang, Adv. Mater., 2015, 27, 5895–5900 CrossRef CAS PubMed .
40. E. H. Wang, W. Xiang, R. Rajagopalan, Z. G. Wu, J. Yang, M. Z. Chen, B. H. Zhong, S. X. Dou, S. L. Chou, X. D. Guo and Y. M. Kang, J. Mater. Chem. A, 2017, 5, 9833 CAS .
41. G. D. Park, J. S. Cho and Y. C. Kang, Nanoscale, 2015, 7, 16781–16788 RSC .
42. J. W. Wang, A. C. J. Peck and J. B. Tracy, Chem. Mater., 2009, 21, 4462–4467 CrossRef CAS .
43. M. Lin, B. L. Chen, X. Wu, J. S. Qian, L. F. Fei, W. Lu, L. W. H. Chan and J. K. Yuan, Nanoscale, 2016, 8, 1856–1860 Search PubMed .
44. C. J. Chen, Y. W. Wen, X. L. Hu, X. L. Ji, M. Y. Yan, L. Q. Mai, P. Hu, B. Shan and Y. H. Huang, Nat. Commun., 2015, 6, 6929 CrossRef CAS PubMed .
45. C. J. Chen, Z. G. Wang, B. Zhang, L. Miao, J. Cai, L. F. Peng, Y. Y. Huang, J. J. Jiang, Y. H. Huang, L. N. Zhang and J. Xie, Energy Storage Materials, 2017, 8, 161–168 CrossRef .
46. C. J. Chen, H. H. Xu, T. F. Zhou, Z. P. Guo, L. N. Chen, M. Y. Yan, L. Q. Mai, P. Hu, S. J. Cheng, Y. H. Huang and J. Xie, Adv. Energy Mater., 2016, 6, 1600322 CrossRef .
47. M. J. Lu, W. H. Yu, J. Shi, W. Liu, S. G. Chen, X. Wang and H. L. Wang, Electrochim. Acta, 2017, 251, 396–406 CrossRef CAS .
48. L. F. Zhang, W. J. He, M. Ling, K. C. Shen, Y. Liu and S. W. Guo, Electrochim. Acta, 2017, 252, 322–330 CrossRef CAS .
49. C. L. Sun, M. Z. Yang, T. L. Wang, Y. L. Shao, Y. Z. Wu and X. P. Hao, ACS Appl. Mater. Interfaces, 2017, 9, 26631–26636 CAS .
50. J. H. Hu, P. Wang, P. P. Liu, G. Q. Cao, Q. Wang, M. Wei, J. Mao, C. H. Liang and G. S. Shao, Electrochim. Acta, 2016, 220, 258–266 CrossRef CAS .
51. Q. Li, J. J. Ma, H. J. Wang, X. Yang, R. Yuan and Y. Q. Chai, Electrochim. Acta, 2016, 213, 201–206 CrossRef CAS .
52. Y. Lu, J. P. Tu, J. Y. Xiang, X. L. Wang, J. Zhang, Y. J. Mai and S. X. Mao, J. Phys. Chem. C, 2011, 115, 23760–23767 CAS .
53. Y. J. Bai, H. J. Zhang, X. Li, L. Liu, H. T. Xu, H. J. Qiu and Y. Wang, Nanoscale, 2015, 7, 1446–1453 RSC .
54. C. Wu, P. Kopold, P. A. V. Aken, J. Maier and Y. Yu, Adv. Mater., 2017, 29, 1604015 CrossRef PubMed .
55. J. Jiang, C. D. Wang, W. Li and Q. Yang, J. Mater. Chem. A, 2015, 3, 23345–23351 CAS .
56. Y. Lu, J. P. Tu, Q. Q. Xiong, J. Y. Xiang, Y. J. Mai, J. Zhang, Y. Q. Qiao, X. L. Wang, C. D. Gu and S. X. Mao, Adv. Funct. Mater., 2012, 22, 3927–3935 CrossRef CAS .
57. G. H. Li, H. Yang, F. C. Li, J. Du, W. Shi and P. Cheng, J. Mater. Chem. A, 2016, 4, 9593–9599 CAS .
58. C. D. Wang, T. Ding, Y. Sun, X. L. Zhou, Y. Liu and Q. Yang, Nanoscale, 2015, 7, 19241–19249 RSC .
59. H. J. Zhang, Y. Y. Feng, Y. Zhang, L. Fang, W. X. Li, Q. Liu, K. Wu and Y. Wang, ChemSusChem, 2014, 7, 2000–2006 CrossRef CAS PubMed .
60. Y. C. Liu, N. Zhang, H. Y. Kang, M. H. Shang, L. F. Jiao and J. Chen, Chem.–Eur. J., 2015, 21, 11882 Search PubMed .

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7ta06843c

This journal is © The Royal Society of Chemistry 2017