Three-dimensional nitrogen-doped holey graphene and transition metal oxide composites for sodium-ion batteries

Abstract

Transition metal oxides (TMOs) such as nickel cobaltite (NCO) and magnetite (Fe₃O₄) with rich electroactive sites are promising anode materials for sodium-ion batteries (NIBs). However, these materials suffer from large volume change during charge/discharge, poor electron conductivity and severe aggregation of TMO nanoparticles. Here, we report an approach to improve the electrochemical performance of NCO and Fe₃O₄ by stabilizing them on three-dimensional (3D) nitrogen-doped holey graphene (N-HG), forming NCO@N-HG and Fe₃O₄@N-HG composite materials, respectively. The thin graphene sheets in both composites facilitate the electron transport and buffer the volume changes, while the interconnected 3D macroporous network with a pore size in the range of several micrometers, combined with the nanopores in the N-HG provide pathways for rapid ion transport. As expected, both composites showed high specific capacities, rate capability and cycling performance in NIBs. The good electrochemical performance of the electrode material indicates that using N-HG to support NCO and Fe₃O₄ particles is an effective approach towards developing high-performance anode materials for NIBs.

---

1. Introduction

Sodium-ion batteries (NIBs) are gaining fast growing attention as an alternative to lithium-ion batteries (LIBs) because of the rich abundance and low cost of sodium sources.^1,2 One of the current grand challenges of the NIB technology is to find suitable anode materials with both high reversible capacities and long cycling life.^3,4 Transition metal oxides (TMOs) such as TiO₂,^5,6 Nb₂O₅,^7,8 NiCo₂O₄ (ref. 9–11) and Fe₃O₄,^12–14 have been demonstrated to be promising anode materials for NIBs. However, the TMOs suffer from large volume change and sluggish kinetics of both ion and electron transport, as well as severe aggregation for TMOs nanoparticles (NPs), resulting in poor rate performance and cycling stability.^15,16 One strategy for solving these problems is to combine TMOs with carbon materials, such as graphene, to prepare composite electrodes.^9,12,17,18 While graphene is an excellent carbon material, graphene sheets tend to stack or agglomerate during electrochemical reactions in NIBs, leading to blockage of some ion transport pathways.^19,20

Using three-dimensional (3D) porous graphene aerogels with 3D ion diffusion channels to support TMOs has been shown to be a more effective approach to fabricating TMO–graphene composite electrode materials.^{12,18,21–23} However, ion transport across the basal plane of individual two-dimensional (2D) graphene sheets has been a great challenge.²⁴ Introducing nanoholes in graphene sheets is regarded as a good approach to solve this problem. Zhao et al.²⁵ reported that graphene can be etched by nitric acid to obtain holy graphene (HG). Jiang et al.²⁶ reported that potassium hydroxide can also be used to create holes for preparing HG. The HG in the above two work exhibited a significantly improved rate capability with an excellent cycling stability as anode for lithium ion batteries. Later, Xu et al.²⁷ reported that graphene can be etched by using a more environmentally friendly agent, namely hydrogen peroxide (H₂O₂) to obtain HG, which showed efficient capacitive energy storage. Recently, Sun and co-workers²⁴ prepared a HG by etching graphene in the presence of H₂O₂, which was then used as a conductive scaffold for Nb₂O₅ to form an 3D architecture electrode. The HG greatly improved ion access to the Nb₂O₅, leading to an ultrahigh-rate lithium ion storage with a capacity of 75 mA h g⁻¹ at 100C. On the other hand, nitrogen (N) doping is an effective approach to further improving the electron conductivity and wettability of carbon materials.^21,28 Therefore, it can be expected that constructing a 3D porous architecture consisting of nitrogen-doped HG (N-HG) and the TMOs could significantly facilitate electron and sodium ion transport throughout the entire architecture.

It has been reported that nickel cobaltite (NCO)^9–11,29 and magnetite (Fe₃O₄)^12–14,30 are typical TMOs with rich electroactive sites are promising electrode materials for electrochemical energy storage due to their low cost, abundance in nature, and high theoretical specific capacities in NIBs (about 890 for NCO and 925 mA h g⁻¹ for Fe₃O₄). However, their severe volume change during electrochemical cycling, inferior electric and ionic conductivity, and severe aggregation of NPs are critical issues that must be solved before such TMOs can find practical applications. In this work, N-HG was used to modify NCO and Fe₃O₄ nanoparticles by forming 3D porous N-HG supported NCO and Fe₃O₄ composites, which were designated as NCO@N-HG and Fe₃O₄@N-HG, respectively. The thin graphene sheets in both composites facilitate the electron transport and buffer the volume changes. In addition, the interconnected 3D macroporous network with a pore size in several micrometers range, combined with the nanoporous channels in the N-HG provide pathways for rapid ion transport. As expected, the composites exhibited enhanced rate capabilities with specific capacities of 403 and 350 mA h g⁻¹ for NCO@N-HG and Fe₃O₄@N-HG at a current rate of 1 A g⁻¹, respectively, and a stable cycling performance when used as anodes in NIBs.

2. Experimental section

Materials preparation

Synthesis of graphene oxide (GO). The GO used in this work was synthesized from natural graphite using the modified Hummers method as described elsewhere.³¹

Synthesis of holey graphene oxide (HGO). The HGO was prepared by etching GO with H₂O₂ using a previously reported method.^24,32 Briefly, 40 mL of 2 mg mL⁻¹ GO aqueous dispersion was mixed with 4 mL of 30% H₂O₂ aqueous solution under stirring. The mixture was heated at 100 °C for 2 h still under stirring. The as-prepared HGO was purified by washing and centrifuging to remove the residual H₂O₂ and then re-dispersed in 30 mL deionized (D.I.) water to obtain a HGO dispersion.

Preparation of graphene-supported TMOs composites. A graphene-supported NCO composite (NCO@G) was prepared using the hydrothermal method. In brief, 4 mmol of Co(NO₃)₂·6H₂O, 2 mmol of NiSO₄·6H₂O and 0.3 g hexamethylenetetramine (HMTA) were dissolved in 10 mL D.I. water under stirring. Then, 20 mL of GO aqueous dispersion (2 mg mL⁻¹) was added under stirring. After 10 min, the mixture was transferred to a teflon-lined stainless steel autoclave and heated at 180 °C for 12 h. The solid product was filtered off and washed with D.I. water, freeze dried, and thermally treated at 300 °C in air for 2 h at a heating rate of 1 °C min⁻¹ to obtain NCO@G sample.

A graphene-supported Fe₃O₄ composite (Fe₃O₄@G) was prepared according to a method reported previously.³³ Briefly, 10 mL of D.I. water containing 2.5 mmol of FeSO₄·7H₂O was added into 15 mL of 2 mg mL⁻¹ GO dispersion. Then, 2.5 mL of D.I. water containing 0.20 g of NaOH was added under stirring. After 1 h, 10 mL of N₂H₄·H₂O was added. The mixture was sealed in a teflon-lined stainless steel autoclave and heated at 180 °C for 8 h. The solid product was filtered off and washed with D.I. water, and freeze dried to obtain Fe₃O₄@G sample.

Preparation of 3D N-HG and TMO composites. An NCO@N-HG composite was prepared by co-assembly of the aforementioned HGO and NCO@G. In brief, 15 mg of NCO@G was dispersed in a mixture of 100 mg urea and 2.5 mL of aqueous HGO suspension under ultrasonication. The mixture was then transferred into a teflon-lined stainless steel autoclave and hydrothermally heated at 120 °C for 4 h. The solid product was filtrated off, washed with D.I. water, freeze-dried, and thermally treated at 350 °C for 2 h in nitrogen atmosphere to obtain a composite sample of NCO@N-HG. Another sample, Fe₃O₄@N-HG was similarly prepared by hydrothermal treatment of Fe₃O₄@G and HGO.

Preparation of other samples. NCO and Fe₃O₄ NPs were synthesized following the same preparation procedure of NCO@G and Fe₃O₄@G composites without adding GO. N-HG was prepared by hydrothermally reducing 2.5 mL HGO suspension in the presence of 100 mg urea at 120 °C for 4 h, followed by thermal treatment at 350 °C for 2 h in nitrogen atmosphere.

Characterization

X-ray diffraction (XRD) patterns were collected on a Bruker D8 X-ray diffractometer with Ni-filtered Cu Kα radiation (λ = 1.54056 Å, 40 kV, 30 mA) at a scan rate of 2° min⁻¹. Field-emission scanning electron microscope (FESEM, JEOL 7100F) and transmission electron microscope (TEM, TECNAI F20) were used to characterise the morphology of samples. X-ray photoelectron spectroscopy (XPS) was acquired on a Kratos Axis ULTRA X-ray photoelectron spectrometer with a 165 mm hemispherical electron energy analyser and a monochromatic Al Kα (1486.6 eV) radiation at 225 W (15 kV, 15 mA). Argon sorption isotherms were measured on Tristar II 3020 at the liquid argon temperature. Before measurement, the samples were degassed at 150 °C for 24 h. The pore size distribution curves were obtained by the density functional theory (DFT) method. Thermal gravimetric analysis (TGA) was conducted on a Perkin-Elmer STA 6000 in air from room temperature up to 700 °C at a heating rate of 5 °C min⁻¹.

Electrochemical measurements

All electrodes were prepared by mixing 70 wt% of an active material, 20 wt% of an electrical conductor (carbon black), and 10 wt% of a binder (carboxymethyl cellulose, CMC) in D.I. water to form a slurry, which was subsequently coated on a copper foil. After drying at 60 °C overnight in a vacuum oven, the copper foil was punched into circular discs. The mass loadings of the samples were all about 1 mg cm⁻². A NIB was assembled using a 2032-type coin cell with the discs as the anode, sodium foil as the counter electrode, glass fibre (Whatman, GF/C, USA) as the separator and 1 M NaPF₆ dissolved in diethylene glycol dimethyl ether solvent as the electrolyte. Both the electrolyte preparation and the coin cell fabrication were performed in an argon-filled glovebox. Galvanostatic charge and discharge (GCD) tests were performed on a Neware battery tester (CT3008). The specific capacities of the composites were calculated based on the weight of the entire composites. Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) measurements were conducted on an electrochemical workstation (CHI-600D).

3. Results and discussion

Scheme 1 illustrates the procedure for preparing of the 3D TMO@N-HG. HGO was firstly prepared by etching the carbon atoms around the defective sites of GO with H₂O₂.^24,34 TMO@G were prepared in a hydrothermal method. In this process, oxygen-containing groups (carboxyl, hydroxyl and epoxy groups) in GO served as the active sites for nucleation and growth of TMO nanoparticles (NPs). In the meantime, GO was reduced.³⁴ TMO@N-HG was then prepared by co-assembly of the aforementioned HGO and TMO@G. Briefly, the obtained TMO@G was crushed and dispersed in a urea dissolved aqueous HGO suspension for a hydrothermal treatment. During this treatment, the HGO linked with TMO@G via oxygen-containing groups to self-assemble into 3D hydrogels. Simultaneously, HGO was reduced and doped with nitrogen by urea. After freeze-drying and thermal treatment, a TMO@N-HG with macroporous channels derived from 3D assembly, as well as nanoporous channels derived from the etched nanoholes in graphene sheets formed. The details of sample preparation are provided in the experimental section.

Scheme 1 The overall synthetic procedure of TMO@N-HG: (1) preparation of HGO by etching GO sheets with H₂O₂; (2) preparation of TMO@G using a hydrothermal method; (3) dispersing TMO@G in a urea-dissolved HGO suspension for a hydrothermal treatment to obtain TMO@N-HG.

Fig. 1a shows the TEM image of N-HG, which was synthesized by etching GO with H₂O₂ followed by a hydrothermal reaction in the presence of urea. The N-HG sheets exhibited abundant in-plane nanoholes of diameters ranging from 1.0 to 3.5 nm. This confirmed that H₂O₂ can partially oxidize and etch GO, leaving behind nanoholes in GO to form HGO.²⁷ The HGO assembled with NCO@G (Fig. S1a†) to prepare 3D NCO@N-HG composite based on the synthetic procedure in Scheme 1. As expected, the composite exhibited various cumulative volumes of pores (Fig. S2†), and the pore size distribution curve indicates the presence of both micropores peaked at 1.1 nm and prominent mesopores peaked at 3.2 nm, which is consistent with the TEM study in Fig. 1a. Fig. 1b–g show the SEM and TEM images of sample NCO@N-HG. A 3D graphene framework with interconnected macrospores ranging from sub-micrometers to several micrometers can be seen from Fig. 1b and c. The TEM image and the high-resolution TEM (HRTEM) image (Fig. 1d and e) of NCO@N-HG show that many NCO NPs with sizes of about 5 nm were on the graphene sheets. The lattice fringes of NCO NP with a space of 0.245 nm agrees well with the d-space of the (311) plane of the NCO crystal.³⁵ Meanwhile, the selected area electron diffraction (SAED) pattern show diffraction rings (Fig. 1f) of (311), (400), (511), and (620) which were well indexed to the crystal planes of NCO.¹⁰ The high-angle annular dark-field scanning TEM (HAADF-STEM) image and elemental mapping results of NCO@N-HG (Fig. 1g) confirmed the existence of elements C, N, Ni, Co, and O, suggesting that N has been doped in the graphene sheets. The unique macro- and nanoporous feature of NCO@N-HG combined with nitrogen doping is favourable for electrolyte accessibility and rapid sodium ions transport.

Fig. 1 (a) TEM images of N-HG sheets with etched pores. (b, c) SEM images of NCO@N-HG at different magnifications. (d and e) TEM and HRTEM images, (f) SAED pattern, and (g) HAADF-STEM image and elemental mapping results of NCO@N-HG.

Fig. 2a shows the XRD pattern of sample NCO@N-HG. The broad diffraction peak at 36.7 degrees two theta confirmed that the size of the NCO crystals is in the nanometer range, which is consistent with the TEM data. The small diffraction hump appearing in the range of 22–28 degrees two theta is attributed to the re-stacked graphene sheets as confirmed by the XRD pattern of sample N-HG shown in Fig. S3,† which compares the XRD patterns of samples N-HG and GO. For sample N-HG, a broad peak in the two theta ranges of 20–30 degrees can be seen. However, the peak at 11.2 degrees two theta disappeared, indicating the GO sheets were reduced after the hydrothermal treatment, leading to sheet stacking.

Fig. 2 (a) XRD pattern of NCO@N-HG and the standard NCO JCPDS card (no. 73-1702). (b) XPS survey of NCO@N-HG. (c, d and e) High resolution XPS spectrum of Ni 2p, Co 2p and N 1s for NCO@N-HG, respectively. (f) High resolution XPS spectrum of C 1s for NCO@N-HG (bottom) and GO (top).

Fig. 2b shows the XPS survey scan of sample NCO@N-HG. The binding energies of 284.8, 399.8, 530.8, 779.8 and 854.8 eV corresponded to C 1s, N 1s, O 1s, Co 2p and Ni 2p, respectively, confirming the integration of NCO and N-HG. Fig. 2c and d present the high-resolution XPS spectra of Ni 2p and Co 2p of NCO@N-HG, respectively. The Ni 2p and Co 2p spectra were deconvoluted with two peaks assigned to Ni²⁺/Ni³⁺ and Co²⁺/Co³⁺, respectively, with two satellite peaks as well. This matched well with previously reported XPS spectra of Ni 2p and Co 2p in NCO.¹¹Fig. 2e shows the high-resolution XPS spectrum of N 1s, which can be resolved into three components centered at 398.2, 399.6 and 401.6 eV, representing the pyridinic, pyrrolic and graphitic types of N atoms, respectively. In addition, the XPS results revealed that the N-doping level was 6.0 at% in N-HG. Doping N in graphene not only enhances its electronic conductivity, but also improves the wettability between graphene and the organic electrolyte.²⁸Fig. 2f compares the C 1s high resolution XPS spectra of GO and NCO@N-HG; the dramatic loss of oxygen-containing functional groups (C–O/C–O–C and C=O) further confirmed the reduction of GO. Therefore, the XRD and XPS results confirmed the formation of NCO and N-HG composite, the reduction of GO and the successful N-doping in graphene.

Fig. 3a shows the CV curves of the NCO@N-HG electrode for the first four cycles in the potential range of 0.005 to 3.000 V (vs. Na⁺/Na) at a scan rate of 0.2 mV s⁻¹. For the first cathodic scan, the strong reduction peak observed at 0.76 V corresponds to the reduction of Ni²⁺/Ni³⁺ and Co²⁺/Co³⁺ to Ni⁰ and Co⁰, and the formation of a solid electrolyte interphase (SEI). For the subsequent anodic scan, two broad peaks appeared at 0.76 V and 1.73 V, which are attributed to the oxidation of Ni⁰/Co⁰ to Ni²⁺/Co²⁺, and Co²⁺ further to Co³⁺, respectively. The electrochemical conversion reaction of NCO in a NIB during charge/discharge can be described by the following equations:¹⁰

Fig. 3 (a) CV curves at a scan rate of 0.2 mV s⁻¹, (b) GCD profiles at various current densities from 0.1 to 5.0 A g⁻¹, and (c) cycling performance at 1 A g⁻¹ of NCO@N-HG. (d) Rate performance from current densities from 0.05 to 5.0 A g⁻¹, (e) cycling performance at 0.1 A g⁻¹ and (f) Nyquist plots of NCO@N-HG, NCO@G and NCO electrodes.

In the first discharge:

NiCo₂O₄ + 8Na⁺ + 8e → Ni + 2Co + 4Na₂O

In the following charge/discharge:

Ni + Na₂O ↔ NiO + 2Na⁺ + 2e

Fig. 3b shows the GCD profiles of NCO@N-HG at current densities ranging from 0.1 to 5.0 A g⁻¹. NCO@N-HG produced GCD profiles without a noticeable plateau at all current densities. This indicates that the reaction mechanism between sodium ions and NCO@N-HG not only involves in phases conversion but also has a large contribution of surface capacitive behaviour.³⁶ The NCO@N-HG electrode also displayed a long cycling performance, as shown in Fig. 3c. After activated at 0.1 A g⁻¹ for the first few cycles, NCO@N-HG remained the capacity of 348 mA h g⁻¹ at 1 A g⁻¹ at the 500^th cycle with a capacity retention of 85.7%.

An electrochemical study compared with NCO@G (Fig. S1a†) and NCO (Fig. S1b†) electrodes was performed in Fig. 3d–f. Fig. 3d shows the rate performance of samples NCO@N-HG, NCO@G and NCO. All electrodes exhibited continuously decrease in capacity in the first few cycles. The capacity loss can be attributed to the formation of solid electrolyte interphase (SEI) on the electrode surface. In the subsequent cycles, electrode NCO@N-HG delivered capacities of 639, 446, 442, 427, 403, 370, and 310 mA h g⁻¹ at current rates of 0.05, 0.1, 0.2, 0.5, 1, 2 and 5 A g⁻¹, respectively. When the current rate was reversed back to 0.1 A g⁻¹, the electrode capacity was about 443 mA h g⁻¹, very close to 446 mA h g⁻¹, indicating an excellent rate performance of this electrode. In contrast, the capacities of NCO and NCO@G electrodes dropped fast in the first 20 cycles, and the specific capacities in the subsequent cycles at different current rates were much lower than that of electrode NCO@N-HG.

Fig. 3e compares the cycling performance of the three electrodes measured at 0.1 A g⁻¹. Despite the decay in the initial few cycles, NCO@N-HG exhibited the best cycling performance with a capacity which was slowly increased to 510 mA h g⁻¹ at the 100^th cycle. This capacity-increasing trend with cycling was related to the electrode structure. With cycling going on, the electrolyte gradually wet the surfaces of small cavities formed between graphene sheets,³⁷ allowing the electrolyte to access to more and more electrochemically active sites for redox reactions with sodium ions. Thus, the capacity of electrode NCO@N-HG gradually increased in the first 100 cycles. In contrast, the capacities of NCO@G and NCO were rapidly decreased to 221 and 150 mA h g⁻¹ under the same measurement conditions, respectively. For comparison, the electrochemical performance of N-HG was also measured and the data was included in Fig. S4.† At a current rate of 0.1 A g⁻¹, the N-HG electrode contributed a reversible specific capacity of 228 mA hg⁻¹ at the 100^th cycle in NIBs, which was higher than that of previously reported nitrogen-doped graphene (205 mA h g⁻¹)³⁶ and graphene (146 mA h g⁻¹)³⁸ electrodes by our group. Although the content of graphene in NCO@N-HG was higher than that in NCO@G (31.9% vs. 12.9%) (see the TGA data in Fig. S5†), the capacity of the former was higher than the latter in all cycles. This is because the presence of N-HG promoted electrolyte diffusion, thus facilitating ion transport during charge/discharge.

Fig. 3f compares the Nyquist plots of electrodes, along with a fitted circuit. Both plots showed a single semicircle in the high-to-medium frequency region and the charge-transfer kinetics at the electrode/electrolyte interface (R_ct), as well as a slope in the low frequency region which stands for the diffusive Warburg impedance (W). R_c is the resistance of the electrolyte. According to the fitting results, NCO@N-HG showed a smaller resistance of R_ct (63 Ω) than that of NCO@G and NCO (128 and 227 Ω, respectively), further demonstrating that addition of N-HG indeed improved the electron transport kinetics of the electrode materials.

The superior rate and cycling performance of NCO@N-HG was attributed to the structure robustness of 3D N-HG foam, which could buffer the volume change of NCO during charge/discharge process. Meanwhile, the high electronic conductivity of the graphene enhanced the electron transport, and the unique pore structure consists of macrospores in the 3D network and nanopores in the graphene sheets facilitated the sodium ions transport.

The 3D Fe₃O₄@N-HG architecture was also constructed based on the synthetic procedure in Scheme 1. Fe₃O₄@N-HG (Fig. 4a and b) showed a similar morphology to that of NCO@N-HG, exhibiting the macroporous structure arising from the 3D graphene architecture. DFT analysis (Fig. S6a and b†) of Fe₃O₄@N-HG confirmed the presence of the nanopores peaking at about 1.0 and 3.3 nm which were mainly derived from the addition of holey graphene. The percentage by mass of graphene in Fe₃O₄@G and Fe₃O₄@N-HG composites were about 13.5 wt% and 34.3 wt%, respectively (Fig. S5†). TEM images (Fig. 4c and d) show that graphene sheets were anchored with nanoparticles with sizes about 20 to 35 nm. The lattice fringes of NP (Fig. 4d) with a space of 0.253 nm corresponds to the d-space of the (311) plane in Fe₃O₄, and the diffraction rings of (220), (311), (400), (422), (511) and (440) in the SAED patterns (Fig. 4e) further confirmed that the nanoparticles were Fe₃O₄ crystals. This is in consistent with the XRD patterns in Fig. S6c.† The HAADF-STEM image and elemental mapping results of Fe₃O₄@N-HG (Fig. 4f) confirmed the existence of elements C, N, Fe and O, suggesting that N has been doped in the graphene sheets.

Fig. 4 (a, b) SEM images, (c, d) TEM images, (e) SAED pattern and (f) HAADF-STEM image with elemental mapping results of the Fe₃O₄@N-HG.

Fig. 5a shows the CV curves of the Fe₃O₄@N-HG electrode for the first four cycles in the potential range of 0.005 to 3.000 V (vs. Na/Na⁺) at a scan rate of 0.2 mV s⁻¹. For the first cathodic scan, the strong reduction peak observed at 0.65 V corresponds to the reduction of Fe²⁺/Fe³⁺ to Fe⁰, and the formation of a SEI.^12,30 For the anodic scan, two broad peaks appeared at 0.78 V and 1.50 V, which are attributed to the oxidation of Fe⁰ to Fe²⁺, and further to Fe³⁺, respectively.¹⁸ In the subsequent cycles, the full sodiation potential is characterized by a high voltage at 0.75 V. This potential change from 0.65 to 0.75 V is mainly in virtue of the improved kinetics of Fe₃O₄@N-HG, originating from the inherent nanosize effects in the TMO electrode during cycling.³⁹

Fig. 5 (a) CV curves at a sweep rates of 0.2 mV s⁻¹ and (b) GCD profiles at various current densities from 0.1 to 5.0 A g⁻¹ of Fe₃O₄@N-HG. (c) Rate performance from current densities from 0.05 to 5.0 A g⁻¹ and (d) cycling performance of Fe₃O₄@N-HG, Fe₃O₄@G and Fe₃O₄ electrodes at 0.1 A g⁻¹.

The electrochemical conversion reaction of Fe₃O₄ in a NIB during charge/discharge can be described by the following equation:^12,30

Fe₃O₄ + 8Na⁺ + 8e ↔ 3Fe + 4Na₂O

Fig. 5b shows the GCD profiles of Fe₃O₄@N-HG at current densities ranging from 0.1 to 5.0 A g⁻¹. The slope charge–discharge curves reveal that the sodium storage in Fe₃O₄@N-HG involved in both the capacitive and diffusion-controlled process.⁴⁰

Fig. 5c compares the rate performance of the Fe₃O₄@N-HG, Fe₃O₄@G and Fe₃O₄ electrodes. Electrode Fe₃O₄@N-HG delivered capacities of about 448, 442, 425, 390, 350, 310, and 220 mA h g⁻¹ at current rates of 0.05, 0.1, 0.2, 0.5, 1, 2 and 5 A g⁻¹, respectively. When the current rate was reversed back to 0.1 A g⁻¹, the electrode capacity was about 444 mA h g⁻¹, very close to 442 mA h g⁻¹. By contrast, the specific capacities of Fe₃O₄@G and Fe₃O₄ at different current rates were much lower than those of Fe₃O₄@N-HG. For example, Fe₃O₄@G and Fe₃O₄ showed the low capacities of only 205 and 245 mA h g⁻¹ at 0.1 A g⁻¹, respectively. At a current rate of 2 A g⁻¹, low capacities of 155 and 130 mA h g⁻¹ were exhibited, respectively. The low capacities were mainly due to the aggregation of Fe₃O₄@G (Fig. S1c†) and Fe₃O₄ (Fig. S1d†) during electrode preparation, thus lowering the active sites accessible to the electrolyte. The cycling performance of the three samples is shown in Fig. 5d. The capacity of Fe₃O₄@N-HG was also superior to that of Fe₃O₄@G and Fe₃O₄ electrodes. The Fe₃O₄@N-HG sample kept a capacity-increasing trend with a capacity of 443 mA h g⁻¹ at the 100^th cycle at 0.1 A g⁻¹, which was higher than the reported Fe₃O₄ based materials for NIBs.^12,13,18,30 The enhanced electrochemical performance of Fe₃O₄@N-HG was mainly attributed to the bi-continuous pathways for both sodium ions and electrons constructed by the N-HG framework.

4. Conclusions

We have demonstrated an effective approach to preparing high-performance sodium-ion battery anode materials by loading nickel cobaltite and magnetite nanoparticles in three-dimensional nitrogen holey graphene framework. The thin graphene sheets buffer the volume change and the unique macroporous architecture in combination with the presence of nanopores in graphene sheets facilitate ion transport during charge/discharge, contributing to the observed high capacity and good cycling performance. Specifically, charge storage capacities of 510 mA h g⁻¹ and 443 mA h g⁻¹ were achieved with electrodes NCO@N-HG and Fe₃O₄@N-HG, respectively, at the 100^th cycle at 0.1 A g⁻¹ in NIBs, manifesting the NCO@N-HG and Fe₃O₄@N-HG composite materials as promising anodes for NIBs. While only NCO and Fe₃O₄ were investigated in this work, it is believed that the approach can be extended to modify other transition metal oxides for sodium ion storage, such as Co₃O₄, NiO, Fe₂O₃, and TiO₂.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This research was supported by The Australian Research Council (ARC) under the ARC Laureate Fellowship Program (FL170100101). DY wishes to thank the China Scholarship Council for providing a scholarship. The authors gratefully acknowledge the facilities and technical assistance of the Australian Microscopy and Microanalysis Research Facility at the UQ Centre for Microscopy and Microanalysis.

Notes and references

1. H. Pan, Y.-S. Hu and L. Chen, Energy Environ. Sci., 2013, 6, 2338–2360.
2. S.-W. Kim, D.-H. Seo, X. Ma, G. Ceder and K. Kang, Adv. Energy Mater., 2012, 2, 710–721.
3. H. Hou, X. Qiu, W. Wei, Y. Zhang and X. Ji, Adv. Energy Mater., 2017, 7, 1602898.
4. D. A. Stevens and J. R. Dahn, J. Electrochem. Soc., 2000, 147, 1271–1273.
5. M. Zhou, Y. Xu, C. Wang, Q. Li, J. Xiang, L. Liang, M. Wu, H. Zhao and Y. Lei, Nano Energy, 2017, 31, 514–524.
6. G. Longoni, R. L. Pena Cabrera, S. Polizzi, M. D'Arienzo, C. M. Mari, Y. Cui and R. Ruffo, Nano Lett., 2017, 17, 992–1000.
7. L. Yang, Y.-E. Zhu, J. Sheng, F. Li, B. Tang, Y. Zhang and Z. Zhou, Small, 2017, 13, 1702588.
8. F. Liu, X. Cheng, R. Xu, Y. Wu, Y. Jiang and Y. Yu, Adv. Funct. Mater., 2018, 28, 1800394.
9. Y. Wang, H. Huang, Q. Xie, Y. Wang and B. Qu, J. Alloys Compd., 2017, 705, 314–319.
10. D. Yang, X. Sun, K. Lim, R. Ranganathan Gaddam, N. Ashok Kumar, K. Kang and X. S. Zhao, J. Power Sources, 2017, 362, 358–365.
11. R. Alcántara, M. Jaraba, P. Lavela and J. L. Tirado, Chem. Mater., 2002, 14, 2847–2848.
12. H. Liu, M. Jia, Q. Zhu, B. Cao, R. Chen, Y. Wang, F. Wu and B. Xu, ACS Appl. Mater. Interfaces, 2016, 8, 26878–26885.
13. X. Ding, X. Huang, J. Jin, H. Ming, L. Wang and J. Ming, Electrochim. Acta, 2018, 260, 882–889.
14. S. Hariharan, K. Saravanan, V. Ramar and P. Balaya, Phys. Chem. Chem. Phys., 2013, 15, 2945–2953.
15. C. Yuan, H. B. Wu, Y. Xie and X. W. D. Lou, Angew. Chem., Int. Ed., 2014, 53, 1488–1504.
16. Y. Jiang, M. Hu, D. Zhang, T. Yuan, W. Sun, B. Xu and M. Yan, Nano Energy, 2014, 5, 60–66.
17. D. Li, D. Yan, X. Zhang, T. Lu, G. Yang and L. Pan, J. Mater. Sci.: Mater. Electron., 2017, 28, 10411–10419.
18. Y. Fu, Q. Wei, X. Wang, G. Zhang, H. Shu, X. Yang, A. C. Tavares and S. Sun, RSC Adv., 2016, 6, 16624–16633.
19. M. Pumera, Energy Environ. Sci., 2011, 4, 668–674.
20. R. Atif and F. Inam, Beilstein J. Nanotechnol., 2016, 7, 1174–1196.
21. B. Wang, W. A. Abdulla, D. Wang and X. S. Zhao, Energy Environ. Sci., 2015, 8, 869–875.
22. W. Wei, S. Yang, H. Zhou, I. Lieberwirth, X. Feng and K. Müllen, Adv. Mater., 2013, 25, 2909–2914.
23. B. Gill Choi, S.-J. Chang, Y. Boo Lee, J. Seong Bae, H. Jin Kim and Y. Suk Huh, Nanoscale, 2012, 4, 5924–5930.
24. H. Sun, L. Mei, J. Liang, Z. Zhao, C. Lee, H. Fei, M. Ding, J. Lau, M. Li and C. Wang, Science, 2017, 356, 599–604.
25. X. Zhao, C. M. Hayner, M. C. Kung and H. H. Kung, ACS Nano, 2011, 5, 8739–8749.
26. Z. Jiang, B. Pei and A. Manthiram, J. Mater. Chem. A, 2013, 1, 7775–7781.
27. Y. Xu, Z. Lin, X. Zhong, X. Huang, N. O. Weiss, Y. Huang and X. Duan, Nat. Commun., 2014, 5, 4554.
28. J. Xu, M. Wang, N. P. Wickramaratne, M. Jaroniec, S. Dou and L. Dai, Adv. Mater., 2015, 27, 2042–2048.
29. X. Q. Zhang, Y. C. Zhao, C. G. Wang, X. Li, J. D. Liu, G. H. Yue and Z. D. Zhou, J. Mater. Sci., 2016, 51, 9296–9305.
30. P. R. Kumar, Y. H. Jung, K. K. Bharathi, C. H. Lim and D. K. Kim, Electrochim. Acta, 2014, 146, 503–510.
31. Z. Xiong, L. Li Zhang, J. Ma and X. S. Zhao, Chem. Commun., 2010, 46, 6099–6101.
32. Y. Xu, C.-Y. Chen, Z. Zhao, Z. Lin, C. Lee, X. Xu, C. Wang, Y. Huang, M. I. Shakir and X. Duan, Nano Lett., 2015, 15, 4605–4610.
33. Q. Wang, L. Jiao, H. Du, Y. Wang and H. Yuan, J. Power Sources, 2014, 245, 101–106.
34. Z.-S. Wu, G. Zhou, L.-C. Yin, W. Ren, F. Li and H.-M. Cheng, Nano Energy, 2012, 1, 107–131.
35. Y. Fu, C. Peng, D. Zha, J. Zhu, L. Zhang and X. Wang, Electrochim. Acta, 2018, 271, 137–145.
36. D. Yang, Q. Zhao, L. Huang, B. Xu, A. K. Nanjundan and X. S. Zhao, J. Mater. Chem. A, 2018, 6, 14146–14154.
37. B. Xu, J. Zhang, Y. Gu, Z. Zhang, W. Al Abdulla, N. A. Kumar and X. S. Zhao, Electrochim. Acta, 2016, 212, 473–480.
38. A. K. Nanjundan, R. Gaddam, S. Varanasi, D. Yang, S. K. Bhatia and X. S. Zhao, Electrochim. Acta, 2016, 214, 319–325.
39. B. Xu, X. Guan, L. Y. Zhang, X. Liu, Z. Jiao, X. Liu, X. Hu and X. S. Zhao, J. Mater. Chem. A, 2018, 6, 4048–4054.
40. S. Li, J. Qiu, C. Lai, M. Ling, H. Zhao and S. Zhang, Nano Energy, 2015, 12, 224–230.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c8ta09188a

---