Reduced Graphene Oxide-Wrapped FeS2 Composite as Anode for High-Performance Sodium-Ion Batteries

Iron disulfide is considered to be a potential anode material for sodium-ion batteries due to its high theoretical capacity. However, its applications are seriously limited by the weak conductivity and large volume change, which results in low reversible capacity and poor cycling stability. Herein, reduced graphene oxide-wrapped FeS2 (FeS2/rGO) composite was fabricated to achieve excellent electrochemical performance via a facile two-step method. The introduction of rGO effectively improved the conductivity, BET surface area, and structural stability of the FeS2 active material, thus endowing it with high specific capacity, good rate capability, as well as excellent cycling stability. Electrochemical measurements show that the FeS2/rGO composite had a high initial discharge capacity of 1263.2 mAh g−1 at 100 mA g−1 and a high discharge capacity of 344 mAh g−1 at 10 A g−1, demonstrating superior rate performance. After 100 cycles at 100 mA g−1, the discharge capacity remained at 609.5 mAh g−1, indicating the excellent cycling stability of the FeS2/rGO electrode.

Abstract Iron disulfide is considered to be a potential anode material for sodium-ion batteries due to its high theoretical capacity. However, its applications are seriously limited by the weak conductivity and large volume change, which results in low reversible capacity and poor cycling stability. Herein, reduced graphene oxide-wrapped FeS 2 (FeS 2 /rGO) composite was fabricated to achieve excellent electrochemical performance via a facile two-step method. The introduction of rGO effectively improved the conductivity, BET surface area, and structural stability of the FeS 2 active material, thus endowing it with high specific capacity, good rate capability, as well as excellent cycling stability. Electrochemical measurements show that the FeS 2 /rGO composite had a high initial discharge capacity of 1263.2 mAh g -1 at 100 mA g -1 and a high discharge capacity of 344 mAh g -1 at 10 A g -1 , demonstrating superior rate performance. After 100 cycles at 100 mA g -1 , the discharge capacity remained at 609.5 mAh g -1 , indicating the excellent cycling stability of the FeS 2 /rGO electrode.
Keywords FeS 2 Á Reduced graphene oxide (rGO) Á Enwrapping structure Á Anode material Á Sodium-ion battery Sodium-ion batteries (SIBs) have been considered to be promising candidates for large-scale energy storage systems, electric vehicles, and other portable devices because of their outstanding electrochemical performance and inexpensive characterization. As the most important parts of SIBs, electrolytes and electrodes have been extensively investigated. It has been demonstrated that polymer electrolytes (such as gel polymer electrolytes and ceramic electrolytes) may be desirable alternatives for liquid electrolytes because of their modularity and reliability in electrochemical devices [1,2]. Various cathodes, including oxides (such as tunnel structure oxides and layered transition metal oxides), transition metal fluorides (typically FeF 3 ), polyanionic compounds (such as NaFePO 4 , NaVPO 4 F, Na 3 M 2 (PO 4 ) 3 NASICON compounds), Prussian blue analogues, and organic electrodes, have displayed outstanding sodium storage performance [3]. Meanwhile, significant progress has been achieved in the area of anodes for SIBs. It has been revealed that carbon materials, metals, alloys, and metal oxides/sulfides are promising anode materials for SIBs [4][5][6]. Among these materials, iron sulfides, such as FeS 2 [7][8][9], FeS [10] and Fe 1-x S [11,12], have attracted much attention due to their advantages of abundant reserves, non-toxicity, low cost, and high theoretical capacity (894 mAh g -1 for FeS 2 , based on the conversion reaction of FeS 2 ? 4Na $ 2Na 2 S ? Fe). However, the intrinsically low conductivity and notable volume change during the charge-discharge process greatly restrain its rate performance and cycling stability, restricting its further commercialization.
To solve these problems, Chen's group tuned an electrolyte and applied a higher voltage cutoff to improve the electrochemical performance of Na/FeS 2 and Li/FeS 2 cells [13,14]. Numerous studies have revealed that constructing nanostructured materials can greatly reduce the electron/ ion transport pathways and effectively buffer the large volume expansion during electrochemical processes, thus improving the reversibility and rate capability of FeS 2 anode materials [15,16]. Moreover, the combination of a carbon or a polypyrrole (ppy) modification strategy to form a coating or an embedded structure (such as FeS 2 /C [17][18][19][20], ppy@MoO 3 [21][22][23], ppy@V 2 O 5 [24]) would help prevent aggregation and improve the conductivity of the electrode materials, thus enhancing the cycling and rate performance. For example, Liu et al. designed FeS 2 @C nanoboxes and obtained discharge capacity of 511 mAh g -1 at 100 mA g -1 after 100 cycles [25]. Graphene is a highly conductive ultrathin nanosheet, with a large surface area and high flexibility, which is commonly used as modification material. In previous studies, FeS/ reduced graphene oxide (rGO) [26], FeS 2 /rGO [27], Fe 3 O 4 / rGO [28], Fe 2 O 3 /rGO [29], and LiFePO 4 /rGO [30,31] composites have been fabricated and used in lithium-ion batteries (LIBs). It is demonstrated that enwrapping anode materials in a conductive network is a favorable strategy to enhance the rate capability. However, the FeS 2 /graphene composite with an enwrapping structure for SIBs has not been reported.
Here, we report a two-step method for the preparation of a novel rGO-wrapped FeS 2 (FeS 2 /rGO) composite for SIBs. Structural and morphological characterization revealed that the FeS 2 nanoparticles are evenly surrounded in the interconnected rGO networks. The composite displayed superior sodium storage performance even at high charge-discharge current densities.

Materials Synthesis
FeS 2 /rGO was synthesized via a hydrothermal method, followed by a sulfurization process. All chemicals were of analytical grade and used without further purification.
Synthesis of Fe 3 O 4 /rGO composite In a typical synthesis, 0.04 g of graphene oxide was dispersed in 65 mL of deionized water by sonication. Then, 0.4 g of Fe(NO 3 ) 3-9H 2 O was dissolved in the above suspension and stirred for 4 h at 70°C. Following this, 5 mL of N 2 H 4 ÁH 2 O was added to the above system, and the solution was sealed in a 100-mL Teflon-lined stainless-steel autoclave for hydrothermal reaction at 150°C for 6 h. Finally, the rGOwrapped Fe 3 O 4 composite was collected by centrifugation, washed with water and ethanol three times, and dried at 70°C in a vacuum for 12 h.
Synthesis of FeS 2 /rGO Composite The as-prepared Fe 3 O 4 /rGO and sulfur powder in a weight ratio of 1:2 were mixed and pressed into a small tablet and sealed in a small quartz tube under Ar atmosphere. Then, the quartz tube was heated at 150°C for 2 h and subsequently at 550°C for 6 h in a quartz tube reactor. After cooling down and washing with CS 2 to remove the residual sulfur powder, the final FeS 2 /rGO composite was obtained. For comparison, FeS 2 nanoparticles were prepared using the same method without the addition of rGO.

Materials Characterization
The crystal structures of the as-prepared samples were characterized by powder X-ray diffraction (XRD) using Cu Ka radiation. The morphologies were investigated using field-emission scanning electron microscopy (SEM) on a JEOL JSM-7500FA system and transmission electron microscopy (TEM) on a Philips Tecnai 20 (200 kV). TG thermal nitrogen adsorption-desorption isotherms of the samples were obtained on a Quantachrome Autosorb-IQ2 analyzer at 77 K. Specific surface areas were measured by Brunauer-Emmett-Teller (BET) analysis.

Electrochemical Measurements
Electrochemical measurements were conducted using CR2032 two-electrode coin cells, with sodium metal as the counter and reference electrodes and glass fiber as the separator. The working electrodes were made by pasting a slurry on copper foil, followed by drying in vacuum at 80°C for 12 h. The slurry was prepared by mixing active materials, Super P, and carboxymethyl cellulose in the weight ratio of 8:1:1. A solution of 1 M NaClO 4 in ethylene carbonate/propylene carbonate (v/v = 1/1) with 5 wt% fluoroethylene carbonate additive was used as the electrolyte.
Cyclic voltammetry (CV, 0-2.5 V, 0.1 mV s -1 ) tests and electrochemical impendence spectroscopy (EIS, with 5 mV amplitude in a frequency range from 100 kHz to 0.01 Hz at open-circuit potential) tests were conducted on a Biologic VMP-3 electrochemical workstation. The galvanostatic charge-discharge curves, cycling performance, and rate capabilities of the electrode materials were tested on a LAND Battery Test System, in the voltage range of 0.01-2.3 V. All the tests were carried out at room temperature.

Results and Discussion
The crystal structures and morphologies of the as-prepared  Figure 1c shows that each nanoparticle is surrounded by thin graphene nanosheets. The TEM image shown in Fig. 1d further confirms the enwrapped structure of the composite. Moreover, it is obvious that each nanoparticle is connected by rGO to form an integral 3D network. Figure 2 shows the crystal structure and morphologies of the as-prepared FeS 2 and FeS 2 /rGO samples. As shown in Fig. 2a, both the samples show high-intensity XRD peaks, all of which can be indexed to pyrite FeS 2 (JCPDS card No. 06-0710), demonstrating the high purity and good crystallinity of the two samples. Figure 2b, c shows that the FeS 2 sample is composed of irregular particles like the Fe 3 O 4 precursor. However, it can be observed that the particles tend to aggregate and become larger than the precursor, which is caused by the sulfuration process. Figure 2d shows that the general morphology of the FeS 2 / rGO composite is similar to that of the Fe 3 O 4 /rGO precursor. The TEM image shown in Fig. 2e further reveals that the FeS 2 nanoparticles continue to be evenly dispersed in the graphene networks and that the particle size remains largely unchanged, compared to its precursor. Figure 2f shows that each nanoparticle is surrounded by graphene, which effectively prevents the aggregation of the FeS 2 nanoparticles. The thickness of the graphene layer is 2-3 nm (Fig. 2g). High-resolution TEM images (HRTEM, Fig. 2g, h) display clear lattice fringes with an interplane distance of 0.16 nm, corresponding to the (311) plane of pyrite FeS 2 . The selected-area electron diffraction (SAED) pattern of FeS 2 /rGO (Fig. 2i) shows well-defined rings, indicating that the as-prepared FeS 2 is polycrystalline.
According to the N 2 adsorption-desorption measurements (Fig. 3b), the specific surface areas of FeS 2 and FeS 2 /rGO are 25.6 and 58.1 m 2 g -1 , respectively, indicating that the introduction of rGO significantly increases the surface areas. To determine the rGO content in the composite, thermogravimetric analysis is carried out in an air atmosphere (Fig. 3a). Both the samples display a minor weight loss (* 6-8%) under 200°C, which is due to the vapor of the residual water in the materials. Then a large weight loss of about 35% is observed in the range 400-500°C for pure FeS 2 , which corresponds to the conversion of FeS 2 to Fe 2 O 3 . (The theoretical weight loss is * 33.3%.) For FeS 2 /rGO, a more significant weight loss of about 42% is observed between 400 and 600°C, which may be caused by the phase change of FeS 2 to Fe 2 O 3 and rGO to carbon dioxide. Based on the thermogravimetric analysis, the weight content of FeS 2 in the FeS 2 /rGO composite can be calculated to be about 79.1%. According to the above analysis, the FeS 2 /rGO composite contains integral nanostructures, with the FeS 2 nanoparticles enwrapped in the 3D rGO networks. This unique structure endows the composite with high structural stability and super electron conductivity, which may be beneficial for the cycling stability and rate performance of the FeS 2 electrode material for sodium storage. Figure 4a, b shows the cyclic voltammetry (CV) curves of pure FeS 2 and the FeS 2 /rGO composite at a scan rate of 0.1 mV s -1 between 0 and 2.5 V (vs. Na/Na ? ). During the initial cathodic scan, a large peak appears at 1.0 V and a broad peak appears at 0.25 V for the FeS 2 electrode, which corresponds to Na ? intercalation and the formation of the Na x FeS 2 (x \ 2) phase, Fe and Na 2 S, and the formation of a solid-electrolyte interface (SEI) layer [13,16,32]. For the FeS 2 /rGO electrode, a large peak at * 0.65 V and a small peak at * 0.1 V are detected, which may be due to a similar electrochemical process with the FeS 2 electrode. The differences in the peaks of the two samples may be caused by the nanostructure and the introduction of rGO. During the subsequent anodic scan, the peaks observed at * 1.4 and * 1.8 V can be attributed to be the desodiation process, with the formation of Na 2 FeS 2 and Na 2-x FeS 2 [19]. During the subsequent cycles, the CV curves are quite different from those in the initial cycle, which may be due to the irreversible formation of the SEI layer and the decomposition of the electrolyte [19,[33][34][35]. It can be observed that the FeS 2 /rGO electrode shows much better repeatability and a larger closed curve area than those of the pure FeS 2 electrode, demonstrating its much better cycling stability and higher specific capacities. Figure 5a presents the charge-discharge curves of FeS 2 / rGO electrode at a current density of 100 mA g -1 . An initial discharge plateau at * 1.0 V (vs. Na/Na ? ) and charge plateau at * 1.3 V are observed, which are in good agreement with the CV curves. In the subsequent cycles, the charge-discharge curves do not change much, showing good electrochemical reversibility. The cycling performances of the two samples are further evaluated at 100 mA g -1 . As shown in Fig. 5b, both the electrodes have quite good cycling stability. However, the FeS 2 /rGO electrode has obviously higher specific capacities than does the pure FeS 2 electrode, which may be due to the higher utilization of the active materials after the introduction of rGO. The FeS 2 /rGO composite displays a high initial discharge capacity of 1263.2 mAh g -1 and charge capacity of 759.4 mAh g -1 , showing a low coulombic efficiency of 60.1%, which is mainly caused by the irreversible formation of the SEI layer and electrolyte decomposition in the initial cycle. Moreover, the dissolution of sodium polysulfides into organic liquid electrolytes causes a parasitic redox shuttle, leading to unfavorable side reactions with sodium, reducing the charging efficiency and resulting in serious capacity decay [36][37][38]. In the following cycles, the coulombic efficiency increases over 95%. From the second cycle, the discharge and charge capacities are stable and remain at 609.5 and 581.7 mAh g -1 , respectively, after 100 cycles.
The rate capability of the two FeS 2 electrodes is evaluated using varying current densities from 0.1 to 10 A g -1 and back to 0.1 A g -1 . As shown in Fig. 5c, the average specific capacities for FeS 2 /rGO electrodes are 705, 672, 613, 555, 496, 426, and 344 mAh g -1 at 0.1, 0.2, 0.5, 1, 2, 5, and 10 A g -1 , respectively, which are remarkably higher than those for pure FeS 2 electrode, demonstrating its superior rate performance. When the current density is altered back to 0.1 A g -1 , the reversible capacity remains at * 655 mAh g -1 after 90 cycles, further confirming the excellent cycling stability of the FeS 2 /rGO composite. We further investigate the electrode process kinetics of the two materials through EIS. As shown in Fig. 5d, both the Nyquist spectra are composed of a semicircle in the highfrequency region and an inclined line in the low-frequency region. The bigger semicircle for the FeS 2 electrode illustrates the poor electrical conductivity of the active 2 Theta (degree) 10 20 Intensity (a.u.) 30   materials. According to the Z-view program in the Sai software set, R ct for FeS 2 and FeS 2 /rGO electrodes is 1055.1 and 291.9 X, respectively, illustrating the better charge transfer kinetics of the FeS 2 /rGO electrode. The FeS 2 /rGO composite displays much higher specific capacity and better rate capability than does the pure FeS 2 electrode. It is inferred that several features may contribute to the excellent electrochemical properties. First, the intimate contact of the FeS 2 nanoparticles with rGO and the integral conductive rGO networks provide a facile electron transport pathway, ensuring good rate performance [27,30]. Second, the unique enwrapping structure can effectively improve the structural stability and buffer the volume change of FeS 2 during the charge-discharge process [26,28]. To investigate the structural stability, the nanostructures of the freshly prepared FeS 2 /rGO electrode and the FeS 2 /rGO electrode after 100 cycles are investigated by SEM and TEM. From Fig. 6a, it can be seen that the morphology of the FeS 2 /rGO composite does not change. After 100 sodiation-desodiation cycles, the nanoparticles are not very regular but are still enwrapped in the graphene networks (Fig. 6b, c). The high-resolution TEM test shows that the nanoparticles transform into smaller nanocrystals (Fig. 6d), which are still surrounded It is obvious that the graphene network can effectively prevent the collapse of the structure and the aggregation of FeS 2 nanoparticles, thus improving the cycling stability of the FeS 2 /rGO composite. Moreover, the improvement of the BET surface area increases the contact area between the active material and the electrolyte, which helps improve the utilization of active materials, endowing the FeS 2 /rGO composite with high specific capacitance.

Conclusions
In summary, an rGO-wrapped FeS 2 composite has been successfully synthesized via a hydrothermal method, followed by sulfuration, and used as an anode for SIBs. The well-dispersed rGO constructs 3D conductive networks and markedly increases the BET surface area and conductivity of the FeS 2 nanoparticles. Thus, the FeS 2 /rGO composite displays an initial discharge capacity of 1263.2 mAh g -1 at 100 mA g -1 and a high discharge capacity of 344 mAh g -1 at 10 A g -1 . Moreover, the enwrapping structure helps in preventing the aggregation of the FeS 2 nanoparticles during the electrochemical process, contributing to the excellent cycling stability. After 100 cycles, the discharge capacity is 609.5 mAh g -1 . We believe that our strategy could be extended to the fabrication of other high-performance metal sulfide/rGO composites for LIBs or SIBs.