Self-supporting Co3O4/Graphene Hybrid Films as Binder-free Anode Materials for Lithium Ion Batteries

A self-supporting Co3O4/graphene hybrid film has been constructed via vacuum filtration of Co(OH)2 nanosheet and graphene, followed by a two-step thermal treatment. Within the hybrid film, Co3O4 nanoparticles with size of 40~60 nm uniformly in-situ grew on the surface of graphene, forming a novel porous and interleaved structure with strong interactions between Co3O4 nanoparticles and graphene. Such fascinating microstructures can greatly facilitate interfacial electron transportation and accommodate the volume changes upon Li ions insertion and extraction. Consequently, the binder-less hybrid film demonstrated extremely high reversible capacity (1287.7 mAh g−1 at 0.2 A g−1), excellent cycling stability and rate capability (1110 and 800 mAh g−1 at 0.5 and 1.0 A g−1, respectively).

Herein, we report the synthesis of a self-supporting Co 3 O 4 /graphene hybrid film, which was directly used as anode materials for LIBs without any binder or additives. Co 3 O 4 nanoparticles (40~60 nm) in-situ grew on the surface of GS, forming a novel interleaved structure with strong interfacial interactions. The hybrid electrode showed extremely high reversible capacity (1287.7 mAh g −1 at 0.2 A g −1 ), good cycling stability (capacity retention of 85.5% after 100 cycles) and excellent rate capability, owing to the fascinating microstructure and binder-free electrode nature, demonstrating great potential to be used in energy storage filed. Figure 1 displayed the synthesis process of Co 3 O 4 /GS hybrid films. The surface of GO, functionalized with oxygen-containing functional groups 17 , was measured to be a negatively charged surface (zeta potential: −60 mV) in this study. Co(OH) 2 colloid, which was tested to be with a zeta potential of +22 mV, was a positively charged dispersion with high stability. Learning from colloid science, there are strong electrostatic attractive interactions between two colloid with opposite charges. As a result, a flocculent solution was produced immediately when mixing GO colloid with Co(OH) 2 colloid, indicating an excellent self-assembly of the two dispersions driven by the strong electrostatic attractive interaction 18 . Afterwards, the flocculent solution can be easily vacuum-filtrated to construct a self-supporting Co(OH) 2 /GO hybrid film. Further two-step heat treatments will lead to the construction of a porous and free-standing Co 3 O 4 /GS hybrid film which can be used as anode electrode directly.

Results
XRD was employed to characterize the phase transformation of Co 3 O 4 during preparation procedures. As shown in XRD pattern of Co(OH) 2 /GO hybrid films (Figure 2a), the well-defined diffraction peaks can be successfully indexed to hexagonal β-Co(OH) 2 (JCPDS no. 74-1057). No obvious peaks of GO were detected, because graphene oxide sheets were highly separated by Co(OH) 2 nanosheets. After annealed in Ar atmosphere at 600 °C for 1 h, Co(OH) 2 /GO was first converted to Co/GS as shown in Figure 2b, in which three main XRD peaks can be well assigned to Co phase (JCPDS no. 15-0806). Noteworthy, there are also no peaks of GS been detected, indicating graphene sheets were well dispersed without any aggregation. After the second step oxidation, clearly, Co phase was successfully transformed to face-centered cubic Co 3 O 4 (Figure 2c, JCPDS no. . No impurity peaks were observed, which manifests that the Co(OH) 2 was completely transformed to pure Co 3 O 4 via a two-step heat treatments. The absence of peaks for GS suggest that graphene maintained the highly dispersibility and the restacking of graphene was well controlled during the whole preparation.
The chemical changes from precursors to Co 3 O 4 /GS hybrid films were analyzed by FTIR spectra. As displayed in the FTIR spectrum of Co(OH) 2 /GO (Figure 3a), the visible peaks at about 3437, 1640 and 1077 cm −1 correspond to the stretching vibration of -OH in water, stretching vibration of C=C and C-O, respectively, while the peak locating at 1390 cm −1 represents the bending vibration of C-OH 19,20 . Apart from these characteristic peaks of oxygen functional groups originating from GO 19,20 , the spectrum of Co(OH) 2 /GO also exhibited stretching vibrations bands of Co-O bond centered at 489.8 cm −1 and O-OH bond in Co(OH) 2 (~3631 cm −1 ) 21,22 . After thermal treatment, the intensities of peaks belonging to oxygen-containing functional groups significantly decreased for Co 3 O 4 /GS, suggesting the effective reduction of GO due to the thermal decomposition of these oxygen-containing functional group 23,24 . Meanwhile, the characteristic peak of υ(O-OH) in Co(OH) 2 disappeared, instead with two distinctive peaks appeared at 563.1 and 657.6 cm −1 , respectively. The first band υ1 at 563.1 cm −1 corresponds to the BOB 3 vibrations in the spinel lattice, in which B is associated with the Co cations in an octahedral position, i.e. Co 3+ ions 4,8,10,25 . The second bands υ2 at 657.6 cm −1 is resulted from the ABO 3 vibrations (A: the metal ions in a tetrahedral position). These two stretching band can be well assigned to the characteristic peaks of Co 3 26,27 . While the spectrum of Co 3 O 4 /GS displayed two distinctive peaks at 1347 and 1592 cm −1 apart from those for Co 3 O 4 , corresponding to the D and G band of graphene 28,29 , respectively. This further proves the successful preparation of Co 3 O 4 /GS composites, in accordance with the XRD and FTIR results. As shown in Figure 3d, the enlarged Raman spectra, the characteristic peaks representing Co 3 O 4 adsorption bands in the hybrid film showed  The microstructures of the hybrid films were illustrated by SEM. The as synthesized Co(OH) 2 shows a sheet-like morphology and will be well dispersed on the matrix of GO via electrostatic attractive force ( Figure S1) 22 . After vacuum-filtration, a free-standing hybrid film with a compact layer-by-layer structure was obtained subsequently(shown in Figure S2). Interestingly, after annealing at 600 °C in Ar for 1 h, Co(OH) 2 nanosheets firstly decomposed and then in-situ reduced to cobalt nanoparticles instead of nanosheets (Figure 4a and 4b).
More importantly, Co nanoparticles tightly and uniformly decorated on both surface sides of GS, suggesting strong interfacial interactions, presumably covalent bond, have formed between them. The cobalt was further in-situ transformed to Co 3 O 4 nanoparticles with size of 40~60 nm via oxidation treatment, bringing in a self-supporting film with a porous interleaved structure (Figure 4c and 4d). The diameter of the hybrid film was about 38 mm ( Figure S3), with a thickness of ~10μm (Figure 4c). The novel interleaved porous structure of Co 3 O 4 /GS hybrid film will greatly facilitate fast Li-ion diffusion within the electrode, and also supply abundant buffer space to allow the volume expansion of Co 3 O 4 during charge/discharge processes.

Discussion
The hybrid film was directly used as electrodes, in which no polymer binder or carbon additives was added. Figure 5 illustrates the electrochemical properties of the binder-free Co 3 O 4 /GS hybrid electrode. As shown in the CV curves, a strong cathodic peak at about 0.6 V is observed in the first discharge process, corresponding to the multi-step electrochemical reduction between Li ions and Co 3 O 4 . Two anodic peaks appeared at ~1.5 and 2.3 V, owing to the oxidation of Co atoms and in consistent with the reported literatures 30,31 . The reactions can be described as: In the subsequent cycles, the cathodic peaks shifted to ~0.75 V, showing a tendency toward stabilization. The anodic peaks were almost the same and overlapped together, indicating the good electrochemical reversibility of the hybrid electrode.
The lithium-storage performance of the binder-free Co 3 O 4 /GS film was characterized by galvanostatic charge/ discharge at 0.2 A g −1 (Figure 5b). In the discharge curve of first cycle, the potential quickly falls to a long potential plateau at ~1.5 V and then gradually decreased to 0.01 V (the cutoff voltage), which is in analogy with the behavior of Co 3 O 4 anode. The first discharge/charge capacities reached 1718.5 and 1287.7 mAh g −1 , respectively, corresponding to the initial coulombic efficiency (CE) of 74.9%, which was comparable to that of Co 3 O 4 /graphene hybrid electrodes reported previously 12,32 . The relative low initial CE value may be ascribed to the inevitable formation of a solid electrolyte interface (SEI) film over the electrode during the charge/discharge process, which led to the insufficient release of capacity. A well-established conductive network may help to further increase the CE 32 , such as surface modification 13 , pre-doping lithium metal 33 and encapsulation 34 . Compared with the theoretical capacity of graphene (372 mAhg −1 ) and Co 3 O 4 (890 mAh g −1 ), the extra capacity may owe to the formation of SEI film or interfacial Li-ion storage 35,36 . After fifty charge/discharge cycles, the discharge capacity still remains up to 1184.2 mAh g −1 , demonstrating the good structural stability of the Co 3 O 4 /graphene anode. Figure 5c highlights the cycling stability of the Co 3 O 4 /GS anode at 0.2 A g −1 . The hybrid anode retains a reversible capacity of 1095.1 mAhg −1 after 100 cycles with the capacity retention of 85.5%. The rate capacity of the self-supported Co 3 O 4 /GS electrode has also been studied in Figure 5d. The capacities of Co 3 O 4 /GS heterostructures were about 1480, 1300, 1110, 920, 800, 620, 530 and 410 mAh g −1 at the current densities of 0.1, 0.2, 0.5, 0.8, 1.0, 1.3, 1.5 and 2.0 A g −1 , respectively, manifesting an excellent rate capability. Importantly, the capacity re-increased to 1601.2 mAh g −1 when the current density returns back to 0.1 A g −1 , further demonstrating the good reversibility. Compared with the capacities and cycling performance of other Co 3 O 4 -based anode materials reported previously (shown in Table 1), the constructedCo 3 O 4 /GS anode exhibited higher capacity and better stability than Co 3 O 4 /nitrogen doped graphene 13 39 and so on. For the as-prepared Co 3 O 4 /GS hybrid film, the strong interfacial interactions between Co 3 O 4 and GS can greatly facilitate the interfacial charges transportations, which is beneficial to enhance the lithium-storage capacity at high current densities. In addition, more active sites were achieved owing to the small particle size of Co 3 O 4 and porous interleaved structure, accounting for the enhanced capacity. The flexible graphene substrate and porous structure, can provide buffer space for volume change of Co 3 O 4 nanoparticles and effectively prevent their aggregation, which contributes to the excellent

Electrode Material Capacityretention
Specific capacity (mA h g −1 ) Reference cycling stability. Overall, the Co 3 O 4 /GS hybrid film delivered superior electrochemical performance benefitting from the synergistic effects of strong interfacial interactions between Co 3 O 4 and graphene, small particle size of Co 3 O 4 , the interleaved porous structure and binder-free electrode nature, suggesting the superiority of using the hybrid films as an anode material for LIBs. Moreover, the two-step fabrication process is simple, controllable and low-cost, showing great promise to be used in practical applications

Conclusion
In summary, a vacuum filtration procedure combined with two-step heat treatment has been developed to construct self-supporting Co 3 O 4 /GS hybrid films, with Co 3 O 4 nanoparticles (40-60 nm) uniformly and tightly decorated on both surface of GS. The formed porous and interleaved microstructure exhibits critical characters as desired anode materials for LIBs, such as strong interfacial interactions, short transport length for Li-ion, and sufficient space for stress relaxation. Consequently, the constructed Co 3 O 4 /GS hybrid film as binder-free electrode exhibited a high capacity of 1287.7 mAh g −1 at 0.2 A g −1 , good cycling stability (capacity retention of 85.5% after 100 cycles at 0.2 A g −1 ) and superior rate capability, making the hybrid films competitive as LIBs anode materials.

Materials and Methods
β-Co(OH) 2 was synthesized based on previous report and further diluted to a homogeneous dispersion (0.5 mg mL −1 ) 40,41 . Graphene oxide (GO) was prepared using a modified Hummers method with natural graphite as raw materials 42 , which was also diluted to 0.5 mg mL −1 for further use. For preparation of Co 3 O 4 /GS film, the β-Co(OH) 2 and GO dispersion were mixed together by sonication; and then vacuum filtered to form a self-supporting Co(OH) 2 /GO hybrid film. Afterwards, two-step heat treatment approach was carried out: First, the obtained Co(OH) 2 /GO film was heat-treated in Ar at 600 °C for 1 h to produce Co/GS. Second, Co/GS was further oxidized to Co 3 O 4 /GS by being calcined in air (300 °C, 2 h).

Materials Characterizations.
The morphologies and microstructures of the samples were characterized via field-emission scanning electron microscopy(FE-SEMJSM-6700F). The crystalline phase of the materials was analyzed from X-ray diffraction measurements (Rigaku D/max 2550 V diffractometer). Raman spectroscopy was conducted on DXR Raman Microscope (Thermal Scientific Corporation, USA, wavelength 532 nm). Fourier transform infrared spectroscopy (FTIR) were examined on Nicolet 7000-C.
Electrochemical Measurements. In this paper, the as-synthesized self-supporting Co 3 O 4 /GS hybrid films were directly used as an electrode without adding any binder or additive. The cell was assembled in glove box, and Li foil was adopted as the counter electrode. The 1 M LiPF 6 in a mixture of ethylene carbonate (EC)/dimethyl carbonate (DMC)(50:50, by volume) was used as electrolyte. The electrochemical properties were measured using a CT2001 battery tester at room temperature. Cyclic voltammetry (CV) was performed via electrochemical workstation (CHI760E) at a scan rate of 0.5 mV s −1 within a voltage range of 0-3.0 V. The mass loading was about 0.8 mg for each Co 3 O 4 /GS electrode.