Introduction

Graphene oxide (GO) is commonly made by reacting graphite powders with strong oxidation agents and acids.12 During this process, graphene is oxidized and derivatized with oxygen containing functional groups, and consequently is readily exfoliated from graphite into single GO layer in water.34 However, the presence of functional groups includ-ing hydroxyl, epoxide, and carboxylic acid groups on a carbon sheet and amorphous carbon regions on basal plane causes this carbon sheet to be electronically insulating. Therefore, considerable works have been focused on re-covering the poor electrical properties of GO through the development of efficient reducing strategies or the exclusion of harsh oxidation steps for realizing the highly motivated properties of pristine graphite.5-7 Despite those disadvant-ageous properties, however, GO has attracted increasing research interest owing to its unique properties including large surface area, high water solubility, facile processibility, and easy large-scale production.8-11 In particular, the func-tional groups of GO may play an important role in faci-litating molecular organization into functional hybrid nano-composites.12

A variety of metal oxides, such as MnOx, RuOx, and CuOx, with pseudo-capacitive properties have been used to form nanocomposites with graphene and have been utilized in supercapacitors because of the improved capability to store more charges than carbon-only electrodes.13-16 Up to date, the preparation methods including hydrothermal synthesis or a polyol process have been successfully achieved to create nanocomposites.17-20 However, the development of a scalable, robust synthetic process is still required to achieve the practical utilization of graphene into high performance supercapacitors.

In this work, we introduce a straightforward preparation approach to prepare a hybrid of graphene and SnO2 via the directed hybridization. SnO2 has attracted much attention as an excellent electrode material for supercapacitors and secondary Li-ion batteries due to its interesting properties of high theoretical capacity, good electroconductivity and low cost.2122 GO was first prepared through chemical exfoliation of graphite and was mixed with a SnO2 precursor solution. During solution mixing, the reduction of GO and the con-version of SnO2 nanoparticles from Sn2+ occur simultane-ously. After further reduction of GO by adding hydrazine and thermal annealing, nanocomposites (rGO-SnO2) were created. Hybrid materials showed the increased super-capacitance values and the more stable electrochemical per-formance over a wide range of voltage scan rates, compared with the rGO-only electrode.

Experimental

Materials. Graphite powders were purchased from Bay carbon (SP-1). Other chemicals, including hydrazine, hydro-chloride, sulfuric acid, hydro peroxide, and tin acetate (Sn(CH3COO)2), were purchased from Sigma-Aldrich. All chemicals were used as received without additional puri-fication.

Synthesis of GO. GO was prepared using a modified Hummers method from graphite powder as reported else-where. 10-12 An aqueous GO dispersion was extensively wash-ed and filtered with 1 M HCl and then was dialyzed with a dialysis membrane (Spectra Dialysis Membrane, MWCO: 6-8,000) to remove the salt byproduct and excess acid. After dialysis, the viscous GO solution was diluted in deionized water (DI) and was put in a water-bath-sonicator for mono-layer exfoliation. The concentration of the resulting GO solution was 3.2 mg/mL.

Synthesis of Graphene-SnO2 Nanocomposites. Tin acetate (Sn(CH3COO)2) was dissolved in a 1:10 DI/MeOH mixture at 20 mg/mL. Then, 3 mL of the SnO2 precursor solution was added dropwise into 40 mL of the GO solution (3.2 mg/mL) under vigorous stirring at room temperature. To mea-sure sheet resistivity of hybrid, the nanocomposites film of 10 μm thickness was prepared with a vacuum filtration kit using a hydrophilic PTFE membrane (Millipore, pore size: 450 nm). For further reduction of composites, 30 μL of hydrazine (Sigma-Aldrich, 64%) was added into a GO-SnO2 solution and mechanically stirred at 70 °C for 2 h. The finally obtained rGO-SnO2 nanocomposites were annealed at 400 °C for 2 h under air.

Characterization. The morphologies of the nanocompo-sites were characterized with a scanning electron microscopy (JEOL JSM-6701F) and a transmission electron microscopy (JEOL JEM-2100F). TEM samples were prepared by disper-sing a small amount of the nanocomposite powder in MeOH with sonication and then applying a few drops of the dispersion onto a lacey-carbon TEM grid (Ted Pella, Inc.). Thermogravimetric analysis (TGA) was performed on SDT Q600 (TA Instruments) in the temperature range between 25 and 800 °C at a heating rate of 10 °C/min under air atmos-phere. X-ray diffraction (XRD) was carried out using Cu Kα radiation (λ = 1.5418 Å). X-ray photoelectron spectroscopy (XPS) measurements were carried out using a Thermo Scientific Theta probe with monochromatic Al Kα radiation. Sheet resistivity of hybrid films was measured with Advan-ced Instrument Technology (CMT-Series).

Electrochemical Measurements. The electrochemical behavior of rGO and rGO-SnO2 was characterized by cyclic voltammetry (CV) and galvanostatic charge-discharge mea-surements using a Bio-Logic (SP-200) in 1 M H2SO4 elec-trolyte. The working electrode was prepared by dropping 10 μL of active material solution (5.0 mg/mL) in NMP onto the glassy carbon electrodes. Prior to experiments, the electro-lyte was purged with pure N2 gas for 30 min to remove dissolved oxygen. Experiments were carried out in a three-electrode glass cell at room temperature. Platinum foil was used as a counter electrode and an Ag/AgCl electrode as the reference electrode.

Results and Discussion

As shown in Figure 1(a), the GO solution was brown in color, and there was no precipitation. To prepare a GO-SnO2 hybrid, a tin acetate (Sn(CH3COO)2) precursor solution was added dropwise under vigorous mechanical mixing. As soon as the tin acetate solution was added, the GO solution became viscous, as shown in images of Figure 1(b) and (c). When the concentration of GO was higher than 4.0 mg/mL or the precursor solution of 20 mg/mL was added more than 3.0 mL, the complex easily formed a gel. This change in the fluidity of the mixture solution is due to the formation of the molecular networks between carbon sheets and SnO2 nanoparticles, as often exhibited in the percolated solution of hydro-/organo-gels.23-25 In this work, the concentration of GO and tin acetate solution was kept consistently under the gel-forming critical concentration to avoid heterogeneous mixing. In addition, the color of GO suspension changed to dark brown, which further indicates the deoxygenation of GO. Compared with insulating GO paper, thin films of GO-SnO2 hybrid showed a much lower sheet resistivity of 8.2 × 106 Ω/sq. due to the reduction effect of GO and the presence of the electroconductive SnO2 nanoparticles on GO sheet. The further reduction of GO was carried out by loading a small amount of hydrazine solution, and the reduced GO (rGO)-SnO2 hybrid composites were annealed to enhance the crystallinity of SnO2 nanoparticles at 400 °C. The weight ratio of SnO2 nanoparticles in the hybrid was verified with TGA (Figure 1(d)). A significant weight loss between 400 and 700 °C is attributed to the decomposition of the carbon sheets.26 Based on the TGA result, it is confirmed that the hybrid contains SnO2 nanoparticles of approximately 6.8 wt %.

Figure 1.Hybrid assembly of GO-SnO2 hybrid. a-c) Photographs of the sample: aqueous GO suspension (a) and GO-SnO2 complex (b) after mixing. GO solution became viscous and dark brown color (c). By adding a small amount of hydrazine GO was reduced, and then thermally annealed to enhance the crystallinity of SnO2. d) TGA curves of the reduced GO (rGO) and rGO-SnO2. e and f) SEM images of rGO-SnO2 nanocomposites before (e) and after thermal annealing (f).

The morphology of the rGO-SnO2 hybrid was charac-terized by field emission scanning electron microscopy (FE-SEM) before and after annealing, as shown in Figure 1(e) and (f), respectively. We note that, though the surface of hybrid composites is quite roughened and crumpled before annealing (Figure 1(e)), the rough surface wasn’t entirely flattened and the nanocomposites didn’t agglomerate even after high temperature annealing (Figure 1(f)).

Figure 2.Crystalline morphology of the hybrid. (a) TEM image of isolated rGO-SnO2 nanocomposites. (b) A selected area electron diffraction (SAED) pattern of hybrid (a). (c) A magnified HRTEM image corresponding to the pointed region with a white arrow in (a). The inset image of (c) shows the crystal lattice fringes of highly crystalline SnO2 nanoparticles. The crystalline lattice spacing is 3.35 Å corresponding to the (110) lattice plane of a cassiterite SnO2 crystal. (d) X-ray diffraction of rGO-SnO2 nanocomposites. XRD pattern of hybrid has six well-resolved peaks assigned as the diffraction peaks of cassiterite of SnO2 phase (JCPDS No. 41-1445). The peaks of the sample annealed at 400 °C for 2 h were greatly enhanced.

The crystalline structure of the annealed SnO2 nano-particles and their spatial distribution over carbon sheets were examined with the high-resolution transmission elec-tron microscopy (HR-TEM). Figure 2(a) shows the isolated single layer of rGO-SnO2 hybrid. The rGO was uniformly coated with SnO2 nanoparticles and this nanocomposite was not densely aggregated. As shown in the higher magnifi-cation image (Figure 2(c)) of the indicated region with a white arrow in Figure 2(a), the size of SnO2 particles was 3-5 nm in diameter and evenly dispersed throughout the rGO sheet. It should be noted that the structure of well-distributed nanoparticles over rGO sheets improves electron transfer, resulting in enhanced electrochemical properties. Figure 2(b) shows the selected area electron diffraction (SAED) pattern of the rGO-SnO2 hybrid; this result is consistent with a typical electron diffraction pattern for cassiterite SnO2 (Figure 2(d)). As shown in Figure 2(c), the crystal lattice fringes were observed over the entire sample, demonstrating the high crystallinity of SnO2 nanoparticles. As indicated in the inset image of Figure 2(c), the crystalline lattice spacing of 3.35 Å is distinct. This corresponds to the (110) lattice plane of a cassiterite SnO2 crystal.

Figure 2(d) presents the powder X-ray diffraction (XRD) patterns of rGO-SnO2 nanocomposites with six well-re-solved diffraction peaks. Those peaks are typical diffraction peaks from the (110), (101), (200), (211), (112) and (301) crystalline lattice planes of the cassiterite of SnO2 phase (JCPDS 41-1445, a = 4.7382 Å, c = 3.1871 Å). An average crystallite size of 3.3 nm was calculated using the Scherrer equation:

where L is the crystallite size and B(2θ) is the line width. The value is completely consistent with the results of TEM characterization. The intensity of the crystalline SnO2 peaks increased after thermal annealing. The most intense peak of rGO (at around 2θ = 16.9°) indicates much larger interlayer spacing (0.41 nm), compared with the pristine graphite (0.34 nm).

Figure 3.XPS patterns of the GO (a) and rGO-SnO2 nano-composites (b). c) The Sn 3d spectrum of the prepared rGO-SnO2 nanocomposite. The C 1s XPS spectra of the GO (d), GO-SnO2 nanocomposites (e), and rGO-SnO2 nanocomposites (f).

Figure 3(a) and (b) show the full range XPS spectra of GO and rGO-SnO2 hybrid, respectively. As presented in Figure 3(b), the Sn 3p, 3d, 4s, 4p, and 4d peaks appear due to the presence of SnO2 nanoparticles, and the peak of C 1s is attributed to rGO.27 In the high-resolution Sn 3d XPS spectrum in Figure 3(c), peaks of Sn 3d3/2 and Sn 3d5/2 are at 487.3 and 495.7 eV, with an 8.4 eV peak-to-peak separation, which indicates the formation of SnO2 nanoparticles on graphene.28 The C 1s XPS spectra of GO and hybrid nanocomposites are shown in Figure 3(d)-(f). The bands at 287-289 eV corresponding to the oxygenated functional groups were definitely shown in Figure 3(d).

Compared with GO, the C 1s XPS spectrum of the GO-SnO2 nanocomposites (Figure 3(e)) shows a significant increase in the intensity of C-C at 284.7 eV and the much smaller amount of oxygen-containing functionalities, indicat-ing a successful deoxygenation. It implies that the GO plays a key role for hybridization as an oxygen-provider to form SnO2 nanoparticles. In addition, as confirmed in Figure 2(a) and (c), the even distribution of SnO2 on graphene sheet indicates the presence of abundant oxygen-containing defect sites throughout the basal plane and edges of GO.11 Adding hydrazine in a GO-SnO2 composite suspension resulted in suppressed oxygen-related peaks, indicating the further reduction of GO (Figure 3(f)). However, at the end of the reduction, these peaks did not completely disappear. This is due to the screening effect of SnO2 nanoparticles attached to carbon sheets for an access of hydrazine to some partially negative epoxy and hydroxyl groups.29 Note that the XPS characterization results indeed indicate that simple mixing of a GO and SnO2 precursor solution not only leads to the reduction of GO to rGO, but also simultaneously results in the formation of SnO2 nanoparticles. Based on our observa-tion, the possible reaction mechanism30 is proposed as follow: Sn(CH3COO)2 + GO + H2O → SnO2 + rGO + 2CH3COOH.

Owing to the effect of strong pseudocapacitive character of SnO2 and the highly electro-conductive rGO support, hybrids are expected to be an excellent electrode material for supercapacitors. The performance of supercapacitor elec-trodes was analyzed using cyclic voltammetry (CV) and gavanostatic charge-discharge at room temperature. Typical three-electrode configuration was employed in 1 M H2SO4 electrolyte. Figure 4(a) and (b) show the CVs of the rGO and rGO-SnO2 electrodes, scanned at various scan rates in a potential range of 0 to 1.0 V. The voltammograms of the rGO and rGO-SnO2 exhibit a nearly rectangular shape, indicating a good capacitive behavior in the electrochemical supercapacitors.31 The higher capacitive current and redox peaks are observed in the CV for the rGO-SnO2, which are attributed to the efficient electric double layer capacitor performance of the carbon support and pseudocapacitive properties from redox reactions on tin oxide. The specific capacitances of electrode can be calculated according to the following equation:

Where I is the response current (A), V is the potential (V), v is the potential scan rate (mV/s), and m is the mass of the electroactive materials in the electrodes (g). The specific capacitance values of rGO and rGO-SnO2 electrodes at 50 mV/s are 241 and 389 F/g, respectively. Specific capacitance values of rGO and rGO-SnO2 at various scan rates are compared in Figure 4(c). The rGO-SnO2 electrode shows the enhanced capacitance performance for various scan rates. The rGO-SnO2 composites exhibit high specific capacitance values from 477 to 362 F/g as the scan rate increases from 5 to 100 mV/s and maintain 75.9% of its specific capacitance at a high rate. The capacitance retention of 52.9% from 411 to 218 F/g at the rates from 5 to 100 mV/ s of the rGO electrode is shown in Figure 4(d). These results are due to the enhanced electrolyte accessibility, facilitated charge pro-pagation along the hybrid electrode and the high electro-chemical stability for the redox transition of electrodes at higher scan rates.

Figure 4.CV curves of annealed rGO (a) and rGO-SnO2 nano-composites (b) at various scan rates in 1 M H2SO4 electrolyte. (c) Specific capacitance at different scan rate of rGO and rGO-SnO2 nanocomposites. (d) The capacitance retention ratio vs. scan rates of rGO and rGO-SnO2. (e) Gavanostatic charge-discharge curves of rGO and rGO-SnO2 in 1 M H2SO4 at 5 A/g. (f) Variation of the capacitance vs. cycle number of rGO and rGO-SnO2.

Figure 4(e) shows the galvanostatic charge-discharge curves of rGO and rGO-SnO2 electrodes at a current density of 5 A/g in a potential rage between 0 and 1.0 V. The galvanostatic curve of the hybrid is similar to that of rGO, which is linear and symmetrical, and shows no obvious iR drop, indicating low inter-resistance and instant I-V response of the hybrid.32 Also, from the charge-discharge curves, the specific capacitance values were calculated according to the following equation:

Where I is the constant current during discharge (A), and ΔV/Δt is the slope of the discharge curve (V/s). The specific capacitance of rGO and rGO-SnO2 nanocomposites at 5 A/g was about 190 and 362 F/g, respectively. To further investigate the cyclability of rGO and the rGO-SnO2 hybrid, the galvanostatic charge-discharge was performed for 500 cycles at 5 A/g (Figure 4(f)). During first 100 cycles the specific capacitance of rGO and rGO-SnO2 increased about 2.6 and 2.4%, respectively. Such an increase in capacitance can be ascribed to an activation process, in which the active materials become fully utilized in the electrochemical reaction.33 After the entire activation, rGO and rGO-SnO2 sustained the constant specific capacitance values, indicating the excellent cycle stability and high reversibility.

Conclusion

In summary, we have demonstrated the preparation of rGO-SnO2 hybrid materials through a straightforward hybridization method. By simple mixing of GO and a SnO2 precursor solution, a hybrid complex of GO-SnO2 was achieved. In this hybrid process, the oxygen functional groups in GO played a key role in conversion of Sn2+ into SnO2 nanoparticles. Furthermore, hybrids of rGO and SnO2 produced via a facile solution-processing route showed enhanced electrochemical properties. Owing to the syner-gistic effects from carbon and metal oxide components, those hybrids demonstrate remarkable performance as super-capacitor electrodes. Therefore, this work has shown that GO is an excellent nanoplatform for hybrid nanocomposites for electrode applications. Our hybrid strategy not only provides a high throughput and scalable synthetic route, but also illustrates the effectiveness of oxygen functionalized graphene sheet (GO) for the synthesis of functional nano-materials.