Journal of Colloid and Interface Science 417 (2014) 270–277

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science www.elsevier.com/locate/jcis

A general approach for fabrication of nitrogen-doped graphene sheets and its application in supercapacitors Dewei Wang, Yonggang Min ⇑, Youhai Yu, Bo Peng Institute of Advanced Materials, State Key Laboratory of Transient Optics and Photonics, Xi’an Institute of Optics and Precision Mechanics, Chinese Academy of Sciences, Xi’an 710119, People’s Republic of China

a r t i c l e

i n f o

Article history: Received 5 August 2013 Accepted 11 November 2013 Available online 21 November 2013 Keywords: Doped Graphene Graphene oxide Electrochemical performance Supercapacitors

a b s t r a c t In this paper, a general and efficient strategy has been developed to produce nitrogen-doped graphene sheets (NGs) based on hard and soft acids and bases (HSAB) theory. Under hydrothermal conditions, any salt with amphiprotic character have a strong tendency to hydrolysis, it is possible to provide reducing agent and nitrogen source simultaneously. It is worth noting that, NGs can be prepared under hydrothermal conditions by using some common ammonium salts with hard acid-soft base pairs as nitrogen-doping agents. The morphology, structure and composition of the as-prepared NGs were studied in detail. The results demonstrated that large amount of nitrogen was incorporated into the nanocarbon frameworks at the same time as the graphene oxide (GO) sheets were reduced. The electrochemical behavior of the synthesized NGs as supercapacitor electrodes was evaluated in a symmetric two-electrode cell configuration with 1 M H2SO4 as the electrolytes. It was found that the nitrogen groups making the as-prepared NGs exhibited remarkably enhanced electrochemical performance when used as electrode materials in supercapacitors. The supercapacitor based on the NGs exhibited a high specific capacitance of 242 F g1 at a current density of 1 A g1, and remains a relatively high capacitance even at a high current density. This work will put forward to understand and optimize heteroatom-doped graphene in energy storage systems. Ó 2013 Elsevier Inc. All rights reserved.

1. Introduction With the fast-growing energy demand in electrically powered vehicles and mobile electronics, great efforts have been made to exploit various energy-related devices and to improve the efficiency of conversion and storage [1,2]. Supercapacitors, also known as electrochemical capacitors, are the most promising candidates for various portable systems and automotive applications owing to their high power density, wide operating temperature and longer lifetime even in harsh conditions [3–6]. Since it was demonstrated and patented by General Electric in 1957, supercapacitors can be found a wide range of applications including, but not limited to, back-up power supply, portable electronics, and braking system [7–10]. Consequently, supercapacitors have attracted tremendous attention in recent years and became one of the most intense research focuses in the electrical energy storage field. As a unique two-dimensional carbon nanostructure, graphene has attracted intense interest for its promising applications in supercapacitor electrodes, mainly due to its superior electrical conductivity, exceptional large specific surface area, and excellent structural stability [11–14]. Ruoff and coworkers first demon⇑ Corresponding author. Fax: +86 29 88865829. E-mail address: [email protected] (Y. Min). 0021-9797/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcis.2013.11.021

strated that graphene based supercapacitors has a specific values of 135 F g1 and 99 F g1 in aqueous and organic electrolytes, respectively [15]. However, the specific energy and capacitance of graphene based supercapacitors is several orders of magnitude lower than that of batteries and fuel cells, which limits its adoption for numerous possible applications. Generally, the capacitive behavior of graphene electrode materials can be improved largely by introducing faradic redox reactions involved components (e.g. transition metal oxides, conducting polymers). Nonetheless, these multi-component graphene materials suffer from poor intrinsic conductivity of these oxide materials and less satisfactory cyclability. Moreover, elaborate procedures to fabricate electrode materials with metal oxide and/or polymer and graphene are complex and expensive to scale up for widespread commercialization. Alternatively, chemical doping with electron-donating or electronwithdrawing elements, such as N, B, and O, is a promising route to improve the electrochemical properties of the graphene-based materials [16–18]. The specific capacitance of the graphene electrode in aqueous electrolytes can be enhanced by incorporating heteroatoms such as N, B and O in the carbon framework because of the redox reactions between the electrolyte ions and the heteroatom-containing functional groups as well as the surface wettability, and electronic conductivity has been improved [19–21]. These unique features endow heteroatoms-doped

D. Wang et al. / Journal of Colloid and Interface Science 417 (2014) 270–277

graphene as a promising electrode for supercapacitors with high capacity, excellent rate capability, and long-term stability. It has been demonstrated that the heteroatoms in graphene derivatives, such as N and O, played a critical role in redox reactions to increase their charge-storage capacity because the oxygen functional groups such as quinone/hydroquinone and nitrogen functional groups such as pyridinic N and pyrrolic N groups can result in a pseudocapacitive response in aqueous electrolytes, and as a result, the electrochemical performances of graphene-based electrode materials have been improved significantly [22–24]. Recent studies have shown that the nitrogen-doped graphene electrode exhibit comparable capacitances to those of pseudocapacitors but still utilize the robust electrical double layers near the electrode surfaces [25]. Hence the nitrogen-doped graphene exhibited superior electrochemical performances to its non-incorporated counterparts when used as supercapacitor electrodes. For example, Jeong et al. reported the synthesis of nitrogen-doped graphene sheets (NGs) by using nitrogen plasma treatment of graphene and found that it exhibited excellent cycle life and high power capability for supercapacitors [25]. Feng and coworkers develop an efficient and facile strategy to fabricate highly crumpled NGs and the as-prepared NGs exhibit significant improvement in terms of various performance parameters of supercapacitors (e.g. capacity, rate, cycling) [26]. Very recently, Rao and coworkers carried out the first preparation of heavily nitrogenated graphene oxide by microwave synthesis [27]. The specific capacitance of the as-prepared NGs reached a value of 320 F g1 at 0.3 A g1 in a two-electrode system. Thereupon, nitrogen-doped graphene is a promising electrode material for supercapacitors because of its excellent capacitive behavior. Up to now, several strategies, including chemical-vapor deposition in the presence of N-containing precursors, arc discharge in N-containing atmosphere, nitrogen plasma treatment, and postsynthesis treatment have been proposed to introduce nitrogen into the graphene framework for high performance supercapacitors [20,28–31]. Of course, these processes suffered from either toxic precursors or sophisticated equipment, accompanied by high cost and labor-intensive preparations. Therefore, it will be of great significance to develop effective and facile methods to fabricate N-doped graphene materials for highenergy density supercapacitors. Recently, hydrothermal and solvothermal reactions have been used as effective approaches to fabricate NGs with a high nitrogen level for supercapacitors. For example, urea, hexamethylenetetramine and ammonia as well as organic amine have been studied for use as nitrogen-doping agents for graphene [32–36]. Even though the capacitance of NGs has been enhanced to some extent by these nitrogen-doping agents, there is still a great of interest to further improvement the capacitive of NGs by introducing new doping agents. Moreover, nitrogen-doping agents will react with oxygen functional groups first before introduce nitrogen into the graphene framework. In this process, nitrogen-doping agents also play a role as reducing agents, the nitrogen groups introduced in at the cost of oxygen groups. Inspired by those ideas, a general method has been developed to produce nitrogen-doped graphene sheets (NGs) based on HSAB theory. Different from the previous report process, there exists an additional reducing agent after hydrolysis. For example, ammonium oxalate is a typical amphoteric salt with hard acid–soft base pairs, which has a strong tendency to hydrolysis to produce ammonia and oxalic acid, where ammonia is an effective nitrogen doping agent and oxalic acid is a well-know reducing agent. Therefore, it is reasonable to produce high quality NGs in the presence of in situ produced reducing agent. Moreover, we found other ammonium salt with amphoteric character can also be used as an effective doping agent for the synthesis of NGs, which provide a new and general concept for the preparation of NGs.

271

In this work, we report nitrogen-doped graphene synthesized by a facile hydrothermal reaction with the assistance of amphoteric salt. During the hydrothermal reaction, amphoteric salt can gradually release NH3 that continually reacts with the oxygen functional groups of GO, which is favorable for doping of high-level nitrogen into graphene skeleton. Moreover, the in situ produced reducing agent can further remove the oxygen functional groups of GO. The high nitrogen content in this approach could be achieved under lower mass ratio of the salt and GO. Significantly, such NGs contained abundant of nitrogen component exhibited outstanding pseudocapacitive behavior in terms of highly specific capacitance, good rate capability, and excellent cyclability when used as electrode materials for supercapacitors. The purpose of this work is not only provide a simple and versatile approach to obtain NGs for high-performance supercapacitors but also put forward the presence of heteroatoms functional groups is an effective approach to maximum the performance of the NGs materials. 2. Experimental 2.1. GO preparation GO sheets were attained by oxidation and exfoliation of graphite powder using a well-known modified Hummers method as described previously [37]. GO (20 mg) and water (20 mL) was ultrasonicated in a beaker for 90 min, and then centrifuged at 4000 rpm for 10 min to remove any unexfoliated aggregations. In this way, a homogeneous GO aqueous dispersion (1 mg mL1) was obtained (see Fig. S1). 2.2. Synthesis of NGs For preparation of the NGs, 18 mL of GO suspensions and 150 mg of (NH4)2C2O4H2O were mixed in a breaker first. After being vigorously stirred for 5 min, the mixture was sealed into a 25 mL Teflon autoclave and maintained at 180 °C for 8 h. Then the autoclave was naturally cooled to room temperature and the as-prepared black product was washed with deionized water to remove residual inorganic compounds and freeze drying for further characterization. The same procedures are applied to synthesize other NGs samples when (NH4)2SO3H2O or NH4H2PO2H2O was used as nitrogen doping agent, respectively. For comparison, reduced GO (RGO) was also prepared under the same experimental parameters but without adding the nitrogen doping agents into the GO aqueous dispersion. 2.3. Characterization The phase structure of the products was measured by powder X-ray diffraction (XRD) experiments on a RigaKu D/max-RB diffractometer with Ni-filtered graphite–monochromatized Cu Ka radiation (k = 1.54056 Å). Transmission electron microscopy (TEM) studies were characterized using a transmission electron microscope (TEM, FEI) with an accelerating voltage of 100 kV. Samples were prepared by first dispersing the final powder in ethanol through ultrasonic treatment, and the dispersion was dropped on a carbon-coated copper grid, drying the samples in air for observation. Scanning electron microcopy (SEM) studies were carried on FEI, Quanta 200. The specific surface area was estimated by the Brunauer–Emmet–Teller (BET) method based on nitrogen absorption–desorption (Micromeritics ASAP 2020). X-Ray photoelectron spectrometer (XPS, X-Ray monochromatisation, Thermon Scientific) was carried out with Al Ka as the excitation source; the binding energies obtained in the XPS analysis were calibrated against the C 1s peak at 285.0 eV. All chemicals were of analytical grade

272

D. Wang et al. / Journal of Colloid and Interface Science 417 (2014) 270–277

and used without further purification. Ultrapure water (Millipore, 18.25 MX cm1) was used throughout. 2.4. Electrochemical measurements The supercapacitor test cells were fabricated with the two-electrode configuration. The fabrication of working electrodes was carried out as follows. NGs were mixed with acetylene black and poly (tetrafluoroethylene) at a mass ratio of 85:10:5 in ethanol to form homogeneous slurry. Finally, processed this slurry between two stainless steel mesh electrodes upon pressing, and dried at 60 °C overnight. Then two electrodes with identical or very close masses were separated by a glass fiber film and were sandwiched in a CR2016-type coin cell. Several drops of 1 M H2SO4 as the electrolytes were added before the cell was pressed to seal for measurement. Finally, the working electrodes were aged for 12 h to ensure good soakage of the electroactive material by the electrolyte [38]. Cyclic voltammetry (CV) tests were done between 0 and 1 V at different scan rates. The long-term galvanostatic charge–discharge was evaluated with a LAND CT2001A (Wuhan, China) multichannel galvanostat in the potential range of 0–1 V at a current density of 3 A g1. CV and chronopotentiometry tests were employed on an electrochemical workstation (CHI 660E, Shanghai, China). Electrochemical impedance spectroscopy (EIS) measurements were carried out by applying an AC voltage with 5 mV amplitude over a frequency range from 0.01 Hz to 100 kHz at the open circuit potential. The mass specific capacitances (Cs) were calculated by using the equations Cs = 2(IDt/mDV), where I is the applied discharge current (A), Dt is the discharging time (s), m is the mass of one electrode (g), and DV is the voltage drop upon discharging (V). Factor of 2 was used because the series capacitance was formed in a two-electrode system [39]. The mass of the electrode materials was determined using a high-precision microbalance (Sartorius, BT25s). The mass of NGs loaded on the working electrodes was around 3.0 mg of each. The current density is calculated based on the weight of active material on one electrode. The energy density (E) and the power density (P) of the devices were calculated by using the equations E = (1/8) CsDV2 and P = E/Dt, respectively. The RGO based device was also assembled under the same conditions for comparison. 3. Results and discussion GO are highly negatively charged result from numerous oxygen-containing functional groups attached to the surface and edge of carbon sheets [39]. They were readily exfoliated in water to yield a stable dispersions consisting mostly of single-layer sheets [41], and subsequent reduction of GO by (NH4)2C2O4H2O under hydrothermal conditions could result in the formation of NGs. As revealed by Fig. 1a, the well-defined and interconnected 3D porous networks with a number of hierarchical pores were observed clearly. From the high-magnification SEM image (Fig. 1b), it can be seen that the pore walls consist of thin layers of stacked graphene sheets with some corrugations and scrolling. TEM image shows that the as-prepared NGs possess several thin two-dimensional sheets with a large amount of wrinkles (Fig. 1c and d), which is consistent with the SEM image. On the contrary, RGO sample only consist of randomly aggregated sheets that are very thick (Fig. S2). The structure of pristine graphite, GO, RGO and NGs were further characterized by XRD. The XRD pattern of pristine graphite appeared a basal reflection (0 0 2) with the strong and sharp peak at 2h = 26.6° (d spacing = 3.347 Å) (Fig. S3a). Upon oxidation of pristine graphite, the typical diffraction peak of GO shifted negatively to the lower angle (2h = 11.7°, d spacing = 7.55 Å) (Fig. S3b). The

larger d spacing of GO than that of the pristine graphite was attributed to the formation of oxygen-containing functional groups and the interaction of water molecules and between the layers of the graphite [38,39]. The XRD pattern of the NGs, as shown in Fig. 2, exhibits a broad peak centered at 2h = 24.9°, corresponding to the graphitic (0 0 2) profile with an interlayer spacing of 3.57 Å. This value is much smaller than that of the GO precursor (7.55 Å), while still being higher than that of graphite (3.347 Å). The interlayer spacing of RGO is estimated to be 3.67 Å which is slightly larger than that of NGs, indicating that the GO is more effective reduction in the presence of (NH4)2C2O4H2O. XPS characterizations were further performed to analyze the elemental compositions and nitrogen bonding configurations in NGs. As shown in Figs. 3 and S4, for GO and RGO samples, only the peaks corresponding to C1s (284.5 eV) and O1s (531.8 eV) signals are detected, while an additional signal centered at 399.0 eV ascribed to N1s can be observed for NGs sample. Moreover, the O1s and C1s peaks of the carbons binding to oxygen of NGs became significantly weak and the C1s peak became much stronger as compared with that of GO, suggesting that the recovery of p-conjugated structure from GO sheets upon hydrothermal reduction. GO has a considerable degree of oxidation with four peaks located at the binding energies of 284.5, 285.2, 286.7, 287.3 and 288.5 eV, which could be assigned to the non-oxygenated ring C atoms (CAC, C@C), the C atoms in the hydroxyl (CAOH), epoxy/ether groups (CAO), the carbonyl C structure (C@O), and the carboxyl C structure (OAC@O), respectively [35,39,40]. In the case of NGs, the intensities of all of the C1s peaks of the carbons binding to oxygen and the peak of sp3 carbon, especially the peak of CAO (epoxy and alkoxy), decreased dramatically revealing that the most of the oxygen containing functional groups are successfully removed after hydrothermal reaction. The asymmetric N1s XPS peak of NGs (Fig. 2c) can be fitted to four peaks at 398.5 (pyridinic N), 399.6 (pyrrolic N), 401.3 eV (graphitic/quaternary N) and 405.2 eV (oxidized nitrogen) [25–30]. Both of pyridinic N and pyrrolic N were the main component in the prepared NGs. The calculated atomic percentage of N in NGs was 6.85 atom%. In addition, four distinct peaks in the O 1s spectrum revealed the presence of diverse oxygen groups in the NGs, as can be seen in Fig. 3d. These results suggest the p-conjugated graphene networks are partially restored and also the presence of abundant nitrogen and oxygen containing functional groups on surface of the NGs after hydrothermal reduction in the presence of (NH4)2C2O4H2O. Moreover, we found other amphoteric salt with hard acid–soft base pairs such as (NH4)2SO3H2O or NH4H2PO2H2O can be also used as an effective doping agent for the preparation of NGs (Figs. S5 and S6). In order to investigate the porous structure and surface area of the as-prepared NGs, N2 adsorption/desorption isotherms were carried out, as shown in Fig. S7. According to the IUPAC classification, this isotherm can be classified as type IV with H2 type hysteresis behavior with an evident hysteresis loop. The BET surface area and total pore volume are 476.46 m2 g1 and 0.3504 cm3 g1, respectively. The large specific surface areas and pore volume make NGs an idea active material for application in energy storage/conversion devices. In order to provide the most accurate measurement for practical application, the electrochemical performance of the NGs as electrode materials for supercapacitors were examined in a symmetrical two-electrode cell configuration using an aqueous H2SO4 solution (1 M) as the electrolyte [39]. Fig. 4a gives the representative cyclic voltammetry (CV) curves of the NGs with various sweep rates ranging from 5 to 50 mV s1. At a low scan rate of 5 mV s1, the NGs present well-symmetric and rectangular CV curves with obvious redox characteristic peaks, indicating a low resistance of NGs electrodes to mass transfer and good charge propagation of ions at the interfaces between the electrolyte and the NGs material [42]. With an increase in scan rate, the current response increased

D. Wang et al. / Journal of Colloid and Interface Science 417 (2014) 270–277

273

Fig. 1. SEM (a and b) and TEM (c and d) images of the as-obtained NGs after freeze drying.

Fig. 2. XRD patterns of the as-prepared NGs and RGO.

accordingly, without any significant changes in the shape of the CV curve, indicating a good rate performance. The rectangular and symmetric shape of the CVs was also observed at high scan rates of 200 mV s1 (Fig. S8), further supporting the suggestion of highly capacitive nature and rapid charge–discharge behavior. [43] For RGO supercapacitor, the appearance of humps in the CV profile indicated the combination of both double-layer capacitive from graphene and pseudocapacitive behavior from the oxidation/ reduction reactions of the oxygen containing functional groups (Fig. S9). However, there has an obvious distortion from rectangular shape when the scan rate increased to 50 mV s1 in the RGO

electrode (Fig. S10). Additionally, we found that the area surrounded by CV curves for NGs electrode is apparently bigger than that of the RGO electrode at the same san rate (Fig. S9), implying the higher specific capacitance of NGs electrode, which indicates a remarkable contribution of nitrogen doping in raising the capacitance of graphene. Fig. 4b shows the galvanostatic charge–discharge curves of the NGs electrode collected at different current densities. The symmetry of the charge and discharge characteristics and the almost constant slope of these curves further support that the NGs based capacitors have high electrochemical reversibility and excellent capacitive characteristics. Moreover, NGs electrode shows a longer charge–discharge time and lower IR drop at the same current density, implying a larger capacitance and super high rate performance than that of RGO, which agree well with the CV results (Fig. S11). The calculated specific capacitance of the NGs are 242, 240, 234, 230 and 224 F g1 at current densities of 1, 2, 3, 5 and 7 A g1, respectively. The present values are comparable and even higher than those of some typical carbon-based supercapacitors reported recently (Table S1). The relationship between the specific capacitance and the current density is also observed in Fig. 4c according to the galvanostatic charge and discharge results collected at different current densities. As shown in Fig. 4c, NG sheets show increased capacitances compared to that of RGO over a range of current density. The specific capacitance of the NGs is calculated to be 220 F g1 at the current density of 10 A g1, which is larger than that of RGO with the specific capacitance of 80 F g1, and is comparable to those values reported for laser reduction of graphite oxide films [44], and heteroatom-enriched electrospun carbon nanofibers based supercapacitors [45]. Importantly, the NGs exhibits excellent rate capability. A specific capacitance of 200 F g1 can be achieved at 20 A g1, and even at a high current density of 30 A g1, the specific capacitance can still remain at a

274

D. Wang et al. / Journal of Colloid and Interface Science 417 (2014) 270–277

Fig. 3. XPS spectra of NGs (a) wide scan; (b) C 1s; (c) N 1s and (d) O 1s spectra of the NGs.

high level of 162 F g1 with a good retention above 67% (Fig. 4c). On the contrary, RGO had poor performance at large current density. For example, when the current density was increased to 5 A g1, the specific capacitance was 130 F g1, accompanying with an obvious IR drop (Fig. S11). The long-term cycling stability of the supercapacitor was tested through a galvanostatic charge–discharge process at a moderate current density of 3 A g1, which is shown in Fig. 4d. The supercapacitor device still remains at 97.6% of the initial capacitance after 5000 continuous charge–discharge cycles, demonstrating its excellent long-term cycling stability. Fig. 4e shows the Ragone plots of the NGs and the RGO devices. The NGs supercapacitor showed significant enhancement the energy density at the same power density. It delivered an energy density of 8.4 W h kg1 at the power density of 250 W kg1. The RGO device delivered energy density as low as 5.7 W h kg1 at the same power density because of its low specific capacitance. More importantly, the energy density was relatively stable with the increase in the power density. The energy density reached up to 5.6 W h kg1 even at a power density as high as 75,000 W kg1, which was much higher than most of current commercial supercapacitors. Encouragingly, these values can meet the criteria (a power density of 5–10 kW kg1 and energy density of approximately 5 W h kg1) for next-generation supercapacitors [46,47]. Furthermore, in order to test its feasibility for practical application of the NGs-based supercapacitors, three fabricated devices were connected in series to power a red light-emitting diodes (LED, working voltage, 2 V). After charging at 5 A g1 for 25 s, the device could light the LED

for over 5 min, which can be seen in Fig. 4f. This result vividly demonstrates the potential of the fabricated supercapacitor device in energy storage. Electrochemical impedance spectroscopy (EIS) was further employed to monitor the electrochemical behavior of the electrode and the impedance characteristics. Fig. 5a reveals the Nyquist impedance plots of the NGs and RGO electrodes. At high frequencies, from the Nyquist plot, the intercept at the real part (Z0 ) is a combined resistance of the ionic resistance of the electrolyte, the intrinsic resistance of the substrate, and the contact resistance at the active material/current collector interface. The semicircle in the high–medium frequency range corresponds to the Faradaic charge transfer resistance (Rc). In the low frequency range, the Warburg tail is expected to occur at 45°, corresponding to the capacitor’s diffusive resistance of the electrolyte in the electrode pores and the ion diffusion in the host material [23,26,42–44]. The Nyquist plot is almost a vertical line, indicating a nearly ideal capacitive behavior of the EDLC. The inset in the Fig. 5a shows the expanded high frequency region of impedance. The electrode series resistance (Rs) was derived from the high frequency intersection of the Nyquist plot in the real axis. It was found that the NGs had smaller series resistance to RGO, manifesting the good conductivity of the electrolyte and very low internal resistance of the NGs electrode. The NGs electrode exhibited almost vertical line at the low frequency region, indicating the desired capacitive behavior. In addition, the semi-arc in the impedance spectrum for NGs electrode is much smaller than RGO electrode, indicating the fast

D. Wang et al. / Journal of Colloid and Interface Science 417 (2014) 270–277

275

Fig. 4. (a) Cyclic voltammogram of the NGs electrode at the scan rate from 5 to 50 mV s1; (b) charge–discharge curves of the NGs electrode at different current densities; (c) plot of specific capacitance as a function of current density; (d) capacitance retention of the NGs electrode as a function of cycle number; (e) Ragone plot (energy density vs. power density) of the NGs and RGO supercapacitors; (f) a red LED were driven by three fabricated NGs supercapacitors in series. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

charge transfer process of the NGs electrode. The frequency-dependent response of supercapacitors can be analyzed from the EIS spectra. Fig. 5b shows the dependence of the phase angle with the frequency. For frequencies up to 0.01 Hz, the phase angle is very close to 90 °, which evidences the good capacitive property of the NGs supercapacitor. Comparing the frequencies at which

phase angle of 45° (f0) [48], we clearly see that the NGs device demonstrate a much faster frequency response with f0 = 1 Hz than RGO with f0 = 0.12 Hz. The faster performance of the NGs supercapacitor correlates with its better capacitance retention at high sweep rates in the CV measurements (Fig. 4a) or higher current densities in galvanostatic charge–discharge tests (Fig. 4b). The

276

D. Wang et al. / Journal of Colloid and Interface Science 417 (2014) 270–277

Fig. 5. (a) Nyquist plot of NGs and RGO supercapacitors; Inset is the enlarged plot of the high-frequency regions and (b) evolution of the Bode phase angle plot vs. frequency analyzed from the EIS spectra.

response time (s = 1/f0), was estimated to be 1 s for the NGs electrode, which was smaller than that (8.3 s) of the RGO electrode. The increase in the response time was due to a larger Faradaic charge transfer resistance in the RGO electrodes. The remarkable performance of NGs could be attributed to the appropriate combination of high electrochemical activity of graphene materials by heteroatoms doping: the co-existing abundant nitrogen and oxygen containing functional groups played as important in the charge–discharge process. On the one hand, different N-types in graphene not only improve the wettability of NGs in electrolyte, but also decrease the intrinsic resistance of graphene to improve the capacitive performance at high current density [23,26]. Besides, pyridinic and pyrrolic N are electrochemically active functional groups, they can participate in redox reactions in alkaline aqueous solution to give additional pseudocapacitance [16,19,20,23–26]. On the other hand, for RGO supercapacitor cannot able to remove all of the oxygen groups, a lot of oxygen functional groups (especially AOH, C@O and ACOOH), which are electrochemically active for electrochemical reactions will further boost the pseudocapacitance [11,45]. Moreover, the continuous 3D network nanostructure, which provides a high accessible surface area and allows abundant adsorption of ions as well as efficient ion migration and charge transportation [44]. Thus, an exceptional specific capacitance can be realized, which is much higher than that of the reduced graphene materials. Overall, the NGs possess a significantly improved capacitance compared with RGO, and also exhibit good rate and stability performance.

faster frequency response with f0 = 1 Hz and a smaller response time constant (1 s). The superior electrochemical performance of the NGs is probably attributed to its combining the unique nanostructure and abundant electrochemically active functional groups on the graphene sheets. In virtue of their superior electrochemical performance, they will be promising electrode materials for highperformance supercapacitors applications. Furthermore, the present method shows a great promise towards the design and fabrication of chemical doped graphene materials for high performance energy storage systems. Acknowledgments The authors would like to acknowledge the financial support of the Director Foundation of Xi’an Institute of Optics and Precision Mechanics (No. Y255F81ZZ0) and Western Light Program of Chinese Academy of Sciences (No. Y329181213). Appendix A. Supplementary material Additional TEM and SEM images, AFM image, XRD analysis, galvanostatic charge–discharge curves, cyclic voltammetry (CV) curves, and Bode phase angle plot of the NGs device. Table that summarizes the specific capacitance of some carbonaceous materials in the recent literatures. Supplementary data associated with this article can be found, in the online version, at http:// dx.doi.org/10.1016/j.jcis.2013.11.021. References

4. Conclusion In summary, NGs with a nitrogen content of 6.85 atom% was prepared by a facile and effective hydrothermal method. After freeze drying, NGs could be obtained, which could be used as electrode materials for supercapacitors. The electrochemical properties of the NGs were measured by cyclic voltammogram and galvanostatic charge–discharge studies. Results show that the NGs exhibits excellent electrochemical performance as an electrode for supercapacitors in terms of specific capacitance (242 F g1 at the current density of 1 A g1), rate capability (67% capacitance retention at current densities up to 30 A g1), cyclability (97.6% of the initial capacitance after 5000 cycles), and energy density (8.4 W h kg1 at the power density of 250 W kg1). Additionally, the NGs have

[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15]

L. Dai, D.W. Chang, J.B. Baek, W. Lu, Small 8 (2012) 1130. N.G. Sahoo, Y.Z. Pan, L. Li, S.H. Chan, Adv. Mater. 24 (2012) 4203. E. Frackowiak, F. Beguin, Carbon 39 (2001) 937. C. Meng, C. Liu, L. Chen, C. Hu, S. Fan, Nano Lett. 10 (2010) 4025. B.E. Conway, Electrochemical Supercapacitors, Kluwer Academic/Plenum Publishers, New York, 1999. M. Winter, R.J. Brodd, Chem. Rev. 104 (2004) 4245. J.R. Miller, P. Simon, Science 321 (2008) 651. A.G. Pandolfo, A.F. Hollenkamp, J. Power Sourc. 157 (2006) 11. P.J. Hall, M. Mirzaeian, S.I. Fletcher, F.B. Sillars, A.J. Rennie, G.O. Shitta-Bey, G. Wilson, A. Cruden, R. Carter, Energy Environ. Sci. 3 (2010) 1238. H. Jiang, J. Ma, C.Z. Li, Adv. Mater. 24 (2012) 4197. Y. Huang, J.J. Liang, Y.S. Chen, Small 8 (2012) 1805. Y.Q. Sun, Q.O. Wu, G.Q. Shi, Energy Environ. Sci. 4 (2011) 1113. L.L. Zhang, R. Zhou, X.S. Zhao, J. Mater. Chem. 20 (2010) 5983. T. Kuila, A.K. Mishra, P. Khanra, N.H. Kim, J.H. Lee, Nanoscale 5 (2013) 52. M.D. Stoller, S.J. Park, Y.W. Zhu, J.H. An, R.S. Ruoff, Nano Lett. 8 (2008) 3498.

D. Wang et al. / Journal of Colloid and Interface Science 417 (2014) 270–277 [16] Y. Zhao, C.G. Hu, Y. Hu, H.H. Cheng, G.Q. Shi, L.T. Qu, Angew. Chem., Int. Ed. 51 (2012) 11371. [17] Z.S. Wu, A. Winter, L. Chen, Y. Sun, A. Turchanin, X.L. Feng, K. Müllen, Adv. Mater. 24 (2012) 5130. [18] D.W. Wang, F. Li, Z.G. Chen, G.Q. Lu, H.M. Cheng, Chem. Mater. 20 (2008) 7195. [19] C.O. Ania, V. Khomenko, E. Raymundo-Pinero, J.B. Parra, F. Beguin, Adv. Funct. Mater. 17 (2007) 1828. [20] H.B. Wang, T. Maiyalagan, X. Wang, ACS Catal. 2 (2012) 781. [21] J. Han, L.L. Zhang, S. Lee, J. Oh, K.S. Lee, J.R. Potts, J.Y. Ji, X. Zhao, R.S. Ruoff, S. Park, ACS Nano 7 (2013) 19. [22] E. Raymundo-Pinero, F. Leroux, F. Beguin, Adv. Mater. 18 (2006) 1877. [23] L.F. Chen, X.D. Zhang, H.W. Liang, M.G. Kong, Q.F. Guan, P. Chen, Z.Y. Wu, S.H. Yu, ACS Nano 6 (2012) 7092. [24] L. Zhao, L.Z. Fan, M.Q. Zhou, H. Guan, S.Y. Qiao, M. Antonietti, M.M. Titirici, Adv. Mater. 22 (2010) 5202. [25] H.M. Jeong, J.W. Lee, W.H. Shin, Y.J. Choi, H.J. Shin, J.K. Kang, J.W. Choi, Nano Lett. 11 (2011) 2472. [26] Z.H. Wen, X.C. Wang, S. Mao, Z. Bo, H. Kim, S.M. Cui, G.H. Lu, X.L. Feng, J.H. Chen, Adv. Mater. 24 (2012) 5610. [27] K. Gopalakrishnan, A. Govindaraj, C.N.R. Rao, J. Mater. Chem. A 1 (2013) 7563. [28] L. Qu, Y. Liu, J.B. Baek, L. Dai, ACS Nano 4 (2010) 1321. [29] L.S. Panchakarla, K.S. Subrahmanyam, S.K. Saha, A. Govindaraj, H.R. Krishnamurthy, U.V. Waghmare, C.N.R. Rao, Adv. Mater. 21 (2009) 4726. [30] Y. Wang, Y. Sha, D.W. Matson, J. Li, Y. Lin, ACS Nano 4 (2010) 1790. [31] Z.H. Sheng, L. Shao, J.J. Chen, W.J. Bao, F.B. Wang, X.H. Xia, ACS Nano 5 (2011) 4350.

277

[32] Z.B. Lei, L. Lu, X.S. Zhao, Energy Environ. Sci. 5 (2012) 6391. [33] J.W. Lee, J.M. Ko, J.D. Kim, Electrochim. Acta 85 (2012) 459. [34] D. Li, M.B. Muller, S. Gilje, R.B. Kaner, G.G. Wallace, Nat. Nanotech. 3 (2008) 101. [35] L. Lai, L. Chen, D. Zhan, L. Sun, J. Liu, S.H. Lim, C.K. Poh, Z. Shen, J. Lin, Carbon 49 (2011) 3250. [36] P. Chen, J.J. Yang, S.S. Li, Z. Wang, T.Y. Xiao, Y.H. Qian, S.H. Yu, Nano Energy 2 (2013) 249. [37] W.S. Hummers, R.E. Offeman, J. Am. Chem. Soc. 80 (1958) 1339. [38] D.W. Wang, Q.H. Wang, T.M. Wang, Inorg. Chem. 50 (2011) 6482. [39] M.D. Stoller, R.S. Ruoff, Energy Environ. Sci. 3 (2010) 1294. [40] V. Singh, D. Joung, L. Zhai, S. Das, S. Khondaker, S. Seal, Prog. Mater Sci. 56 (2011) 1178. [41] M.J. Allen, V.C. Tung, R.B. Kaner, Chem. Rev. 110 (2010) 132. [42] Y. Zhu, S. Murali, M.D. Stoller, K.J. Ganesh, W. Cai, P.J. Ferreira, A. Pirkle, R.M. Wallace, K.A. Cychosz, M. Thommes, D. Su, E.A. Stach, R.S. Ruoff, Science 332 (2011) 1537. [43] L.L. Zhang, X.S. Zhao XS, Chem. Soc. Rev. 38 (2009) 2520. [44] M.F. El-Kady, V. Strong, S. Dubin, R.B. Kaner, Science 335 (2012) 1326. [45] C. Ma, Y. Song, J.L. Shi, D.Q. Zhang, Q.G. Guo, L. Liu, J. Colloid Interface Sci. 395 (2013) 217. [46] D.W. Wang, Y.Q. Li, Q.H. Wang, T.M. Wang, Eur. J. Inorg. Chem. (2012) 628. [47] X. Zhao, B.M. Sanchez, P. Dobson, P. Grant, Nanoscale 3 (2011) 839. [48] V. Presser, L.F. Zhang, J.J. Niu, J. McDonough, C. Perez, H. Fong, Y. Gogotsi, Adv. Energy Mater. 1 (2011) 423.