PCCP View Article Online
Published on 30 April 2015. Downloaded by North Dakota State University on 25/05/2015 07:13:44.
PAPER
Cite this: Phys. Chem. Chem. Phys., 2015, 17, 14028
View Journal | View Issue
Insight into the formation mechanism of graphene quantum dots and the size effect on their electrochemical behaviors† Yonghuan Liu,ab Rutao Wang,a Junwei Langa and Xingbin Yan*a To study the formation mechanism and influencing factors of graphene quantum dots (GQDs), GQDs with different average sizes were prepared using a modified hydrothermal method with hydrogen peroxide (H2O2) as an etching agent and ammonia as an assistant. It is found that size-controlled GQDs were prepared by adjusting the amount of ammonia and porous reduced graphene oxide (PRGO) debris can be synthesized by reducing the hydrothermal reaction time. Structural changes of final products were mainly attributed to the changes in the etching ability of the hydroxyl radical (OH ) against the reduction ability of the hydroxyl group (OH) in different alkaline environments regulated by ammonia.
Received 2nd February 2015, Accepted 30th April 2015 DOI: 10.1039/c5cp00646e
Furthermore, we studied the electrochemical properties of GQDs and PRGO. The results showed that the specific capacitance of all samples increases linearly with the size and the smallest GQDs can work at the highest scan rate of as high as 5000 V s1 with an ultra-fast power response (t0 = 63.3 ms). Thus, these findings elucidate the formation mechanism of GQDs and demonstrate that GQDs are applicable
www.rsc.org/pccp
in microelectronic devices with high power response requirements.
1. Introduction Graphene quantum dots (GQDs), which are graphene nanosheets with lateral dimensions of o100 nm, exhibit numerous novel chemical and physical properties because of quantum confinement and edge effects associated with graphene and QDs.1–5 To date, GQDs have demonstrated great potential for various applications, such as bio-imaging devices,6 sensors,7 catalysts,8 light-emitting diodes,9 fuel cells,10 lithium ion batteries,11 and photovoltaic devices.12 The primary approaches for synthesizing GQDs are the bottom-up and top-down methods.4,5 The former mainly involves pyrolysis of polycyclic aromatic hydrocarbons for a subsequent selfassembly or cyclodehydrogenation of polyphenylene precursors, but often entails a serious drawbacks concerning low purity;13 by contrast, the latter primarily depends on chemical cutting of large graphene or graphene oxide (GO) into nanometer-sized pieces (GQDs),5,14 which often requires abundant strong acids or organic solvents that are very difficult to remove. Recently, the preparation of GQDs by cutting GO sheets in a hydrothermal ambience by using hydrogen peroxide (H2O2) and ammonia has gained much a
Laboratory of Clean Energy Chemistry and Materials, State Key Laboratory of Solid Lubrication, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, Lanzhou 730000, P. R. China. E-mail: [email protected]; Fax: +86 931 4968055; Tel: +86 931 4968055 b Graduate University of Chinese Academy of Sciences, Beijing 100080, P. R. China † Electronic supplementary information (ESI) available. See DOI: 10.1039/ c5cp00646e
14028 | Phys. Chem. Chem. Phys., 2015, 17, 14028--14035
attention owing to the high purity and strong photoluminescence of as-produced GQDs, as well as the simplicity of the preparation process.5,15 However, the formation mechanism of GQDs, especially the respective roles of H2O2 and ammonia, remains unclear. Moreover, as-prepared GQDs showed a simple size range of only several nanometers, thereby limiting the applications of GQDs. Supercapacitors are promising candidates in energy storage systems because of their high power density, fast charge/ discharge rate and excellent cycle life.16–18 Our previous reports first demonstrated GQDs as promising electrode materials to build micro-supercapacitors with a superior power response and rate capability.16,17 Subsequently, Zhu et al. reported that adding GQDs can remarkably enhance the rate capability of RuO2 nanoparticles.18 Hu et al. also reported that depositing GQDs on carbon nanotubes (CNTs) can significantly increase the capacitance of CNT arrays.19 Furthermore, Chen et al. also reported the enhanced capacitance of 3D graphene by depositing GQDs.20 These results indicated the great potential of GQDs as supercapacitors. However, no report is available regarding the intrinsic electrochemical characteristics of GQDs. In the current research, we demonstrated a modified hydrothermal approach to prepare GQDs with controllable sizes by regulating the ratio of ammonia to H2O2. We systematically studied the formation mechanism and influencing factors. The results revealed that the average size of GQDs is strongly dependent upon the amount of ammonia, suggesting that aside from H2O2, ammonia also plays an important role in GQD formation.
This journal is © the Owner Societies 2015
View Article Online
Published on 30 April 2015. Downloaded by North Dakota State University on 25/05/2015 07:13:44.
Paper
PCCP
GQDs with different sizes can be prepared by adjusting the amount of ammonia in H2O2, and porous reduced graphene oxide (PRGO) can be synthesized by shortening the hydrothermal reaction time. The structure changes were mainly attributed to the alteration of the etching ability of the hydroxyl radical (OH ) against the reduction ability of the hydroxyl group (OH) in different alkaline environments. Based on this, we studied the electrochemical properties of the final GQDs and PRGO with different sizes. The specific capacitance as a function of electrode material size showed a nearly linear increase. GQDs with the smallest size exhibited an ultra-high scan rate of 5000 V s1, an ultra-fast power response (t0 = 63.3 ms) and the lowest specific capacitance of 3.3 F g1.
and RGO-T (reduced graphene oxide prepared by thermal reduction at 300 1C for 2 h) as raw materials in place of GO. 2.2
Morphological analysis of the samples was conducted on a transmission electron microscope (TEM; JEOL 2100 FEG) and an atomic force microscope (AFM; Veeco, Nanoscope IIIA). The chemical bonds and components were investigated using a Fourier transform infrared spectrometer (FTIR; IFS120HR) and an X-ray photoelectron spectrometer (XPS; ESCALAB250x). Raman spectra were measured on a Raman spectrometer (JY-HR800) with an excitation wavelength of 532 nm. The fluorescence spectra were obtained using a fluorescence spectrophotometer (PerkinElmer Lambda 35). 2.3
2. Experimental section 2.1
Preparation of GQDs and PRGO with different sizes
GO sheets were prepared from natural graphite by using a modified Hummers method, as described previously.21,22 GQDs and PRGO with different sizes were then prepared using a hydrothermal synthesis method with H2O2 as an etching agent and ammonia as an assistant. In a typical synthesis, 0.5 ml of ammonia (28 wt%) and 6 ml of H2O2 (30 wt%) were added in 40 ml of GO aqueous dispersion (2 mg ml1) with the aid of stirring. The resulting dispersion was diluted to 160 ml, then transferred to a 200 ml Teflon autoclave and maintained at 180 1C for 8 h. After cooling to room temperature, a light yellow supernatant was obtained by a filtration process through a porous cellulose ester membrane (220 nm). The supernatant was heated to remove unreacted ammonia and H2O2. The yellow powder (denoted as GQDs-1) was finally collected by freeze-drying, and the yield was about 25%. Other samples of GQDs with large sizes and PRGO were produced by changing either the volume ratio of ammonia to H2O2 or the hydrothermal reaction time at the same temperature. The experimental parameters of these samples are listed in Table 1. To analyze the formation mechanism of GQDs, a series of contrast experiments were conducted, including the synthesis with only H2O2 in GO dispersion (without ammonia), and the synthesis using sodium hydroxide (NaOH) to replace ammonia, as well as the experiments by using N-RGO (reduced graphene oxide obtained using hydrazine reduction with the aid of reflux at 95 1C for 1 h)
Table 1 Reaction conditions for synthesizing GQD series and PRGO, the mean size or the size range, and the contents of N and O in the final samples
Ammonia Mean volume H 2 O2 Reaction size Samples (ml) volume (ml) time (h) (nm) GQDs-1 GQDs-2 GQDs-3 PRGO-1 PRGO-2 PRGO-3
0.5 2 3 3 3 3
6 6 6 6 6 6
8 8 8 6 4 2
This journal is © the Owner Societies 2015
Yield N O (%) (at%) (at%)
4.8 25 5.80 13.1 37 3.31 34.7 490 2.76 100–150 4.43 200–300 4.02 500–700 7.28
30.00 28.28 25.57 30.06 26.49 31.00
Structural characterization
Electrode preparation and electrochemical measurements
To study the intrinsic electrochemical performance of the as-prepared samples, the working electrodes were fabricated by dipping GQDs or PRGO dispersion with determined concentration on polished glassy carbon electrodes (5 mm in diameter) without additives. The active material on each glassy carbon electrode, which was about 0.0185 mg, was calculated using the calibrated concentration and the adjusted volume of each size sample. After drying at 60 1C, the electrode was coated with 8 ml of naphthol solution (5 wt%) to prevent the dissolution of the active material in the electrolyte. The electrode was then dried at 60 1C for 1 h. Cyclic voltammetry (CV), galvanostatic charge/discharge (GCD), and electrochemical impedance spectroscopy (EIS) measurements were investigated in a three-electrode system with a saturated calomel electrode (SCE) as a reference electrode, a platinum foil (with area of 1 cm2) as a counter electrode and Na2SO4 solution (1 M L1) as an electrolyte. CV and GCD tests were conducted in a potential window of 0.4 V to 0.6 V (vs. SCE). EIS measurements were performed in a frequency range of 0.01 Hz to 100 kHz with 5 mV ac amplitude. All tests were conducted on an electrochemical workstation (CHI660D, Shanghai, China) at room temperature. The specific capacitance was calculated from the GCD curves based on the following equation: C = IDt/mDV (F g1) where I is the current density (A), Dt is the discharge time (s), DV is the potential window of discharging (V), and m is the active material mass.
3. Results and discussion GQDs with different sizes were successfully prepared in this research. Fig. 2a and the inset show that GQDs (GQDs-1) with a homogeneous size (a mean size of 4.8 nm) were synthesized. The inset also depicts the HRTEM image of GQDs-1 with a relatively high crystallinity. The AFM image of GQDs-1 illustrated a uniform size distribution, and the inset shows that their topographic heights are mostly o2 nm, revealing one to three layers (Fig. S3, ESI†).
Phys. Chem. Chem. Phys., 2015, 17, 14028--14035 | 14029
View Article Online
Published on 30 April 2015. Downloaded by North Dakota State University on 25/05/2015 07:13:44.
PCCP
Fig. 1 Schematic of the formation processes of GQDs: (a) a hydrothermal process by adding H2O2 only; (b) a hydrothermal process by adding both ammonia and H2O2.
To understand the formation processes and the mechanism of GQDs, H2O2 (3 ml) was added to the GO (80 mg) dispersion. As shown in the route Fig. 1a, we can roughly observe the etching processes of GO by using H2O2. Generally, H2O2 may be dissociated into hydroxyl radicals (OH ) under a hydrothermal process at 180 1C, and the as-generated OH was considered as one of the most powerful oxidizing species resembling chemical scissors.4,22,23 Thus, OH can react with the defect sites of GO, which include abundant oxygen-containing functional groups, such as OH, C–O–C and CQO/COOH on their surfaces. Therefore, the large GO sheets are broken down into small fragments/debris along with the release of CO2.24,25 The generated GO fragments reacting with OH again are converted to smaller debris and CO2. Ultimately, with the increase of etching time, the GO debris can thoroughly transform into CO2, CO, and H2O.24,25 It is visually seen that the color of the mixture gradually faded and disappeared with an etching time of 2 h. (Fig. S1a–e, ESI†), revealing the high etching ability of H2O2. The TEM images of the samples etched by using only H2O2 for different time (Fig. S2d–f, ESI†) showed that the GO sheets were gradually rived off, however, GQDs infrequently appeared.
Paper
However, tiny-sized GQDs were prepared after simultaneously adding a small amount of ammonia (0.5 ml) and H2O2 (3 ml) to GO dispersion under the same conditions as shown in the route of Fig. 1b. Our analysis indicated that OH can act as an initiator, similar to the role of Fe2+ in the photo-Fenton reaction, to induce H2O2 and decompose into more OH in alkaline medium.4,26,27 The presence of GO fragments and GQDs (Fig. S2c, ESI†) in the reaction system even after the increase in reaction time can be attributed to the ammonia atmosphere. To our knowledge, GO suspension can be partly reduced under alkaline conditions at a moderate temperature.28,29 Therefore, the as-produced small GO fragments and GQDs can be reduced and protected in the reaction system, thus the tiny-sized GQDs (GQDs-1) were obtained by removal from the GO (Fig. S2a, ESI†). As a verification, a reaction was hardly observed between H2O2 and N-RGO, as well as RGO-T (to replace GO) at 180 1C for 8 h (Fig. S1g and h, ESI†). Moreover, by adding NaOH to replace ammonia with a similar pH value at 180 1C for 8 h, we can achieve a similar result, in which a brown dispersion can be obtained by filtration (Figs. S1f and b, ESI†). As the volume of ammonia increased, larger GQDs (GQDs-2) with uniform size distribution (average size of 13.1 nm) and GQDs with a mean size of 34.7 nm (GQDs-3) were obtained (Fig. 2b and c). The inset shows the corresponding size distribution of GQDs-2 and GQDs-3. The preparation of larger GQDs benefits from the addition of more ammonia, which results in more rapid OH release and more active participation in reacting with GO at the initial stage. Meanwhile, the oxygencontaining functional groups of GO can react with ammonia to generate amide and develop N-doping products.30–32 Based on the above observations, we can naturally summarize that ammonia mainly plays a role in three aspects: first, OH can facilitate decomposition of H2O2 into more OH ; second, OH can react with oxygen-containing functional groups of residual fragments and GQDs to passivate their surfaces and play the role of a ‘‘protector’’; third, the volume of ammonia determines the structural changes in the etching ability of OH against the reduction ability of OH. The total reaction processes under ammonia conditions may be presented as follows:
Fig. 2 TEM images of the as-prepared samples with different sizes: (a) GQDs-1 and the inset showing its HRTEM, (b) GQDs-2, (c) GQDs-3, (d) PRGO-1, (e) PRGO-2, and (f) PRGO-3. The inset shows the corresponding size distribution for GQDs.
14030 | Phys. Chem. Chem. Phys., 2015, 17, 14028--14035
This journal is © the Owner Societies 2015
View Article Online
Published on 30 April 2015. Downloaded by North Dakota State University on 25/05/2015 07:13:44.
Paper
PCCP
H2O2 + H2O - H3O+ + HO2
(1)
H3O+ + OH - 2H2O
(2)
H2O2 + HO2 - HO2 + HO + OH
(3)
HO + C - CO2
(4)
Based on the above discussion, we proposed that the reaction processes in the ammonia medium are those as shown in the route of Fig. 1b. Moreover, by reducing the reaction time, the larger-sized samples of the PRGO debris (including PRGO-1 with a size range of 100–150 nm, PRGO-2 with a size range of 200–300 nm, and PRGO-3 with a size range of 500–700 nm; Table 1) were successfully prepared (Fig. 2d–f). We can only show the size ranges of these samples because their irregular porous structure complicates the evaluation of their mean sizes. Fig. 3a shows the Raman spectra of GQDs with different sizes (denoted GQD series) and PRGO debris. Two unconspicuous peaks existed at around 1360 and 1600 cm1, corresponding to the D and G bands of the GQD series, respectively. The abnormal phenomenon is mainly attributed to the size effect and strong fluorescence disturbance of the Raman characterization.31,33 However, with an increase in sample size, the D and G bands gradually appeared and exhibited characteristic peaks of a graphene-based material for PRGO-3. The FTIR spectra showed similar chemical species for GQDs and PRGO debris samples of different sizes (Fig. 3b). The spectra indicated the presence of oxygen functional groups, including C–O (n C–O at 1110 cm1), C–OH (n C–O at 1394 cm1), CQO mainly in carbonyl moieties (n CQO at 1645 cm1) and the
H-bond associated OH (a broad band at around 3470 cm1). In addition, the peak at 3180 cm1 was assigned to the N–H stretching of the amine group, indicating a successful reaction between ammonia and GO.30,34 The difference among the functional group peak centers of all the samples was caused by their diverse surrounding environments. For comparison, the FTIR spectra of GO, RGO-H obtained by hydrothermal treatment of GO at 180 1C for 8 h without adding ammonia, N-RGO and RGO-T were investigated (Fig. S4, ESI†). A relatively weak peak at about 1630 cm1 (n CQO) was found for RGO-H and RGO-T compared with GO and a peak at about 3180 cm1 (n N–H) was observed for N-RGO after reduction, which can explain the scarce reaction between H2O2 and RGO (including N-RGO and RGO-T). In view of the rich oxygencontaining functional groups of final products, the GQD series demonstrates an excellent stability in water without any change for three months. XPS measurements were performed to determine the composition of the GQD series and PRGO debris with different sizes (Fig. 3c). The XPS survey spectra of all the products showed similar element compositions, such as the predominant graphitic C1s peak at about 284 eV, the O1s peak at about 532 eV and the N1s peak at about 400 eV. Table 1 summarizes the contents of O and N elements, exhibiting similar results with relatively high values in all the samples. Briefly, the GQDs-1 sample was selected as a model for further analysis. The highresolution C1s XPS spectra of the GQDs-1 can be divided into five components (Fig. 3d), corresponding to the signals of C–C (284.7 eV), C–N (286.1 eV), C–O (286.6 eV), CQO (288.5 eV), and O–CQO (289.1 eV). The oxygen species of the GQDs-1 can be analyzed from the high-resolution O1s XPS spectra (Fig. 3e),
Fig. 3 (a) Raman spectra, (b) FTIR spectra, and (c) XPS patterns of GQDs and PRGO debris with different sizes. (d) C1s, (e) O1s, and (f) N1s spectra of GQDs-1.
This journal is © the Owner Societies 2015
Phys. Chem. Chem. Phys., 2015, 17, 14028--14035 | 14031
View Article Online
Published on 30 April 2015. Downloaded by North Dakota State University on 25/05/2015 07:13:44.
PCCP
which revealed three components corresponding to quinine (531.3 eV), CQO (531.8 eV), and C–O (533.2 eV). Fig. 3f shows the high-resolution N1s XPS spectra, consisting of amide (399.8 eV), pyridine-like (401.4 eV), and pyrrolic (402.1 eV) N atoms.30,31 Based on the above results, we roughly concluded the successfully introduced N atoms in the as-prepared GQDs series and PRGO debris. The main chemical groups on their surfaces were OH, epoxy/ether, CQO, and CO–NH2. GQDs have demonstrated promising properties for photoelectric applications. To our knowledge, the size, shape, and function group can contribute to the energy gaps and the fluorescence property.35–37 In our case, the chemical groups were similar for the series of GQDs, with an increase in size for GQDs-1 to GQDs-3. The photoluminescence (PL) spectra of the GQDs series (Fig. S5a–c, ESI†) showed that the as-prepared GQDs exhibited an excitation dependent PL behavior. When the excitation wavelength changed from 360 nm to 480 nm, the PL peak shifted from 480 nm to 530 nm. PL intensity initially increased and then decreased, reaching the strongest emission at an excitation wavelength of 440 nm. Importantly, the PL spectra of the GQDs series (GQDs-1, GQDs-2, and GQDs-3) at the same excitation wavelength of 440 nm showed that the PL intensity gradually decreased (Fig. S5d, ESI†), and the PL peak center gradually red shifted with the increase of the GQDs size (the peak centers of 504 nm, 512 nm, and 518 nm for GQDs-1, GQDs-2, and GQDs-3, respectively). Generally, this red shift behavior is mainly ascribed to the decrease of band-gaps.36,37,39 Compared with the size and functional groups, the band gaps (HOMO–LUMO) are relatively insensitive to the shape, and the dependence of the band gaps on size shows 1/L2, approximately.38 Therefore, this red shift phenomenon mainly resulted from the increase of the sample size. In our case, the PL
Paper
behavior almost disappeared for the large sized samples of PRGO debris owing to their exceeding size. The photographs of different size products in the aqueous solution captured under white light and 365 nm UV light (Fig. S6, ESI†) clearly showed that the fluorescence color gradually receded; moreover, a negligible fluorescence color appeared for the PRGO debris, which is mainly ascribed to the residual GQDs (Fig. S2c, ESI†). Fig. 4a shows the CV curves of the smallest sized product (GQDs-1) in a potential range of 0.4 V to 0.6 V at various scan rates from 1 V s1 to 500 V s1. The CV curve area increased gradually with the increase of the scan rate, and the CV curve at a scan rate of 500 V s1 still maintained the rectangular shape, exhibiting an ideal capacitive behavior. Furthermore, an ultrahigh scan rate of up to 5000 V s1 (Fig. S7, ESI†) can be achieved only with a little shape variance, reaching the highest level value compared with a similar study.19,20,40 The superior results of GQDs-1 are attributed to its ultrasmall size, which shortens the diffusion time of electrolyte ions in the electrode material bulk, because the diffusion time of ions (t) is proportional to the square of the diffusion length (L) (t E L2/D).41,42 To further study the effects of sizes on the electrochemical performance, the CV curves of the large GQDs (GQDs-2, and GQDs-3) at different scan rates are depicted in Fig. 4b and c. Compared with GQDs-1, the polarization in CV curves gradually became distinct at the same scan rate because of the longer ion diffusion time. Furthermore, as shown in Fig. 4d–f, the CV curves of large-sized samples (PRGO debris, 4100 nm) exhibited relatively low scan rates (the highest scan rate was only up to 10 V s1), and the CV curves of the largest size product (PRGO-3) showed evident variation in shape, resulting from its large size that limits ion diffusion. Generally, the specific capacitance of the electrode is directly proportional
Fig. 4 CV curves at different scan rates for samples with different sizes: (a) GQDs-1, (b) GQDs-2, (c) GQDs-3, (d) PRGO-1, (e) PRGO-2, and (f) PRGO-3.
14032 | Phys. Chem. Chem. Phys., 2015, 17, 14028--14035
This journal is © the Owner Societies 2015
View Article Online
Published on 30 April 2015. Downloaded by North Dakota State University on 25/05/2015 07:13:44.
Paper
PCCP
Fig. 5 Electrochemical measurements for samples with different sizes: CV curves at different scan rates of (a) 1 V s1 and (b) 10 V s1; (c) GCD curves at a current density of 0.5 A g1, and (d) Nyquist plots.
to its CV area. Thus, we can easily understand from Fig. 5a that the capacitance gradually increased along with the increase in size of the active material. When the scan rate reached 10 V s1 (Fig. 5b), the CV curve of PRGO-3 showed obvious polarization and an area similar to that of PRGO-2. This phenomenon is attributed to the greater capacitance loss of larger electrode materials at higher scan rates. Moreover, the GCD curves of the samples with different sizes exhibited an increasing trend in discharge time at a current density of 0.5 A g1 (Fig. 5c). Furthermore, Fig. 6 also shows that the specific capacitance (calculated by GCD curves at a current density of 0.5 A g1) as a function of electrode material size presented a nearly linear increase in the total size range (about 5–700 nm), with the largest size sample of PRGO-3 displaying the highest specific capacitance of about 88 F g1. However, this value is still lower than that of the specific capacitance of origin RGO-H (107 F g1) obtained by hydrothermal reduction at 180 1C for 8 h (Fig. S8, ESI†).
Fig. 6 The dependence relationship of specific capacitance on the active material size.
This journal is © the Owner Societies 2015
The Nyquist plots (Fig. 5d) were used to display the EIS of GQDs and PRGO debris. Generally, an arc shape in the highfrequency region and a vertical line in the low-frequency region represent charge transfer resistance and ion diffusion resistance, respectively. The Nyquist plot of GQDs-1 showed a huge arc in the high-frequency region, which may be explained by the poor conductivity related to the ultra-small size of GQDs-1. Moreover, inconspicuous arcs were observed in the highfrequency region for PRGO debris (Fig. S9, ESI†), revealing a high electrical conductivity associated with the variation trend, in which the arc gradually diminished with the increase in the size of the active material. To our knowledge, the capacitance (C) can also be calculated using equation C = (–1)/(2pf Z00 ), where f is the frequency and Z00 is the imaginary part of the impedance spectrum.16,40 In view of the above analyses, we can obtain approximately similar results for the smallest-sized GQDs (GQDs-1), which exhibited the lowest capacitance; and the change trend for all samples still existed, keeping the consistency with the results evaluated by the CV and GCD measurements. The frequency response of the samples with different sizes can be evaluated by a comparison between the characteristic frequency (f0), which is the frequency at a 451 phase angle, and its corresponding relaxation time constant (t0 = 1/f0), which is the minimum time needed to discharge all the energy from the device with an efficiency of 450%.17,43 The t0 value for GQDs-1 is 63.3 ms, which is significantly less than those of reported graphene-based materials, such as rGO (t0 = 33 ms),43 onionlike carbon (t0 = 26 ms),44 G-CNTs (t0 = 0.82 ms),45 and graphene films (t0 = 13.3 ms),46 indicating an excellent power response of GQDs-1 (Fig. S10, ESI†). However, the t0 value gradually increased when the size of the active material increased, with the t0 value of 52.0 ms (almost 1000 times than
Phys. Chem. Chem. Phys., 2015, 17, 14028--14035 | 14033
View Article Online
Published on 30 April 2015. Downloaded by North Dakota State University on 25/05/2015 07:13:44.
PCCP
the value of GQDs-1) for the largest product (PRGO-3), revealing the weak power response ability. Cycle performance is crucial for the application in supercapacitors, hence, GQDs-1 and PRGO-3 were further tested by repeating CV tests at a scan rate of 10 V s1 (Fig. S11, ESI†). The capacitance of GQDs-1 and PRGO-3 showed about 95% and 93% retention after 10 000 cycles, respectively. The insets show little difference between the CV curves of the 1st and 10 000th (10 V s1 for GQDs-1, 1 V s1 for PRGO-3), indicating great cycle stability.
4. Conclusions We studied the formation mechanism of GQDs from GO sheets by using H2O2 as an etching agent and ammonia as an assistant, as well as the size effect on the electrochemical behaviors of nanosized graphene-based materials for supercapacitors. GQDs were hardly obtained without ammonia. A small amount of ammonia in H2O2 favored the formation of the smallest sized GQDs, and bigger sized GQDs and PRGO debris were successfully synthesized with the increase of ammonia or the decrease of hydrothermal time. The size variation was mainly attributed to the changes in the etching ability of the hydroxyl radical (OH ) against the reduction ability of the hydroxyl group (OH) in different alkaline environments. In view of the different sizes of GQDs, the fluorescence properties demonstrated an excitation-dependent fluorescence emission feature and a red shift feature with an increase in the GQDs size. Electrochemical tests showed that the smallest sized sample of GQDs-1 exhibited an ultra-high scan rate of up to 5000 V s1, an ultra-fast power response (t0 = 63.3 ms), and the least specific capacitance with only 3.3 F g1. In addition, the specific capacitance as a function of electrode material size revealed a nearly linear increase, whereas the power response also demonstrated an obvious regulation in the total size range. The present results can provide useful information to elucidate the formation mechanism of GQDs and the etching behavior of GO sheets in the liquid phase. These findings can also offer an important reference for synthesizing porous graphene materials through similar methods for supercapacitor applications, as well as demonstrate that GQDs are applicable in microelectronic devices with high power response requirements.
Acknowledgements This work was supported by the National Defense Basic Research Program of China (B1320133001) and the National Nature Science Foundations of China (21203223).
References 1 L. A. Ponomarenko, F. Schedin, M. I. Katsnelson, R. Yang, E. W. Hill, K. S. Novoselov and A. K. Geim, Science, 2008, 320, 356–358. 2 K. A. Ritter and J. W. Lyding, Nat. Mater., 2009, 8, 235–242.
14034 | Phys. Chem. Chem. Phys., 2015, 17, 14028--14035
Paper
3 P. Nguyen, J. W. Li, T. S. Sreeprasad, K. Jasuja, N. Mohanty, M. Ikenberry, K. Hohn, V. B. Shenoy and V. Berry, Small, 2013, 9, 3823–3828. 4 X. J. Zhou, Y. Zhang, C. Wang, X. C. Wu, Y. Q. Yang, B. Zheng, H. X. Wu, S. W. Guo and J. Y. Zhang, ACS Nano, 2012, 6, 6592–6599. 5 F. Jiang, D. Q. Chen, R. M. Li, Y. C. Wang, G. Q. Zhang, S. M. Li, J. P Zheng, N. Y. Huang, Y. Gu, C. R. Wang and C. Y. Shu, Nanoscale, 2013, 5, 1137–1142. 6 X. Michalet, F. F. Pinaud, L. A. Bentolila, J. M. Tsay, S. Doose, J. J. Li, G. Sundaresan, A. M. Wu, S. S. Gambhir and S. Weiss, Sicence, 2005, 307, 538–544. 7 Y. Y. Liu and D. Y. Kim, Chem. Commun., 2015, 51, 4176–4179. 8 L. Cao, S. Sahu, P. Anilkumar, C. E. Bunker, J. Xu, K. A. S. Fernando, P. Wang, E. A. Guliants, K. N. Tackett and Y. P. Sun, J. Am. Chem. Soc., 2011, 133, 4754–4757. 9 L. B. Tang, R. B. Ji, X. K. Cao, J. Y. Lin, H. X. Jiang, X. M. Li, K. S. Teng, C. M. Luk, S. J. Zeng, J. H. Hao and S. P. Lau, ACS Nano, 2012, 6, 5102–5110. 10 Y. Li, Y. Zhao, H. H. Cheng, Y. Hu, G. Q. Shi, L. M. Dai and L. T. Qu, J. Am. Chem. Soc., 2012, 134, 15–18. 11 D. L. Chao, C. R. Zhu, X. H. Xia, J. L. Liu, X. Zhang, J. Wang, P. Liang, J. Y. Lin, H. Zhang, Z. X. Shen and H. J. Fan, Nano Lett., 2015, 15, 565–573. 12 V. Gupta, N. Chaudhary, R. Srivastava, G. D. Sharma, R. Bhardwaj and S. Chand, J. Am. Chem. Soc., 2011, 133, 9960–9963. 13 S. N. Baker and G. A. Baker, Angew. Chem., Int. Ed., 2010, 49, 6726–6744. 14 J. Peng, W. Gao, B. K. Gupta, Z. Liu, R. R. Aburto, L. Ge, L. Song, L. B. Alemany, X. B Zhan, G. H. Gao, S. A. Vithayathil, B. A. Kaipparettu, A. A. Marti, T. Hayashi, J. J. Zhu and P. M. Ajayan, Nano Lett., 2012, 12, 844–849. 15 X. H. Zhu, X. X. Zuo, R. P. Hu, X. Xiao, Y. Liang and J. M. Nan, Mater. Chem. Phys., 2014, 147, 963–967. 16 W. W. Liu, X. B. Yan, J. T. Chen, Y. Q. Feng and Q. J. Xue, Nanoscale, 2013, 5, 6053–6062. 17 W. W. Liu, Y. Q. Feng, X. B. Yan, J. T. Chen and Q. J. Xue, Adv. Funct. Mater., 2013, 33, 4111–4122. 18 Y. R. Zhu, X. B. Ji, C. C. Pan, Q. Q. Sun, W. X. Song, L. B. Fang, Q. Y. Chen and C. E. Banks, Energy Environ. Sci., 2013, 6, 3665–3675. 19 Y. Hu, Y. Zhao, G. W. Lu, N. Chen, Z. P. Zhang, H. Li, H. B. Shao and L. T. Qu, Nanotechnology, 2013, 24, 195401. 20 Q. Chen, Y. Hu, C. G. Hu, H. H. Cheng, Z. P. Zhang, H. B. Shao and L. T. Qu, Phys. Chem. Chem. Phys., 2014, 16, 19307–19313. 21 A. Lerf, H. Y. He, M. Forster and J. Klinowski, J. Phys. Chem. B, 1998, 102, 4477–4482. 22 H. Bai, W. T. Jiang, G. P. Kotchey, W. A. Saidi, B. J. Bythell, J. M. Jarvis, A. G. Marshall, R. A. S. Robinson and A. Star, J. Phys. Chem. C, 2014, 118, 10519–10529. 23 Y. X. Xu, Z. Y. Lin, X. Zhong, X. Q. Huang, N. O. Weiss, Y. Huang and X. F. Duan, Nat. Commun., 2014, 5, 4554. 24 T. Palaniselvam, M. O. Valappil, R. Illathvalappil and S. Kurungot, Energy Environ. Sci., 2014, 7, 1059–1067.
This journal is © the Owner Societies 2015
View Article Online
Published on 30 April 2015. Downloaded by North Dakota State University on 25/05/2015 07:13:44.
Paper
25 F. Yang, M. L. Zhao, B. Z. Zheng, D. Xiao, L. Wu and Y. Guo, J. Mater. Chem., 2012, 22, 25471–25479. 26 H. Jin, D. A. Heller, J. H. Kim and M. S. Strano, Nano Lett., 2008, 8, 4299–4304. ´brega, J. S. S. Oliveira, 27 C. A. Bizzi, E. L. M. Flores, J. A. No L. Schmidt and S. R. Mortari, J. Anal. At. Spectrom., 2014, 29, 332–338. 28 C. Chen, W. X. Kong, H. M. Duan and J. Zhang, Phys. Chem. Chem. Phys., 2014, 16, 12858–12864. 29 Y. C. Bai, R. B. Rakhi, W. Chen and H. N. Alshareef, J. Power Sources, 2013, 233, 313–319. 30 K. N. Wood, R. O’Hayre and S. Pylypenko, Energy Environ. Sci., 2014, 7, 1212–1249. 31 D. S. Geng, Y. Chen, Y. G. Chen, Y. L. Li, R. Y. Li, X. L. Sun, S. Y. Ye and S. Knights, Energy Environ. Sci., 2011, 4, 760–764. 32 X. S. Du, C. F. Zhou, H. Y. Liu, Y. W. Mai and G. X. Wang, J. Power Sources, 2013, 241, 460–466. ´ski and M. Połomska, Micro33 S. Łos´, L. Duclaux, W. Kempin porous Mesoporous Mater., 2010, 130, 21–25. 34 N. Daems, X. Sheng, I. F. J. Vankelecom and P. P. Pescarmona, J. Mater. Chem. A, 2014, 2, 4085–4110. ¨rfer, Phys. Rev. B: 35 F. Libisch, C. Stampfer and J. Burgdo Condens. Matter Mater. Phys., 2009, 79, 115423.
This journal is © the Owner Societies 2015
PCCP
36 M. A. Sk, A. Ananthanarayanan, L. Huang, K. H. Lim and P. Chen, J. Mater. Chem. C, 2014, 2, 6954–6960. 37 A. M. Silva, M. S. Pires, V. N. Freire, E. L. Albuquerque, D. L. Azevedo and E. W. S. Caetano, J. Phys. Chem. C, 2010, 114, 17472–17485. 38 Z. Z. Zhang, K. Chang and F. M. Peeters, Phys. Rev. B: Condens. Matter Mater. Phys., 2008, 77, 235411. 39 J. Kim and J. S. Suh, ACS Nano, 2014, 8, 4190–4196. 40 T. M. Dinh, K. Armstrong, D. Guay and D. Pech, J. Mater. Chem. A, 2014, 2, 7170–7174. 41 S. B. Yang, Y. J. Gong, Z. Liu, L. Zhan, D. P. Hashim, L. L. Ma, R. Vajtai and P. M. Ajayan, Nano Lett., 2013, 13, 1596–1601. 42 P. G. Bruce, B. Scrosati and J. M. Tarascon, Angew. Chem., Int. Ed., 2008, 47, 2930–2946. 43 M. Beidaghi and C. L. Wang, Adv. Funct. Mater., 2012, 22, 4501–4510. 44 D. Pech, M. Brunet, H. Durou, P. H. Huang, V. Mochalin, Y. Gogotsi, P. L. Taberna and P. Simon, Nat. Nanotechnol., 2010, 5, 651–654. 45 J. Lin, C. G. Zhang, Z. Yan, Y. Zhu, Z. W. Peng, R. H. Hauge, D. Natelson and J. M. Tour, Nano Lett., 2013, 13, 72–78. 46 X. W. Yang, J. W. Zhu, L. Qiu and D. Li, Adv. Mater., 2011, 23, 2833–2838.
Phys. Chem. Chem. Phys., 2015, 17, 14028--14035 | 14035
Published on 30 April 2015. Downloaded by North Dakota State University on 25/05/2015 07:13:44.
PAPER
Cite this: Phys. Chem. Chem. Phys., 2015, 17, 14028
View Journal | View Issue
Insight into the formation mechanism of graphene quantum dots and the size effect on their electrochemical behaviors† Yonghuan Liu,ab Rutao Wang,a Junwei Langa and Xingbin Yan*a To study the formation mechanism and influencing factors of graphene quantum dots (GQDs), GQDs with different average sizes were prepared using a modified hydrothermal method with hydrogen peroxide (H2O2) as an etching agent and ammonia as an assistant. It is found that size-controlled GQDs were prepared by adjusting the amount of ammonia and porous reduced graphene oxide (PRGO) debris can be synthesized by reducing the hydrothermal reaction time. Structural changes of final products were mainly attributed to the changes in the etching ability of the hydroxyl radical (OH ) against the reduction ability of the hydroxyl group (OH) in different alkaline environments regulated by ammonia.
Received 2nd February 2015, Accepted 30th April 2015 DOI: 10.1039/c5cp00646e
Furthermore, we studied the electrochemical properties of GQDs and PRGO. The results showed that the specific capacitance of all samples increases linearly with the size and the smallest GQDs can work at the highest scan rate of as high as 5000 V s1 with an ultra-fast power response (t0 = 63.3 ms). Thus, these findings elucidate the formation mechanism of GQDs and demonstrate that GQDs are applicable
www.rsc.org/pccp
in microelectronic devices with high power response requirements.
1. Introduction Graphene quantum dots (GQDs), which are graphene nanosheets with lateral dimensions of o100 nm, exhibit numerous novel chemical and physical properties because of quantum confinement and edge effects associated with graphene and QDs.1–5 To date, GQDs have demonstrated great potential for various applications, such as bio-imaging devices,6 sensors,7 catalysts,8 light-emitting diodes,9 fuel cells,10 lithium ion batteries,11 and photovoltaic devices.12 The primary approaches for synthesizing GQDs are the bottom-up and top-down methods.4,5 The former mainly involves pyrolysis of polycyclic aromatic hydrocarbons for a subsequent selfassembly or cyclodehydrogenation of polyphenylene precursors, but often entails a serious drawbacks concerning low purity;13 by contrast, the latter primarily depends on chemical cutting of large graphene or graphene oxide (GO) into nanometer-sized pieces (GQDs),5,14 which often requires abundant strong acids or organic solvents that are very difficult to remove. Recently, the preparation of GQDs by cutting GO sheets in a hydrothermal ambience by using hydrogen peroxide (H2O2) and ammonia has gained much a
Laboratory of Clean Energy Chemistry and Materials, State Key Laboratory of Solid Lubrication, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, Lanzhou 730000, P. R. China. E-mail: [email protected]; Fax: +86 931 4968055; Tel: +86 931 4968055 b Graduate University of Chinese Academy of Sciences, Beijing 100080, P. R. China † Electronic supplementary information (ESI) available. See DOI: 10.1039/ c5cp00646e
14028 | Phys. Chem. Chem. Phys., 2015, 17, 14028--14035
attention owing to the high purity and strong photoluminescence of as-produced GQDs, as well as the simplicity of the preparation process.5,15 However, the formation mechanism of GQDs, especially the respective roles of H2O2 and ammonia, remains unclear. Moreover, as-prepared GQDs showed a simple size range of only several nanometers, thereby limiting the applications of GQDs. Supercapacitors are promising candidates in energy storage systems because of their high power density, fast charge/ discharge rate and excellent cycle life.16–18 Our previous reports first demonstrated GQDs as promising electrode materials to build micro-supercapacitors with a superior power response and rate capability.16,17 Subsequently, Zhu et al. reported that adding GQDs can remarkably enhance the rate capability of RuO2 nanoparticles.18 Hu et al. also reported that depositing GQDs on carbon nanotubes (CNTs) can significantly increase the capacitance of CNT arrays.19 Furthermore, Chen et al. also reported the enhanced capacitance of 3D graphene by depositing GQDs.20 These results indicated the great potential of GQDs as supercapacitors. However, no report is available regarding the intrinsic electrochemical characteristics of GQDs. In the current research, we demonstrated a modified hydrothermal approach to prepare GQDs with controllable sizes by regulating the ratio of ammonia to H2O2. We systematically studied the formation mechanism and influencing factors. The results revealed that the average size of GQDs is strongly dependent upon the amount of ammonia, suggesting that aside from H2O2, ammonia also plays an important role in GQD formation.
This journal is © the Owner Societies 2015
View Article Online
Published on 30 April 2015. Downloaded by North Dakota State University on 25/05/2015 07:13:44.
Paper
PCCP
GQDs with different sizes can be prepared by adjusting the amount of ammonia in H2O2, and porous reduced graphene oxide (PRGO) can be synthesized by shortening the hydrothermal reaction time. The structure changes were mainly attributed to the alteration of the etching ability of the hydroxyl radical (OH ) against the reduction ability of the hydroxyl group (OH) in different alkaline environments. Based on this, we studied the electrochemical properties of the final GQDs and PRGO with different sizes. The specific capacitance as a function of electrode material size showed a nearly linear increase. GQDs with the smallest size exhibited an ultra-high scan rate of 5000 V s1, an ultra-fast power response (t0 = 63.3 ms) and the lowest specific capacitance of 3.3 F g1.
and RGO-T (reduced graphene oxide prepared by thermal reduction at 300 1C for 2 h) as raw materials in place of GO. 2.2
Morphological analysis of the samples was conducted on a transmission electron microscope (TEM; JEOL 2100 FEG) and an atomic force microscope (AFM; Veeco, Nanoscope IIIA). The chemical bonds and components were investigated using a Fourier transform infrared spectrometer (FTIR; IFS120HR) and an X-ray photoelectron spectrometer (XPS; ESCALAB250x). Raman spectra were measured on a Raman spectrometer (JY-HR800) with an excitation wavelength of 532 nm. The fluorescence spectra were obtained using a fluorescence spectrophotometer (PerkinElmer Lambda 35). 2.3
2. Experimental section 2.1
Preparation of GQDs and PRGO with different sizes
GO sheets were prepared from natural graphite by using a modified Hummers method, as described previously.21,22 GQDs and PRGO with different sizes were then prepared using a hydrothermal synthesis method with H2O2 as an etching agent and ammonia as an assistant. In a typical synthesis, 0.5 ml of ammonia (28 wt%) and 6 ml of H2O2 (30 wt%) were added in 40 ml of GO aqueous dispersion (2 mg ml1) with the aid of stirring. The resulting dispersion was diluted to 160 ml, then transferred to a 200 ml Teflon autoclave and maintained at 180 1C for 8 h. After cooling to room temperature, a light yellow supernatant was obtained by a filtration process through a porous cellulose ester membrane (220 nm). The supernatant was heated to remove unreacted ammonia and H2O2. The yellow powder (denoted as GQDs-1) was finally collected by freeze-drying, and the yield was about 25%. Other samples of GQDs with large sizes and PRGO were produced by changing either the volume ratio of ammonia to H2O2 or the hydrothermal reaction time at the same temperature. The experimental parameters of these samples are listed in Table 1. To analyze the formation mechanism of GQDs, a series of contrast experiments were conducted, including the synthesis with only H2O2 in GO dispersion (without ammonia), and the synthesis using sodium hydroxide (NaOH) to replace ammonia, as well as the experiments by using N-RGO (reduced graphene oxide obtained using hydrazine reduction with the aid of reflux at 95 1C for 1 h)
Table 1 Reaction conditions for synthesizing GQD series and PRGO, the mean size or the size range, and the contents of N and O in the final samples
Ammonia Mean volume H 2 O2 Reaction size Samples (ml) volume (ml) time (h) (nm) GQDs-1 GQDs-2 GQDs-3 PRGO-1 PRGO-2 PRGO-3
0.5 2 3 3 3 3
6 6 6 6 6 6
8 8 8 6 4 2
This journal is © the Owner Societies 2015
Yield N O (%) (at%) (at%)
4.8 25 5.80 13.1 37 3.31 34.7 490 2.76 100–150 4.43 200–300 4.02 500–700 7.28
30.00 28.28 25.57 30.06 26.49 31.00
Structural characterization
Electrode preparation and electrochemical measurements
To study the intrinsic electrochemical performance of the as-prepared samples, the working electrodes were fabricated by dipping GQDs or PRGO dispersion with determined concentration on polished glassy carbon electrodes (5 mm in diameter) without additives. The active material on each glassy carbon electrode, which was about 0.0185 mg, was calculated using the calibrated concentration and the adjusted volume of each size sample. After drying at 60 1C, the electrode was coated with 8 ml of naphthol solution (5 wt%) to prevent the dissolution of the active material in the electrolyte. The electrode was then dried at 60 1C for 1 h. Cyclic voltammetry (CV), galvanostatic charge/discharge (GCD), and electrochemical impedance spectroscopy (EIS) measurements were investigated in a three-electrode system with a saturated calomel electrode (SCE) as a reference electrode, a platinum foil (with area of 1 cm2) as a counter electrode and Na2SO4 solution (1 M L1) as an electrolyte. CV and GCD tests were conducted in a potential window of 0.4 V to 0.6 V (vs. SCE). EIS measurements were performed in a frequency range of 0.01 Hz to 100 kHz with 5 mV ac amplitude. All tests were conducted on an electrochemical workstation (CHI660D, Shanghai, China) at room temperature. The specific capacitance was calculated from the GCD curves based on the following equation: C = IDt/mDV (F g1) where I is the current density (A), Dt is the discharge time (s), DV is the potential window of discharging (V), and m is the active material mass.
3. Results and discussion GQDs with different sizes were successfully prepared in this research. Fig. 2a and the inset show that GQDs (GQDs-1) with a homogeneous size (a mean size of 4.8 nm) were synthesized. The inset also depicts the HRTEM image of GQDs-1 with a relatively high crystallinity. The AFM image of GQDs-1 illustrated a uniform size distribution, and the inset shows that their topographic heights are mostly o2 nm, revealing one to three layers (Fig. S3, ESI†).
Phys. Chem. Chem. Phys., 2015, 17, 14028--14035 | 14029
View Article Online
Published on 30 April 2015. Downloaded by North Dakota State University on 25/05/2015 07:13:44.
PCCP
Fig. 1 Schematic of the formation processes of GQDs: (a) a hydrothermal process by adding H2O2 only; (b) a hydrothermal process by adding both ammonia and H2O2.
To understand the formation processes and the mechanism of GQDs, H2O2 (3 ml) was added to the GO (80 mg) dispersion. As shown in the route Fig. 1a, we can roughly observe the etching processes of GO by using H2O2. Generally, H2O2 may be dissociated into hydroxyl radicals (OH ) under a hydrothermal process at 180 1C, and the as-generated OH was considered as one of the most powerful oxidizing species resembling chemical scissors.4,22,23 Thus, OH can react with the defect sites of GO, which include abundant oxygen-containing functional groups, such as OH, C–O–C and CQO/COOH on their surfaces. Therefore, the large GO sheets are broken down into small fragments/debris along with the release of CO2.24,25 The generated GO fragments reacting with OH again are converted to smaller debris and CO2. Ultimately, with the increase of etching time, the GO debris can thoroughly transform into CO2, CO, and H2O.24,25 It is visually seen that the color of the mixture gradually faded and disappeared with an etching time of 2 h. (Fig. S1a–e, ESI†), revealing the high etching ability of H2O2. The TEM images of the samples etched by using only H2O2 for different time (Fig. S2d–f, ESI†) showed that the GO sheets were gradually rived off, however, GQDs infrequently appeared.
Paper
However, tiny-sized GQDs were prepared after simultaneously adding a small amount of ammonia (0.5 ml) and H2O2 (3 ml) to GO dispersion under the same conditions as shown in the route of Fig. 1b. Our analysis indicated that OH can act as an initiator, similar to the role of Fe2+ in the photo-Fenton reaction, to induce H2O2 and decompose into more OH in alkaline medium.4,26,27 The presence of GO fragments and GQDs (Fig. S2c, ESI†) in the reaction system even after the increase in reaction time can be attributed to the ammonia atmosphere. To our knowledge, GO suspension can be partly reduced under alkaline conditions at a moderate temperature.28,29 Therefore, the as-produced small GO fragments and GQDs can be reduced and protected in the reaction system, thus the tiny-sized GQDs (GQDs-1) were obtained by removal from the GO (Fig. S2a, ESI†). As a verification, a reaction was hardly observed between H2O2 and N-RGO, as well as RGO-T (to replace GO) at 180 1C for 8 h (Fig. S1g and h, ESI†). Moreover, by adding NaOH to replace ammonia with a similar pH value at 180 1C for 8 h, we can achieve a similar result, in which a brown dispersion can be obtained by filtration (Figs. S1f and b, ESI†). As the volume of ammonia increased, larger GQDs (GQDs-2) with uniform size distribution (average size of 13.1 nm) and GQDs with a mean size of 34.7 nm (GQDs-3) were obtained (Fig. 2b and c). The inset shows the corresponding size distribution of GQDs-2 and GQDs-3. The preparation of larger GQDs benefits from the addition of more ammonia, which results in more rapid OH release and more active participation in reacting with GO at the initial stage. Meanwhile, the oxygencontaining functional groups of GO can react with ammonia to generate amide and develop N-doping products.30–32 Based on the above observations, we can naturally summarize that ammonia mainly plays a role in three aspects: first, OH can facilitate decomposition of H2O2 into more OH ; second, OH can react with oxygen-containing functional groups of residual fragments and GQDs to passivate their surfaces and play the role of a ‘‘protector’’; third, the volume of ammonia determines the structural changes in the etching ability of OH against the reduction ability of OH. The total reaction processes under ammonia conditions may be presented as follows:
Fig. 2 TEM images of the as-prepared samples with different sizes: (a) GQDs-1 and the inset showing its HRTEM, (b) GQDs-2, (c) GQDs-3, (d) PRGO-1, (e) PRGO-2, and (f) PRGO-3. The inset shows the corresponding size distribution for GQDs.
14030 | Phys. Chem. Chem. Phys., 2015, 17, 14028--14035
This journal is © the Owner Societies 2015
View Article Online
Published on 30 April 2015. Downloaded by North Dakota State University on 25/05/2015 07:13:44.
Paper
PCCP
H2O2 + H2O - H3O+ + HO2
(1)
H3O+ + OH - 2H2O
(2)
H2O2 + HO2 - HO2 + HO + OH
(3)
HO + C - CO2
(4)
Based on the above discussion, we proposed that the reaction processes in the ammonia medium are those as shown in the route of Fig. 1b. Moreover, by reducing the reaction time, the larger-sized samples of the PRGO debris (including PRGO-1 with a size range of 100–150 nm, PRGO-2 with a size range of 200–300 nm, and PRGO-3 with a size range of 500–700 nm; Table 1) were successfully prepared (Fig. 2d–f). We can only show the size ranges of these samples because their irregular porous structure complicates the evaluation of their mean sizes. Fig. 3a shows the Raman spectra of GQDs with different sizes (denoted GQD series) and PRGO debris. Two unconspicuous peaks existed at around 1360 and 1600 cm1, corresponding to the D and G bands of the GQD series, respectively. The abnormal phenomenon is mainly attributed to the size effect and strong fluorescence disturbance of the Raman characterization.31,33 However, with an increase in sample size, the D and G bands gradually appeared and exhibited characteristic peaks of a graphene-based material for PRGO-3. The FTIR spectra showed similar chemical species for GQDs and PRGO debris samples of different sizes (Fig. 3b). The spectra indicated the presence of oxygen functional groups, including C–O (n C–O at 1110 cm1), C–OH (n C–O at 1394 cm1), CQO mainly in carbonyl moieties (n CQO at 1645 cm1) and the
H-bond associated OH (a broad band at around 3470 cm1). In addition, the peak at 3180 cm1 was assigned to the N–H stretching of the amine group, indicating a successful reaction between ammonia and GO.30,34 The difference among the functional group peak centers of all the samples was caused by their diverse surrounding environments. For comparison, the FTIR spectra of GO, RGO-H obtained by hydrothermal treatment of GO at 180 1C for 8 h without adding ammonia, N-RGO and RGO-T were investigated (Fig. S4, ESI†). A relatively weak peak at about 1630 cm1 (n CQO) was found for RGO-H and RGO-T compared with GO and a peak at about 3180 cm1 (n N–H) was observed for N-RGO after reduction, which can explain the scarce reaction between H2O2 and RGO (including N-RGO and RGO-T). In view of the rich oxygencontaining functional groups of final products, the GQD series demonstrates an excellent stability in water without any change for three months. XPS measurements were performed to determine the composition of the GQD series and PRGO debris with different sizes (Fig. 3c). The XPS survey spectra of all the products showed similar element compositions, such as the predominant graphitic C1s peak at about 284 eV, the O1s peak at about 532 eV and the N1s peak at about 400 eV. Table 1 summarizes the contents of O and N elements, exhibiting similar results with relatively high values in all the samples. Briefly, the GQDs-1 sample was selected as a model for further analysis. The highresolution C1s XPS spectra of the GQDs-1 can be divided into five components (Fig. 3d), corresponding to the signals of C–C (284.7 eV), C–N (286.1 eV), C–O (286.6 eV), CQO (288.5 eV), and O–CQO (289.1 eV). The oxygen species of the GQDs-1 can be analyzed from the high-resolution O1s XPS spectra (Fig. 3e),
Fig. 3 (a) Raman spectra, (b) FTIR spectra, and (c) XPS patterns of GQDs and PRGO debris with different sizes. (d) C1s, (e) O1s, and (f) N1s spectra of GQDs-1.
This journal is © the Owner Societies 2015
Phys. Chem. Chem. Phys., 2015, 17, 14028--14035 | 14031
View Article Online
Published on 30 April 2015. Downloaded by North Dakota State University on 25/05/2015 07:13:44.
PCCP
which revealed three components corresponding to quinine (531.3 eV), CQO (531.8 eV), and C–O (533.2 eV). Fig. 3f shows the high-resolution N1s XPS spectra, consisting of amide (399.8 eV), pyridine-like (401.4 eV), and pyrrolic (402.1 eV) N atoms.30,31 Based on the above results, we roughly concluded the successfully introduced N atoms in the as-prepared GQDs series and PRGO debris. The main chemical groups on their surfaces were OH, epoxy/ether, CQO, and CO–NH2. GQDs have demonstrated promising properties for photoelectric applications. To our knowledge, the size, shape, and function group can contribute to the energy gaps and the fluorescence property.35–37 In our case, the chemical groups were similar for the series of GQDs, with an increase in size for GQDs-1 to GQDs-3. The photoluminescence (PL) spectra of the GQDs series (Fig. S5a–c, ESI†) showed that the as-prepared GQDs exhibited an excitation dependent PL behavior. When the excitation wavelength changed from 360 nm to 480 nm, the PL peak shifted from 480 nm to 530 nm. PL intensity initially increased and then decreased, reaching the strongest emission at an excitation wavelength of 440 nm. Importantly, the PL spectra of the GQDs series (GQDs-1, GQDs-2, and GQDs-3) at the same excitation wavelength of 440 nm showed that the PL intensity gradually decreased (Fig. S5d, ESI†), and the PL peak center gradually red shifted with the increase of the GQDs size (the peak centers of 504 nm, 512 nm, and 518 nm for GQDs-1, GQDs-2, and GQDs-3, respectively). Generally, this red shift behavior is mainly ascribed to the decrease of band-gaps.36,37,39 Compared with the size and functional groups, the band gaps (HOMO–LUMO) are relatively insensitive to the shape, and the dependence of the band gaps on size shows 1/L2, approximately.38 Therefore, this red shift phenomenon mainly resulted from the increase of the sample size. In our case, the PL
Paper
behavior almost disappeared for the large sized samples of PRGO debris owing to their exceeding size. The photographs of different size products in the aqueous solution captured under white light and 365 nm UV light (Fig. S6, ESI†) clearly showed that the fluorescence color gradually receded; moreover, a negligible fluorescence color appeared for the PRGO debris, which is mainly ascribed to the residual GQDs (Fig. S2c, ESI†). Fig. 4a shows the CV curves of the smallest sized product (GQDs-1) in a potential range of 0.4 V to 0.6 V at various scan rates from 1 V s1 to 500 V s1. The CV curve area increased gradually with the increase of the scan rate, and the CV curve at a scan rate of 500 V s1 still maintained the rectangular shape, exhibiting an ideal capacitive behavior. Furthermore, an ultrahigh scan rate of up to 5000 V s1 (Fig. S7, ESI†) can be achieved only with a little shape variance, reaching the highest level value compared with a similar study.19,20,40 The superior results of GQDs-1 are attributed to its ultrasmall size, which shortens the diffusion time of electrolyte ions in the electrode material bulk, because the diffusion time of ions (t) is proportional to the square of the diffusion length (L) (t E L2/D).41,42 To further study the effects of sizes on the electrochemical performance, the CV curves of the large GQDs (GQDs-2, and GQDs-3) at different scan rates are depicted in Fig. 4b and c. Compared with GQDs-1, the polarization in CV curves gradually became distinct at the same scan rate because of the longer ion diffusion time. Furthermore, as shown in Fig. 4d–f, the CV curves of large-sized samples (PRGO debris, 4100 nm) exhibited relatively low scan rates (the highest scan rate was only up to 10 V s1), and the CV curves of the largest size product (PRGO-3) showed evident variation in shape, resulting from its large size that limits ion diffusion. Generally, the specific capacitance of the electrode is directly proportional
Fig. 4 CV curves at different scan rates for samples with different sizes: (a) GQDs-1, (b) GQDs-2, (c) GQDs-3, (d) PRGO-1, (e) PRGO-2, and (f) PRGO-3.
14032 | Phys. Chem. Chem. Phys., 2015, 17, 14028--14035
This journal is © the Owner Societies 2015
View Article Online
Published on 30 April 2015. Downloaded by North Dakota State University on 25/05/2015 07:13:44.
Paper
PCCP
Fig. 5 Electrochemical measurements for samples with different sizes: CV curves at different scan rates of (a) 1 V s1 and (b) 10 V s1; (c) GCD curves at a current density of 0.5 A g1, and (d) Nyquist plots.
to its CV area. Thus, we can easily understand from Fig. 5a that the capacitance gradually increased along with the increase in size of the active material. When the scan rate reached 10 V s1 (Fig. 5b), the CV curve of PRGO-3 showed obvious polarization and an area similar to that of PRGO-2. This phenomenon is attributed to the greater capacitance loss of larger electrode materials at higher scan rates. Moreover, the GCD curves of the samples with different sizes exhibited an increasing trend in discharge time at a current density of 0.5 A g1 (Fig. 5c). Furthermore, Fig. 6 also shows that the specific capacitance (calculated by GCD curves at a current density of 0.5 A g1) as a function of electrode material size presented a nearly linear increase in the total size range (about 5–700 nm), with the largest size sample of PRGO-3 displaying the highest specific capacitance of about 88 F g1. However, this value is still lower than that of the specific capacitance of origin RGO-H (107 F g1) obtained by hydrothermal reduction at 180 1C for 8 h (Fig. S8, ESI†).
Fig. 6 The dependence relationship of specific capacitance on the active material size.
This journal is © the Owner Societies 2015
The Nyquist plots (Fig. 5d) were used to display the EIS of GQDs and PRGO debris. Generally, an arc shape in the highfrequency region and a vertical line in the low-frequency region represent charge transfer resistance and ion diffusion resistance, respectively. The Nyquist plot of GQDs-1 showed a huge arc in the high-frequency region, which may be explained by the poor conductivity related to the ultra-small size of GQDs-1. Moreover, inconspicuous arcs were observed in the highfrequency region for PRGO debris (Fig. S9, ESI†), revealing a high electrical conductivity associated with the variation trend, in which the arc gradually diminished with the increase in the size of the active material. To our knowledge, the capacitance (C) can also be calculated using equation C = (–1)/(2pf Z00 ), where f is the frequency and Z00 is the imaginary part of the impedance spectrum.16,40 In view of the above analyses, we can obtain approximately similar results for the smallest-sized GQDs (GQDs-1), which exhibited the lowest capacitance; and the change trend for all samples still existed, keeping the consistency with the results evaluated by the CV and GCD measurements. The frequency response of the samples with different sizes can be evaluated by a comparison between the characteristic frequency (f0), which is the frequency at a 451 phase angle, and its corresponding relaxation time constant (t0 = 1/f0), which is the minimum time needed to discharge all the energy from the device with an efficiency of 450%.17,43 The t0 value for GQDs-1 is 63.3 ms, which is significantly less than those of reported graphene-based materials, such as rGO (t0 = 33 ms),43 onionlike carbon (t0 = 26 ms),44 G-CNTs (t0 = 0.82 ms),45 and graphene films (t0 = 13.3 ms),46 indicating an excellent power response of GQDs-1 (Fig. S10, ESI†). However, the t0 value gradually increased when the size of the active material increased, with the t0 value of 52.0 ms (almost 1000 times than
Phys. Chem. Chem. Phys., 2015, 17, 14028--14035 | 14033
View Article Online
Published on 30 April 2015. Downloaded by North Dakota State University on 25/05/2015 07:13:44.
PCCP
the value of GQDs-1) for the largest product (PRGO-3), revealing the weak power response ability. Cycle performance is crucial for the application in supercapacitors, hence, GQDs-1 and PRGO-3 were further tested by repeating CV tests at a scan rate of 10 V s1 (Fig. S11, ESI†). The capacitance of GQDs-1 and PRGO-3 showed about 95% and 93% retention after 10 000 cycles, respectively. The insets show little difference between the CV curves of the 1st and 10 000th (10 V s1 for GQDs-1, 1 V s1 for PRGO-3), indicating great cycle stability.
4. Conclusions We studied the formation mechanism of GQDs from GO sheets by using H2O2 as an etching agent and ammonia as an assistant, as well as the size effect on the electrochemical behaviors of nanosized graphene-based materials for supercapacitors. GQDs were hardly obtained without ammonia. A small amount of ammonia in H2O2 favored the formation of the smallest sized GQDs, and bigger sized GQDs and PRGO debris were successfully synthesized with the increase of ammonia or the decrease of hydrothermal time. The size variation was mainly attributed to the changes in the etching ability of the hydroxyl radical (OH ) against the reduction ability of the hydroxyl group (OH) in different alkaline environments. In view of the different sizes of GQDs, the fluorescence properties demonstrated an excitation-dependent fluorescence emission feature and a red shift feature with an increase in the GQDs size. Electrochemical tests showed that the smallest sized sample of GQDs-1 exhibited an ultra-high scan rate of up to 5000 V s1, an ultra-fast power response (t0 = 63.3 ms), and the least specific capacitance with only 3.3 F g1. In addition, the specific capacitance as a function of electrode material size revealed a nearly linear increase, whereas the power response also demonstrated an obvious regulation in the total size range. The present results can provide useful information to elucidate the formation mechanism of GQDs and the etching behavior of GO sheets in the liquid phase. These findings can also offer an important reference for synthesizing porous graphene materials through similar methods for supercapacitor applications, as well as demonstrate that GQDs are applicable in microelectronic devices with high power response requirements.
Acknowledgements This work was supported by the National Defense Basic Research Program of China (B1320133001) and the National Nature Science Foundations of China (21203223).
References 1 L. A. Ponomarenko, F. Schedin, M. I. Katsnelson, R. Yang, E. W. Hill, K. S. Novoselov and A. K. Geim, Science, 2008, 320, 356–358. 2 K. A. Ritter and J. W. Lyding, Nat. Mater., 2009, 8, 235–242.
14034 | Phys. Chem. Chem. Phys., 2015, 17, 14028--14035
Paper
3 P. Nguyen, J. W. Li, T. S. Sreeprasad, K. Jasuja, N. Mohanty, M. Ikenberry, K. Hohn, V. B. Shenoy and V. Berry, Small, 2013, 9, 3823–3828. 4 X. J. Zhou, Y. Zhang, C. Wang, X. C. Wu, Y. Q. Yang, B. Zheng, H. X. Wu, S. W. Guo and J. Y. Zhang, ACS Nano, 2012, 6, 6592–6599. 5 F. Jiang, D. Q. Chen, R. M. Li, Y. C. Wang, G. Q. Zhang, S. M. Li, J. P Zheng, N. Y. Huang, Y. Gu, C. R. Wang and C. Y. Shu, Nanoscale, 2013, 5, 1137–1142. 6 X. Michalet, F. F. Pinaud, L. A. Bentolila, J. M. Tsay, S. Doose, J. J. Li, G. Sundaresan, A. M. Wu, S. S. Gambhir and S. Weiss, Sicence, 2005, 307, 538–544. 7 Y. Y. Liu and D. Y. Kim, Chem. Commun., 2015, 51, 4176–4179. 8 L. Cao, S. Sahu, P. Anilkumar, C. E. Bunker, J. Xu, K. A. S. Fernando, P. Wang, E. A. Guliants, K. N. Tackett and Y. P. Sun, J. Am. Chem. Soc., 2011, 133, 4754–4757. 9 L. B. Tang, R. B. Ji, X. K. Cao, J. Y. Lin, H. X. Jiang, X. M. Li, K. S. Teng, C. M. Luk, S. J. Zeng, J. H. Hao and S. P. Lau, ACS Nano, 2012, 6, 5102–5110. 10 Y. Li, Y. Zhao, H. H. Cheng, Y. Hu, G. Q. Shi, L. M. Dai and L. T. Qu, J. Am. Chem. Soc., 2012, 134, 15–18. 11 D. L. Chao, C. R. Zhu, X. H. Xia, J. L. Liu, X. Zhang, J. Wang, P. Liang, J. Y. Lin, H. Zhang, Z. X. Shen and H. J. Fan, Nano Lett., 2015, 15, 565–573. 12 V. Gupta, N. Chaudhary, R. Srivastava, G. D. Sharma, R. Bhardwaj and S. Chand, J. Am. Chem. Soc., 2011, 133, 9960–9963. 13 S. N. Baker and G. A. Baker, Angew. Chem., Int. Ed., 2010, 49, 6726–6744. 14 J. Peng, W. Gao, B. K. Gupta, Z. Liu, R. R. Aburto, L. Ge, L. Song, L. B. Alemany, X. B Zhan, G. H. Gao, S. A. Vithayathil, B. A. Kaipparettu, A. A. Marti, T. Hayashi, J. J. Zhu and P. M. Ajayan, Nano Lett., 2012, 12, 844–849. 15 X. H. Zhu, X. X. Zuo, R. P. Hu, X. Xiao, Y. Liang and J. M. Nan, Mater. Chem. Phys., 2014, 147, 963–967. 16 W. W. Liu, X. B. Yan, J. T. Chen, Y. Q. Feng and Q. J. Xue, Nanoscale, 2013, 5, 6053–6062. 17 W. W. Liu, Y. Q. Feng, X. B. Yan, J. T. Chen and Q. J. Xue, Adv. Funct. Mater., 2013, 33, 4111–4122. 18 Y. R. Zhu, X. B. Ji, C. C. Pan, Q. Q. Sun, W. X. Song, L. B. Fang, Q. Y. Chen and C. E. Banks, Energy Environ. Sci., 2013, 6, 3665–3675. 19 Y. Hu, Y. Zhao, G. W. Lu, N. Chen, Z. P. Zhang, H. Li, H. B. Shao and L. T. Qu, Nanotechnology, 2013, 24, 195401. 20 Q. Chen, Y. Hu, C. G. Hu, H. H. Cheng, Z. P. Zhang, H. B. Shao and L. T. Qu, Phys. Chem. Chem. Phys., 2014, 16, 19307–19313. 21 A. Lerf, H. Y. He, M. Forster and J. Klinowski, J. Phys. Chem. B, 1998, 102, 4477–4482. 22 H. Bai, W. T. Jiang, G. P. Kotchey, W. A. Saidi, B. J. Bythell, J. M. Jarvis, A. G. Marshall, R. A. S. Robinson and A. Star, J. Phys. Chem. C, 2014, 118, 10519–10529. 23 Y. X. Xu, Z. Y. Lin, X. Zhong, X. Q. Huang, N. O. Weiss, Y. Huang and X. F. Duan, Nat. Commun., 2014, 5, 4554. 24 T. Palaniselvam, M. O. Valappil, R. Illathvalappil and S. Kurungot, Energy Environ. Sci., 2014, 7, 1059–1067.
This journal is © the Owner Societies 2015
View Article Online
Published on 30 April 2015. Downloaded by North Dakota State University on 25/05/2015 07:13:44.
Paper
25 F. Yang, M. L. Zhao, B. Z. Zheng, D. Xiao, L. Wu and Y. Guo, J. Mater. Chem., 2012, 22, 25471–25479. 26 H. Jin, D. A. Heller, J. H. Kim and M. S. Strano, Nano Lett., 2008, 8, 4299–4304. ´brega, J. S. S. Oliveira, 27 C. A. Bizzi, E. L. M. Flores, J. A. No L. Schmidt and S. R. Mortari, J. Anal. At. Spectrom., 2014, 29, 332–338. 28 C. Chen, W. X. Kong, H. M. Duan and J. Zhang, Phys. Chem. Chem. Phys., 2014, 16, 12858–12864. 29 Y. C. Bai, R. B. Rakhi, W. Chen and H. N. Alshareef, J. Power Sources, 2013, 233, 313–319. 30 K. N. Wood, R. O’Hayre and S. Pylypenko, Energy Environ. Sci., 2014, 7, 1212–1249. 31 D. S. Geng, Y. Chen, Y. G. Chen, Y. L. Li, R. Y. Li, X. L. Sun, S. Y. Ye and S. Knights, Energy Environ. Sci., 2011, 4, 760–764. 32 X. S. Du, C. F. Zhou, H. Y. Liu, Y. W. Mai and G. X. Wang, J. Power Sources, 2013, 241, 460–466. ´ski and M. Połomska, Micro33 S. Łos´, L. Duclaux, W. Kempin porous Mesoporous Mater., 2010, 130, 21–25. 34 N. Daems, X. Sheng, I. F. J. Vankelecom and P. P. Pescarmona, J. Mater. Chem. A, 2014, 2, 4085–4110. ¨rfer, Phys. Rev. B: 35 F. Libisch, C. Stampfer and J. Burgdo Condens. Matter Mater. Phys., 2009, 79, 115423.
This journal is © the Owner Societies 2015
PCCP
36 M. A. Sk, A. Ananthanarayanan, L. Huang, K. H. Lim and P. Chen, J. Mater. Chem. C, 2014, 2, 6954–6960. 37 A. M. Silva, M. S. Pires, V. N. Freire, E. L. Albuquerque, D. L. Azevedo and E. W. S. Caetano, J. Phys. Chem. C, 2010, 114, 17472–17485. 38 Z. Z. Zhang, K. Chang and F. M. Peeters, Phys. Rev. B: Condens. Matter Mater. Phys., 2008, 77, 235411. 39 J. Kim and J. S. Suh, ACS Nano, 2014, 8, 4190–4196. 40 T. M. Dinh, K. Armstrong, D. Guay and D. Pech, J. Mater. Chem. A, 2014, 2, 7170–7174. 41 S. B. Yang, Y. J. Gong, Z. Liu, L. Zhan, D. P. Hashim, L. L. Ma, R. Vajtai and P. M. Ajayan, Nano Lett., 2013, 13, 1596–1601. 42 P. G. Bruce, B. Scrosati and J. M. Tarascon, Angew. Chem., Int. Ed., 2008, 47, 2930–2946. 43 M. Beidaghi and C. L. Wang, Adv. Funct. Mater., 2012, 22, 4501–4510. 44 D. Pech, M. Brunet, H. Durou, P. H. Huang, V. Mochalin, Y. Gogotsi, P. L. Taberna and P. Simon, Nat. Nanotechnol., 2010, 5, 651–654. 45 J. Lin, C. G. Zhang, Z. Yan, Y. Zhu, Z. W. Peng, R. H. Hauge, D. Natelson and J. M. Tour, Nano Lett., 2013, 13, 72–78. 46 X. W. Yang, J. W. Zhu, L. Qiu and D. Li, Adv. Mater., 2011, 23, 2833–2838.
Phys. Chem. Chem. Phys., 2015, 17, 14028--14035 | 14035
