Graphene oxide derived graphene quantum dots with different photoluminescence properties and peroxidase-like catalytic activity

Abstract

Graphene quantum dots (GQDs), as a new kind of carbon nanomaterial, have been widely prepared with graphene oxide (GO) as precursor via various methods. However, little work has been done to detail the structural relationship between GQDs and pristine GO. Herein, we synthesized GQDs through acidic oxidation of GO and separated blue-photoluminescent GQDs (b-GQDs) and green-photoluminescent GQDs (g-GQDs) by a simple dialysis technique. Although the transmission electron microscopy (TEM) and atomic force microscopy (AFM) images reveal their similar morphology, the results of X-ray photoelectron spectroscopy (XPS), Fourier-transform infrared (FTIR) spectroscopy, Raman spectroscopy and zeta potential measurements reveal their distinct structures with different origins from the GO. The b-GQDs may originate from the intact sp² cluster of GO, while the g-GQDs are derived from the relaxed carbon backbone with numerous oxygen-containing functional groups. Besides photoluminescence (PL) properties, the peroxidase-like catalytic activity of the two GQDs was also compared. Interestingly, the g-GQDs exhibit higher peroxidase-like catalytic activity and can be used to detect H₂O₂ with a detection limit of 87 nM, which is lower than most other reported methods. We believe this work provides important insights into the structure, PL properties and potential applications of GO-derived GQDs.

---

1. Introduction

Graphene quantum dots (GQDs), as a new kind of quantum dots, have attracted increasing attention in nanoscience and nanotechnology due to their remarkable properties, such as high surface area, excellent solubility, low cytotoxicity, stable fluorescence and adjustable bandgap.¹ These properties make GQDs fascinating for applications in bioimaging,^2–7 bio- and metal sensing,^8–12 photovoltaics^13–15 and photocatalysts.^16,17

At present, a variety of methods have been developed to prepare GQDs, most of which emit blue photoluminescence (PL) or green PL (here, we label them as b-GQDs and g-GQDs respectively). These methods are usually classified into two main categories, namely, “bottom-up” and “top-down” synthetic strategies.¹ The bottom-up strategy involves the assembly of some small molecules into GQDs.^5,8,13,15 Conversely, the “top-down” route is cutting some big-size carbon sources (such as graphite,^18,19 graphite nanoparticles,^9,20 GO,^{2,7,12,21–27} carbon nanotubes,¹⁹ activated carbon,²⁸ carbon fibers,⁴ carbon black,²⁹ C₆₀ (ref. 30)) into small pieces. Among these carbon sources, GO is probably the most popular one, which has the advantage of low cost, easy processing and mass production. Additionally, due to the abundant oxygen-containing functional groups on GO, GO-derived GQDs are naturally entitled to high hydrophilicity and easy to functionalize. Therefore, GO serves as an excellent precursor and is widely used in preparing GQDs. For example, in 2010, Pan et al. first reported the hydrothermal synthesis of b-GQDs with micrometer-sized GO sheets as the starting material.²¹ Later in 2011, they improved this hydrothermal approach to prepare g-GQDs with excellent photostability.⁷ Besides the hydrothermal approach, Zhu et al. developed a solvothermal method for the synthesis of strongly luminescent g-GQDs from GO.² Li et al. proposed a facile microwave-assisted approach for the preparation of stabilizer-free g-GQDs from GO nanosheets. When the g-GQDs were further reduced with NaBH₄, bright b-GQDs were obtained.²² Subsequently, Zhu et al. reported a highly efficient ultrasonic strategy to synthesize the b-GQDs with high quantum yield from GO.¹² Teng et al. found that g-GQDs exhibiting excitation-wavelength independent PL property could be produced by mildly oxidizing GO sheets.²⁶ Very recently, in the work reported by Lin et al., b-GQDs with tunable PL intensity have also been fabricated by pulsed laser ablation in GO solution.²⁵ All these works illustrate that GO is extensively used and plays a significant role in preparing GQDs with various PL properties. However, many related fundamental issues are still waiting for further exploration. First, although various methods have been developed, the PL colour of the GO-derived GQDs is always blue or green. What is the disparity in structure of these two PL emitting GQDs? Secondly, how do we interpret their distinct PL properties brought by their unique structure? In other words, what is the origin of their PL? Thirdly, is there any other difference between them except their different PL properties?

To figure out these questions, we prepared and separated the b-GQDs and g-GQDs using GO as precursor. After that, the morphology and structure of them were systematically studied. Based on these data, we gave a possible explanation for the origin of the b-GQDs and g-GQDs. Furthermore, we investigated their different PL properties which can be explained by their structural and compositional variations. Finally, for the first time, the peroxidase-like catalytic activity of these two GQDs and their precursor was compared. It was found that, compared to b-GQDs and GO, g-GQDs have an excellent peroxidase-like activity, which can be used to detect H₂O₂.

2. Experimental

2.1 Chemicals and materials

Graphite powder was obtained from Sigma-Aldrich Chemical Co. (USA). H₂SO₄, HNO₃, Na₂CO₃ and H₂O₂ were purchased from Sinopharm Chemical Reagent Co., Ltd. (China). 3,3′,5,5′-Tetramethylbenzidine (TMB) and dialysis bags (material: regenerated cellulose, retained molecular weight: 1000 Da and 100 Da) were purchased from Shanghai Sangon Biotechnology Co., Ltd. (China). All reagents were of analytical grade and used as received without any further purification. Milli-Q ultrapure water with a resistivity of 18.2 MΩ cm was used in all experiments.

2.2 Preparation and separation of b-GQDs and g-GQDs

GO was synthesized from purified natural graphite powder by a modified Hummers method.^10,31 To prepare the GQDs, GO (0.200 g) was added into a mixture of concentrated H₂SO₄ (30 mL) and HNO₃ (10 mL). The mixture was sonicated for two hours and stirred for 3 hours at 100 °C. After being cooled down to room temperature, the product was diluted with ultrapure water and the pH was tuned to 7 with Na₂CO₃ in an ice-bath. Then the mixture was filtered through the polytetrafluoroethylene (PTFE) membrane with average pore size of 0.22 μm to remove the residual GO. Subsequently, a light yellow solution was obtained. Afterwards, the solution was dialyzed in a dialysis bag (MWCO: 100 Da) for 2 days to remove the NaNO₃ and Na₂SO₄ formed during the neutralization process. Finally, the GQDs water dispersion were separated into two portions by further dialysis (MWCO: 1000 Da) for another 2 days. Interestingly, the part of the GQDs retained in the dialysis bag exhibit green photoluminescence under UV light (g-GQDs, inside the 1000 Da dialysis bag), while the other part of the GQDs collected outside the dialysis bag emit blue emission (b-GQDs, outside the 1000 Da dialysis bag).

2.3 Characterization

Transmission electron microscopy (TEM) and high resolution TEM (HRTEM) measurements were performed using a JEOL-2010 (Jeol, Japan) at an accelerating voltage of 200 kV. Atomic force microscopy (AFM) images were taken out using a Nanoscope III a multimode atomic force microscope (Veeco Instruments, USA) in tapping mode to simultaneously collect height and phase data. X-ray photoelectron spectroscopy (XPS) data was obtained on a Thermo VG ESCALAB 250 spectrometer (East Grinsted, UK) with Al Kα radiation at 1486.6 eV. Fourier-transform infrared (FTIR) spectra were obtained on a Vertex70 analyzer (Bruker, Co., Ltd., Germany) with KBr pellet in the 400–4000 cm⁻¹ region. Raman spectra were collected from a LabRAM HR Raman Spectrometer (HORIBA Jobin-Yvon, France) using a 633 nm laser. UV-vis absorption spectra were measured on a Perkin-Elmer Lambda 35 UV-vis spectrophotometer (Perkin-Elmer, USA). PL spectra were collected by using a Hitachi F-4600 fluorometer (Hitachi Co. Ltd., Japan). PL lifetimes were measured with an Edinburgh Instrument FLS920 spectrometer (Edinburgh Instruments, UK). Zeta potential of the nanoparticles was measured using a BI-200SM instrument (Brookhaven Instruments, USA).

2.4 Comparing peroxidase-like catalytic activity of GO, b-GQDs and g-GQDs

For comparing peroxidase-like catalytic activity of GO, b-GQDs and g-GQDs, we performed the kinetic measurements under the optimal conditions as previously reported.¹¹ In a typical experiment, 10 μL of 0.5 mg mL⁻¹ GO and 20 μL of 500 mM H₂O₂ were added into 150 μL of 0.2 M acetate buffer (pH = 4.0). The measurement was started by addition of 20 μL of 8 mM TMB. The variations in the characteristic absorption of oxidized TMB at 652 nm with time were recorded on a Synergy™ 4 multi-mode microplate reader (Bio-Tek, Burleigh, Australia). For the case of b-GQDs and g-GQDs, similar procedures were performed and absorption–time curves were recorded for the purpose of comparison.

2.5 Detection of H₂O₂ based on the catalytic activity of g-GQDs

10 μL of 0.5 mg mL⁻¹ g-GQDs and 20 μL of different concentrations of H₂O₂ were added into 150 μL of 0.2 M acetate buffer (pH = 4.0). The measurements were also started by addition of 20 μL of 8 mM TMB. The absorption–time curves were recorded as described in the above section. The final concentrations of H₂O₂ in the system were varied from 100 nM to 0.5 mM.

3. Results and discussion

3.1 Characterization of the morphology of the GO, b-GQDs and g-GQDs

First, the original GO was characterized by atomic force microscopy (AFM) (Fig. S1†). The height profiles along the lines in Fig. S1b and c† show that the thickness of the GO is about 1.5 (±0.3) nm, suggesting the sheet is bi-layered.^32–34 The result can be also confirmed by the high resolution transmission electron microscope (HRTEM) analysis of the GO edges (Fig. S2†).

TEM and AFM characterization techniques were conducted for the obtained GQDs as well. As shown in the TEM images (Fig. 1a and d), both GQDs are round shape, but the b-GQDs are slightly smaller in size than that of g-GQDs. Statistical analysis indicates that the b-GQDs have the average diameter of 3.0 (±1.0) nm (Fig. 1c), while the g-GQDs possess the average diameter of 4.1 (±1.9) nm (Fig. 1f). In addition, HRTEM image and the corresponding fast Fourier transform (FFT) pattern of the b-GQDs reveal that they are crystalline with a lattice distance of 0.21 nm, corresponding to the (100) plane of graphite (Fig. 1b).^10,35 In contrast, most of the g-GQDs are amorphous carbon nanoparticles without any discernible lattice fringes (Fig. 1e).

Fig. 1 (a) and (d) TEM images of the b-GQDs and g-GQDs. (b) and (e) High resolution TEM images of the b-GQDs and g-GQDs and corresponding 2D FFT images. (c) and (f) Diameter distribution of the b-GQDs and g-GQDs.

Fig. 2a–f show the AFM images, height profiles and height distributions of b-GQDs and g-GQDs. It can be clearly seen that the height of the majority of these GQDs is among 1–2 nm, which is similar to that of GO. However, the average height of b-GQDs (1.3 nm) and g-GQDs (1.6 nm) is different. Interestingly, the b-GQDs are a little thinner than GO, while the g-GQDs are just thicker. Since the interlayer spacing of GO is about 0.7 nm,^32–34 such a little difference (0.3 nm) in height between the b-GQDs and g-GQDs should not be ascribed to their different number of layers, and according to the work reported by Huang et al., it may result from the disparity of surface functional groups.¹⁵

Fig. 2 (a) and (d) AFM images of the b-GQDs and g-GQDs. (b) and (e) The height profiles along the line of panel (a) and (d). (c) and (f) Height distribution of the b-GQDs and g-GQDs.

3.2 Characterization of the composition and structure of the GO, b-GQDs and g-GQDs

To verify the difference in functionalization of the two samples, the chemical composition of them was characterized by X-ray photo-electron spectroscopy (XPS) technique. As shown in Fig. 3a, the XPS results show a predominant C 1s peak at ca. 284.5 eV and an O 1s peak at ca. 532.0 eV for all the three samples,^4,27 and the C/O atomic ratio of the GO, b-GQDs and g-GQDs are estimated to be 2.04, 3.82 and 1.89, respectively. In the meanwhile, the deconvoluted C 1s XPS spectra in Fig. 3b–d show that the GO and the two GQDs all have three components with varied proportions at ca. 284.5 eV, 286.3 eV and 288.2 eV, corresponding to C–C/C=C in aromatic rings, C–O (epoxy and alkoxy), and C=O (carbonyl and carboxyl) groups, respectively.^8,24,27 It can be seen in Table 1 that the pristine GO is functionalized with a large amount of C–O and C–O–C groups (40.49%). In contrast, the content of these groups in both GQDs decrease dramatically (23.78% for g-GQDs and 18.13% for b-GQDs), implying C–O or C–O–C can serve as chemically reactive sites during the oxidative cutting of GO.^21,22 On the other hand, the proportion of C–C and C=C in b-GQDs (69.04%) is much higher than that in g-GQDs (53.39%), while the amount of COOH and C=O groups is lower (12.83% for b-GQDs and 23.23% for g-GQDs). This is in accordance with the results of C/O atomic ratio of the two samples, demonstrating the b-GQDs are less oxidized.

Fig. 3 (a) XPS survey spectra of the GO, b-GQDs and g-GQDs. (b)–(d) High-resolution XPS C 1s spectra of the GO, b-GQDs and g-GQDs.

Table 1 XPS analysis and zeta potentials of GO, g-GQDs and b-GQDs

Samples	C–C and C=C (%)	C–O and C–O–C (%)	C=O and COOH (%)	Zeta potential (mV)
GO	50.57	40.49	8.94	−26.0
g-GQDs	53.39	23.78	23.23	−18.8
b-GQDs	69.04	18.13	12.83	−8.9

---

To further probe the difference in oxygen-related functional groups between the GO and the two kinds of GQDs, we performed Fourier transform infrared spectroscopy (FTIR) on these samples. In the FTIR spectra (Fig. 4), GO shows the stretching vibrations of O–H (3420 cm⁻¹), C–H (2928 and 2850 cm⁻¹), COOH (1730 cm⁻¹), C=O (1630 cm⁻¹), C–O–C (1231 cm⁻¹), C–OH (1056 cm⁻¹) and bending vibrations of CH₂ (1389 cm⁻¹), C–O–C (858 cm⁻¹).^36,37 In comparison with the GO, the FTIR spectra of the two GQDs display obvious change. On the one hand, the absorption band of C–O–C nearly disappears, corresponding to the decrease of amount of C–O or C–O–C in XPS measurements. On the other hand, the relative intensity of C=O and C–OH absorption band (versus C–H absorption band) increases, suggesting that the GQDs are obtained via cutting C–O–C group of GO into C=O or C–OH. Additionally, there is a distinct difference in FTIR spectra between the two GQDs. For g-GQDs, they retain the COOH group absorption band (1730 cm⁻¹) from the GO, instead, that of b-GQDs is nearly absent. Since the COOH group is basically distributed on the edge of GO,^38,39 it can be concluded that the b-GQDs should stem from the interior zone of GO.

Fig. 4 FTIR data for GO, b-GQDs and g-GQDs.

The variance in the oxygen-containing functional groups among the GO, g-GQDs and b-GQDs strongly implies their difference in structure. To further confirm this, Raman spectroscopy was performed. Commonly, there are two characteristic bands in the Raman spectrum of carbon-based material: D-band and G-band. The former is related to sp³ hybridized carbon atoms and to the presence of structural defects, while the latter is thought to be associated with sp² hybridized carbon atoms. Thus the intensity of D-band to G-band (I_D/I_G) can be used to estimate the crystalline quality of these materials.^30,40 As shown in Fig. 5, all the three samples display both the two bands (the D-band at ∼1340 cm⁻¹ and G-band at ∼1585 cm⁻¹), and the I_D/I_G (integrated intensity) ratio of the GO, g-GQDs and b-GQDs are calculated to be 1.08, 1.26 and 1.00, respectively. This result indicates all of these samples are composed of the sp³ and sp² hybridized carbon with different ratio. Apparently, the b-GQDs have the lowest I_D/I_G ratio and the highest level of crystallinity among the three samples, which can also be supported by the distinct lattice fringe image and hexagonal 2D FFT pattern above (Fig. 1b).

Fig. 5 Raman data for the GO, b-GQDs and g-GQDs.

3.3 The derivation of b-GQDs and g-GQDs

Based on the above data and corresponding analysis, we attempt to trace the derivation of the two GQDs. Generally, GQDs made by chemical oxidative method will have higher degree of oxidation than their precursors.^{4,9,18,20,29,30} However, our XPS results indicate that the as-prepared b-GQDs exhibit higher C/O atomic ratio than that of the pristine GO. A similar phenomenon was also observed by Li and Zhou et al., who suggested that reduction process occurred during the microwave or UV irradiation treatment.^22,24 Since neither of these treatments was involved in our experiment, we propose another possible explanation that the b-GQDs and g-GQDs may be produced from different parts of the GO. Specifically, the b-GQDs should be “extracted” from the intact sp² cluster of GO, while the g-GQDs should be derived from its matrix where the carbon backbone were covered by numerous oxygen-containing functional groups. In this way, the GO would display a higher oxidation degree than the b-GQDs, but lower than the g-GQDs. For the same reason, the g-GQDs would show higher I_D/I_G ratio than GO, while that of the b-GQDs is lower, which is in agreement with our experimental data. In addition, the distinguished difference in the degree of crystallinity (Fig. 1b and e) and functional groups (Fig. 4) of the two samples can also support our statements. Nevertheless, another fundamental question still needs to be answered, how can the two GQDs be effectively separated by simple dialysis technique? We believe there are two main reasons. On the one hand, as shown in the TEM image, the lateral size of the g-GQDs (∼4.1 nm) is slightly bigger than that of the b-GQDs (∼3.0 nm), thus they tend to be retained in the dialysis bag. On the other hand, XPS results reveal that the g-GQDs (with C/O atomic ratio of 1.89) have much higher level of oxidation degrees than the b-GQDs (with C/O atomic ratio of 3.82), so there will be more static charge present on the surface of them, which can be proved by the zeta potential values of −18.8 mV for g-GQDs and −8.9 mV for b-GQDs (Table 1). Accordingly, when dispersed in a polar solvent like water, individual g-GQDs will be surrounded by a thicker solvation shell, and the effective size (hydrodynamic radius) of g-GQDs will be even larger than b-GQDs. For these two main factors, the g-GQDs can be kept inside the dialysis bag, while the smaller b-GQDs are more likely to penetrate through the dialysis membrane. It is noteworthy that Liu et al. also obtained two kinds of GQDs adopting the similar separation methods, although the precursor (carbon nano-onions) used in preparing GQDs is different from ours. In a word, there are two kinds of GQDs exist in the product of chemical oxidative cutting of GO. The b-GQDs with high crystallinity and little defects are inherited from the sp² cluster of the GO, while the poorly crystallized and highly oxidized g-GQDs are the derivates from the matrix of the GO. The successful separation of them by dialysis can be attributed to their difference in size and oxidation degree. And the schematic representation of the preparation route is presented in the Scheme 1.

Scheme 1 Schematic representations of the preparation and separation route for the g-GQDs and b-GQDs.

3.4 Optical properties of the b-GQDs and g-GQDs

To study the optical properties of the two GQDs, UV-visible (UV-vis) absorbance, PL emission, and PL excitation (PLE) spectroscopy were investigated. As shown in Fig. 6, both GQDs display strong absorbance in deep UV range (<240 nm), corresponding to π–π* transition of aromatic C=C bonds.⁹ Besides, the b-GQDs exhibit two vague absorption shoulders at ∼265 nm and ∼330 nm, while the g-GQDs only show a small shoulder at around 300 nm, which is previously ascribed to n–π* transition of C=O or COOH groups.^9,19

Fig. 6 Comparison between UV-vis absorbance for the b-GQDs and g-GQDs.

In PL spectra (Fig. 7a), b-GQDs exhibit an optimal emission wavelength at 425 nm, and the PL excitation (PLE) spectrum recorded with this wavelength shows two distinct peaks at 265 nm and 335 nm, which corresponds to the 265 nm and 330 nm absorption band of the b-GQDs, respectively. It is worth noting that the GO reduced by hydrazine-vapor and the GQDs prepared by hydrothermal cutting reduced graphene sheets can also show similar PL and PLE spectrum to that of the as-prepared b-GQDs,^21,34 implying their PL origin is identical and must be related to their high-crystalline structure. In the case of g-GQDs (Fig. 7b), the value of optimal emission wavelength red shift to 500 nm, and a broad excitation band with the maximum at 295 nm can be observed when detected with this emission wavelength. The 295 nm PLE peak corresponds to the vague absorption shoulder at 300 nm, suggesting the C=O or COOH groups play a significant role in the PL of the g-GQDs.

Fig. 7 (a) and (b) Comparison between PL excitation and emission spectra of the b-GQDs and g-GQDs. The inset in (a) and (c) are photographs of the g-GQDs and b-GQDs aqueous solution taken under visible light and 365 nm UV light. (c) and (d) Excitation dependent emission spectra of the b-GQDs and g-GQDs.

As many previous studies reported,^{2–4,7,21–23,29,35} the excitation-dependent PL behaviour is one of the most common features for GQDs, which is different from that of the traditional fluorescent dye molecule. However, recently, more and more work reported that the excitation-independent PL behaviour could also been observed from GQDs.^9,17,26 Interestingly, the as-prepared b-GQDs and g-GQDs just exhibit excitation dependent/independent PL behaviour, respectively.

When the excitation wavelength is changed from 230 to 430 nm, the PL peak of the b-GQDs shifts to longer wavelength constantly (Fig. 7c), while, the peak position of the g-GQDs is almost invariable at ca. 500 nm (Fig. 7d). Generally, the origin of the excitation-dependent PL behaviour can be explained by the inhomogeneity of the size and emissive sites of GQDs.^2,16,29 Whereas, our statistical results in Fig. 1c and f show that the g-GQDs (2–6 nm) have wider size distribution compared with b-GQDs (2–4 nm), but display a molecular type of PL, spectrally invariable with the change of particle size or excitation energy.

Interestingly, unlike the traditional fluorescent molecule, g-GQDs can also show an excellent photostability. As shown in Fig. S3a,† upon 295 nm excitation, no photobleaching is observed for 1200 s in a fluorescence spectrophotometer. According to the study reported by Wang et al., such a prominent photostability can be attributed to the graphitic carbon backbone, which plays a significant role of temporary reservoir for photogenerated carriers, enhancing the capability of resistance to photobleaching.³⁶ Although our HRTEM results reveal that the g-GQDs are poorly crystalline, the appearance of G-band at ∼1580 cm⁻¹ in Raman spectra can definitely verify the existence of graphitic carbon in them. As a result, the g-GQDs display such a high photostability. As the b-GQDs had a higher degree of crystallinity than g-GQDs, it stands to reason that the b-GQDs possess a remarkable capability to resist photobleaching as well (Fig. S3b†).

3.5 PL mechanism of the b-GQDs and g-GQDs

As is well-known, the origin of luminescence from GQDs is still an open question, and it is tentatively suggested to be related to excitons of carbon,⁴¹ emissive traps,⁴² aromatic structures,² free zig-zag sites^21,35 and an integration of emissions from the aromatic core and functional groups.⁴³ Additionally, GQDs prepared via various methods are demonstrated to possess different mechanism.¹ Interestingly, the above optical characterization results imply that the g-GQDs and b-GQDs prepared by same methods can also have different luminescence mechanism. To further examine this disparity, we performed time-resolved photoluminescence (TRPL) spectroscopy (see Fig. 8 and Table 2). The decays were monitored at their optimal emission wavelengths, and the best-fit parameters were listed in Table 2. It can be seen both samples show bi-exponential decay curves with fast (<1 ns) and slow (>4 ns) decay components, though the proportion of them is just opposite. For b-GQDs, the fast decay component is dominant (0.977 ns for 80.3%), while the slow decay component dominates the emission from g-GQDs (7.42 ns for 84.8%). Since the emission that originates from intrinsic states usually shows a shorter recombination lifetime than that from defect states,^3,15,20,36 it can be stated that the luminescence of the b-GQDs mainly originates from intact sp² carbon core, while that of the g-GQDs can be basically attributed to surface defect.

Fig. 8 TRPL spectra of the b-GQDs and g-GQDs.

Table 2 Multi-exponential fitting for time-resolved PL

Samples	τ1 (ns)	τ2 (ns)	τave (ns)
b-GQDs	0.977(80.3%)	4.71(19.7%)	1.71
g-GQDs	0.956(15.2%)	7.42(84.8%)	6.44

---

By piecing together the conclusion from TRPL measurements with the direct experimental evidence from structural and compositional characterization, we give an interpretation for their disparity in PL properties. For b-GQDs, they are well crystalline with more intact sp² clusters (Fig. 1b), thus the sp² clusters (the intrinsic states) play the leading roles in their emission. Since the size of these sp² clusters in different GQDs is varying, the b-GQDs show excitation-dependent PL behaviour overall. By comparison, the XPS (Fig. 3) and FTIR (Fig. 4) results reveal that g-GQDs are covered with more oxygen-containing functional groups (especially for C=O and COOH) which would introduce disorder-induced localized states within the gap of the sp² cluster, making the emission red-shift.⁴⁴ Meanwhile, these functional groups, as Wang et al. reported, would also result in a molecular type of photoluminescence.⁴³ Therefore, the g-GQDs display green emission with excitation-independent PL behaviour. In short, the luminescence of the b-GQDs should originate from intrinsic states in the high-crystalline structure, while the luminescence of g-GQDs could be attributed to defect states with oxygen-containing functional groups.

3.6 Peroxidase-like catalytic activity of the GO, b-GQDs and g-GQDs

Apart from the optical properties, we also studied the peroxidase-like catalytic ability of the two GQDs. It was reported that GO and carbon nanodots possess significant peroxidase-like activity to produce a blue color reaction in the presence of H₂O₂ and a peroxidase substrate, 3,3′,5,5′-tetramethylbenzidine (TMB).^11,45,46 However, to our best knowledge, little work has been done to investigate the activity of different GQDs. Inspired by this, the catalytic abilities of our b-GQDs, g-GQDs and GO were compared, under the optimal conditions as previously reported.¹¹

Fig. 9 shows the time-dependent absorbance (monitored at 652 nm) changes of the TMB–H₂O₂ system under different conditions. And the corresponding photographs of different systems recorded at 600 s are presented in Fig. S4.† For the blank TMB/H₂O₂ system (Fig. 9, curve a), a small absorbance change can be observed within 600 s due to their low reaction rate.¹¹ In contrast, upon the addition of b-GQDs, g-GQDs and GO, the absorbance intensity of the corresponding systems (Fig. 9, curve b, c, d) increases respectively, but the rate of rise in absorbance varied for different systems. Obviously, the change of TMB/H₂O₂/g-GQDs system is most remarkable. Moreover, other three control experiments, including TMB/b-GQDs, TMB/g-GQDs and TMB/GO systems (Fig. 9, curve e, f, g), were conducted as well. These results show that all of them display negligible change in absorbance within 600 s.

Fig. 9 The time-dependent absorbance changes of the system under different conditions.

The fast colorimetric responses for TMB/H₂O₂/b-GQDs (Fig. 9, curve b) and TMB/H₂O₂/g-GQDs (Fig. 9, curve c) system indicate that both GQDs possess the peroxidase-like activity. However the catalytic ability of g-GQDs is apparently higher than that of the b-GQDs. Such an obvious difference is related to their different structure. While the TMB tend to interact with the aromatic backbone on the surface of GQDs through π–π interactions, the oxygen-containing functional groups on the edge of GQDs, such as COOH and C=O groups, should play a more significant role in facilitating the interaction between the GQDs and H₂O₂.^45,46 Via the C 1s XPS spectra of them (Fig. 3), we can find the proportion of COOH and C=O groups in g-GQDs (23.23%) is much higher than that in b-GQDs (12.83%). In the meanwhile, as the FTIR results in Fig. 4 revealed, the typical COOH peak at 1730 cm⁻¹ are observed from the g-GQDs, while there are no distinct corresponding peak on the b-GQDs. Thus, the functional groups on edge of the g-GQDs, especially the COOH group, may play an important role in exhibiting the catalytic property.

On the basis of this property of g-GQDs, we explored the colorimetric method for the detection of H₂O₂, using the g-GQDs catalyzed blue colour reaction. As shown in Fig. 10a, the time-dependent absorbance curves display a H₂O₂ concentration-dependent behaviour. To be specific, as the concentration of H₂O₂ is increased from 0.1 μM to 500 μM, the absorbance intensity (recorded at 600 s) increase gradually (Fig. 10b). The corresponding photograph is presented in Fig. S5.† Meanwhile, the Fig. 10b inset shows that the linear range is from 100 nM to 10 μM. And the detection limit is calculated to be 87 nM, which is much lower than that of many other nanomaterials-catalyzed methods.^6,8,45 Therefore, the g-GQDs may have tremendous potential for applications in biosensing field.

Fig. 10 (a) The time-dependent absorbance changes upon analyzing different concentrations of H₂O₂ (0.1, 0.2, 0.5, 1, 2, 5, 10, 20, 50, 100, 200, and 500 μM). (b) Calibration curve of the H₂O₂ sensing using the absorbance intensity at 10 min of the system as the ordinate and the concentrations of H₂O₂ as the abscissa. The inset is magnification of the plot in the range 0.1–10 μM.

4. Conclusions

In summary, we have proved that there were basically two kinds of GO-derived GQDs with different structures through the analysis of HRTEM images, XPS, FTIR and Raman spectra etc. In this work, the b-GQDs should be inherited from high-crystalline sp² cluster of GO, while the highly oxidized g-GQDs should be derived from the other part of GO with rich oxygen-containing functional groups. Optical characterization results reveal that the luminescence of the b-GQDs actually originates from intrinsic states of graphene, while that of the g-GQDs may be attributed to defect states of graphene related to oxygen groups. Due to the unique structure, g-GQDs exhibit much higher peroxidase-like catalytic activity than b-GQDs and can be used to detect H₂O₂ with the detection limit of 87 nM, which is lower than that of most other reported methods. We believe our work will provide important insights into the structure, PL properties and potential applications of GO-derived GQDs.

Acknowledgements

The authors gratefully acknowledge the financial from the National Natural Science Foundation of China (No. 21252001, 21473204), the Special Project of National Major Scientific Equipment Development of China (No. 2012YQ120060), the Natural Science Foundation of Fujian Province (2015J01070), Science and Technology Planning Project of Fujian Province, Grant No. 2014H2008.

References

1. L. L. Li, G. H. Wu, G. H. Yang, J. Peng, J. W. Zhao and J. J. Zhu, Nanoscale, 2013, 5, 4015–4039.
2. S. J. Zhu, J. H. Zhang, C. Y. Qiao, S. J. Tang, Y. F. Li, W. J. Yuan, B. Li, L. Tian, F. Liu and R. Hu, Chem. Commun., 2011, 47, 6858–6860.
3. S. J. Zhu, J. H. Zhang, S. J. Tang, C. Y. Qiao, L. Wang, H. Y. Wang, X. Liu, B. Li, Y. F. Li and W. L. Yu, Adv. Funct. Mater., 2012, 22, 4732–4740.
4. J. Peng, W. Gao, B. K. Gupta, Z. Liu, R. Romero-Aburto, L. H. 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.
5. L. Wang, Y. L. Wang, T. Xu, H. B. Liao, C. J. Yao, Y. Liu, Z. Li, Z. W. Chen, D. Y. Pan, L. T. Sun and M. H. Wu, Nat. Commun., 2014, 5, 5357.
6. X. Wu, F. Tian, W. X. Wang, J. Chen, M. Wu and J. X. Zhao, J. Mater. Chem. C, 2013, 1, 4676–4684.
7. D. Y. Pan, L. Guo, J. C. Zhang, C. Xi, Q. Xue, H. Huang, J. H. Li, Z. W. Zhang, W. J. Yu and Z. W. Chen, J. Mater. Chem., 2012, 22, 3314–3318.
8. S. Umrao, M. H. Jang, J. H. Oh, G. Kim, S. Sahoo, Y. H. Cho, A. Srivastva and I. K. Oh, Carbon, 2015, 81, 514–524.
9. Y. Y. Liu and D. YoungáKim, Chem. Commun., 2015, 51, 4176–4179.
10. J. J. Liu, Z. T. Chen, D. S. Tang, Y. B. Wang, L. T. Kang and J. N. Yao, Sens. Actuators, B, 2015, 212, 214–219.
11. A. X. Zheng, Z. X. Cong, J. R. Wang, J. Li, H. H. Yang and G. N. Chen, Biosens. Bioelectron., 2013, 49, 519–524.
12. Y. H. Zhu, G. F. Wang, H. Jiang, L. Chen and X. J. Zhang, Chem. Commun., 2015, 51, 948–951.
13. X. Yan, X. Cui, B. S. Li and L. S. Li, Nano Lett., 2010, 10, 1869–1873.
14. Y. Li, Y. Hu, Y. Zhao, G. Q. Shi, L. Deng, Y. B. Hou and L. T. Qu, Adv. Mater., 2011, 23, 776–780.
15. Z. C. Huang, Y. T. Shen, Y. Li, W. J. Zheng, Y. J. Xue, C. Q. Qin, B. Zhang, J. J. Hao and W. Feng, Nanoscale, 2014, 6, 13043–13052.
16. H. T. Li, X. D. He, Z. H. Kang, H. Huang, Y. Liu, J. L. Liu, S. Y. Lian, C. H. A. Tsang, X. B. Yang and S. T. Lee, Angew. Chem., Int. Ed., 2010, 49, 4430–4434.
17. S. J. Zhuo, M. W. Shao and S. T. Lee, ACS Nano, 2012, 6, 1059–1064.
18. Y. H. Shin, J. Lee, J. Yang, J. Park, K. Lee, S. Kim, Y. Park and H. Lee, Small, 2014, 10, 866–870.
19. L. X. Lin and S. W. Zhang, Chem. Commun., 2012, 48, 10177–10179.
20. F. Liu, M. H. Jang, H. D. Ha, J. H. Kim, Y. H. Cho and T. S. Seo, Adv. Mater., 2013, 25, 3657–3662.
21. D. Y. Pan, J. C. Zhang, Z. Li and M. H. Wu, Adv. Mater., 2010, 22, 734–738.
22. L. L. Li, J. Ji, R. Fei, C. Z. Wang, Q. Lu, J. R. Zhang, L. P. Jiang and J. J. Zhu, Adv. Funct. Mater., 2012, 22, 2971–2979.
23. J. H. Shen, Y. H. Zhu, X. L. Yang, J. Zong, J. M. Zhang and C. Z. Li, New J. Chem., 2012, 36, 97–101.
24. 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.
25. T. N. Lin, K. H. Chih, C. T. Yuan, J. L. Shen, C. A. J. Lin and W. R. Liu, Nanoscale, 2015, 7, 2708–2715.
26. C. Y. Teng, T. F. Yeh, K. I. Lin, S. J. Chen, M. Yoshimura and H. Teng, J. Mater. Chem. C, 2015, 3, 4553–4562.
27. V. Štengl, S. Bakardjieva, J. Henych, K. Lang and M. Kormunda, Carbon, 2013, 63, 537–546.
28. Y. Q. Dong, N. N. Zhou, X. M. Lin, J. P. Lin, Y. W. Chi and G. N. Chen, Chem. Mater., 2010, 22, 5895–5899.
29. Y. Q. Dong, C. Q. Chen, X. T. Zheng, L. L. Gao, Z. M. Cui, H. B. Yang, C. X. Guo, Y. W. Chi and C. M. Li, J. Mater. Chem., 2012, 22, 8764–8766.
30. C. K. Chua, Z. k. Sofer, P. Šimek, O. e. Jankovský, K. i. Klímová, S. Bakardjieva, S. t. p. n. Hrdličková Kučková and M. Pumera, ACS Nano, 2015, 9, 2548–2555.
31. W. S. Hummers Jr and R. E. Offeman, J. Am. Chem. Soc., 1958, 80, 1339.
32. X. M. Sun, Z. Liu, K. Welsher, J. T. Robinson, A. Goodwin, S. Zaric and H. J. Dai, Nano Res., 2008, 1, 203–212.
33. X. M. Sun, D. C. Luo, J. F. Liu and D. G. Evans, ACS Nano, 2010, 4, 3381–3389.
34. G. Eda, Y. Y. Lin, C. Mattevi, H. Yamaguchi, H. A. Chen, I. Chen, C. W. Chen and M. Chhowalla, Adv. Mater., 2010, 22, 505–509.
35. S. J. Zhu, J. R. Shao, Y. B. Song, X. H. Zhao, J. L. Du, L. Wang, H. Y. Wang, K. Zhang, J. H. Zhang and B. Yang, Nanoscale, 2015, 7, 7927–7933.
36. L. Wang, S. J. Zhu, H. Y. Wang, S. N. Qu, Y. L. Zhang, J. H. Zhang, Q. D. Chen, H. L. Xu, W. Han and B. Yang, ACS Nano, 2014, 8, 2541–2547.
37. M. Acik, G. Lee, C. Mattevi, M. Chhowalla, K. Cho and Y. Chabal, Nat. Mater., 2010, 9, 840–845.
38. A. Lerf, H. He, M. Forster and J. Klinowski, J. Phys. Chem. B, 1998, 102, 4477–4482.
39. T. Szabó, O. Berkesi, P. Forgó, K. Josepovits, Y. Sanakis, D. Petridis and I. Dékány, Chem. Mater., 2006, 18, 2740–2749.
40. L. G. Cançado, A. Jorio, E. M. Ferreira, F. Stavale, C. Achete, R. Capaz, M. Moutinho, A. Lombardo, T. Kulmala and A. Ferrari, Nano Lett., 2011, 11, 3190–3196.
41. J. G. Zhou, C. Booker, R. Y. Li, X. T. Zhou, T. K. Sham, X. L. Sun and Z. F. Ding, J. Am. Chem. Soc., 2007, 129, 744–745.
42. V. N. Mochalin and Y. Gogotsi, J. Am. Chem. Soc., 2009, 131, 4594–4595.
43. S. J. Wang, I. S. Cole, D. Y. Zhao and Q. Li, Nanoscale, 2016, 8, 7449–7458.
44. C. T. Chien, S. S. Li, W. J. Lai, Y. C. Yeh, H. A. Chen, I. Chen, L. C. Chen, K. H. Chen, T. Nemoto and S. Isoda, Angew. Chem., Int. Ed., 2012, 51, 6662–6666.
45. Y. J. Song, K. G. Qu, C. Zhao, J. S. Ren and X. G. Qu, Adv. Mater., 2010, 22, 2206–2210.
46. W. B. Shi, Q. L. Wang, Y. J. Long, Z. L. Cheng, S. H. Chen, H. Z. Zheng and Y. M. Huang, Chem. Commun., 2011, 47, 6695–6697.

---

Footnote

† Electronic supplementary information (ESI) available: The topographic analysis of GO; the TEM and HRTEM image of GO; the photostability tests of b-GQDs and g-GQDs; the photographs of the TMB–H₂O₂ system under different conditions. See DOI: 10.1039/c5ra26279h

---