Effects of elemental doping on the photoluminescence properties of graphene quantum dots

Abstract

Heteroatom doping can endow graphene quantum dots (GQDs) with enhanced photoluminescence (PL) properties. With the aim of understanding the PL mechanism of GQDs and improving their PL properties, a simple strategy was used for the synthesis of GQDs and four kinds of element-doped GQDs (N/S/P/B). Compared to GQDs, nitrogen and sulfur-doping result in a blue-shifted PL emission, but phosphorus and boron doping lead to a red-shifted emission. Experimental results indicate that element doping can change structural defects, surface functional groups (especially carboxyl groups), and interactions among carbon atoms with their neighboring atoms of carbon, oxygen, and doped atoms. All the changes above can tailor the PL properties of GQDs for various applications, such as biomedical sensing and imaging.

---

1. Introduction

In recent years, great attention has been paid to graphene quantum dots (GQDs) because of their unique structural and optoelectronic properties and their great potential in various applications¹ such as bioimaging,^2–4 medical diagnosis,⁵ catalysis,⁶ and photovoltaic devices.⁷ In particular, these carbon nanoparticles can serve as superior fluorescent reporters for bio-imaging and optical sensing. GQDs are expected to outperform current fluorescent reporters (fluorescent proteins, organic dyes, and semiconductor quantum dots) in several applications because of their unique combination of several key merits, including tunable photoluminescence (PL), superior photostability, excellent biocompatibility in physiological conditions, and ease of bioconjugation. However, the widespread use of GQDs is currently hindered by the lack of controllable synthesis methods and the poor understanding of their tunable PL properties.⁸

Primary approaches for synthesizing GQDs can be classified into two main categories: bottom-up and top-down methods, where the former is limited by complex processes, severe synthetic conditions and difficulty of mass production.⁹ Top-down approach refers to cleavage of carbonaceous materials, including electron beam lithography, oxygen plasma treatment, acidic exfoliation and oxidation, hydrothermal or solvothermal synthesis, and electrochemical oxidation. Among them, acidic exfoliation and oxidation is considered to be an efficient approach to produce GQDs in bulk quantity from abundant raw materials.¹⁰ Moreover, these GQDs are rich in oxygen-related functional groups on their surfaces, which can be helpful to tune their optical properties by surface modification and edge effects.¹¹ However, this method faces some challenges, such as low quantum yield and product yield and the difficulty of removing excess strong acid from the solution. Thus, a facile and efficient synthetic route to producing high-quality GQDs on a large scale is imminently desired. Dong et al.⁴ developed a unique strategy to simultaneously prepare single-layer and multi-layer GQDs by refluxing Vulcan CX-72 carbon black with concentrated nitric acid and separate them by centrifugation. Two types of solutions were obtained, which exhibited green and yellow PL, respectively, under excitation at 365 nm. Nitric acid can be removed from the solution by evaporating it at 200 °C. This facile method shows prominent advantages of low-cost, high yield and mass manufacturability.

To further improve PL performance and enlarge application scope of GQDs, two kinds of modifications, surface functionalization and heteroatom doping, were proposed. It has been proven that surface modification of GQDs with polymers or small organic molecules¹² can greatly increase their fluorescence quantum yields. However, it is usually complex to purify functionalized GQDs despite the fact that these groups would occupy functional positions for specific analytical and sensor purposes.¹³

Along with surface modification, heteroatom doping (most commonly nitrogen doping thus far) is another effective method to fine-tune or obtain new PL of GQDs.⁸ It has been shown that N-doping on GQDs increases quantum yield and causes blue shift in emission due to the electron-withdrawing ability of nitrogen atoms.^14–17 Other elements (e.g., B^18–21 and S²²) have also been doped into GQDs to change PL characteristics or gain catalytic properties. Fan et al.¹⁸ prepared water-soluble B-GQDs by electrochemical exfoliation of graphite in borax electrolyte, and B-GQDs feature intriguing rich fluorescence. Recently, a few studies showed successful preparation of N/P co-doped GQDs,²³ S/N co-doped GQDs^24,25 and N/F/S co-doped GQDs²⁶ with high quantum yield. More studies are urgently needed in this direction, as there are limited studies on heteroatom doping of GQDs despite their potential significance.

Moreover, contradictory hypotheses sometimes arise from inconsistent experimental observations because of large heterogeneity in GQDs synthesized by current methods and the fact that the PL properties of GQDs are intriguingly determined by numerous parameters. The emission of GQDs can be widely tuned from deep ultraviolet to near infrared by size, edge configuration, shape, functional groups, defects, and heterogeneous hybridization of carbon network.⁸

In this study, in order to understand the PL mechanism of heteroatom doped GQDs, we employed the same approaches and similar process parameters to synthesize four different types of elemental (N/S/P/B)-doped GQDs. GQDs were obtained from thermally exfoliated graphite oxide by refluxing with concentrated nitric acid. Four different elemental (N/S/P/B)-doped GQDs were further synthesized using a pyrolysis method with GQDs as C source and melamine, dibenzyl disulfide, triphenylphosphine, and boric acid as N, S, P, and B sources, respectively. For the first time to our knowledge, we conducted systematic experiments to reveal the PL mechanism, especially the effects of elemental doping on PL properties of GQDs.

2. Experimental

2.1. Preparation of GQDs

Graphene oxide (GO) was first dried in a vacuum oven at 60 °C for 24 h just before use. GO powder (0.2 g) was placed in a quartz tube, which was sealed at one end. The other end was closed using a rubber stopper, and an argon (Ar) inlet was inserted through the stopper. The sample was flushed with Ar for 10 min, and then the quartz tube was quickly inserted into a tube furnace preheated to 1050 °C and held in the furnace for 30 s. This was followed by refluxing the powder in HNO₃ (6 M) for 24 h. The solution was then cooled and centrifuged at 5000 rpm for 10 min. The supernatant was heated at 200 °C to evaporate water and HNO₃ and reddish-brown solid was obtained. The resulting powder was subsequently dispersed in deionized water and dialyzed to remove ions and other impurities.

2.2. Preparation of elemental (N/S/P/B)-doped GQDs

N-doped GQDs (N-GQDs) were prepared by pyrolysis of a homogeneous mixture consisting of GQDs and melamine. Briefly, GQDs (50 mg) were dispersed in 100 mL water by ultrasonication. Then, melamine (0.25 g) was added to the GQDs suspension and stirred until significant agglomeration was observed. The resulting mixture was heated at 80 °C to evaporate water, then it was dried in a vacuum oven at 60 °C overnight. The solid GQD–melamine complex was grinded into fine powder using a mortar. The resulting complex was put in a quartz boat in the center of a tube furnace and pyrolysed at 750 °C for 45 min under an Ar atmosphere. The resulting powder was subject to refluxing in HNO₃ (6 M) for 8 h. After that, centrifugation, evaporation and dialysis treatments were conducted the same as for GQDs. S-doped GQDs (S-GQDs), P-doped GQDs (P-GQDs) and B-doped GQDs (B-GQDs) were prepared using similar methods to N-GQDs, but with the addition of their corresponding doping precursor: dibenzyl disulfide (30 mg), triphenylphosphine (TPP, 200 mg), and boric acid (50 mg), respectively.

2.3. Characterization

The morphology and structure of GQDs and doped GQDs were characterized by high-resolution transmission electron microscopy (HRTEM, Tecnai G2 F20) and X-ray diffraction (XRD, Bruker D8 Advance). X-ray photoelectron spectroscopy (XPS) measurements were performed on a Thermo ESCALAB 250XI spectrometer equipped with mono-chromatized Al Kα excitation. Raman spectra were obtained using a laser confocal micro-Raman microscope (Horiba Co., Lab HR800) with an excitation wavelength of 532 nm. UV-Vis spectra were recorded on a UV-3900 spectrophotometer (Hitachi High Technologies). PL spectra were obtained using an F-4600 fluorescence spectrophotometer (Hitachi High Technologies). The quantum yield (Q) was calculated with the following equation. Quinine sulfate in 0.1 M H₂SO₄ (Q = 0.54 at 340 nm) was chose as a standard. Q is quantum yield, I is measured integrated emission intensity, n is refractive index, and A is optical density. The subscript R refers to the reference fluorophore with known quantum yield.

3. Results and discussion

Transmission electron microscopy (TEM) and microanalysis were used to explore atomic structures of GQDs and elemental doped-GQDs (Fig. 1 and Fig. S1 in ESI†).²⁷ As given in Fig. 1, both GQDs and N-GQDs are well dispersed with narrow size distributions and average diameters of 3.7 nm and 2.3 nm, respectively, as measured from 100 individual particles. Compared to GQDs, N-GQDs exhibit smaller lateral size, which can be due to the pyrolysis process. Compared to most GQDs obtained by acidic exfoliation and oxidation method reported previously,^4,28,29 in this study, GQDs exhibit much smaller lateral sizes. Discernible lattice structures of GQDs and N-GQDs indicate that resultant nanoparticles have crystal structures. Representative HRTEM images (inset of Fig. 1a and b) display a lattice spacing distance of 0.21 nm and 0.22 nm, which matches the lattice constant of (100). The results above suggest that the nanoparticles are composed of crystal cores of graphitic sp² carbon atoms. Representative HRTEM images of S/P/B-doped GQDs (insets of Fig. 1c, d and e) demonstrate lattice spacing distances of 0.32 and 0.27/0.25 nm, which can be indexed to graphitic (002) and (100) planes, respectively.

Fig. 1 TEM images of GQDs (a), N-GQDs (b), S-GQDs (c), P-GQDs (d) and B-GQDs (e) and their HRTEM images (insets).

Fig. S2† shows typical XRD patterns of GQDs and doped-GQDs. The characteristic graphitic (002) peaks of GQDs, N-GQDs, S-GQDs, P-GQDs and B-GQDs locate around 22.77, 25.91, 25.03, 24.28 and 25.85 degrees, corresponding to 0.391, 0.344, 0.356, 0.367 and 0.345 nm interlayer spacing, respectively. The interlayer spacing of GQDs is larger than that of graphite (0.334 nm), which can be attributed to oxygen-containing groups introduced in the exfoliation and oxidation process.^4,10,30 The interlayer spacing of N/S/P/B-doped GQDs is slightly larger than that of graphite due to the presence of chemical groups and doping atoms enlarging the basal plane spacing, which is similar to those of doped GQDs prepared by other methods.^{14,16,31–33} Comparing with GQDs, the more compact interlayer spacing in the N/S/P/B-doped GQDs can be due to a more thorough high temperature reduction during the pyrolysis doping process.

Raman spectroscopy was carried out to evaluate the degree of structural deformations in GQDs, N-GQDs, S-GQDs, P-GQDs and B-GQDs. As shown in Fig. 2, all samples exhibit typical D and G bands. In graphene-based materials, the G band corresponds to the sp² hybridized carbon atoms in the hexagonal framework, and the D band is indicative of the sp³ hybridized carbon atoms on the defects and edges of graphene sheets.³⁴ The ratio of the D to G band intensities (I_D/I_G) is generally used as a measure of carbon disorder, and a higher I_D/I_G value indicates more defects in the carbon structure.³⁵ As given in Table S1,† the I_D/I_G value is 1.01, 0.88, 0.86, 0.98 and 1.21 for GQDs, N-GQDs, S-GQDs, P-GQDs and B-GQDs, respectively, with the order for I_D/I_G value being B-GQDs > GQDs > P-GQDs > N-GQDs > S-GQDs. This means the order for the number of defects in carbon structures is B-GQDs > GQDs > P-GQDs > N-GQDs > S-GQDs.

Fig. 2 Raman spectra of GQDs, N-GQDs, S-GQDs, P-GQDs and B-GQDs.

To further study chemical compositions and elemental binding states, XPS measurements were carried out. Fig. S3† shows the survey spectra of GQDs and doped GQDs, demonstrating a predominant graphitic C1s peak at about 284.0 eV and an O1s peak at about 531.2 eV. Apart from C1s and O1s peaks, the appearance of N1s (399.1 eV), S2p (164.3 eV), P2p (133.8 eV), and B1s (192.4 eV) in the XPS spectra of N-GQDs, S-GQDs, P-GQDs, and B-GQDs indicates the success of doping N, S, P, and B atoms into the structure of GQDs, respectively. The atomic ratio of N/C, S/C, P/C and B/C is 17.36%, 1.60%, 1.78% and 40.8%, respectively. Compared to N-GQDs and B-GQDs, the concentration of P and S is rather low, which may be due to the relatively large atomic radius of P and S¹². On the other hand, the B/C atomic ratio is much higher than those reported before with other methods.^18–21 Moreover, the O1s intensity from surface oxygen groups increases through the doping of B or P. This is because B and P prefer to bond with oxygen and the oxygen cannot be eliminated via heat-treatment.³⁶

For high-resolution C1s spectrum of GQDs (Fig. 3a), the peaks at 284.6, 285.3, 286.5 and 288.6 eV can be assigned to C=C, C–O, C=O and O–C=O bonds, respectively, which reveals that GQDs contain abundant graphitic structures and carboxyl, carbonyl, and hydroxyl groups. In addition to C–N bond (285.1 eV), the C1s spectrum of N-GQDs (Fig. 3b) confirms the presence of O-rich groups, such as C=C (284.6 eV), C–O (285.5 eV) and C=O (286.8 eV). The N1s spectrum (Fig. 3c) reveals that nitrogen atoms are mainly in the form of pyridine-like bonding configuration in N-GQDs. The peaks at 397.6, 398.7, 400.0, 402.8 and 405.9 eV correspond to pyridinic N, pyrrolic N, graphitic-like N, pyridinic-oxide, and NO₂ groups, respectively. The NO₂ groups are likely introduced at the edges during the exfoliation with HNO₃.²³ The XPS analyses clearly demonstrate that nitrogen atoms are successfully doped into the framework of GQDs.

Fig. 3 C1s spectra of GQDs (a) and N-GQDs (b); N1s spectrum of N-GQDs (c); C1s spectrum of S-GQDs (d), P-GQDs (e) and B-GQDs (f); S2p spectrum of S-GQDs (g); P2p spectrum of P-GQDs (h), and B1s spectrum of B-GQDs (i).

The C1s spectrum of S-GQDs exhibits four main peaks (Fig. 3d). The binding energy peak at 284.6 eV confirms C=C and the peak at about 285.2 eV suggests the presence of C–O and C–S. The peaks around 288.6 and 289.0 eV can be assigned to C=O and O–C=O bonds, respectively. Fig. 3g shows the S2p spectrum with two peaks at 163.8 and 165.0 eV, belonging to C–S–C structure. The peaks at 167.3, 169.0 and 170.0 eV can be assigned to –SO_n– (n = 2, 3, 4), indicating that S elements are partially doped into GQDs.

The C1s spectrum of P-GQDs can be well-fitted with characteristic peaks of 284.6, 285.1, 286.4 and 288.6 eV, corresponding to C=C, C–O/C–P, C=O, and O–C=O (Fig. 3e). The P2p peak can be de-convoluted into two sub-peaks located at 132.6 and 133.4 eV (Fig. 3h), which can be attributed to P–O and P–C bonding. This confirms the successful incorporation of P atoms into GQDs.

Fig. 3f shows the C1s spectrum with sub-peaks at 283.7, 284.6, 285.5, 286.5 and 288.3 eV, which correspond to C–B, C=C, C–O, C=O, and O–C=O bonds, respectively. Fig. 3i shows the B1s spectrum with three peaks at 189.6, 190.2 and 191.0 eV, corresponding to B–C, BC₂O, and BCO₂.²⁰ All these results indicate the formation of C–B bonds in the reaction. Moreover, the binding energy of the B1s peak in B-GQDs (192.4 eV) is higher than that of pure boron (188 eV), suggesting that boron atoms are partially bound to carbon atoms in the sp² graphene network.²⁰

The contents of C=C bond of GQDs and doped GQDs were calculated by integrating the fitting curve area of the C1s spectrum, which are 32.0%, 35.6%, 33.3%, 29.9% and 26.5% for GQDs, N-GQDs, S-GQDs, P-GQDs and B-GQDs, respectively. This data indicates that the order for defects in graphitic structure is B-GQDs > P-GQDs > GQDs > S-GQDs > N-GQDs, which does not correspond well with the results of Raman spectroscopy, being B-GQDs > GQDs > P-GQDs > N-GQDs > S-GQDs. This can be due to different analysis properties of Raman spectroscopy and XPS. Table S2† shows the binding energy of C=C (284.6 eV), C–N (285.05 eV), C–S (285.2 eV), C–P (285.05 eV) and C–B (283.7 eV) for GQDs, N-GQDs, S-GQDs, P-GQDs and B-GQDs, respectively, with the order for binding energy being C–S > C–N=C–P > C=C > C–B. It was reported that N-doping into carbon structure induces positively charged carbon atoms due to higher electronegativity of N (3.04) than that of C (2.55)^15,35. However, the doped S and P atoms show a more pronounced electron affinity, despite lower electronegativity of S (2.58) and P (2.19) compared to that of N. As shown in Fig. 3g and h, except for C–S and C–P bonds, there are –SO_n– and P–O in S-GQDs and P-GQDs, respectively. This means an oxygen bridge is formed between S and C as well as P and C, and oxygen atoms with higher electronegativity (3.44) can enhance electron poverty in carbon atoms.

For GQDs, one of the most fascinating features is their photoluminescence (PL). Until now, variously sized GQDs with different PL colors, ranging from deep ultraviolet to near infrared region, have been prepared via various synthetic approaches. However, the PL of GQDs sensitively depends on numerous parameters, such as their size, edge configuration, functional groups, defects, and heterogeneous hybridization of carbon network. The exact mechanism of PL for GQDs is still a matter of current debate and requires further clarification.^8,10 UV-Vis absorption and PL emission spectra were used to evaluate PL properties of GQDs, N-GQDs, S-GQDs, P-GQDs and B-GQDs. Fig. 4a shows the UV-Vis spectra of GQDs and doped GQDs. GQDs show a broad UV-Vis absorption below 600 nm with one shoulder at 324 nm, which is due to n–π* transition of C=O. UV-Vis spectra of both P-GQDs and B-GQDs show a broad absorption at about 350 nm. For N-GQDs and S-GQDs, UV-Vis spectra show two shoulders at 270 nm and 350 nm, assigned to C=O n–π* transition, which can result in fluorescence.⁸ Compared to GQDs, the n–π* absorption bands of N-GQDs and S-GQDs are blue-shifted. However, the n–π* absorption bands of P-GQDs and B-GQDs are red-shifted. GQDs (Fig. 4b), P-GQDs (Fig. 4e) and B-GQDs (Fig. 4f) only exhibit a slight excitation dependent emission with the maximum emission at 513.8 nm, 526.8 nm and 525.2 nm, which correspond to PL excitation peaks centered at 460, 360 and 360 nm, respectively. Compared to GQDs, P-GQDs and B-GQDs show red-shifted emission. As shown in Fig. 4c and d, both N-GQDs and S-GQDs exhibit an obvious excitation dependent emission and show a maximum emission with an excitation wavelength of 360 nm. The excitation wavelength dependence of the emission wavelength and intensity is a common phenomenon observed in carbon-based fluorescent materials,²⁴ which may arise from different particle sizes and different emission sites of N-GQDs and S-GQDs. Moreover, in the PL spectra of N-GQDs and S-GQDs, two wavelength peaking regions were detected when excited at wavelengths from 340 to 420 nm. However, when excitation wavelengths were changed from 440 to 480 nm, the PL spectra of N-GQDs and S-GQDs have one emission peak. N-GQDs shows an intense peak at 442 nm accompanied by a shoulder located at 490 nm when excited at wavelengths from 340 to 360 nm. When excitation wavelengths were changed from 380 to 420 nm, both emission wavelengths red-shifted and the emission shoulder at longer wavelength becomes more obvious. N-GQDs shows an intense peak at 492 nm accompanied by a shoulder located at 475 nm when excited at a wavelength of 420 nm. There is only one emission peak in the PL excitation spectrums of N-GQDs when excitation wavelength changed from 440 to 480 nm and the emission wavelength shows more obvious red-shift from 503 to 532 nm. The PL spectra of S-GQDs are very similar to N-GQDs which show a prominent peak at 450 nm and a shoulder peak at 490 nm with an excitation wavelength of 360 nm. Under irradiation with a 365 nm UV lamp, the colors of N-GQDs, S-GQDs, GQDs, P-GQDs and B-GQDs (Fig. 4, inset) varied from blue to green to yellow-green to bright yellow to yellow. The PL quantum yield of GQDs is 0.8%, while that of N-GQDs, S-GQDs, P-GQDs and B-GQDs are 15.8%, 1.28%, 1.52% and 1.68%, respectively. It is obviously that all elemental-doped GQDs show higher quantum yield than GQDs. Fig. S4† shows the location of PL prominent emission peak of GQDs, N-GQDs, S-GQDs, P-GQDs and B-GQDs with an excitation wavelength of 360 nm. Compared to GQDs (513.8 nm), N-GQDs (443.6 nm) and S-GQDs (447.8 nm) show blue-shifted emission, while P-GQDs (526.8 nm) and B-GQDs (525.2 nm) show red-shifted emission. The order of emission wavelength is B-GQDs > P-GQDs > GQDs > S-GQDs > N-GQDs. The phenomenon of PL emission is consistent with the result of UV-Vis absorption spectra. This can also be explained from the view point of defect effects. The disruption of graphitic carbon lattice can cause more drastic red-shift of emission peak,⁸ so the defect order of B-GQDs > P-GQDs > GQDs > S-GQDs > N-GQDs, as evident from XPS, corresponds well with the PL results. On the other hand, from the order for binding energy of C–S > C–N=C–P > C=C > C–B, we also can explain the blue- or red-shifted emission phenomenon of N-GQDs, S-GQDs and B-GQDs. The red-shifted emission of P-GQDs is probably due to more influence of defects than the binding environments.

Fig. 4 (a) UV-Vis absorption spectra of GQDs, N-GQDs, S-GQDs, P-GQDs and B-GQDs. UV-Vis and PL spectra of GQDs (b), N-GQDs (c), S-GQDs (d), P-GQDs (e) and B-GQDs (f) under different excitation wavelengths of 340–480 nm. The insets in (b–f) are optical images of GQDs, N-GQDs, S-GQDs, P-GQDs and B-GQDs solution under a UV beam of 365 nm, respectively.

We further studied the influence of nitric acid reflux time on the PL of doped GQDs. Fig. 5 shows the PL emission spectra of N-GQDs and B-GQDs obtained via refluxing in HNO₃ for 32 h. As shown in Fig. 5a, N-GQDs show a maximum emission at 555.4 nm with an excitation wavelength of 500 nm. However, the PL emission spectra of the N-GQDs via refluxing in HNO₃ for 8 h (Fig. 4c) show a maximum emission at 443.6 nm with an excitation wavelength of 360 nm. Details of the changes are listed in Table S3.† Both maximum emissions and excitation wavelengths exhibit red-shift with the increasing of refluxing time. The changes in PL spectra of B-GQDs are very similar to N-GQDs. As shown in Fig. 5b, B-GQDs obtained from refluxing in HNO₃ for 32 h show a maximum emission at 557.4 nm with an excitation wavelength of 500 nm, whereas those obtained from refluxing in HNO₃ for 8 h (Fig. 4f) show a maximum emission at 525.2 nm with an excitation wavelength of 360 nm. The red-shifted phenomenon may be caused by the increase of the oxygenated groups resulting from the HNO₃ oxidation processes with the increasing of refluxing time. This conjecture was confirmed by XPS results (Fig. S5†), where the species and quantity of the signal peaks did not change. However, the integral area of O–C=O bond increased remarkably. This means the quantity of –COOH functional group in the structure of B-GQDs increase with the increasing of refluxing time. Existing research⁸ shows that the oxidation of GQDs by –COOH functional groups red-shifts emission peaks because of band gap reduction in a coverage dependent manner. Moreover, the content of C=C bond varied from 26.5% to 21.8% when the fluxing time increased from 8 h to 32 h. This means the quantity of defects in the structure of B-GQDs increased, which can also red-shift emission peaks.

Fig. 5 PL spectra of N-GQDs (a) and B-GQDs (b) obtained via refluxing in HNO₃ for 32 h.

4. Conclusions

In conclusion, we have synthesized GQDs and N/S/P/B-doped GQDs by a simple chemical route. GQDs were prepared from thermally exfoliated graphite oxide by refluxing with concentrated nitric acid. Then, N/S/P/B-doped GQDs were synthesized using a pyrolysis method with melamine, dibenzyl disulfide, triphenylphosphine, and boric acid as N, S, P, and B sources, respectively. The elemental doping alters PL properties of GQDs. It is found that all N/S/P/B-doped GQDs had higher quantum yield than GQDs. Moreover, compared to the emission of GQDs at 513.8 nm, N-GQDs (443.6 nm) and S-GQDs (447.8 nm) show blue-shifted emission, while P-GQDs (526.8 nm) and B-GQDs (525.2 nm) show red-shifted emission under an excitation wavelength of 360 nm, which can be attributed to the change of structural defects, surface functional groups, and interactions among carbon atoms with oxygen and doping atoms. Therefore, the PL mechanism of GQDs and N/S/P/B-doped GQDs is related to both structural defects and surface functional groups, especially carboxyl groups.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (no. 51472174). L. F. Dong thanks financial support from the Malmstrom Endowment Fund at Hamline University.

Notes and references

1. X. T. Zheng, A. Ananthanarayanan, K. Q. Luo and P. Chen, Small, 2015, 11, 1620–1636.
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, R. Hu, H. N. Gao, H. T. Wei, H. Zhang, H. C. Sun and B. Yang, Chem. Commun., 2011, 47, 6858–6860.
3. D. Y. Pan, L. Guo, J. C. Zhang, C. Xi, Q. Xue, H. Huang, J. H. Li, Z. W. Zhang, W. J. Yu, Z. W. Chen, Z. Li and M. H. Wu, J. Mater. Chem., 2012, 22, 3314–3318.
4. 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.
5. S. Pandey, M. Thakur, A. Mewada, D. Anjarlekar, N. Mishra and M. Sharon, J. Mater. Chem. B, 2013, 1, 4972–4982.
6. H. T. Li, R. H. Liu, S. Y. Lian, Y. Liu, H. Huang and Z. H. Kang, Nanoscale, 2013, 5, 3289–3297.
7. V. Gupta, N. Chaudhary, R. Srivastava, G. D. Sharma, R. Bhardwaj and S. Chand, J. Am. Chem. Soc., 2011, 133, 9960–9963.
8. M. A. Sk, A. Ananthanarayanan, L. Huang, K. H. Lim and P. Chen, J. Mater. Chem. C, 2014, 2, 6954–6960.
9. 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.
10. L. L. Li, G. H. Wu, G. H. Yang, J. Peng, J. W. Zhao and J. J. Zhu, Nanoscale, 2013, 5, 4015–4039.
11. J. L. Wang and J. J. Qiu, Sci. Adv. Mater., 2014, 6, 1–11.
12. Z. S. Qian, J. J. Ma, X. Y. Shan, L. X. Shao, J. Zhou, J. R. Chen and H. Feng, RSC Adv., 2013, 3, 14571–14579.
13. Z. S. Qian, X. Y. Shan, L. J. Chai, J. J. Ma, J. R. Chen and H. Feng, ACS Appl. Mater. Interfaces, 2014, 6, 6797–6805.
14. C. F. Hu, Y. L. Liu, Y. H. Yang, J. H. Cui, Z. R. Huang, Y. L. Wang, L. F. Yang, H. B. Wang, Y. Xiao and J. H. Rong, J. Mater. Chem. B, 2013, 1, 39–42.
15. J. Ju, R. Z. Zhang, S. J. He and W. Chen, RSC Adv., 2014, 4, 52583–52589.
16. 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.
17. Z. Li, H. J. Yu, T. Bian, Y. F. Zhao, C. Zhou, L. Shang, Y. H. Liu, L. Z. Wu, C. H. Tung and T. R. Zhang, J. Mater. Chem. C, 2015, 3, 1922–1928.
18. Z. T. Fan, Y. C. Li, X. H. Li, L. Z. Fan, S. X. Zhou, D. C. Fang and S. H. Yang, Carbon, 2014, 70, 149–156.
19. S. Dey, A. Govindaraj, K. Biswas and C. N. R. Rao, Chem. Phys. Lett., 2014, 595–596, 203–208.
20. X. Hai, Q. X. Mao, W. J. Wang, X. F. Wang, X. W. Chen and J. H. Wang, J. Mater. Chem. B, 2015, 3, 9109–9114.
21. T. T. Zhang, H. M. Zhao, G. F. Fan, Y. X. Li, L. Li and X. Quan, Electrochim. Acta, 2016, 190, 1150–1158.
22. W. Kwon, J. Lim, J. Lee, T. Park and S. W. Rhee, J. Mater. Chem. C, 2013, 1, 2002–2008.
23. A. Ananthanarayanan, Y. Wang, P. Routh, M. A. Sk, A. Than, M. Lin, J. Zhang, J. Chen, H. D. Sun and P. Chen, Nanoscale, 2015, 7, 8159–8165.
24. D. Qu, M. Zheng, P. Du, Y. Zhou, L. G. Zhang, D. Li, H. Q. Tan, Z. Zhao, Z. G. Xie and Z. C. Sun, Nanoscale, 2013, 5, 12272–12277.
25. Y. Q. Dong, H. C. Pang, H. B. Yang, C. X. Guo, J. W. Shao, Y. W. Chi, C. M. Li and T. Yu, Angew. Chem., Int. Ed., 2013, 52, 7800–7804.
26. S. Kundu, R. M. Yadav, T. N. Narayanan, M. V. Shelke, R. Vajtai, P. M. Ajayan and V. K. Pillai, Nanoscale, 2015, 7, 11515–11519.
27. H. Zhang, J. Geng, R. T. Ott, M. F. Besser and M. J. Kramer, Metall. Mater. Trans. A, 2015, 46, 4078–4085.
28. J. Shen, Y. Zhu, C. Chen, X. Yang and C. Li, Chem. Commun., 2011, 47, 2580–2582.
29. J. Shen, Y. Zhu, X. Yang, J. Zong, J. Zhang and C. Li, New J. Chem., 2012, 36, 97–101.
30. J. Peng, S. Wang, P. H. Zhang, L. P. Jiang, J. J. Shi and J. J. Zhu, J. Biomed. Nanotechnol., 2013, 9, 1679–1685.
31. B. J. Zheng, Y. F. Chen, P. J. Li, Z. G. Wang, B. Q. Cao, F. Qi, J. B. Liu, Z. W. Qiu and W. L. Zhang, Nanophotonics, 2016 DOI:10.1515/nanoph-2016-0102.
32. S. H. Li, Y. C. Li, J. Cao, J. Zhu, L. Z. Fan and X. H. Li, Anal. Chem., 2014, 86, 10201–10207.
33. L. Zhang, Z. Y. Zhang, R. P. Liang, Y. H. Li and J. D. Qiu, Anal. Chem., 2014, 86, 4423–4430.
34. S. W. Yang, J. Sun, C. Zhu, P. He, Z. Peng and G. Q. Ding, Analyst, 2015, 141, 1–26.
35. C. H. Choi, S. H. Park and S. I. Woo, ACS Nano, 2012, 8, 7084–7091.
36. C. H. Choi, M. W. Chung, H. C. Kwon, S. H. Park and S. I. Woo, J. Mater. Chem. A, 2013, 1, 3694–3699.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra19673j

---