Using hydrogen peroxide to mediate through a one-step hydrothermal method for the fast and green synthesis of N-CDs

Abstract

A facile and green approach for preparation of photoluminescent nitrogen-doped carbon dots (N-CDs) is reported, using cheap and extensively available cocoon silk as raw material. The use of H₂O₂ leads to the enhanced formation of N-CDs. The N-CDs have been synthesized in a short time (50 minutes), and have a high fluorescence quantum yield (24.0%). A small amount of H₂O₂ solution plays a vital role in shortening the reaction time, which is a brand new way to synthesize N-CDs with the help of H₂O₂, with both reactants and products being environmentally friendly. The prepared N-CDs are collected by removing the insoluble materials through simple filtration rather than dialysis. As far as we know, this is the first report of N-CDs being synthesized simply with the help of H₂O₂ solution. The prepared N-CDs solution is sensitive to pH changes, which means they have promise for application in the pH detection area.

---

1 Introduction

Fluorescent carbon-based materials have attracted more and more research attention in recent years due to their exciting properties such as high optical absorptivity, biocompatibility, chemical stability, and low toxicity.¹ These materials primarily include carbon dots,² nanodiamonds,³ carbon nanotubes,⁴ fullerene⁵ and fluorescent graphene.⁶ They are a new type of carbon materials within the nano-size range and are promising candidates for numerous applications due to their superior properties, which successfully distinguish them from traditional fluorescent materials.²⁷ They have been broadly used in the areas of biomarkers, photocatalysis, photovoltaic devices, biological sensing agents, surface-enhanced Raman scattering and so on.^7–9 Among all of these materials, carbon quantum dots (CDs) have drawn the most extensive attention.

We can say that CDs are the bridge between carbon materials and luminescent materials.¹⁰ In the late 20th century, CDs were firstly developed and their advantageous properties attracted more and more attention.¹⁰ CDs are a kind of environmentally friendly fluorescent nanomaterial. They not only have properties including excellent optical performance, small size and so on, which are similar to the properties of traditional semiconductor materials, but also have the properties of low cytotoxicity, great biocompatibility, functionalization, low cost, and low demand of reaction conditions.^11–14

At present, there are many problems which need to be resolved about the preparation of CDs, such as fluorescence quantum yield being far behind that of semiconductor quantum dots. Generally speaking, CDs prepared without any decoration are not fluorescent or have a low fluorescence quantum yield; this aspect seriously influences the promotion and application of CDs. The preparation methods are diverse, which can be summarized into top-down and bottom-up. But efficient one-step strategies for the fabrication of CDs on a large scale are still a challenge in this field. In order to improve the quantum yield, some other particular groups can be included, such as amino and sulfhydryl. After the treatment of decoration in the surface, CDs can be better used in other fields.¹⁵ But there is still a need for in-depth and systematic research about both the graphite process and the formation mechanism.¹⁶ In order to enrich the potential applications, the doping of other elements into CDs is an efficient method of tuning electronic structures. It can be postulated that the introduction of nitrogen (N) into carbon dots (N-doped carbon dots, N-CDs) will further enhance their versatile properties.³²

Recently it has become a growing trend that cocoon silk be used as the carbon source to synthesize CDs, on account of it being easy to obtain, nontoxic, and so on.^17,18 The silkworm cocoon silk fiber is composed of two cores of fibroin surrounded by a cementing layer of sericin in a structure known as a bave (each individual fibroin core is known as a brin).¹⁹ Fibroin and sericin account for about 75 and 25 wt%, respectively. A typical biomass material, more than 480 000 tons per year of cocoon silk produced by the Bombyx mori silkworm is cultivated all over the world.²⁰ Cocoon silk consists mainly of two fibroin fibers coated by sericin, a water-soluble protein.^21,22

In this experiment, N-CDs which contain amino groups (such as –NH₂, –NH–) mainly coming from a peptide which is a kind of polymer made of GAGAGS (glycine–alanine–glycine–alanine–glycine–serine) existing in natural cocoon silk can be successfully prepared by using the cocoon silk as the carbon source.²⁹ In order to make this conclusion more convincing and even more universal, a parallel experiment has been carried out, when the raw material is replaced by peptide, and sample 3 and sample 4 refer to the obtained products after the reaction of peptide with H₂O₂ and deionized water, respectively. The data of this parallel experiment are shown in the ESI including Fig. S1 and S6–S11.† When the raw material is cocoon silk and peptide, respectively, the obtained experimental phenomena and experimental data are highly similar. The prepared carbon dot solution can exhibit bright blue fluorescence under the excitation of UV light (at 365 nm). This preparation method possesses many advantages, mainly including easily obtained and cheap raw materials, and simple procedures of the experiment, and with no need to conduct surface passivation. We characterized the prepared N-CDs solution in detail. Also, when 30 mL of H₂O₂ (mass concentration: 3%) was replaced by 30 mL of deionized water, we amazedly found that the quantum yields decreased seriously. So in summary, the H₂O₂ solution plays an indispensable role in the improvement of fluorescence quantum yield.

2 Experimental method

Preparation of cocoon silk, peptide and 3% H₂O₂ solution

The best reaction conditions which need to be optimized include the reaction temperature, reaction time, amount of the reactants and so on. In the experiment, 3% H₂O₂ was obtained by diluting 30% H₂O₂ which was purchased from Aladdin. Cocoon silk was purchased online, and then it should be correctly dealt with, which is the key to the success of the experiment. The specific preparation of the cocoon silk was as described elsewhere.²³ Briefly, cocoons of Bombyx mori were boiled for 45 min in an aqueous solution of Na₂CO₃ (5 wt%) to remove silk sericin and wax, and then cooled to room temperature. The silks were obtained after being washed several times using water and ethanol, and then dried at room temperature. After those steps, the preparation of the cocoon silks was accomplished. Peptide was purchased from Sangon Biotech (Shanghai) Co. Ltd.

Synthesis of nitrogen-doped carbon quantum dots

The prepared cocoon silk (0.5 g) was mixed with H₂O₂ (30 mL, mass concentration: 3%), then added into a 50 mL Teflon-lined autoclave and heated at 260 °C for 50 min. The N-CDs were collected by removing the insoluble materials through simple filtration with a filter membrane (D_m = 0.22 μm). And the corresponding solid products were obtained through freeze-drying.

Instrumentation

IR spectra were obtained with a Nicolet Avatar 360 FT-IR spectrophotometer. Transmission electron microscopy (TEM) images were recorded with a FEI Tecnai 12 transmission electron microscope. Fluorescence spectra were measured with a fluorescence spectrophotometer (Hitachi model F-7000) equipped with a 150 W xenon lamp as the excitation source. The fluorescence lifetime of the samples was recorded using a photo-counting PicoHarp 300 system (PicoPico-Quant, Berlin, Germany). High-resolution TEM observations were performed with an electron microscope (Tecnai G2 F20S-TWIN, 200 kV). UV-visible (UV-Vis) absorption spectra were obtained with a UV/Vis/NIR spectrophotometer (Lambda 750). X-ray photoelectron spectroscopy (XPS) was conducted by employing an X-ray photoelectron spectrometer (AXIS ULTRA DLD, Kratos).

3 Results and discussion

The quantum yield of the N-CDs was calculated by the following equation:where Q is the quantum yield, I is the measured integrated emission intensity, n is the refractive index, and OD is the optical density, which is measured with a UV-Vis spectrophotometer. The subscript R refers to the reference fluorophore of known quantum yield; for example, quinine sulfate was used in the present work. The quinine sulfate (literature Φ = 0.54) was dissolved in 0.1 M H₂SO₄ (n = 1.33) and the N-CDs were dissolved in distilled water (n = 1.33).

The quantum yield (Fig. 1) results show that when the reaction temperature and reaction time are respectively 200 °C and 50 minutes, the highest quantum yields can reach 4.19%. When the reaction temperature and reaction time are respectively 220 °C and 70 minutes, the highest quantum yields can be 9.20%. When the temperature is 240 °C, the quantum yields increase constantly with increasing reaction time. Furthermore, as the temperature reaches 260 °C, the quantum yields firstly increase and then decrease evidently, reaching a maximum of about 24.0% when reaction time is 50 minutes. Therefore, in consideration of low power consumption and quick preparation, the best reaction temperature and reaction time are 260 °C and 50 minutes, respectively. And 24.0% is the highest quantum yield in the whole experiment. In order to illustrate the function of H₂O₂, we did a parallel experiment.

Fig. 1 The specific quantum yields when cocoon silk (0.5 g) was reacted with 30 mL of H₂O₂ (mass concentration: 3%) at different reaction times and reaction temperatures.

In order to shorten the reaction time, we added some H₂O₂ into the system (Fig. 2). Photos (a), (c) and (e) show the results when cocoon silk (0.5 g) reacts with H₂O₂ (30 mL, mass concentration: 3%); photos (b), (d) and (f) show the result of the reaction of cocoon silk (0.5 g) with 30 mL of deionized water; photos (f) and (e) show the as-prepared N-CDs solution under UV excitation at 365 nm. When H₂O₂ is not added, a lot of the unreacted cocoon silk remained on the filter membrane. But if we add some H₂O₂ solution, few reactants remained on the filter membrane, and the intensity of fluorescence was much stronger, as seen in Fig. 2(e) and (f). So we found that H₂O₂ plays a vital role during the process of the reaction, not only improving the efficiency of the reaction, but also obviously leading to a large quantum yield increase. Furthermore, the N-CDs solution which is prepared with adding H₂O₂ is freely and easily dispersible in water and exhibits good photostability.

Fig. 2 Comparison of results of adding H₂O₂ and not adding H₂O₂: (a) and (b) show residues on the filter membrane; (c) and (d) represent the reaction products under daylight; (e) and (f) show their blue fluorescence under UV light (λ_ex = 365 nm).

Comparison was made of UV-Vis spectra and excitation/emission spectra of sample 1 and sample 2, which refer to liquid samples, prepared without adding H₂O₂ and adding H₂O₂, respectively. To explore further the effect of surface groups on the optical properties of these two samples, their photoluminescence (PL) and UV-Vis absorption spectra were studied. The UV-Vis absorption spectra are shown in Fig. 3(a) and (c). For sample 1 (not adding H₂O₂), a typical absorption peak at ca. 252 nm was observed, which is assigned to the π → π* transition of C=C; another peak is located at 275 nm which is attributed to n → π* transition from C=O.³¹ For sample 2 (adding H₂O₂), the absorption spectrum shows a clear adsorption feature at ca. 268 nm, which is ascribed to n → π* transition along with a typical characteristic of fluorescent CDs.^29,30 However, besides the strong n → π* absorption peak, a new absorption band at ca. 320 nm was also observed. These absorption features thus arose from the reaction of hydrogen peroxide with the carbon source (cocoon silk), which broke the π → π carbon bonds to form carboxyl or hydroxyl groups.²⁸ So the formation of the absorption band at ca. 320 nm can greatly increase the carbon quantum yields, and this peak is in accordance with the excitation spectrum as shown in Fig. 3(d). On the other hand, the PL spectra showed typical excitation-dependent behavior: when the excitation wavelength is changed from 310 nm to 430 nm, the PL peak shifts to longer wavelengths as seen in Fig. 3(b) and (d). The PL excitation spectrum recorded with the strongest luminescence shows two peaks at 320 and 370 nm in Fig. 3(d). The 320 nm PL excitation peak corresponds to the 320 nm absorption band. The 252 nm and 268 nm PL peaks should have a corresponding absorption band that can hide in the strong background absorption from the π → π* transition. The PL excitation spectrum clearly demonstrates that the observed luminescence from the samples is directly correlated with the new transition at 320 nm rather than the commonly observed π → π* transition. Fourier transform infrared (FTIR) spectroscopy is utilized to investigate the functional groups and chemical bonding state (Fig. S3†). After the addition of H₂O₂, the PL lifetime increased from 1.24 ns to 3.99 ns (Fig. S2†). A quantitative comparison is shown in Table S2.†

Fig. 3 UV-Vis absorption spectra of liquid sample 1 (a) and sample 2 (c); (b) and (d) are the PL spectra excited by different excitation wavelengths and their best excitation spectra (blue line) at 445 nm emission and 438 nm emission of sample 1 and sample 2, respectively. The excitation wavelength was varied from 310 nm to 430 nm. Colors correspond to different excitation wavelengths: cyan, magenta, yellow, dark yellow, navy, purple, wine, and olive correspond to 310, 330, 350, 360, 370, 390, 410, and 430 nm.

All these changes show that the hydrothermal treatment under the effect of H₂O₂ exerts a strong influence on the formation, microstructure, and optical properties of the CDs. So using H₂O₂ can effectively optimize the preparation of N-CDs.

The strategy for the formation mechanism of N-CDs (Fig. 4) can be summarized as consisting of two parts including top-down and bottom-up. The top-down mechanism is as follows. Under high temperature and high pressure, H₂O₂ is almost decomposed into ˙OH. The prepared cocoon silks are cut into short microrods under the effect of ˙OH,²⁴ and after a longer time, the short microrods are degraded into many fragments; subsequently, these fragments are degraded and hydrolyzed into much smaller molecules. The bottom-up mechanism can be clearly presented: the small molecules by way of dehydration and polymerization can successfully combine with each other, forming some nanospheres. Subsequently, the small nanospheres with the action of intermolecular dehydration can further polymerize together, finally producing the N-CDs. From the results of experiments, we think that H₂O₂ plays an important role in the synthesis of N-CDs. H₂O₂ enhances the electro-conduction ability of fluorescent carbon nanoparticles by breaking C–H and N–H bonds and doping N (N–H) into CDs to form pyrrole-like N at the same time. Secondly, when H₂O₂ is at high temperature and high pressure, it can be decomposed into ˙OH rapidly, ˙OH effectively combining with C=C, accelerating the reaction. Of course, the detailed mechanism needs further discussion, and the specific formation mechanism is still a great challenge for this field.²⁵ The structure of precursor polymer can be found in Fig. S1.†

Fig. 4 Top-down and bottom-up strategy for the formation mechanism of N-CDs.

In Fig. 5(a), a TEM image shows that the N-CDs are regular in size with an average diameter of about 7.4 nm. Their size distribution is shown in Fig. S4.† The image in Fig. 5(b) clearly reveals that diffraction contrast of the N-CDs is very low and nearly without any obvious lattice fringes, which strongly indicates their polymer-like amorphous structure; however, without adding H₂O₂, the diameter is nearly 167 nm as shown in Fig. 5(c).

Fig. 5 (a) TEM image. (b) High-resolution TEM image of the N-CDs formed on adding H₂O₂. (c) TEM image of the product without adding H₂O₂.

XPS was used to investigate the surface states of the N-CDs. The survey spectrum (Fig. 6) of the N-CDs shows three typical peaks of C_1s, N_1s, and O_1s, and the corresponding content of each element is displayed in the inset in Fig. 6(a). The spectrum of C_1s in the N-CDs in Fig. 6(b) can be deconvoluted into two peaks that correspond to C–C (284.5 eV) and C=N/C=O (288.0 eV) functional groups, which are consistent with the FTIR results. The N_1s spectrum in Fig. 6(c) reveals two relative nitrogen species of C–N–C (399.5 eV) and N–H (401.5 eV), which have been observed in the case of C_1s and FTIR spectra. Moreover, the spectrum of O_1s in Fig. 6(d) further shows these observations with two characteristic oxygen states of C=O (531.6 eV) and C–O–C (533.0 eV). The elemental analysis results further reveal that the CDs are composed of C (56.34%), H (5.63%), N (12.82%), and O (calculated, 25.21%) atoms,²⁶ which is consistent with the XPS results. Compared with cocoon silk, the increase in carbon content and decrease in oxygen content indicate adequate carbonization during the high-temperature treatment. Therefore, we can conclude that the as-prepared N-CDs are mainly composed of polycyclic aromatic polymers of amino acids, as well as possessing abundant hydroxyl, amino, and carbonyl/carboxylate groups on their surface. In order to make sure that N-CDs can also be obtained from peptide, we did parallel experiments, involving peptide reacting with each of H₂O₂ and deionized water. Subsequent characterization confirmed that the results are much the same as those obtained when the raw material is cocoon silk. Related data can be found in Fig. S7–S11† including yield, TEM, XPS and so on. So we suggest that N-CDs can be obtained from any natural sources containing peptides. With the addition of H₂O₂, the content of oxygen and nitrogen elements increased a lot; at the same time, the content of carbon element decreased, and the position of the peaks did not change significantly comparing to Fig. 7.

Fig. 6 XPS spectra of the N-CDs when cocoon silk reacts with H₂O₂. (a) Survey spectrum. (b) C_1s spectrum. (c) N_1s spectrum. (d) O_1s spectrum.

Fig. 7 XPS spectra of nanoparticles when cocoon silk reacts with H₂O. (a) Survey spectrum. (b) C_1s spectrum. (c) N_1s spectrum. (d) O_1s spectrum.

As shown in Fig. 8, an increase in the fluorescence intensity of the N-CDs solution can be observed with an increase in the solution pH under 370 nm UV, and there is a good linear relationship between the fluorescence intensity of the N-CDs and solution pH within the range 4–9 with a correlation coefficient of 0.97052 (as follows):

Intensity = 5.95086 [pH] + 51.6727 (2)

Fig. 8 pH-dependent PL spectra. Fluorescence spectra of the N-CDs for different pH values of Tris-HCl buffer (10 mM) solution with excitation wavelength at 370 nm. Inset: a linear relationship between PL intensity of the N-CDs solution and pH value.

Based on the above results, it can be deduced that the protonation state of the N-CDs plays an important role in the fluorescence intensity.¹⁷ For an increase in solution pH from 4 to 9, the deprotonation degree of the N-CDs gradually increases, producing higher concentrations of deprotonated nitrogen and oxygen atoms on the surface of the N-CDs, which change the electron distributions. In addition, the variation in fluorescence intensity of the N-CDs may be attributed to the protonation and deprotonation of amide and carboxylic functional groups on the N-CDs' surface. The above results show that the N-CDs have potential as fluorescent sensors for pH measurement. And N-CDs solution under UV light (λ_ex = 365 nm) shows great stability (Fig. S5†). This point means a wider range of applications.

4 Conclusions

In summary, highly fluorescent N-CDs have been prepared by a one-step hydrothermal reaction through 50 minutes using cheap and extensively available cocoon silk as raw material. The prepared N-CDs have a polymer-like amorphous structure (7.4 nm average diameter) with abundant amino groups on their surfaces, and show strong fluorescence activity with a high quantum yield (24.0%). And when the raw material of cocoon silk was replaced by peptide (GAGAGS), the obtained product is much the same, suggesting that N-CDs can be obtained from any natural sources containing peptides. The N-CDs solution is sensitive to pH changes giving a very sensitive signal response, showing their potential for use in pH measurement. At the same time, at room temperature, the N-CDs have considerable stability.

Acknowledgements

This research was supported by the National Science Foundation of China (51372091, 21371061, 21571067 and 21371062), the Science and Technology Plan Projects of Guangdong Province (2012B010200030), the National Science Foundation of Guangdong Province (S2013030012842) and Mei Le Shi Printing Ink Co. Ltd of Qing Yuan city of China.

References

1. L. Cao, M. J. Meziani, S. Sahu and Y. P. Sun, Acc. Chem. Res., 2013, 46, 171–180.
2. J. Wang, C. F. Wang and S. Chen, Angew. Chem., Int. Ed., 2012, 37, 9297–9301.
3. A. Krueger, Adv. Mater., 2008, 20, 2445–2449.
4. K. Welsher, Z. Liu, S. P. Sherlock, J. T. Robinson, Z. Chen, D. Daranciang and H. Dai, Nat. Nanotechnol., 2009, 4, 773–780.
5. Y. T. Ke, B. Garg and Y. C. Ling, RSC Adv., 2014, 102, 58329–58336.
6. S. Zhu, J. Zhang, S. Tang, C. Qiao, L. Wang, H. Wang, X. Liu, B. Li, Y. Li, W. Yu, X. Wang, H. Sun and B. Yang, Adv. Funct. Mater., 2012, 22, 4732–4740.
7. S. Zhu, S. Tang, J. Zhang and B. Yang, Chem. Commun., 2012, 48, 4527–4539.
8. S. Zhuo, M. Shao and S. T. Lee, ACS Nano, 2012, 6, 1059–1064.
9. X. Yan, X. Cui and L. Li, J. Am. Chem. Soc., 2010, 132, 5944–5945.
10. Q. Q. Dou, X. T. Fang, S. Jiang, P. L. Chee, T. C. Lee and X. J. Loh, RSC Adv., 2015, 5, 46817–46822.
11. S. J. Zhu, Y. B. Song, X. H. Zhao, J. R. Shao, J. H. Zhang and B. Yang, Nano Res., 2015, 8, 355–381.
12. L. Cao, X. Wang, M. J. Meziani, F. S. Lu, H. F. Wang, P. J. G. Luo, Y. Lin, B. A. Harruff, L. M. Veca, D. Murray, S. Y. Xie and Y. P. Sun, J. Am. Chem. Soc., 2007, 129, 11318–11319.
13. X. J. Mao, H. Z. Zheng, Y. J. Long, J. Du, J. Y. Hao, L. L. Wang and D. B. Zhou, Spectrochim. Acta, Part A, 2015, 151, 1–10.
14. S. J. Zhu, Q. N. Meng, L. Wang, J. H. Zhang, Y. B. Song, H. Jin, K. Zhang, H. C. Sun, H. Y. Wang and B. Yang, Angew. Chem., 2013, 125, 4045–4049.
15. Z. Yang, Z. H. Li, M. H. Xu, Y. J. Ma, J. Zhang, Y. J. Su, F. Gao, H. Wei and L. Y. Zhang, Nano-Micro Lett., 2013, 4, 247–259.
16. F. Yang, S. T. Yang, S. K. Sonkar, L. J. Yang, J. J. Broglie, Y. Liu and Y. P. Sun, RSC Adv., 2014, 4, 10791–10807.
17. L. X. Zhao, F. Di, D. B. Wang, L. H. Guo, Y. Yang, B. Wan and H. Zhang, Nanoscale, 2013, 7, 2055–2058.
18. B. Hu, K. Wang, L. H. Wu, S. H. Yu, M. Antonietti and M. M. Titirici, Adv. Mater., 2010, 22, 813–828.
19. P. Jiang, H. F. Liu, C. H. Wang, L. Z. Wu, J. G. Huang and C. Guo, Mater. Lett., 2006, 60, 919–925.
20. Y. S. Yun, S. Y. Cho, J. Y. Shim, B. H. Kim, S. J. Chang, S. J. Baek, Y. S. Huh, Y. Tak, Y. W. Park, S. J. Park and H. J. Jin, Adv. Mater., 2013, 14, 1993–1998.
21. H. J. Jin, J. Park, V. Karageorgiou, U. J. Kim, R. Valluzzi, P. Cebe and D. L. Kaplan, Adv. Funct. Mater., 2005, 8, 1241–1247.
22. R. Nazarov, H. J. Jin and D. L. Kaplan, Biomacromolecules, 2004, 5, 718–726.
23. W. Li, Z. H. Zhang, B. Kong, S. S. Feng, J. X. Wang, L. Z. Wang, J. P. Yang, F. Zhang, P. Y. Wu and D. Y. Zhao, Angew. Chem., Int. Ed., 2013, 52, 8151–8155.
24. D. M. Phillips, L. F. Drummy, D. G. Conrady, D. M. Fox, R. R. Naik, M. O. Stone, P. C. Trulove, H. C. de Long and R. A. Mantz, J. Am. Chem. Soc., 2004, 126, 14350–14351.
25. S. J. Zhu, J. H. Zhang, S. J. Tang, C. Y. Qiao, L. Wang, H. Y. Wang, X. Liu, B. Li, Y. F. Li, W. L. Yu, X. F. Wang, H. C. Sun and B. Yang, Adv. Funct. Mater., 2012, 22, 4732–4740.
26. W. Li, Z. H. Zhang, B. Kong, S. S. Feng, J. X. Wang, L. Z. Wang, J. P. Yang, F. Zhang, P. Y. Wu and D. Y. Zhao, Angew. Chem., Int. Ed., 2013, 52, 8151–8155.
27. Z. Gao, L. B. Wang, R. X. Su, R. L. Huang, W. Qi and Z. M. He, Biosens. Bioelectron., 2015, 15, 232–238.
28. S. Wang, P. He, J. M. Zhang, H. Jiang and S. Z. Zhu, Synth. Commun., 2005, 35, 1803–1808.
29. X. M. Li, S. L. Zhang, S. A. Kulinich, Y. L. Liu and H. B. Zeng, Sci. Rep., 2014, 4, 4976.
30. C. Z. Zhu, J. F. Zhai and S. J. Dong, Chem. Commun., 2012, 48, 9367–9369.
31. M. X. Gao, C. F. Liu, Z. L. Wu, Q. L. Zeng, X. X. Yang, W. B. Wu, Y. F. Li and C. Z. Huang, Chem. Commun., 2013, 49, 8015–8017.
32. H. Li, W. Q. Kong, J. Liu, N. Y. Liu, H. Huang, Y. Liu and Z. H. Kang, Carbon, 2015, 91, 66–75.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra12976a

---