Nitrogen-doped photoluminescent carbon nanospheres: green, simple synthesis via hair and application as a sensor for Hg²⁺ ions

Abstract

In this paper, hair, which is composed of 99% keratin and 1% other elements, is chosen as a carbon source for nitrogen-doped carbon nanomaterials. By hydrothermal treatment of hair in water without any additives, such as salts, acids, or bases, carbon nanospheres with a photoluminescent quantum yield of 24.8% have been prepared. When the excitation wavelength changes from 300 to 480 nm, the photoluminescent peak shifts from 422 (violet) to 520 nm (green). Fourier transform infrared and X-ray photoelectron spectroscopy spectra analysis shows the carbon nanospheres are functionalized with hydroxyl, amino, carbonyl, and carboxylic acid groups. The carbon nanospheres have been further used as a novel probe for label-free detection of Hg²⁺ ions. The method possesses high sensitivity and selectivity. The linear range for Hg²⁺ ions is 10 to 100 nM. Meanwhile, the detection can be easily accomplished with a one-step rapid operation. This sensing system has been successfully used for the analysis of river water samples.

---

1. Introduction

Photoluminescent carbon nanomaterials have attracted tremendous attention owing to their unique properties and potential applications in bioimaging,^1–4 disease detection,⁵ catalysis,⁶ and sensors.^7–9 Compared to organic dyes and semiconductor quantum dots, carbon nanomaterials are superior in chemical inertness and possess distinct benefits, such as no optical blinking, low photobleaching, low cytotoxicity, and excellent biocompatibility.^10,11 Up to date, great efforts have been made to synthesize photoluminescent carbon nanomaterials including arc discharge,¹² laser irradiation,¹³ electrochemical synthesis,^7,14 pyrolysis,^10,15 and hydrothermal methods.¹⁶ It has been demonstrated that functionalizing fluorescent carbon nanomaterials with nitrogen groups can significantly enhance their properties and expand their novel applications.¹⁷ Recently, there has been a trend to synthesize carbon nanomaterials from biomass materials, as these are inexpensive, easy to obtain, and nontoxic.¹⁸ Carbon nanomaterials with excellent fluorescent properties from some protein-rich biomass materials, such as soybean,¹⁹ egg²⁰ and silk^21,22 have been reported. Keratin, as a kind of protein, exists widely in human and animal organs including epidermis, hair, wool, feather, hoof and horn. At present a majority of keratin are discarded, which result in resource waste and environmental pollution. So, we expect to obtain fluorescent carbon nanomaterials from keratin. Hair, which is composed of 99% keratin and 1% other elements, is chosen as a representative of keratin-rich material to carry out the research. It has been reported in 2014 that hair can be thermally decomposed at 300 °C to form carbon dots (average diameters: 2.3 nm, quantum yield: 17.3%, fluorescence lifetime: 2.83 ns).²³ Herein, we present a simple hydrothermal treatment of hair in water to prepare water-soluble, nitrogen-doped, photoluminescent carbon nanospheres (CNSs), without any additives, such as salts, acids, or bases. The CNSs possess excellent and stable fluorescent properties with a quantum yield of 24.8% and a fluorescence lifetime of 5.39 ns, which provide the further practical possibility in many fields. The lower temperature during hydrothermal synthesis results in less nitrogen loss, so the CNSs have a high nitrogen content, which might account for the high PL quantum yield. Furthermore, the CNSs have been successfully applied as a fluorescent probe for the detection of Hg²⁺ ions.

2. Experimental section

2.1. Materials

Hg(Ac)₂, AgNO₃, NiCl₂, CuCl₂, CoCl₂, MgCl₂, MnCl₂, Zn(OAc)₂, CaCl₂, FeCl₃, Pb(NO₃)₂, and CdCl₂ were purchased from Tianjin Basf Chemical Co., Ltd. (China). NaH₂PO₄ and Na₂HPO₄ were purchased from Aladdin Ltd. (Shanghai, China). All chemicals were used as received without further purification. The water used throughout all the experiments was purified through redistillation process. The hair was obtained from a local barbershop. River water samples were from the Naihe River, Taian, Shandong province, China. The river water was filtered through a 0.22 μm microporous membrane.

2.2. Synthesis of carbon nanospheres

The hair was first washed with water, and then dried at room temperature. The prepared hair (1.0 g) was mixed with H₂O (40 mL), then added into a 80 mL Teflon-lined autoclave and heated at 200 °C for 6 h. After cooling to room temperature, the resulting black suspension was filtered through a 0.22 μm microporous membrane and a brown filter solution was collected. By freeze-dried method, brown powder was obtained. The yield was calculated to be ca. 14%. The powder was dispersed in distilled water at a concentration of 1 mg mL⁻¹ for further characterization and use.

2.3. Characterization

The morphology and microstructures of CNSs were analyzed by transmission electron microscopy (Tecnai G20, USA) and high resolution transmission electron microscopy (Tecnai G2 F20, USA) operated at 200 kV. Zeta potential was measured after suitable dilution of the CNSs solution at 25.0 ± 0.5 °C, using a Zetasizer analyzer (ZS 90, Malvern). Fourier transform infrared (FTIR) spectra were obtained on a Thermo Nicolet-380 IR spectrophotometer (USA) with the KBr pellet technique ranging from 400 to 4000 cm⁻¹. X-ray photoelectron spectroscopy (XPS) was investigated using Al Kα excitation source (1486.6 eV) in a Thermo ESCALAB 250XI apparatus (USA). Fluorescence spectroscopy was performed with a Cary Eclipse spectrophotometer at different excitation wavelength ranging from 300 to 480 nm (USA, VARIAN). UV-vis absorption spectra were obtained using a UV-2450 Shimadzu Vis-spectrometer (Japan). X-ray powder diffraction (XRD) data were collected using a Bruker D8 Advance. Fluorescence lifetime was measured by an F900 time-resolved spectroscope (Edinburgh) with excitation and emission wavelengths of 340 and 443 nm respectively.

2.4. Quantum yield calculations

The quantum yield (Φ) of the carbon nanospheres was calculated using quinine sulfate as reference. For calculation of quantum yield, five concentrations of each compound were made, all of which had absorbance less than 0.1 at 340 nm. Quinine sulfate (literature²⁴ Φ = 0.54) was dissolved in 0.1 M H₂SO₄ (refractive index (η) of 1.33) while the carbon sample was dissolved in water (η = 1.33). Their fluorescence spectra were recorded at same excitation of 340 nm. Then by comparing the integrated photoluminescence intensities (excited at 340 nm) and the absorbency values (at 340 nm) of the carbon sample with quinine sulfate, the quantum yield of the carbon sample was determined. The quantum yield was calculated using the eqn (1), where Φ is the quantum yield, m is slope, η is the refractive index of the solvent, ST is the standard and X is the sample.

Φ_x = Φ_ST(m_x/m_ST)(η_x/η_ST) (1)

2.5. Assay procedure

16 μL of CNSs solution was mixed with 1 mL of 0.2 M PBS 8 buffer. 5 μL of different concentration of Hg²⁺ ions was added, and equilibrated for 6 min at room temperature before the spectral measurements. The PL spectra were recorded under excitation at 340 nm.

3. Results and discussion

3.1. Synthesis and microscopic studies of CNSs

The synthesis procedure has been illustrated in Scheme 1 Hydrothermal treatment of hair at 200 °C leads to a yellow dispersion of the CNSs. Keratin, which is insoluble at room temperature, can degrade into small amino acids under hydrothermal condition, which may be further carbonized to form carbon nanomaterials. The dispersion exhibits a blue colour under a UV lamp (365 nm), illustrating that the CNSs exhibit a strong blue fluorescence.

Scheme 1 Illustration of formation of CNSs from hair.

The transmission electron microscopy (TEM) images (Fig. 1a and b) show that the CNSs are well dispersed in a spherical shape with diameter ranging from 10 to 100 nm. The corresponding size distribution histogram is plotted in Fig. 1c. The majority particle size is distributed between 20 and 80 nm and the average diameter is calculated to be 49 nm. The reaction time can influence the size of CNSs. Fig. S1, ESI† shows the TEM images of CNSs under different reaction time. When the time is 2 h, the image shows the hair is partial decomposed, and the nanoparticles have not formed. When the time is 24 h, the average diameter of CNSs is calculated to be 94 nm. The high resolution TEM (HRTEM) image (inset of Fig. 1a) shows the lattice spacing of 0.23 nm which corresponds to the [100] facet of graphitic carbon. Fig. 1d represents typical XRD profiles of CNSs. A broad peak round 21.4°(0.43 nm), which is close to the [002] facet of graphite. The corresponding interlayer spacing in graphite (0.34 nm) becomes larger in CNSs, which can be attributed to the introduction of more oxygen containing groups.

Fig. 1 TEM micrographs at different magnifications (a) 400 nm (with HRTEM image, inset) and (b) 40 nm, (c) the corresponding particle size distribution histogram (the error bars represent the standard deviation of three measurements), and (d) XRD patterns of CNSs.

3.2. FTIR and XPS spectra of CNSs

The FTIR studies are carried about to obtain chemical and structural information about the CNSs (Fig. 2). The characteristic absorption band of O–H around 3418 cm⁻¹, and the stretching vibration band of C=O at 1702 cm⁻¹ indicate the presence of carboxylic acid and other oxygen-containing functional groups.^15,19 Furthermore, the broad absorption around 3418 cm⁻¹ and one sharp peak at 1629 cm⁻¹ are assigned to the stretching vibration and bending vibration bands of N–H, suggesting the existence of amino-containing functional groups. Moreover, three obvious absorption peaks at 2959, 1401 and 1304 cm⁻¹ are associated with the stretching vibration C–H, C=C and C–C, suggesting the presence of alkyl and aryl groups.²⁵

Fig. 2 FTIR spectrum of CNSs.

X-ray photoelectron spectroscopy is used to investigate the surface states of the CNSs. The survey spectrum (Fig. 3a) of the CNSs shows four typical peaks of C1s, N1s, O1s and S2p. The content of N is up to 11.88%, which is much higher than the previous reported.²³ The spectrum of C1s (Fig. 3b) can be deconvoluted into several single peaks that correspond to C–C (284.5 eV), C–N (285.1 eV), C–O (285.8 eV), and C=N/C=O (287.9 eV) functional groups,²⁵ which are consistent with the FTIR results. The N1s spectrum (Fig. 3c) reveals three relative nitrogen species of C–N–C (399.7 eV), N–C₃ (400.6 eV), and N–H (401.3 eV).^25–27 The spectrum of O1s (Fig. 3d) further confirms these observations with two characteristic oxygen states of C=O (531.4 eV) and C–O (532.2 eV).²⁵ Presence of C=C, C–O, C–N, N–H, COOH bonds clearly indicates that the CNSs are functionalized with hydroxyl, amino, carbonyl, and carboxylic acid groups. The presence of plentiful hydrophilic groups on the surface, endows the CNSs with high water solubility and a Zeta potential of −20.5 mV.

Fig. 3 XPS spectra of CNSs. (a) Survey spectrum. (b) C1s spectrum. (c) N1s spectrum. (d) O1s spectrum.

3.3. UV-vis absorption and photoluminescence spectra

The absorption spectrum (Fig. 4a) shows a clear adsorption at ca. 276 nm, which is a typical characteristic of fluorescent carbon dots.¹⁹ Fig. 4a shows photoluminescence (PL) emission spectra of the CNSs, when the excitation wavelength changes from 300 to 480 nm, the PL peak correspondingly shifts from 422 (violet) to 520 nm (green). The exact mechanisms of PL responsible for carbon nanomaterials have not yet been confirmed. Previous reports confirmed that excitons of carbon, emissive traps, the quantum confinement effect, aromatic structures, oxygen-containing groups, edge defects, and free zigzag sites can contribute to the fluorescence.^26–31 In carbon-based fluorescent materials, the excitation wavelength usually depends on emission wavelength.^{28,29,32–35} Pang et al. claimed that the wavelength shifts are mainly caused by the particle surfaces rather than by a size effect based on their electrochemical tuning experiments.³⁶ The CNSs exhibit different emission colours including blue, green and red, when they are excited by different excitation wavelengths, as investigated under a fluorescent microscope (Fig. 4c–e). The quantum yield (QY) is calculated to be as high as 24.8% at 340 nm excitation, which is higher than the previous reported (17.3%). High nitrogen content and plentiful groups on the surface of CNSs maybe account for the high quantum yield of the CNSs. Table S1 ESI† shows the relationship between quantum yield of CNSs and the reaction temperature. The quantum yield of the CNSs prepared at 200 °C is the highest. When the reaction temperature is over 200 °C, the quantum yield of the CNSs decreases with the increase of temperature.

Fig. 4 (a) UV-vis absorption spectrum, (b) PL emission spectra at different excitation wavelengths and PL images of CNSs excited at (c) UV, (b) blue, and (c) green light. All scale bars are 100 μm.

The PL stability of CNSs to the effects of the pH and ionic strength of solutions, and to UV exposure is investigated. Generally speaking, the fluorescence intensity of the nanoparticles decreases significantly upon changing from an acidic to a basic solution (Fig. 5a). The mechanism of this pH-dependent PL behaviour is not very clear. Maybe the functional groups take place significant change with the change of pH. The same phenomenon has been found in carbon dots passivated using amine-terminated compounds.³⁷ However, most of previous reported carbon nanoparticles show weak fluorescence in a strong acidic solution, which are attributed to the protonation and deprotonation of the functional groups on the surface of the nanoparticles.⁴ There are no obvious changes in PL intensity or peak characteristics at different ionic strengths (Fig. 5b), which is significant for CNSs to be used in biological system. The measurement of PL intensity with time indicates that photo bleaching effect of UV on the CNSs can be negligible (Fig. 5c).

Fig. 5 (a) and (b) Effect of pH, ionic strengths (ionic strengths were controlled by various concentrations of NaCl), (c) and time on the PL intensity of CNSs (0.025 mg mL⁻¹).

3.4. CNSs probe for Hg²⁺ detection

Initial experiments demonstrate that Hg²⁺ ions can obvious quench the PL of CNSs. To perform the experiment of Hg²⁺ ions determination, the pH effect on the quenched PL efficiency in the 0.2 M PBS buffer solution is studied by monitoring the dependence of PL intensity on the pH of the solution (Fig. 6a). As shown in Fig. 6a, the pH of the solution has some effect on the PL quenched efficiency. In buffer of pH 8, the quenched efficiency is the highest. We have also studied the effect of response time on the PL quenched efficiency. The results reveal that the PL quenching reaches a stable value within 6 min. In order to obtain a lower detection limit, the buffer of pH 8 is chosen in the following experiment and six minutes is chosen as the response time.

Fig. 6 Plots of the quenched efficiency under (a) different pH and (b) different response time by 100 nM Hg²⁺ ions (the error bars represent the standard deviation of three measurements).

Fig. 7a shows a gradual decrease in PL intensity with increasing Hg²⁺ concentration at room temperature, revealing that the sensing system is sensitive to the Hg²⁺ concentration. The Hg²⁺ ion dependence plot (Fig. 7c) of I/I₀ (I₀ and I are the PL intensities of CNSs excited at 340 nm in the absence and presence of Hg²⁺ ion, respectively) shows good linearity with concentrations of Hg²⁺ ions in the range of 10–100 nM. The detection limit is estimated to be 3.7 nM at a signal-to-noise ratio of 3, which is lower than the maximum level (10 nM, 2 ppb) for mercury in drinking water permitted by the United States Environmental Protection Agency. Table 1 shows the comparison of the different fluorescent probes for Hg²⁺ detection, suggesting the sensitivity of our sensing system is comparable or superior to that of previously reported sensing systems.

Fig. 7 (a) PL emission spectra of CNSs in the presence of different concentrations of Hg²⁺ ions (0, 10, 20, 40, 60, 80, 100, 500, 1 × 10³, 5 × 10³, 1 × 10⁴, 5 × 10⁴,1 × 10⁵ nM). (b) The relationship between I/I₀ and concentrations of Hg²⁺ ions from 0 to 100 μM. (c) The linear region between I/I₀ and concentrations of Hg²⁺ ions (the error bars represent the standard deviation of three measurements).

Table 1 Comparison of different fluorescent probes for Hg²⁺ detection

Fluorescent probes	Line arrange (nM)	Detection limit (nM)	Ref.
CdS-encapsulated DNA	10–110	4.3	38
CdTe quantum dots	2–14	1.55	39
Surface-modified CdTe quantum dots	1.2 × 10^3–1.5 × 10^6	4	40
LysVI-AuNCs	0.01–5	0.003	41
Fluorescent Ag clusters	10–5 × 10^3	10	42
Au@Ag core–shell nanoparticles	10–450	9	43
DNA-functionalized gold nanoparticles	50–2.5 × 10^3	25	44
Mononucleotides-stabilized gold nanoparticles	20–6.0 × 10^3	50	45
CDs	0–3 × 10^3	4.2	46
CPs	0.5–10	0.23	47
CNSs	10–100	3.7	This work

---

We have also investigated the PL intensity changes of CNSs in the presence of various metal ions under the same conditions, including Ag⁺, Ni²⁺, Cu²⁺, Co²⁺, Mg²⁺, Mn²⁺, Zn²⁺, Ca²⁺, Fe³⁺, Pb²⁺ and Cd²⁺ (Fig. 8a). No tremendous decrease is observed upon addition of these ions into the CNSs dispersion, thus indicating their negligible influence on our fluorescent CNSs as a probe towards Hg²⁺ ions. Furthermore, the selectivity of this nanoprobe in the presence of all possible interference ions was evaluated considering the cross reactivity. As demonstrated in Fig. 8b, the present method can still detect Hg²⁺ in the presence of all possible interference ions. The sensor retains 97% of its initial response after it has kept in refrigerator at 4 °C for one month, which indicate that the sensor has good stability. The reproducibility has investigated by five parallel experimental. The relative standard deviation is 2.3%, which suggest that the sensor displays good reproducibility.

Fig. 8 (a) Quenched efficiency of metal ions and (b) quenched efficiency of Hg²⁺ ion under different condition, M stands for the competing ions. The error bars represent the standard deviation of three measurements. The concentration of all competing ion solutions is 100 nM.

In order to evaluate the feasibility of the CNSs for a practical application, a standard addition method is applied to test water samples from a river. As presented in Table 2, the recoveries of water samples range from 102% to 103%, and the relative standard deviations (RSDs) range from 1.9% to 2.5%. These results imply that the method can be used as Hg²⁺ detection in water samples.

Table 2 Determination of Hg²⁺ ions in river water samples

Samples	Added Hg^2+ (nM)	Measured (nM)	Recovery (%)	RSD (n = 3%)
1	10	10.2	102	2.4
2	40	41.2	103	2.5
3	80	82.7	103	1.9

---

To get further insight into the PL quenching mechanism, time-correlated single photon counting (TCSPC) is used to study the excitation behaviour of CNSs in the presence and absence of Hg²⁺ (Fig. 9). The CNSs in water exhibit double exponential decay, using eqn (2).

Y(t) = α₁ exp(−t/τ₁) + α₂ exp(−t/τ₂) (2)

Fig. 9 The TCSPC plots of CNSs and CNSs/Hg²⁺.

In the eqn (2), α₁ and α₂ are the fractional contributions of time-resolved decay lifetimes of τ₁ and τ₂. The average lifetime could be concluded from eqn (3).

The decay time of CNSs is 5.39 ns and has two components: 7.1 ns (68.6%) and 1.6 ns (31.4%). After mixing with Hg²⁺ ions, the CNSs/Hg²⁺ decay time decrease to 3.9 ns. Moreover, the fast decay component increases sharply: 6.6 ns (20%) and 0.4 ns (80%). More than one lifetime may be either attributed to complex energy level, or complex mechanism of fluorescence carbon-based materials. The obviously reduced lifetime indicates an ultra fast CNSs/Hg²⁺ electron-transfer process and leads to dynamic quenching. So, the PL quenching phenomenon can be attributed to the effective coordination/chelation interactions between Hg²⁺ ions and the plentiful hydroxyl, amino, and carbonxylate groups of the CNSs.^25,48

4. Conclusions

In conclusion, we have demonstrated a facile, green and large scale synthesis of carbon nanoparticles from a cheap and readily available natural precursor. These carbon nanoparticles show strong and stable PL, which is dependent on excitation wavelength and pH. The CNSs have been further used as a novel sensing probe for label-free detection of Hg²⁺ ions. The method possesses high sensitivity and good selectivity. Meanwhile, the detection can be easily accomplished with one-step rapid (within 10 min) operation. This sensing system has been successfully used for the analysis of river water samples.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (no. 21375079, 21105056) and Project of Development of Science and Technology of Shandong Province, China (no. 2013GZX20109).

Notes and references

1. L. Cao, X. Wang, M. J. Meziani, F. Lu, H. Wang, P. G. Luo, Y. Lin, B. A. Harruff, L. M. Veca and D. Murray, J. Am. Chem. Soc., 2007, 129, 11318–11319.
2. Y. Fang, S. Guo, D. Li, C. Zhu, W. Ren, S. Dong and E. Wang, ACS Nano, 2010, 6, 400–409.
3. C. Fowley, B. McCaughan, A. Devlin, I. Yildiz, F. M. Raymo and J. F. Callan, Chem. Commun., 2012, 48, 9361–9363.
4. S. Sahu, B. Behera, T. K. Maiti and S. Mohapatra, Chem. Commun., 2012, 48, 8835–8837.
5. L. Bao, Z. Zhang, Z. Tian, L. Zhang, C. Liu, Y. Lin, B. Qi and D. Pang, Adv. Mater., 2011, 23, 5801–5806.
6. S. Zhuo, M. Shao and S.-T. Lee, ACS Nano, 2012, 6, 1059–1064.
7. L. Zhou, Y. Lin, Z. Huang, J. Ren and X. Qu, Chem. Commun., 2012, 48, 1147–1149.
8. W. Lu, X. Qin, S. Liu, G. Chang, Y. Zhang, Y. Luo, A. M. Asiri, A. O. Al-Youbi and X. Sun, Anal. Chem., 2012, 84, 5351–5357.
9. Y. Guo, Z. Wang, H. Shao and X. Jiang, Carbon, 2013, 52, 583–589.
10. R. Liu, D. Wu, X. Feng and K. Müllen, J. Am. Chem. Soc., 2011, 133, 15221–15223.
11. S. N. Baker and G. A. Baker, Angew. Chem., Int. Ed., 2010, 49, 6726–6744.
12. X. Xu, R. Ray, Y. Gu, H. J. Ploehn, L. Gearheart, K. Raker and W. A. Scrivens, J. Am. Chem. Soc., 2004, 126, 12736–12737.
13. S.-L. Hu, K.-Y. Niu, J. Sun, J. Yang, N.-Q. Zhao and X.-W. Du, J. Mater. Chem., 2009, 19, 484–488.
14. H. Li, X. He, Z. Kang, H. Huang, Y. Liu, J. Liu, S. Lian, C. H. A. Tsang, X. Yang and S. Lee, Angew. Chem., Int. Ed., 2010, 49, 4430–4434.
15. H. Liu, T. Ye and C. Mao, Angew. Chem., Int. Ed., 2007, 46, 6473–6475.
16. S. Qu, X. Wang, Q. Lu, X. Liu and L. Wang, Angew. Chem., 2012, 124, 12381–12384.
17. H. Peng and J. Travas-Sejdic, Chem. Mater., 2009, 21, 5563–5565.
18. M.-M. Titirici and M. Antonietti, Chem. Soc. Rev., 2010, 39, 103–116.
19. C. Zhu, J. Zhai and S. Dong, Chem. Commun., 2012, 48, 9367–9369.
20. J. Wang, C. Wang and S. Chen, Angew. Chem., 2012, 124, 9431–9435.
21. W. Li, Z. Zhang, B. Kong, S. Feng, J. Wang, L. Wang, J. Yang, F. Zhang, P. Wu and D. Zhao, Angew. Chem., Int. Ed., 2013, 52, 8151–8155.
22. Z. L. Wu, P. Zhang, M. X. Gao, C. F. Liu, W. Wang, F. Leng and C. Z. Huang, J. Mater. Chem. B, 2013, 1, 2868–2873.
23. S.-S. Liu, C.-F. Wang, C.-X. Li, J. Wang, L.-H. Mao and S. Chen, J. Mater. Chem. C, 2014, 2, 6477–6483.
24. J. Q. Umberger and V. K. LaMer, J. Am. Chem. Soc., 1945, 67, 1099–1109.
25. S. Liu, J. Tian, L. Wang, Y. Zhang, X. Qin, Y. Luo, A. M. Asiri, A. O. Al-Youbi and X. Sun, Adv. Mater., 2012, 24, 2037–2041.
26. J. Zhou, C. Booker, R. Li, X. Zhou, T.-K. Sham, X. Sun and Z. Ding, J. Am. Chem. Soc., 2007, 129, 744–745.
27. H. Li, X. He, Z. Kang, H. Huang, Y. Liu, J. Liu, S. Lian, C. H. A. Tsang, X. Yang and S. T. Lee, Angew. Chem., Int. Ed., 2010, 49, 4430–4434.
28. K. P. Loh, Q. Bao, G. Eda and M. Chhowalla, Nat. Chem., 2010, 2, 1015–1024.
29. L. Bao, Z. L. Zhang, Z. Q. Tian, L. Zhang, C. Liu, Y. Lin, B. Qi and D. W. Pang, Adv. Mater., 2011, 23, 5801–5806.
30. V. N. Mochalin and Y. Gogotsi, J. Am. Chem. Soc., 2009, 131, 4594–4595.
31. H. X. Zhao, L. Q. Liu, Z. De Liu, Y. Wang, X. J. Zhao and C. Z. Huang, Chem. Commun., 2011, 47, 2604–2606.
32. R. Liu, D. Wu, S. Liu, K. Koynov, W. Knoll and Q. Li, Angew. Chem., 2009, 121, 4668–4671.
33. X. Wang, L. Cao, S.-T. Yang, F. Lu, M. J. Meziani, L. Tian, K. W. Sun, M. A. Bloodgood and Y.-P. Sun, Angew. Chem., Int. Ed., 2010, 49, 5310–5314.
34. K. Welsher, Z. Liu, S. P. Sherlock, J. T. Robinson, Z. Chen, D. Daranciang and H. Dai, Nat. Nanotechnol., 2009, 4, 773–780.
35. J. Jeong, M. Cho, Y. T. Lim, N. W. Song and B. H. Chung, Angew. Chem., Int. Ed., 2009, 48, 5296–5299.
36. L. Bao, Z.-L. Zhang, Z.-Q. Tian, L. Zhang, C. Liu, Y. Lin, B. Qi and D.-W. Pang, Adv. Mater., 2011, 23, 5801–5806.
37. Z.-A. Qiao, Y. Wang, Y. Gao, H. Li, T. Dai, Y. Liu and Q. Huo, Chem. Commun., 2010, 46, 8812–8814.
38. Y. Fan, Y. F. Long and Y. F. Li, Anal. Chim. Acta, 2009, 653, 207–211.
39. T. Li, Y. Zhou, J. Sun, D. Tang, S. Guo and X. Ding, Microchim. Acta, 2011, 175, 113–119.
40. X. Yunsheng and Z. Changqing, Talanta, 2008, 75, 215–221.
41. Y.-H. Lin and W.-L. Tseng, Anal. Chem., 2010, 82, 9194–9200.
42. C. Guo and J. Irudayaraj, Anal. Chem., 2011, 83, 2883–2889.
43. F. Au, Langmuir, 2011, 27, 13198–13205.
44. C.-W. Liu, C.-C. Huang and H.-T. Chang, Langmuir, 2008, 24, 8346–8350.
45. Y. Xu, L. Deng, H. Wang, X. Ouyang, J. Zheng, J. Li and R. Yang, Chem. Commun., 2011, 47, 6039–6041.
46. L. Zhou, Y. Lin, Z. Huang, J. Ren and X. Qu, Chem. Commun., 2012, 48, 1147–1149.
47. W. Lu, X. Qin, S. Liu, G. Chang, Y. Zhang, Y. Luo, A. M. Asiri, A. O. Al-Youbi and X. Sun, Anal. Chem., 2012, 84, 5351–5357.
48. A. Zhu, Q. Qu, X. Shao, B. Kong and Y. Tian, Angew. Chem., 2012, 124, 7297–7301.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra04209c

---