Synthesis of gold nanoclusters/glucose oxidase/graphene oxide multifunctional catalyst with surprisingly enhanced activity and stability and its application for glucose detection

Abstract

The paper reports the synthesis of gold nanoclusters/glucose oxidase/graphene oxide (GNC/GOD/GO) multifunctional catalyst and its application for glucose detection. First, an amine group was introduced into the GO sheet through the 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide/N-hydroxysuccinimide mediated reaction between carboxylic acid and ethylenediamine to form a positively charged GO sheet (GO-NH₃⁺). Then, negatively charged GNC and GOD were assembled on the surface of GO-NH₃⁺ with electrostatic interaction. Finally, the hybrid was dried by freeze drying. The resulting hybrid gives high fluorescence intensity and optical stability. More importantly, the hybrid exhibits largely enhanced activities towards the decomposition of hydrogen peroxide and oxidation of glucose at neutral pH. Owing to the greatly synergistic catalytic effect between GNC and GOD, the nanosensor based on the hybrid displays an ultrasensitive fluorescence response to glucose. The fluorescence peak intensity linearly decreases with the increase of glucose concentration in the range of 1.1 × 10⁻² to 1.6 × 10⁻⁷ M with a detection limit of 6.8 × 10⁻⁸ M (S/N = 3). The method provides the advantage of sensitivity, repeatability and stability compared with present glucose sensors. It has been successfully applied in detection of glucose in real serum samples. The study also opens a new avenue for the fabrication of GO-based catalytic materials that holds great promise in potential applications such as biocatalyst, nanosensors and molecular carriers.

---

1 Introduction

Metal nanoclusters consist of several to tens of atoms with the diameter of below 2 nm and give the properties regulated by their subnanometer dimensions.^1–4 Gold nanoclusters (GNC) have increasingly received great attention because of their fascinating features, including ease of preparation and conjugation, biocompatibility, good water solubility, excellent stability and large stokes shifts.^5,6 The synthesis of GNC is typically based on the polymer template,^7–10 or relies on the monolayer protection in presence of the molecule with thiol ligands.^11,12 In order to meet the needs of different applications, some enzymes, amino acids and proteins have been developed for the fabrication of various fluorescent GNC in recent years. For example, Ma et al. reported an alkanethiol-stabilized GNC based on simply placing histidine, 11-mercaptoundcanoic acid and chloroauric acid together at room temperature.¹³ The resulting GNC exhibits intense fluorescence, long fluorescence lifetime, considerable stability and large Stoke's shift. Moreover, it also indicates negligible effect in altering cell proliferation or triggering apoptosis. Yang et al. reported a lysine-stabilized GNC.¹⁴ Here, Cu²⁺ coordination with –COOH and –NH₂ of both lysine-stabilized GNC and BSA-stabilized GNC leads to obvious fluorescence quenching. It has been successfully applied to the determination of Cu²⁺ ions with high sensitivity and selectivity. Qi et al. reported a folic acid-functionalized fluorescent GNC for the cancer cell imaging.¹⁵ The GNC shows high fluorescence intensity, good photostability, and near-infrared fluorescence spectrum. Recently, some proteins have also been proven to be excellent scaffolds and capping/reducing agents for the formation of GNC. The fluorescent properties of GNC make them potential labels for biologically motivated studies.^16,17 However, present GNCs can only offer single function with relatively low activity when these were employed as the catalyst for chemical and biological processes. This limits their applications that require high activity and multistep reaction. There is a growing need for the development of multifunctional GNC catalyst with high activity and stability for biocatalysis and biosensor.

Graphene has been a hotbed of research since the experimental isolation and characterization of graphene in 2004.¹⁸ Great effort has been made to develop graphene hybrids and explore their applications in sensor,^19–21 imaging,²² chemiluminescence,²³ and catalysts.²⁴ Due to good biocompatibility, graphene hybrids were also used for the controlled anticancer.^25,26 Graphene/gold hybrid integrates unique properties of two classes of materials and exhibits some novel functions induced by the synergistic effects between graphene and gold nanoparticles.²⁷ However, graphene is not suitable as a carrier for fluorescent GNC, because it is an excellent electron acceptor, which will result in obvious fluorescence quenching. Different from graphene, graphene oxide (GO) possesses good water solubility and provides fertile opportunities for the construction of special fluorescent GNCs. Recently, Ren et al. reported the synthesis of GNC/GO and its application for cancer cell detection.²⁸ The investigation reveals that GO plays an important role in modulating catalytic activity of the supported GNC.

The study focuses on the synthesis of gold nanoclusters/glucose oxidase/graphene oxide (GNC/GOD/GO) and its application for glucose detection. The as-prepared hybrid exhibits strong fluorescence and largely enhanced catalytic activities towards the decomposition of hydrogen peroxide and oxidation of glucose. It was used as the cascade of enzymes for the detection of glucose in serum samples by fluorescence method. The analytical method provides a better sensitivity, repeatability and stability compared with present glucose sensors.

2 Experimental

2.1 Materials

Glucose, glucose oxidase, hydrogen tetrachloroaurate (HAuCl₄), 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC), hydrogen peroxide (H₂O₂), N-hydroxysuccinimide (NHS) and bovine serum albumin (BSA) were purchased from Sigma-Aldrich (Mainland, China). Natural flake graphite was obtained from Qingdao Henlide Graphite Company (Qingdao, China) with an average particle size of 20 μm. Phosphate-buffered saline (PBS, pH 7.4, Na₂HPO₄–NaH₂PO₄, 0.05 M) was prepared in the laboratory. GO was synthesized from natural graphite flakes according to modified Hummers method.²⁹ The GNC/GOD/GO solution was prepared by dissolving 20 mg of solid GNC/GOD/GO product in a 20 ml of ultrapure water. Other reagents employed were of analytical regent grade and were purchased from Shanghai Chemical Company (Shanghai, China). Ultrapure water (18.2 MΩ cm) purified from Milli-Q purification system was used throughout the experiment.

2.2 Apparatus

Transmission electron microscope (TEM) analysis was conducted on a JEOL 2010 FEG microscope at 200 keV. The sample was prepared by dispensing a small amount of dry powder in the PBS. Then, one drop of the suspension was dropped on 300 mesh copper TEM grids covered with thin amorphous carbon films. EDX mappings were obtained with an FEI Tecnai G2 F30 S-TWIN field emission transmission electron microscope equipped with an EDAX energy-dispersive X-ray spectroscopy operating at 300 kV. Fluorescence spectra were recorded by a Cary Eclipse fluorescence spectrophotometer (Agilent, Japan) with a excitation wavelength of 365 nm. Circular dichroism analysis was performed with a MOS-450 circular dichroism spectrometer using a 0.01 cm quartz cell at room temperature. The zeta potential measurements were carried out in a ZETASIZER2000 Zeta Potential Analyzer.

2.3 Synthesis of GNC

BSA-stabilized GNC was prepared by the reported method.³⁰ In a typical experiment, all glassware used in the experiments were cleaned in a bath of freshly prepared aquaregia (HCl : HNO₃ volume ratio = 3 : 1), and rinsed thoroughly in water. A 50 ml of HAuCl₄ solution (20 mM) was added to an equal volume of the BSA solution (50 mg ml⁻¹) under the magnetic stirring. Then, 3.0 ml of the NaOH solution (1.0 M) was introduced and the mixture was allowed to incubate at 37 °C for 24 h under vigorous stirring. The color of the solution changed from light yellow to deep brown. The solution was then dialyzed in double distilled water for 48 h to remove unreacted HAuCl₄ and NaOH. The obtained GNC solution was stored at 4 °C in refrigerator before use, which the final concentration of gold is 2.5 mM.

2.4 Synthesis of GNC/GOD/GO

To obtain positively charged GO sheets (GO-NH₃⁺), 20 mg of graphite oxide was dispersed in 20 ml of the PBS by ultrasonication for 1 h to form homogeneous GO dispersion. Added 1 ml of the mixed EDC/NHS solution (50 mM) to the dispersion. After 30 min incubation, 0.2 ml of ethylenediamine was added. After another 2 h incubation, the solution was kept overnight at 4 °C and then was dialyzed by the PBS to remove unreacted ethylenediamine. The final concentration of GO is 0.5 mg ml⁻¹. Next, the GO solution (4.0 ml) was mixed with the GNC solution (50 ml) and GOD (20 ml, 2 mg ml⁻¹). After 2 h incubation under vigorous shaking at room temperature, the mixed solution was dried by freeze drying to obtain GNC/GOD@GO product. The product was stored at 4 °C for further use.

2.5 Glucose detection

The glucose solution of known concentration or real serum sample was mixed with 1 ml of the GNC/GOD/GO solution. After 20 min incubation, the mixed solution was subjected to fluorescence measurements on the fluorescence spectrophotometer with an excitation wavelength of 365 nm.

3 Results and discussion

3.1 Synthesis and characterization

The synthesis of GNC/GOD/GO mainly includes three assemble processes, i.e. the preparation of GO-NH₃⁺, formation of GNC/GOD/GO and freeze drying of the hybrid (shown in Fig. 1). First, graphite oxide was dispersed in the PBS by ultrasonication to form homogeneous GO dispersion. Then, amine group was introduced on the surface of GO sheet through the EDC/NHS mediated reaction between carboxylic acid and ethylenediamine to produce GO-NH₃⁺. Zeta potential analysis reveals that zeta potential of the as-prepared GO-NH₃⁺ is +65.2 ± 0.98 mV, indicating strong positive charge. Next, GO-NH₃⁺ was mixed with GNC and GOD. Because the zeta potentials of GNC and GOD are −25.2 ± 0.52 mV and −32.91 ± 0.36 mV respectively, verifying both GNC and GOD have strong negative charge, the mixture of GO-NH₃⁺, GNC and GOD will lead to produce GNC/GOD/GO hybrid due to their electrostatic attraction. Further, we observed a considerable decrease in zeta potential of GO-NH₃⁺ after the addition of GNC and GOD. This confirms again the formation of GNC/GOD/GO through electrostatic adsorption. Final, the hybrid was dried by freeze drying to improve its long-term stability.

Fig. 1 Procedure for the synthesis of GNC/GOD/GO.

The as-prepared GNC/GOD/GO was characterized by TEM, EDX and XPS. Fig. 2 shows that the hybrid contains many wrinkled GO sheets, and the GNCs were well dispersed on the surface of GO sheets (Fig. 2a). The BSA-modified GNC has a narrow particle size distribution (shown in Fig. S1†) with an average diameter 3–4 nm (Fig. 2b). Gold cores in the GNCs are about 1–2 nm with a narrow size distribution, and no aggregated GNCs could be observed (Fig. 2c). XPS technology is a powerful tool for studying on the surface chemical properties of materials. The XPS spectrum of GNC/GOD/GO contains five peaks at 83.5, 87.3, 284.4, 399.6 and 531.9 eV. The peaks at 284.4, 399.6 and 531.9 eV could be assigned to C_1s, N_1s, and O_1s, which come from BSA, GO and GOD. The peaks at 83.5 and 87.3 eV could be assigned to Au_4f, verifying the success of GNCs modification on the surface of GO sheets. Moreover, the hybrid was also characterized by IR spectrum and the result was shown in Fig. S2.†

Fig. 2 TEM (a), enlarged TEM (b), EDX images (c) and XPS patterns (d) of the GNC/GOD/GO.

3.2 Optical and catalytic properties

The hybrid can act as fluorescence probe and be applied in bioanalysis and bioimaging due to its good biocompatibility. However, the application requires strong and stable fluorescence signal. Herein, we investigated the effects of GO and GOD on the fluorescence intensity of GNC. Fig. 3 shows that UV absorption spectrum of the GNC gives a maximum absorption at 256 nm, and corresponding fluorescence emission peak lies at 645 nm. After the addition of GO into the system, the fluorescence intensity can remain almost unchanged. The fact demonstrates that the introduction of GO does not reduce the fluorescence intensity of GNC. This is because GO is different from graphene, its good insulation and rich oxygen-containing groups is beneficial to improve optical property of the GNC. Further, effect of the GO concentration on optical stability of the GNC was also examined and the results were shown in Fig. 4. It can be seen that in the absence of GO the fluorescence intensity will irregularly change with the increase of standing time, indicating a relatively poor optical stability. However, the optical stability will be obviously improved after the addition of GO. With the increase of GO concentration the optical stability will further be improved. When the GO concentration is more than 1.0 μg ml⁻¹, an ideal optical stability was obtained. The improvement should be attributed that GNC interacts with GO sheets to form stable GNC/GO due to electrostatic attraction. However, the fluorescence intensity will rapidly decrease with the increase of GO concentration owing to reduce of the GNC ratio in the hybrid. To obtain strong and stable fluorescence signal, a 20 μg ml⁻¹ of GO was employed for synthesis of the GNC/GOD/GO.

Fig. 3 UV absorption spectrum (a) of the GNC solution and fluorescence spectra of the GNC before (b) and after (c) added GO.

Fig. 4 Fluorescence changes of the GNC/GOD/GO prepared in different concentration of GO with standing time.

The zeta potential measurement was used to study on the effect of GOD on fluorescence intensity of the GNC. Since GNC and GOD are negatively charged, GNC does not directly combine with the GOD to create a stable hybrid due to electrostatic repulsion. However, this is very beneficial to keep original natures of GNC and GOD in the hybrid. The study demonstrates that the introduction of GOD leads to no change in fluorescence intensity and stability of the GNC. To investigate on the long-term optical stability, the GNC/GOD/GO was stored at 4 °C for the period and then measured its fluorescence intensity. The result shows that the fluorescence intensity can remain 98.6% at least after the period of six months, indicating an excellent long-term stability.

The hybrid can act as a mimic enzyme to catalyze the decomposition of hydrogen peroxide (H₂O₂).²⁸ Kinetic study confirms that GNC/GOD/GO exhibits excellent catalytic activity at neutral pH. Because single GO and single GOD show almost no activity for H₂O₂ under the same conditions, observed peroxidase-like activity of GNC/GOD/GO is attributed to GNC in the hybrid. The fluorescence response plateau was also observed when H₂O₂ concentration is more than 1 mM, showing the characteristics of Michaelis–Menten kinetic mechanism. The apparent Michaelis–Menten constants (K_m) can be calculated according to the Lineweaver–Burk eqn (1):

Here, ΔF_ss is change value of the fluorescence intensity after the addition of H₂O₂, C is the buck concentration of H₂O₂ and ΔF_max is the maximum fluorescence intensity change value measured under the saturated H₂O₂. The apparent K_m was calculated for H₂O₂ according to the slope of curves in Fig. 5. The K_m is 3.16 mM. The value is obviously lower than that of single GNC (5.3 mM), implying that the GNC/GOD/GO has a better affinity for H₂O₂ than single GNC. The significant enhancement in the activity demonstrates that GO played an important role in modulating the catalytic activity of the supported GNC.

Fig. 5 The relationship between the reciprocal of fluorescence change and the reciprocal of H₂O₂ concentration using GNC/GOD/GO (a) or single GNC (b) as a catalyst.

To understand the effect of GO on catalytic activity of the GNC, circular dichroism technique was used for studying on the structure change of BSA on the surface of GNC before and after the addition of GO. Then, the ratio of each secondary structure in the BSA was calculated from the circular dichroism spectrum. Fig. 6 shows the ratios of various secondary structures in the GNC and GNC/GO. It can be seen that the introduction of GO results in the increase of α-helix and reduce of β-sheet, β-turn and random. The change in secondary structure will remarkably increase the exposure degree of gold atoms in the GNC, the interactions between GO sheet and GNC makes gold atom can fully contact with H₂O₂, which can improve the catalytic activity.

Fig. 6 The secondary structures of BSA on the surface of GNC and GNC/GO.

The hybrid can act as glucose oxidase to catalyze the oxidation of glucose. The study reveals that GNC/GOD/GO offers an enhanced activity at neutral pH towards glucose oxidation. In the absence of GOD, both GNC and GO show almost no activity for the glucose oxidization under the same conditions. Thus, the observed glucose oxidase-like activity of GNC/GOD/GO is creditable to GOD in the hybrid. Further, a fluorescence response plateau was observed when the glucose concentration is more than 0.8 mM, showing the characteristics of Michaelis–Menten kinetic mechanism. The K_m is 0.51 mM, indicating high catalytic activity toward the glucose oxidation. The value is obviously lower than that of the GOD/GNC (3.7 mM) and single GOD (5.6 mM). The results demonstrate that the introduction of GO and GNC can largely enhance the catalytic activity of GOD in the hybrid towards the glucose oxidization.

To understand the effects of GO and GNC, the secondary structures of GOD before and after the additions of GO and GNC were measured. From Table 1, we can draw two conclusions. The one conclusion is that the introduction of GO will result in an obvious decrease of the α-helix, and increase of the β-sheet compared with single GOD. The change may increase exposure degree of the active sites in the GOD, which leads to enhance the catalytic activities. The another conclusion is that the introduction of GNC does not influence on the secondary structures of GOD. However, we noted that GOD, GNC/GOD and GNC/GOD/GO have very similar secondary structures. This is because the effects of GO on the secondary structures of GNC and GOD shows an opposite change trend. Because of this, no substantial change was observed among GOD, GNC/GOD and GNC/GOD/GO.

Table 1 The secondary structures of various GOD hybrids

GOD hybrid	α-Helix	β-Sheet	β-Turn	Random
GOD	0.346	0.111	0.232	0.324
GNC/GOD	0.347	0.114	0.237	0.322
GOD/GO	0.211	0.242	0.191	0.367
GNC/GOD/GO	0.346	0.115	0.236	0.324

---

The catalytic mechanism of the hybrid towards glucose oxidation was suggested in Fig. 7. First, GOD in the hybrid catalyzes to oxidate β-D glucose into δ-lacttonic glucose with O₂, hydrolyze into acid radical glucose and H₂O₂. Next, GNC in the hybrid catalyzes to decompose H₂O₂ into O₂. At the same time, gold–sulfur bond was damaged by H₂O₂ oxidization to make free gold atoms, which will result in the decomposition of GNC and sensitively decrease of the fluorescence intensity.³¹ More importantly, the rapid removal of H₂O₂ from the reaction system will largely accelerate the oxidation of glucose. In addition, the introduction of GO also improves the catalytic activities of GNC and GOD. Based on the above consideration, we think that the excellent catalytic activity of GOD/GNC/GO towards glucose should be attributed to prominent synergistic effect between GOD, GNC and GO.

Fig. 7 Suggested catalytic mechanism for the glucose oxidation.

3.3 Analytical characteristics

The nanosensor based on the GNC/GOD/GO was used for glucose detection. Fig. 8 exhibits fluorescence response of the nanosensor in different concentration of glucose and the calibration plot of logarithmic glucose concentration vs. corresponding maximum fluorescence intensity. The fluorescence intensity will rapidly decrease with the increase of glucose concentration. When the glucose concentration is in the range of 1.1 × 10⁻² to 1.6 × 10⁻⁷ M, the fluorescence intensity linearly decreases with the increase of glucose concentration. The linear equation was F = −163.76C − 97.03, with a statistically significant correlation coefficient of 0.9965, which F is the fluorescence intensity at 645 nm and C is logarithmic glucose concentration. The detection limit was found to be 6.8 × 10⁻⁸ M that was obtained from the signal-to-noise characteristics of these data (S/N = 3). The nanosensor was repeatedly measured ten times in a 2.0 × 10⁻⁴ M of glucose standard solution under the same conditions. A relative standard deviation of 1.6% for the measurements was obtained, indicating good precision. The GNC/GOD/GO product was stored in at 4 °C and its sensitivity for glucose was checked every week. The fluorescence intensity remains the initial response of 99.2% at least after the period of twenty-eight weeks, indicating excellent long-term stability. The above analytical parameters are better than that of glucose sensors reported in literatures for such as the sensor based on mesoporous Fe₂O₃–graphene³² and Fe-doped graphitic carbon nitride.³³

Fig. 8 (A) Fluorescence response of GNC/GOD/GO in 1.6 × 10⁻⁷, 5 × 10⁻⁷, 1 × 10⁻⁶, 5 × 10⁻⁶, 1 × 10⁻⁵, 5 × 10⁻⁵, 1 × 10⁻⁴, 5 × 10⁻⁴, 1 × 10⁻³, 5 × 10⁻³, 1 × 10⁻², 5 × 10⁻² and 1 × 10⁻¹ M of glucose (from a to m). (B) Calibration plots of concentration of logarithmic glucose vs. fluorescence intensity at 645 nm.

To study effect of GO on the sensitivity for glucose detection, the nanosensor based on the GNC/GOD was fabricated and used for glucose detection. The fluorescence response of the nanosensor in different concentration of glucose was measured under the same conditions. The linear equation was obtained from the calibration plot of GNC/GOD in Fig. 8B. The result was F = −127.43C + 144.29. From the equation, we observed that slope of the curve is obviously smaller than that of the nanosensor based on the GNC/GOD/GO, indicating a lower fluorescence response to glucose. This confirms again that the introduction of GO can enhance the catalytic activity of GNC/GOD towards glucose oxidation, which will largely improve the sensitivity.

3.4 Application to detection of glucose in serum samples

The feasibility of the newly developed method for possible applications was investigated by analyzing real serum samples. The concentration of glucose in serum sample was determined from the calibration curve and the value can be used to calculate the concentration in the original sample. The mean ± SD of each sample were calculated, and the values are reported in Table 2. This table also shows that there is a very good agreement between the results obtained by the proposed method and those obtained by application of a routine enzymatic method (using hexokinase method) in a local hospital. These indicated proposed method has a good accuracy and precision.

Table 2 Determination of glucose in serum samples (n = 5)^a

Samples	Glucose found by proposed method (mM)	Glucose found by hexokinase method (mM)
a Results expressed as: , where X is the mean of n observations of x, s is the standard deviation, t is distribution value chosen for the desired confidence level, the t- and F-values refer to comparison of the proposed method with the hexokinase method. Theoretical values at 95% confidence limits: F = 6.39, t = 2.78.
Serum 1	4.52 ± 0.13	4.25 ± 0.15
	F = 1.33, t = 0.31
Serum 2	5.51 ± 0.14	5.62 ± 0.18
	F = 1.65, t = 1.55
Serum 3	7.82 ± 0.11	7.80 ± 0.12
	F = 1.19, t = 0.36
Serum 4	6.09 ± 0.17	5.97 ± 0.15
	F = 1.28, t = 1.58
Serum 5	4.78 ± 0.10	4.66 ± 0.12
	F = 1.44, t = 2.37

---

4 Conclusions

GNC/GOD/GO has been successfully synthesized and applied for glucose detection. The hybrid is a multifunctional catalyst and can simultaneously catalyze the decomposition of hydrogen peroxide and oxidation of glucose. Since the use of GO increases exposure degree of the active site in the hybrid, the hybrid largely enhances the catalytic activities when compared with single GNC and single GOD. Further, the hybrid was developed as a nanosensor for the detection of glucose by fluorescence method. The proposed method has higher long-term stability, sensitivity, repeatability and stability when compared with present glucose sensors. It can be widely used for the glucose detection in routine analysis. The study also provides an approach for design and synthesis of GO-based multifunctional catalyst with high catalytic activity, which can meet the needs of different application that requires high activity and multistep reaction.

Acknowledgements

The authors acknowledge the financial support from the National Natural Science Foundation of China (no. 21176101), the Fundamental Research Funds for the Central Universities (no. JUSRP51314B), MOE & SAFEA for the 111 Project (B13025) and the country “12th Five-Year Plan” to support science and technology project (no. 2012BAK08B01).

Notes and references

1. L. A. Peyser, A. E. Vinson, A. P. Bartko and R. M. Dickson, Science, 2001, 291, 103.
2. A. A. Herzing, C. J. Kiely, A. F. Carley, P. Landon and G. J. Hutchings, Science, 2008, 321, 1331.
3. Z. Y. Li, N. P. Young, M. Di Vece, S. Palomba, R. E. Palmer, A. L. Bleloch, B. C. Curley, R. L. Johnston, J. Jiang and J. Yuan, Nature, 2008, 451, 46.
4. M. M. Alvarez, J. T. Khoury, T. G. Schaaff, M. N. Shafigullin, I. Vezmar and R. L. Whetten, J. Phys. Chem. B, 1997, 101, 3706.
5. I. Diez and R. H. A. Ras, Nanoscale, 2011, 3, 1963.
6. H. C. Yeh, J. Sharma, J. J. Han, J. S. Martinez and J. H. Werner, Nano Lett., 2010, 10, 3106.
7. J. Zheng, C. W. Zhang and R. M. Dickson, Phys. Rev. Lett., 2004, 93, 077402.
8. J. P. Xie, Y. G. Zheng and J. Y. Ying, J. Am. Chem. Soc., 2009, 131, 888.
9. C. L. Liu, H. T. Wu, Y. H. Hsiao, C. W. Lai, C. W. Shih, Y. K. Peng, K. C. Tang, H. W. Chang, Y. C. Chien, J. K. Hsiao, J. T. Cheng and P. T. Chou, Angew. Chem., Int. Ed., 2011, 50, 7056.
10. T. A. C. Kennedy, J. L. MacLean and J. W. Liu, Chem. Commun., 2012, 48, 6845.
11. M. Zhu, C. M. Aikens, F. J. Hollander, G. C. Schatz and R. Jin, J. Am. Chem. Soc., 2008, 130, 5883.
12. Z. K. Wu and R. C. Jin, Nano Lett., 2010, 10, 2568.
13. P. P. Bian, J. Zhou, Y. Y. Liu and Z. F. Ma, Nanoscale, 2013, 5, 6161.
14. X. M. Yang, L. Yang, Y. Dou and S. S. Zhu, J. Mater. Chem. C, 2013, 1, 6748.
15. J. Qiao, X. Y. Mu, L. Qi, J. J. Deng and L. Q. Mao, Chem. Commun., 2013, 49, 8030.
16. H. Kawasaki, K. Yoshimura, K. Hamaguchi and R. Arakawa, Anal. Sci., 2011, 27, 591.
17. Y. Chen, Y. Wang, C. X. Wang, W. Y. Li, H. P. Zhou, H. P. Jiao, Q. Lin and C. Yu, J. Colloid Interface Sci., 2013, 396, 63.
18. D. L. Miller, K. D. Kubista, G. M. Rutter, M. Ruan, W. A. D. Heer, P. N. First and J. A. Stroscio, Science, 2009, 324, 924.
19. R. Y. Li, Q. F. Xia, Z. J. Li, X. L. Sun and J. K. Liu, Biosens. Bioelectron., 2013, 44, 235.
20. Z. J. Li, X. L. Sun, Q. F. Xia, R. Y. Li, Y. J. Fang, S. P. Yang and J. K. Liu, Electrochim. Acta, 2012, 85, 42.
21. L. T. Zhou, R. Y. Li, Z. J. Li, Q. F. Xia, Y. J. Fang and J. K. Liu, Sens. Actuators, B, 2012, 174, 359.
22. C. S. Wang, J. Y. Li, C. Amatore, Y. Chen, H. Jiang and X. M. Wang, Angew. Chem., Int. Ed., 2011, 50, 11644.
23. Y. Chen, Y. Y. Shen, D. Sun, H. Y. Zhang, D. B. Tian, J. R. Zhang and J. J. Zhu, Chem. Commun., 2011, 47, 11733.
24. A. R. Biris, S. Ardelean, M. D. Lazar, E. Dervishi, F. Watanabe, A. Ghosh, A. Biswas and A. S. Biris, Carbon, 2012, 50, 2252.
25. S. X. Li, J. Z. Zheng, D. J. Chen, Y. J. Wu, W. X. Zhang, F. Y. Zheng, J. Chao, H. R. Ma and Y. L. Liu, Nanoscale, 2013, 5, 11718.
26. C. A. Diaz, Y. N. Xia, M. Rubino, R. Auras, K. Jayaraman and J. Hotchkiss, Nanoscale, 2013, 5, 164.
27. C. Z. Zhu and S. J. Dong, Nanoscale, 2013, 5, 10765.
28. Y. Tao, Y. H. Lin, Z. Z. Huang, J. S. Ren and X. G. Qu, Adv. Mater., 2013, 25, 2594.
29. W. S. Hummers and R. E. Offeman, J. Am. Chem. Soc., 1958, 80, 1339.
30. H. Lin, L. J. Li, C. Y. Lei, X. H. Xu, Z. Nie, M. L. Guo, Y. Huang and S. Z. Yao, Biosens. Bioelectron., 2013, 41, 256.
31. X. D. Xia, Y. F. Long and J. X. Wang, Anal. Chim. Acta, 2013, 772, 81.
32. Z. C. Xing, J. Q. Tian, A. M. Asiri, A. H. Qusti, A. O. Al-Youbi and X. P. Sun, Biosens. Bioelectron., 2014, 52, 452.
33. J. Q. Tian, Q. Liu, A. M. Asiri, A. H. Qusti, A. O. Al-Youbi and X. P. Sun, Nanoscale, 2013, 5, 11604.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c3ra47553k

---