Ultrasensitive electrochemiluminescence aptasensor based on a graphene/polyaniline composite film modified electrode and CdS quantum dot coated platinum nanostructured networks as labels

Abstract

A simple electrochemiluminescence (ECL) aptasensor for adenosine triphosphate (ATP) based on graphene/polyaniline (GR/PANI) composite films modified glassy carbon electrode (GCE) was successfully fabricated. The GR/PANI composite film using graphite oxide (GO) and aniline as the starting materials was obtained through a new one-step large-scale electrochemical synthesis for the immobilization of the first probe. PNNs could be synthesized through the chemical reduction of H₂PtCl₆ by benzyl alcohol under microwave irradiation without the introduction of any surfactants, templates, or seeds. PNNs@CdS QDs were used as an excellent label with amplification techniques. Taking advantage of dual-amplification effects, the aptasensor could detect ATP quantitatively, in the range from 0.5 pM to 20 nM, with a limit of detection as low as 0.1 pM. The proposed ECL aptasensor should have wide applications in the diagnosis of genetic diseases due to its simplicity, low cost, and high sensitivity at extremely low concentrations.

---

1. Introduction

The ubiquitous function of adenosine triphosphate (ATP) as an energy equivalent in nature has resulted in a common folding pattern of ATP-binding proteins. Because of its common use as an energy source in nature, ATP seems to be the most prominent candidate.¹ Currently, aptamers have been extensively utilized to fabricate various biosensors owing to their simple synthesis and excellent selectivity and stability.^2,3 ATP has been used in the aptamer-based techniques.⁴ To date, several methods have been developed for the detection of ATP, such as electrochemical assays,⁵ electrochemiluminescence (ECL) biosensors,⁶ and fluorometry.⁷ Therefore, ECL is a valuable and powerful analytical technique owing to its inherent features, such as low cost, rapid determination, wide range of analytes and high sensitivity.⁸

Graphene (GR), a two-dimensional hexagonal packed sheet of carbon atoms, has attracted a lot of attention from different research fields due to its remarkable electronic, mechanical, and thermal properties and large variety of applications in nanoelectronics, sensors, absorber, supercapacitors, and hydrogen storage.^9–13 The physical and chemical stability of graphene in such an environment is of critical importance. In particular, graphene based nanocomposites are of scientific and industrial interest because of their enhanced properties arising from the high electrical and thermal conductivities,¹⁴ large specific surface area,¹⁵ and good biocompatibility of graphene.¹⁶ It is known that exfoliated graphene oxide (GO) precursor can be electrochemically reduced to graphene at cathodic potentials.^17,18 Meanwhile, aniline monomer can be polymerized at anodic potentials.¹⁹ Therefore, an electrochemical synthesis of GR/polyaniline (PANI) composite film in large scale by using GO and aniline as the starting materials was reported.²⁰ In this work, GR/PANI composite film were prepared successfully for the immobilization of first probe.

Nanoparticle-based amplification, one of the successful signal amplification strategies, has attracted special interests due to the outstanding optical, electronic, and biocompatible performance of nanoparticles.^21,22 Recently, 3D nanostructured materials have attracted widespread attention owing to their high surface areas and abundant active sites.^23,24 Platinum nanostructured networks (PNNs) possessed superior electrochemical activity and stability because of the unique 3D nanostructured networks and active site formed during the assembly process, thus it could be excellent carriers in the transduction amplification of recognition events.²⁵ As well known, semiconductor quantum dots (QDs) have been widely used for constructing biosensors due to their numerous advantageous features, such as broad excitation spectra for multicolor imaging, robust and narrowband emissions, and feasibility for surface modification.^26–28 They have been widely used as ECL and luminescence labels for bioassays and bioimagings.^29,30 In this work, PNNs were successfully prepared and employed as an ECL signal material carrier to obtain a novel biocompatible ECL signal amplifier PNNs@CdS QDs composites.

In this paper, a sensitive ECL aptasensor based on a graphene/polyaniline (GR/PANI) composite film modified glassy carbon electrode and CdS quantum dot (QD) coated platinum nanostructured networks (PNNs@CdS QDs) was developed for the detection of adenosine triphosphate (ATP). With a high surface area, PNNs were employed as carriers for immobilization of CdS QDs. The PNNs@CdS QDs labels could be brought to the surface of GR/PANI composite film modified electrode through the first probe (ssDNA1)/ATP/the other one (ssDNA2) sandwich composites layer in the presence of ATP. The experimental results indicated that it exhibited good performance for detection of ATP with a wide linear range and a low detection limit.

2. Experimental

2.1. Chemicals and materials

Graphite powder (KS-10, 99.95%) were purchased from Alfa Aesar China Ltd. Adenosine 5′-triphosphate (ATP) cytidine triphosphate (CTP), guanosine triphosphate (GTP), uridine triphosphate (UTP) were purchased from Aladdin Chemistry Co., Ltd. N-succinimidyl-4-(N-maleimidome)cyclo-hexanecarboxylate (SMCC), dimethyl sulfoxide (DMSO), bovine serum albumin (BSA), and chitosan were obtained from Sigma-Aldrich Chemical Co. (St. Louis, MO). Thioglycolic acid (TGA), 1-ethyl-(3-dimethyl aminopropyl) carbodiimide hydrochloride (EDC), chloroplatinic acid, benzyl alcohol, cadmium chloride, sodium hydroxide and sodium sulfide were purchased from Alfa Aesar China Ltd. Phosphate buffered solutions (PBS) were prepared using 0.01 M KH₂PO₄ and 0.01 M Na₂HPO₄. The Ultrapure water obtained from a Lichun water purification system (≥18 ΜΩ cm, Jinan, China) was used in all assays and solutions. All other reagents were of analytical reagent grade and used without further purification. The sequences of the two oligonucleotides were as follows:

-ssDNA1: 5′-HS-(CH₂)6-ACCTGGGGGAGTAT-3′;

-ssDNA2: 5′-TGCGGAGGAAGGT-NH₂-3′.

2.2. Apparatus

The ECL experiments were carried out on a MPI-Bmultiple-parameter chemiluminescence analytical testing system equipped with a MPI-A/B multifunction chemiluminescence detector. (Xi'An Remax Electronic Science & Technology Co. Ltd. Xi'an, Changchun Institute of Applied Chemistry Chinese Academy Sciences, China) at room temperature. Cyclic voltammetric (CV) measurements were performed with a CHI760D electrochemical workstation (Shanghai CH Instruments, China). All experiments were carried out with a conventional three-electrode system with GCE as the working electrode (WE), a platinum counter electrode (CE) and an Ag/AgCl (sat. KCl) reference electrode (RE). GCE was cleaned by using the electrochemical technique in 0.1 M H₂SO₄ with a high scan rate of 0.1 V s⁻¹ and potential scanning between −0.2 V and 1.4 V until a reproducible CV was obtained. Electrochemical impedance spectroscopy (EIS) was carried out on an IM6x electrochemical station (Zahner, Germany). UV-vis spectra were recorded on a UV-3101 spectrophotometer (Shimadzu, Japan). The PL characterization was achieved on a LS-55 spectrofluorometer (P. E. USA). Transmission electron microscopy (TEM) images were obtained from a Hitachi H-800 microscope (Japan). Scanning electron microscopy (SEM) images were recorded using a QUANTA FEG 250 thermal field emission SEM (FEI Co., USA).

2.3. Synthesis of the GR/PANI composite film

The GR/PAIN composite film was synthesized according to a reported method.²⁰ 10 mg of GO was firstly dispersed in 1 M H₂SO₄, and then 100 μL of aniline monomer was added to give a brown dispersion, which was stirred for 30 min at room temperature. After that, the dispersion was centrifuged and rinsed with distilled water to remove the loosely adsorbed aniline monomer. The obtained GO/aniline composite was dispersed again into 10 mL distilled water to form 1 mg mL⁻¹ suspension. The electrochemical synthesis of GR/PANI composite film was carried out in a three-electrode system. The GO/aniline suspension was dropped onto GCE. The electrode was dried in a vacuum oven. Electrodeposition of the GR/PANI composite film was conducted by scanning the potential of the electrode between −1.3 and +1.0 V versus a saturated calomel electrode at 50 mV s⁻¹ in 1 M H₂SO₄. After deposition, the electrode was rinsed with distilled water. All experiments were carried out at room temperature.

2.4. Synthesis of CdS QDs

CdS QDs were synthesized according to the method similar to the reported previously.³¹ Thioglycolate and a solution of cadmium chloride were added to a flask. After adjusting pH to 11, the liquid was heated to 110 °C and refluxed for 30 min with the protection of N₂. Then, 5.5 mL of 0.1 M Na₂S solution was added into the mixture, which was refluxed at 110 °C with the protection of N₂ for 1 h. After the reaction, the solution was allowed to cool down naturally to room temperature.

2.5. Synthesis of PNNs

The PNNs were prepared as described previously.²⁵ An aqueous solution of H₂PtCl₆ (0.5 mL, 100 mM) was dropped in benzyl alcohol at room temperature, and the mixture was stirred for a few minutes for a transparent solution to form. A 50 mL flask containing the solution was then placed in a microwave oven and heated to 160 °C for a given time under magnetic stirring. After the reaction, the solution was allowed to cool down naturally to room temperature and the product was centrifuged, washed thoroughly with anhydrous ethanol. Finally, it was dried in a vacuum oven at 100 °C.

2.6. Preparation of PNNs@CdS QDs composites conjugated ssDNA2

Firstly, EDC was added into 500 μL of chitosan solution. After that, the mixture was dispersed in 5 mL of CdS QDs. Then, the PNNs were mixed with above solution and stirred for 2 h. Subsequently, the mixture was mixed with ssDNA2 for 1 h, followed by centrifuged and washed with distilled water. Finally, the PNNs@CdS composites conjugated ssDNA2 were obtained.

2.7. Preparation of ECL aptasensor

The steps for constructing the ECL aptasensor were shown in Scheme 1. A GCE of 3 mm diameter was polished carefully with 1.0, 0.3 and 0.05 μm alumina powder on the fine abrasive paper, separately. The alumina powder was sonicated in water to remove any residues. After removal of the trace alumina from the electrode surface, the electrode was rinsed thoroughly with water and cleaned by ethanol ultrasonically and then allowed to dry at room temperature. After the electrode was fabricated with GR/PANI composite film, 5 μL of SMCC was dropped onto the electrode surface. Subsequently, 5 μL of 2 μM ssDNA1 was dropped into the modified electrode. Afterwards, the electrode was incubated with 1% BSA for 1 h at the room temperature to block the possible nonspecific binding. After that, 5 μL of sample solution contained a varying concentration of ATP was mixed with PNNs@CdS labeled ssDNA2 (10 μL) at 30 °C for 0.5 h in buffer solution. Finally, 5 μL of the mixed solution was placed on the electrode, followed by washing with PBS to remove the nonspecific binding. Each step was washed by PBS thoroughly for three times.

Scheme 1 Schematic representation of the fabrication of the ECL aptasensor.

2.8. ECL measurements

ECL measurements were carried out in a solution of buffer (pH 7.4) containing 0.1 M K₂S₂O₈ as the coreactant at room temperature and the potential swept from −1.5 to 0 V. The ECL signals related to the ATP concentrations could be measured.

3. Results and discussion

3.1. Characterization GR/PANI composite film and CdS QDs@PNNs composite

As is showed in the Fig. 1A, the SEM image of dispersed single GO sheets showed corrugated flakelike shapes, which indicated GO was synthesized successfully. The GR/PANI composite film has a layered structure as shown in Fig. 1B, which is probably caused by the self-assembly effect of GR sheets during reduction or polymerization of aniline monomer. The electrochemical properties of the GR/PANI composite film were investigated by CVs. As shown in Fig. 1C, the bare GCE exhibited one set of well defined redox peaks (curve a). After the GR/PANI composite film was modified on GCE, the current increased obviously (curve b). In addition, the GR/PANI composite film had a significantly higher peak current and larger CV area compared with the bare one (curve b), indicating that the effective surface area of the GR/PANI composite film increased gradually in the process.

Fig. 1 SEM images of (A) GO, (B) the GR/PANI composite film; (C) CVs of (a) GCE and (b) the GR/PANI composite film under condition of 2 mM of K₃[Fe(CN)₆].

Fig. 2A showed the SEM image of PNNs and their size distributions. The image demonstrates that the networks are actually composed of many rough ligaments, and these ligaments contain numerous nanoparticles of several nanometers in size. TEM measurements further confirmed that the networks of ligaments are not solid (Fig. 2B). Individual nanoparticles could not be observed under TEM, suggesting that all the nanoparticles were tightly assembled together into networks.

Fig. 2 (A) SEM image, (B) TEM image of PNNs; (C) UV-vis absorption spectra of (a) CdS, (c) PNNs, (d) PNNs@CdS, and PL (λ_ex = 480 nm) (b) spectra of the as-prepared CdS QDs.

UV-vis and PL spectra were used to characterize the formation of CdS QDs. The UV-vis spectrum of the as-prepared CdS QDs showed a clear absorption at 431 nm (Fig. 2C, curve a). The size of the CdS QDs could be estimated to be 4.1 nm from the TEM image of CdS QDs (Fig. S1A†). The PL spectrum (excited at 480 nm) of the CdS QDs solution showed a strong emission peak with a maximum intensity at 580 nm (Fig. 2C, curve b). The TEM image of CdS QDs@PNNs (Fig. S1B†) was used to demonstrate the successful preparation of composites. A strong absorption at approximately 280 nm was obtained in the UV-vis spectrum of PNNs (curve c). CdS QDs@PNNs exhibited a absorption at about 434 nm and 324 nm, indicating the successful preparation of CdS QDs@PNNs composite (curve d).

3.2. EIS and ECL behaviors

EIS is an effective method for probing the surface features of modified electrodes. The impedance spectra include a semicircle portion and a linear portion. The semicircle diameter at higher frequencies corresponds to the electron-transfer resistance and the linear part at lower frequencies corresponds to the diffusion process. Fig. 3A showed the EIS at different stages. The bare GCE showed a relatively small electron-transfer resistance (Ret) (curve a). After the GCE was coated by the GR/PANI composite film, the film decreased the impedance, thus showed a smaller Ret (curve b). Remarkable increase in the Ret value was observed after the immobilization of thiol-modified ssDNA1 (curve c), this was because of the electrostatic repulsion between negative charged phosphate backbone of the oligonucleotides and the [Fe(CN)₆]^3−/4−. For aptamer in the presence of ATP (curve e), the impedance spectra included a semicircle portion at higher frequencies compared with aptamer in the absence of ATP. Stable and nearly insulating ssDNA1/ATP/ssDNA2 sandwich composites layer was formed through the ATP-induced split aptamer chips combination on the surface of the electrode when ATP was present.

Fig. 3 (A) EIS of (a) bare GCE, (b) GCE/GR/PANI composite film, (c) GCE/GR/PANI composite film/ssDNA1, (d) GCE/GR/PANI composite film/ssDNA1 after the addition of ssDNA2 in absence of 1 nM of ATP, and (e) GCE/the GR/PANI composite film/ssDNA1 after the addition of ssDNA2 in the presence of 1 nM of ATP in 0.01 M PB (10 mM Fe(CN)₆^3−/4− + 0.5 M KCl, pH 7.0); (B) cyclic ECL-potential curves obtained at (a) bare GCE, (b) GCE/GR/PANI composite film, (c) GCE/GR/PANI composite film/ssDNA1, (d) GCE/GR/PANI composite film/ssDNA1 after the addition of ssDNA2 in absence of 1 nM of ATP, and (e) GCE/the GR/PANI composite film/ssDNA1/CdS QDs in the presence of 1 nM of ATP and (f) GCE/the GR/PANI composite film/ssDNA1/PNNs@CdS QDs in the presence of 1 nM of ATP in the solution of 0.1 M, pH 7.4 PBS with 0.1 M of S₂O₈²⁻ as coreactant. (C) Cyclic ECL-potential curves obtained at (a) GCE/the GR/PANI composite film/ssDNA1/CdS QDs and (b) GCE/the GR/PANI composite film/ssDNA1/PNNs@CdS QDs in the absence of ATP.

To investigate the amplification technique of the PNNs@ CdS QDs composite for ECL analysis, a control experiment was carried out. In Fig. 3B, curve a showed ECL potential curves of bare GCE, and curve b, curve c, and curve d showed the ECL-potential curves of the GR/PANI composite film, ssDNA1 and ssDNA2 in the absence of ATP, respectively. No obvious increase of ECL emission could be observed from the above curve. Curve e showed the ECL-potential curves of the pure CdS QDs labeled ssDNA2, and curve f showed the ECL-potential curves of the PNNs@CdS QDs composite labeled ssDNA2. The quantity of the CdS QDs and PNNs@ CdS QDs composite was equal in both labels. As can be seen in Fig. 3B, the PNNs@ CdS QDs composite revealed excellent ECL performance compared with pure CdS QDs. The comparative results (Fig. 3C) indicated ECL intensity would be affected greatly by the PNNs@ CdS QDs composite. Therefore, the PNNs@ CdS QDs arrays showed a superior ECL performance than CdS QDs arrays. The results indicated that the enhanced ECL emission intensity was attributed to attached CdS QDs labeled ssDNA2 and PNNs@ CdS QDs labeled ssDNA2 on the electrode surface, which could react with S₂O₈²⁻ and enhance the ECL signal. In conclusion, the PNNs@CdS QDs composite had highly effective ECL properties, which were promising for the construction of the ECL biosensor.

3.3. Optimization of PH and the incubation time

To achieve the maximal ECL signal, the pH value and the incubation time were investigated using ATP concentration of 1 nM as follows. As shown in Fig. 4A, when the pH of buffer solution reached 7.4, the optimal response was achieved. To further evaluate the sensitivity of the assay, it is necessary to optimize the reaction time of the aptasensor. As shown in Fig. 4B, in the 30–120 min time period, the ECL value of the aptasensor was increased with the increasing time. After more than 120 min, the degree of ECL become stable with the reaction time prolonged. Consequently, considering the optimal analytical performance, the incubation time of 2 h was selected in the further study.

Fig. 4 Effect of (A) pH value and (B) incubation time on ECL intensity in the presence of 1 nM of ATP.

3.4. Analytical performance

Furthermore, ECL behaviors of aptasensor were studied in the solution of pH 7.4 PBS with 0.1 M of S₂O₈²⁻ as coreactant. In this system, ssDNA2 was functionalized with signal amplification composites (PNNs@CdS QDs) as detection probe, which provide a promising platform for the development of high-performance ECL aptamer sensor. In the presence of ATP, owing to the formation of a stable complex between the ssDNA1 and PNNs@CdS QDs modified ssDNA2, PNNs@CdS QDs were close to the electrode surface, resulting in ECL signal. Under the optimal conditions, the relationship between ECL intensity and ATP concentrations was shown in Fig. 5. It could be found that the ECL intensity increased linearly with the increased logarithm of the ATP concentration. The linear regression equation was I_ECL = 372.33 lg[ATP] (nM) + 2141.28 with a correlation coefficient of 0.98959, indicative of an acceptable quantitative behavior. The linear range for detection of ATP was 0.5 pM–20 nM, and the limit of detection (LOD) was 0.1 pM and the detection limit was estimated to be 0.1 pM at a signal-to-noise ratio of 3. Compared with other methods, the sensor has a relatively large linear range and low detection limit (Table 1). The results demonstrated that the proposed method could be used for the determination of ATP.

Fig. 5 Relationship between ECL intensity and ATP concentration (n = 8). (a) ECL signals of aptasensor under different concentrations of ATP; (b) logarithmic calibration curve of the ECL signals for ATP.

Table 1 Comparisons of other reported techniques for ATP detection

Detection methods	Linear range/nM	Detection limit/nM	Ref.
Fluorescence	0.1–10	1.3	32
Chemiluminescence	2–80	1.4	33
Differential pulse voltammetry	0.1–100	0.1	34
ECL	0.05–10	0.01	35
ECL	0.0005–20	0.0001	This work

---

3.5. Specificity, reproducibility, stability, and regeneration of the aptasensor

The specificity of the proposed ECL aptasensor was examined by detecting the ECL intensity of the functionalized electrode in the presence of three other ATP analogs: CTP, GTP, UTP. Fig. 6A exhibited the ECL signals of the aptasensor after it reacted with 1.0 nM ATP, 10 nM CTP, 10 nM GTP, 10 nM UTP, and the mixture, respectively. As can be seen in Fig. 6A, no significant increase of the ECL intensity was observed after incubation of CTP, GTP, and UTP. As comparison, the ECL signal was greatly increased after the incubation of ATP. Similarly, a mixed sample did not also exhibit apparent signal difference compared with that of ATP alone, indicating the coexisting species could not cause the observable interference. The results demonstrated that the proposed ECL aptasensor has a good selectivity for discriminating ATP from others.

Fig. 6 (A) The specificity of aptasensor to ATP (1 nM); (B) ECL vs. time curve of this aptasensor at an ATP concentration of 1 nM under continuous cyclic scans.

After the sensor was stored in pH 7.4 PBS at 4 °C over 30 days, it was used to detect the same ATP concentration. The ECL intensity of the sensor using PNNs@CdS QDs composite labeled ssDNA2 decreased to about 93% of its initial value, demonstrating that the sensor had good stability. Ten measurements of ECL emission upon cyclic scans of the ATP aptasensor in the presence of 200 pM ATP were shown in Fig. 6B. ECL measurements of the sensor upon continuous cyclic scans in PBS showed constant signals with relative standard deviation (RSD) of 2.3%, indicating the excellent stability. Thus, the constructed biosensor possessed good stability and was feasible for ECL detection.

The reproducibility of the DNA biosensor was estimated with intra- and inter-assay precision. The intra-assay precision was evaluated by assaying one ATP level for four replicate measurements. The inter-assay precision was estimated by determining one ATP level with four sensors made at the same electrode. The intra- and inter-assay variation coefficients (CVs) obtained from 1 nM ATP were 4.6% and 7.3%, respectively. Obviously, the inter-assay CV showed a good electrode-to-electrode reproducibility of the fabrication protocol, while the low value of intra-assay CV indicated that the sensor could be reproduced and used repeatedly.

Regeneration is an extremely important feature for biosensors in practical applications, so the regeneration of the aptasensor was also performed. In our test, the DNA biosensor could be regenerated by incubation of the modified electrode in hot water (90 °C) for 2 min, by which hybridized DNA was removed via thermal denaturation. The electrode was evaluated by five replicative measurements, showing accepted reusability. The consecutive measurements were repeated five times.

3.6. Possible ECL mechanism of CdS QDs

From the ECL mechanism for carbon dots, the ECL mechanism of CdS QDs system could be inferred.³⁶ Then S₂O₈²⁻ as coreactant could react with the negative charged CdS-QDs⁻ through electron transfer, which produced the excited state (CdS-QDs*) to emit light. The possible ECL mechanisms could be inferred as the following eqn (1)–(4)

CdS-QDs + e⁻ → CdS-QDs⁻* (1)

S₂O₈²⁻ + e⁻ → SO₄²⁻ + SO₄⁻* (2)

CdS-QDs⁻* + SO₄⁻* → CdS-QDs* + SO₄²⁻ (3)

CdS-QDs* → CdS-QDs + hv (4)

4. Conclusions

In summary, a novel aptasensor here by ECL measurement was proposed based on GR/PANI composite film. PNNs@CdS QDs composites were employed as an ideal ECL reagent, which had excellent property and ECL activity with amplification techniques. Taking advantage of the dual amplification, the ECL sensor exhibited excellent performances, including the good stability, high sensitivity, and satisfied stability and selectivity. This method was stable and provides a promising platform to fabricate cost-effective, simple, robust biosensors, which has great potential in point-of care applications for accurate gene diagnostics.

Acknowledgements

This work was financially supported by National Natural Science Foundation of China (21475052, 51273084, 21175058); Natural Science Foundation of Shandong Province, China (ZR2012BZ002).

References

1. L. Schmitt and R. Tampe, J. Am. Chem. Soc., 1996, 118, 5532–5543.
2. C. L. A. Hamula, H. Q. Zhang, L. L. Guan, X. F. Li and X. C. Le, Anal. Chem., 2008, 80, 7812–7819.
3. J. C. Lin and D. Thirumalai, J. Am. Chem. Soc., 2008, 130, 14080–14081.
4. J. Wang, Y. X. Jiang, C. S. Zhou and X. H. Fang, Anal. Chem., 2005, 77, 3542–3546.
5. L. Y. Feng, A. Sivanesan, Z. Z. Lyu, A. Offenhäusser and D. Mayer, Biosens. Bioelectron., 2015, 66, 62–68.
6. W. Yao, L. Wang, H. Y. Wang, X. L. Zhang and L. Li, Biosens. Bioelectron., 2009, 24, 3269–3274.
7. X. D. Zeng, X. L. Zhang, W. Yang, H. Y. Jia and Y. M. Li, Anal. Biochem., 2012, 424, 8–11.
8. J. Wang, Y. Shan, W. W. Zhao, J. J. Xu and H. Y. Chen, Anal. Chem., 2011, 83, 4004–4011.
9. R. S. Sundaram, M. Steiner, H. Y. Chiu, M. Engel, A. A. Bol, R. Krupke, M. Burghard, K. Kern and P. Avouris, Nano Lett., 2011, 11, 3833–3837.
10. J. C. Reed, H. Zhu, A. Y. Zhu, C. Li and E. Cubukcu, Nano Lett., 2012, 12, 4090–4094.
11. W. Cai, S. Z. Jiang, S. C. Xu, Y. Q. Li, J. Liu, C. Li, L. H. Zheng, L. B. Su and J. Xu, Opt. Laser Technol., 2015, 65, 1–4.
12. N. B. Trung, T. V. Tam, H. R. Kim, S. H. Hur, E. J. Kim and W. M. Choi, Chem. Eng. J., 2014, 255, 89–96.
13. C. X. Guo, Y. Wang and C. M. Li, ACS Sustainable Chem. Eng., 2013, 1, 14–18.
14. X. Li, Y. Chen, S. P. Mo, L. S. Jia and X. F. Shao, Thermochim. Acta, 2014, 595, 6–10.
15. D. Sfyris, G. I. Sfyris and C. Galiotis, Int. J. Non. Lin. Mech., 2014, 67, 186–197.
16. M. Gkikas, G. V. Theodosopoulos, B. P. Das, M. Tsianou, H. Iatrou and G. Sakellariou, Eur. Polym. J., 2014, 60, 106–113.
17. M. Lei, N. Wang, L. H. Zhu, C. S. Xie and H. Q. Tang, Chem. Eng. J., 2014, 241, 207–215.
18. P. A. Denis, C. P. Huelmo and F. Iribarne, Theoretical characterization of sulfur and nitrogen dual-doped graphene, Comput. Theor. Chem., 2014, 1049, 13–19.
19. M. Oh and S. Kim, Electrochim. Acta, 2012, 78, 279–285.
20. R. M. Li, Y. W. Ma, R. F. Chen, N. E. Shi, Q. L. Fan and W. Huang, Adv. Funct. Mater., 2011, 21, 2989–2996.
21. J. Qian, C. Y. Zhang, X. D. Cao and S. Q. Liu, Anal. Chem., 2010, 82, 6422–6429.
22. S. Pinijsuwan, P. Rijiravanich, M. Somasundrum and W. Surareungchai, Anal. Chem., 2008, 80, 6779–6784.
23. R. Edla, N. Patel, M. Orlandi, N. Bazzanella, V. Bello, C. Maurizio, G. Mattei, P. Mazzoldi and A. Miotello, Appl. Catal., B, 2015, 166–167, 475–484.
24. X. W. Li, D. Li, Z. W. Wei, X. N. Shang and D. Y. He, Electrochim. Acta, 2014, 121, 415–420.
25. F. L. Jia, F. F. Wang, Y. Lin and L. Z. Zhang, Chem. - Eur. J., 2011, 17, 14603–14610.
26. Z. X. Zou, D. Du, J. Wang, J. N. Smith, C. Timchalk, Y. Q. Li and Y. H. Lin, Anal. Chem., 2010, 82, 5125–5133.
27. C. Burda, X. B. Chen, R. Narayanan and M. A. El-Sayed, Chem. Rev., 2005, 105, 1025–1102.
28. B. J. M. Klostranec and W. C. W. Chan, Adv. Mater., 2006, 18, 1953–1964.
29. X. F. Hu, R. Y. Wang, Y. Ding, X. L. Zhang and W. R. Jin, Talanta, 2010, 80, 1737–1743.
30. J. Li, X. W. Zhao, Y. J. Zhao and Z. Z. Gu, Chem. Commun., 2009, 17, 2329–2331.
31. D. J. Lin, J. Wu, F. Yan, S. Y. Deng and H. X. Ju, Anal. Chem., 2011, 83, 5214–5221.
32. K. Y. Wang, J. Liao, X. Y. Yang, M. Zhao, M. Chen, W. R. Yao, W. H. Tan and X. P. Lan, Biosens. Bioelectron., 2015, 63, 172–177.
33. Y. H. Song, X. Yang, Z. Q. Li, Y. J. Zhao and A. P. Fan, Biosens. Bioelectron., 2014, 51, 232–237.
34. Y. Du, B. L. Li, F. A. Wang and S. J. Dong, Biosens. Bioelectron., 2009, 24, 1979–1983.
35. H. Chen, Q. Chen, Y. Y. Zhao, F. Zhang, F. Yang, J. Tang and P. G. He, Talanta, 2014, 121, 229–233.
36. Y. Zhang, W. Y. Liu, S. G. Ge, M. Yan, S. W. Wang, J. H. Yu, N. Q. Li and X. R. Song, Biosens. Bioelectron., 2013, 41, 684–690.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra12674f

---