A novel chiral electrochemiluminescence sensor that can discriminate proline enantiomers

Abstract

A sensitive, stable and stereoselective electrochemiluminescence (ECL) sensor has been designed to enantioselectively discriminate proline enantiomers by immobilizing Ru(bpy)₃²⁺–gold nanoparticles (Ru–AuNPs) and β-cyclodextrin–reduced graphene oxide (β-CD–rGO) on glassy carbon electrode. More Ru(bpy)₃²⁺ could be immobilized on the surface of electrode stably via preparing Ru–AuNPs and better stereoselectivity could be introduced to the sensor via the synthesis of β-CD–rGO. When the developed sensor interacted with proline enantiomers, obvious difference of ECL intensities towards L- and D-proline was observed, and a larger intensity was obtained from D-proline. As a result, ECL technique might act as a promising method to chiral recognition of amino acids enantiomers or chiral drugs.

---

Introduction

Chirality is one of the most fundamental and crucial properties of various natural systems,¹ and chiral recognition plays a very important role in understanding the interactions of biological molecules. Amino acids, as the building blocks for peptides, possess chirality.² Furthermore, amino acids enantiomers have identical physical and chemical properties but may exhibit different physiological and biochemical behaviors in biological processes. For example, L-proline (L-Pro) is a pivotal neuronal modulator or transmitter candidate in the central nervous system,³ while D-proline (D-Pro) is also present in mammals.⁴ But the D-Pro in liver might cause periportal fibrosis and necrosis of liver cells, and severe proximal tubular dystrophy and necrosis in kidney.⁵ So it is especially important to enantioselectively detect Pro enantiomers by using effective methods.

Electrochemiluminescence (ECL), as a new developing and powerful analytical tool for detection, has been widely used to chemical and biological analysis based on its distinct advantages of simplicity, rapidity, sensitivity, controllability and low background.^6–10 Among the ECL systems, Ru(bpy)₃²⁺ has received considerable attention due to high ECL efficiency.¹¹ What's more, solid-state Ru(bpy)₃²⁺ ECL by immobilizing Ru(bpy)₃²⁺ on electrodes surface provides several advantages over solution-phase ECL sensors, for it can reduce consumption of expensive reagents, simplify experimental design and enhance the ECL signal.^12–14 Recently, the appliance of nanoparticles to immobilize Ru(bpy)₃²⁺ has been widespread.^15–21 Graphene, due to the high surface areas, low cost, and high conductivity,²² holds great promise for potential applications in many technological fields.^23,24 In order to capitalize on the properties of graphene, diverse methods of chemical modification/doping have been developed.^25,26 Accordingly, ECL sensors based on graphene have faster response, better stability and larger ECL intensities.^27,28

The key to chiral recognition is the fabrication of the chiral interface materials. There are various approaches to construct effective chiral sensing interface.²⁹ β-cyclodextrin (β-CD), a cyclic oligosaccharide consisting of seven glucose units, has been widely used as chiral selector to construct chiral materials for specific recognition of guest molecules.^30,31 There are few reports depicting ECL technique in enantioselective detection amino acids enantiomers or chiral drugs.^32,33 In this work, the composites of Ru(bpy)₃²⁺–gold nanoparticles, graphene and β-cyclodextrin has been prepared to interact with D-/L-Pro.

Experimental

Reagents and materials

Ruthenium(II)-tris (2,2′-bipyridyl) chloride·6H₂O was purchased from SunaTech Inc (Zhengzhou, China). β-cyclodextrin was gotten from Aladdin. D- and L-proline (99%), gold chloride (HAuCl₄), hydrazine hydrate (Nafion (5 wt%)) were bought from J&k Chemical (Beijing, China). Ammonia (25–28%), sodium citrate were purchased from Chemical Reagent Co. (Chongqing, China). Graphene oxide (GO) was obtained from Nanjing Xianfeng nano Co. (Nanjing, China). 0.1 M phosphate buffer solution (PBS) at various pH values were prepared with KH₂PO₄ and Na₂HPO₄ containing 0.1 M KCl. Other chemical reagents were analytical grade and used without further purification. Double distilled water was used in all experiments.

Apparatus

The ECL measurements were carried out with a MPI-A electrochemiluminescence analyzer (Xi'an Remax Analysis Instrument Co., Ltd.) with the voltage of the photomultiplier tube (PMT) set at 800 V. A conventional three-electrode setup containing a modified glassy carbon electrode (GCE, 4.0 mm in diameter) as working electrode, a platinum auxiliary electrode, and a Ag/AgCl (saturated KCl) reference electrode was adopted. Morphologies of the nanomaterials were determined by a scanning electron microscope (SEM, S-4800, Hitachi, Tokyo, Japan). Confocal Laser Raman Spectrometer (Kenishaw Trading Co Ltd, Shanghai) was used to carry out the Raman spectroscopy. The FTIR spectrograms were obtained from Fourier Transform Infrared Spectrometer (Perkinelmer, USA).

Preparation of the Ru–AuNPs composites

AuNPs with a diameter of about 12 nm were prepared by sodium citrate reduction of HAuCl₄ in aqueous solution.³⁴ For preparing Ru–AuNPs, 2 mL of 0.5 mg mL⁻¹ Ru(bpy)₃²⁺ aqueous solution was added into 5 mL of AuNPs solution under vigorously stirring at room temperature.³⁵ Several minutes later, large amount of black precipitate was formed. The resulting precipitate was collected by centrifugation and washed with ultrapure water for three times. The obtained Ru–AuNPs composites were redissolved in 1 mL ultrapure water.

Synthesis of the β-CD–rGO nanosheets

The β-CD–rGO nanosheets were synthetized according to the literature.³⁶ In brief, 5 mL homogeneous graphene oxide dispersion (0.5 mg mL⁻¹), 5 mL of 80 mg mL⁻¹ β-cyclodextrin (β-CD) aqueous solution and 75 μL of ammonia solution was mixed completely, followed by the addition of 5 μL of hydrazine solution. And the mixed solution was well-distributed by vigorously stirring for a few minutes at room temperature. Then the vial was put in a water bath (60 °C) for 3.5 h. Stable black dispersion was obtained. The β-CD–rGO nanosheets were collected by centrifugation, and carefully washed with ultrapure water. The obtained nanosheets were redispersed in 5 mL ultrapure water. When not in use, it was stored at 4 °C.

Fabrication of the modified electrodes

Before modification, the glass carbon electrodes (GCE) (Ø = 4 mm) were first polished with 1.0, 0.3, 0.05 μm alumina slurry respectively, followed by rinsing thoroughly with ultrapure water. After that, the electrodes were successively sonicated in ultrapure water, ethanol, ultrapure water separately and dried at the room temperature.

The obtained Ru–AuNPs composites and β-CD–rGO nanosheets and 2.5 wt% Nafion were mixed completely at the aid of sonication (about 0.25 mg mL⁻¹ β-CD–rGO) to get the stereoselective ECL nanohybrids (denoted as CGRA). For fabrication of the modified electrodes, 10 μL resulting nanohybrids was dropped onto the glass carbon electrode (GCE) and dried at the room temperature (denoted as CGRA/Nafion/GCE). When not in use, the modified electrodes were stored at 4 °C.

Results and discussion

Characteristics

Raman spectroscopy is widely used to study the ordered and disordered crystal structures and scanning-electron-microscope (SEM) to characterize the morphologies and distribution of the nanomaterials, they were used to characterize β-CD–rGO nanosheets, Ru–AuNPs composites, and CGRA nanohybrids. Pristine graphite (Fig. 1A, a) displayed a prominent G band at 1580 cm⁻¹, corresponding to the first-order scattering of the E_2g mode observed for sp² domains, in addition to a tiny D band at 1352 cm⁻¹, which was ascribed to edge planes and disordered structures.³⁷ While the graphene oxide (GO) (Fig. 1A, b) showed a broadened and shifted G band to 1595 cm⁻¹, and the strongly strengthened D band at 1352 cm⁻¹ indicating the shrinking of the in-plane sp² domains, which might be due to the extensive oxidation.³⁸ For β-CD–rGO (Fig. 1A, c), the G band and D band (at 1590 and 1352 cm⁻¹, respectively) were coexisted still, whereas there was a slightly change of the intensity ratio of D/G comparing to GO. This change indicated that the average size of the sp² domains decreased after the reduction of GO and that the number of edge planes and the degree of disorder in rGO both increased, indicating the creation of numerous new graphitic domains in β-CD–rGO.^36,39

Fig. 1 (A) Raman spectra of graphite (a), GO (b), β-CD–rGO (c); SEM images of the Ru–AuNPs composites (B); CGRA nanohybrids (C).

Fig. 1B showed the appearance of Ru–AuNPs composites existed in aggregated form, which was attributed to the electrostatic interactions between Ru(bpy)₃²⁺ and negatively charged AuNPs. The SEM image of CGRA nanohybrids was presented in Fig. 1C, the dense coverage of Ru–AuNPs could be observed on the wrinkled β-CD–rGO surface, hinting the high surface areas of rGO may act as an effective matrix to immobilize more Ru–AuNPs.

The discussion of the reaction mechanism

Amino acids could take part in the ECL reactions with Ru(bpy)₃²⁺, which would act as oxidative-reduction co-reactants of Ru(bpy)₃²⁺. The ECL mechanism of reaction between Ru(bpy)₃²⁺ and proline has been discussed: at the electrode surface, the immobilized Ru(bpy)₃²⁺ was oxidized to Ru(bpy)₃³⁺ in eqn (1), while the negatively charged proline (pH 8.5, isoelectric point is 6.3)^40,41 was also oxidized to produce free radical in eqn. (2).^42,43 As described by eqn (3) and (4), the radical would undergo a deprotonation process to produce an intermediate radical ion, which could react with Ru(bpy)₃³⁺ to produce the excited state Ru(bpy)₃²⁺* resulting in light emission.^44–46 Following that, the amino group presented in by-product of proline might generate a secondary amine following by a dealkylation process in the presence of water.^45,47,48 The mechanism of Ru(bpy)₃²⁺–proline might be presumably analogous to that of the TPrA–Ru(bpy)₃²⁺ system.^42,49,50 The reaction processes were illustrated as follows:

Ru(bpy)₃²⁺ − e⁻ → Ru(bpy)₃²⁺ (1)

Ru(bpy)₃²⁺* → Ru(bpy)₃²⁺ + hv (5)

In our work, Pro enantiomers as the analyte participated in the ECL of Ru(bpy)₃²⁺, which would yield more ECL emission. As illustrated in Fig. 2, when Pro absent from the detection solution, obvious decreased ECL intensity of Ru(bpy)₃²⁺ was obtained, which could further confirm the feasibility of ECL sensor based on Ru(bpy)₃²⁺ in the detection of proline.

Fig. 2 Reaction scheme of ECL on surface of the proposed sensor.

Enantioselective recognition for Pro enantiomers via ECL

Based on the theories above, the designed electrodes were tested with the potential sweeping from 0.2 V to 1.25 V in 0.1 M phosphate buffer solution (PBS, pH 8.5) containing 5 mM D-/L-Pro. As shown in Fig. 3A, a great difference was observed on the ECL intensities between the D-Pro (curve a) and L-Pro (curve b), and the intensity of D-Pro was higher than L-Pro, which might be caused by the synergistic effects of the nano-materials.

Fig. 3 (A) ECL intensity–potential curves of CGRA/Nafion/GCE in 5 mM D-Pro (a) or L-Pro (b); (B) ECL intensity–potential curves of GRA/Nafion/GCE in 5 mM D-Pro (a) and L-Pro (b) 0.1 M PBS (pH 8.5, containing 0.1 M KCl) was used.

In order to examine ECL performances of the modified electrode without β-CD (denoted as GRA/nafion/GCE), corresponding measurements have been carried out. Fig. 3B showed that the ECL intensity decreased sharply and the value of difference (ΔI) between L-/D-Pro was also decreased without β-CD. In our study, β-CD was used as a chiral selector which might play a key role in the different ECL intensities from D- or L-Pro. It may capture proline inside the cavity to react with Ru(bpy)₃²⁺ on the surface of modified electrode, which could explain the sharp decrease of ECL intensity.^51,52 In addition, it was also a good dispersing reagent for graphene.^53,54 Therefore, the combination of β-CD and rGO could improve the solubility and electrochemical sensing performance comparing to unmodified graphene.^36,39

Mechanism of chiral recognition

The mechanism of enantiomeric selectivity may be due to the combination of different molecular interactions such as H-bonding and dipolar and steric interactions between hydroxyl groups of β-CD and proline.^5,52,55 Hydrogen bonding of polar –(NH)s– of proline to the –(HO)s on β-CD rims would take place when forming the inclusion complex of proline within the hydrophobic central cavity. What's more, the steric interactions between the D-/L-proline and the chiral carbons of D-glucose units of β-CD may also play a role in the chiral recognition.⁵⁶ FT-IR spectra was used to investigate the hydrogen bonds between the proline and the proposed chiral interface based β-CD.⁵⁷ Based on the results of the FT-IR spectra (shown in Fig. 4), the O–H stretching vibration peaks for CGRA (curve a) and L-Pro/CGRA (curve c) appeared at 3343 cm⁻¹ and 3341 cm⁻¹ respectively, and a slightly red shift peak was located at 3324 cm⁻¹ from D-Pro/CGRA (curve b).^36,58 From the understanding of the molecular level, the O–H stretching vibration peak could exhibit typical red-shift when hydrogen bonding is formed.⁵⁹ The results showed that more hydrogen bonds would be formed between D-proline and chiral sensing platform than L-proline. In this case, the –NH– in the L-enantiomers positioned less favorably for hydrogen bonding, leading to less inclusion complexation with chiral sensing platform, which may explain the chiral recognition.

Fig. 4 FT-IR spectra of CGRA (a), D-Pro/CGRA (b), and L-Pro/CGRA (c).

The influence of pH

The pH of the working buffer is an important influencing factor for the ECL intensities of Ru(bpy)₃²⁺ and the sensitivity of the modified electrodes. Because OH⁻ takes participate in the reaction of Ru(bpy)₃²⁺ and proline. The influence of the pH values ranging from 5.0 to 10.0 was investigated. As shown in Fig. 5, the highest ECL intensity and optimum chiral discrimination appeared at pH 8.5. Thus, the buffer solution of pH 8.5 was selected for further experiments.

Fig. 5 The effect of pH on the ECL intensities of the CGRA/Nafion/GCE in 5 mM D-Pro (a) or L-Pro (b).

Application of the enantioselective sensor

Fig. 6 showed the designed sensor was performed in quantitative analysis, for the ECL intensities increased with the increasing concentrations of L-Pro or D-Pro and displayed a linear with D-Pro or L-Pro in the range from 1.0 × 10⁻⁴ to 5.0 × 10⁻³ M. The regression equation was I_D- = 66.86 + 1056.86c or I_L- = 73.84 + 805.46c, and the detection limit was 2.5 × 10⁻⁵ M and 3.3 × 10⁻⁵ (S/N = 3), respectively. Comparing with L-Pro, the D-Pro obtained a higher ECL intensity with the designed ECL sensor. All of above results demonstrated the proposed stereoselective ECL sensor may provide a new perform on chiral discrimination of Pro enantiomers.

Fig. 6 The calibration plots with increasing concentrations of 0.1, 0.5, 1.0, 2.0, 3.0, 4.0 and 5.0 mM for D-Pro (a) and L-Pro (b) determination in 0.1 M PBS buffer (pH 8.5) [insets: ECL intensities of D-Pro and L-Pro with different concentrations, respectively].

The stability and reproducibility of the sensor

In present work, at the optimum pH 8.5, the investigation of stability of the sensors was carried out by immersing the sensors in 5 mM D-/L-Pro. After consecutive cyclic potential scans for 11 cycles at a scan rate of 100 mV s⁻¹, the intensities of sensor were observed remaining stable (in Fig. 7). The reproducibility of the sensors was examined in 5 mM D-Pro (pH 8.5) by preparing ten electrodes. The ECL intensities of the ten independent electrodes were similar and the relative standard deviation (RSD) was 4.9%, which achieved an acceptable target.

Fig. 7 ECL intensity–time curves of CGRA/Nafion/GCE in PBS (pH 8.5) containing 5 mM D-Pro (a) and L-Pro (b) under continuous cyclic potential scans for 11 cycles.

The selectivity of the sensor

In order to examine the selectivity of the designed sensor, the experiments for determination of more amino acids with the concentration of 5 mM were carried out. The amino acids used to study the interference were arginine (Arg), phenylalanine (Phe), lysine (Lys), tyrosine (Tyr), histidine (His), valine (Val), alanine (Ala), glutamic acid (Glu), cysteine (Cys), serine (Ser), tryptophan (Trp) and glutamine (Gln). As shown in Fig. 8, proline exhibited substantially greater chemiluminescence responses than other amino acids. The selectivity observed was due to proline containing the only secondary amine group of the naturally occurring amino acids.⁶⁰ For the other primary amino acids, the CL intensity of the reaction of primary amino acids with Ru(bpy)₃³⁺ varies with the electron-withdrawing character of the R-group attached to the α-carbon of the amino acid. Brune and Bobbitt have previously shown that amino acids with poorer electron-withdrawing characteristics could exhibit higher CL intensity.⁶¹ This could be well explained the good selectivity of the proposed sensor towards Pro enantiomers.

Fig. 8 Comparison of enantioselectivity of 13 kinds of amino acids (5 mM) on the CGRA/Nafion/GCE in PBS (pH 8.5).

Conclusions

In summary, a novel stereoselective ECL sensor was constructed based on β-CD–rGO and Ru–AuNPs, which was applied to discriminate Pro enantiomers. When the D-/L-Pro involved in the ECL of Ru(bpy)₃²⁺, the ECL response was enhanced greatly. Because of the secondary amine group of proline, it has the better ability of enhancing the ECL emission than the other naturally occurring amino acids containing primary amine group. The results of the experiments revealed that the ECL sensor exhibited obviously different ECL intensities towards Pro enantiomers, and a larger ECL intensity was obtained from D-Pro. Moreover, the experimental results also indicated that the selectivity, sensitivity, stereoselectivity, stability of the ECL sensor was acceptable. Hence, such an ECL sensor provides a promising method for the detection of amino acids or chiral drugs based on the same principle.

Acknowledgements

The authors gratefully acknowledge financial support for this study by the National Natural Science Foundation of China (No. 21272188).

Notes and references

1. J. Labuta, S. Ishihara, T. Šikorský, Z. Futera, A. Shundo, L. Hanyková, J. V. Burda, K. Ariga1 and J. P. Hill, Nat. Commun., 2013, 4, 1–8.
2. M. X. Zhang, G. Y. Qing and T. L. Sun, Chem. Soc. Rev., 2012, 41, 1972–1984.
3. R. T. Fremeau, M. G. Caron and R. D. Blakely, Neuron, 1992, 8, 915–926.
4. K. Hamase, S. Takagi, A. Morikawa, R. Konno, A. Niwa and K. Zaitsu, Anal. Bioanal. Chem., 2006, 386, 705–711.
5. D. Kampel, R. Kupferschmidt and G. Lubec, Amino Acids, 1990, 1164–1171.
6. P. Bertoncello, http://www.sciencedirect.com/science/article/pii/S0956566309000980-cor1, mailto: E-mail: paolo.bertoncello@dcu.ie, R. J. Forster, Biosens. Bioelectron., 2009, 24, 3191–3200.
7. A. J. Bard, Electrogenerated Chemiluminescence, Marcel Dekker, New York, 2004.
8. W. J. Miao, Chem. Rev., 2008, 108, 2506–2553.
9. M. M. Richte, Chem. Rev., 2004, 104, 3003–3036.
10. Z. Y. Liu, W. J. Qi and G. B. Xu, Chem. Soc. Rev., 2015, 44, 311.
11. B. A. Gorman, P. S. Francis and N. W. Barnett, Analyst, 2006, 131, 616–639.
12. L. H. Zhang, X. P. Sun and S. J. Dong, Biosens. Bioelectron., 2007, 22, 1097–1100.
13. L. L. Li, K. Liu, G. H. Yang, C. M. Wang, J. R. Zhang and J. J. Zhu, Adv. Funct. Mater., 2011, 21, 869–878.
14. Z. H. Guo, Y. Shen, M. K. Wang, F. Zhao and S. J. Dong, Anal. Chem., 2004, 76, 184–191.
15. Y. He, Y. Q. Chai and R. Yuan, Biosens. Bioelectron., 2013, 50, 294–299.
16. E. Rampazzo, S. Bonacchi, D. Genovese, R. Juris, M. Marcaccio, M. Montalti, F. Paolucci, M. Sgarzi, G. Valenti, N. Zaccheroni and L. Prodi, Coord. Chem. Rev., 2012, 256, 1664–1681.
17. G. Valenti, E. Rampazzo, S. Bonacchi, T. Khajvand, R. Juris, M. Montalti, M. Marcaccio, F. Paolucci and L. Prodi, Chem. Commun., 2012, 48, 4187–4189.
18. K. Kadimisetty, S. Malla, N. P. Sardesai, A. A. Joshi, R. C. Faria, N. H. Lee and J. F. Rusling, Anal. Chem., 2015, 87, 4472–4478.
19. D. F. Wang, L. H. Guo, R. Huang, B. Qiu, Z. Y. Lin and G. N. Chen, Sci. Rep., 2015, 5, 7954.
20. X. P. Sun, Y. Du, S. J. Dong and E. K. Wang, Anal. Chem., 2005, 77, 8166–8169.
21. G. F. Gui, Y. Zhuo, Y. Q. Chai, N. Liao, M. Zhao, J. Han, Y. Xiang and R. Yuan, RSC Adv., 2014, 4, 1955–1960.
22. A. K. Geim and K. S. Novoselov, Nat. Mater., 2007, 6, 183–191.
23. A. K. Geim, Science, 2009, 324, 1530–1534.
24. B. Seger and P. V. Kamat, J. Phys. Chem. C, 2009, 113, 7990–7995.
25. U. N. Maiti, W. J. Lee, J. M. Lee, Y. Oh, J. Y. Kim, J. E. Kim, J. Shim, T. H. Han and S. O. Kim, Adv. Mater., 2014, 26, 40–67.
26. S. Bai and X. P. Shen, RSC Adv., 2012, 2, 64–98.
27. Y. Y. Su and Y. Lv, RSC Adv., 2014, 4, 29324–29339.
28. W. H. Gao, Y. S. Chen, J. Xi, A. Zhang, Y. W. Chen, F. S. Lu and Z. G. Chen, Sens. Actuators, B, 2012, 171, 1159–1165.
29. K. Ariga, Y. Yamauchi, G. Rydzek, Q. Min Ji, Y. Yonamine, K. C.-W. Wu and J. P. Hill, Chem. Lett., 2014, 43, 36–68.
30. D. W. Armstrong, T. J. Ward, R. D. Armstrong and T. E. Beesley, Science, 1986, 232, 1132–1135.
31. J. Zhou, Y. Dai, S. Y. Wang, E. W. Zhu, J. F. Hai, Y. Liu, J. Tang and W. H. Tang, RSC Adv., 2012, 2, 5088–5093.
32. C. Z. Zhao, S. Y. Xu, Y. Su and G. L. Zhao, Analyst, 2002, 127, 889–891.
33. L. Wu, C. Q. Zhao, J. S. Ren and X. G. Qu, Chem.–Eur. J., 2014, 20, 11675–11679.
34. G. Frens, Nature, Phys. Sci., 1973, 241, 20.
35. M. Li, M. Zhang, S. G. Ge, M. Yan and J. H. Yu, Sens. Actuators, B, 2013, 181, 50–56.
36. Y. J. Guo, S. J. Guo, J. T. Ren, Y. M. Zhai, S. J. Dong and E. K Wang, ACS Nano, 2010, 4, 4001–4010.
37. M. S. Dresselhaus, A. Jorio, M. Hofmann, G. Dresselhaus and R. Saito, Nano Lett., 2010, 10, 751–758.
38. Y. W. Zhu, S. Murali, W. W. Cai, X. S. Li, J. W. Suk, J. R. Potts and R. S. Ruoff, Adv. Mater., 2010, 22, 3906–3924.
39. L. Tan, K. G. Zhou and Y. H. Zhang, Electrochem. Commun., 2010, 12, 557–560.
40. L. Pogliani, J. Pharm. Sci., 1992, 81, 334–336.
41. S. Yamada, C. Hongo and I. Chibata, Agric. Biol. Chem., 1977, 41, 2413–2416.
42. M. Hosseini, M. R. Moghaddam, F. Faridbod, P. Norouzi, M. R. K. Purb and M. R. Ganjali, RSC Adv., 2015, 5, 64669.
43. Y. F. Zhuang and H. X. Ju, Electroanalysis, 2004, 16, 1401–1405.
44. H. Hosono, W. Satoh and H. Suzuki, Sens. Mater., 2007, 19, 191–201.
45. R. J. Forster and C. F. Hogan, Anal. Chem., 2000, 72, 5576–5582.
46. J. G. Li, Q. Y. Yan, Y. L. Gao and H. X. Ju, Anal. Chem., 2006, 78, 2694–2699.
47. X. Chen, C. Q. Yi, M. J. Li, X. Lu, Z. Li, P. W. Li and X. R. Wang, Anal. Chim. Acta, 2002, 466, 79–86.
48. A. W. Knight and G. M. Greenway, Analyst, 1996, 121, 101R–106R.
49. H. W. Sun, M. Su and L. Q. Li, J. Chromatogr. Sci., 2010, 48, 49–54.
50. W. J. Miao, Chem. Rev., 2008, 108, 2506–2553.
51. H. Dai, C. P. Yang, X. L. Ma, Y. Y. Lin and G. N. Chen, Chem. Commun., 2011, 47, 11915–11917.
52. L. Szente and J. Szeman, Anal. Chem., 2013, 85, 8024–8030.
53. K. S. Liu, H. G. Fu, Y. Xie, L. L. Zhang, K. Pan and W. Zhou, J. Phys. Chem. C, 2008, 112, 951–957.
54. B. Konkena and S. Vasudevan, Langmuir, 2012, 28, 12432–12437.
55. Y. X. Tao, J. Y. Dai, Y. Kong and Y. Sha, Anal. Chem., 2014, 86, 2633–2639.
56. D. W. Armstrong, T. Ward, R. D. Armstrong and T. Beesley, Science, 1986, 232, 1132–1135.
57. X. Zhang, J. Yin and J. Yoona, Chem. Rev., 2014, 114, 4918–4959.
58. M. Yoshida, Y. Hatate, K. Uezu, M. Goto and S. Furusaki, Colloids Surf., A, 2000, 169, 259–269.
59. L. F. Scatena, M. G. Brown and G. L. Richmond, Science, 2001, 292, 908–912.
60. W. Jackson and D. R. Bobbitt, Anal. Chim. Acta, 1994, 285, 309–320.
61. S. N. Brune and D. R. Bobbitt, Anal. Chem., 1992, 64, 166.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra13528a

---