An ultrasensitive electrochemical immunosensor for determination of estradiol using coralloid Cu₂S nanostructures as labels

Abstract

Herein, coralloid Cu₂S nanoparticles were prepared through a facile etching method and used as labels for the first time to fabricate an electrochemical immunosensor. A novel competitive immunoassay was then proposed using Cu₂S covalent conjugation with bovine serum albumin (BSA)–estradiol (E2) for the sensitive detection of trace E2 concentrations. Without using enzyme-label and acid dissolution, the Cu₂S can generate excellent electrochemical signals. The quantitative detection was based on the competitive binding of the E2 antibody with Cu₂S-labeled E2 or free E2. The redox signal decreased with increasing concentration of the free E2 as the amount of Cu₂S–BSA–E2 labels decreased at the immunosensor probe. Cyclic voltammetry and electrochemical impedance spectroscopy techniques were used to characterize the immunosensor. Square wave voltammetry was used to monitor the electrochemical response. The immunosensor exhibited a wide linear response from 25 to 7500 pg mL⁻¹ with a detection limit of 7.5 pg mL⁻¹. The proposed method showed good precision, broad linear range, and acceptable stability and could be used for the detection of E2 in real samples, which showed promising application in field research.

---

Introduction

Although estradiol (E2) is endogenously produced by all mammalian species, several studies have indicated that E2 is frequently released into the water environment through human activities.^1,2 As a typical endocrine disrupting compound, E2 is toxic to the environment because of its potential effects on animal reproductive efficiency and the immune functions of aquatic systems.^3,4 What is more, the presence of low concentrations of E2 in the environment would act on humans, cause abnormal sexual development and decrease the average number of human spermatozoa.⁵ Therefore, it is very important to establish a simple, sensitive and rapid method for the detection of E2 in the environment to protect the health of animals and human.

Various methods have become powerful for E2 analysis, such as high performance liquid chromatography,⁶ molecular imprinting technology,⁷ chemiluminescence⁸ and surface plasmon resonance sensing.⁹ Although determination of E2 by these methods is promising because of their high selectivity and sensitivity, these instruments are expensive, the operating processes are complicated and long, and pretreatment of samples is involved. Recently, several studies have focused on fabricating electrochemical sensors for ultrasensitive and fast detection of E2. From the molecular structure, E2 has electrochemical active due to its phenolic hydroxyl group, which can be oxidized on the electrode surface. However, direct electrochemical detection of E2 is very limited. He et al. used nano-Al₂O₃ film modified glassy carbon electrode (GCE) to detect E2 and got detection limit of 21.79 ng mL.¹⁰ By combining immunoassay with electrochemistry, high analytical speed, excellent selectivity, high sensitivity and economics, especially low detection limit can be obtained. Based on the highly specific molecular recognition of the immunoreaction, the E2 detection limit could generally achieve at picogram level. Many groups have fabricated electrochemical immunosensors for the determination of E2. Liu et al. developed an electrochemical immunosensor with underpotential deposition Cu|DTBP-Protein G scaffold to detect E2. The determination of E2 was based on the electrochemical signal produced by square wave voltammetry (SWV) oxidation of ferrocenemethanol in the detection solution. The increased amounts of sample E2 decreased the BSA–E2 conjugation on the immunosensor surface, which resulted in charge transfer resistance decrease of the immunosensor.¹¹ Ojeda et al. investigated an immunosensor which performed by applying a competitive immunoassay with peroxidase-labelled E2 (HRP-E2) and measured the amperometric response at −200 mV using hydroquinone (HQ) as redox mediator.¹² Chaisuwan et al. developed a competitive format to determine trace E2 concentrations using CdSe quantumdots (QDs) conjugation with BSA–E2. They used the in situ bismuth-coated carbon electrodes for detecting the cadmium ions (Cd²⁺) which was released from CdSe QDs during the acid dissolution step.¹³ The above mentioned methods have obtained favorable results, whereas development of an enzyme-free and without acid dissolution immunosensor is necessary. Cuprous sulfide (Cu₂S) nanomaterial has widely utilized in solar cells, optical filters, photoelectron slice and sensors.¹⁴ As a novel nanomaterial, Cu₂S has been used in sensors to detect hydrogen peroxide,¹⁵ glucose,¹⁶ polyphenols¹⁷ or ammonia gas.¹⁸ Cu₂S used in fabricating sensors were based on different synthetic methods and showed different morphology. And to the best of our knowledge, there are few reports based on Cu₂S for fabrication of immunosensor. In the present work, coralloid Cu₂S was prepared by a simple etching method. The process of preparing the Cu₂S is cheap, convenient and environmentally friendly. Due to the good electroactivity of Cu₂S, it could produce obvious redox signal in moderate environment. Using Cu₂S as labels could effectively avoid the shortcoming of enzyme labels. In addition, it does not need to release ions with HCl or HNO₃, which is time-consuming and the acid may lead to the denaturizing of the biomolecules.

Herein, we report on a novel electrochemical immunosensor for the determination of E2 by using Cu₂S incubated BSA–E2 conjugates modified gold electrode based on competitive immunoassay mode. The electrochemical signal was generated by the redox reaction of Cu₂S under SWV method. The concentration of E2 can be reflected directly because the decreased free E2 concentration in sample will increase the binding site of Cu₂S–BSA–E2 captured with the E2 antibody. The method has been successfully applied to analyze E2 in water samples and shows good recovery and accuracy, demonstrating its potential use in real sample analysis.

Experimental

Reagent and materials

Mouse monoclonal estradiol antibody (E2 Ab), E2 and BSA–E2 were purchased from Donglinchangsheng Biotechnology Co., Ltd. (Beijing, China). Mercaptoacetic acid, 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDC), N-hydroxysuccinimide (NHS), glutaraldehyde, high purity copper sheet, thiourea, ethylenediamine and other chemicals were obtained from Sinopharm Chemical Reagent Co., Ltd (Beijing, China). Phosphate buffered saline (PBS, 1/15 M Na₂HPO₄ and KH₂PO₄) was used as an electrolyte for all electrochemistry measurement. Ultrapure water was used throughout the experiment. All the chemicals were of analytical reagent grade.

All electrochemical behaviors were performed by a traditional three-electrode system. It contained a gold electrode (4 mm diameter) as the working electrode, a platinum wire electrode as the auxiliary electrode and a saturated calomel electrode (SCE) as the reference electrode. All electrochemical measurements were carried out using a CHI 760D electrochemical workstation (Chenhua Instrument Shanghai Co. Ltd, China). Transmission electron microscope (TEM) images were obtained from an H-800 microscope (Hitachi, Japan). Fourier transform infrared spectroscopy (FTIR) spectrum was obtained from VERTEX 70 (Germany). X-ray powder diffraction (XRD) was gathered from a Bruker D8 Focus diffractometer (Germany) using CuKα radiation (40 kV, 30 mA) of wavelength 0.154 nm.

Preparation of coralloid Cu₂S

Coralloid Cu₂S was synthesized as described by Zhang et al. with slight modifications.¹⁴ Briefly, copper sheet (1 cm²) was firstly polished by fine sandpaper. Then a clean copper sheet was prepared from consecutive ultrasonication in ethanol. Subsequently, it was immersed in a 5 mL test tube, which contains 0.8 mL 0.01 M thiourea and 0.1 mL ethylenediamine at room temperature for 3 days. Ethylenediamine has strong basicity and corrosivity. It could regulate the morphologies of nanomaterials. Thiourea could provide S element which constituted Cu₂S nanoparticles. Furthermore, ethylenediamine and thiourea both contain amine groups, which may result in the amine groups linking on Cu₂S. Finally, the nanoparticles were released from the Cu sheet by ultrasonication and washed with distilled water and ethanol for several times to remove the impurities before characterization.

Cu₂S conjugation to BSA–E2

The conjugation procedure of Cu₂S to BSA–E2 was as follows: Cu₂S was first pretreated by ultrasonication to disperse evenly. Then it was mixed with 0.5 mL of 2% glutaraldehyde solution and stirred for 2 h at room temperature. Subsequently, residual glutaraldehyde was removed by washing three times in PBS. The products were then centrifuged and dispersed in 0.5 mL of pH 7.4 PBS. Finally, 4 μg of BSA–E2 was added into the aforesaid products and shaken for 12 h at 4 °C. The amine groups on the Cu₂S surface were activated with glutaraldehyde to give a stable intermediate derivative which is used for covalent binding BSA–E2.¹⁹ The reaction mixture was then centrifuged and the supernatant was removed. Cu₂S–BSA–E2 bioconjugates were treated with 1 mL 1% BSA solution for 30 min to block non-specific binding. The mixture was centrifuged using PBS to remove any free E2 and BSA. The final mixture was redispersed in 1.0 mL PBS (pH 7.4) and stored at 4 °C before use. The synthesis of Cu₂S–BSA–E2 conjugates is schematically summarized in Fig. 1A.

Fig. 1 (A) Cu₂S incubate with BSA–E2 by covalent immobilization using glutaraldehyde and (B) schematic diagram of the detection principle of E2 with competitive immunoassay mode.

Preparation of competitive immunoassay

A gold electrode was first polished to a mirror-like finish with 0.05 μm alumina slurries to remove adsorbed impurities. After removal of the trace alumina from the electrode surface, the electrode was cleaned by ultrapure water and ethanol in an ultrasonic bath. Then the pretreated electrode was transferred to an electrochemical cell for cleaning by cyclic voltammetry (CV) between −0.4 and +1.5 V versus SCE at 50 mV s⁻¹ in 0.1 M H₂SO₄ until a stable CV profile was obtained. After rinsing, the mercaptoacetic acid covered electrodes were prepared by immersing the cleaned gold electrodes in an ethanol solution containing 10 mM mercaptoacetic acid in the darkness for 12 h.²⁰ E2 Ab was immobilized onto gold electrode by cross-linking E2 Ab onto the mercaptoacetic acid surface via the EDC/NHS-mediated amine coupling reaction. Briefly, mercaptoacetic acid covered electrodes were soaked in a solution comprised of 0.2 M EDC and 0.1 M NHS for 15 min.²¹ After cleaning, 4 μL E2 Ab (100 ng mL⁻¹) was dropped on the pretreated gold electrode, and incubated for 2 h at 4 °C. In order to minimize nonspecific binding, 2 μL 1% wt BSA was then added and incubated for 2 h at 4 °C. After drying, the electrodes were washed with pH 7.4 PBS. Then, a solution was prepared by mixing Cu₂S–BSA–E2 bioconjugation and an unknown amount of free E2. Finally, 6 μL of aforesaid mixture solution was dropped onto electrode and reaction for another 1 h at 4 °C. The procedures used for the construction of the immunosensor are depicted in Fig. 1B.

Measurement procedure

The PBS (pH 6.6) was used for all the electrochemical measurements. Electrochemical signals were measured using SWV technique. For SWV measurement of the immunosensor, it was scanned from −0.3 V to +0.6 V with a potential step of 5 mV, a frequency of 25 Hz, and amplitude of 25 mV.

Results and discussion

Characterization of Cu₂S and Cu₂S–BSA–E2 bioconjugates

Under the simple etching method, a layer of dense and black Cu₂S was grown onto pure copper sheet (Fig. 2A). A SEM image in Fig. 2B shows the morphology of the prepared Cu₂S and predicts that they were coralloid in shape. XRD analysis was used to characterize the crystalline structure of Cu₂S. According to Fig. 2C, the strong peaks at 37.43°, 45.77°, 48.48° and 53.67° are originated from (102), (110), (103) and (112) lattice planes, respectively. Moreover, the weak peaks at 26.49°, 29.20°, 54.56° and 55.22° are related to the lattice planes of (002), (101), (004) and (201), respectively.²² All the diffraction peaks matched well with those from the Jade PDF card (26-1116) for Cu₂S. Energy dispersive spectrometer (EDS) analysis of Cu₂S was further confirmed that the prepared nanomaterial was Cu₂S. Cu₂S was dropped on the aluminum foil to test. From Fig. 2D, it shows three major elements in the sample: Cu, S and Al. Obviously, Cu and S came from the Cu₂S, while Al originated from the aluminum foil. FTIR was used to characterize the functional groups on Cu₂S (Fig. 2E). A band at about 3129 cm⁻¹ in the spectrum showed N–H stretching vibration and a band at about 1623 cm⁻¹ showed N–H bending vibration.²³ The results indicated that amino groups have been grafted onto Cu₂S. The Cu₂S–BSA–E2 bioconjugates were characterized by UV-visible spectroscopy (Fig. 2F). All samples were dispersed in ultrapure water and mixed thoroughly. After laying aside for 2 min at room temperature, the UV-visible spectroscopy was monitored from 200 nm to 800 nm. For Cu₂S (curve a), there was no obvious absorption in visible region.²⁴ For BSA–E2 solution, there was a series of fluctuation at approximately 220–500 nm (curve b) and the peak at 280 nm was the characteristic absorption peak of protein BSA. When BSA–E2 was bonded onto Cu₂S, it was obvious that the spectrum of Cu₂S–BSA–E2 bioconjugates (curve c) was a superposition of curve a and curve b. Electrochemical methods are also suitable for investigating the interaction between biological molecules and other materials.²⁵ Due to the proteins are intrinsically unable to act as redox partners, the combined biological molecules will weaken the electrochemical signals obviously. From ESI Fig. S1,† the Cu₂S shows an enormous peak (curve a). However, after BSA–E2 combined on Cu₂S, the Cu₂S–BSA–E2 bioconjugation shows a relatively small peak (curve b). All the results show that BSA–E2 was conjugated onto Cu₂S successfully.

Fig. 2 Photograph (A) of (a) clean copper sheet and (b) covered Cu₂S; SEM images (B), XRD spectrum (C), EDS (D) and FTIR (E) spectrum of Cu₂S and UV-visible spectrum (F) of (a) Cu₂S, (b) E2–BSA and (c) Cu₂S–E2–BSA.

Electrochemical behavior of the immunosensor

Some parameters and optimization of conditions were shown in ESI.† The electrochemical behavior of Cu₂S–BSA–E2 bioconjugates was investigated before such competitive immunoassay mode established. When Cu₂S–BSA–E2 bioconjugates without free E2 coupled to the anti-E2 modified electrode, an obvious SWV peak can be seen from Fig. 3A curve b. However, there is no signal generated when only the free E2 coupled to the anti-E2 modified electrode (curve a). Thus, Cu₂S exhibited strong redox electrochemical signal and this performance was also detected under CV technique. After Cu₂S–BSA–E2 bioconjugates combined on the immunosensor, it shows a pair of strong redox signal (Fig. 3B curve b). But the free E2 does not show any peak (curve a). A reversible system can be seen from the CV. The reaction occurred on the electrode can be speculated according to the Nernst equation:²⁶
where R is constant term, R = 8.314 J K⁻¹ mol⁻¹; F is faraday constant, F = 96 485 J mol⁻¹ V⁻¹; n is electron transfer number. n can be calculated and equals about 1. Based on that, the redox reaction and the generated electrical signals may be produced by the mutual transformation of cuprous ion and cupric ion. Based on such appearance, Cu₂S can be used as a commendable label in fabrication of immunosensor.

Fig. 3 SW (A) and cyclic (B) voltammograms of the immunosensor, (a) without Cu₂S labels, (b) Cu₂S labels.

Electrochemical impedance spectroscopy (EIS) characterization of the immunosensor

EIS is one of the most powerful tools to investigate the interface properties of surface-modified electrodes and it was carried out to illustrate that all fabricating steps of the immunosensor were effective. It is well known that the high frequency region of the impedance plot shows a semicircle related to the redox probe Fe(CN)₆^3−/4−, followed by a Warburg line in the low frequency region which corresponds to the diffusion step of the over all process.²⁷ Curve a in Fig. 4 shows EIS of the bare gold electrode. It presented a straight line, implying that a very low electron transfer resistance in the electrolyte solution. After mercaptoacetic acid and anti-E2 covalently binding, an increased semicircle at high frequency region was observed (curve b). The reason of R_et increasing was that the nonconductive antibody obstructed the electron transfer of the redox probe from solution to electrode surface. The adsorption of BSA at the E2 Ab/gold electrode resulted in a dramatically increased diameter of the semicircle (curve c), indicating a higher electron transfer resistance at the electrode interface. After modified Cu₂S–BSA–E2 and 50 pg mL⁻¹ E2 on the electrode, the electron transfer resistance decreased (curve d). This decrease is attributed to the good electrical conductivity of Cu₂S, which presumably accelerates electron transfer at the electrochemical probe. The above results clearly confirm that the immunosensor had been fabricated successfully.²⁸

Fig. 4 EIS of gold electrode (a), E2 Ab/gold electrode (b), BSA/E2 Ab/gold electrode (c) and Cu₂S–BSA–E2 with 50 pg mL⁻¹ E2/BSA/E2 Ab/gold electrode (d) in 0.1 mol L⁻¹ KCl solution containing 10 mmol L⁻¹ [Fe(CN)₆]^3−/4− (1 : 1).

Analytical performance characteristics

SWV was used to evaluate the performance of the immunosensor. The peak currents of SWV were decreased with increasing concentration of free E2 under optimized conditions. Fig. 5A respectively shows the SWV response of 25 pg mL⁻¹, 50 pg mL⁻¹, 75 pg mL⁻¹, 100 pg mL⁻¹, 250 pg mL⁻¹, 500 pg mL⁻¹, 750 pg mL⁻¹, 1000 pg mL⁻¹, 2500 pg mL⁻¹, 5000 pg mL⁻¹ and 7500 pg mL⁻¹ (a–k, in order) of free E2 concentrations. The currents changed linearly with the concentrations of E2 in the range from 25 pg mL⁻¹ to 7500 pg mL⁻¹ with a detection limit of 7.5 pg mL⁻¹. The linear regression equations was i (μA) = 28.8843 − 4.9857 log c (pg mL⁻¹) with a statistically significant correlation coefficient of 0.9935 (Fig. 5B). The analytical performance of this fabricated immunosensor was compared with previously reported competitive immunoassay for the determination of estradiol.^13,29–31 The comparison results are presented in Table S1.† It can be seen that the developed immunosensor is more comparable, convenient, inexpensive and even exhibits better analytical performance toward estradiol than previously reported other methods.

Fig. 5 SW voltammogram (A) of 25, 50, 75, 100, 250, 500, 750, 1000, 2500, 5000 and 7500 pg mL⁻¹ (a–k, inorder); the calibration curve (B) of the current values vs. the E2 concentration. Error bar = RSD (n = 5).

Moreover, the reproducibility, selectivity, and stability of the immunosensor were also investigated. To evaluate the reproducibility of the immunosensor, a series of five electrodes were prepared for the detection of 50 pg mL⁻¹ E2. The relative standard deviations (RSD) of the measurements for the five electrodes were 4.7%. The results indicated that the precision and reproducibility of the proposed immunosensor was quite well.

A series of small organic chemicals (each 5 ng mL⁻¹) were selected, saccharose, ascorbic acid, bisphenol A and glucose, owing to their structural similarity to E2, as negative controls, in order to assess the selectivity of this immunosensor system for E2 (50 pg mL⁻¹). No remarkable difference of currents was observed in comparison with the result obtained in presence of only E2 detection. The current variation due to the interfering substances was less than 5.0% of that without interferences, indicating that the selectivity of the immunosensor was acceptable (Fig. 6A). Because of the specificity of the antigen–antibody binding complex, the interferents had almost no effect on the determination of analytes.

Fig. 6 Selectivity and stability of the immunosensor. Error bar = RSD (n = 5).

Stability of immunosensors is also a key factor in their application and development. The stability of the immunosensor was examined by checking periodically its current response. When the immunosensor was prepared and not in use, it was stored in a refrigerator at 4 °C. The current response of the as-prepared immunosensor decreased 4.2% after 7 day storage. Ten days later, the current response of the immunosensor decreased to about 88.6% of its initial value (Fig. 6B). The slow decrease in the current response may be due to the gradual denaturation of antibodies.

Application in analysis of water sample

In order to investigate the possibility of the immunosensor to be applied for practical analysis, the detection of E2 in water samples was performed using the proposed immunosensor with standard addition methods. 50 pg mL⁻¹, 100 pg mL⁻¹ and 250 pg mL⁻¹ of E2 solution were added into water samples. The recovery of detection E2 was from 94% to 104% and the RSD was in the range of 1.5–3.4% (Table S2†). The facts showed that the developed immunoassay methodology could be applied to the determination of E2 in water samples.

Conclusions

In summary, a novel electrochemical immunosensor for rapid determination of E2 by using Cu₂S–BSA–E2 bioconjugates was fabricated. Based on a competitive immunoassay strategy and the excellent redox electrochemical signal of CuS₂, the electrochemical immunosensor shows a linear response in the range of 25–7500 pg mL⁻¹ with a detection limit of 7.5 pg mL⁻¹. Due to the good precision, high sensitivity, acceptable stability and excellent selectivity, the electrochemical immunosensor has potential application for the rapid and simple measurement of E2 in real samples.

Acknowledgements

This study was supported by the Natural Science Foundation of China (no. 21175057, 21375047 and 21377046), the Science and Technology Development Plan of Shandong Province (no. 2014GSF120004), the Science and Technology Plan Project of Jinan (no. 201307010) and QW thanks the Special Foundation for Taishan Scholar Professorship of Shandong Province and UJN (no. ts20130937).

References

1. W. Jiang, Y. Yan, M. Ma, D. Wang, Q. Luo, Z. Wang and S. K. Satyanarayanan, J. Environ. Sci., 2012, 24, 320–328.
2. S. K. Atkinson, V. L. Marlatt, L. E. Kimpe, D. R. Lean, V. L. Trudeau and J. M. Blais, Sci. Total Environ., 2012, 430, 119–125.
3. M. Andrew-Priestley, W. O'Connor, R. Dunstan, L. Van Zwieten, T. Tyler, A. Kumar and G. MacFarlane, Aquat. Toxicol., 2012, 120, 99–108.
4. L. Sun, X. Shao, X. Hu, J. Chi, Y. Jin, W. Ye and Z. Fu, Aquat. Toxicol., 2011, 105, 629–639.
5. H.-S. Chang, K.-H. Choo, B. Lee and S.-J. Choi, J. Hazard. Mater., 2009, 172, 1–12.
6. M. Sun, L. Du, S. Gao, Y. Bao and S. Wang, Steroids, 2010, 75, 400–403.
7. L. Yuan, J. Zhang, P. Zhou, J. Chen, R. Wang, T. Wen, Y. Li, X. Zhou and H. Jiang, Biosens. Bioelectron., 2011, 29, 29–33.
8. L. Zhao, J.-M. Lin, Z. Li and X. Ying, Anal. Chim. Acta, 2006, 558, 290–295.
9. M. Miyashita, T. Shimada, H. Miyagawa and M. Akamatsu, Anal. Bioanal. Chem., 2005, 381, 667–673.
10. Q. He, S. Yuan, C. Chen and S. Hu, Mater. Sci. Eng. C, 2003, 23, 621–625.
11. X. Liu, X. Wang, J. Zhang, H. Feng, X. Liu and D. K. Wong, Biosens. Bioelectron., 2012, 35, 56–62.
12. I. Ojeda, J. López-Montero, M. Moreno-Guzmán, B. Janegitz, A. González-Cortés, P. Yáñez-Sedeño and J. Pingarrón, Anal. Chim. Acta, 2012, 743, 117–124.
13. N. Chaisuwan, H. Xu, G. Wu and J. Liu, Biosens. Bioelectron., 2013, 46, 150–154.
14. X. Zhang, L. Wang, R. Ji, L. Yu and G. Wang, Electrochem. Commun., 2012, 24, 53–56.
15. X. Bo, J. Bai, L. Wang and L. Guo, Talanta, 2010, 81, 339–345.
16. H. Lee, S. W. Yoon, E. J. Kim and J. Park, Nano Lett., 2007, 7, 778–784.
17. M. Xu, L. Cui, R. Han and S. Ai, J. Solid State Electrochem., 2012, 16, 2547–2554.
18. M. Shinde, D. Patil and R. Patil, Indian J. Pure Appl. Phys., 2013, 51, 713–716.
19. S. Eissa, C. Tlili, L. L'Hocine and M. Zourob, Biosens. Bioelectron., 2012, 38, 308–313.
20. L. Liu, Q. He, F. Zhao, N. Xia, H. Liu, S. Li, R. Liu and H. Zhang, Biosens. Bioelectron., 2014, 51, 208–212.
21. L. Liu, D. Deng, Y. Xing, S. Li, B. Yuan, J. Chen and N. Xia, Electrochim. Acta, 2013, 89, 616–622.
22. Y. Wang, X. Ai, D. Miller, P. Rice, T. Topuria, L. Krupp, A. Kellock and Q. Song, CrystEngComm, 2012, 14, 7560–7562.
23. T. Thongtem, A. Phuruangrat and S. Thongtem, J. Mater. Sci., 2007, 42, 9316–9323.
24. A. Kassim, H. S. MINa, L. K. Siang and S. Nagalingam, Chalcogenide Lett., 2011, 8, 405–410.
25. M. Guo, L. He, P. Strong and H. Wang, Chemosphere, 2014, 112, 472–480.
26. A. Bard, Electrochemical Methods, Wiley, New York, 1980.
27. M. A. Panagopoulou, D. V. Stergiou, I. G. Roussis and M. I. Prodromidis, Anal. Chem., 2010, 82, 8629–8636.
28. R. Feng, Y. Zhang, H. Yu, D. Wu, H. Ma, B. Zhu, C. Xu, H. Li, B. Du and Q. Wei, Biosens. Bioelectron., 2013, 42, 367–372.
29. D. Butler and G. G. Guilbault, Sens. Actuators, B, 2006, 113, 692–699.
30. X. Liu, P. A. Duckworth and D. K. Wong, Biosens. Bioelectron., 2010, 25, 1467–1473.
31. J. Li, S. Liu, J. Yu, W. Lian, M. Cui, W. Xu and J. Huang, Sens. Actuators, B, 2013, 188, 99–105.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra13066a

---