Highly sensitive and stabilized sensing of 6-benzylaminopurine based on NiCo₂O₄ nanosuperstructures

Abstract

NiCo₂O₄ nanosuperstructures that were synthesized by a facile hydrothermal method were developed for the determination of 6-benzylaminopurine (6-BAP). As a novel electric material, NiCo₂O₄ can be utilized to structure an ultrasensitive and prominently stable electrochemical sensor. The electrochemical investigation confirmed that NiCo₂O₄ nanosuperstructures stabilized via polyethyleneimine (PEI) could remarkably boost the electrocatalytic activity for the oxidation response of 6-BAP owing to its unique properties of large pore volume, good electrical conductivity and metal catalysis. The detailed electrochemical behavior of 6-BAP at PEI-NiCo₂O₄/GCE exhibited a higher peak current and more negative oxidation potential than the bare glassy carbon electrode (GCE). Simultaneously, its kinetic and thermodynamic parameters were further calculated, indicating that this sensing platform accomplished excellent electrochemical catalysis towards 6-BAP. The fabricated sensor of PEI-NiCo₂O₄/GCE for 6-BAP presented a wide linear range from 10⁻¹⁰ to 10⁻⁵ M with a low detection limit of 0.1 nM. The modified electrode exhibited long-term stability and high sensitivity and showed good practicability for determining contamination in food and even broader applications.

---

Introduction

6-Benzylaminopurine (6-BAP) is a class of chronic toxic substance; receiving extra doses of it could cause irritation to mucous membranes, the upper respiratory tract, eyes, and skin and even cause cancer.¹ Therefore, it is necessary to develop an effective and ultrasensitive approach to determine 6-BAP. To date, many methods have been developed to determine 6-BAP such as ultraviolet-visible (UV-Vis) spectrophotometry,² high performance liquid chromatography (HPLC),³ and electrochemical methods.^4,5 However, the sensitivity of UV-Vis spectrophotometry is limited, and not suitable for trace levels of 6-BAP. HPLC, its mobile phases almost were poisonous. Thus, an electrochemical method was a better choice for 6-BAP analysis, because it facilitates many remarkable advantages, especially speediness, cost-saving, and obtaining real-time detection.⁴ To improve the stability and sensitivity of 6-BAP determination, various modified materials were used for constructing ultrasensitive electrochemical sensors. Porous materials have received considerable attention because of their high porosity and large surface area.

Recently, the spinel nickel cobaltite (NiCo₂O₄) was one kind of highly conductive binary metal oxide dramatically regarded as a desirable electrode material due to its inherent conductivity characteristics.⁶ Typically, the nanosuperstructure⁷ of NiCo₂O₄ was composed of uniform small granules, resulting in increasing the specific area and offering an open three-dimensional (3D) structure to accommodate a large amount of superficial electroactive species for participation in the electrochemical reaction.^8,9 Nevertheless, there are hardly any studies reporting research about the electrocatalytic activity of NiCo₂O₄ stabilized by polyethyleneimine (PEI) applied to the analysis of practical samples.

The nanosuperstructure of NiCo₂O₄ gets tightly anchored to the surface of the electrode by PEI, offering good conductivity and large surface area, which can strongly improve the performance of fast electron transport pathways and electrochemical activity.⁸ Thus, the PEI-NiCo₂O₄ modified film is an excellent electro-oxidation catalyst sensor for the determination of 6-BAP with high sensitivity. Consequently, this electrochemical sensor showed good performance, easy construction, low cost, stability, speed, and sensitive response as well as a wide linear range. Furthermore, this PEI-NiCo₂O₄ modifier was a good prospect to develop a new type of chemical sensor applied for other contaminants in food with high sensitivity and stability.

Experimental

Synthesis of NiCo₂O₄ nanosheets

Preparation of the porous NiCo₂O₄ nanosuperstructure was undertaken as follows: 4.5 mM of hexamethylenetetramine, 2 mM of Co(NO₃)₂·6H₂O and 1 mM of Ni(NO₃)₂·6H₂O were dissolved in solution containing 40 mL deionized water and 20 mL ethanol. Then, the mixture solution was transferred into a 100 mL Teflon-lined stainless steel autoclave and maintained at 150 °C for 12 h. After cooling to room temperature, the precipitates were collected by centrifugation and washed with ethanol and water repeatedly. The product was dried in the blast oven at 60 °C overnight. Finally, the product was annealed at 350 °C for 2 h in air to obtain porous NiCo₂O₄ nanosheets.

Preparation of modified electrodes

A bare glass carbon electrode (GCE, 3 mm in diameter) was obtained by burnishing on chamois with 0.3 μm alumina particles, then washed with ethanol and water repeatedly and dried naturally in air before use. 3 μL of 0.2 mg mL⁻¹ PEI was drop-cast on the prepared bare GCE and air dried to obtain PEI/GCE. In addition, for preparation of the PEI-NiCo₂O₄ modified electrode (PEI-NiCo₂O₄/GCE), 0.2 mg mL⁻¹ of PEI was added into 3 mg mL⁻¹ NiCo₂O₄ solution. 3 μL of PEI-NiCo₂O₄ solution was deposited on the freshly prepared GCE surface with a microinjector, and placed under an infrared lamp to dry. When cooled to room temperature the PEI-NiCo₂O₄/GCE was obtained.

Results and discussion

Characterization of NiCo₂O₄

The morphological property of the spinel NiCo₂O₄ nanosuperstructures is shown by the scanning electron microscopy (SEM) image in Fig. 1A. Obviously, the resulting product was composed of nanosheets with leaf-like morphologies and lots of serrations that increased interconnected sites to form a typical three-dimensional structure. Typically, these nanosheets also exhibited lots of open space and electroactive surface sites between them, which could enhance the contact area with the electrolyte and enable a fast electron transport process. Transmission electron microscopy (TEM) was further carried out to investigate the structure of the NiCo₂O₄ nanosheets, as shown in Fig. 1B. It indicated that this nanosuperstructure may be constructed with granules of approximately 10–15 nm diameter, the bright area seems like some mesopores,¹⁰ providing tremendous active sites for electro-catalysis during the redox reaction. To further understand the composition and structure of the NiCo₂O₄ samples, the phases of the NiCo₂O₄ were identified by X-ray diffraction (XRD), as shown in Fig. 1D, wherein the typical XRD patterns were assigned to (111), (220), (311), (222), (400), (422), (511), and (440) plane reflections of the spinel NiCo₂O₄ crystalline structure, which are consistent with the standard JCPDS (card no. 73-1702) suggesting that this NiCo₂O₄ was formed of a homogeneous spinel phase with different exposed crystal planes.¹¹ The porous characteristics of NiCo₂O₄ were also investigated by nitrogen adsorption–desorption isotherm measurement shown in Fig. 1C. It was observed that the typical hysteresis loop in the relative pressure range of 0.6–1.0 P/P₀ was present and demonstrated the presence of mesopores.¹² As shown in the inset of Fig. 1C, from the Barrett–Joyner–Halenda (BJH) pore size distribution curve, a wide ranging distribution of mesopores¹³ could be observed. Therefore, it could be concluded that the NiCo₂O₄ nanosuperstructures exhibited large pore volume, abundant open space and electroactive surface sites,¹⁴ which could contribute to the activity, selectivity and stability of the electrode.

Fig. 1 (A) SEM image of NiCo₂O₄. (B) TEM image of NiCo₂O₄. (C) N₂ adsorption–desorption isotherm of NiCo₂O₄. Inset shows the corresponding BJH pore size distribution of NiCo₂O₄. (D) XRD pattern of NiCo₂O₄.

Electrochemical characterization of GCE, PEI/GCE, PEI-NiCo₂O₄/GCE

Under the optimal conditions (Fig. S1†), the electrochemical characterization of different modified electrodes was investigated by cyclic voltammetry (CV) in 5 mM potassium hexacyanoferrate containing 0.1 M potassium chloride, as shown in Fig. 2A. Both the modified and unmodified electrodes have good reversibility towards K₃[Fe(CN)₆] solution and the redox peak current of PEI-NiCo₂O₄/GCE (curve c) was higher than those of GCE (curve a) and PEI/GCE (curve b), revealing the best conductivity and catalytic activity of PEI-NiCo₂O₄/GCE among the other electrodes. Moreover, owing to PEI containing a high positive charge,¹⁵ when NiCo₂O₄ is added to PEI solution as the modified film, synergistic effects occur at the modified layer, which might be caused by the fact that NiCo₂O₄ is a mesoporous material with large surface area, good electrochemical activity, and the possibility of promoting electron transfer reactions at a lower overpotential.¹¹ The peak current of PEI increased with the scan rates from 25 mV s⁻¹ to 300 mV s⁻¹ shown in Fig. 2B. As shown in Fig. 2C, the relationship of the peak currents and the square of scan rates was highly linear, indicating that this electrochemical reaction process on the PEI-NiCo₂O₄/GCE sensing interface was a diffusion controlled process. To investigate the electrochemical surface active area, chronocoulometry of K₃[Fe(CN)₆] at the GCE (curve a), PEI/GCE (curve b) and PEI-NiCo₂O₄/GCE (curve c) was conducted and the results are shown in Fig. 2E. Based on the slope of relationship of Q and t^1/2 shown in the Fig. 2F, the electrode active area could be calculated according to the following equation:¹⁶where n is the number of transferred electron, D is the diffusion coefficient of K₃[Fe(CN)₆], Q_ads is the faradic charge, c is the concentration of K₃[Fe(CN)₆], F and π have the usual values. By calculation, the value of the electrode active area of GCE was 0.071 cm², PEI/GCE was 0.111 cm² and PEI-NiCo₂O₄/GCE was 0.152 cm², suggesting the PEI-NiCo₂O₄/GCE had the largest electroactive surface area compared with the bare GCE and PEI/GCE. Furthermore, the capacitance ability of the modified layer was investigated by CV shown in Fig. 2D and its value could be calculated following the equation:¹⁷ C_CV = i/(sυ), where i, s, and υ are average charge or discharge current (A), surface area (cm²) and scan rate (V s⁻¹), respectively. Therefore, the capacitance of GCE (curve a) and PEI-NiCo₂O₄/GCE (curve b) were calculated to be 107.4 μF cm⁻² and 352.9 μF cm⁻², respectively, that is to say that the capacitance of PEI-NiCo₂O₄/GCE was 3 times higher than that of GCE; therefore, the sensor PEI-NiCo₂O₄/GCE had an excellent capacitance electrode performance. Moreover, the leaf-like NiCo₂O₄ nanosheets with different exposed crystal planes exhibited prominent electrochemical catalysis,¹⁸ because PEI-NiCo₂O₄ had the feasibility of a conduction pathway for electrons between the analyte and the electrode.

Fig. 2 (A) Cyclic voltammograms of bare GCE (a), PEI/GCE (b), PEI-NiCo₂O₄/GCE (c) in 5 mM K₃[Fe(CN)₆] and 0.1 M KCl containing 0.1 M PBS (pH 7.0) at 100 mV s⁻¹ scan rate. (B) Cyclic voltammograms of PEI-NiCo₂O₄/GCE in 5 mM K₃[Fe(CN)₆] containing 0.1 M KCl at various scan rates from 25 mV s⁻¹ to 300 mV s⁻¹. (C) The linear relationship between the peak currents (line (a) is the anode peak current, line (b) is the cathode peak current) and the square root of the scan rates. (D) Cyclic voltammograms of bare GCE (a, black line) and PEI-NiCo₂O₄/GCE (b, red line) in 1.0 M KCl solution range from 0.3 V to 0.6 V versus Ag/AgCl as reference electrode at 100 mV s⁻¹ scan rate. (E) Chronocoulometry of bare GCE (a), PEI/GCE (b), PEI-NiCo₂O₄/GCE (c) in 5 mM K₃[Fe(CN)₆] and 0.1 M KCl containing 0.1 M PBS (pH 7.0). (F) The linear relationship between the charge and the square root of scan time on bare GCE (a), PEI/GCE (b), PEI-NiCo₂O₄/GCE (c).

Electrocatalytic activity of PEI-NiCo₂O₄/GCE towards 6-BAP

To the best of our knowledge, the current density is an important value for obtaining qualitative information about the electrochemical reactions.¹⁹ According to the equation: J = I/S, where J is the current density, I is the current, and S is the surface area of electrode. The electrochemical characterization of different modified electrodes towards 6-BAP was investigated by current density shown in Fig. 3A. A weak oxidation peak was observed at bare GCE with potential about 1.01 V (curve a). In contrast, it was clear that the peak current density recorded at PEI/GCE (curve b) was higher than the bare GCE, and the oxidation peak potential was about 0.99 V, which was less positive than the bare GCE. PEI-NiCo₂O₄/GCE showed the highest oxidation peak; the peak current density was 6.5 A m⁻² and the most negative potential was about 0.93 V, less than the other electrodes, owing to the spinel NiCo₂O₄ nanosheets possessing good electrical conductivity and high electrochemical activity.^16,20 Based on these advantages of PEI-NiCo₂O₄/GCE, it was expected to provide an excellent analytical performance, sensitivity, and detectability towards 6-BAP.

Fig. 3 (A) Current density of bare GCE (a), PEI/GCE (b), and PEI-NiCo₂O₄/GCE (c) at 100 mV s⁻¹ scan rate. (B) LSVs of PEI-NiCo₂O₄/GCE at different scan rates. (C) The linear relationship of E–ln υ. (D) The linear relationship of I–υ^1/2. (E) Chronocoulometry of PEI-NiCo₂O₄/GCE. (F) The linear relationship of Q–t^1/2 in 0.1 mM 6-BAP containing 0.1 M PBS (pH 8.5).

For further investigating the reaction characteristics of 6-BAP at PEI-NiCo₂O₄/GCE, the effect of scan rates (υ) on the electrochemical behavior of 6-BAP was validated by the linear sweep voltammogram (LSV) shown in Fig. 3B. It was found that peak potential (E_pa) shifts positively accompanied with the increase of peak current (I_pa) upon the increase of scan rates. As shown in Fig. 3C, the linear relationship of E_pa and natural logarithm of scan rate was cited as the calibration equation of E_pa (V) = 0.0242 ln υ + 1.5756 (R² = 0.9987), which followed the following equation:²¹

Moreover, according to eqn (2), the value of α was 0.5 in the totally irreversible electrochemical reaction. Thus, the number of electrons transferred (n) could be easily found to be 2 from the slope of E vs. ln(υ) in Fig. 3C. Based on the abovementioned results and an earlier published report,²² a tentative overall 6-BAP oxidation reaction at PEI-NiCo₂O₄/GCE could be described as follows:

In addition, the good linear relationship of current and the square root of scan rate (I vs. υ^1/2) is shown in Fig. 3D. From this I_p = −24.135υ^1/2 + 91.43 (R² = 0.9903) which indicates the 6-BAP oxidation on PEI-NiCo₂O₄/GCE is a diffusion-controlled transfer process, which involves a two-electron and a two-proton transfer (Fig. S2†).

To obtain the value of the diffusion constant of 6-BAP, it was observed that the chronocoulometry of 6-BAP had a good response based on the PEI-NiCo₂O₄/GCE shown in Fig. 3E. Furthermore, the relationship of Q vs. t^1/2 from the Fig. 3F fitted to the Anson equation:¹⁶

In this system, the slope of Q vs. t^1/2 was 34 from the calibration equation as Q (μC) = −34.048t^1/2 − 2.1258 (R² = 0.9985), n = 2, F = 96 500, A = 0.152 cm², from which D_6-BAP = 1.08 × 10⁻⁴ cm² s⁻¹ could be easy calculated. Moreover, the value of the standard heterogeneous rate constant (k_s) of 6-BAP could be calculated from the following equation:²³

herein, the value of the E_p/2 was 0.72 V, when the current was half of the peak current. Thus, the k_s was 3.16 × 10⁻³ cm s⁻¹ for 6-BAP, which was larger than the literature reports,⁵ suggesting the oxidation progress of 6-BAP on the PEI-NiCo₂O₄/GCE has a faster sensing rate, exhibiting high electrocatalytic activity.

The apparent activation energy (E_a) of 6-BAP

The thermodynamic parameter of the apparent activation energy (E_a) was determined by examining the effect of temperature on the LSV behavior of 6-BAP ranging from 0 V to 1.5 V versus Ag/AgCl as a reference electrode at 100 mV s⁻¹ scan rate. The linear relationship between the reciprocal temperature and the natural logarithm of peak current on PEI/GCE (curve a) and PEI-NiCo₂O₄/GCE (curve b) are shown in the Fig. 4. The current intensity of peaks was increased with elevation of temperature, and the slope of curve (a) was larger than the curve (b) showing that the calibration equations: I_PEI/GCE (μA) = −1296/T + 8.4 (R² = 0.9893) and I_{PEI-NiCo2O4/GCE} (μA) = −973/T + 7.2 (R² = 0.9985), suggest that the oxidation of 6-BAP on PEI/GCE needs more energy than on PEI-NiCo₂O₄/GCE. The values of E_a could be calculated by the following equation:²⁴

Fig. 4 The relationship of PEI/GCE (a) and PEI-NiCo₂O₄/GCE (b) between natural logarithm of peak current and 1/T in 0.1 mM 6-BAP containing 0.1 M PBS (pH 8.5) in the range from 0 V to 1.5 V versus Ag/AgCl as the reference electrode at 100 mV s⁻¹ scan rate.

Thus, the apparent activation energies of 6-BAP on PEI/GCE and PEI-NiCo₂O₄/GCE were 10.77 and 8.09 kJ mol⁻¹, respectively. The lower energy on PEI-NiCo₂O₄/GCE revealed that the electrochemical oxidation process of 6-BAP on PEI-NiCo₂O₄/GCE was easier to occur than on PEI/GCE.²⁵ It was suggested that PEI-NiCo₂O₄ had high electrochemical activity and good electron transfer ability, due to the characteristics of superior electronic conductivity and metal catalysis of NiCo₂O₄ nanosuperstructures.

The catalytic rate constant (k_h) by chronoamperometry investigations

The relationship between I_C/I_L and t^1/2 under the optimal potential of 0.93 V was investigated, as shown in Fig. 5B. The catalytic rate constant for the oxidation of 6-BAP at PEI-NiCo₂O₄/GCE could be determined according to the following equation:²⁶

I_C/I_L = γ^1/2[π^1/2 erf(γ^1/2) + exp(−γ)/γ^1/2] (7)

where I_C and I_L are the currents at PEI-NiCo₂O₄/GCE in the presence and absence of 6-BAP, γ = k_hc₀t is the argument of error function, and the other parameters have their usual meanings. In our cases, the error function was almost equal to 1 and therefore the abovementioned equation could be reduced to the following equation:

I_C/I_L = π^1/2γ^1/2 = π^1/2(k_hc₀t)^1/2 (8)

Utilizing the slope of the calibration equation as I_C/I_L = 32.59t^1/2 − 3.4924 (R² = 0.9955) in Fig. 5B, the k_h value was calculated as 3.38 × 10⁶ cm³ mol⁻¹ s⁻¹, which indicated that the electro-oxidation of 6-BAP had a high catalytic rate on this modified electrode of PEI-NiCo₂O₄/GCE based on the fact that PEI-NiCo₂O₄ material exhibits good catalytic activity and inherent electrical conductivity. Eventually, an expected conclusion could be drawn that the excellent electrocatalytic activity and high electrochemical performances towards 6-BAP were achieved at PEI-NiCo₂O₄/GCE.

Fig. 5 (A) The plot of chronoamperometry current against the logarithm of 6-BAP concentrations at 0.93 V versus Ag/AgCl as the reference electrode at 100 mV s⁻¹ scan rate. (B) The linear relationship between I_C/I_L and t^1/2. (C) Interference test of PEI-NiCo₂O₄/GCE in 0.1 M PBS (pH 8.5) with 5 × 10⁻⁶ M 6-BAP in presence of 5 × 10⁻⁵ M VB₁ (a), VB₂ (b), AA (c), valine (d) and proline (e); 5 × 10⁻⁴ M Fe³⁺ (f), K⁺ (g) and Zn²⁺ (h).

Calibration curve

The relationship between I_pa and the logarithm of the concentration of 6-BAP is illustrated in Fig. 5A. The peak current is proportional to the concentration of 6-BAP in the range from 10⁻¹⁰ to 10⁻⁵ M, and the calibration equation is I_pa (μA) = 4.966 − 0.3324 log C (R² = 0.9905) with a low detection limit of 0.1 nM, which is lower than those mentioned in recent reports (shown in Table S1†). This sensor of PEI-NiCo₂O₄/GCE exhibited very high sensitivity in this application.

Selectivity, reproducibility and stability

Selectivity of PEI-NiCo₂O₄/GCE for the determination of 6-BAP was investigated by chronoamperograms to evaluate the effects of possible interference in bean sprout samples. 6-BAP often occurs with some electroactive biomolecules in fruits and vegetables, including inorganic metal ions and antioxidants such as vitamin B₁ (VB₁), vitamin B₂ (VB₂), ascorbic acid (AA), valine, proline, Fe³⁺, Zn²⁺ and K⁺. Therefore, the selectivity of PEI-NiCo₂O₄/GCE was examined by monitoring the influence of these interfering molecules in the presence of 5 × 10⁻⁶ M of 6-BAP, as shown in Fig. 5C. Even a 10-fold concentration of vitamin B₁, vitamin B₂, AA, valine, proline and a 100-fold concentration of Fe³⁺, Zn²⁺ and K⁺ had no influence on the determination of 6-BAP, meaning that this constructed method was adequate for the determination of 6-BAP in practical samples (Table S2†) with good selectivity and which could contribute to the further detection of other contaminants.

The reproducibility and repeatability of PEI-NiCo₂O₄/GCE for the established current response in 0.1 mM 6-BAP was investigated and is shown in Fig. S3.† The peak current of 6-BAP was determined by five PEI-NiCo₂O₄ modified electrodes with the same conditions after three days, showing that the relative standard deviation (RSD) was 4.6%. This result suggested the high level of repeatability between different electrodes. To investigate the storage stability of the senor, the PEI-NiCo₂O₄/GCE was stored in PBS (pH 8.5) at 4 °C in the refrigerator when not in use. It was found that the current responses remained about 90% of the initial values. Therefore, this sensor of PEI-NiCo₂O₄/GCE has good storage stability in this application.

Conclusion

Herein, NiCo₂O₄ nanosuperstructures obtained by a novel synthesis method were used as a fantastic electric material. The electrochemical sensor based on NiCo₂O₄ nanosuperstructures offered excellent electric activity, owing to the porous characteristics of NiCo₂O₄, resulting in great electrocatalytic activity towards the oxidation process of 6-BAP with a low detection limit of 0.1 nM. The fabricated PEI-NiCo₂O₄/GCE electrode exhibited strong catalytic activity and excellent electrochemical stability by its systemic kinetic and thermodynamic parameters. The experimental results demonstrated that a PEI-NiCo₂O₄/GCE electrochemical sensor offers a feasible approach to detect rudimental contaminants in vegetables and fruit with high sensitivity, stability, reproducibility and selectivity.

Acknowledgements

This project was financially supported by the NSFC (21205016, 21575024, 51502038), the National Science Foundation of Fujian Province (2011J05020), the Education Department of Fujian Province (JA14071, JB14036, JA13068), and the Foundation of Fuzhou Science and Technology Bureau (2015-S-160).

Notes and references

1. Y. F. Zhang, X. J. Bo and L. P. Guo, Anal. Methods, 2012, 4, 736.
2. G. Woldemichael, T. Tulu and G. U. Flechsig, Microchim. Acta, 2012, 179, 99.
3. S. Lee, G. Y. Kim and J. H. Moon, Anal. Methods, 2013, 5, 961.
4. G. Zhao, K. Z. Liu, S. Lin, J. Liang, X. Y. Guo and Z. J. Zhang, Microchim. Acta, 2003, 143, 255.
5. D. Sun and H. J. Zhang, Anal. Chim. Acta, 2006, 557, 64.
6. H. C. Chen, J. J. Jiang, L. Zhang, D. D. Xia, Y. D. Zhao, D. Q. Guo, T. Qi and H. Z. Wan, J. Power Sources, 2014, 248, 28.
7. T. J. Zhu, Q. S. Wu, P. Chen and Y. P. Ding, J. Organomet. Chem., 2009, 694, 21.
8. T. Y. Wei, C. H. Chen, H. C. Chien, S. Y. Lu and C. C. Hu, Adv. Mater., 2010, 22, 347.
9. C. Yuan, J. Li, L. Hou, X. Zhang, L. Shen and X. W. Lou, Adv. Funct. Mater., 2012, 22, 4592.
10. C. H. Wang, X. Zhang, D. C. Zhang, C. Yao and Y. W. Ma, Electrochim. Acta, 2012, 63, 220.
11. N. Padmanathan and S. Selladurai, RSC Adv., 2014, 4, 8341.
12. H. Wang, J. L. Guo, C. Qing, D. M. Sun, B. X. Wang and Y. W. Tang, Chem. Commun., 2014, 50, 8697.
13. M. U. A. Prathap and R. Srivastava, Electrochim. Acta, 2013, 108, 145.
14. Y. Li, L. L. Zou, J. Li, K. Guo, X. W. Dong, X. W. Li, X. Z. Xue, H. F. Zhang and H. Yang, Electrochim. Acta, 2014, 129, 14.
15. J. A. Syed, H. B. Lu, S. C. Tang and X. K. Meng, Appl. Surf. Sci., 2015, 325, 160.
16. Y. H. Zhang, L. H. Wu, W. Lei, X. F. Xia, M. Z. Xia and Q. L. Hao, Electrochim. Acta, 2014, 146, 568.
17. L. H. Tang, Y. Wang, Y. M. Li, H. B. Feng, J. Lu and J. H. Li, Adv. Funct. Mater., 2009, 19, 2782.
18. R. J. Zou, K. B. Xu, T. Wang, G. J. He, Q. Liu, X. J. Liu, Z. Y. Zhang and J. Q. Hu, J. Mater. Chem. A, 2013, 1, 8560.
19. B. Cheng and X. X. Zhang, Electrochim. Acta, 2014, 130, 785.
20. F. Wang, Y. J. Wu, K. Lu, L. Gao and B. X. Ye, Electrochim. Acta, 2014, 141, 82.
21. M. H. Pournaghi-Azar and R. Sabzi, J. Electroanal. Chem., 2003, 543, 115.
22. S. S. Lu, Y. P. Wen, L. Bai, G. B. Liu, Y. X. Chen, H. Du and X. Q. Wang, J. Electroanal. Chem., 2015, 750, 89.
23. J. Velasco, Electroanalysis, 1997, 9, 880.
24. R. B. H. Kacemm, N. Ouerfelli, J. V. Herraez, M. Guettari, H. Hamda and M. Dallel, Fluid Phase Equilib., 2014, 383, 11.
25. Y. Wang, X. Wang and C. M. Li, Appl. Catal., B, 2010, 99, 229.
26. J. H. Li, H. B. Feng, J. Li, Y. L. Feng, Y. Q. Zhang and J. B. Jiang, Electrochim. Acta, 2015, 167, 22.

---

Footnotes

† Electronic supplementary information (ESI) available: Fig. S: further analysis on optimization of detection conditions; Table S1: comparable figures of determining 6-benzylaminopurine; Table S2: the results of addition recovery test. See DOI: 10.1039/c5ra23343g

‡ These authors contributed equally to this study.

---