Chinese Journal of Chemical Physics  2016, Vol. 29 Issue (6): 729-734

#### The article information

Li-wen Wang, Shou-guo Wu, Tian-yang Shi, Wei Wei, Pan Pan

Electrocatalytic Oxidation of Calcium Folinate on Carboxyl Graphene Modified CuxO/Cu Electrode

Chinese Journal of Chemical Physics, 2016, 29(6): 729-734

http://dx.doi.org/10.1063/1674-0068/29/cjcp1604079

### Article history

Received on: April 17, 2016
Accepted on: July 25, 2016
Electrocatalytic Oxidation of Calcium Folinate on Carboxyl Graphene Modified CuxO/Cu Electrode
Li-wen Wang, Shou-guo Wu, Tian-yang Shi, Wei Wei, Pan Pan
Dated: Received on April 17, 2016; Accepted on July 25, 2016
Department of Chemistry, University of Science and Technology of China, Hefei 230026, China
*Author to whom correspondence should be addressed. Shou-guo Wu, E-mail:sgwu@ustc.edu.cn
Abstract: Carboxyl graphene modified CuxO/Cu electrode was fabricated. The bare copper electrode was firstly anodic polarized in 1.0 mol/L NaOH solution in order to get CuxO nanoparticles, then the carboxyl graphene (CG) was electrodeposited on the CuxO/Cu electrode by cyclic potential sweeping. The electrocatalytic oxidation behaviors of calcium folinate (CF) at the graphene modified CuxO/Cu electrode were investigated by cyclic voltammetry. A positive scan polarization reverse catalytic voltammetry was used to obtain the pure catalytic oxidation current. The graphene modified CuxO/Cu electrode was served as the electrochemical sensor of CF, a highly sensitivity of 22.0 μA·(μmol/μL)-1cm-2 was achieved, and the current response was linear with increasing CF concentration in the range of 2.0×10-7 mol/L to 2.0×10-5 mol/L, which crossed three orders of magnitude, and the detection limit was found 7.6×10-8 mol/L (S/N=3). In addition, the proposed sensor was successfully applied in determination of CF in drug sample.
Key words: Calcium folinate    Carboxyl graphene    CuxO/Cu electrode    Chemically modified electrode    Positive scan polarization reverse catalytic voltammetry
Ⅰ. INTRODUCTION

Calcium folinate (CF), also known as leucovorin calcium, is an active metabolic intermediate of folic acid. In clinical, CF commonly served as an antidote [1, 2] to methotrexate, an antineoplastic agent [3], in order to alleviate the side effect of high concentration of methotrexate [4]. Another application of CF is in cooperation with fluorouracil to improve the reactivity of fluorouracil [1]. Thus, a fast, sensitive and reliable assay method for the determination of CF is necessary. The proposed methods that have been reported for the detection of calcium folinate were mainly high-performance liquid chromatography (HPLC) which was combined with solid-phase extraction and detection with fluorescence [1] or ultraviolet (UV) [2], capillary electrophoresis (CE) [5] and so on. There also have developed many methods for simultaneous determination of CF, methotrexate and other similar structure compounds, such as HPLC with fluorescence or UV [1, 3, 6], CE with UV [7], LC with electrospray ionization mass spectrometry (ESI-MS) [4], LC-MS [8], etc. [9]. These methods are often time-consuming, high cost, and need tedious pretreatment, professional operators and multistage steps. In contrast, electrochemical methods are low-cost, time-saving, easy to operate and high sensitivity. Up to now, the reported electrochemical methods for detection of CF [10] are very few.

In the electrochemical biosensors, copper electrode is often chosen as a work electrode for its high conductivity and low cost. But the biggest challenge is to protect it from inactivation. Usually, there are two kinds of methods, the first one is to form a stable protection layer on the surface of copper [11, 12], and the second one is to form an oxidation layer. If a layer of copper oxide or other forms of films form on the surface of copper electrode, it can not only protect electrode from corrosion, but also improve the surface area and enhance the electrochemical properties [13]. Copper oxide is often chosen as sensing materials for its high catalytic activity and easy to prepare. In this work, we built a graphene protected CuxO/Cu electrode for CF detection. CuxO nanoparticles are generated on the surface of copper electrode by anodic polarization, and a layer of carboxyl graphene (CG) was subsequently covered on the surface of the CuxO nanoparticles by electrodeposition. With the protection of CG, the CG/CuxO/Cu electrode showed an excellent performance for CF detection.

Ⅱ. EXPERIMENTS

NaH2PO4·2H2O, Na2HPO4·12H2O, Na2CO3, NaOH, and NaHCO3 are analytical grade and purchased from Sinopharm Chemical Reagent Co., Ltd. and used without purification. CG was purchased from XF NANO, INC (Nanjing, China). CF standard substance (Chinese institute for food and drug control), and CF injection was purchased from Jiangsu Hengrui Medicine Co., Ltd. 0.5 mg/mL carboxyl graphene was dispersed in carbonate buffer solution (pH=10.0), 0.2 mol/L phosphate buffer (pH=10.4) was prepared by mixing appropriate amount of NaH2PO4·2H2O and Na2HPO4·12H2O, adjusted by adding NaOH solution and then used as supporting electrolyte. All solutions were prepared with deionized water.

All electrochemical measurements were performed on a LK2005 electrochemical work station (Tianjin LANLIKE Co., China) with a conventional three-electrode system. The modified copper electrode (2 mm) served as the working electrode, a platinum wire as the auxiliary electrode and a saturated calomel electrode as the reference electrode. Other commonly used instruments include KQ118 ultrasonic cleaner (Kunshan Ultrasonic Instruments Co., Ltd.), pH meter (Shanghai Precision & Scientific Instrument Co., Ltd.). X-ray photoelectron spectroscopy (XPS) was made at Thermo ESCALAB 250, Al Kα. Raman spectrum was obtained at LabRamHR (JY, France). All experiments were performed at room temperature.

Prior to modification, the bare copper electrode was firstly burnished with a fine SiC paper, then polished carefully with aqueous slurries of alumina powder of 1.0, 0.30, and 0.05 µm, successively, and then cleaned by ultra sonication for 5 min in acetone, ethanol and water respectively, so that the absorbed substances on the electrode surface can be removed thoroughly. After that, the copper electrode was anodic polarized at 0.8 V for 200 s in 1.0 mol/L NaOH solution to form the CuxO nanoparticles at the surface of the copper electrode. The CG/CuxO/Cu electrode was prepared by cyclic voltammetry (CV) in a carboxyl graphene suspended (0.5 mg/mL) Na2CO3-NaHCO3 solution (pH=10.0) from 0.4 V to -0.8 V at the rate of 50 mV/s for 10 cycles.

The CF injection was added directly into 0.2 mol/L phosphate buffer solution (pH=10.4), using the CG/CuxO/Cu electrode as working electrode, then the CF can be detected quantitatively by employing the positive scan polarization reverse catalytic voltammetry (PSPRCV) [14].

Ⅲ. RESULTS AND DISCUSSION A. The electrodeposition of the carboxylgraphene on the CuxO/Cu

Figure 1 shows the typical cyclic voltammograms of the electrodeposition of carboxyl graphene on the CuxO/Cu electrode. Two pairs of redox peaks can be observed. Compared with the literature [15, 16, 17], the first anodic peak at -0.19 V is attributed to the oxidation of Cu to Cu (Ⅰ). In alkaline solution, Cu (Ⅰ) hydroxide is the main product and then transforms to the corresponding oxide upon aging, the reaction can be described as,

 FIG. 1 Successive cyclic voltammograms for the electrodeposition of CG in 0.5 mg/mL CG carbonate buffer solution (pH=10.0) at the scan rate of 50 mV/s.
 ${\text{Cu}} + {\text{O}}{{\text{H}}^-} \to {\text{CuOH}} + {{\text{e}}^-}$ (1)
 $2{\text{CuOH}} \to {\text{C}}{{\text{u}}_2}{\text{O}} + {{\text{H}}_2}{\text{O}}$ (2)

The second anodic peak at about 0 V is attributed to Cu (Ⅰ) to Cu (Ⅱ),

 ${\text{CuOH}} + {\text{O}}{{\text{H}}^-} \to {\text{Cu}}{({\text{OH}})_2} + {{\text{e}}^-}$ (3)

which then transforms through the following process,

 ${\text{Cu}}{({\text{OH}})_2} \to {\text{CuO}} + {{\text{H}}_2}{\text{O}}$ (4)

There are no Cu (Ⅲ) species produced under the present conditions [18, 19].

Similarly, the cathodic peaks at -0.22 and -0.55 V belong to the reduction of Cu (Ⅱ) to Cu (Ⅰ), and Cu (Ⅰ) to Cu respectively. Both of the anodic and cathodic currents reduce gradually with the progress of successive cyclic scanning, which demonstrates the formation of CG layer on the surface of the CuxO/Cu electrode [20].

B. Characterization of the modified electrode

The chemical environments of both the CuxO/Cu and CG/CuxO/Cu electrodes were characterized by XPS. The curve fits of Cu 2p spectrum are shown in Fig. 2(A), which is assigned to 2p3/2 (930.8 eV) and 2p1/2 (952.2 eV). Figure 2(B) shows the main peaks assigned to the carbon bond, the peak at approximately 284.9 eV is assigned to C-C/C-H, whereas the peaks at 286.6 and 288.7 eV are attributed to C-O and C=O respectively [21]. These peaks of the CG/CuxO/Cu electrode mainly attribute to the CG. For the CuxO/Cu electrode, the intensities of peaks decrease greatly, these weak peaks may come from the adsorption of the small organic molecules in the environment. The chemical compositions of the film are summarized in Table Ⅰ. Carbon content increases from 30.22% to 56.35%, whereas copper decreases from 36.89% to 16.79%. All of the results indicate the effective deposition of CG.

 FIG. 2 XPS survey spectrum of the modified electrodes; (A) Cu 2p, (B) C 1s. (a) CuxO/Cu electrode, (b) CG/ CuxO/Cu electrode.
Table Ⅰ Element analysis of CuxO/Cu and CG/CuxO/Cu electrode by XPS.

The surface Raman spectroscopy was also made to characterize the formation of the graphene layer on the surface of the CuxO/Cu electrode. From Fig. 3, three strong peaks appear in the spectrum of the CG/CuxO/Cu electrode. These peaks present at 1356 cm-1 (D band), 1585 cm-1 (G band), and 2646 cm-1 (2D band) respectively are the characteristic peaks of graphene, thus further confirm the existence of CG layer. The D band is induced by disordered crystal structure of carbon, which attributes to the in-phase vibration of the carbon atoms. The G band namely the main characteristic peak belongs to in-plane bond stretching motion of sp2 carbon atoms. The 2D band, also called G' band, originates from the stacking order of the nanoplates [22, 23, 24, 25].

 FIG. 3 Raman spectrum of the modified electrodes, (a) CuxO/Cu electrode, (b) CG/CuxO/Cu electrode.
C. Electrocatalytic oxidation of CF at the CG/CuxO/Cu electrode

The electrochemical behaviors of CF at the modified electrodes were studied by cyclic voltammetry. Figure 4 shows the cyclic voltammograms in the absence and presence of 5 µmol/L CF in 0.2 mol/L PBS (pH=10.4) at the CuxO/Cu and CG/CuxO/Cu electrodes. The potential is scanned between -0.8 and 0.4 V at the scan rate of 50 mV/s. From Fig. 4(A), we can see that, in the absence of CF, when scanning from -0.8 V to 0.4 V, there is an anodic peak at -0.23 V (curve a), which attributes to the oxidation of Cu (0) to Cu (Ⅰ) or the direct oxidation of Cu (0) to Cu (Ⅱ). This reaction may correspond to either competing or stepwise processes [26]. In the reversal scan, two cathodic peaks at -0.40 and -0.59 V are observed, which can be assigned to the reduction of Cu (Ⅱ) to Cu (Ⅰ) and Cu (Ⅰ) to Cu (0), respectively. After adding CF (curve b), there is only slight decrease of current in the positive scanning. But in the reverse scanning, it shows a significant change. For the CuxO/Cu electrode (see Fig. 4(B)), the first reduction peak is almost unchanged both in potential and current, only the current of the second reduction peak decreases obviously. For the CG/CuxO/Cu electrode (see Fig. 4(A)), the potentials of two cathodic peaks shift from -0.4 V to -0.44 V and -0.59 V to -0.57 V respectively, and the relative peak current of Cu (Ⅱ) to Cu (Ⅰ) increases while decrease of Cu (Ⅰ) to Cu (0). The decrease in peak current of Cu (Ⅰ) to Cu (0) is ascribed to the catalysis of Cu (Ⅰ) to the oxidation of CF.

 FIG. 4 Cyclic voltammograms of (A) CG/CuxO/Cu, (B) CuxO/Cu modified electrodes in (a) the absence and (b) the presence of 5 µmol/L CF in 0.2 mol/L PBS (pH=10.4).

In positive scanning:

 ${\text{Cu}}({\text{0}}) \to {\text{Cu}}({\text{I}}) + {{\text{e}}^{\text{-}}}$ (5)
 ${\text{Cu}}({\text{I}}){\text{ + C}}{{\text{F}}_{{\text{red}}}} \to {\text{Cu}}({\text{0}}){\text{ + C}}{{\text{F}}_{{\text{ox}}}}$ (6)
 ${\text{Cu}}({\text{I}}) \to {\text{Cu}}({\text{II}}) + {{\text{e}}^-}$ (7)

In negative scanning:

 ${\text{Cu}}({\text{II}}) + {{\text{e}}^{\text{-}}} \to {\text{Cu}}({\text{I}})$ (8)
 ${\text{Cu}}({\text{I}}){\text{ + C}}{{\text{F}}_{{\text{red}}}} \to {\text{Cu}}({\text{0}}){\text{ + C}}{{\text{F}}_{{\text{ox}}}}$ (9)
 ${\text{Cu}}({\text{I}}) + {{\text{e}}^-} \to {\text{Cu}}({\text{0}})$ (10)

CFox and CFred represent the oxidative and reductive state of CF respectively. Reaction (6) makes reaction (5) easier and reaction (7) weaker, similarly reaction (9) makes reaction (8) easier and reaction (10) weaker. It can also be seen that there are two well-separated cathodic peaks and stronger catalytic activity due to the existence of CG layer. Meanwhile, the anodic peak of Cu (Ⅰ) to Cu (Ⅱ) (for reaction (7)) almost disappeares due to the catalytic reaction (6).

Figure 5 shows the CV response of the CG/CuxO/Cu electrode in 0.2 mol/L PBS solution with different concentration of CF. In positive scan, the peak currents have a slightly decrease with the increase of CF concentration. In reverse scan, both the reduction peak potentials and currents have a big change, the two peaks move closer and closer and almost overlap completely when CF concentration is high enough, which contribute to the oxidation of CF by Cu (Ⅰ). All the Cu (Ⅰ) ions reduced by Cu (Ⅱ) are used for the oxidation of CF, so the cathodic peak seems to be Cu (Ⅱ) directly reduction to Cu (0).

 FIG. 5 Cyclic voltammograms of modified CG/CuxO/Cu electrodes in different CF concentration solutions. The CF concentrations are 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, and 20 µmol/L as arrows show respectively.

We have also studied the influence of the potential scan rate. Figure 6(a) displays the influence of the scan rate on the peak currents of the CG/CuxO/Cu electrode in 0.2 mol/L PBS (pH=10.4). With the scan rate increasing, both the cathodic and anodic peak currents increased. The anodic peak current of Cu (Ⅰ) to Cu (Ⅱ) is obviously weak. The oxidation peak current of Cu (0) to Cu (Ⅰ) (ipa), reduction peak currents of Cu (Ⅱ) to Cu (Ⅰ) (ipc1) and Cu (Ⅰ) to Cu (0) (ipc2) are proportional to the square root of the scan rate (v) in the range from 10 mV/s to 100 mV/s (see Fig. 6(b)), with the correlation coefficients R2 of 0.9927, 0.9976, and 0.9984 respectively. This indicates the redox process of different states of copper ions is diffusion controlled. Furthermore, the oxidation peak shifted positively and the reduction peaks shifted negatively with the scan rate increasing, which means the redox reactions of the modified electrode is quasi-reversible.

 FIG. 6 (a) Cyclic voltammograms of the CG/CuxO/Cu electrode in PBS. The potential scan rates are 10, 20, 30, 40, 50, 60, 70, 80, 90, and 100 mV/s as an arrow shows. (b) The calibration plot of peak currentsvs. square root of the scan rate (v1/2).
D. Quantitative detection of CF by using the CG/CuxO/ Cu electrode with PSPRCV

The PSPRCV was proposed based on cyclic voltammetry by our group [14]. Like cyclic voltammetry, the potential first scans positively from initial to switching potential, the working electrode is anodic polarized, then the potential scans reversely (negatively) to the initial, the voltammogram is recorded. Finally, the blank data (without analyte) is deducted from the sample data (with analyte). The difference value peak current presents the pure catalytic current and is proportional to the bulk concentration of the analyte.

From Fig. 5, we can see that both the oxidation peaks and reduction peaks have regular change with increase of CF concentration. In the positive scan, the peak current has a slight change. However, in the negative (reverse) scan, the two reduction peaks change very much with increase of CF concentration. Based on PSPRCV, in the absence of CF, the recorded voltammogram is stored as the blank data, the two reduction peaks present the reduction of Cu (Ⅱ) to Cu (Ⅰ) and Cu (Ⅰ) to Cu (0) respectively. In the presence of CF, the recorded voltammograms stored as the sample data, the reduction peak current is amplified by the catalytic reaction. The difference value voltammogram is achieved by subtracting the blank data from the sample data, which presents the catalytic current caused by the oxidation of CF.

 FIG. 7 (a) The voltammograms of different concentrations of CF in the reverse scans, (b) the corresponding different value voltammograms of PSPRCV from 0.2, 0.6, 1, 2, 4, 6, 8, 10, 12, 14, 16, 18 to 20 µmol/L as arrows show. (c) The calibration curve of CF with the concentration range of 0.2-20 µmol/L.

Figure 7(a) shows the voltammograms of different concentrations of CF in the reverse scans. Figure 7(b) is the corresponding difference value voltammograms of PSPRCV. It appears two positive-negative peaks at -0.40 and -0.56 V respectively. Figure 7(c) shows the linear response of the first difference value peak currents to the concentration of CF, which is easier to be measured than the second peak. The linear regression equation, in the range of the CF concentration from 0.2 µmol/L to 20 µmol/L, is

 $i{\text{ = 1}}{\text{.27 + 0}}{\text{.69}}c$ (11)

where R2=0.9981. The limit of detection (LOD) is 0.076 µmol/L at the signal to noise ratio (S/N=3), and the current sensitivity of the modified electrode is 22.0 µA·μmol/L-1·cm-2. Table Ⅱ gives a comparision of the analytical performance between our work and the reported in literatures.

Table 2 The analytical performance of different methods for detection of CF. Liner range and LOD are in µmol/L.

To illustrate the reproducibility and repeatability of the CG/CuxO/Cu electrode, control tests were performed. Ten successive measurements for 5 µmol/L CF of 0.2 mol/L PBS resolution, the peak currents of the modified electrode gave a relative standard deviation of 1.55%, while the CuxO/Cu electrode gave a RSD 12.8%, which is much larger than the CG protected CuxO/Cu electrode. Also, seven different solutions containing 5 µmol/L of CF were tested, the CG/CuxO/Cu electrode gave a RSD of 8.82%, whereas the RSD of the CuxO/Cu electrode was 29.6%. These results illustrate the protective effect of the CG to stabilize the CuxO nanoparticles.

For long-term stability, the CG/CuxO/Cu electrode was repeatedly used in 0.2 mol/L PBS for one week, the peak current remained 85.7% of the initial value, while for the CuxO/Cu electrode, the surface of the electrode had a severe oxidation and was unable to use for CF detection. As mentioned above, the CG protected CuxO/Cu electrode possesses better reproducibility and stability in alkaline solution for CF detection.

In order to further illustrate the practical application of the CG/CuxO/Cu electrode, the calcium folinate injection sample was detected for analytical assay. Standard addition method and standard curve method were all performed. The labeled concentration of CF of the injection sample is 0.01 g/mL, under a confidence level of 95%, the detection results were 10 mL: (0.108±0.001) g and 10 mL: (0.101±0.001) g by standard addition method and standard curve method, and the RSDs were 4.60% and 4.95% for six measurements, respectively. Therefore, the proposed method is competent for determination of CF in real samples.

Ⅳ. CONCLUSION

In this work, we have successfully developed a simple and sensitive CF electrochemical sensor based on the graphene protected copper oxide nanoparticles. The carboxyl graphene could be electrodeposited on the surface of the copper oxide nanoparticles due to the complexation of the carboxyl groups with the copper ions and the reductive reaction of the double bond of the graphene. The graphene layer is stable, effective, and permitting the inorganic ions passing through, to participate in the electrode reaction. This means the potential ability of the graghene in construction of biosensors. Here, it should be stressed that the positive scan polarization reverse catalytic voltammetry will become a brand new and useful tool for investigation of the electro catalytic reactions, and the experiments showed that PSPRCV could not be replaced by the anodic polarization then cathodic scan method, like cathodic stripping voltammetry.

Ⅴ. ACKNOWLEDGMENTS

This work is supported by the National Basic Research Program of China (No.2013CB933900).

Reference
 [1] E. Schleyer, J. Reinhardt, M. Unterhalt, and W. Hiddemann, J. Chromatogr. B 669 , 319 (1995). DOI:10.1016/0378-4347(95)00118-3 [2] G. L. Duan, L. X. Zheng, J. Chen, W. B. Cheng, and D. Li, Biomed. Chromatogr. 16 , 282 (2002). DOI:10.1002/(ISSN)1099-0801 [3] S. Belz, C. Frickel, C. Wolfrom, H. Nau, and G. Henze, J. Chromatogr. B 661 , 109 (1994). DOI:10.1016/0378-4347(94)00328-9 [4] P. Koufopantelis, S. Georgakakou, M. Kazanis, C. Giaginis, A. Margeli, S. Papargiri, and I. Panderi, J. Chromatogr. B 877 , 3850 (2009). DOI:10.1016/j.jchromb.2009.09.034 [5] F. Süß, V. Harang, C. E. Sanger-van de Griend, and G. K. Scriba, Electrophoresis 25 , 766 (2004). DOI:10.1002/(ISSN)1522-2683 [6] O. van Tellingen, H. R. van Derwoude, J. H. Beijnen, V. J. T. van Beers, and W. J. Nooyen, J. Chromatogr. B, 488 , 379 (1989). DOI:10.1016/S0378-4347(00)82962-X [7] G. Hempel, F. Sczesny, J. Boos, and G. Blaschke, J. Chromatogr. B 718 , 177 (1998). DOI:10.1016/S0378-4347(98)00367-3 [8] K. Liu, X. Dai, D. Zhong, P. Deng, J. Ma, and X. Chen, J. Chromatogr. B 877 , 902 (2009). DOI:10.1016/j.jchromb.2009.02.046 [9] A. Espinosa, - Mansilla, I. D. Meras, M. J. R. Gomez, A. M. de la Pena, and F. Salinas, Talanta 58 , 255 (2002). DOI:10.1016/S0039-9140(02)00243-6 [10] Z. Zhu, F. Wang, F. Wang, and L. Xi, J. Electroanal. Chem. 708 , 13 (2013). DOI:10.1016/j.jelechem.2013.09.004 [11] Y. Yamamoto, H. Nishihara, and K. Aramaki, J. Electrochem. Soc. 140 , 436 (1993). DOI:10.1149/1.2221064 [12] H. Nishihara, M. Itoh, and K. Aramaki, J. Electrochem. Soc. 142 , 1839 (1995). DOI:10.1149/1.2044203 [13] G. Karim, - Nezhad, and P. Seyed Dorraj, Electrochim. Acta 55 , 3414 (2010). DOI:10.1016/j.electacta.2010.01.057 [14] S. G. Wu, Z. X. Zhang, Q. P. Zhao, L. Zhou, and Y. Yao, Chin. J. Chem. Phys. 27 , 600 (2014). DOI:10.1063/1674-0068/27/05/600-606 [15] S. M. Abd El Haleem, and B. G. Ateya, J. Electroanal. Chem. 117 , 309 (1981). DOI:10.1016/S0022-0728(81)80091-5 [16] L. D. Burke, M. J. G. Ahern, and T. G. Ryan, J. Electrochem. Soc. 137 , 553 (1990). DOI:10.1149/1.2086496 [17] W. Z. Le, and Y. Q. Liu, Sens. Actuat. B 141 , 147 (2009). DOI:10.1016/j.snb.2009.05.037 [18] T. R. L. C. Paix, D. Corbo, and M. Bertotti, Anal. Chim. Acta 472 , 123 (2002). DOI:10.1016/S0003-2670(02)00942-X [19] J. M. Marioli, and T. Kuwana, Electrochim. Acta 37 , 1187 (1992). DOI:10.1016/0013-4686(92)85055-P [20] M. Pumera, Electrochem. Commun. 36 , 14 (2013). DOI:10.1016/j.elecom.2013.08.028 [21] S. Watcharotone, D. A. Dikin, S. Stankovich, R. Piner, I. Jung, G. H. B. Dommett, G. Evmentnko, S. Wu, S. F. Chen, C. P. Liu, and S. T. N. A. R. S. Ruoff, Nano Lett. 7 , 1888 (2007). DOI:10.1021/nl070477+ [22] J. Wu, H. Xu, and J. Zhang, Acta Chim. Sin. 72 , 301 (2014). DOI:10.6023/A13090936 [23] F. Yan, Y. Zhang, S. Zhang, J. Zhao, S. Liu, L. He, X. Feng, H. Zhang, and Z. Zhang, Microchim Acta 182 , 855 (2015). DOI:10.1007/s00604-014-1399-y [24] A. A. Balandin, S. Ghosh, W. Bao, I. Calizo, D. Teweldebrhan, and F. M. A. C. N. Lau, Nano Lett. 8 , 902 (2008). DOI:10.1021/nl0731872 [25] P. Sharma, S. K. Tuteja, V. Bhalla, G. Shekhawat, V. P. Dravid, and C. R. Suri, Biosens. Bioelectron 39 , 99 (2013). DOI:10.1016/j.bios.2012.06.061 [26] G. Karim-Nezhad, R. Jafarloo, and P. S. Dorraji, Electrochim. Acta 54 , 5721 (2009). DOI:10.1016/j.electacta.2009.05.019