- Email: [email protected]
ferrocene-branched chitosan composites
Talanta 77 (2008) 366–371
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Amperometric sulfite sensor based on multiwalled carbon nanotubes/ferrocene-branched chitosan composites Hong Zhou, Weiwei Yang, Changqing Sun ∗ College of Chemistry, Jilin University, Changchun 130012, PR China
a r t i c l e
i n f o
Article history: Received 17 April 2008 Received in revised form 23 June 2008 Accepted 23 June 2008 Available online 1 July 2008 Keywords: Carbon nanotubes Ferrocene-branched chitosan Sulfite Amperometric sensor Electrocatalysis
a b s t r a c t A novel amperometric sensor for the determination of sulfite was fabricated based on multiwalled carbon nanotubes (MWCNTs)/ferrocene-branched chitosan (CHIT-Fc) composites-covered glassy carbon electrode (GCE). The electrochemical behavior of the sensor was investigated in detail by cyclic voltammetry. The apparent surface electron transfer rate constant (Ks ) and charge transfer coefficient (˛) of the CHITFc/MWCNTs/GCE were also determined by cyclic voltammetry, which were about 1.93 cm s−1 and 0.42, respectively. The sensor displayed good electrocatalytic activity towards the oxidation of sulfite. The peak potential for the oxidation of sulfite was lowered by at least 330 mV compared with that obtained at CHIT/MWCNTs/GCE. In optimal conditions, linear range spans the concentration of sulfite from 5 M to 1.5 mM and the detection limit was 2.8 M at a signal-to-noise ratio of 3. The proposed method was used for the determination of sulfite in boiler water. In addition, the sensor has good stability and reproducibility. © 2008 Elsevier B.V. All rights reserved.
1. Introduction As is known, sulfite is a typical example of sulfur oxoanions. The main interest in sulfite lies in its reducing properties. Along with ascorbate, they are well established and play an important part in the anti-oxidant defence. Generally, sulfite is widely used as an additive in food and beverages to prevent oxidation and bacterial growth and to control enzymatic reactions during production and storage. Despite these great advantages, the sulfite content in food and beverages should be strictly limited due to its potential toxicity and harmful effects towards hypersensitive people [1]. Therefore, it is important to develop a rapid, reliable and sensitive detection method for determination of sulfite for food and beverages industry in order to control product quality. In the past decade, the reported analytical methods for sulfite determination include mainly spectrophotometry [2,3], chromatography [4,5], electrochemical methods [6–14] and biosensors [15–17]. Carbon nanotubes (CNTs) have been intensively investigated since Iijima’s discovery [18] due to their unique electrical, mechanical and structural properties. As electrode materials, one promising application of carbon nanotubes involves their use in the construction of biosensors and chemical sensors. CNTs represent a new kind of carbon-based material and are superior to other kinds of carbon materials commonly used in electrochemistry, such as glassy car-
∗ Corresponding author. E-mail address: [email protected] (C. Sun). 0039-9140/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2008.06.036
bon, graphite and diamond, mainly in the special structural features and unique electronic properties [19]. Such potential applications would greatly benefit from the ability that the CNTs exhibit a high ability to promote some type of the electron transfer reactions between electroactive species and electrodes, minimize fouling of electrode surfaces, enhance electrocatalytic activity, and facilitate the immobilization of molecules such as enzymes or antibodies on their surface with a view to developing biosensors [20]. Recent electrochemical studies revealed that the unique properties of the CNTs make them very promising in electrochemical application, for example, for protein electrochemistry [21–24], development of novel electrochemical sensors and biosensors [25–29]. Chitosan is a linear -1,4-linked polysaccharide (similar to cellulose) that is obtained by the partial deacetylation of chitin [30]. Because chitosan containing abundant amino groups with pKa 6.3 is soluble in slightly acidic solution due to the protonation and insoluble in solution above pH 6.3 for the deprotonation, it exhibits robust film-forming ability. In addition, chitosan displays nontoxicity, biocompatibility, cheapness and a susceptibility to chemical modification. Because of its desirable properties, chitosan has been widely used as an immobilization matrix for biosensors and bioreactors. Recently, our research group has successfully synthesized ferrocene-branched chitosan derivatives (CHIT-Fc) and reagentless enzyme-based biosensors had been fabricated by the redox polymer [31]. The redox polymers have been used for mediated electron transfer in biosensors since they were reported by Heller’s and Calvo’s groups [32,33]. Recently, Gorski and Schmicltke reported
H. Zhou et al. / Talanta 77 (2008) 366–371
a new electrode system that utilized synergy between carbon nanotubes and redox mediators, and their results suggested that the integration of redox and carbon nanotubes in a polymeric matrix was able to provide a remarkable synergistic augmentation of sensor performance [27,29,34]. In the present work, a novel amperometric sulfite sensor was prepared by using multiwalled carbon nanotubes/ferrocene-branched chitosan composites-covered glassy carbon electrode. The integration of CHIT-Fc and MWCNTs for the development of electrochemical sulfite sensor has not been explored thus far. The electrochemical behavior and electrocatalytic activity towards the oxidation of sulfite for the sensor were studied in detail. Due to the excellent electrocatalytic ability of CHIT-Fc and the unique physiochemical properties of MWCNTs and especially the synergistic augmentation of MWCNTs and CHIT-Fc, the sensor showed very good performance characteristics towards electrocatalytic determination of sulfite. 2. Experimental 2.1. Apparatus Electrochemical measurements were performed with a CHI 660A electrochemical workstation (CH Instruments, USA). Threeelectrode systems were employed in this study. A platinum wire and a saturated calomel electrode (SCE) were used as auxiliary and reference electrode, respectively. All potentials were referred to the latter. CHIT-Fc/MWCNTs composites electrode employed as working electrode was prepared in our laboratory according to the procedure described below. A magnetic Teflon stirrer provided the convective transport during the amperometric measurements. All the experiments were performed at room temperature. 2.2. Reagents Sodium sulfite was obtained from Changchun Reagent Co. Ltd. (Changchun, China). Chitosan with a degree of deacetylation of 92% was purchased from Sanland-chem International Inc. (Xiamen, China). Sodium cyanoborohydride (NaCNBH3 , 98%) and ferrocenecarboxaldehyde (Fc-CHO, 98%) were obtained from Acros and Fluka, respectively. Multiwall carbon nanotubes (MWCNTs, diameter: 10–20 nm, average length: 1–2 m, purity: ≥95%) were purchased from Shenzhen Nanotech. Port. Co. Ltd. (Shenzhen, China). Before use, the MWCNTs were treated with mixed acid according to a method already described [35] and the oxidized MWCNTs (MWCNTsCOOH) were formed. Ferrocene-branched chitosan (CHIT-Fc) was prepared according to the method described in the literature [31]. All other reagents were of analytical grade and used without further purification. All aqueous solutions were prepared with doubly distilled water. The solutions containing sulfite bubbled with ultrapure N2 and kept under the nitrogen atmosphere during the electrochemical experiments.
367
The CHIT-Fc/MWCNTs composites were prepared by mixing above two solutions by sonicating the mixture for 30 min. Then a 6 l aliquot of this solution was cast on the surface of cleaned GC electrode, dried at room temperature for 2 h. The obtained electrode, the CHIT-Fc/MWCNTs/GCE, dipped into 0.1 M (pH 7.0) phosphate buffer solution (PBS) for 5 min, was ready for use. The composition of the layer of the electrode was 3 g CHIT-Fc and 3 g MWCNTs, respectively. The fabricated method of different composition of the layer of electrodes was similar to that above. In the comparing test, fabrication process of the sensor was similar to that of CHIT-Fc/MWCNTs/GCE by substituting MWCNTs with graphite. 3. Results and discussion 3.1. Electrochemical behavior of the CHIT-Fc/MWCNTs/GCE To investigate whether MWCNTs on GCE could provide a remarkable synergistic augmentation of the sensor performance, cyclic voltammetry (CV) was performed with the different kinds of electrodes in 0.1 M PBS (pH 8.0). Fig. 1 shows cyclic voltammograms (CVs) at a CHIT-Fc/GCE and the CHIT-Fc/MWCNTs/GCE, respectively. In the initial comparing experiment, at the GCE modified with CHITFc film alone, a pair of well-defined redox peaks corresponding to the oxidation and reduction of the CHIT-Fc were observed at +0.32 and +0.26 V vs. SCE, respectively (Fig. 1(a)). The formal potential (E◦ ) of +0.29 V was calculated from the average value of the anodic and cathodic peak potentials and the peak-to-peak separation (Ep ) was 60 mV at a scan rate of 0.05 V s−1 . Obviously this is a one-electron redox reaction of CHIT-Fc+ /CHIT-Fc. For the case of CHIT-Fc/MWCNTs composites-coated GCE modified with the same redox polymer, although no change in the shape and peak potentials of CV was observed, yet there was an obvious increase in the redox peak currents (Fig. 1(b)–(d)). This suggests that in the case of the electrode with the CHIT-Fc alone, not all of redox centers were in electrical communication with the electrode surface and that the MWCNTs increased efficiency in mediating electron transfer for the CHIT-Fc/MWCNTs composites-coated GCE. Inset of Fig. 1 shows the relationship between oxidation peak currents and the amounts of MWCNTs in the composites film. As can be seen that the oxidation peak current response increased linearly with the amounts of MWCNTs and maximum oxidation
2.3. Sensor preparation Firstly, a GCE (3 mm-diameter) was polished with emery paper followed by polishing it with alumina (1.0, 0.5, and 0.3 m), and then thoroughly washed with twice-distilled water, sonicated in ethanol, washed again with twice-distilled water and ethanol, and finally dried in nitrogen at room temperature. 0.5 mg CHIT-Fc was dissolved in 0.5 ml 0.1 M acetic acid and 0.5 mg oxidized MWCNTs was dispersed in 0.1 M acetic acid with ultrasonication for 15 min, respectively.
Fig. 1. Cyclic voltammograms of GCE modified with different composite film. (a) 3 g CHIT-Fc; (b) 3 g CHIT-Fc and 1.5 g MWCNTs; (c) 3 g CHIT-Fc and 3 g MWCNTs; (d) 3 g CHIT-Fc and 6 g MWCNTs. Conditions: 0.1 M PBS (pH 8.0); scan rate: 50 mV s−1 . Inset shows the relationship between oxidation peak currents (Ipa ) and the amounts of MWCNTs in the composite film.
368
H. Zhou et al. / Talanta 77 (2008) 366–371
Fig. 2. Cyclic voltammograms of the CHIT-Fc/MWCNTs/GCE in 0.1 M PBS (pH 8.0) at scan rates (from inner to outer) 10, 20, 30, 40, 50, 60, 70, 80, 90 and 100 mV s−1 , respectively. Inset A: plot of Ip vs. . Inset B: Ip vs. 1/2 . Inset C: Epa vs. log .
current was obtained at 3 g of MWCNTs. At higher amounts of MWCNTs, the plot of oxidation peak currents vs. plot of amounts of MWCNTs deviated from linearity. Therefore, 3 g of MWCNTs was chosen as the optimum amount of MWCNTs in the composites for all following electrochemical experiments. Fig. 2 shows the cyclic voltammograms of CHIT-Fc/MWCNTs/ GCE at different scan rates in potential range of 0.00–0.60 V in 0.1 M PBS (pH 8.0). As shown in the inset A of Fig. 2, the oxidation peak currents increased linearly with the scan rate between 10 and 100 mV s−1 as expected for a surface-controlled electrode process. At higher sweep rates, the plot of oxidation peak currents vs. scan rate plot deviated from linearity and the oxidation peak currents became proportional to the square root of the scan rate (inset B of Fig. 2), showing a diffusion controlled process. At higher sweep rates, the oxidation peak potentials (Epa ) were proportional to the logarithm of the scan rate (inset C of Fig. 2) and the slop of the ∂Epa /∂ log was about 140.0 mV. According to Lavirou’s theory [36], when Ep > 200 mV, using the equation Ep = k − 2.303(RT/˛nF) log and one electron transferred for CHITFc+ /CHIT-Fc redox couple, charge transfer coefficient, ˛ = 0.42, was obtained. Introducing this ˛ value in the following equation: log Ks =˛ log(1 − ˛) + (1 − ˛) log ˛− log
RT nF
−
˛(1 − ˛)nFEp 2.303RT
Adding sulfite to the cell produced a dramatic change in the cyclic voltammogram with an increase in anodic current and decrease in cathodic current, which indicated that the electrocatalytic oxidation of sulfite occurred at the modified electrode. The oxidation peak potential of sulfite at the modified electrode was negatively shifted about 330 mV compared with that at CHIT/MWCNTs/GCE. This suggests that CHIT-Fc/MWCNTs composites exhibit high electrocatalytic activity towards sulfite. The effect of buffer solution pH on the electrocatalytic activity of CHIT-Fc/MWCNTs composites towards sulfite was investigated. The experimental results show that the catalytic currents increase and the peak potentials were negatively shifted as pH increase in 2.0–10.0 pH range and maximum growth of catalytic current occurred at pH 8.0, as shown in Fig. 4. The catalytic response increased with pH and reached a platform at pH larger than 8.0. Therefore, an optimum pH for the electrocatalytic system of 8.0 was selected. Fig. 5 shows the cyclic voltammograms of 2 mM sulfite at the CHIT-Fc/MWCNTS/GCE in 0.1 M PBS (pH 8.0) at different scan rates in potential range 0.0–0.6 V. The oxidation peak currents were proportional to the square root of the scan rate (inset (A) of Fig. 5). This result indicates that the reaction system was controlled by
An apparent surface electron transfer rate constant, Ks = 1.93 cm s−1 , was estimated. For the case of the electrode with the CHIT-Fc alone, the obtained Ks was 0.63 cm s−1 by similar method (date not shown here). The above results also show that the integration of CHIT-Fc and MWCNTs can provide a remarkable synergistic augmentation of sensor performance. 3.2. Electrocatalytic oxidation of sulfite at CHIT-Fc/MWCNTs/GCE The electrocatalytic oxidation of sulfite by water-soluble ferrocene derivatives in homogeneous solution has been reported [13]. Here, we take sulfite as an example of sulfur oxoanions and investigated its electrocatalytic oxidation through the CHITFc/MWCNTs/GCE. At CHIT/MWCNTs/GCE a little response was obtained in the range from 0.00 to 0.90 V (vs. SCE) in 0.1 M PBS (pH 8.0) containing 2 mM sulfite, as shown in Fig. 3, curve (a) and (b). However, the catalytic oxidation of sulfite at the CHITFc/MWCNTs/GCE can be seen clearly in Fig. 3, curve (c) and (d).
Fig. 3. Cyclic voltammograms in 0.1 M PBS (pH 8.0) at scan rate of 5 mV s−1 : CHIT/MWCNTs/GCE in buffer solution containing no sulfite (a) and containing 2 mM sulfite (b); CHIT-Fc/MWCNTs/GCE in buffer solution containing no sulfite (c) and containing 2 mM sulfite (d).
H. Zhou et al. / Talanta 77 (2008) 366–371
369
3.3. Amperometric determination of sulfite at the CHIT-Fc/MWCNTs/GCE
Fig. 4. Cyclic voltammograms of the CHIT-Fc/MWCNTs/GCE in the presence of 2 mM sulfite in 0.1 M PBS at various pH: (a) 2, (b) 4 and (c) 6, (d) 8, (e) 10. Scan rate: 5 mV s−1 .
sulfite diffusion. It can also be noted in Fig. 5 that by increasing the scan rate the peak potential for the catalytic oxidation of sulfite shifted to more positive values, suggesting a kinetic limitation in the reaction system. Based on the above results, the catalytic process (EC catalytic mechanism) could be expressed as follows: CHIT-Fc ↔ CHIT-Fc+ + e
2CHIT-Fc+ + SO3 2− + H2 O → 2CHIT-Fc + SO4 2− + 2H+ For information on the rate-determining step, a Tafel plot was obtained from the linear relationship observed for Epa vs. log (inset (B) of Fig. 5) by using the following equation [37]. Ep =
(b log ) 2 + constant
On the basis of the above equation, the slop of Ep vs. log is b/2, where b is the Tafel slop. Thus b = 2 × ∂Ep /∂ log = 139.6 mV. The result is close to that obtained from ferrocenemonocarboxylic acid used as a homogeneous mediator to catalyze the electrooxidation of sulfite [13].
Firstly, in order to optimize the working potential, the dependence of the sulfite response on the applied potential was studied by the amperometry over the potential range from +0.20 to +0.45 V. The current response was measured as a function of applied potential on exposure to 0.1 mM sulfite in a stirring 0.1 M PBS (pH 8.0). At the CHIT-Fc/MWCNTs/GCE, the current response increased with the increase in potential and the optimal value was observed at +0.35 V (data not shown here). Therefore, a potential of +0.35 V was selected for constant potential amperometry to study the amperometric response of the sensor on sulfite. Fig. 6 displays the typical steady-state catalytic current-time response of the CHIT-Fc/MWCNTs/GCE with successive injection of sulfite at an applied potential of +0.35 V. As shown during the successive additions of 0.15 mM (Fig. 6(A), high concentration section) and 0.02 mM (Fig. 6(B), low concentration section) of sulfite, a well-defined response was observed, respectively. And from Fig. 6 it can be seen that the time required to reach 95% of the steady-state current was less than 10 s after the addition of sulfite, which showed that the current response of the sensor was rapid. The insets in Fig. 6(A) and (B) illustrate the calibration curves of the sensor under the optimal experimental conditions. The linear range spanned the concentration of sulfite from 5 M to 1.5 mM. In the linear range, the sensor had a high sensitivity of 13.08 A mM−1 . The detection limit of the sensor was determined to be 2.8 M at a signal-to-noise ratio of 3. Obviously, the analytical performance parameters of the proposed sensor are better than that obtained by water-soluble ferrocene derivatives in homogeneous solution [13]. High sensitivity can be attributed to the synergistic augmentation of MWCNTs and CHIT-Fc towards sulfite response. In the past decade, the detection limit, linear concentration range and sensitivity of other related modified electrodes for sulfite detection have been reported in Table 1. In order to demonstrate the electrocatalytic oxidation of sulfite in a real sample, the determination of sulfite in boiler-water samples was carried out by the standard addition method. The results obtained were compared with those come from a standard iodometric method [38] and were shown to be in good agreement, as shown in Table 2. In addition, interference of coexisting species come from boilerwater samples was studied. The experimental results showed that
Fig. 5. Cyclic voltammograms of the CHIT-Fc/MWCNTs/GCE in 0.1 M PBS (pH 8.0) containing 2 mM sulfite at scan rates (from inner to outer) 10, 20, 30, 40, 50 and 60 mV s−1 , respectively. Inset A: Ipa vs. 1/2 . Inset B: Epa vs. log .
370
H. Zhou et al. / Talanta 77 (2008) 366–371
Table 1 The comparison of the performance of present sensor and others reported in the literatures for sulfite detection No. 1 2 3 4 5 6 7 8 9 10 11 12
Type of film modified electrode
Linear range
NiPCNF/AI CoPCNF/GCE Fe/CCE In situ plated copper modified gold ultramicroelectrode arrays Fc/CPE PB/GCE NiPPIX/GCE NiPCNF/CCEs Sulfite oxidase Sulfite oxidase Sulfite oxidase CHIT-Fc/MWCNTs/GCE
−6
Sensitivity −4
4 × 10 to 2 × 10 M 5 × 10−6 to 1 × 10−4 M 0.73–95.42 mg/l 20–500 M 8.7 × 10−5 to 1.1 × 10−2 M 0.0–4.0 mM ∼9.0 g/ml 2 M–2.0 mM 0.01–1.0 mM 0.04–5.9 mM 0.2–1.8 mM 5.0 M–1.5 mM
Detection limit −6
13.5 nA/M
3 × 10 M 3 × 10−6 M 0.59 mg/l 6 M 5.3 × 10−6 M 80 M 0.15 g/ml 0.5 M
13.08 A/mM
4.0 ppm 0.2 mM 2.8 M
0.35 nA/M 2.18 A/mM
Ref. [10] [13] [14] [15] [16] [17] [18] [20] [21] [22] [23] This work
atmosphere, the current response remained almost unchanged for at least a month. In addition, after 30 days the sensor remained 95% of its initial response to electrocatalytic oxidation of sulfite. The good stability of the CHIT-Fc/MWCNTs/GCE can be attributed primarily to electrostatic interaction between MWCNTs contained negative charges and CHIT-Fc contained positive charges, and robust film-forming ability of CHIT-Fc. The sensor has good repeatability. The relative standard deviation (R.S.D.) is 6.7% for seven determination of 30 M sulfite in 0.1 M PBS (pH 8.0). 4. Conclusions In this work, we developed a novel way to fabricate amperometric sulfite sensor by CHIT-Fc/MWCNTs composites-covered GCE. The CHIT-Fc and MWCNTs composites showed obvious synergistic augmentation of the sensor performance compared with that obtained by CHIT-Fc alone. The sensor showed good electrocatalytic activity for the oxidation of sulfite in 0.1 M PBS (pH 8.0). Because of electrostatic interaction between MWCNTs and CHIT-Fc and robust film-forming ability of the latter, the sensor also exhibited very good reproducibility and stability. The above results suggest that the proposed sensor can be used as amperometric sensor for the determination of sulfite. Acknowledgement This work was supported by the State Key Laboratory for Supramolecular Structure and Materials, Jilin University. References Fig. 6. An amperometric response at the CHIT-Fc/MWCNTs/GCE for successive addition of (A) 0.15 mM and (B) 0.02 mM sulfite. Conditions: 0.1 M PBS (pH 8.0); applied potential: 0.35 V (vs. SCE). Insets show the relationship between catalytic currents and sulfite concentrations.
a 600-fold excess of Ca2+ , Mg2+ , Ba2+, PO4 3− , NO3 − , CO3 2− and Cl− did not interfere in the determination of sulfite. The stability of the CHIT-Fc/MWCNTs/GCE was investigated. In the potential range from 0.00 to 0.60 V at a scan rate of 50 mV s−1 and in pH 8.0 0.1 M PBS, after about 50 scanning cycles the redox currents decreased about 3.0%. When the sensor was stored in Table 2 The results of determination of boiler-water sample Sample number
1 2 3
Concentration of sulfite (mg/l) Proposed method
Iodometric method
10.3 16.3 20.9
9.7 15.8 19.8
[1] E. Dinckaya, M.K. Sezgintürk, E. Akyilmaz, F.N. Ertas, Food Chem. 101 (2007) 1540. [2] I.V. Pulyayeva, N.L. Yegorova, L.P. Experiandova, A.B. Blank, Anal. Chim. Acta 357 (1997) 239. [3] S.S.M. Hassan, M.S.A. Hamza, A.H.K. Mohamed, Anal. Chim. Acta 570 (2006) 232. [4] Y. Zuo, H. Chen, Talanta 59 (2003) 875. [5] G.A. Perfetti, G.W. Diachenko, J. AOAC Int. 86 (2003) 544. [6] M.H. Pournaghi-Azar, M. Hydarpour, H. Dastangoo, Anal. Chim. Acta 497 (2003) 133–141. [7] M.H. Pouranghi-Azar, R.E. Sabzi, Electroanalysis 16 (2004) 860. [8] D.R. Shankaran, K.I. Iimura, T. Kato, Electroanalysis 16 (2004) 556. ˜ [9] O. Ordeig, C.E. Banks, F.J.d. Campo, F.X. Munoz, J. Davis, R.G. Compton, Electroanalysis 18 (2006) 247. [10] J.B. Raoof, R. Ojani, H. Karimi-Maleh, Int. J. Electrochem. Sci. 2 (2007) 257. [11] T. García, E. Casero, E. Lorenzo, F. Pariente, Sens. Actuators B 106 (2005) 803. [12] R. Carballo, V.C. DalÎOrto, A.L. Balbo, I. Rezzano, Sens. Actuators B 88 (2003) 155. [13] Z.N. Gao, J.F. Ma, W.Y. Liu, Appl. Organomet. Chem. 19 (2005) 1149. [14] A. Salimi, K. Abdi, G.R. Khayatiyan, Electrochim. Acta 49 (2004) 413. [15] M. Zhao, D.B. Hibbert, J.J. Gooding, Anal. Chim. Acta 556 (2006) 195. [16] A.K. Abass, J.P. Hart, D. Cowell, Sens. Actuators B 62 (2000) 148. ˝ [17] M.K. Sezginturk, E. Dinckaya, Talanta 65 (2005) 998. [18] S. Iijima, Nature 354 (1991) 56.
H. Zhou et al. / Talanta 77 (2008) 366–371 [19] G.A. Rivas, M.D. Rubianes, M.C. Rodríguez, N.F. Ferreyra, G.L. Luque, M.L. Pedano, S.A. Miscoria, C. Parrado, Talanta 74 (2007) 291. [20] M. Valcárcel, S. Cárdenas, B.M. Simonet, Anal. Chem. 79 (2007) 4788. [21] J.X. Wang, M.X. Li, Z.J. Shi, N.Q. Li, Z.N. Gu, Anal. Chem. 74 (2002) 1993. [22] J.C. Cai, J. Chen, Anal. Biochem. 325 (2004) 285. [23] G. Zhao, L. Zhang, X. Wei, Z. Yang, Electrochem. Commun. 5 (2003) 825. [24] J.A. Guiseppi-Elie, C. Lei, R. Baughman, Nanotechnology 13 (2002) 559. [25] J. Wang, M. Musameh, Anal. Chem. 75 (2003) 2075. [26] L. Qian, X. Yang, Talanta 68 (2006) 721. [27] M. Zhang, W. Gorski, Anal. Chem. 77 (2005) 3960. [28] Y. Liu, J. Lei, H. Ju, Talanta 74 (2008) 965. [29] M. Zhang, W. Gorski, J. Am. Chem. Soc. 127 (2005) 2058.
371
[30] H. Yi, L. Wu, W.E. Bentley, R. Ghodssi, G.W. Rubloff, J.N. Culver, G.F. Payne, Biomacromolecules 6 (2005) 2881. [31] W. Yang, H. Zhou, C. Sun, Macromol. Rapid Commun. 28 (2007) 265. [32] A. Heller, Acc. Chem. Res. 23 (1990) 128. [33] J. Hodak, R. Etchenique, E.J. Calvo, Langmuir 13 (1997) 2708. [34] P.P. Joshi, S.A. Merchant, Y. Wang, D.W. Schmidtke, Anal. Chem. 77 (2005) 3183. [35] M. Zhang, Y. Yan, K. Gong, L. Mao, Z. Guo, Y. Chen, Langmuir 20 (2004) 8781. [36] E. Laviron, J. Electroanal. Chem. 52 (1974) 355. [37] J.A. Harrison, Z.A. Khan, J. Electroanal. Chem. 28 (1970) 131. [38] Chinese Institute of Boiler Water Treatment, Water Quality for Industrial Boilers, Chinese Standard Publishing Company, Beijing, 2004.
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Amperometric sulfite sensor based on multiwalled carbon nanotubes/ferrocene-branched chitosan composites Hong Zhou, Weiwei Yang, Changqing Sun ∗ College of Chemistry, Jilin University, Changchun 130012, PR China
a r t i c l e
i n f o
Article history: Received 17 April 2008 Received in revised form 23 June 2008 Accepted 23 June 2008 Available online 1 July 2008 Keywords: Carbon nanotubes Ferrocene-branched chitosan Sulfite Amperometric sensor Electrocatalysis
a b s t r a c t A novel amperometric sensor for the determination of sulfite was fabricated based on multiwalled carbon nanotubes (MWCNTs)/ferrocene-branched chitosan (CHIT-Fc) composites-covered glassy carbon electrode (GCE). The electrochemical behavior of the sensor was investigated in detail by cyclic voltammetry. The apparent surface electron transfer rate constant (Ks ) and charge transfer coefficient (˛) of the CHITFc/MWCNTs/GCE were also determined by cyclic voltammetry, which were about 1.93 cm s−1 and 0.42, respectively. The sensor displayed good electrocatalytic activity towards the oxidation of sulfite. The peak potential for the oxidation of sulfite was lowered by at least 330 mV compared with that obtained at CHIT/MWCNTs/GCE. In optimal conditions, linear range spans the concentration of sulfite from 5 M to 1.5 mM and the detection limit was 2.8 M at a signal-to-noise ratio of 3. The proposed method was used for the determination of sulfite in boiler water. In addition, the sensor has good stability and reproducibility. © 2008 Elsevier B.V. All rights reserved.
1. Introduction As is known, sulfite is a typical example of sulfur oxoanions. The main interest in sulfite lies in its reducing properties. Along with ascorbate, they are well established and play an important part in the anti-oxidant defence. Generally, sulfite is widely used as an additive in food and beverages to prevent oxidation and bacterial growth and to control enzymatic reactions during production and storage. Despite these great advantages, the sulfite content in food and beverages should be strictly limited due to its potential toxicity and harmful effects towards hypersensitive people [1]. Therefore, it is important to develop a rapid, reliable and sensitive detection method for determination of sulfite for food and beverages industry in order to control product quality. In the past decade, the reported analytical methods for sulfite determination include mainly spectrophotometry [2,3], chromatography [4,5], electrochemical methods [6–14] and biosensors [15–17]. Carbon nanotubes (CNTs) have been intensively investigated since Iijima’s discovery [18] due to their unique electrical, mechanical and structural properties. As electrode materials, one promising application of carbon nanotubes involves their use in the construction of biosensors and chemical sensors. CNTs represent a new kind of carbon-based material and are superior to other kinds of carbon materials commonly used in electrochemistry, such as glassy car-
∗ Corresponding author. E-mail address: [email protected] (C. Sun). 0039-9140/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2008.06.036
bon, graphite and diamond, mainly in the special structural features and unique electronic properties [19]. Such potential applications would greatly benefit from the ability that the CNTs exhibit a high ability to promote some type of the electron transfer reactions between electroactive species and electrodes, minimize fouling of electrode surfaces, enhance electrocatalytic activity, and facilitate the immobilization of molecules such as enzymes or antibodies on their surface with a view to developing biosensors [20]. Recent electrochemical studies revealed that the unique properties of the CNTs make them very promising in electrochemical application, for example, for protein electrochemistry [21–24], development of novel electrochemical sensors and biosensors [25–29]. Chitosan is a linear -1,4-linked polysaccharide (similar to cellulose) that is obtained by the partial deacetylation of chitin [30]. Because chitosan containing abundant amino groups with pKa 6.3 is soluble in slightly acidic solution due to the protonation and insoluble in solution above pH 6.3 for the deprotonation, it exhibits robust film-forming ability. In addition, chitosan displays nontoxicity, biocompatibility, cheapness and a susceptibility to chemical modification. Because of its desirable properties, chitosan has been widely used as an immobilization matrix for biosensors and bioreactors. Recently, our research group has successfully synthesized ferrocene-branched chitosan derivatives (CHIT-Fc) and reagentless enzyme-based biosensors had been fabricated by the redox polymer [31]. The redox polymers have been used for mediated electron transfer in biosensors since they were reported by Heller’s and Calvo’s groups [32,33]. Recently, Gorski and Schmicltke reported
H. Zhou et al. / Talanta 77 (2008) 366–371
a new electrode system that utilized synergy between carbon nanotubes and redox mediators, and their results suggested that the integration of redox and carbon nanotubes in a polymeric matrix was able to provide a remarkable synergistic augmentation of sensor performance [27,29,34]. In the present work, a novel amperometric sulfite sensor was prepared by using multiwalled carbon nanotubes/ferrocene-branched chitosan composites-covered glassy carbon electrode. The integration of CHIT-Fc and MWCNTs for the development of electrochemical sulfite sensor has not been explored thus far. The electrochemical behavior and electrocatalytic activity towards the oxidation of sulfite for the sensor were studied in detail. Due to the excellent electrocatalytic ability of CHIT-Fc and the unique physiochemical properties of MWCNTs and especially the synergistic augmentation of MWCNTs and CHIT-Fc, the sensor showed very good performance characteristics towards electrocatalytic determination of sulfite. 2. Experimental 2.1. Apparatus Electrochemical measurements were performed with a CHI 660A electrochemical workstation (CH Instruments, USA). Threeelectrode systems were employed in this study. A platinum wire and a saturated calomel electrode (SCE) were used as auxiliary and reference electrode, respectively. All potentials were referred to the latter. CHIT-Fc/MWCNTs composites electrode employed as working electrode was prepared in our laboratory according to the procedure described below. A magnetic Teflon stirrer provided the convective transport during the amperometric measurements. All the experiments were performed at room temperature. 2.2. Reagents Sodium sulfite was obtained from Changchun Reagent Co. Ltd. (Changchun, China). Chitosan with a degree of deacetylation of 92% was purchased from Sanland-chem International Inc. (Xiamen, China). Sodium cyanoborohydride (NaCNBH3 , 98%) and ferrocenecarboxaldehyde (Fc-CHO, 98%) were obtained from Acros and Fluka, respectively. Multiwall carbon nanotubes (MWCNTs, diameter: 10–20 nm, average length: 1–2 m, purity: ≥95%) were purchased from Shenzhen Nanotech. Port. Co. Ltd. (Shenzhen, China). Before use, the MWCNTs were treated with mixed acid according to a method already described [35] and the oxidized MWCNTs (MWCNTsCOOH) were formed. Ferrocene-branched chitosan (CHIT-Fc) was prepared according to the method described in the literature [31]. All other reagents were of analytical grade and used without further purification. All aqueous solutions were prepared with doubly distilled water. The solutions containing sulfite bubbled with ultrapure N2 and kept under the nitrogen atmosphere during the electrochemical experiments.
367
The CHIT-Fc/MWCNTs composites were prepared by mixing above two solutions by sonicating the mixture for 30 min. Then a 6 l aliquot of this solution was cast on the surface of cleaned GC electrode, dried at room temperature for 2 h. The obtained electrode, the CHIT-Fc/MWCNTs/GCE, dipped into 0.1 M (pH 7.0) phosphate buffer solution (PBS) for 5 min, was ready for use. The composition of the layer of the electrode was 3 g CHIT-Fc and 3 g MWCNTs, respectively. The fabricated method of different composition of the layer of electrodes was similar to that above. In the comparing test, fabrication process of the sensor was similar to that of CHIT-Fc/MWCNTs/GCE by substituting MWCNTs with graphite. 3. Results and discussion 3.1. Electrochemical behavior of the CHIT-Fc/MWCNTs/GCE To investigate whether MWCNTs on GCE could provide a remarkable synergistic augmentation of the sensor performance, cyclic voltammetry (CV) was performed with the different kinds of electrodes in 0.1 M PBS (pH 8.0). Fig. 1 shows cyclic voltammograms (CVs) at a CHIT-Fc/GCE and the CHIT-Fc/MWCNTs/GCE, respectively. In the initial comparing experiment, at the GCE modified with CHITFc film alone, a pair of well-defined redox peaks corresponding to the oxidation and reduction of the CHIT-Fc were observed at +0.32 and +0.26 V vs. SCE, respectively (Fig. 1(a)). The formal potential (E◦ ) of +0.29 V was calculated from the average value of the anodic and cathodic peak potentials and the peak-to-peak separation (Ep ) was 60 mV at a scan rate of 0.05 V s−1 . Obviously this is a one-electron redox reaction of CHIT-Fc+ /CHIT-Fc. For the case of CHIT-Fc/MWCNTs composites-coated GCE modified with the same redox polymer, although no change in the shape and peak potentials of CV was observed, yet there was an obvious increase in the redox peak currents (Fig. 1(b)–(d)). This suggests that in the case of the electrode with the CHIT-Fc alone, not all of redox centers were in electrical communication with the electrode surface and that the MWCNTs increased efficiency in mediating electron transfer for the CHIT-Fc/MWCNTs composites-coated GCE. Inset of Fig. 1 shows the relationship between oxidation peak currents and the amounts of MWCNTs in the composites film. As can be seen that the oxidation peak current response increased linearly with the amounts of MWCNTs and maximum oxidation
2.3. Sensor preparation Firstly, a GCE (3 mm-diameter) was polished with emery paper followed by polishing it with alumina (1.0, 0.5, and 0.3 m), and then thoroughly washed with twice-distilled water, sonicated in ethanol, washed again with twice-distilled water and ethanol, and finally dried in nitrogen at room temperature. 0.5 mg CHIT-Fc was dissolved in 0.5 ml 0.1 M acetic acid and 0.5 mg oxidized MWCNTs was dispersed in 0.1 M acetic acid with ultrasonication for 15 min, respectively.
Fig. 1. Cyclic voltammograms of GCE modified with different composite film. (a) 3 g CHIT-Fc; (b) 3 g CHIT-Fc and 1.5 g MWCNTs; (c) 3 g CHIT-Fc and 3 g MWCNTs; (d) 3 g CHIT-Fc and 6 g MWCNTs. Conditions: 0.1 M PBS (pH 8.0); scan rate: 50 mV s−1 . Inset shows the relationship between oxidation peak currents (Ipa ) and the amounts of MWCNTs in the composite film.
368
H. Zhou et al. / Talanta 77 (2008) 366–371
Fig. 2. Cyclic voltammograms of the CHIT-Fc/MWCNTs/GCE in 0.1 M PBS (pH 8.0) at scan rates (from inner to outer) 10, 20, 30, 40, 50, 60, 70, 80, 90 and 100 mV s−1 , respectively. Inset A: plot of Ip vs. . Inset B: Ip vs. 1/2 . Inset C: Epa vs. log .
current was obtained at 3 g of MWCNTs. At higher amounts of MWCNTs, the plot of oxidation peak currents vs. plot of amounts of MWCNTs deviated from linearity. Therefore, 3 g of MWCNTs was chosen as the optimum amount of MWCNTs in the composites for all following electrochemical experiments. Fig. 2 shows the cyclic voltammograms of CHIT-Fc/MWCNTs/ GCE at different scan rates in potential range of 0.00–0.60 V in 0.1 M PBS (pH 8.0). As shown in the inset A of Fig. 2, the oxidation peak currents increased linearly with the scan rate between 10 and 100 mV s−1 as expected for a surface-controlled electrode process. At higher sweep rates, the plot of oxidation peak currents vs. scan rate plot deviated from linearity and the oxidation peak currents became proportional to the square root of the scan rate (inset B of Fig. 2), showing a diffusion controlled process. At higher sweep rates, the oxidation peak potentials (Epa ) were proportional to the logarithm of the scan rate (inset C of Fig. 2) and the slop of the ∂Epa /∂ log was about 140.0 mV. According to Lavirou’s theory [36], when Ep > 200 mV, using the equation Ep = k − 2.303(RT/˛nF) log and one electron transferred for CHITFc+ /CHIT-Fc redox couple, charge transfer coefficient, ˛ = 0.42, was obtained. Introducing this ˛ value in the following equation: log Ks =˛ log(1 − ˛) + (1 − ˛) log ˛− log
RT nF
−
˛(1 − ˛)nFEp 2.303RT
Adding sulfite to the cell produced a dramatic change in the cyclic voltammogram with an increase in anodic current and decrease in cathodic current, which indicated that the electrocatalytic oxidation of sulfite occurred at the modified electrode. The oxidation peak potential of sulfite at the modified electrode was negatively shifted about 330 mV compared with that at CHIT/MWCNTs/GCE. This suggests that CHIT-Fc/MWCNTs composites exhibit high electrocatalytic activity towards sulfite. The effect of buffer solution pH on the electrocatalytic activity of CHIT-Fc/MWCNTs composites towards sulfite was investigated. The experimental results show that the catalytic currents increase and the peak potentials were negatively shifted as pH increase in 2.0–10.0 pH range and maximum growth of catalytic current occurred at pH 8.0, as shown in Fig. 4. The catalytic response increased with pH and reached a platform at pH larger than 8.0. Therefore, an optimum pH for the electrocatalytic system of 8.0 was selected. Fig. 5 shows the cyclic voltammograms of 2 mM sulfite at the CHIT-Fc/MWCNTS/GCE in 0.1 M PBS (pH 8.0) at different scan rates in potential range 0.0–0.6 V. The oxidation peak currents were proportional to the square root of the scan rate (inset (A) of Fig. 5). This result indicates that the reaction system was controlled by
An apparent surface electron transfer rate constant, Ks = 1.93 cm s−1 , was estimated. For the case of the electrode with the CHIT-Fc alone, the obtained Ks was 0.63 cm s−1 by similar method (date not shown here). The above results also show that the integration of CHIT-Fc and MWCNTs can provide a remarkable synergistic augmentation of sensor performance. 3.2. Electrocatalytic oxidation of sulfite at CHIT-Fc/MWCNTs/GCE The electrocatalytic oxidation of sulfite by water-soluble ferrocene derivatives in homogeneous solution has been reported [13]. Here, we take sulfite as an example of sulfur oxoanions and investigated its electrocatalytic oxidation through the CHITFc/MWCNTs/GCE. At CHIT/MWCNTs/GCE a little response was obtained in the range from 0.00 to 0.90 V (vs. SCE) in 0.1 M PBS (pH 8.0) containing 2 mM sulfite, as shown in Fig. 3, curve (a) and (b). However, the catalytic oxidation of sulfite at the CHITFc/MWCNTs/GCE can be seen clearly in Fig. 3, curve (c) and (d).
Fig. 3. Cyclic voltammograms in 0.1 M PBS (pH 8.0) at scan rate of 5 mV s−1 : CHIT/MWCNTs/GCE in buffer solution containing no sulfite (a) and containing 2 mM sulfite (b); CHIT-Fc/MWCNTs/GCE in buffer solution containing no sulfite (c) and containing 2 mM sulfite (d).
H. Zhou et al. / Talanta 77 (2008) 366–371
369
3.3. Amperometric determination of sulfite at the CHIT-Fc/MWCNTs/GCE
Fig. 4. Cyclic voltammograms of the CHIT-Fc/MWCNTs/GCE in the presence of 2 mM sulfite in 0.1 M PBS at various pH: (a) 2, (b) 4 and (c) 6, (d) 8, (e) 10. Scan rate: 5 mV s−1 .
sulfite diffusion. It can also be noted in Fig. 5 that by increasing the scan rate the peak potential for the catalytic oxidation of sulfite shifted to more positive values, suggesting a kinetic limitation in the reaction system. Based on the above results, the catalytic process (EC catalytic mechanism) could be expressed as follows: CHIT-Fc ↔ CHIT-Fc+ + e
2CHIT-Fc+ + SO3 2− + H2 O → 2CHIT-Fc + SO4 2− + 2H+ For information on the rate-determining step, a Tafel plot was obtained from the linear relationship observed for Epa vs. log (inset (B) of Fig. 5) by using the following equation [37]. Ep =
(b log ) 2 + constant
On the basis of the above equation, the slop of Ep vs. log is b/2, where b is the Tafel slop. Thus b = 2 × ∂Ep /∂ log = 139.6 mV. The result is close to that obtained from ferrocenemonocarboxylic acid used as a homogeneous mediator to catalyze the electrooxidation of sulfite [13].
Firstly, in order to optimize the working potential, the dependence of the sulfite response on the applied potential was studied by the amperometry over the potential range from +0.20 to +0.45 V. The current response was measured as a function of applied potential on exposure to 0.1 mM sulfite in a stirring 0.1 M PBS (pH 8.0). At the CHIT-Fc/MWCNTs/GCE, the current response increased with the increase in potential and the optimal value was observed at +0.35 V (data not shown here). Therefore, a potential of +0.35 V was selected for constant potential amperometry to study the amperometric response of the sensor on sulfite. Fig. 6 displays the typical steady-state catalytic current-time response of the CHIT-Fc/MWCNTs/GCE with successive injection of sulfite at an applied potential of +0.35 V. As shown during the successive additions of 0.15 mM (Fig. 6(A), high concentration section) and 0.02 mM (Fig. 6(B), low concentration section) of sulfite, a well-defined response was observed, respectively. And from Fig. 6 it can be seen that the time required to reach 95% of the steady-state current was less than 10 s after the addition of sulfite, which showed that the current response of the sensor was rapid. The insets in Fig. 6(A) and (B) illustrate the calibration curves of the sensor under the optimal experimental conditions. The linear range spanned the concentration of sulfite from 5 M to 1.5 mM. In the linear range, the sensor had a high sensitivity of 13.08 A mM−1 . The detection limit of the sensor was determined to be 2.8 M at a signal-to-noise ratio of 3. Obviously, the analytical performance parameters of the proposed sensor are better than that obtained by water-soluble ferrocene derivatives in homogeneous solution [13]. High sensitivity can be attributed to the synergistic augmentation of MWCNTs and CHIT-Fc towards sulfite response. In the past decade, the detection limit, linear concentration range and sensitivity of other related modified electrodes for sulfite detection have been reported in Table 1. In order to demonstrate the electrocatalytic oxidation of sulfite in a real sample, the determination of sulfite in boiler-water samples was carried out by the standard addition method. The results obtained were compared with those come from a standard iodometric method [38] and were shown to be in good agreement, as shown in Table 2. In addition, interference of coexisting species come from boilerwater samples was studied. The experimental results showed that
Fig. 5. Cyclic voltammograms of the CHIT-Fc/MWCNTs/GCE in 0.1 M PBS (pH 8.0) containing 2 mM sulfite at scan rates (from inner to outer) 10, 20, 30, 40, 50 and 60 mV s−1 , respectively. Inset A: Ipa vs. 1/2 . Inset B: Epa vs. log .
370
H. Zhou et al. / Talanta 77 (2008) 366–371
Table 1 The comparison of the performance of present sensor and others reported in the literatures for sulfite detection No. 1 2 3 4 5 6 7 8 9 10 11 12
Type of film modified electrode
Linear range
NiPCNF/AI CoPCNF/GCE Fe/CCE In situ plated copper modified gold ultramicroelectrode arrays Fc/CPE PB/GCE NiPPIX/GCE NiPCNF/CCEs Sulfite oxidase Sulfite oxidase Sulfite oxidase CHIT-Fc/MWCNTs/GCE
−6
Sensitivity −4
4 × 10 to 2 × 10 M 5 × 10−6 to 1 × 10−4 M 0.73–95.42 mg/l 20–500 M 8.7 × 10−5 to 1.1 × 10−2 M 0.0–4.0 mM ∼9.0 g/ml 2 M–2.0 mM 0.01–1.0 mM 0.04–5.9 mM 0.2–1.8 mM 5.0 M–1.5 mM
Detection limit −6
13.5 nA/M
3 × 10 M 3 × 10−6 M 0.59 mg/l 6 M 5.3 × 10−6 M 80 M 0.15 g/ml 0.5 M
13.08 A/mM
4.0 ppm 0.2 mM 2.8 M
0.35 nA/M 2.18 A/mM
Ref. [10] [13] [14] [15] [16] [17] [18] [20] [21] [22] [23] This work
atmosphere, the current response remained almost unchanged for at least a month. In addition, after 30 days the sensor remained 95% of its initial response to electrocatalytic oxidation of sulfite. The good stability of the CHIT-Fc/MWCNTs/GCE can be attributed primarily to electrostatic interaction between MWCNTs contained negative charges and CHIT-Fc contained positive charges, and robust film-forming ability of CHIT-Fc. The sensor has good repeatability. The relative standard deviation (R.S.D.) is 6.7% for seven determination of 30 M sulfite in 0.1 M PBS (pH 8.0). 4. Conclusions In this work, we developed a novel way to fabricate amperometric sulfite sensor by CHIT-Fc/MWCNTs composites-covered GCE. The CHIT-Fc and MWCNTs composites showed obvious synergistic augmentation of the sensor performance compared with that obtained by CHIT-Fc alone. The sensor showed good electrocatalytic activity for the oxidation of sulfite in 0.1 M PBS (pH 8.0). Because of electrostatic interaction between MWCNTs and CHIT-Fc and robust film-forming ability of the latter, the sensor also exhibited very good reproducibility and stability. The above results suggest that the proposed sensor can be used as amperometric sensor for the determination of sulfite. Acknowledgement This work was supported by the State Key Laboratory for Supramolecular Structure and Materials, Jilin University. References Fig. 6. An amperometric response at the CHIT-Fc/MWCNTs/GCE for successive addition of (A) 0.15 mM and (B) 0.02 mM sulfite. Conditions: 0.1 M PBS (pH 8.0); applied potential: 0.35 V (vs. SCE). Insets show the relationship between catalytic currents and sulfite concentrations.
a 600-fold excess of Ca2+ , Mg2+ , Ba2+, PO4 3− , NO3 − , CO3 2− and Cl− did not interfere in the determination of sulfite. The stability of the CHIT-Fc/MWCNTs/GCE was investigated. In the potential range from 0.00 to 0.60 V at a scan rate of 50 mV s−1 and in pH 8.0 0.1 M PBS, after about 50 scanning cycles the redox currents decreased about 3.0%. When the sensor was stored in Table 2 The results of determination of boiler-water sample Sample number
1 2 3
Concentration of sulfite (mg/l) Proposed method
Iodometric method
10.3 16.3 20.9
9.7 15.8 19.8
[1] E. Dinckaya, M.K. Sezgintürk, E. Akyilmaz, F.N. Ertas, Food Chem. 101 (2007) 1540. [2] I.V. Pulyayeva, N.L. Yegorova, L.P. Experiandova, A.B. Blank, Anal. Chim. Acta 357 (1997) 239. [3] S.S.M. Hassan, M.S.A. Hamza, A.H.K. Mohamed, Anal. Chim. Acta 570 (2006) 232. [4] Y. Zuo, H. Chen, Talanta 59 (2003) 875. [5] G.A. Perfetti, G.W. Diachenko, J. AOAC Int. 86 (2003) 544. [6] M.H. Pournaghi-Azar, M. Hydarpour, H. Dastangoo, Anal. Chim. Acta 497 (2003) 133–141. [7] M.H. Pouranghi-Azar, R.E. Sabzi, Electroanalysis 16 (2004) 860. [8] D.R. Shankaran, K.I. Iimura, T. Kato, Electroanalysis 16 (2004) 556. ˜ [9] O. Ordeig, C.E. Banks, F.J.d. Campo, F.X. Munoz, J. Davis, R.G. Compton, Electroanalysis 18 (2006) 247. [10] J.B. Raoof, R. Ojani, H. Karimi-Maleh, Int. J. Electrochem. Sci. 2 (2007) 257. [11] T. García, E. Casero, E. Lorenzo, F. Pariente, Sens. Actuators B 106 (2005) 803. [12] R. Carballo, V.C. DalÎOrto, A.L. Balbo, I. Rezzano, Sens. Actuators B 88 (2003) 155. [13] Z.N. Gao, J.F. Ma, W.Y. Liu, Appl. Organomet. Chem. 19 (2005) 1149. [14] A. Salimi, K. Abdi, G.R. Khayatiyan, Electrochim. Acta 49 (2004) 413. [15] M. Zhao, D.B. Hibbert, J.J. Gooding, Anal. Chim. Acta 556 (2006) 195. [16] A.K. Abass, J.P. Hart, D. Cowell, Sens. Actuators B 62 (2000) 148. ˝ [17] M.K. Sezginturk, E. Dinckaya, Talanta 65 (2005) 998. [18] S. Iijima, Nature 354 (1991) 56.
H. Zhou et al. / Talanta 77 (2008) 366–371 [19] G.A. Rivas, M.D. Rubianes, M.C. Rodríguez, N.F. Ferreyra, G.L. Luque, M.L. Pedano, S.A. Miscoria, C. Parrado, Talanta 74 (2007) 291. [20] M. Valcárcel, S. Cárdenas, B.M. Simonet, Anal. Chem. 79 (2007) 4788. [21] J.X. Wang, M.X. Li, Z.J. Shi, N.Q. Li, Z.N. Gu, Anal. Chem. 74 (2002) 1993. [22] J.C. Cai, J. Chen, Anal. Biochem. 325 (2004) 285. [23] G. Zhao, L. Zhang, X. Wei, Z. Yang, Electrochem. Commun. 5 (2003) 825. [24] J.A. Guiseppi-Elie, C. Lei, R. Baughman, Nanotechnology 13 (2002) 559. [25] J. Wang, M. Musameh, Anal. Chem. 75 (2003) 2075. [26] L. Qian, X. Yang, Talanta 68 (2006) 721. [27] M. Zhang, W. Gorski, Anal. Chem. 77 (2005) 3960. [28] Y. Liu, J. Lei, H. Ju, Talanta 74 (2008) 965. [29] M. Zhang, W. Gorski, J. Am. Chem. Soc. 127 (2005) 2058.
371
[30] H. Yi, L. Wu, W.E. Bentley, R. Ghodssi, G.W. Rubloff, J.N. Culver, G.F. Payne, Biomacromolecules 6 (2005) 2881. [31] W. Yang, H. Zhou, C. Sun, Macromol. Rapid Commun. 28 (2007) 265. [32] A. Heller, Acc. Chem. Res. 23 (1990) 128. [33] J. Hodak, R. Etchenique, E.J. Calvo, Langmuir 13 (1997) 2708. [34] P.P. Joshi, S.A. Merchant, Y. Wang, D.W. Schmidtke, Anal. Chem. 77 (2005) 3183. [35] M. Zhang, Y. Yan, K. Gong, L. Mao, Z. Guo, Y. Chen, Langmuir 20 (2004) 8781. [36] E. Laviron, J. Electroanal. Chem. 52 (1974) 355. [37] J.A. Harrison, Z.A. Khan, J. Electroanal. Chem. 28 (1970) 131. [38] Chinese Institute of Boiler Water Treatment, Water Quality for Industrial Boilers, Chinese Standard Publishing Company, Beijing, 2004.