- Email: [email protected]
organically modified silicate films
Analytica Chimica Acta 594 (2007) 169–174
Tris(2,2-bipyridyl)ruthenium(II) electrochemiluminescence sensor based on carbon nanotube/organically modified silicate films Ying Tao a , Zhi-Jie Lin a , Xiao-Mei Chen a , Xi Chen a,b,∗ , Xiao-Ru Wang a a
Department of Chemistry and Key Laboratory of Analytical Sciences of the Ministry of Education, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005, China b State Key Laboratory of Marine Environmental Science, Xiamen University, Xiamen 361005, China Received 10 March 2007; received in revised form 8 May 2007; accepted 15 May 2007 Available online 21 May 2007
Abstract In this paper, a novel electrochemiluminescence (ECL) sensor was constructed to determine herring sperm (HS) double-stranded (ds) DNA. Tetramethoxysilane and dimethyldimethoxysilane were selected as co-precursors to form an organically modified silicate (ORMOSIL) film for the immobilization of multiwall carbon nanotubes (MWNTs) wrapped by poly(p-styrenesulfonate) (PSS), and then Tris(2,2 -bipyridyl)ruthenium(II) (Ru(bpy)3 2+ ) was successfully immobilized on a glassy carbon electrode via ion-association. PSS was employed to increase the conductivity of the ORMOSIL film and disperse the cut MWNTs, which were cut and shortened in a mixture of concentrated sulfuric and nitric acids, in the film. It was found that MWNTs could adsorb Ru(bpy)3 2+ and acted as conducting pathways to connect Ru(bpy)3 2+ sites to the electrode. MWNTs also played a key role as materials for the mechanical and thermal properties. The ECL performance of this modified electrode was evaluated in a flow injection analysis (FIA) system, and the detection limit (S/N = 3) for HS ds-DNA was 2.0 × 10−7 g mL−1 with a linear range from 1.34 × 10−6 to 6.67 × 10−4 g mL−1 (R2 = 0.9876). In addition, the ECL sensor presented excellent characteristics in terms of stability, reproducibility and application life. © 2007 Elsevier B.V. All rights reserved. Keywords: Electrochemiluminescence sensor; Carbon nanotubes; Tris(2,2 -bipyridyl)ruthenium(II); Nucleic acid
1. Introduction Electrochemiluminescence (ECL) is a well-known detection method that provides high sensitivity and selectivity through the generation of an optical signal triggered by an electrochemical reaction [1,2]. This detection method has been utilized in a number of bioanalytical arenas, and commercial systems have been developed which use ECL to detect many clinically important analytes, including application in immunoassays and DNA analyses by employing ECL-active species as labels of biological molecules [3]. Sensitive and selective detection of DNA is central to clinical tests, pathogen detection, and other methods utilizing the polymerase chain reaction, to genetic disease screening based on oligonucleotide hybridization, and to molecular genotoxicity studies [4,5]. ECL using
∗
Corresponding author. Tel.: +86 592 2184530; fax: +86 592 218 6401. E-mail address: [email protected] (X. Chen).
0003-2670/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.aca.2007.05.025
Tris(2,2 -bipyridyl)ruthenium(II) (Ru(bpy)3 2+ ) could become such a detection method. It provides the necessary sensitivity and low background when utilized in a competitive hybridization assay format that would circumvent labeling of the sample DNA [6–8]. The reaction mechanism which has been generally accepted whereby ECL is generated by electrochemical catalytic pathways between Ru(bpy)3 2+ and oxidized guanine bases in DNA is shown in Scheme 1 [9,10]. Initial oxidation at sufficiently positive potentials gives the RuIII oxidant (Eq. (1)), which reacts with guanines (G) in DNA to give the guanine radical (Eq. (2)) [11,12]. This radical may produce RuII * sites (Eq. (3)), representing the excited state complex, by directly reducing the RuIII sites. G2ox in Eq. (3) represents a guanine oxidized by two electrons, a reaction observed [13] in ss-DNA oxidized by dissolved Ru(bpy)3 3+ . In general, the conventional flow injection analytical method based on Ru(bpy)3 2+ -ECL, requiring Ru(bpy)3 2+ as a component of the mobile phase has a high cost. ECL was generated
170
Y. Tao et al. / Analytica Chimica Acta 594 (2007) 169–174
2. Experimental 2.1. Materials
Scheme 1. Possible pathways for the ECL reaction between Ru(bpy)3 2+ and DNA.
under an oxidation-reduction process of Ru(bpy)3 2+ [13], which revealed that Ru(bpy)3 2+ could be electrochemically regenerated during the ECL reaction. As a result, to develop regenerable ECL-based sensors and detection devices, and to simplify the detection system, Ru(bpy)3 2+ -immobilization approaches have been extensively studied [14–26]. For example, Ru(bpy)3 2+ has been immobilized in Langmuir–Blodgget films [15], selfassembled films [16,17], carbon paste[18,19], polymer films [14,20–22], and sol–gel composites [23–26]. The effort on the immobilization approaches has resulted in rapid development of Ru(bpy)3 2+ ECL sensors and detection devices and is very important for further development of new immobilization approaches. In our laboratory, we developed a Ru(bpy)3 2+ –ORMOSILfilm modified electrode and applied it as an ECL sensor for the determination of methamphetamine [26], but some unsatisfactory points, such as the leaching of Ru(bpy)3 2+ and the weak mechanical strength of the film, still remained. Singlewall (SWNTs) and multiwall (MWNTs) carbon nanotubes have attracted extensive attention owing to their properties of high mechanical strength [27], remarkable electrical conductivity [28] and highly pi-conjunctive and hydrophobic side walls [29,30]. In ECL sensor construction, these characteristics of carbon nanotubes could be applied to improve the performance of the modified electrode, since carbon nanotubes acted as a film-skeleton to increase the mechanical strength and conductivity of the ORMOSIL film [31–33]. Their highly pi-conjunctive and hydrophobic sidewalls enabled carbon nanotubes to be a potential support for Ru(bpy)3 2+ by pi pi bond hydrophobic interactions. Although carbon nanotubes easily aggregated in the ORMOSIL film preparation, fortunately the problem could be avoided based on the approaches reported [34]. In this paper, we report a novel ECL sensor approach using poly(p-styrenesulfonate) (PSS) wrapped MWNTs [34] to fabricate an MWNTs–PSS–ORMOSIL film on a glassy carbon electrode (GCE), and then Ru(bpy)3 2+ immobilized in the film followed by an ion-exchange process. Ru(bpy)3 2+ was easily concentrated in the film due to the strong interaction between Ru(bpy)3 2+ and the MWNTs, resulting in a higher concentration of Ru(bpy)3 2+ in the ORMOSIL film, and this give the modified electrode better stability for longtime usage. The addition of MWNTs improved further the mechanical strength of the film. The significant fact was that Ru(bpy)3 2+ retained its excellent redox ability in the film and showed good ECL responses to herring sperm (HS) DNA. The functioning of film components was studied further by fabricating MWNTs–ORMOSIL and PSS–ORMOSIL films on GCEs.
Ru(bpy)3 Cl2 ·6H2 O, Si(OCH3 )4 (TMOS) and (CH3 )2 Si(OCH3 )2 (DiMe–DiMOS) were obtained from Fluka AG (Buchs, Switzerland) and used as received. Poly(pstyrenesulfonate) (PSS, MW ∼ 70,000) and HS double-stranded (ds) DNA were purchased from Aldrich (Milwaukee, WI, USA). HS ds-DNA stock solution was prepared by dissolving HS dsDNA in a phosphate buffer solution stored at 4 ◦ C, and could be in use for 2 days. The concentration of the stock solution was confirmed by its absorbance measured at 260 nm using a biophotometer (Eppendorf AG, German). The MWNTs with a ∼90% purity and typical lengths of 5–15 m, were obtained from Shenzhen Nanotech. Port. Co. Ltd. All other reagents were of analytical reagent grade and were used without further purification. Distilled water was purified with a Millipore system (Millipore Co., USA), and was used in all aqueous solution preparations and in washing. 2.2. Instrumentation SEM (Oxford Co., UK) was used to study the surface state of the ORMOSIL film under an accelerating voltage at 20 kV. IR spectra were obtained with a FT-IR7400SX spectrometer (Nicolet, USA). Cyclic voltammetry (CV) and electronic impedance spectra (EIS) measurements were performed with a CHI 660B Electrochemical Analyzer (CHI Co. Shanghai, China). A conventional three-electrode system including a modified electrode as the working electrode, a platinum auxiliary electrode and a saturated calomel reference electrode was applied. ECL experiments were carried out on a CHI 660B Electrochemical Analyzer and an IFFM-D FIA Luminescence Analyzer (Ruimai Co., China) at room temperature. The flow system was also equipped with a flow cell and a Rheodyne 7125 sample injector (CA, USA, 50 L). The main body of the thin layer electrolysis flow cell was constructed from two pieces of Diflon blocks separated by a 50 m thick Teflon spacer. The threeelectrode system was composed of a modified electrode as the working electrode, a stainless steel pipe as the auxiliary electrode and an Ag/AgCl reference electrode. The cell was set in front of the detection window of photomultiplier tube. The mobile phase was 0.1 mol L−1 phosphate buffer solution (pH 9.0) and the flow rate was kept at 0.50 mL min−1 . 2.3. Procedures Before preparation of the Ru(bpy)3 2+ –MWNTs–PSS– ORMOSIL modified electrode, a glassy carbon electrode (diameter 3.0 mm, BAS Co. Ltd, Tokyo, Japan) was sequentially polished with 1, 0.3 and 0.05 m alpha-Al2 O3 , followed by ultrasonic concussion after each polish and then dried at room temperature. MWNTs were immersed in a mixture of concentrated sulfuric and nitric acids (3:1, 98% and 70%, respectively) at 70 ◦ C for 4 h under continuous ultrasonic con-
Y. Tao et al. / Analytica Chimica Acta 594 (2007) 169–174
cussion [35]. The resulting MWNTs had typical lengths of several hundred nanometers, and were terminated by carboxylic and hydroxyl groups. The MWNTs prepared under these experimental conditions were negatively charged [36]. The cut MWNTs were dispersed in water using a sufficient ultrasonic concussion, and then a suitable amount of PSS was added to the mixture to provide a 5% solution by weight. The solution was incubated at 50 ◦ C for 12 h, and then ultrasonically concussed to produce a stable solution of PSS wrapped MWNTs (MWNTs–PSS). Finally, 200 L TMOS, 360 L DiMe–DiMOS, 100 L 0.01 mol L−1 HCl and 200 L MWNTs–PSS were mixed together. The mixture was stirred for about 24 h until a gel solution appeared. A suitable amount (6 L) of the gel solution was dropped onto the GCE surface and dried in the dark at 25 ◦ C for 24 h. The aged MWNTs–PSS–ORMOSIL film modified electrode was then immersed in a 1.0 mmol L−1 Ru(bpy)3 2+ aqueous solution for 0.5 h. The modified GCE was washed thoroughly with water, stored in the ambient state and employed as an ECL sensor. Ru(bpy)3 2+ –PSS–ORMOSIL and Ru(bpy)3 2+ –MWNTs–ORMOSIL modified electrodes were prepared using the same process without the addition of MWNTs or PSS. Under the above conditions, the thickness of the film was around 20 m based on SEM measurement. 3. Results and discussion 3.1. Voltammetric characterization of Ru(bpy)3 2+ in MWNTs–PSS–ORMOSIL film Analogous with the CV behavior of Ru(bpy)3 2+ in aqueous solution (Fig. 1c), a pair of redox waves for MWNTs–PSS–ORMOSIL film appeared with an oxidation potential at 1.10 V and a reduction potential at 1.04 V (Fig. 1b). The interval of redox peaks indicated an excellent electrochemical reversibility of Ru(bpy)3 2+ immobilized in MWNTs–PSS–ORMOSIL film. In general, an electrically insulative ORMOSIL material presented a fully inert
171
ability in the electrochemical reaction of Ru(bpy)3 2+ , but, interestingly, the addition of a suitable amount of PSS and MWNTs obviously improved the electrical character of ORMOSIL film [26,37]. The apparent diffusion coefficient, Dapp , for Ru(bpy)3 2+ in the MWNTs–PSS–ORMOSIL composite films was estimated according to the description of Anson [38]. The calculated Dapp is ca. 2.7 × 10−9 cm2 s−1 , which is almost one order of magnitude larger than the previously reported diffusion coefficient for the Ru(bpy)3 2+ in the PSS–ORMOSIL films (7.6 × 10−10 cm2 s−1 ) [26] and in pure Nafion films (5.2 × 10−10 cm2 s−1 ) [39], and even larger than that previously reported Dapp for the Ru(bpy)3 2+ in the MWNTs–titania–Nafion composite films (7.9 × 10−10 cm2 s−1 ) or the SWNTs–titania–Nafion composite films (1.2 × 10−9 cm2 s−1 ) [37]. An explanation for the phenomenon could be supposed that Ru(bpy)3 2+ remained in a more free state in ORMOSIL–MWNTs–PSS film and the MWNTs–PSS provided a better medium for the electron transfer in the reaction. 3.2. Characterizing the functioning of PSS in ORMOSIL film The SEM images of representative transect of MWNTs–ORMOSIL and MWNTs–PSS–ORMOSIL films are shown in Fig. 2. Although MWNTs were prepared using a mixture of concentrated sulfuric and nitric acids and cut into several hundred nanometers (Fig. 2a), the aggregation of MWNTs that was brought about by the strong inter-forces between MWNTs [34] made it difficult to embed MWNTs well in ORMOSIL films. In Fig. 2b, plot shape MWNT clusters wrapped by ORMOSIL with a diameter of around 40 nm were found inside the film, and these resulted in large size apertures inside and on the surface of the film. Comparatively, in the presence of PSS, the cut MWNTs were homogeneously dispersed in the ORMOSIL film (Fig. 2c) resulting in a much smoother film surface was obtained. In addition, PSS was found to greatly increase the conductivity of ORMOSIL films and EIS measurement showed that the impedance of ORMOSIL films was about 3 × 107 ohm, but the impedance was decreased more than 2 magnitudes with the addition of PSS (Fig. 3a). This could be attributed to the ionic property of PSS. Although MWNTs were immobilized in an insulator matrix without contacting the GCE directly, MWNTs acted as conducting pathways to connect Ru(bpy)3 2+ sites to the electrode. The addition of MWNTs to the ORMOSIL film also caused conductivity increase of the film (Fig. 3b). 3.3. Characterizing the functioning of MWNTs in ORMOSIL film
Fig. 1. Cyclic voltammograms of a MWNTs–PSS–ORMOSIL-modified GCE (a), Ru(bpy)3 2+ immobilized on a MWNTs–PSS–ORMOSIL-modified GCE (b) and 1 × 10−3 mol L−1 Ru(bpy)3 2+ at a bare GCE (c) in 0.1 mol L−1 phosphate buffer solution (pH 9.0), with a scan rate of 0.05 V s−1 .
After immersion in 1.0 mmol L−1 Ru(bpy)3 2+ aqueous solution for 0.5 h, the cyclic voltammograms of the GCEs with PSS–ORMOSIL or MWNTs–PSS–ORMOSIL film modification are presented in Fig. 4. Obviously, the redox currents were synchronously increased by about 10-fold in the pres-
172
Y. Tao et al. / Analytica Chimica Acta 594 (2007) 169–174
Fig. 3. EIS spectra of PSS–ORMOSIL (a) and MWNTs–PSS–ORMOSIL (b) films in the presence of 1.0 × 10−2 mol L−1 [Fe(CN)6]3−/4− (1:1 mixture). The frequency range from 100 kHz to 0.1 Hz at an applied constant bias potential of +0.23 V, with amplitude of alternating voltage at 5 mV in 0.1 mol L−1 KCl solution.
the cutting process also made the MWNTs negatively charged, and therefore increased the interaction between MWNTs and Ru(bpy)3 2+ . The electronic conduction of cut MWNTs and strong absorption of Ru(bpy)3 2+ ions increased the electrochemical activity of the film and resulted in the increased redox currents. 3.4. ECL sensor behavior Fig. 5b shows the ECL signal of the Ru(bpy)3 2+ –PSS– MWNTs–ORMOSIL sensing film in the presence of HS dsDNA upon the oxidation process. The onset luminescence occurred near +1.05 V and then the ECL intensity rose steeply. The ECL intensity reached its maximal value at +1.15 V, which is coincident with the oxidation of the immobilized Ru(bpy)3 2+ [19,21]. Clearly, the ECL response with the addition of HS dsDNA in the buffer solution showed a greater increase than that in the absence of HS ds-DNA. The performance of Ru(bpy)3 2+ ECL in the presence of HS ds-DNA accorded well with its CV result, which indicated that the ECL sensor obtained by Fig. 2. SEM images of the cut MWNTs (a), a representative transect of MWNTs–ORMOSIL film (b), and a representative transect of MWNTs–PSS–ORMOSIL film (c). Inserted SEM images are the surfaces of MWNTs–ORMOSIL and MWNTs–PSS–ORMOSIL film.
ence of MWNTs rather than in their absence, indicating that the addition of MWNTs greatly improved the concentration of Ru(bpy)3 2+ in the MWNTs–PSS–ORMOSIL film. Generally, MWNTs had highly pi-conjunctive and hydrophobic side walls and Ru(bpy)3 2+ had a bipyridine ligand with a large conjugated pi bond, so that there could have been a strong pi pi bond interaction between the MWNTs and Ru(bpy)3 2+ which led to effective absorption of Ru(bpy)3 2+ on the MWNTs [30]. This absorption could be enhanced by the electrostatic interaction of the negative charge on MWNTs and the positively charged Ru(bpy)3 2+ . Furthermore, some functional groups (such as hydroxyl or carboxyl) generated on the ends and sides in
Fig. 4. Cyclic voltammograms of Ru(bpy)3 2+ immobilized on PSS–ORMOSIL (a) and MWNTs–PSS–ORMOSIL (b) modified GCEs in 0.1 mol L−1 phosphate buffer solution, at pH 9.0, with a scan rate of 50 mV s−1 .
Y. Tao et al. / Analytica Chimica Acta 594 (2007) 169–174
Fig. 5. Corresponding ECL-potential curves of the Ru(bpy)3 2+ –PSS–MWNTs– ORMOSIL sensing film in the absence (a) and presence (b) of 1.0 × 10−5 g mL−1 HS ds-DNA in phosphate buffer solution (pH 9.0).
Ru(bpy)3 2+ immobilized MWNTs–PSS–ORMOSIL was sensitive and also favorable in its response to HS ds-DNA. 3.5. Evaluation of the ECL sensor ECL intensities of Ru(bpy)3 2+ and HS ds-DNA were measured as a function of the PSS and MWNT contents at the MWNTs-PSS-ORMOSIL sensing film. The experimental results showed that both their ECL intensities increased rapidly as the PSS and MWNT contents increased, and the ECL intensities appeared at their respective maximal value when 5% PSS and 0.05% MWNTs (w/w) were applied. Higher contents of PSS or MWNTs brought a brighter ECL emission, but the film was crumbly and difficult to be modified on the GCE, which accordingly caused a shorter application life. The ECL intensity of Ru(bpy)3 2+ in MWNTs–PSS– ORMOSIL sensing film greatly depended on the solution pH. Correspondingly, a weak ECL response was obtained before the pH of Ru(bpy)3 2+ in HS ds-DNA phosphate buffer solution reached 6.0, and the maximum signals were obtained at pH 9.0. With further increase of pH, the ECL intensity slightly declined. A tentative reason for this decrease could be the reduction of the available concentration of Ru(bpy)3 2+ at the high pH value, due to the competitive reaction between HS DNA and OH− or its reactive intermediates products. Additionally, noise increased obviously beyond pH 9.0. As previously reported for analytical applications [40,41], a phosphate buffer solution with pH 9.0 was employed. The ECL response of HS ds-DNA was depended on the flow rate of carrier solution in a FIA system. The luminescence intensity decreased with an increase in the flow rate over 0.50 mL min−1 , since the reaction between HS ds-DNA and Ru(bpy)3 2+ occurred as a result of electrode reaction. The luminescence intensity increased at lower flow rates, but the blank response also increased. Consequently, an appropriate flow rate of 0.50 mL min−1 represents a good compromise among maximum sensitivity, peak broadening, and analysis time. The ECL of HS ds-DNA depended significantly on the applied potential (Fig. 6). The ECL response was observed when
173
Fig. 6. Effect of voltage on ECL intensity at Ru(bpy)3 2+ –MWNTs– PSS–ORMOSIL film modified GCE in a FIA system. Experimental conditions: flow rate, 0.5 mL min−1 ; HS ds-DNA concentration, 1.0 × 10−5 g mL−1 ; mobile phase, 0.1 mol L−1 phosphate buffer solution (pH 9.0).
the potential reached +1.05 V, and increased apparently with the increasing of potential over the range of +1.10–1.50 V. Finally, the ECL peak height tended to become constant after reaching +1.50 V. Further increasing the potential of the working electrode did not result in higher ECL peak height but led to bigger background. This might be due to the formation and aggregation of oxygen bubbles on the working electrode surface, causing the obvious ECL noise. The highest ratio of signal-to-noise is obtained at +1.50 V, so we select this value as the working potential. Typically, the ECL sensor constructed by the Ru(bpy)3 2+ – PSS–MWNTs–ORMOSIL modified GCE was stored in the ambient state at room temperature. The sensor was immersed in the phosphate buffer solution for several minutes, and was ready for use after obtaining six ECL signals by CV scan with a standard deviation (S.D.) less than 1%. As shown in Fig. 7, a typical ECL sensing profile was obtained by a successive injection of 1.0 × 10−5 g mL−1 HS ds-DNA in a flow
Fig. 7. Typical ECL responses of the ECL sensor based on Ru(bpy)3 2+ –PSS– MWNTs–ORMOSIL modification in a FIA system. Experimental conditions: flow rate, 0.5 mL min−1 ; applied potential, +1.50 V; HS ds-DNA concentration, 1.0 × 10−5 g mL−1 ; mobile phase, 0.1 mol L−1 phosphate buffer solution (pH 9.0).
174
Y. Tao et al. / Analytica Chimica Acta 594 (2007) 169–174
system. The maximal ECL signal was achieved under conditions of 0.1 mol L−1 phosphate buffer solution (pH 9.0) with a flow rate of 0.5 mL min−1 at the applied potential of +1.5 V. The relative S.D. was 2.2%. A calibration curve was plotted on logarithmic axes to show the wide dynamic ranges. The detection limit (S/N = 3) was 2 × 10−7 g mL−1 with a linear range from 1.3 × 10−6 to 6.67 × 10−4 g mL−1 (R2 = 0.9876). The stability and reproducibility of the ECL sensor for 1.0 × 10−5 g mL−1 HS ds-DNA standard solution during a 1.5 h continuous detection was investigated. The S.D. was less than ±5%. After one month’s storage, the initial ECL responses of the sensor decreased by 6.3% compared with the newly made one, but the characteristics of response linearity remained the same. Although the production reproducibility of the sensor greatly depended on the amounts of PSS and MWNTs, the concentration of immobilized Ru(bpy)3 2+ , the characteristics of the ORMOSIL material, as well as the situation of the experienced sensor producer, the characteristics of the ECL sensor could be controlled well, based on the description in the experimental section. In the preparation of the ECL sensor in our laboratory, involving 10 different GCE modifications, reproducibility within a mean value of ±4.5% was generally achieved for 1.0 × 10−5 g mL−1 HS ds-DNA ECL determination. 4. Conclusions In this paper, we presented a novel Ru(bpy)3 2+ –MWNTs– PSS–ORMOSIL sensing film for ECL sensor construction. It was found that MWNTs played a key role as materials for the conduction and adsorption of Ru(bpy)3 2+ , which obviously increased the Ru(bpy)3 2+ concentration in the ORMOSIL sensing film. The addition of PSS led to MWNTs being dispersed homogeneously in the ORMOSIL film and also improved its conductivity. The Ru(bpy)3 2+ in the sensing film retained well its electrochemical redox activity and ECL response. The detection limit (S/N) of HS ds-DNA using this approach was 2 × 10−7 g mL−1 with a linear range from 1.34 × 10−6 to 6.67 × 10−4 g mL−1 (R2 = 0.9876). In addition, the ECL sensor presented excellent characteristics in terms of stability, reproducibility and application life. Acknowledgements This research work was financially supported by the National Nature Scientific Foundation of China (no. 20375033), the Science and Technology Project of Fujian Province (no. 2005I-030), the Program for New Century Excellent Talents in University of China (NCET) and the Japan Society for Promotion of Science (JSPS), which are gratefully acknowledged. Furthermore, we would like to extend our thanks to Professor John Hodgkiss of The University of Hong Kong for his assistance with English.
References [1] N.E. Tokel, A.J. Bard, J. Am. Chem. Soc. 94 (1972) 2862. [2] B.A. Gorman, P.S. Francis, N.W. Barnett, Analyst 131 (2006) 616. [3] A.J. Bard, J.D. Debad, J.K. Leland, G.B. Sigal, J.L. Wilbur, J.N. Wohlstadter, in: R.A. Meyers (Ed.), Encyclopedia of Analytical Chemistry: Applications, Theory and Instrumentation, John Wiley & Sons, New York, 2000, p. 9842. [4] J. Cadet, M. Weinfeld, Anal. Chem. 65 (1993) 675A. [5] R.M. Santella, Cancer Epidem. Biomar. 8 (1999) 733. [6] R.D. Gerardi, N.W. Barnett, P. Jones, Anal. Chim. Acta 388 (1999) 1. [7] W.D. Cao, J.P. Ferrance, J. Demas, J.P. Landers, J. Am. Chem. Soc. 128 (2006) 7572. [8] G. Firrao, Int. J. Environ. An. Ch. 85 (2008) 609. [9] I. Rubinstein, A.J. Bard, J. Am. Chem. Soc. 102 (1980) 6641. [10] L. Dennany, R.J. Forster, J.F. Rusling, J. Am. Chem. Soc. 125 (2003) 5213. [11] D.H. Johnston, K.C. Glasgow, H.H. Thorp, J. Am. Chem. Soc. 117 (1995) 8933. [12] S.C. Weatherly, I.V. Yang, H.H. Thorp, J. Am. Chem. Soc. 123 (2001) 1236. [13] P.M. Armistead, H.H. Thorp, Anal. Chem. 73 (2001) 558. [14] I. Rubinstein, A.J. Bard, J. Am. Chem. Soc. 103 (1981) 5007. [15] X. Zhang, A.J. Bard, J. Phys. Chem. 92 (1988) 5566. [16] Y.S. Obeng, A.J. Bard, Langmuir 7 (1991) 195. [17] H. Wei, E.K. Wang, Chem. Lett. 36 (2007) 210. [18] N. Egashire, N. Kondoh, Y. Kurauchi, K. Ohga, Denki Kagaku 60 (1992) 1148. [19] Y.F. Zhuang, D.M. Zhang, H.X. Ju, Analyst 130 (2005) 534. [20] A.N. Khramov, M.M. Collinson, Anal. Chem. 72 (2000) 2943. [21] Z.H. Guo, S.J. Dong, Anal. Chem. 76 (2004) 2683. [22] S.N. Ding, J.J. Xu, H.Y. Chen, Electrophoresis 26 (2005) 1737. [23] M.M. Collinson, B. Novak, S.A. Martin, J.S. Taussig, Anal. Chem. 72 (2000) 2914. [24] H.Y. Wang, L.H. Nie, G.B. Xu, S.J. Dong, Analyst 126 (2001) 1095. [25] H.N. Choi, S.H. Cho, W.Y. Lee, Anal. Chem. 75 (2003) 4250. [26] C.Q. Yi, Y. Tao, B. Wang, X. Chen, Anal. Chim. Acta 541 (2005) 75. [27] P.M. Ajayan, Chem. Rev. 99 (1999) 1787. [28] T.W. Ebbesen, H.J. Lezec, H. Hiura, J.W. Bennett, H.F. Ghaemi, T. Thio, Nature 382 (1996) 54. [29] L. Su, F. Gao, L.Q. Mao, Anal. Chem. 78 (2006) 2651. [30] K.P. Gong, X.Z. Zhu, R. Zhao, S.H. Xiong, L.Q. Mao, C.F. Chen, Anal. Chem. 77 (2005) 8158. [31] R.S. Ruoff, D.C. Lorents, Carbon 33 (1995) 925. [32] V.S. Tripathi, V.B. Kandimalla, H.X. Ju, Biosens. Bioelectron. 21 (2006) 1529. [33] V.B. Kandimalla, V.S. Tripathi, H.X. Ju, Biomaterials 27 (2006) 1167. [34] M.J. O’Connell, P. Boul, L.M. Ericson, C. Huffman, Y.H. Wang, E. Haroz, C. Kuper, J. Tour, K.D. Ausman, R.E. Smalley, Chem. Phys. Lett. 342 (2001) 265. [35] J. Liu, A.J. Rinzler, H.J. Dai, J.H. Hafner, R.K. Bradley, P.J. Boul, A. Lu, T. Iverson, K. Shelimov, C.B. Huffman, F. Rodriguez-Macias, Y.S. Shon, T.R. Lee, D.T. Colbert, R.E. Smalley, Science 280 (1998) 1253. [36] O. Matarredona, H. Rhoads, Z.R. Li, J.H. Harwell, L. Balzano, D.E. Resasco, J. Phys. Chem. B 107 (2003) 13357. [37] H.N. Choi, J.Y. Lee, Y.K. Lyu, W.Y. Lee, Anal. Chim. Acta 565 (2006) 48. [38] F.C. Anson, Anal. Chem. 36 (1964) 975. [39] C.R. Martin, I. Rubinstein, A.J. Bard, J. Am. Chem. Soc. 104 (1982) 4817. [40] Z. Massimo, P. Paoio, M. Franco, Anal. Chem. 72 (2000) 4934. [41] J. Matrin, E.J. Hart, A.W. Adamson, J. Halpern, J. Am. Chem. Soc. 94 (1972) 9238.
Tris(2,2-bipyridyl)ruthenium(II) electrochemiluminescence sensor based on carbon nanotube/organically modified silicate films Ying Tao a , Zhi-Jie Lin a , Xiao-Mei Chen a , Xi Chen a,b,∗ , Xiao-Ru Wang a a
Department of Chemistry and Key Laboratory of Analytical Sciences of the Ministry of Education, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005, China b State Key Laboratory of Marine Environmental Science, Xiamen University, Xiamen 361005, China Received 10 March 2007; received in revised form 8 May 2007; accepted 15 May 2007 Available online 21 May 2007
Abstract In this paper, a novel electrochemiluminescence (ECL) sensor was constructed to determine herring sperm (HS) double-stranded (ds) DNA. Tetramethoxysilane and dimethyldimethoxysilane were selected as co-precursors to form an organically modified silicate (ORMOSIL) film for the immobilization of multiwall carbon nanotubes (MWNTs) wrapped by poly(p-styrenesulfonate) (PSS), and then Tris(2,2 -bipyridyl)ruthenium(II) (Ru(bpy)3 2+ ) was successfully immobilized on a glassy carbon electrode via ion-association. PSS was employed to increase the conductivity of the ORMOSIL film and disperse the cut MWNTs, which were cut and shortened in a mixture of concentrated sulfuric and nitric acids, in the film. It was found that MWNTs could adsorb Ru(bpy)3 2+ and acted as conducting pathways to connect Ru(bpy)3 2+ sites to the electrode. MWNTs also played a key role as materials for the mechanical and thermal properties. The ECL performance of this modified electrode was evaluated in a flow injection analysis (FIA) system, and the detection limit (S/N = 3) for HS ds-DNA was 2.0 × 10−7 g mL−1 with a linear range from 1.34 × 10−6 to 6.67 × 10−4 g mL−1 (R2 = 0.9876). In addition, the ECL sensor presented excellent characteristics in terms of stability, reproducibility and application life. © 2007 Elsevier B.V. All rights reserved. Keywords: Electrochemiluminescence sensor; Carbon nanotubes; Tris(2,2 -bipyridyl)ruthenium(II); Nucleic acid
1. Introduction Electrochemiluminescence (ECL) is a well-known detection method that provides high sensitivity and selectivity through the generation of an optical signal triggered by an electrochemical reaction [1,2]. This detection method has been utilized in a number of bioanalytical arenas, and commercial systems have been developed which use ECL to detect many clinically important analytes, including application in immunoassays and DNA analyses by employing ECL-active species as labels of biological molecules [3]. Sensitive and selective detection of DNA is central to clinical tests, pathogen detection, and other methods utilizing the polymerase chain reaction, to genetic disease screening based on oligonucleotide hybridization, and to molecular genotoxicity studies [4,5]. ECL using
∗
Corresponding author. Tel.: +86 592 2184530; fax: +86 592 218 6401. E-mail address: [email protected] (X. Chen).
0003-2670/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.aca.2007.05.025
Tris(2,2 -bipyridyl)ruthenium(II) (Ru(bpy)3 2+ ) could become such a detection method. It provides the necessary sensitivity and low background when utilized in a competitive hybridization assay format that would circumvent labeling of the sample DNA [6–8]. The reaction mechanism which has been generally accepted whereby ECL is generated by electrochemical catalytic pathways between Ru(bpy)3 2+ and oxidized guanine bases in DNA is shown in Scheme 1 [9,10]. Initial oxidation at sufficiently positive potentials gives the RuIII oxidant (Eq. (1)), which reacts with guanines (G) in DNA to give the guanine radical (Eq. (2)) [11,12]. This radical may produce RuII * sites (Eq. (3)), representing the excited state complex, by directly reducing the RuIII sites. G2ox in Eq. (3) represents a guanine oxidized by two electrons, a reaction observed [13] in ss-DNA oxidized by dissolved Ru(bpy)3 3+ . In general, the conventional flow injection analytical method based on Ru(bpy)3 2+ -ECL, requiring Ru(bpy)3 2+ as a component of the mobile phase has a high cost. ECL was generated
170
Y. Tao et al. / Analytica Chimica Acta 594 (2007) 169–174
2. Experimental 2.1. Materials
Scheme 1. Possible pathways for the ECL reaction between Ru(bpy)3 2+ and DNA.
under an oxidation-reduction process of Ru(bpy)3 2+ [13], which revealed that Ru(bpy)3 2+ could be electrochemically regenerated during the ECL reaction. As a result, to develop regenerable ECL-based sensors and detection devices, and to simplify the detection system, Ru(bpy)3 2+ -immobilization approaches have been extensively studied [14–26]. For example, Ru(bpy)3 2+ has been immobilized in Langmuir–Blodgget films [15], selfassembled films [16,17], carbon paste[18,19], polymer films [14,20–22], and sol–gel composites [23–26]. The effort on the immobilization approaches has resulted in rapid development of Ru(bpy)3 2+ ECL sensors and detection devices and is very important for further development of new immobilization approaches. In our laboratory, we developed a Ru(bpy)3 2+ –ORMOSILfilm modified electrode and applied it as an ECL sensor for the determination of methamphetamine [26], but some unsatisfactory points, such as the leaching of Ru(bpy)3 2+ and the weak mechanical strength of the film, still remained. Singlewall (SWNTs) and multiwall (MWNTs) carbon nanotubes have attracted extensive attention owing to their properties of high mechanical strength [27], remarkable electrical conductivity [28] and highly pi-conjunctive and hydrophobic side walls [29,30]. In ECL sensor construction, these characteristics of carbon nanotubes could be applied to improve the performance of the modified electrode, since carbon nanotubes acted as a film-skeleton to increase the mechanical strength and conductivity of the ORMOSIL film [31–33]. Their highly pi-conjunctive and hydrophobic sidewalls enabled carbon nanotubes to be a potential support for Ru(bpy)3 2+ by pi pi bond hydrophobic interactions. Although carbon nanotubes easily aggregated in the ORMOSIL film preparation, fortunately the problem could be avoided based on the approaches reported [34]. In this paper, we report a novel ECL sensor approach using poly(p-styrenesulfonate) (PSS) wrapped MWNTs [34] to fabricate an MWNTs–PSS–ORMOSIL film on a glassy carbon electrode (GCE), and then Ru(bpy)3 2+ immobilized in the film followed by an ion-exchange process. Ru(bpy)3 2+ was easily concentrated in the film due to the strong interaction between Ru(bpy)3 2+ and the MWNTs, resulting in a higher concentration of Ru(bpy)3 2+ in the ORMOSIL film, and this give the modified electrode better stability for longtime usage. The addition of MWNTs improved further the mechanical strength of the film. The significant fact was that Ru(bpy)3 2+ retained its excellent redox ability in the film and showed good ECL responses to herring sperm (HS) DNA. The functioning of film components was studied further by fabricating MWNTs–ORMOSIL and PSS–ORMOSIL films on GCEs.
Ru(bpy)3 Cl2 ·6H2 O, Si(OCH3 )4 (TMOS) and (CH3 )2 Si(OCH3 )2 (DiMe–DiMOS) were obtained from Fluka AG (Buchs, Switzerland) and used as received. Poly(pstyrenesulfonate) (PSS, MW ∼ 70,000) and HS double-stranded (ds) DNA were purchased from Aldrich (Milwaukee, WI, USA). HS ds-DNA stock solution was prepared by dissolving HS dsDNA in a phosphate buffer solution stored at 4 ◦ C, and could be in use for 2 days. The concentration of the stock solution was confirmed by its absorbance measured at 260 nm using a biophotometer (Eppendorf AG, German). The MWNTs with a ∼90% purity and typical lengths of 5–15 m, were obtained from Shenzhen Nanotech. Port. Co. Ltd. All other reagents were of analytical reagent grade and were used without further purification. Distilled water was purified with a Millipore system (Millipore Co., USA), and was used in all aqueous solution preparations and in washing. 2.2. Instrumentation SEM (Oxford Co., UK) was used to study the surface state of the ORMOSIL film under an accelerating voltage at 20 kV. IR spectra were obtained with a FT-IR7400SX spectrometer (Nicolet, USA). Cyclic voltammetry (CV) and electronic impedance spectra (EIS) measurements were performed with a CHI 660B Electrochemical Analyzer (CHI Co. Shanghai, China). A conventional three-electrode system including a modified electrode as the working electrode, a platinum auxiliary electrode and a saturated calomel reference electrode was applied. ECL experiments were carried out on a CHI 660B Electrochemical Analyzer and an IFFM-D FIA Luminescence Analyzer (Ruimai Co., China) at room temperature. The flow system was also equipped with a flow cell and a Rheodyne 7125 sample injector (CA, USA, 50 L). The main body of the thin layer electrolysis flow cell was constructed from two pieces of Diflon blocks separated by a 50 m thick Teflon spacer. The threeelectrode system was composed of a modified electrode as the working electrode, a stainless steel pipe as the auxiliary electrode and an Ag/AgCl reference electrode. The cell was set in front of the detection window of photomultiplier tube. The mobile phase was 0.1 mol L−1 phosphate buffer solution (pH 9.0) and the flow rate was kept at 0.50 mL min−1 . 2.3. Procedures Before preparation of the Ru(bpy)3 2+ –MWNTs–PSS– ORMOSIL modified electrode, a glassy carbon electrode (diameter 3.0 mm, BAS Co. Ltd, Tokyo, Japan) was sequentially polished with 1, 0.3 and 0.05 m alpha-Al2 O3 , followed by ultrasonic concussion after each polish and then dried at room temperature. MWNTs were immersed in a mixture of concentrated sulfuric and nitric acids (3:1, 98% and 70%, respectively) at 70 ◦ C for 4 h under continuous ultrasonic con-
Y. Tao et al. / Analytica Chimica Acta 594 (2007) 169–174
cussion [35]. The resulting MWNTs had typical lengths of several hundred nanometers, and were terminated by carboxylic and hydroxyl groups. The MWNTs prepared under these experimental conditions were negatively charged [36]. The cut MWNTs were dispersed in water using a sufficient ultrasonic concussion, and then a suitable amount of PSS was added to the mixture to provide a 5% solution by weight. The solution was incubated at 50 ◦ C for 12 h, and then ultrasonically concussed to produce a stable solution of PSS wrapped MWNTs (MWNTs–PSS). Finally, 200 L TMOS, 360 L DiMe–DiMOS, 100 L 0.01 mol L−1 HCl and 200 L MWNTs–PSS were mixed together. The mixture was stirred for about 24 h until a gel solution appeared. A suitable amount (6 L) of the gel solution was dropped onto the GCE surface and dried in the dark at 25 ◦ C for 24 h. The aged MWNTs–PSS–ORMOSIL film modified electrode was then immersed in a 1.0 mmol L−1 Ru(bpy)3 2+ aqueous solution for 0.5 h. The modified GCE was washed thoroughly with water, stored in the ambient state and employed as an ECL sensor. Ru(bpy)3 2+ –PSS–ORMOSIL and Ru(bpy)3 2+ –MWNTs–ORMOSIL modified electrodes were prepared using the same process without the addition of MWNTs or PSS. Under the above conditions, the thickness of the film was around 20 m based on SEM measurement. 3. Results and discussion 3.1. Voltammetric characterization of Ru(bpy)3 2+ in MWNTs–PSS–ORMOSIL film Analogous with the CV behavior of Ru(bpy)3 2+ in aqueous solution (Fig. 1c), a pair of redox waves for MWNTs–PSS–ORMOSIL film appeared with an oxidation potential at 1.10 V and a reduction potential at 1.04 V (Fig. 1b). The interval of redox peaks indicated an excellent electrochemical reversibility of Ru(bpy)3 2+ immobilized in MWNTs–PSS–ORMOSIL film. In general, an electrically insulative ORMOSIL material presented a fully inert
171
ability in the electrochemical reaction of Ru(bpy)3 2+ , but, interestingly, the addition of a suitable amount of PSS and MWNTs obviously improved the electrical character of ORMOSIL film [26,37]. The apparent diffusion coefficient, Dapp , for Ru(bpy)3 2+ in the MWNTs–PSS–ORMOSIL composite films was estimated according to the description of Anson [38]. The calculated Dapp is ca. 2.7 × 10−9 cm2 s−1 , which is almost one order of magnitude larger than the previously reported diffusion coefficient for the Ru(bpy)3 2+ in the PSS–ORMOSIL films (7.6 × 10−10 cm2 s−1 ) [26] and in pure Nafion films (5.2 × 10−10 cm2 s−1 ) [39], and even larger than that previously reported Dapp for the Ru(bpy)3 2+ in the MWNTs–titania–Nafion composite films (7.9 × 10−10 cm2 s−1 ) or the SWNTs–titania–Nafion composite films (1.2 × 10−9 cm2 s−1 ) [37]. An explanation for the phenomenon could be supposed that Ru(bpy)3 2+ remained in a more free state in ORMOSIL–MWNTs–PSS film and the MWNTs–PSS provided a better medium for the electron transfer in the reaction. 3.2. Characterizing the functioning of PSS in ORMOSIL film The SEM images of representative transect of MWNTs–ORMOSIL and MWNTs–PSS–ORMOSIL films are shown in Fig. 2. Although MWNTs were prepared using a mixture of concentrated sulfuric and nitric acids and cut into several hundred nanometers (Fig. 2a), the aggregation of MWNTs that was brought about by the strong inter-forces between MWNTs [34] made it difficult to embed MWNTs well in ORMOSIL films. In Fig. 2b, plot shape MWNT clusters wrapped by ORMOSIL with a diameter of around 40 nm were found inside the film, and these resulted in large size apertures inside and on the surface of the film. Comparatively, in the presence of PSS, the cut MWNTs were homogeneously dispersed in the ORMOSIL film (Fig. 2c) resulting in a much smoother film surface was obtained. In addition, PSS was found to greatly increase the conductivity of ORMOSIL films and EIS measurement showed that the impedance of ORMOSIL films was about 3 × 107 ohm, but the impedance was decreased more than 2 magnitudes with the addition of PSS (Fig. 3a). This could be attributed to the ionic property of PSS. Although MWNTs were immobilized in an insulator matrix without contacting the GCE directly, MWNTs acted as conducting pathways to connect Ru(bpy)3 2+ sites to the electrode. The addition of MWNTs to the ORMOSIL film also caused conductivity increase of the film (Fig. 3b). 3.3. Characterizing the functioning of MWNTs in ORMOSIL film
Fig. 1. Cyclic voltammograms of a MWNTs–PSS–ORMOSIL-modified GCE (a), Ru(bpy)3 2+ immobilized on a MWNTs–PSS–ORMOSIL-modified GCE (b) and 1 × 10−3 mol L−1 Ru(bpy)3 2+ at a bare GCE (c) in 0.1 mol L−1 phosphate buffer solution (pH 9.0), with a scan rate of 0.05 V s−1 .
After immersion in 1.0 mmol L−1 Ru(bpy)3 2+ aqueous solution for 0.5 h, the cyclic voltammograms of the GCEs with PSS–ORMOSIL or MWNTs–PSS–ORMOSIL film modification are presented in Fig. 4. Obviously, the redox currents were synchronously increased by about 10-fold in the pres-
172
Y. Tao et al. / Analytica Chimica Acta 594 (2007) 169–174
Fig. 3. EIS spectra of PSS–ORMOSIL (a) and MWNTs–PSS–ORMOSIL (b) films in the presence of 1.0 × 10−2 mol L−1 [Fe(CN)6]3−/4− (1:1 mixture). The frequency range from 100 kHz to 0.1 Hz at an applied constant bias potential of +0.23 V, with amplitude of alternating voltage at 5 mV in 0.1 mol L−1 KCl solution.
the cutting process also made the MWNTs negatively charged, and therefore increased the interaction between MWNTs and Ru(bpy)3 2+ . The electronic conduction of cut MWNTs and strong absorption of Ru(bpy)3 2+ ions increased the electrochemical activity of the film and resulted in the increased redox currents. 3.4. ECL sensor behavior Fig. 5b shows the ECL signal of the Ru(bpy)3 2+ –PSS– MWNTs–ORMOSIL sensing film in the presence of HS dsDNA upon the oxidation process. The onset luminescence occurred near +1.05 V and then the ECL intensity rose steeply. The ECL intensity reached its maximal value at +1.15 V, which is coincident with the oxidation of the immobilized Ru(bpy)3 2+ [19,21]. Clearly, the ECL response with the addition of HS dsDNA in the buffer solution showed a greater increase than that in the absence of HS ds-DNA. The performance of Ru(bpy)3 2+ ECL in the presence of HS ds-DNA accorded well with its CV result, which indicated that the ECL sensor obtained by Fig. 2. SEM images of the cut MWNTs (a), a representative transect of MWNTs–ORMOSIL film (b), and a representative transect of MWNTs–PSS–ORMOSIL film (c). Inserted SEM images are the surfaces of MWNTs–ORMOSIL and MWNTs–PSS–ORMOSIL film.
ence of MWNTs rather than in their absence, indicating that the addition of MWNTs greatly improved the concentration of Ru(bpy)3 2+ in the MWNTs–PSS–ORMOSIL film. Generally, MWNTs had highly pi-conjunctive and hydrophobic side walls and Ru(bpy)3 2+ had a bipyridine ligand with a large conjugated pi bond, so that there could have been a strong pi pi bond interaction between the MWNTs and Ru(bpy)3 2+ which led to effective absorption of Ru(bpy)3 2+ on the MWNTs [30]. This absorption could be enhanced by the electrostatic interaction of the negative charge on MWNTs and the positively charged Ru(bpy)3 2+ . Furthermore, some functional groups (such as hydroxyl or carboxyl) generated on the ends and sides in
Fig. 4. Cyclic voltammograms of Ru(bpy)3 2+ immobilized on PSS–ORMOSIL (a) and MWNTs–PSS–ORMOSIL (b) modified GCEs in 0.1 mol L−1 phosphate buffer solution, at pH 9.0, with a scan rate of 50 mV s−1 .
Y. Tao et al. / Analytica Chimica Acta 594 (2007) 169–174
Fig. 5. Corresponding ECL-potential curves of the Ru(bpy)3 2+ –PSS–MWNTs– ORMOSIL sensing film in the absence (a) and presence (b) of 1.0 × 10−5 g mL−1 HS ds-DNA in phosphate buffer solution (pH 9.0).
Ru(bpy)3 2+ immobilized MWNTs–PSS–ORMOSIL was sensitive and also favorable in its response to HS ds-DNA. 3.5. Evaluation of the ECL sensor ECL intensities of Ru(bpy)3 2+ and HS ds-DNA were measured as a function of the PSS and MWNT contents at the MWNTs-PSS-ORMOSIL sensing film. The experimental results showed that both their ECL intensities increased rapidly as the PSS and MWNT contents increased, and the ECL intensities appeared at their respective maximal value when 5% PSS and 0.05% MWNTs (w/w) were applied. Higher contents of PSS or MWNTs brought a brighter ECL emission, but the film was crumbly and difficult to be modified on the GCE, which accordingly caused a shorter application life. The ECL intensity of Ru(bpy)3 2+ in MWNTs–PSS– ORMOSIL sensing film greatly depended on the solution pH. Correspondingly, a weak ECL response was obtained before the pH of Ru(bpy)3 2+ in HS ds-DNA phosphate buffer solution reached 6.0, and the maximum signals were obtained at pH 9.0. With further increase of pH, the ECL intensity slightly declined. A tentative reason for this decrease could be the reduction of the available concentration of Ru(bpy)3 2+ at the high pH value, due to the competitive reaction between HS DNA and OH− or its reactive intermediates products. Additionally, noise increased obviously beyond pH 9.0. As previously reported for analytical applications [40,41], a phosphate buffer solution with pH 9.0 was employed. The ECL response of HS ds-DNA was depended on the flow rate of carrier solution in a FIA system. The luminescence intensity decreased with an increase in the flow rate over 0.50 mL min−1 , since the reaction between HS ds-DNA and Ru(bpy)3 2+ occurred as a result of electrode reaction. The luminescence intensity increased at lower flow rates, but the blank response also increased. Consequently, an appropriate flow rate of 0.50 mL min−1 represents a good compromise among maximum sensitivity, peak broadening, and analysis time. The ECL of HS ds-DNA depended significantly on the applied potential (Fig. 6). The ECL response was observed when
173
Fig. 6. Effect of voltage on ECL intensity at Ru(bpy)3 2+ –MWNTs– PSS–ORMOSIL film modified GCE in a FIA system. Experimental conditions: flow rate, 0.5 mL min−1 ; HS ds-DNA concentration, 1.0 × 10−5 g mL−1 ; mobile phase, 0.1 mol L−1 phosphate buffer solution (pH 9.0).
the potential reached +1.05 V, and increased apparently with the increasing of potential over the range of +1.10–1.50 V. Finally, the ECL peak height tended to become constant after reaching +1.50 V. Further increasing the potential of the working electrode did not result in higher ECL peak height but led to bigger background. This might be due to the formation and aggregation of oxygen bubbles on the working electrode surface, causing the obvious ECL noise. The highest ratio of signal-to-noise is obtained at +1.50 V, so we select this value as the working potential. Typically, the ECL sensor constructed by the Ru(bpy)3 2+ – PSS–MWNTs–ORMOSIL modified GCE was stored in the ambient state at room temperature. The sensor was immersed in the phosphate buffer solution for several minutes, and was ready for use after obtaining six ECL signals by CV scan with a standard deviation (S.D.) less than 1%. As shown in Fig. 7, a typical ECL sensing profile was obtained by a successive injection of 1.0 × 10−5 g mL−1 HS ds-DNA in a flow
Fig. 7. Typical ECL responses of the ECL sensor based on Ru(bpy)3 2+ –PSS– MWNTs–ORMOSIL modification in a FIA system. Experimental conditions: flow rate, 0.5 mL min−1 ; applied potential, +1.50 V; HS ds-DNA concentration, 1.0 × 10−5 g mL−1 ; mobile phase, 0.1 mol L−1 phosphate buffer solution (pH 9.0).
174
Y. Tao et al. / Analytica Chimica Acta 594 (2007) 169–174
system. The maximal ECL signal was achieved under conditions of 0.1 mol L−1 phosphate buffer solution (pH 9.0) with a flow rate of 0.5 mL min−1 at the applied potential of +1.5 V. The relative S.D. was 2.2%. A calibration curve was plotted on logarithmic axes to show the wide dynamic ranges. The detection limit (S/N = 3) was 2 × 10−7 g mL−1 with a linear range from 1.3 × 10−6 to 6.67 × 10−4 g mL−1 (R2 = 0.9876). The stability and reproducibility of the ECL sensor for 1.0 × 10−5 g mL−1 HS ds-DNA standard solution during a 1.5 h continuous detection was investigated. The S.D. was less than ±5%. After one month’s storage, the initial ECL responses of the sensor decreased by 6.3% compared with the newly made one, but the characteristics of response linearity remained the same. Although the production reproducibility of the sensor greatly depended on the amounts of PSS and MWNTs, the concentration of immobilized Ru(bpy)3 2+ , the characteristics of the ORMOSIL material, as well as the situation of the experienced sensor producer, the characteristics of the ECL sensor could be controlled well, based on the description in the experimental section. In the preparation of the ECL sensor in our laboratory, involving 10 different GCE modifications, reproducibility within a mean value of ±4.5% was generally achieved for 1.0 × 10−5 g mL−1 HS ds-DNA ECL determination. 4. Conclusions In this paper, we presented a novel Ru(bpy)3 2+ –MWNTs– PSS–ORMOSIL sensing film for ECL sensor construction. It was found that MWNTs played a key role as materials for the conduction and adsorption of Ru(bpy)3 2+ , which obviously increased the Ru(bpy)3 2+ concentration in the ORMOSIL sensing film. The addition of PSS led to MWNTs being dispersed homogeneously in the ORMOSIL film and also improved its conductivity. The Ru(bpy)3 2+ in the sensing film retained well its electrochemical redox activity and ECL response. The detection limit (S/N) of HS ds-DNA using this approach was 2 × 10−7 g mL−1 with a linear range from 1.34 × 10−6 to 6.67 × 10−4 g mL−1 (R2 = 0.9876). In addition, the ECL sensor presented excellent characteristics in terms of stability, reproducibility and application life. Acknowledgements This research work was financially supported by the National Nature Scientific Foundation of China (no. 20375033), the Science and Technology Project of Fujian Province (no. 2005I-030), the Program for New Century Excellent Talents in University of China (NCET) and the Japan Society for Promotion of Science (JSPS), which are gratefully acknowledged. Furthermore, we would like to extend our thanks to Professor John Hodgkiss of The University of Hong Kong for his assistance with English.
References [1] N.E. Tokel, A.J. Bard, J. Am. Chem. Soc. 94 (1972) 2862. [2] B.A. Gorman, P.S. Francis, N.W. Barnett, Analyst 131 (2006) 616. [3] A.J. Bard, J.D. Debad, J.K. Leland, G.B. Sigal, J.L. Wilbur, J.N. Wohlstadter, in: R.A. Meyers (Ed.), Encyclopedia of Analytical Chemistry: Applications, Theory and Instrumentation, John Wiley & Sons, New York, 2000, p. 9842. [4] J. Cadet, M. Weinfeld, Anal. Chem. 65 (1993) 675A. [5] R.M. Santella, Cancer Epidem. Biomar. 8 (1999) 733. [6] R.D. Gerardi, N.W. Barnett, P. Jones, Anal. Chim. Acta 388 (1999) 1. [7] W.D. Cao, J.P. Ferrance, J. Demas, J.P. Landers, J. Am. Chem. Soc. 128 (2006) 7572. [8] G. Firrao, Int. J. Environ. An. Ch. 85 (2008) 609. [9] I. Rubinstein, A.J. Bard, J. Am. Chem. Soc. 102 (1980) 6641. [10] L. Dennany, R.J. Forster, J.F. Rusling, J. Am. Chem. Soc. 125 (2003) 5213. [11] D.H. Johnston, K.C. Glasgow, H.H. Thorp, J. Am. Chem. Soc. 117 (1995) 8933. [12] S.C. Weatherly, I.V. Yang, H.H. Thorp, J. Am. Chem. Soc. 123 (2001) 1236. [13] P.M. Armistead, H.H. Thorp, Anal. Chem. 73 (2001) 558. [14] I. Rubinstein, A.J. Bard, J. Am. Chem. Soc. 103 (1981) 5007. [15] X. Zhang, A.J. Bard, J. Phys. Chem. 92 (1988) 5566. [16] Y.S. Obeng, A.J. Bard, Langmuir 7 (1991) 195. [17] H. Wei, E.K. Wang, Chem. Lett. 36 (2007) 210. [18] N. Egashire, N. Kondoh, Y. Kurauchi, K. Ohga, Denki Kagaku 60 (1992) 1148. [19] Y.F. Zhuang, D.M. Zhang, H.X. Ju, Analyst 130 (2005) 534. [20] A.N. Khramov, M.M. Collinson, Anal. Chem. 72 (2000) 2943. [21] Z.H. Guo, S.J. Dong, Anal. Chem. 76 (2004) 2683. [22] S.N. Ding, J.J. Xu, H.Y. Chen, Electrophoresis 26 (2005) 1737. [23] M.M. Collinson, B. Novak, S.A. Martin, J.S. Taussig, Anal. Chem. 72 (2000) 2914. [24] H.Y. Wang, L.H. Nie, G.B. Xu, S.J. Dong, Analyst 126 (2001) 1095. [25] H.N. Choi, S.H. Cho, W.Y. Lee, Anal. Chem. 75 (2003) 4250. [26] C.Q. Yi, Y. Tao, B. Wang, X. Chen, Anal. Chim. Acta 541 (2005) 75. [27] P.M. Ajayan, Chem. Rev. 99 (1999) 1787. [28] T.W. Ebbesen, H.J. Lezec, H. Hiura, J.W. Bennett, H.F. Ghaemi, T. Thio, Nature 382 (1996) 54. [29] L. Su, F. Gao, L.Q. Mao, Anal. Chem. 78 (2006) 2651. [30] K.P. Gong, X.Z. Zhu, R. Zhao, S.H. Xiong, L.Q. Mao, C.F. Chen, Anal. Chem. 77 (2005) 8158. [31] R.S. Ruoff, D.C. Lorents, Carbon 33 (1995) 925. [32] V.S. Tripathi, V.B. Kandimalla, H.X. Ju, Biosens. Bioelectron. 21 (2006) 1529. [33] V.B. Kandimalla, V.S. Tripathi, H.X. Ju, Biomaterials 27 (2006) 1167. [34] M.J. O’Connell, P. Boul, L.M. Ericson, C. Huffman, Y.H. Wang, E. Haroz, C. Kuper, J. Tour, K.D. Ausman, R.E. Smalley, Chem. Phys. Lett. 342 (2001) 265. [35] J. Liu, A.J. Rinzler, H.J. Dai, J.H. Hafner, R.K. Bradley, P.J. Boul, A. Lu, T. Iverson, K. Shelimov, C.B. Huffman, F. Rodriguez-Macias, Y.S. Shon, T.R. Lee, D.T. Colbert, R.E. Smalley, Science 280 (1998) 1253. [36] O. Matarredona, H. Rhoads, Z.R. Li, J.H. Harwell, L. Balzano, D.E. Resasco, J. Phys. Chem. B 107 (2003) 13357. [37] H.N. Choi, J.Y. Lee, Y.K. Lyu, W.Y. Lee, Anal. Chim. Acta 565 (2006) 48. [38] F.C. Anson, Anal. Chem. 36 (1964) 975. [39] C.R. Martin, I. Rubinstein, A.J. Bard, J. Am. Chem. Soc. 104 (1982) 4817. [40] Z. Massimo, P. Paoio, M. Franco, Anal. Chem. 72 (2000) 4934. [41] J. Matrin, E.J. Hart, A.W. Adamson, J. Halpern, J. Am. Chem. Soc. 94 (1972) 9238.