2301

Full Paper

Disposable Screen-Printed Carbon Electrodes for Dual Electrochemiluminescence/Amperometric Detection: Sequential Injection Analysis of Oxalate Mei-Hsin Chiu, Han Wu, Jyh-Cheng Chen, Govindan Muthuraman, Jyh-Myng Zen* Department of Chemistry, National Chung Hsing University, Taichung 402, Taiwan *e-mail: [email protected] Received: June 4, 2007 Accepted: July 25, 2007 Abstract This work describes the development and application of an electrochemical cell specifically designed for disposable screen printed carbon electrodes (SPCE) suitable for simultaneous electrochemiluminescence (ECL) and amperometric detection in sequential injection analysis. The flow system with facility for photomultiplier tube via a fiber optic facing the SPCE is user-friendly and makes the detection process very easy to operate. Instead of the need to constant deliver the chemiluminescence (CL) reagents to the reaction zone, sequential injection analysis allows a considerable reduction in the consumption of the sample and expensive CL reagents (such as Ru(bpy)2þ 3 salts). The utility of the analyzer was demonstrated for the detection of oxalate based on the ECL reaction with 2þ 2þ Ru(bpy)3 . Under optimized conditions, in the presence of 100 mM Ru(bpy)3 , the linear ranges of peak current and ECL light intensity for oxalate distinctly varied from 10 mM to 5 mM and 0.1 mM to 100 mM, respectively. In other words, the linear detection can be covered over a four-order range with the combination of these two signals. Keywords: Electrochemiluminescence, Sequential injection analysis, Screen-printed electrodes, Oxalate DOI: 10.1002/elan.200703984

1. Introduction Electrochemiluminescence (ECL) is a powerful analytical technique with high sensitivity and has been reviewed in detail [1 – 3]. In the past, static electrochemical cells for ECL systems have been used to determine the analytes [4]. Due to its limitation to supply luminescence reagent to the reaction zone, modified electrode [5] and flow injection system [6] have received much attention to partially solve the problem. Nevertheless the usefulness of modified electrodes is relatively limited because of inhibited mass transport of reagent and poor analyte permeation into the film. Flow injection systems coupled with ECL generation using Ru(bpy)2þ 3 have been used to facilitate the regeneration of reagent with higher mass transport to detect alkylamine, amino acids, peptides, and ruthenium-labeled proteins and nucleic acid probe assays [7 – 9]. Regardless of the availability of some commercial instruments for ECL assays, most researchers still develop their own ECL cells to fit their specific requirements and to get valid performance and compatibility. Recent works have focused mainly on miniaturization of cell volume [10 – 12] and electrode size and type [13 – 15]. However the drawback that the working electrode is not easy to renew still restricts its application. Blum and co-workers recently demonstrated that it is possible to use a disposable screen-printed electrode (SPCE) to trigger luminol ECL in a batch system [16]. Previously, Forster and Hogan reported an ECL metalloElectroanalysis 19, 2007, No. 22, 2301 – 2306

polymer coating that combined light and current detection in flow injection analysis [17]. The dual-detection approach provides important additional information about the composition of complex solutions. Our group has developed an electrochemical cell with valid compatibility for disposable SPCE for use in flow injection analysis (FIA) [18]. The goal of this work is therefore to modify the above-mentioned electrochemical cell for simultaneous ECL and amperometric detection. 2 The Ru(bpy)2þ 3 /C2O4 system is one of the most popular 2þ Ru(bpy)3 -based ECL reaction systems and is thus selected as a model analyte in this study. Oxalate is an important marker for the diagnosis of a number of medical conditions including renal failure, vitamin deficiencies, intestinal diseases and hyperoxaluria. Its concentration in urine appears to be a key factor in the formation of kidney stones. In addition the determination of oxalate content in foodstuffs is also important. In this paper, we describe the ability of the proposed electrochemical cell with facility for disposable SPCE to sensitively determine oxalate in aqueous solution with combined light and current detection. Instead of the need to constant deliver the CL reagents to the reaction zone, sequential injection analysis allows a considerable reduction in the consumption of sample and expensive CL salts). These results are reagents (such as Ru(bpy)2þ 3 envisaged to provide a new electrochemical method for sensitive determination of biological analytes and offer a user-friendly setup for electroanalytical applications. C 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

2302

M.-H. Chiu et al.

2. Experimental 2.1. Materials Tris(2,2’-bipyridyl) ruthenium(II) dichloride hexahydrate (Ru(bpy)3Cl2 · 6 H2O) and sodium oxalate were purchased from Aldrich. Sodium dihydrogen phosphate, sodium hydrogen phosphate, sodium hydroxide and phosphoric acid were bought from Showa and used as received. Aqueous solutions were prepared with Millipore de-ionized water (18 MW/cm) throughout this investigation. A 10 mM stock solutions of Ru(bpy)3Cl2 · 6 H2O and Na2C2O4 were prepared with 0.1 M, pH 8 PBS and stored under refrigeration. Carrier and sample solutions were prepared by suitable dilution of the stock solutions before experiments.

on the top folder and a cavity for disposable SPCE on the bottom folder is depicted in Figure 1. In this type of cell device, counter and reference as well as inlet and outlet can be fixed easily on the top folder using capillary holes. An O ring with various sizes is suitably fixed on the center to cover the SPCE not only to prevent leakage of the carrier buffer but also to control the experimental volume. Finally, locking-type screws are used to further tighten the folded device. During the measurement, the PBS was pumped continuously and the merged stream was mixed when the sample solutions that contain Ru(bpy)3Cl2 · 6 H2O and Na2C2O4 were injected from a valve injector. The final stream was introduced into the cell for measuring. While a static potential was applied at working electrode, both the ECL intensity and oxidation current were recorded at the same time.

2.2. Apparatus and Measurement Hydrodynamic amperometric measurements and cyclic voltammetric experiments were carried out with a CHI 703 electrochemical workstation (Austin, TX, USA). The FIA system consisted of a Cole-Palmer microprocessor pump drive, a Rheodyne model 7125 sample injection valve (20 mL loop) and the proposed electrochemical flow cell. The three-electrode system consists of an SPCE working electrode with geometric area 0.07 cm2 (Zensor R&D, Taiwan), an Ag/AgCl reference (Model RE-S, BAS), and a platinum auxiliary electrode. Before each experiment, the SPCE working electrode was subjected to repeat scanning in the potential ranges of 0.0 to 1.4 V (vs. Ag/AgCl) in 0.1 M, pH 8 PBS. The ECL signals were captured using a photomultiplier tube (PMT, Hamamatsu H7421) installed along with the electrochemical workstation. The detailed fabrication of electrochemical cell is described below. Schematic sketches of the proposed button-and-lock-type assembly coupled with a quartz window for PMT detector

3. Results and Discussion 3.1. Characterization of the Reaction Process As described in the experimental section, the proposed ECL cell coupled with a disposable SPCE holds many advantages. In conventional ECL cells, the working electrode is fixed and thus the removal for activation or even the change of new electrode is difficult [15]. In our cell design, the fact that both reference and counter electrodes are connected through a capillary hole on the top of the working electrode results in a very small iR drop. Simply by using various size of O ring, the cell volume can easily be varied from 5 to 100 mL. Here, we used a 100 mL cell volume for demonstration and oxalate was chosen as a model analyte for investigation. Figure 2 illustrates the cyclic voltammetric behavior of 2 the Ru(bpy)2þ 3 /C2O4 system at a bare SPCE in 0.1 M PBS (pH 8). As can be seen, no peak was detected in the blank

Fig. 1. Schematic diagram of the proposed device specifically designed for a disposable SPCE. Electroanalysis 19, 2007, No. 22, 2301 – 2306

www.electroanalysis.wiley-vch.de

C 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

2303

Sequential Injection Analysis of Oxalate

Fig. 2. Cyclic voltammograms recorded at a bare SPCE in different working solutions: a) 0.1 M, pH 8 PBS, b) 1 mM Na2C2 2þ O4, c) 1 mM Ru(bpy)2þ 3 , and d) 1 mM Ru(bpy)3 þ 1 mM Na2C2 O4. Background electrolyte, 0.1 M PBS (pH 8); scan rate, 50 mV/s.

PBS within the scan range of 0.6 – 1.6 V. After adding 1 mM C2O2 4 , an oxidation current started at ca. 1.1 V with a peak at ca. 1.4 V (vs. Ag/AgCl) corresponding to the oxidation of oxalate was observed. As to the electrochemical behavior of Ru(bpy)2þ 3 , a redox couple appeared at ca. 1.1 V similar to that expected for a reversible one-electron-transfer reaction was observed. Most importantly, Figure 2 also shows that there is a sharp electrocatalytic oxidative peak at ca. 1.1 V in the presence of C2O2 and Ru(bpy)2þ 3 . This is a typical 4 example of mediated catalysis through the reaction of C2O2 4 and Ru(bpy)3þ 3 . Since the oxidation of oxalate is accompanied by emission from the cell, indicating a proper ECL reaction mechanism occurs in the proposed system. Note that solution phase mediated catalysis of oxalate using Ru(bpy)2þ in various aqueous solution has been well 3 documented and the mechanism of the ECL reaction is shown below [19].  3þ Ru(bpy)2þ 3 ! Ru(bpy)3 þ e

(1)

2þ 2  Ru(bpy)3þ 3 þ C2O4 ! Ru(bpy)3 þ C2O4

(2)

C2O4 _ ! CO2 _ þ CO2

(3)

2þ  Ru(bpy)3þ 3 þ CO2 _ ! Ru(bpy)3 * þ CO2

(4)

2þ Ru(bpy)2þ 3 * ! Ru(bpy)3 þ hn

(5)

3þ The Ru(bpy)2þ 3 is first oxidized to Ru(bpy)3 on the working 3þ electrode. The electrogenerated Ru(bpy)3 then reacts with 2þ C2O2 4 to form the excited species Ru(bpy)3 *. Finally, the 2þ excited Ru(bpy)3 * species comes back to its ground state Ru(bpy)2þ by emitting light while the CL intensity is 3 monitored and related to the oxalate concentration in the sample. According to mediated catalysis reaction and ECL mechanism, in contrast to the traditional approach in which only the emission is monitored, we simultaneously detect

Electroanalysis 19, 2007, No. 22, 2301 – 2306

Fig. 3. Both ECL and amperometric signals for triplicate injections of 0.5, 1, and 3 mM oxalate þ 100 mM Ru(bpy)2þ 3 at a bare SPCE. Conditions: detection potential ¼ 1.2 V, flow rate ¼ 0.3 mL/ min with pH 8 PBS as mobile phase.

both the amperometric and ECL signals in this study. As shown in Figure 3, both light and current signals can be detected precisely after the sequence injection of 2 Ru(bpy)2þ 3 þ C2O4 . By using 0.1 M, pH 8 PBS as the mobile phase, the advantage of effectively save the usage of Ru(bpy)2þ solution by the proposed system is achieved 3 through the demonstration. Furthermore, the onset of light emission for each peak occurs simultaneously with the appearance of the current transient also indicates that the emission response reaches steady state in a very short time period. The most significant difference in the two responses lies in the stability of the baseline. Due to the intrinsic selectivity of the ECL signals, the much more stable baseline for the ECL signals is as expected. In contrast, any redoxactive species with a formal potential more negative than ca. 1.1 V is capable of generating an amperometric signal. Moreover, the amperometric response is also sensitive to any changes in the double-layer capacitance. The fact that the current response is not as sensitive as the ECL intensity also indicates that the current or emission is controlled by the rate of reaction between oxalate and Ru3þ centers and not by the generation rate of Ru(bpy)3þ 3 [20]. If 2þ  this is the case, the Ru(bpy)3þ 3 þ CO2 _ ! Ru(bpy)3 * þ CO2 step can be regarded as the rate-determining step. Note that those compounds with capability to quench the electrochemically exited state by energy or electron transfer are

www.electroanalysis.wiley-vch.de

C 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

2304

M.-H. Chiu et al.

potential interference [17]. In these respects, we have carried out experiments with various gases such as air, Ar, CO2 and O2. Combined current and light detections in the presence of the above mentioned gases are shown in Figure 4. As can be seen, the observed current responses are almost same in the presence of air, O2, and Ar with the exception of CO2. This is as expected, since Reaction (3), i.e., oxalate ion to produce the strong reductant CO2 . upon oxidation in aqueous solution, becomes slower in the presence of CO2. As to the ECL signals, the greatly inhibited light intensity in the presence of O2 is again as expected due to the efficient exited quencher of Ru(bpy)2þ 3 * [17]. The obviously enhanced ECL signal in the presence of CO2 is somewhat surprising and needs more study to clarify the mechanism.

3.3. Analytical Performance The optimization of analytical parameters, such as detection potential and flow rate, that can affect the analytical performance in sequential injection analysis were studied next. As can be seen in Figure 5, the amperometric signals with a maximum at ca. 1.3 V match very well with the cyclic voltammogram observed for the oxidation of oxalate.

Fig. 4. Interferences of ECL and amperometric signals on the detection of 100 mM Na2C2O4 in presence of 100 mM Ru(bpy)2þ 3 at various gases of a) air, b) Ar, c) CO2, and d) O2. Conditions: detection potential ¼ 1.2 V, flow rate ¼ 0.3 mL/min with pH 8 PBS as mobile phase. Electroanalysis 19, 2007, No. 22, 2301 – 2306

Meanwhile the ECL intensity for the injection of 2 Ru(bpy)2þ 3 þ C2O4 reached a maximum at ca. 1.1 V. The observation is indeed also in good agreement with the electrocatalytic responses observed in Figure 2. Since the detection at 1.2 V resulted in a relatively smaller signal deviation, a detection potential of 1.2 V was thus used as the optimal for the following studies. As to the effect of flow rate, Figure 6 shows the results obtained for two different concentrations of Ru(bpy)2þ 3 in the detection of 1 mM Na2C2 O4. At a higher concentration of 1 mM Ru(bpy)2þ 3 , the ECL intensity was found to increase with the increase in flow rate up to 0.4 mL/min; after that a plateau like response was obtained. In the case for a lower concentration of 100 mM Ru(bpy)2þ 3 , the highest intensity was observed at a flow rate of 0.3 mL/min and started to decrease at higher flow rate. The decrease in ECL intensity at higher flow rate while using a lower Ru(bpy)2þ 3 concentration is presumably due to the longitudinal diffusion of Ru(bpy)3þ 3 . While the flow rate and hence sample throughput can certainly be increased for relatively concentrated Ru(bpy)2þ 3 , a flow rate of 0.3 mL/ min represents a good compromise among the reduction in the consumption of expensive Ru(bpy)2þ and maximum 3 sensitivity. It was thus chosen in the subsequent experiments. Under the optimum parameters: i.e., detection potential ¼ 1.2 V, flow rate ¼ 0.3 mL/min, different concentrations of 1 mM and 100 mM of Ru(bpy)2þ 3 were used to detect oxalate. It was found a relatively wider linear range of 0.5 mM – 1 mM was obtained for 1 mM Ru(bpy)2þ 3 compared to that of 0.1 mM – 100 mM for 100 mM Ru(bpy)2þ 3 . Based on is a limiting reagent and the results, we infer that Ru(bpy)2þ 3 reacts with oxalate by 1 : 1 in the ECL system. As to the amperometric current, more concentrated Ru(bpy)2þ 3 causes a higher background current and hence a relatively poor detection limit. Typical calibration plots of ECL and amperometric peak intensity versus oxalate concentration are shown in Figure 7. As can be seen, in the presence of 100 mM Ru(bpy)2þ 3 ,

Fig. 5. Effect of detection potential on current and light variation for the detection of 1 mM Na2C2O4 using 1 mM and 0.1 M PBS (pH 8) as carrier solution. Flow Ru(bpy)2þ 3 rate ¼ 0.2 mL/min.

www.electroanalysis.wiley-vch.de

C 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

2305

Sequential Injection Analysis of Oxalate

Fig. 6. Effect of flow rate to the ECL signal by using a) 1 mM for the detection of 1 mM Na2C2O4. and b) 100 mM Ru(bpy)2þ 3 Detection potential ¼ 1.2 V with pH 8 PBS as mobile phase.

the ECL light intensity increased linearly in the oxalate concentration range of 0.1 mM to 100 mM with a slope value of 113 (counts/mM). As to the amperometric current responses, a linear range between 10 mM to 5 mM with a slope value of 0.01 mA/mM was observed. It is interesting that the obtained linear range of theses two signals is very different. The precision of the proposed system was further evaluated by repeating injections of the same oxalate concentration at a level of 10 mM. As shown in Figure 8, both signals are very consistent and the relative standard deviation (n ¼ 10) for the ECL and amperometric responses was 1.68% and 3.26%, respectively. The utility of the proposed system was demonstrated for the detection of oxalate with satisfactory results. Most importantly the combined ECL and amperometric detection can be used to detect oxalate from nanomolar to millimolar ranges by using 100 mM Ru(bpy)2þ 3 as the reactant.

Fig. 7. Calibration plots for A) ECL light intensity, B) amperometric current responses with increasing concentration of Na2C2O4. Other conditions are the same as in Figure 3. Electroanalysis 19, 2007, No. 22, 2301 – 2306

www.electroanalysis.wiley-vch.de

C 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

2306

M.-H. Chiu et al.

to detect oxalate from nanomolar to millimolar ranges by using this new cell with the help of 100 mM Ru(bpy)2þ 3 as the reactant. Most importantly, it paves the way for the usage of disposable electrodes in ECL analysis.

5. Acknowledgement The authors gratefully acknowledge financial support from the National Science Council of Taiwan.

6. References

Fig. 8. The precision of A) ECL light intensity, B) amperometric current responses by repeating injections of 10 mM Na2C2O4. Other conditions are same as in Figure 3.

4. Conclusions A user-friendly electrochemical cell with compatibility for SPCE was successfully developed and applied to ECL analysis with Ru(bpy)2þ 3 /oxalate standard system in sequence injection analysis. The proposed cell allows to simultaneously measuring of both current and light signals and gives a superior analytical performance. Sequential injection analysis allows a considerable reduction in the consumption of sample and expensive Ru(bpy)2þ 3 reagent. The combined ECL and amperometric signals can be used

Electroanalysis 19, 2007, No. 22, 2301 – 2306

[1] R. D. Gerardi, N. W. Barnett, S. W. Lewis, Anal. Chim. Acta 1999, 378, 1. [2] X. Yin, S. Dong, E. Wang, TrAC-Trends, Anal. Chem. 2004, 23, 4321. [3] M. M. Richter, Chem. Rev. 2004, 104, 3003. [4] I. Rubinsteln, C. R. Martin, A. J. Bard, Anal. Chem. 1983, 55, 1580. [5] K. M. Maness, R. H. Terrill, T. J. Meyer, R. W. Murray, R. M. Wightman, J. Am. Chem. Soc. 1996, 118, 10609. [6] W.-Y. Lee, T. A. Nieman, Anal. Chem. 1995, 67, 1789. [7] W.-Y. Lee, Mikrochim. Acta 1997, 127, 19. [8] K. A. FOhnrich, M. Pravda, G. G. Guilbault, Talanta 2001, 54, 531. [9] A. W. Knight, Trends, Anal. Chem. 1999, 18, 47. [10] L. L. Shultz, J. S. Stoyanoff, T. A. Nieman, Anal. Chem. 1996, 68, 349. [11] M. M. Collinson, R. M. Wightman, Anal. Chem. 1993, 65, 2576. [12] D. R. Skotty, W.-Y. Lee, T. A. Nieman, Anal. Chem. 1996, 68, 1530. [13] L. S. Kuhn, A. Weber, S. G. Weber, Anal. Chem. 1990, 62, 1631. [14] H. Wang, G. Xu, S. Dong, Electrochem. Commun. 2002, 4, 214. [15] Y. Chi, J. Duan, S. Lin, G. Chen, Anal. Chem. 2006, 78, 1568. [16] B. Leca, L. J. Blum, Analyst 2000, 125, 781. [17] R. J. Forster, C. F. Hogan, Anal. Chem. 2000, 72, 789. [18] C. T. Hsu, H.-H. Chung, H.-J. Lyuu, D.-M. Tsai, A. S. Kumar, J.-M. Zen, Anal. Sci. 2006, 22, 35. [19] I. Rubenstein, A. J. Bard, J. Am. Chem. Soc. 1981, 103, 512. [20] C. F. Hogan, R. J. Forster, Anal. Chim. Acta 1999, 396, 13.

www.electroanalysis.wiley-vch.de

C 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim