Available online at www.sciencedirect.com

Electrochemistry Communications 9 (2007) 2744–2750 www.elsevier.com/locate/elecom

A poly(dimethylsiloxane)-based electrochemical cell coupled with disposable screen printed edge band ultramicroelectrodes for use in flow injection analysis Jen-Lin Chang, Jyh-Myng Zen

*

Department of Chemistry, National Chung Hsing University, 250 Kuo-Kuang Road, Taichung 402, Taiwan Received 23 July 2007; received in revised form 18 September 2007; accepted 19 September 2007 Available online 24 October 2007

Abstract We report here the development of an inexpensive poly(dimethylsiloxane) (PDMS)-based electrochemical cell specifically designed for disposable screen printed edge band ultramicroelectrodes (SPUMEs) for use in flow injection analysis (FIA). The SPUME is fabricated with a built-in three-electrode pattern on a non-conducting polypropylene substrate. The edge of the carbon and/or metal-sandwiched films between the insulator layers can serve as a band type ultramicroelectrode. Fabrication of the cell is straightforward; no micromechanical operation is included. Simply by molding of PDMS with a ‘‘T’’ type channel to fix the SPUME in a confined wall-jet-type con4 figuration, the performance characteristic of the proposed cell was evaluated by using the FeðCNÞ3 6 =FeðCNÞ6 redox couple. The high velocity jet of solution resulted in a mass transport coefficient up to ca. 0.48 cm/s. The proposed FIA system was applied for the detection of nitrite and the current response was linear up to 700 lM with a detection limit of 0.067 lM (S/N = 3). Finally the determination of nitrite in lake and ground waters without the addition of supporting electrolyte was successfully demonstrated.  2007 Elsevier B.V. All rights reserved. Keywords: Ultramicroelectrode; Disposable; Screen printed electrode; Nitrite; PDMS

1. Introduction The application of ultramicroelectrodes (UMEs) has continuously attracted a great deal of attention in various research fields. The advantages include largely increasing in mass transport to the electrode surface, minimizing of iR drop even in the absence of supporting electrolyte or in highly resistive non-aqueous media, and facilitating fast response with a steady state current–potential response [1– 5]. So far, diverse types of UMEs have been developed for ultratrace analysis [2,6–13]; yet, the fabrication of UMEs normally requires tedious and time-consuming procedures and thus limits their widespread use. Our group recently reported a disposable screen printed edge band ultramicro-

*

Corresponding author. E-mail address: [email protected] (J.-M. Zen).

1388-2481/$ - see front matter  2007 Elsevier B.V. All rights reserved. doi:10.1016/j.elecom.2007.09.014

electrode (SPUME), which is low-cost (thus disposable), easy for mass production, and flexible in design [14]. This disposable SPUME with an in-built three-electrode configuration was successfully demonstrated for electroanalysis of nitrite by linear scan voltammetry with a detection limit of 0.38 lM (S/N = 3) [15]. In continuation of our previous investigation, in this study, we further develop an inexpensive poly(dimethylsiloxane) (PDMS)-based electrochemical cell specifically designed for this disposable SPUME for use in flow injection analysis (FIA). Note that the advantages of hydrodynamic UME include remarkable sensitivity, wide linear calibration range, low dead volume, and fast response time. PDMS is a soft material widely used in the field of microfabrication [16,17]. The main course of its fabrication includes preparation of the template and curing of the PDMS precursor on it. By taking this material into the construction of a simplified electrochemical cell, it is

J.-L. Chang, J.-M. Zen / Electrochemistry Communications 9 (2007) 2744–2750

expected that the flexibility of the material can effectively avoid the solution leakage. Micromolding of the material provides an easy way to replicate cells with a high efficiency. Overall, by curing of the PDMS precursor on a designed template, the electrochemical cell fabrication process is simple, cheap, precise, and reproducible. As mentioned earlier, this disposable SPUME was successfully demonstrated for electroanalysis of nitrite by linear scan voltammetry with a detection limit of 0.38 lM (S/N = 3) [15]. For the purpose of comparison, the determination of nitrite is again chosen as the model analyte in this study. The optimization of analytical parameters, such as distance between the SPUME and the wall-jet capillary inlet and flow rate, that can affect the analytical performance in FIA were carefully evaluated. Finally, the proposed system was used in real sample analysis to detect the amount of nitrite in lake and ground waters without the addition of supporting electrolyte. These demonstrations are envisaged to provide a useful electrochemical method as well as a userfriendly setup for electrochemical applications. 2. Experimental 2.1. Reagents and chemicals Potassium ferricyanide and sodium nitrite were obtained from Sigma (St. Louis, MO, USA). Carbon and silver inks were purchased from Acheson (Tokyo, Japan). Conventional SPE (3 mm in diameter) in three-electrode configuration was obtained from Zensor R&D (Taichung, Taiwan). All the other compounds (ACS-certified reagent grade) were used without further purification. Aqueous solutions were made up of de-ionized water prepared from the Mil-

2745

lipore-Q purification system. Natural water samples were collected in a polyethylene bottle from the campus of Chung Hsing University and kept under refrigeration below 4.8 C and were directly used for detection. 2.2. SPUME and cell design The wall-jet cell assembly as well as the incorporation of the SPUME into the wall-jet electrochemical cell is depicted in Fig. 1. By molding of PDMS with a ‘‘T’’ type channel (depth 1.15 mm and width 1.15 mm) the SPUME not only can be easily fixed but also can prevent the solution leakage from inlet and outlet. Since the capillary within the channel has an inner diameter of 0.5 mm, the insertion as well as the alignment of the SPUME into the flow cell is also very simple. Two polymethylmethcrylate (PMMA) plates were then used to cover and strengthen the wall-jet cell. The system is capable of operating under a high pressure flow system of up to 10 cm3/s. The fabrication of the SPUME was the same as reported by our group earlier [14]. In brief, different stencil assemblies were first prepared properly to fabricate the multilayer SPE on a 50 mm · 15 mm polypropylene (PP) base. The layer-by-layer assembly of the built-in three-electrode system contains a stencil format in the order of carbon ink ! insulating polymer ! silver ink ! insulating polymer ! silver ink ! insulating polymer. The as-prepared SPUME was then cured in an UV radiation source at an intensity of 1.85 mW/cm2 for 2 h. The tip edge window was suitably sliced to expose the SPUME with a built-in three-electrode pattern. Note that the procedures can allow for preparing versatile three-electrode SPUME suitably inbetween the insulating polymeric layers.

Fig. 1. Scheme and pictorial representation for the proposed electrochemical cell with the SPUME: (A) arrangement of the cell component, (B) magnification of the T-type interface between the injection inlet and the SPUME, (C) photograph of the proposed system.

2746

J.-L. Chang, J.-M. Zen / Electrochemistry Communications 9 (2007) 2744–2750

2.3. Apparatus

7125 sample injection valve (20 lL loop) with interconnecting Teflon tube, and the proposed SPUME system. The SPUME in three-electrode configuration consists of carbon-working, silver-quasi reference, and silver-counter electrodes. The physical features, such as electrode topography and edges, were characterized by scanning electron

All electrochemical experiments were performed either on a CHI 832 or a CHI 8021 electrochemical workstation (Austin, TX, USA). The FIA system contains a ColeParmer microprocessor pump drive, a Rheodyne Model

60

B

A

40

ip / nA

20

0.1 M KCl 0

10 μM KCl 0.1 M KCl

10 nA

-20

-40 0.4

10 μM KCl

0.3

0.2

0.1

0.0

-0.1

E / V (vs. pseudo Ag)

20 s

Fig. 2. Typical i–t responses (A) and hydrodynamic voltammograms (B) of 300 lM K3Fe(CN)6 at the SPUME in (a) 10 lM KCl and (b) 0.1 M KCl mobile phase. Conditions: detection potential at 0.1 V with a flow rate of 1.4 mL/min.

35

30 5 nA

25

ip / nA

20

5 nA

15

5 nA

D 10

5

0 0.0

0.2

0.4

0.6

0.8

1.0

1.2

D / mm Fig. 3. Influence of distance (D) between capillary injection gateway and the SPUME in 300 lM K3Fe(CN)6 in 0.1 M KCl mobile phase. Conditions: detection potential at 0.1 V with a flow rate of 0.8 mL/min.

J.-L. Chang, J.-M. Zen / Electrochemistry Communications 9 (2007) 2744–2750

microscopy (SEM). Before electrochemical experiments, the working electrode of the SPUME was electrochemically pretreated in 0.1 M KCl within the potential window between oxygen and hydrogen gas evolution reaction until the current becomes constant. It normally took 20 continuous cycles at m = 50 mV/s.

Vf / ( cm3/min ) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 40

A

30

40

3.1. Electrochemical characteristics

30

20

20

10

Hpw / s

50

ip/ nA

3. Results and discussion

2747

10 0

3.2. Parameters optimization in FIA In the FIA experiments, the analytical performance was found to depend on the distance between the capillary inlet tip and SPUME. Fig. 3 shows the current responses in FIA observed by varying the distance between the inlet tip and the SPUME. When the tip of the capillary and SPUME is relatively far away (e.g., 1.15 mm), the increase in mass transfer is rather limited and hence a comparatively smaller current response was observed. As the gap is close down to 0.5 mm, a much higher gradient of electroactive species leads to the increase in current signals [22]. Nevertheless the current starts to decrease a little as the tip is even closer to the electrode surface (e.g., 0.15 mm) presumably due to the domination of turbulent flow rather than laminar flow. As a result, a distance of 0.5 mm between the capillary inlet tip and SPUME is selected in subsequent studies. As to the effect of flow rate, the signal intensity increases initially and remains constant afterward, as shown in Fig. 4A. It indicates a mass transport limited rather than diffusion limited property. Furthermore, the half peak

0 50

B

40

ip / nA

Previously, we have employed the SPUME for voltammetric studies [14,15]. A sigmoidal shape of current–potential response clearly indicates a near ideal behavior of the Fe(CN)63/4 redox species at the SPUME [18]. In this study, the SPUME coupled with the proposed electrochemical cell was first time used for electroanalysis by FIA. Fig. 2A shows the typical i–t curve obtained at the proposed system for 300 lM Fe(CN)63 by using 0.1 M KCl solution as the mobile phase. As can be seen, very consistent current responses for four repeated injections are obtained indicating good reproducibility of the electrodes surface. Most importantly, similar current responses were observed at various working conditions of 0.1 M and 10 lM KCl validating the SPUME for use in electroanalysis even in low concentration of supporting electrolyte. By measuring the peak current at different detection potentials, the hydrodynamic i-E behavior on the SPUME for 300 lM Fe(CN)63 in 0.1 M and 10 lM KCl was further compared and studied. As shown in Fig. 2B, current– potential responses with a sigmoidal shape were indeed observed for both cases. In other words, the SPUME can permit electrochemical measurements in low dielectric permittivity and low concentration supporting electrolyte by the proposed system in FIA [19–21].

30 20 10 0 2.0

2.5

3.0

3.5

4.0

4.5

5.0

Vf1/3 Fig. 4. (A) Effect of flow rate on (a) signal height and (b) half potential width (Hpw) for reduction of 300 lM K3Fe(CN)6 in 0.1 M KCl mobile phase; (B) transport limited current (ilim) versus flow rate (Vf) plot. Other conditions are the same as in Fig. 2.

width (Hpw) decreases sharply initially and remains constant with the increase in flow rate after that. This is as expected for a wall-jet flow system and further confirms the reversible nature of Fe(CN)63/4. Note that, considering the ratio of electrode length (1.15 mm) to nozzle radius (0.5 mm), the SPUME is a wall-jet electrode with laminar flow rather than turbulent flow. Fig. 4B illustrates the dependence of flow rate with mass transport limited current (ilim) for Fe(CN)63. According to our cell design, even though the flow is perpendicular to the SPUME, the outlet with a band shape of SPUME and width of the wall makes the flow considered as axial rather than radial. In that case, Levich has developed an equation for axial flow hydrodynamic microband channel electrode for axial flow, where steady-state transport-limited current were found to be dependent on the cube-root of volume of flow rate (Vf) [23]: 2=3

ilim ¼ 0:925FC½FeðCNÞ xe ðwDÞ

ðV f =h2 dÞ

1=3

ð1Þ

where F is the Faraday constant; D is the diffusion coefficient of the electroactive species Fe(CN)63. Vf is the volume flow rate; xe is the electrode length; w is the electrode width; h is the half-height of the channel; and d is the channel width. By using the values of xe = 1.15 mm, w = 20 lm, 2h = 0.5 mm, and d = 0.5 mm, a value of 8.4 · 1012 m2/s for D was deduced from the slope of Le-

2748

J.-L. Chang, J.-M. Zen / Electrochemistry Communications 9 (2007) 2744–2750

vich plot. This value is fairly close to the reported literature value of the diffusion coefficient of Fe(CN)63 [24]. In order to consider mass transport coefficient (kt), the following equation was adapted: ilim ¼ FAj ¼ 4FAk t C 0

ð2Þ

where j is a mans flux to the electrode of area; A is electrode area; and C0 is concentration of bulk solute. Under a relatively slow flow condition, for an electrode area of ca. 2.3 lm2, the mass transport coefficient (kt) can be as high as 0.46 cm/s. Note that Compton and co-workers have developed a high-pressure flow system in which solution is flowed at high volume flow rates (approaching 10 cm3/s) over a microband electrode in a channel geometry [25], with resulting mass transport coefficients up to ca. 1 cm/s. Furthermore, as depicted in Fig. 5, a wide linear range (1– 600 lM) for the determination of Fe(CN)63 indicates the validity of the SPUME in the proposed cell system. Overall, the results confirm that the SPUME and wall-jet cell can be used for kinetic and analytical assay purpose in low concentration electrolytes. 3.3. Analytical consideration As mentioned earlier, for the purpose of comparison, nitrite is chosen as a model analyte in this study. Fig. 6

shows the hydrodynamic voltammograms of nitrite in the millimolar to sub-micromolar concentration range in low concentration supporting electrolyte of 10 lM H2SO4 at the SPUME. An applied potential of 1.3 V (i.e., in the limiting current region) and a flow rate of 1.4 mL/min was used for the experiments. As can be seen, good linearity up to 700 lM was obtained at the proposed system for the detection of nitrite with a correlation coefficient of 0.995. This result is as good as that observed with a linear relationship from 1–600 lM when 0.1 M KCl containing Fe(CN)63 was measured by continuously injecting the solution into the flow cell (correlation coefficient = 0.999), as shown in Fig. 5. Thus, there appears to be no need of high supporting electrolyte concentration in cases where the detector system response is sufficiently fast to accurately record the peak height at the SPUME. Furthermore, a relative standard deviation of 2.2% (n = 5) for analyzing 1 lM nitrite indicated a detection limit of 0.067 lM (S/ N = 3) by FIA, as shown in Fig. 6. This is a clear improvement compared to the detection limit of 0.38 lM (S/N = 3) by linear scan voltammetry using the same SPUME as reported by our group earlier [15]. From the above results, the SPUME combined with the microfabricated wall-jet cell is suitable as a detector for small-volume flow measurement in the presence of low concentration of supporting electrolyte. Another merit of the fabrication technique

10 nA

1

100 s

600 μM

2 nA 50 s

10

50 μM

0.2 nA 100 s

1

9 μM

Fig. 5. Typical calibration curve for various concentration of K3Fe(CN)6. Conditions are the same as in Fig. 2. Insert shows the linear plot with linear range between 1–600 lM K3Fe(CN)6 with r2 = 0.9998.

J.-L. Chang, J.-M. Zen / Electrochemistry Communications 9 (2007) 2744–2750

Bulk current

1

5

2749

10 30

50

70 100

10 pA 300

100 s

50

10 nA

B

40

500

700 μM

i p / nA

30

20

10

A

0 0

200

400

600

800

[NO2 ] / μM Fig. 6. Typical calibration curve for various concentration of nitrite in 10 lM H2SO4. Conditions: detection potential at 1.3 V with a flow rate of 1.4 mL/ min. Insert shows the linear plot with a linear range up to 700 lM nitrite with r2 = 0.995.

described here, of course, is the convenience for replicating the cells. In our method, after the mold is ready, only a little labor is required if more cells are needed. They can be quickly replicated at a rather low-cost with good reproducibility. 3.4. Real sample assay In order to investigate the applicability of the proposed system to determine nitrite ion in natural waters, experiments were performed in lake and ground waters collected from the campus of Chung Hsing University. Note that the water samples were analyzed without any pretreatment. Standard addition method was used for analysis and the obtained results are summarized in Tables 1 and 2 for ground water and lake water, respectively. Meanwhile the results of recovery test at different nitrite concentration levels from 30 to 700 lM were also demonstrated in the tables. As can be seen in Table 1, the recoveries of hydrodynamic voltammograms of nitrite in ground water are found to perform as good as those obtained by linear scan voltammetry as reported in our previous study [15]. In this study,

Table 1 Results from studies of nitrite determination in ground water by LSV and FIA ½NO 2 /[M] added

½NO 2 /[M] found

Recovery [%]

FIA 30.00 50.00 70.00 90.00 300.00 500.00 700.00 x

28.81 49.63 73.36 100.78 296.59 499.04 701.75 –

96.05 99.26 104.80 100.78 98.86 99.81 100.25 99.97 ± 1.69

LSV 30.00 50.00 70.00 90.00 300.00 500.00 700.00 x

31.83 54.58 67.72 91.71 274.28 474.65 664.54 –

106.1 109.2 96.7 101.9 91.4 94.9 94.9 99.0 ± 5.7

x: arithmetic mean.

2750

J.-L. Chang, J.-M. Zen / Electrochemistry Communications 9 (2007) 2744–2750

Table 2 Results from studies of nitrite determination in lake water by FIA

Acknowledgements

½NO 2 /[M] added

½NO 2 /[M] found

Recovery [%]

30.00 50.00 70.00 90.00 300.00 500.00 700.00 x

32.15 56.98 68.85 94.90 301.24 486.54 709.30 –

107.15 113.96 98.35 94.90 100.41 97.31 101.33 101.92 ± 4.93

We would like to thank the financial support from National Science Council of Taiwan. This work is supported in part by the Ministry of Education, Taiwan under the ATU plan.

a more complicated matrix of lake water was also subjected for analysis by the proposed FIA system. As can be seen in Table 2, even in a more complicated matrix of lake water, the mean recoveries of 101.9% was also found to be very good. These results indicate that the method is applicable to natural water application.

4. Conclusion The newly developed wall-jet flow cell was successfully demonstrated for nitrite determination using the SPUME by FIA. A model compound Fe(CN)63 was studied in the proposed wall-jet flow cell confirming the applicability of detection in low concentration of supporting electrolyte with a mass transfer controlled mechanism. Effects of distance between SPUME and capillary and flow rate indicate the compression of diffusion layer with an axial flow mechanism. The system was successfully applied in the detection of nitrite in natural waters. The performed analytical tests showed good results for precision and accuracy. Since both the SPUME and PDMS-based electrochemical cell are cheap, the electroanalytical setup can offer a new platform to diverse applications.

References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25]

J.V. Macpherson, P.R. Unwin, Anal. Chem. 71 (1999) 4642. F. Zhou, J.T. Aronson, M.W. Ruegnitz, Anal. Chem. 69 (1997) 728. C. Ceballos, H. Ferna´ndez, Food Res. Int. 33 (2000) 357. R. Feeney, S.P. Kounaves, Anal. Chem. 72 (2000) 2222. A.J. Bard, L.R. Faulkner, Electrochemical Methods Fundamentals and Applications, second ed., Wiley, New York, 2001. J.V. Macpherson, Electroanalysis 12 (13) (2000) 1001. J.V. Macpherson, P.R. Unwin, Anal. Chem. 71 (1999) 2939. N. Simjee, J.V. Macpherson, P.R. Unwin, Electroanalysis 15 (18) (2003) 1445. K. Yunus, A.C. Fisher, Electroanalysis 15 (22) (2003) 1782. A.M. Nowicka, M. Donten, M. Palys, Z. Stojek, Anal. Chem. 77 (2005) 5174. D.E. Williams, J.V. Macpherson, Comp. Chem. Kin. 37 (1999) 369. J. Wang, Anal. Chem. 53 (1981) 1528. W.J. Blaedel, S.L. Boyer, Anal. Chem. 43 (1971) 1538. J.-L. Chang, J.-M. Zen, Electrochem. Commun. 8 (2006) 571. J.-L. Chang, J.-M. Zen, Electroanalysis 18 (2006) 941. J. Yan, X. Yang, E. Wang, Anal. Chem. 77 (2005) 5385. Y.S. Heo, S. Chung, K. Cho, C. Chung, D.C. Han, J.K. Chang, J. Chromatogr. A 1013 (2003) 111. C.G. Zoski, Electroanalysis 14 (2002) 1042. M.J. Pena, M. Fieischmann, N.J. Garrard, J. Electroanal. Chem. 220 (1987) 31. C. Amatore, M.R. Deakin, R.M. Wightman, J. Electroanal. Chem. 220 (1987) 49. C. Ciszkowska, Z. Stojek, J. Electroanal. Chem. 213 (1986) 189. K. Borgwarth, D.G. Ebling, J. Heinze, Ber. Bunsen. Phys. Chem. 98 (1994) 1317. J.A. Alden, R.G. Compton, J. Electroanal. Chem. 404 (2004) 27. O.V. Klymenko, R.G. Evans, C. Hardacre, I.B. Svir, R.G. Compton, J. Electroanal. Chem. 571 (2004) 211. N.V. Rees, R.A.W. Dryfe, J.A. Cooper, B.A. Coles, R.G. Compton, S.G. Davies, T.D. McCarthy, J. Phys. Chem. 99 (1995) 7096.