Electrochemistry Communications 8 (2006) 571–576 www.elsevier.com/locate/elecom

Fabrication of disposable ultramicroelectrodes: Characterization and applications Jen-Lin Chang, Jyh-Myng Zen

*

Department of Chemistry, National Chung Hsing University, 250 Kuo-Kuang Road, Taichung 402, Taiwan Received 17 January 2006; received in revised form 31 January 2006; accepted 31 January 2006 Available online 28 February 2006

Abstract We report an easy method for fabricating disposable screen-printed edge band ultramicroelectrodes (designated as SPUMEs) with a built-in three-electrode pattern of alternating printed-layer of carbon, silver, and insulator on a non-conducting polypropylene substrate. Central idea is that the edge of the carbon and/or metal-sandwiched films between the insulator layers can serve as a band type ultramicroelectrode. Simply by slicing the edge of working window, the SPUME dimension can be easily varied (e.g., in the range of 0.18–1.35 mm length with a width of 20 lm in this study). Cyclic voltammograms of ferricyanide in aqueous media displayed low-noise, low-background, sigmoidal responses with virtually no current hysteresis similar to that of a hemicylindrical electrode. The proposed SPUMEs exhibit very low electrical noise and can be reproduced multiple times by repetitively cutting the strip. Since they are cheap and easy for mass production, the disposable nature further offers to application in diverse field of electroanalytical chemistry.  2006 Elsevier B.V. All rights reserved. Keywords: Ultramicroelectrode; Disposable; Screen printed; Iodide; Nitrite

1. Introduction Electrodes of dimensions lower than millimeter size with diffusion-limited process have attracted considerable interest to electrochemist. Their unique electrochemical properties have been used in many applications preferable to electrodes of conventional size [1–5]. The advantages include minimizing of iR drop even in the absence of supporting electrolyte or in highly resistive media, reducing of the double layer capacitance, enhancing of flux and thus current sensitivity, facilitating fast response with a steady state current–potential curve, etc. [6,7]. Indeed these properties can be exploited to many useful research fields, especially in sensor applications [8–18], electron transfer kinetics measurements [2,19–25], and scanning electrochemical microscope (SECM) [26], etc. Nevertheless the fabrication of many types of ultramicroelectrodes (UMEs) *

Corresponding author. Fax: +886 4 22854007. E-mail address: [email protected] (J.-M. Zen).

1388-2481/$ - see front matter  2006 Elsevier B.V. All rights reserved. doi:10.1016/j.elecom.2006.01.028

normally requires tedious and time-consuming instrumental procedures. The as-prepared UMEs are in general expensive and are thus not designed for single-use purpose. In this report, we have utilized the screen printed technique for the fabrication of screen-printed ultramicroelectrodes (SPUMEs). Screen-printed electrode (SPE), in which thin film of metal or carbon ink is printed on a polymeric non-conductive substance, have attracted analytical chemists due to its easy and simple preparation procedures [27,28]. Important advantages of SPEs include low-cost (thus disposable), easy for mass production, and flexible in design. The idea behind our proposed fabrication method is that the edge of the carbon and/or metal-sandwiched film between insulator screen-printed layers can serve as a band type ultramicroelectrode (BUME). Our proposed fabrication method possesses two major advantages. Firstly, the same three-electrode SPUME pattern can be arranged as either: (1) carbon-working, silverquasi reference, silver-counter electrodes (designated as SPUME-1); or (2) silver-working, silver-quasi reference,

572

J.-L. Chang, J.-M. Zen / Electrochemistry Communications 8 (2006) 571–576

carbon-counter electrodes (designated as SPUME-2) for different electroanalytical applications. Note that a trend in the development of sensors for decentralized analysis is to integrate together with the working electrode both reference and counter electrodes on one strip in order to miniaturize and to simplify the instrumentation. In this study, cyclic voltammograms of nitrite and iodide in aqueous media were applied to validate the electrochemical behavior of SPUME-1 and SPUME-2, respectively. Secondly, the dimension, i.e., width (w) and length (l), of the SPUME can be tunable simply by adjusting the screen printed film thickness and breadth, respectively. As reported earlier, band geometry can provide larger currents than disk UME, while maintaining the properties of non-linear diffusion [2,7]. Overall, new design and fabrication of ultramicroelectrodes is a significant and demanding research work in electroanalytical chemistry. Screen printed methods are useful because they provide the ability to reproducibly design multiple-layered electrode systems for a particular UME application. The disposable in nature of the SPUME makes it especially attractive for electroanalysis. 2. Experimental section 2.1. Reagents and chemicals Sodium nitrite, potassium iodide, and potassium ferricyanide were obtained from Sigma (St. Louis, MO, USA). Carbon and silver inks were purchased from Acheson (Tokyo, Japan). Conventional SPEs (3 mm in diameter) in three-electrode configuration were obtained from Zensor R & D (Taichung, Taiwan). All the other compounds (ACS-certified reagent grade) were used without further purification. Aqueous solutions were prepared with deionized water purified using Millipore-Q purification system. 2.2. Apparatus All electrochemical experiments were performed either on a CHI 832 or CHI 8021 electrochemical workstation (Austin, TX, USA). The SPUME in three-electrode configuration consisted of either SPUME-1 (i.e., carbon-working, silver-quasi reference, silver-counter electrodes) or SPUME-2 (i.e., silver-working, silver-quasi reference, carbon-counter electrodes). The physical features, such as electrode topography and edges, were characterized by scanning electron microscopy (SEM). 2.3. Electrode fabrication Different stencil assemblies were first prepared properly to making the multi-layer SPE on a 50 mm · 15 mm polypropylene (PP) base. Fig. 1A shows typical layer-by-layer assembly of the built-in three-electrode system using stencil format in the order of carbon ink ! insulating polymer ! silver ink ! insulating polymer ! silver ink !

Fig. 1. (A) An alternating layer-by-layer pattern structure of the SPUME assembly. (B) Cross-sectional diagram of a typical SPUME with a built-in three-electrode configuration.

insulating polymer. The methodology is similar to the classical SPE preparation method using a semi-automatic screen-printing machine [28]. The as-prepared SPUMEs were 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, as illustrated in Fig. 1B. The procedures can allow for preparing versatile three-electrode SPUME suitably inbetween the insulating polymeric layers. Since the working electrode possesses a triangle-shaped end portion, different cut distance (SE) represents for different SPUME dimensions. 2.4. Electrochemical studies Measurements were all carried out with the SPUME in a three-electrode configuration. The working electrode of the SPUME in three-electrode configuration was electrochemically pretreated in respective pH with the potential window between oxygen and hydrogen gas evolution reaction until the current become constant. It normally took 20 continuous cycles at v = 50 mV/s. Experiments with ferricyanide were studied in pH 2 KCl/HCl using an SPUME-1. Trace analyses with nitrite (in 10 lM H2SO4) and iodide (in pure water) were carried out with SPUME-1 and SPUME-2, respectively.

J.-L. Chang, J.-M. Zen / Electrochemistry Communications 8 (2006) 571–576

3. Results and discussion 3.1. SPUME preparation The SPUME with three-electrode configuration was prepared by cutting the edge of a multi-layer SPE at a position that produced the desired dimension for the analytical purpose. This is an advantageous feature of the screen printed technique for preparing SPUME from a single strip of multi-layer SPE because the SPUME dimension can be easily varied, e.g., in the range of 0.18–1.35 mm length with a width of 20 lm in this study. Of course, the use of machine cutting (instead of manual cutting) can result in a better precision in electrode length. Note that Wang and co-workers reported a relatively complicated SPUME preparation method based on surface-masking and laser photo-ablation technique, where ultramicro-holes were ‘‘drilled’’ on the outer-surface for desired patterns [29]. Williams and co-workers reported microband electrodes fabricated by screen printing of the surface of alumina tiles with conducting and insulating ink materials and by cutting with a diamond saw [30]. The work by Compton and co-workers together with that by Davis and co-workers provided technical details for the fabrication of cheap, disposable ultramicroelectrodes with laminate assembly for a plethora of applications [31,32] Fig. 2A shows the optical picture of a SPUME cut at SE = 3.2 mm. Different layers of the carbon, silver, and silver portions can be clearly seen from the optical microscopic picture. An inbuilt Vernier-Caliper was used for the scaling of the SPUME and the ‘‘l’’ value was measured as 0.18 mm. The length of the bands can be measured by optical spectroscopy; whereas the thickness, i.e., the ‘‘w’’ value, of the microband was determined by high-magnification scanning electron microscopy (SEM). As can be seen in Fig. 2B, the microband electrodes are sandwiched between insulator layers with three electrodes stayed considerably apart from each other in the SEM morphology. The w value of the

573

microband carbon electrode was measured as 20 lM. Note that since the resulting film layer was quite uniform, virtually the same w was observed from SEM analysis of different SE windows with variation in the l values. 3.2. Electrochemical characterization of the SPUME Fig. 3 compares the cyclic voltammograms for the redox reaction of 1 mM ferricyanide on a macroscopic SPE strip in three-electrode configuration and the SPUME-1 at a scan rate of 10 mV/s. As expected, a steady-state response similar to that of the reported metallic UME with a sigmoidal-shaped curve for the SPUME-1 was observed; whereas a more dynamic response with diffusion-controlled characteristics was observed for the macroscopic SPE strip. The half-wave reduction potential of E1/2 = 160 mV observed at the SPUME-1 was found to match with the redox potential of ferricyanide at normal-sized SPE (electrode area = 2.25 mm2). The Nernstian steady-state condition (Tomes criterion) was evaluated with the SPUME-1, in which the (E3/4–E1/4) at steady state voltammetric slope should be 56 mV for a reversible redox reaction [22,33]. Previously, for 1 mM ferrocene redox reaction, the (E3/4– E1/4) values were reported as 53 and 56 mV on platinum (r = 5 lm) and boron-doped diamond (r = 4.72 lm) electrodes, respectively [33]. In this study, based on the sigmoidal-shaped redox curve, the ferricyanide redox reaction showed a value of 54 mV at the SPUME-1 indicating a near ideal behavior on the underlying surface. The macroscopic SPE strip, on the other hand, showed a peak potential separation (DEp) of 95 mV corresponding to the quasireversible electron-transfer with non-ideal Nernstian characteristics. Overall, the Nernstian characteristic with near ideal reversible behavior indicates a proper electrode/electrolyte interface of the proposed SPUME-1 in this study. As reported by both Wightman and Fritsch groups, a band electrode can be modeled more correctly by that of a hemicylindrical electrode [7,34,35]. Theoretical limiting

Fig. 2. Optical (A) and scanning electron (B) microscopic images of the SPUME in three-electrode configuration.

574

J.-L. Chang, J.-M. Zen / Electrochemistry Communications 8 (2006) 571–576

(A) SPE (3 mm dia.)

2 μA

(B) SPUME-1 (w = 20 μm, I = 0.18 mm)

5 nA

-0.1

0.0

0.1

0.2

0.3

0.4

E / V vs pseudo Ag Fig. 3. Cyclic voltammetric responses at a conventional SPE (A) and SPUME-1 (B) for 1 mM ferricyanide in pH 2 KCl/HCl solution at v = 10 mV/s.

current ðiTL Þ of the SPUME-1 can be calculated by using the following expression concerning with hemicylindrical (nonlinear) diffusion model [2,7]: iTL ¼ 2pnFDC FeCN l½1= lnð4hÞ;

ð1Þ

where n is the number of electron involved in the reaction, F is the Faraday constant, D(=7.6 · 106 cm2/s) is the diffusion coefficient of ferricyanide, CFeCN (=1 lmol/cm3) is the concentration of ferricyanide, l (cm) is the length of micro-band, and h = Dt/(w/p)2, where w (cm) and t (=RT/Fv) correspond to the bandwidth and time constant, respectively. Under the standard experimental condition with w = 20 lm and l = 0.18 mm, the iTL value was calculated as 27.93 nA. This is in good agreement to the experimentally observed limiting current (iL) of 27.10 nA for the SPUME-1. The observation particularly supports ideality of the SPUME-1 surface for the hemicylindrical diffusion of the analyte on the carbon microband electrode/electrolyte interface. As mentioned earlier, another advantage of the proposed fabrication method is that different band dimensions can be easily varied by choosing a suitable SE. As shown in Fig. 4, different l value of 1.01, 1.32, and 1.35 mm with w = 20 lM can be obtained by cutting at SE of 4.1, 4.4, and 6.2 mm, respectively. The SE shows a marked alteration initially due to a triangle sharp. Typical CV responses 3=4 of the FeðCNÞ6 redox couple at the as-prepared SPUME-1 with varying band dimension were shown in Fig. 4B. The iL values were found to systematically increase with the increase in l up to SE = 4.4 mm (i.e., the end of the triangular sharp); after that a plateau in the response was noticed. In sum, the SPUME-1 with l ranging from 0.18 to 1.35 mm can be easily prepared with significant non-linear diffusion characteristics. The effect of scan rate (v) on the electrochemical behavior of ferricyanide was further investigated using a same piece of SPUME-1. As reported earlier, at higher scan rates, the observed cyclic voltammograms tends to become more peak shaped due to the increasing contribution from linear diffusion [3,5,14]. This is indeed the result observed

Fig. 4. (A) Plots of limiting current versus SE and l versus SE. (B) Cyclic voltammetric responses of the SPUME-1 (w = 20 lm) with different dimensions for the reduction reaction of 1 mM ferricyanide in pH 2 KCl/HCl at v = 10 mV/s.

J.-L. Chang, J.-M. Zen / Electrochemistry Communications 8 (2006) 571–576

200

0.2

0.4

0.6

0.8

1.0

1.2

(A)

575 0.4

0.8

1.2

1.6

(B)

Limiting current / nA

150

l = 1.32 mm

0.5 μA

20 nA

100

1.01

50

(C)

(D)

0.18

0 0.5

1.0

1.5

2.0

2.5

log (v) / mV/s

on the SPUME-1, as shown in Fig. 5. As can be seen, a 0.18 mm band length SPUME-1 shows almost identical current responses up to v = 50 mV/s; after that a slowly increase in current response is noticed. Note that this trend behavior is quite general for SPUME-1 with different band lengths. Nevertheless, a much longer 1.32 mm band SPUME-1 can retain its ideality only up to v = 20 mV/s. Later a much steeper increase than that of SPUME-1 (l = 0.18 mm) in the current response was noticed due to the change in diffusion mechanism from non-linear to planner process. The characteristics of the SPUME-1 is comparable to those of classical UMEs, e.g., Wightman’s group reported an ideal non-diffusion characteristic with constant iL values for a 5-nm nanoband Au microelectrode in the 3þ scan rate window of 5–1000 mV/s for RuðNH3 Þ6 [7]. But for a 2.3-mm Au-band system a marked alteration in the iL value was observed even at a very slow v of 10 mV/s. Similarly, a 25-lm Pt disc electrodes started to show the diffusion function at v = 100 mV/s for ferrocene [36]. 3.3. Applications We demonstrate here the application of the proposed electrodes in trace analyses. As pointed out by Wightman et al., a decrease in the width of the band electrode should result in a decrease in the residual current with a concurrent improvement in the detection limits [7]. To enhance the current response a relatively longer l = 1.32 mm SPUME with easy-reproducible surface was taken for all analytical experiments. Fig. 6A shows the cyclic voltam-

5 μA

0.1 μA

Fig. 5. Dependence of the maximum current on scan rate for the reduction reaction of 1 mM ferricyanide in pH 2 KCl/HCl at the SPUMEs with different dimensions.

-0.6

-0.4

-0.2

E / V vs pseudo Ag

0.0

-0.4

-0.2

0.0

E / V vs pseudo Ag

Fig. 6. CV responses for the detection of 50 lM nitrite in 10 lM H2SO4 (by using (A) SPUME-1 and (B) conventional SPE) and 300 lM iodide in pure water (by using (C) SPUME-2 and (D) conventional SPE) at v = 50 mV/s.

metric responses for the SPUME-1 with and without 50 lM nitrite at v = 50 mV/s. A well-defined oxidation signal starting from 0.7 V versus pseudo Ag with a limiting current behavior was noticed. The increase in the concentration of nitrite leads to a regular increase in the current oxidation current signal. The voltammogram from the conventional SPE, on the other hand, was largely composed of background current as shown in Fig. 6B. The limiting oxidation current for nitrite at the SPUME-1 showed a wide linear range up to 3 lM. For analyzing 5 lM nitrite, a relative standard deviation of 2.46% (n = 5) indicated a detection limit (S/N = 3) of 0.38 lM. To demonstrate the versatility of the proposed system, the same piece of SPUME was arranged as SPUME-2 (i.e., let silver layer as the working electrode) for iodide determination in pure water. Note that the application of silver electrode for iodide detection has been reported earlier [37]. As performed by cyclic voltammetry in Fig. 6C, a well-defined oxidation signal starting from 0.3 V versus pseudo Ag with a limiting current behavior was again noticed. For comparison, the voltammogram from a conventional screen printed silver electrode in the presence of 0.1 M, pH 7 PBS was also recorded. As can be seen in Fig. 6D, relatively higher

576

J.-L. Chang, J.-M. Zen / Electrochemistry Communications 8 (2006) 571–576

background current was observed. Determination of iodide in the absence of supporting electrolyte using the disposable UMEs is thus possible. Note that the deliberate addition of supporting electrolyte when trace iodide has to be determined introduces contamination from impurities. Because the ohmic losses are negligible, the measurement with UMEs can be performed in solutions of low-ionic strength, for example, natural waters. The other possible application of the disposable UMEs should lie in the area of so-called solid-state voltammetry [38]. Application of disposable UMEs leads to an improvement in the quality of solid-state electrochemical data and provides new diagnostic and analytical possibilities. The analytical aspects of electrochemical studies of solid, rigid or semirigid (non-fluid) systems in the absence of a liquid solution phase can thus be obtained. Research along this line is currently underway in our laboratory. 4. Conclusions A new band type screen-printed ultramicroelectrode (SPUME) with an in-built three-electrode configuration was successfully fabricated by a simple procedure. The working band electrode dimension can be easily tuned with a fixed width simply by slicing the edge of the SPUME. The diffusion characteristic of the new SPUME in threeelectrode configuration was evaluated by using a 3=4 FeðCNÞ6 benchmark system. The SPUME fits very well to the hemicylindrical diffusion model with the obtained experimental value very close to the theoretically predicted limiting current. The same three-electrode SPUME pattern can be arranged as either: (1) carbon-working, silver-quasi reference, silver-counter electrodes; or (2) silver-working, silver-quasi reference, carbon-counter electrodes for sensitive electroanalysis of nitrite and iodide, respectively, by linear scan voltammetry. Since the SPUME is cheap, it can offer a new platform to diverse applications. Acknowledgement The authors gratefully acknowledge financial support from the National Science Council of Taiwan. References [1] M. Fleischmann, S. Pons, D.R. Rolison, P.P. Schmidt, Ultramicroelectrodes, Datatech Systems, Inc. (Edt.), Scienific Publishing Division, 1987.

[2] R.M. Wightman, D.O. Wipf, Voltammetry at Ultramicroelectrodes, in: A.J. Bard (Ed.), Electroanalytical Chemistry, vol. 15, Marcel Dekker, New York, 1988, p. 267. [3] C.G. Zoski, Electroanalysis 14 (2002) 1041. [4] V. Komanicky, W.R. Fawcett, Electrochim. Acta 49 (2004) 1185. [5] J.J. Watkins, B. Zhang, H.S. White, J. Chem. Ed. 82 (2005) 712. [6] J.B. Cooper, A.M. Bond, K.B. Oldham, J. Electroanal. Chem. 331 (1992) 877. [7] K.R. Wehmeyer, M.R. Deakin, R.M. Wightman, Anal. Chem. 57 (1985) 1913. [8] K.T. Kawagoe, J.B. Zimmerman, R.M. Wightman, J. Neurosci. Methods 48 (1993) 225. [9] J.L. Peters, N.V. Kulagina, H. Yang, A.C. Michael, Encyclopedia of Electrochemistry, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany, 2002. [10] M.L.A.V. Heien, M.A. Johnson, R.M. Wightman, Anal. Chem. 76 (2004) 5697. [11] J. Wang, L.H. Wu, R. Li, J. Sanchez, Anal. Chim. Acta 228 (1990) 251. [12] D. Pletcher, E.M. Valdes, Anal. Chim. Acta 246 (1991) 267. [13] J.L. Kawagoe, D.E. Niehaus, R.M. Wightman, Anal. Chem. 63 (1991) 2961. [14] Q. Cheng, T.A. Brajter, Anal. Chem. 68 (1996) 4180. [15] H.M. Koudelka, P.D. van der Wall, Electrochim. Acta 45 (2000) 2437. [16] J. Wang, X. Zhang, Anal. Chem. 73 (2001) 844. [17] B. Fabre, L. Taillebois, Chem. Comm. 24 (2003) 2982. [18] G. Chen, L. Zhang, J. Wang, Talanta 64 (2004) 1018. [19] C.P. Andrieux, P. Hapiot, J.M. Saveant, Chem. Rev. 90 (1990) 723. [20] J.F. Rusling, Z. Wang, A. Owlia, Coll. Surf. 48 (1990) 173. [21] H. Zhang, R.W. Murray, J. Am. Chem. Soc. 113 (1991) 5183. [22] Y. Zhang, C.D. Baer, C. Camaioni-Neto, P. O’Brien, D.A. Sweigart, Inorg. Chem. 30 (1991) 1682. [23] R. Pyati, R.W. Murray, J. Am. Chem. Soc. 118 (1996) 1743. [24] G. Che, S. Dong, Electrochim. Acta 41 (1996) 381. [25] S. Rebouillat, M.E.G. Lyons, A. Flynn, Analyst 125 (2000) 1611. [26] A.J. Bard, M.V. Mirkin (Eds.), Scanning Electrochemical Microscopy, Marcel Dekker, Inc., New York, 2001. [27] J. Wang, Acc. Chem. Res. 35 (2002) 811. [28] J.-M. Zen, A.S. Kumar, in: C.A. Grimes, E.C. Dickey, M.V. Pishko (Eds.), Encyclopedia of Sensors, American Scientific Publisher, 2006, in press. [29] J. Wang, J. Lu, B. Tian, C. Yarnitzky, J. Electroanal. Chem. 361 (1993) 77. [30] D.H. Craston, C.P. Jones, D.E. Williams, Talanta 38 (1991) 17. [31] P.J. Welford, J. Freeman, S.J. Wilkins, J.D. Wadhaman, C.E.W. Hahn, R.G. Compton, Anal. Chem. 73 (2001) 6088. [32] N.G. Karousos, R. Carrington, S.J. Wilkins, C. Livingstone, J. Davis, Electrochem. Commun. 7 (2005) 500. [33] J.B. Cooper, S. Pang, S. Albin, J. Zheng, R.M. Johnson, Anal. Chem. 70 (1998) 464. [34] M.P. Nagale, I. Fritsch, Anal. Chem. 70 (1998) 2902. [35] M.P. Nagale, I. Fritsch, Anal. Chem. 70 (1998) 2908. [36] S. Ching, R. Dudek, E. Tabet, J. Chem. Ed. 71 (1994) 602. [37] H. Parham, B. Zargar, Anal. Chim. Acta 464 (2002) 115. [38] P.J. Kulesza, J.A. Cox, Electroanalysis 10 (1998) 73.