Anal. Chem. 2001, 73, 758-762
Palladium Film Decoupler for Amperometric Detection in Electrophoresis Chips Der-chang Chen,† Feng-Liu Hsu,† Dian-Zhen Zhan,† and Chun-hsien Chen*,†,‡
Department of Chemistry, National Sun Yat-Sen University, Kaohsiung, Taiwan 80424, R.O.C. and Department of Chemistry, National Tsing Hua University, Hsinchu, Taiwan 300, R.O.C.
Demonstrated in this article is that a palladium metal film can be applied to decouple the electric circuitry of electrochemical detection from that of the electrophoretic separation in an electrophoresis chip. The Pd solid-state field decoupler, as well as the working electrodes, is thermally evaporated onto the plastic chip and oriented vertically across the separation channel. After the sample zones flow over the Pd decoupler, their electrochemical response is measured at working electrodes in the downstream pathway. Because the electrodes are on the separation channel, the electrode channel alignment is no longer a problem. For a separation channel of roughly 200 µm in width and 75 µm in depth in 10 mM phosphate (pH 5.1), the noise level at the working electrode is < 15 pA at an electric field of 570 V/cm. For over a decade, studies of CE-EC (capillary electrophoresis coupled with electrochemical detection) have demonstrated the inherent advantages of EC detection with its relatively low cost and high sensitivity.1-3 However, CE-EC is still not commercially available in analytical instrumentation markets. Two concerns impede the commercialization. First, high electric field and electrophoretic current interfere with the measurements of the EC current at the working electrode, resulting in a raised EC background current and shifted redox potential for analytes.4-6 The degree of interference is associated with the concentration and conductivity of the electrolyte within the separation channel. When the electrophoretic current is large, it is necessary to introduce an interface to decouple the EC detection and the electrophoresis circuitry so as to minimize the interference and to prevent the EC detector from damage due to surge spikes.7 Decouplers can be made from porous glass or graphite,7,8 * Corresponding author: Phone: +886 3 573 7009. Fax: +886 3 571 1082. E-mail:
[email protected]. Current address: National Tsing Hua University. † National Sun Yat-Sen University. ‡ National Tsing Hua University. (1) Voegel, P. D.; Baldwin, R. P. Electrophoresis 1997, 18, 2267-2278. (2) Holland, L. A.; Lunte, C. E. Anal. Commun. 1998, 35, 1H-4H. (3) Kappes, T.; Hauser, P. C. J. Chromatogr. A 1999, 834, 89-101. (4) Lu, W. Z.; Cassidy, R. M. Anal. Chem. 1994, 66, 200-204. (5) Matysik, F.-M. J. Chromatogr. A 1996, 742, 229-234. (6) Wallenborg, S. R.; Nyholm, L.; Lunte, C. E. Anal. Chem. 1999, 71, 544549. (7) Wallingford, R. A.; Ewing, A. G. Anal. Chem. 1987, 59, 1762-1766. (8) Yik, Y. F.; Lee, H. K.; Li, S. F. Y.; Khoo, S. B. J. Chromatogr. 1991, 585, 139-144.
758 Analytical Chemistry, Vol. 73, No. 4, February 15, 2001
polymeric materials,9-11 etched capillary,12-14 and Pd-metal tubing.15-17 However, the use of most decouplers is troubled by the inconvenience and difficulty in preparation and by inherent problems such as fragility, leakage of analytes, and dead volume. Alternatively, for capillaries with an inner diameter of 50 cm) leads to potentiostat, and delicate metal films require dexterity to prevent the detection unit from film abrasion, quivering, tension due to lead dragging, or loss of electric contact. The ultimate answer to the problems seems to be miniaturization of CE-EC into a “lab-on-a-chip” format.26-30 In combination with the strength of microfabrication techniques, integrated CE-EC devices has great potential for commercialization because it is robust, compact, user-friendly, and suitable for mass production. Interests in applications of microfluid devices in separations have been intense recently. Fabrication methods of CE chips are well documented for a variety of materials, including glass,30-33 silicon,34 and elastomers.28,35-38 Naturally, for integration of EC detection into CE microchips, the tactic of deposition of the working electrode onto the separation outlet is incorporated quite well with the microfabrication processes.27-30 However, so far, the decoupler has not been developed for CE-EC microchips and, thus, only narrow channels can be utilized. Introduced herein is the first lab-on-a-chip containing a decoupler, which further perfects the commercialization of CE-EC. Similarly to the preparation of working electrodes on CE micro(26) Wang, J.; Tian, B. M.; Sahlin, E. Anal. Chem. 1999, 71, 5436-5440. (27) Wang, J.; Tian, B. M.; Sahlin, E. Anal. Chem. 1999, 71, 3901-3904. (28) Martin, R. S.; Gawron, A. J.; Lunte, L. M.; Henry, C. S. Anal. Chem. 2000, 72, 3196-3202. (29) Henry, C. S.; Zhong, M.; Lunte, S. M.; Kim, M.; Bau, H.; Santiago, J. J. Anal. Commun. 1999, 36, 305-307. (30) Woolley, A. T.; Lao, K. Q.; Glazer, A. N.; Mathies, R. A. Anal. Chem. 1998, 70, 684-688. (31) Jacobson, S. C.; Koutny, L. B.; Hergenroder, R.; Moore, A. W.; Ramsey, J. M. Anal. Chem. 1994, 66, 3472-3476. (32) Effenhauser, C. S.; Manz, A.; Widmer, H. M. Anal. Chem. 1993, 65, 26372642. (33) Harrison, D. J.; Manz, A.; Fan, Z.; Ludi, H.; Widmer, H. M. Anal. Chem. 1992, 64, 1926-1932. (34) Burns, M. A.; Johnson, B. N.; Brahmasandra, S. N.; Handique, K.; Webster, J. R.; Krishnan, M.; Sammarco, T. S.; Man, P. M.; Jones, D.; Heldsinger, D.; Mastrangelo, C. H.; Burke, D. T. Science 1998, 282, 484-487. (35) Martynova, L.; Locascio, L. E.; Gaitan, M.; Kramer, G. W.; Christensen, R. G.; Maccrehan, W. A. Anal. Chem. 1997, 69, 4783-4789. (36) Locascio, L. E.; Perso, C. E.; Lee, C. S. J. Chromatogr. A 1999, 857, 275284. (37) Chen, Y.-H.; Wang, W.-C.; Young, K.-C.; Chang, T.-T.; Chen, S.-H. Clin. Chem. 1999, 45, 1938-1943. (38) Wang, S.-C.; Perso, C. E.; Morris, M. D. Anal. Chem. 2000, 72, 17041706.
chips, a Pd film can be easily deposited across the microfluid channel and serve as the decoupler. The application and rationale of the Pd-metal decoupler have been elucidated by Kok and Sahin.15 In brief, during electrophoretic separation, the high electric field is sufficient for the electrolysis of water, which may cause gas evolution at the ground electrode and interfere with the EC signal.39 For the Pd decoupler onto which hydrogen ion is reduced and adsorbed, the unique property is that hydrogen can diffuse relatively fast on a Pd surface. Therefore, before gas bubbles can develop, hydrogen is removed from the palladium decoupler by the solution due to the electroosmotic flow.15 Taking advantage of the decoupler, single or multiple working electrodes can be placed across the separation channel. Alignment between the outlet of the separation channel and the working electrode is no longer a concern. We will demonstrate in this article that such a decoupler can effectively isolate EC current from electrophoretic current for channels wider than 200 µm. EXPERIMENTAL SECTION CE-EC Chip Device. The scheme displayed in Figure 1A is an example of the device configuration. The separation channel was wire-imprinted on a piece of Plexiglas whose width, length, and thickness were nominally 15 mm, 65 mm, and 2 mm, respectively. The imprinting and bonding procedures developed by Locascio et al.35-37 were adapted and are described briefly in the following. A 90-µm-i.d. tungsten wire (California Fine Wire Co., Grover Beach, CA) was used to imprint the channels. The tungsten wire was placed between a piece of Plexiglas and a glass slide. Heavy stainless steel blocks were utilized to apply imprinting pressure at 175 °C overnight. The imprinted channel was trianglelike, typically 200 µm in width and 75 µm in depth (Figure 1B). The channels generally exhibited a few dent-like defects (Figure 1B), which probably resulted from statically adsorbed dust particles. A second channel for sample loading was imprinted on another piece of Plexiglas with width, length, and thickness of 30 mm, 65 mm, and 2 mm, respectively. The Pd decoupler and the working electrodes were then thermally deposited on the second piece of Plexiglas in a bell-jar evaporator (KV-301, KEY High Vacuum Co., (39) Slater, J. M.; Watt, E. J. Analyst 1994, 119, 2303-2307.
Analytical Chemistry, Vol. 73, No. 4, February 15, 2001
759
Nesconset, NY). A piece of Al foil was used as the mask and was wrapped on the Plexiglas. The pattern on the Al foil for deposition of metal films was trimmed manually with an exacto knife. As depicted in Figure 1A, the sample loading channel and the films were configured perpendicularly to the separation channel. The base pressure, deposition rate, and film thickness were about 5 × 10-6 Torr, 0.5 nm/s, and 150 nm, respectively. Adhesion enhancement, such as placing chromium or titanium interlayers between the electrodes and Plexiglas, was unnecessary. The two pieces of Plexiglas were thermally bonded together.35-37 Prior to bonding, four 2-mm-i.d. holes were drilled through the first piece of Plexiglas. These holes were aligned with the channels at the appropriate position. After bonding, four glass tubes (1 cm in length) were epoxy-glued (Epoxi-Patch, Dexter, Seabrook, NH) with the four holes to create buffer, analyte, and two waste reservoirs. Apparatus. A high-voltage dc power supply (model CZE1000R, Spellman High-Voltage Electronics Corp., Plainview, NY) was used for the electrophoretic separation and sample loading. A digital dual-output power supply (model E3620a, Hewlett-Packard Co.) was used to regulate the output voltage of the high-voltage power supply so as to improve the reproducibility of the experiments. Amperometric measurements were performed using an LC-4C potentiostat (Bioanalytical Systems, West Lafayette, IN) that was connected to a Pentium 100 MHz PC equipped with data acquisition software ChemLab (Scientific Information Service Corp., Taipei, Taiwan, ROC). The reference electrode was a Ag wire coated with chloride as a quasi-reference electrode. The counter-electrode was a Pt wire. Both of the electrodes were placed in the buffer waste reservoir. Electrophoretic current was determined by connecting a multimeter (model 34401A, HewlettPackard Co.) in series with the Pd decoupler and collected by data acquisition software (HP VEE, Hewlett-Packard Co.). Reagents. All of the reagents for the preparation of buffer solutions were analytical reagent grade and were purchased from Sigma or Aldrich. Solutions were prepared with purified water (18 MΩ-cm, Millipore-Q, Millipore Inc.). The electrolyte solutions were MES (2-[N-morpholino]ethanesulfonic acid, Sigma) or phosphate buffered to desired pH by dropwise introduction of 5 M NaOH (semiconductor grade, Aldrich). The reported pH of the solution was carefully measured with a pH meter (model SP2200, Suntex, Taiwan, R.O.C.). Standard solutions of catecholamines of desired concentrations were made fresh daily by serial dilution in the separation electrolyte. Because the channel size was large, filtration of solutions prior to introduction to the microchip reservoirs was unnecessary. Electrophoresis Procedures. Prior to electrophoresis, the channels were flushed with 1 M NaOH for 20 min; with deionized water for 10 min; with 1% HCl for 20 min;27 with deionized water for 10 min; and finally, with the separation electrolyte for 20 min. Before repetitive runs got started, the channel for sample injection (the channel between wells 3 and 4 in Figure 1A) was filled electrokinetically with the sample solution. To ensure that the sample fills the cross section, the initial drive must be long enough. The duration is typically >5 min. During this period, analytes at the cross section diffused into the separation channel. Therefore, the results measured in the first run should be discarded. Thus, a high voltage was then applied between the 760 Analytical Chemistry, Vol. 73, No. 4, February 15, 2001
Figure 2. Electropherograms of (A) 0.10 mM dopamine in 10 mM phosphate (pH 5.2) and (B) 4.0 µM dopamine and 20 µM catechol in 10 mM MES (pH 5.7). Conditions: strength of electric field, 225 V/cm; amperometric detection at +0.4 V vs EAg/AgCl.
buffer reservoir (well 1) and the Pd decoupler until the EC current returned to baseline, which indicated the passage of the analytes over the working electrodes.26-30,37 Subsequent sample loading and separation were repetitively performed by applying an electric field of 225 V/cm across the sample loading channel for 5 s and then applying a separation voltage between the buffer reservoir and the Pd decoupler. During sample loading, the electrodes for electrochemical detection were disconnected by switching the potentiostat to the standby position. RESULTS AND DISCUSSION Dopamine and catechol were chosen to manifest the feasibility of the decoupler-integrated CE-EC microchips, because catecholamines are among the most common substances discussed in CE-EC literatures. An electropherogram of 0.10 mM dopamine in 10 mM phosphate buffer (pH 5.2), shown in Figure 2A, is representative of CE-EC operation in electrolytes having high conductivity. Figure 2B shows amperometric signals of 4.0 µM dopamine and 20 µM catechol obtained in 10 mM MES (pH 5.7), which are representative of electropherograms for analytes at a relatively low conductivity level. The electrophoretic currents for Figure 2A and B are 39 µA and 0.5 µA, respectively, more than 3 orders of magnitude larger than the corresponding EC signal. The peak-to-peak noise levels measured under the conditions of Figure 2A and B are about 100 pA and 10 pA, respectively. Although such detection noise is not so small as literature values,11,12,15 the electropherograms, obtained from a relatively large channel, demonstrate that a Pd film can, to some extent, decouple the circuitry of EC detection from that of the separation electric field. The noise level of the background current in electrolytes having lower conductivity appears to be smaller than those exhibited in more conductive electrolytes. This may be a result of the formation of small hydrogen bubbles in more conductive media, but it may also be attributed to the incomplete decoupling of the electrophoretic current.17 There have been numerous reports illustrating the fact that applied CE voltage causes the shift in half-wave potential of redox species for end-column detection.5,6,40 Hydrodynamic voltammo(40) Gerhardt, G. C.; Cassidy, R. M.; Baranski, A. S. Anal. Chem. 1998, 70, 2167-2173.
Figure 3. Influence of the separation voltage on background amperometric current. From bottom to top, the strength of the electric field increases from 0 to 714 V/cm in 143 V/cm steps. Other conditions are the same as for Figure 2A.
grams (HDVs) are generally employed for demonstration of such an effect by the parallel comparison of half-wave potentials with and without the decoupler. For the Pd-film integrated CE-EC chips, HDVs without using the decoupler are apparently not attainable, nor is the parallel comparison. Nonetheless, to set the working electrode at an ideal potential in this study, HDVs (not shown) are performed by tabulating the amperometric response of the catecholamines at potentials varied from -0.1 V to +0.6 V in 0.1 V steps. The results show that the slopes of the voltammetric waves are generally sharper than HDVs without decouplers depicted in similar studies.6,26,27 Additionally, the diffusion-limited plateaus of the HDVs are reached at less positive potentials. For example, at electric field strengths of 236 V/cm in 10 mM MES (pH 5.7), the current of both HDVs for dopamine rises at 0.0 V and levels off around 0.4 V. The half-wave potential is 0.2 V, ca. 0.1 V less positive than that obtained by on-column detection under similar conditions.27 The HDVs measured at 472 V/cm are essentially identical, which suggests that the working electrode can be decoupled effectively from that of the separation circuitry. Without an exhausting optimization of separation conditions, an amperometric signal of
