Simultaneous Measurement of Dopamine and Ascorbate at Their

Chaoxiong Ma , Nicholas M. Contento , Larry R. Gibson , II , and Paul W. Bohn ..... Zhou Xucheng , Wang Kun , Zou Xiaobo , Shi Jiyong , Huang Xiaowei ...
0 downloads 0 Views 85KB Size
Download PDF
Anal. Chem. 2001, 73, 1196-1202

Simultaneous Measurement of Dopamine and Ascorbate at Their Physiological Levels Using Voltammetric Microprobe Based on Overoxidized Poly(1,2-phenylenediamine)-Coated Carbon Fiber Jian-Wei Mo† and Bozˇidar Ogorevc*

Analytical Chemistry Laboratory, National Institute of Chemistry, P.O. Box 3430, SI-1001 Ljubljana, Slovenia

Overoxidized poly-(1,2-phenylenediamine) (OPPD)-coated carbon fiber microelectrodes (CFMEs) exhibit, in combination with square-wave voltammetry (SWV) detection mode, the attractive ability to simultaneously measure low nM dopamine (DA) and mM ascorbate (AA) in a pH 7.4 medium. The PPD polymer film is electrodeposited onto a carbon fiber at a constant potential of 0.8 V versus Ag/AgCl using a solution containing sodium dodecylsulfate as the dopant. After overoxidation using cyclic voltammetry (CV) in the potential range from 0 to 2.2 V at a scan rate of 10 V/s, the resulting OPPD-CFME displays a high SWV current response to cationic DA at ∼0.2 V and has a favorably low response to anionic AA at ∼0.0 V vs Ag/AgCl. The preparation of the new OPPDsensing film has been carefully studied and optimized. The OPPD properties and behavior were characterized using CV and SWV under various conditions and are discussed with respect to DA and AA detection. The linear calibration range for DA in the presence of 0.3 mM AA is 50 nM to 10 µM, with a correlation coefficient of 0.998 and a detection limit of 10 nM using 45-s accumulation. The detection limit for DA in the absence of AA was estimated to be 2 nM (S/N ) 3). The linear range for AA in the presence of 100 nM DA is 0.2-2 mM, with a correlation coefficient of 0.999 and a detection limit of 80 µM. The reproducibilities of SWV measurements at OPPD-CFCMEs are 1.6% and 2.5% for 100 nM DA and 0.3 mM AA, respectively. Potential interfering agents, such as 3,4dihydroxyphenylacetic acid, uric acid, oxalate, human serum proteins, and glucose, at their physiologically relevant or higher concentrations did not have any effect. These favorable features offer great promise for in vitro and in vivo application of the proposed OPPD-coated microprobe. Dopamine (DA) is one of the most significant catecholamines and belongs to the family of excitatory chemical neurotransmitters.1 It plays a very important role in the functioning of the central * Corresponding author. E-mail: [email protected]. † Present address: Department of Chemistry and Biochemistry, New Mexico State University, Las Cruces, NM. (1) Smith, T. E. In Devlin, T. M., Ed.; Textbook of Biochemistry with Clinical Correlations; Wiley-Liss: New York, 1992; p 929.

1196 Analytical Chemistry, Vol. 73, No. 6, March 15, 2001

nervous, cardiovascular, renal, and hormonal systems, as well as in drug addiction and Parkinson’s disease.2-4 Furthermore, neurobiological investigations have suggested that DA system dysfunction plays a critical role in some clinical manifestations of HIV infection.5 DA has, therefore, been given tremendous consideration in biomedically-oriented research, due to which there is a strong need for sensitive, selective, and reliable methods for the direct measurement of DA. Because DA is an electrochemically active (oxidizable) compound, its electrochemical detection, particularly in combination with the employment of microelectrodes, has long been wellexploited and -explored.6-9 However, the following primary challenges to measuring DA under physiological conditions utilizing electrochemical methods still remain: (i) the very low concentration levels of DA (in the nM range), and (ii) the intensive interference arising from the electroactive ascorbate (AA) that is present at relatively high concentrations (0.2-0.5 mM), as well as from the presence of the 3,4-dihydroxyphenylacetic acid (DOPAC), which is a major DA metabolite (1-20 µM). This interference is due to the very close oxidation potentials of DA, AA, and DOPAC at the unmodified electrodes.10,11 One possible, and up to now the most exploited, strategy to overcome these problems has been to cover the electrode surface with a negatively charged film.12,13 Several reports have demonstrated that films such as Nafion,14,15 clay,16 overoxidized polypyrrole,17,18 poly(2) Velasco, M.; Luchsinger, A. Am. J. Ther. 1998, 5, 37-43. (3) Mascia, A.; AÄ fra, J.; Schoenen, J. Cephalalgia 1998, 18, 174-182. (4) Damier, P.; Hirsch, E. C.; Agid, Y.; Graybiel, A. M. Brain 1999, 122, 14371448. (5) Lopez, O. L.; Smith, G.; Meltzer, C. C.; Becker, J. T. Neuropsy. Neuropsy. Behav. Neurol. 1999, 12, 184-192. (6) Adams, R. N. Anal. Chem. 1976, 48, 1126A-1137A. (7) Voltammetric Methods in Brain Systems; Boulton, A. A., Baker, G. B., Adams, R. N., Eds.; Humana Press: Totowa, NJ, 1995. (8) Kume-Kick, J.; Rice, M. E. J. Neurosci. Methods 1998, 84, 55-62. (9) Runnels, P. L.; Joseph, J. D.; Logman, M. J.; Wightman, R. M. Anal. Chem. 1999, 71, 2782-2789. (10) Parsons, L. H.; Justice, J. B., Jr. J. Neurochem. 1992, 58, 212-218. (11) O’Neill, R. D. Analyst 1994, 119, 767-779. (12) Wiedemann, D. J.; Kawagoe, K. T.; Kennedy, R. T.; Ciolkowski, E. L.; Wightman, R. M. Anal. Chem. 1991, 63, 2965-2970. (13) Kawagoe, K. T.; Zimmerman, J. B.; Wightman, R. M. J. Neurosci. Methods 1993, 48, 225-240. (14) Gerhardt, G. A.; Oke, A. F.; Nagy, G.; Moghaddam, B.; Adams, R. N. Brain Res. 1984, 290, 390-395. (15) Lacroix, M.; Bianco, P.; Lojou, E. Electroanalysis 1999, 11, 1068-1076. (16) Zen, J.-M.; Chen, P.-J. Anal. Chem. 1997, 69, 5087-5093. 10.1021/ac0010882 CCC: $20.00

© 2001 American Chemical Society Published on Web 02/10/2001

thiophene,19 and others20 at a physiological pH of 7.4 attracted and could even preconcentrate the cationic DA while effectively rejecting the negatively charged AA and other anionic interfering agents. However, the dip- and cast-coating procedures used for depositing, for example, Nafion and clay films permit a poor preparation control, which results in low repeatability and nonuniformity of such films. Electropolymerization procedures, on the other hand, allow the precise development of thin and uniform films and enable a high degree of geometrical conformity, which is particularly important for the modification of microelectrodes.17-19,21 However, the physiological interactions between AA and DA have recently been given increased attention by the pharmacological and neurosciences. It is assumed that these interactions can modify the biological activities of AA and DA and, under certain conditions, can induce significant neurophysiological changes such as nerve disorders (e.g., Parkinson’s disease) and DA-induced cytotoxicity.22-24 It is, therefore, obvious that the simultaneous and direct determination of DA and AA in extracellular fluids is a topic of great interest in the context of further studies in this direction. For such in vivo or in vitro measurements, appropriately modified microelectrodes are certainly the “tool of choice”,13 where properly designed and biocompatible substrate microtransducers are also important.25,26 In the early 80s, Gonon et al. reported the application of surface-treated microelectrodes to simultaneously measure AA and catecholamines. By preanodizing carbon fibers at high potentials (up to +3 V) a sufficient separation of voltammetric peaks was achieved, due to a negative potential shift in AA oxidation (ca 150 mV).27,28 A similar approach involving two microelectrodes, one of which was preanodized and the other was an untreated bare carbon fiber, was used for simultaneous amperometric measurement of AA and catecholamine cell secretion.29 In efforts to achieve simultaneous voltammetric measurement of DA and AA, several authors investigated different polymer coatings30-32 or introduced new carbon electrode materials.33 With the conventionally sized electrodes used in all of these studies, the voltammetric resolution obtained was good, but some of the approaches either showed a (17) Zhang, X.; Ogorevc, B.; Tavcar, G.; Grabec Sˇ vegl, I. Analyst 1996, 121, 1817-1822. (18) Pihel, K.; Walker, Q. D.; Wightman, R. M. Anal. Chem. 1996, 68, 20842089. (19) Gao, Z.; Yap, D.; Zhang, Y. Anal. Sci. 1998, 14, 1059-1063. (20) Downard, A. J.; Roddick, A. D.; Bond, A. M. Anal. Chim. Acta 1995, 317, 303-310. (21) Bartlett, P. N.; Cooper, J. M. J. Electroanal. Chem. 1993, 362, 1-12. (22) Sakagami, H.; Satoh, K.; Ida, Y.; Hosaka, M.; Arakawa, H.; Maeda, M. Free Radical Biol. Med. 1998, 25, 1013-1020. (23) Si, F.; Ross, G. M.; Shin, S. H. Exp. Brain Res. 1998, 123, 263-268. (24) Shin, S. H.; Si, F.; Chang, A.; Ross, G. M. Am. J. Physiol. 1997, 273, E593E598. (25) Zhang, X.; Ogorevc, B. Anal. Chem. 1998, 70, 1646-1651. (26) Zhang, X.; Ogorevc, B.; Rupnik, M.; Kreft, M.; Zorec, R. Anal. Chim. Acta 1999, 378, 135-143. (27) Gonon, F.; Buda, M.; Cespuglio, R.; Jouvet, M.; Pujol, J.-F. Nature 1980, 286, 902-904. (28) Gonon, F. G.; Fombarlet, C. M.; Buda, M. J.; Pujol, J. F. Anal. Chem. 1981, 53, 1386-1389. (29) Cahill, P. S.; Wightman, R. M. Anal. Chem. 1995, 67, 2599-2605. (30) Erdogdu, G.; Mark, H. B., Jr.; Karagoezler, E. Anal. Lett. 1996, 29, 221231. (31) Sun, Y.; Ye, B.; Zhang, W.; Zhou, X. Anal. Chim. Acta 1998, 363, 75-80. (32) Ciszewski, A.; Milczarek, G. Anal. Chem. 1999, 71, 1055-1061. (33) Miyazaki, K.; Matsumoto, G.; Yamada, M.; Yasui, S.; Kaneko, H. Electrochim. Acta 1999, 44, 3809-3820.

less favorable detection capability30 or required a nonphysiological pH of the medium.31 Despite these great achievements, it seems that further efforts are needed to introduce a suitable voltammetric microprobe for simple, selective, and reliable simultaneous measurement of DA and AA at their physiological concentrations. Various poly-phenylenediamine electrode coatings have been successfully exploited, mainly in the fabrication of biosensors, due to their superior ability to immobilize enzymes, their good suppression of AA interference, and their excellent influence on the signal-to-noise ratio.34,35 A poly-(1,2-phenylenediamine) (PPD)coated electrode has been used for in vivo detection of nitric oxide.36 Notably, in that study, the PPD-coated sensor was reported to also exhibit a high selectivity for NO against DA. No other papers seem to have reported on the application of PPD for the electrochemical sensing of neurotransmitters. However, our preliminary experiments have revealed that by an appropriate anodization of the electropolymerized PPD coating, a different, overoxidized PPD (OPPD) film is created with completely new sensing capabilities, including simultaneous detection of lownanomolar DA and low-millimolar AA at a physiological pH. Its preparation, characterization and potential applications are described in the following sections. EXPERIMENTAL SECTION Apparatus. Cyclic voltammetry (CV) and square-wave voltammetry (SWV) were performed using a modular electrochemical system (Autolab, Eco Chemie, Utrecht, The Netherlands) that was equipped with PGSTAT10 and ECD modules and driven by GPES software (Eco Chemie). All measurements were conducted in a conventional one-compartment voltammetric cell that was placed in a well-grounded Faraday cage. The three-electrode configuration consisted of a modified or bare carbon-fiber microelectrode (CFME), a platinum wire, and Ag/AgCl,KClsatd as the working, counter, and reference electrodes, respectively. All potentials quoted in this work are referred to versus Ag/AgCl,KClsatd as the reference. All electrochemical experiments were carried out at controlled room temperature (22 ( 1 °C). Chemicals. Dopamine, ascorbic acid, 3,4-dihydroxyphenylacetic acid, 1,2-phenylenediamine (PD), sodium dodecyl-sulfate (SDS) and potassium hexacyanoferrate(III) were obtained from Sigma. All reagents were of analytical reagent grade and were used as received. Deionized water further purified via a Milli-Q unit (Millipore, Bedford, MA) was employed to prepare all solutions. A DA standard solution was freshly prepared for daily use. A 0.05 M phosphate buffer solution was prepared using NaH2PO4 and Na2HPO4 obtained from Merck. Fabrication and Modification of CFMEs. Glass capillary tubes (Euroglass, Ljubljana, Slovenia) were pulled with a PP-830 pipet puller (Narishige, Tokyo, Japan), to form very thin capillaries with a ∼10-µm i.d. The fabrication of a cylinder CFME was carried out according a standard procedure by inserting a cleaned carbon fiber (7 µm in diameter, Goodfellow, Oxford, UK) mounted on the end of a copper wire (0.2 mm in diameter) into a capillary under an inverted microscope (Eclipse, Nikon, Tokyo, Japan) so that ∼0.3 mm of the fiber tip was left protruding. After optical (34) McAteer, K.; O’Neill, R. D. Analyst 1996, 121, 773-777. (35) Dumont, J.; Fortier, G. Biotechnol. Bioeng. 1996, 49, 544-552. (36) Friedemann, M. N.; Robinson, S. W.; Gerhardt, G. A. Anal. Chem. 1996, 68, 2621-2628.

Analytical Chemistry, Vol. 73, No. 6, March 15, 2001

1197

and voltammetric testing and corresponding selection, CFMEs were ready for modification. More details are given elsewhere.37 The poly-(1,2-phenylenediamine) (PPD) film was electrodeposited onto a CFME in a pH 7.0 phosphate buffer solution containing 5 mM PD monomer and 0.1 M SDS, at +0.8 V for 20 min. After being rinsed thoroughly with water, the PPD film was electrochemically overoxidized (anodized) using cyclic voltammetry at a scan rate of 10 V/s in the potential range of 0.0 to +2.2 V for 50 cycles in a pH 7.0 phosphate buffer. After this, the overoxidized PPD (OPPD)-modified CFMEs were immediately ready to simultaneously measure DA and AA. The steady-state signals were obtained without any pretreatment or conditioning. Procedure. The SWV settings for the simultaneous measurement of DA and AA were frequency, 10 Hz; step potential, 10 mV; pulse amplitude, 50 mV; initial potential, -0.15 V; and final potential, +0.50 V. The accumulation potential and time were -0.2 V and 10 s for the evaluation of the OPPD film or 45 s for the simultaneous measurement of nM DA and mM AA, respectively.

Figure 1. Effects of A, anodization (vertex) potential and B, anodization time (in terms of the number of CV runs) during the CV overoxidation treatment of the PPD coating on the SWV measurement parameters: peak currents of a, 1 µM DA, and b, 1 mM AA, and c, peak resolution (∆Ep) at an OPPD-CFME where ∆Ep ) Ep,DA - Ep,AA. Overoxidation conditions: medium, pH 7.0 phosphate buffer; CV scan rate, 10 V/s; no. of CV runs in A, 50; anodic vertex potential in B, 2.2 V. Measurement medium, 0.05 M phosphate buffer (pH 7.4); accumulation conditions, 10 s at -0.2 V.

RESULTS AND DISCUSSION Preparation of the OPPD Coating for DA and AA Sensing. The electrosynthesis of the PPD film onto the CFME surface was studied and optimized in view of its further treatment (overoxidation) and the optimum sensing capability of the final OPPD layer by varying the polymerization potential (from +0.6 to +1.0 V), polymerization time (5-60 min), pH of the polymerization solution (3-8), monomer concentration (2-10 mM), and dopant (e.g., SDS, ferricyanide). The optimum polymerization potential and time, the major factors in tailoring the thickness of the polymer film, were selected to be +0.8 V and 20 min, respectively. To achieve the desired permselectivity of the membrane, that is, its porosity that can be partly controlled through the pinhole size and density, it was essential to choose a suitable dopant, which is a primary impact factor for the creation of pinholes, as well as to carefully modulate the experimental conditions of PPD polymerization and its overoxidation. After intensive experimentation, SDS at a concentration of 0.1 M was found to be the most suitable dopant with respect to optimum DA and AA detection. It has been reported that the anodization at high positive potentials (overoxidation) of some polymer electrode coatings can result in an increase in film porosity and the amount of negative charge, which might eventually enhance the sensitivity and selectivity of DA detection.17-19 This prompted us to investigate the overoxidation of the PPD film for the same purposes. Figure 1A shows the effects of the anodic vertex potential variation at a fixed initial potential (0.0 V), using a CV overoxidation procedure, on the simultaneous voltammetric detection of 1 µM DA and 1 mM AA. SWV mode was employed due to its distinct advantage (see below). Clearly, the sensor did not respond to DA and AA after application of anodization (vertex) potentials below 1.5 V, which indicates an insulating character of the so-treated PPD film. Upon increasing the anodization potential, the SWV peak currents of DA and AA were considerably enhanced and reached a maximum at 2.2 V (Figure 1A, curves a, b). The peak resolution also reached the maximum at 2.2 V (curve c). At anodization potentials beyond 2.3 V, the current of AA increased steeply and

overlapped that of DA. From these results, it is assumed that at the potentials >1.5 V, the PPD film gradually becomes negatively charged (overoxidized) through the introduction of oxygen-rich groups (e.g., carboxylic). This results in the anionic dopant SDS being expelled from the film, which leads to increased porosity in the film, which is in turn reflected in the appearance of the current signals. Moreover, at higher positive potentials, an additional breaking of some fragments in the OPPD frame leads to certain structural defects in the film.17,19 The observed behavior (Figure 1) may be attributed to the specific permeability and accumulation of the cationic DA in the OPPD film. In contrast, a major fraction of anionic AA at the surface is rejected via electrostatic forces while, on the other hand, part of the AA can apparently penetrate through cleavage holes and larger pores into the film and may interact with hydrophilic oxygen-rich groups.19,32,38 However, at overoxidation potentials > 2.4 V, the extent of porosity and/or the OPPD skeletal damage becomes excessive. In addition, too high anodization potentials cause mechanical instability in the polymer, which induces a substantial decrease in the adhesion of the polymer to the carbon fiber, resulting in the film’s pealing from the surface. For example, AA demonstrated similar electrochemical behavior at both the PPD-coated and bare CFMEs after they had been anodized at 2.6 V (not shown). In using cyclic voltammetry, the effective overoxidation time was actually the time when the potential was scanned between 1.5 V and the upper anodic vertex potential of a CV run, that is, 2.2 V at optimum. The anodization time was, therefore, simply expressed in terms of the number of CV runs, and its effects are presented in Figure 1B. As shown, the optimum conditions were reached at 50 CV runs, when the actual overoxidation time, at a scan rate of 10 V/s, is exactly 7 s. The similarity in the behavior displayed in Figure 1A, B implies a complementarity of anodization

(37) Mo, J.-W.; Ogorevc, B.; Zhang, X.; Pihlar, B. Electroanalysis 2000, 12, 4854.

(38) Dalmia, A.; Liu, C. C.; Savinell, R. F. J. Electroanal. Chem. 1997, 430, 205214.

1198 Analytical Chemistry, Vol. 73, No. 6, March 15, 2001

Figure 2. Effect of the solution pH during the CV overoxidation treatment of the PPD coating on the SWV peak current of a, 1 µM DA, and b, 1 mM AA, at an OPPD-CFME. Overoxidation conditions: CV scan rate, 10 V/s; no. of CV runs, 50; anodic vertex potential, 2.2 V. Measurement conditions as in Figure 1.

time and potential. Both of the parameters are obviously strongly interconnected in determining the degree of porosity of the OPPD film or its structural damage. In our overoxidation procedure, we used cyclic voltammetry rather than the anodization at a fixed potential in order to allow a certain degree of relaxation during the process of overoxidation of the doped PPD. A high scan rate of 10 V/s was selected to increase the number of anodization and relaxation periods with regard to a large cycling potential window and a relatively short optimum overoxidation time. As a result, a high anodic vertex potential of 2.2 V, coupled with a short overoxidation time of 7 s (50 CV runs) at a cycling frequency of 2.3 Hz, proved to effectively turn PPD into an OPPD film for optimum simultaneous detection of DA and AA. Figure 2 shows the effect of the pH of the anodization medium. While the SWV peak current response to both DA (curve a) and AA (b) exhibits a narrow maximum with OPPD prepared when the pH of the anodization medium was ∼7, their peak resolution does not change with pH (not shown). From this behavior, it may be assumed that the OH- ions are involved in the PPD overoxidation process, probably through the formation of oxygen-rich groups at oxidizable polymer sites. With increasing OH- ion concentration, a higher rate of oxygenation, in combination with the electro-oxidation, gradually increases its negative charge intensity. At low pHs (below 6), low OH- concentration apparently does not allow overoxidation to start, and at pHs that are too high (beyond 7.5), the degree of overoxidation is too high and, thus, appears to suppress the response signals to both of the analytes at the resulting OPPD film. Characterization of the OPPD Film with Respect to DA and AA Sensing. To demonstrate the difference in electrochemical behavior between the PPD and the new OPPD coating, CV responses to 1 mM potassium hexacyanoferrate(III) at four different electrode surfaces were recorded (Figure 3): (i) at a PPD-coated CFME, (ii) at an OPPD-modified CFME, (iii) at a bare

Figure 3. Cyclic voltammograms of 1 mM potassium hexacyanoferrate(III) in a 0.1 M KCl solution at (a) a PPD-, (b) an OPPD-modified CFME, (c) an overoxidized bare CFME, and (d, inset) an untreated bare CFME. Scan rate, 100 mV/s.

CFME that was overoxidized using the same procedure as for preparing the OPPD film, and (iv) at an untreated bare CFME. The PPD-CFME with no signal at all (curve a) and the untreated bare CFME with a well-defined voltammogram (d, inset) clearly displayed their expected behavior, which is in agreement with the observations of other authors that the PPD, prepared under similar conditions, is a strongly adhering nonconducting polymer.35,39 Obviously, the examined redox species could not penetrate through the insulating PPD film to reach the carbon fiber surface. In contrast, the OPPD film shows a distinct current response (Figure 3, curve b). At the OPPD film, low reversibility behavior with a cathodic-to-anodic peak separation of ∼120 mV is observed. The large difference in current responses that was obtained at the OPPD-CFME and the untreated bare CFME (curves b and d with a cathodic current ratio of ∼1:20) is primarily due to a negative charge at the oxygen-rich OPPD surface repelling the ferricyanide anion. For comparison, a similarly suppressed but ill-defined response at an overoxidized bare CFME is also shown (curve c). These observations indicate that not only the extent of porosity and amount of negative charge but also some other factors (e.g., analyte and/or oxidation product adsorption/desorption processes, van der Waals interactions, ion-pairing, etc.) may play a particular role in governing the electrochemical behavior of the OPPD film. Because the heterogeneous mechanisms of the oxidation of catechols and ascorbic acid at carbon and some coated electrodes have been reported in detail,32,38,40,41 our attention was mainly focused on the particular electrochemical behavior of DA and AA at the OPPD-modified CFME. Figure 4 displays cyclic voltammograms of DA in pH 7.4 phosphate buffer at an overoxidized bare (39) Komura, T.; Funahasi, Y.; Yamaguti, T.; Takahasi, K. J. Electroanal. Chem. 1998, 446, 113-123. (40) Deakin, M. R.; Kovach, P. M.; Stutts, K. J.; Wightman, R. M. Anal. Chem. 1986, 58, 1474-1480. (41) Malem, F.; Mandler, D. Anal. Chem. 1993, 65, 37-41.

Analytical Chemistry, Vol. 73, No. 6, March 15, 2001

1199

Figure 4. Cyclic voltammograms of 1 µM DA at a, an overoxidized bare CFME, and b, an OPPD-modified CFME: medium, pH 7.4 phosphate buffer; a1 and b1, first scan; a2 and b2, second scan; and b3, third scan. Scan rate, 100 mV/s; accumulation conditions, 10 s at -0.2 V.

CFME (curves a1, a2) and an OPPD-CFME (b1-b3), demonstrating a two-electron redox reaction, with anodic and cathodic peaks (at the OPPD-CFME) at 0.19 and 0.14 V, respectively, that is apparently affected by coupled surface interactions of DA. Multiscan recording revealed preconcentration capabilities of both overoxidized surfaces after accumulation of DA at -0.2 V for 10 s. However, the current increase is substantially larger at the OPPD-CFME than at the overoxidized bare CFME pretreated in the same way (b1/b2 versus a1/a2). Even without preconcentration, the performance of the OPPD-CFME is much better (b2 versus a2). The true value and attractiveness of the OPPD coating is demonstrated when both analytes, DA and AA, are present in the solution. As shown in Figure 5, 1 µM DA at the OPPD-CFME exhibits a distinct and much larger current signal at 0.2 V than 1 mM AA at ∼0 V (curves b, c). A remarkable improvement is achieved, as compared to the overoxidized bare CFME (Figure 5, inset) where the current response of AA totally overlaps the signal of DA (curves e, f). A recording in a blank phosphate buffer (Figure 5, curve a) exhibits a low background, which implies excellent signal-to-background characteristics of the OPPD film. It is interesting to note that the peak potentials of DA and AA at the OPPD film are similar to those observed at some other anodized carbon and organic polymer electrodes.9,27,32,42 This implies that these materials must contain much the same functional groups, for example, oxygen-rich groups, particularly carboxylic/carboxylate groups. In fact, the pKa value of a carboxylate group confined to the surface is reported to be ∼7 rather than ∼4, as it is in the bulk of the solution, due to the lower dielectric constant, hydrogen bonding and interfacial potential effects. As a consequence, at physiological pH, confined carboxylic groups are, in large part, incompletely dissociated.38,43 It can, therefore, be supposed that DA and AA are adsorbed through (42) Chen, P.; McCreery, R. L. Anal. Chem. 1996, 68, 3958-3965. (43) Creager, S. E.; Clarke, J. Langmuir 1994, 10, 3675-3683.

1200 Analytical Chemistry, Vol. 73, No. 6, March 15, 2001

Figure 5. Cyclic voltammograms of DA and AA at an OPPD-CFME and (inset) at an overoxidized bare CFME: medium, pH 7.4 phosphate buffer; a and d, blank medium; b and e, 1 mM AA; c and f, 1 mM AA + 1 µM DA. Scan rate, 100 mV/s.

the interactions between the neutral carboxylic and the hydroxyl groups of DA and AA, whereas the dissociated negative carboxylate groups will uptake only cationic DA and will repel anionic AA away from the film. This is a probable reason for the high response to DA and for the extremely low response to AA at the OPPD-modified electrode. Another reason for the good voltammetric resolution at the OPPD film, besides a large suppression of the AA signal, is the shift of the AA oxidation to a less anodic potential of ∼200 mV (Figure 5). This negative shift may occur for different reasons, but one possible explanation is a kinetic effect.41 Although the transport rate of DA in the OPPD film is relatively slow, the anionic AA moves faster due to the expelling forces. As a further consequence of this negative shift, the anodic current of DA has no contribution from AA present at the DA oxidation sites, because all of the AA is readily oxidized. Thus, AA cannot reduce the product of DA anodic oxidation back to DA, which would result in an increase in the DA current signal, dependent on the actual AA concentration. Such behavior was observed at electrodes coated with, for example, a negatively charged Nafion film.15,44 This property of the OPPD film further contributes to the AAinterference-free detection of DA at the OPPD-CFME. Analytical Performance of the OPPD-CFME for Simultaneous Sensing of DA and AA. When the SWV detection mode with the OPPD-coated microprobes was applied, a dramatic improvement in the resolution and heights of DA and AA voltammetric peaks was observed, as clearly displayed in Figure 6, which is of particular importance in the detection of nanomolar DA. The fast-pulsing potential ramp used in SWV, coupled with the advantageous diffusion-profiling properties of microelectrodes, apparently proves its strong ability to effectively eliminate the background and enhance the response currents. On the basis of this observation and backed by the findings of other authors,45 an optimized SWV detection mode was employed instead of CV (44) Maeda, H.; Yamauchi, Y.; Yoshida, M.; Ohmori, H. Anal. Sci. 1995, 11, 947-952. (45) Nuwer, M. J.; Osteryoung, J. Anal. Chem. 1989, 61, 1954-1959.

Figure 6. Square-wave voltammetry, a, and cyclic voltammetry, b, recordings of 100 nM DA and 0.3 mM AA at an OPPD-CFME: medium, 0.05 M phosphate buffer pH 7.4. Scan rate (CV), 100 mV/ s. SWV settings: frequency, 10 Hz; potential step, 10 mV, pulse amplitude, 50 mV; accumulation condition, 45 s at -0.2 V.

throughout this work, except in some characterization experiments, due to the excellent peak resolution, well-shaped peaks, and negligible interference from dissolved oxygen. Because the ability of the OPPD film to accumulate DA at physiological pH (Figure 4) was considered to be an important analytical factor, the preconcentration parameters were carefully optimized. The response to DA greatly increased when accumulation potentials were altered from the positive to negative side but remained constant when applying preconcentration potentials in the range of -0.1 to -0.5 V. In the presence of AA, -0.2 V was selected as an optimum preconcentration potential to enable recording of the AA signal, whereas in the absence of AA, the accumulation potential can be set to -0.1 V. The optimum accumulation time was found to be 45 and 10 s for DA at nanomolar and micromolar levels, respectively. The influence of the pH of the measurement solution on the SWV current response to DA and AA at the OPPD-CFME shows a maximum for DA at pHs from 5 to 7.5 and a constant decrease for AA with rising pH (not shown). These trends roughly fit with the facts that the form of DA changes from cationic to neutral at pHs > 8, that the form of AA changes from neutral to anionic at pHs > 4, and that the dissociation of polymer carboxylic groups occurs at pHs higher than ∼7. The SWV signal of DA at a physiological pH of 7.4 is, therefore, at its optimum and allows an excellent detection limit for DA in the low nanomolar range. With increasing pH, the peak potentials of DA and AA are shifted negatively, with their difference remaining nearly constant. This is a consequence of a deprotonation step involved in both oxidation processes that is facilitated at higher pHs. The analytical (calibration) experiments were carried out either by varying DA concentration in the presence of 0.3 mM AA (an average AA physiological concentration) or by varying AA concentration in the presence of 100 nM DA (a typical DA physiological concentration) in pH 7.4 phosphate buffer. As shown

Figure 7. SWV recordings of DA at an OPPD-CFME in a pH 7.4 phosphate buffer and in the presence of 0.3 mM AA. DA concentration (nM): a, 0; b, 50; c, 80; d, 100; e, 200; f, 300; g, 500; h, 800; i, 1000. Inset: SWV recordings of DA at an OPPD-CFME in a pH 7.4 phosphate buffer in the absence of AA. DA concentration (nM): j, 0; k, 10; l, 20; m, 50. SWV settings and accumulation conditions as in Figure 6.

Figure 8. SWV recordings of AA at an OPPD-modified CFME in a pH 7.4 phosphate buffer containing 100 nM DA. AA concentration (mM): a, 0.1; b, 0.2; c, 0.3; d, 0.5; e, 0.8; f, 1.0. SWV settings and accumulation conditions as in Figure 6.

in Figure 7, DA exhibits excellent SWV response with the AA signal height remaining unchanged, thus proving that the responses to DA and AA at the OPPD microprobe are relatively independent. Using the 45-s accumulation time, the signal-toconcentration relationship for DA is linear from 50 nM to 10 µM in the presence of 0.3 mM AA, with a correlation coefficient of 0.998 and a detection limit of 10 nM. In the absence of AA, the detection limit of DA was estimated (S/N ) 3) to be 2 nM. The inset of Figure 7 clearly demonstrates the ability of the OPPDCFME to measure low nanomolar DA. Figure 8 shows the measurement of AA in the presence of 100 nM DA (note that curves do not have the same zero-current level). The linearity of the AA signal ranges from 0.2 to 2 mM, with a correlation Analytical Chemistry, Vol. 73, No. 6, March 15, 2001

1201

coefficient of 0.999 and a detection limit of 80 µM. The current sensitivities are 2.6 nA/µM and 1.7 nA/mM for DA and AA, respectively. The reproducibilities of measurements using the OPPD-CFMEs are 1.6% (n ) 7) and 2.5% (n ) 7) at 100 nM DA and 0.3 mM AA, respectively. The primary potential interfering agents in electrochemical detection of DA are AA and DOPAC. However, from Figure 8, it is clear that for the measurement of 100 nM DA, AA does not interfere in the concentration range e2 mM. In other words, AA does not represent an interfering factor for DA detection up to a 20 000-fold excess. The other main interfering substances beside AA were examined and it was found that a 100-fold excess of DOPAC, 1000-fold excess of uric acid and oxalate, and 10 000fold excess of glucose did not interfere with the measurement of 100 nM DA. Furthermore, the measurements of DA and AA in Ringer’s solution (39 mM NaCl, 1.3 mM KCl, 4.5 mM CaCl2, 1.5 mM MgCl2, and 6.9 mM NaHCO3; pH 7.4) demonstrated responses completely identical to those obtained in a pH 7.4 phosphate buffer. This indicates that the DA measurements are independent of the influence of the divalent cations.8 Moreover, the addition of human serum proteins up to 0.7% (m/m) presented no interference to DA detection. It is known that the amount of proteins in the cerebrospinal fluid (CSF) does not exceed 0.05% (max, 45 mg/dL), and that the extracellular fluid within the central nervous system communicates directly with the CSF.46 Hence, the proposed OPPD-coated microprobe should prove to be suitable for in vivo and in vitro measurements of DA and AA in brain studies. Concerning the lifetime, maintenance, and storage of the OPPD-CFME, our experience has demonstrated that the OPPD film showed only a minor change in DA response, even after being stored in air for 1 month. The refreshment of the microsensor is very easy to perform by running a certain number of voltammetric (46) Physiology, 4th ed; Berne, R. M., Levy, M. N., Eds.; Mosby: St. Louis, 1998; pp 85, 86.

1202

Analytical Chemistry, Vol. 73, No. 6, March 15, 2001

cycles (e.g., 20 cycles after the detection of 1 µM DA) between -0.2 and +0.8 V in a pH 7.4 phosphate buffer. CONCLUSIONS The new procedure involving electropolymerization of SDSdoped PPD onto CFMEs, and its activation by overoxidazing it into an OPPD film is very simple, reliable, well-controllable, and fast. The resulting microsensor exhibits, in combination with the SWV detection mode, a strong ability to measure DA at nM and AA at mM concentration levels simultaneously at physiological pH. The favorable properties (e.g., amount of negative charge and permeability) achieved through the reported overoxidation procedure permit DA accumulation and AA rejection and cause a cathodic shift of the AA oxidation potential of nearly 200 mV, leading to superior sensitivity and selectivity of the OPPD coating. By using 45 s of accumulation, detection of DA is allowed down to the 10 nM level at a 20 000-fold excess of AA and even down to 2 nM in the absence of AA. In addition, the possible interfering agents for DA, for example, DOPAC, uric acid, oxalate, glucose, and proteins at their physiologically relevant concentrations, presented no interference for nanomolar DA measurement. These attractive features prove that the proposed OPPD-coated microprobe is promising for in vivo and in vitro applications, including measurement of DA and AA in extracellular and cerebrospinal fluids. ACKNOWLEDGMENT This work was supported by the Ministry of Science and Technology of the Republic of Slovenia (P1-0509-0104). J.-W. Mo is grateful for a scholarship provided by the National Institute of Chemistry, Ljubljana. Received for review September 12, 2000. Accepted December 6, 2000. AC0010882