Anal. Chem. 1996, 68, 2084-2089
Overoxidized Polypyrrole-Coated Carbon Fiber Microelectrodes for Dopamine Measurements with Fast-Scan Cyclic Voltammetry Karin Pihel, Q. David Walker, and R. Mark Wightman*
Department of Chemistry, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599-3290
Thin films of overoxidized polypyrrole have been electrochemically coated onto carbon fiber microelectrodes and used for dopamine measurements with backgroundsubtracted, fast-scan cyclic voltammetry at a scan rate of 300 V/s. The films were stable on the electrode surface only when the electrodes were scanned to high potentials (1400 mV vs SSCE) in pH 7.4 aqueous buffer. Dopamine sensitivity and ascorbate and dihydroxyphenylacetic acid (DOPAC) rejection at the overoxidized polypyrrole-coated electrode were compared to those at carbon fiber electrodes coated with Nafion, a perfluorinated ion-exchange material. At 300 V/s, the overoxidized polypyrrole-coated electrode was almost 3 times more sensitive to dopamine than an uncoated disk electrode. Furthermore, the films were as effective as Nafion in the attenuation of the response to ascorbate and DOPAC, common interferences of dopamine in vivo. Overoxidized polypyrrole-coated electrodes maintained a stable response to dopamine for several hours when implanted in the rat brain. The electrochemical deposition procedure was effective at both elliptical and cylindrical electrodes. This is in contrast to the dip-coating procedures employed with Nafion films that lead to nonuniform coatings at cylindrical electrodes. Microelectrodes are useful for measurements of changes in easily oxidized neurotransmitter concentrations in extracellular brain fluid1 because they provide a way to observe the rapid chemical changes associated with release of neurotransmitters from neurons and their subsequent removal from the extracellular fluid. These measurements, however, present several analytical challenges. First, there are many substances present in brain tissue that can foul the electrode surface. Second, the neurotransmitters of interest are present in very low concentrations, while many other electroactive species are present at higher concentrations and interfere with the signals of interest. One approach to reducing the extent of these problems has been to coat electrodes with ion-exchange polymer films.2,3 Most neurotransmitters are cations at physiological pH. Thus, they can be accumulated into a cation-exchange membrane to increase sensitivity. In contrast, many substances that cause electrochemical interference are anions, and the cation exchanger blocks their access to the electrode surface. The film also prevents blockage (1) Kawagoe, K. T.; Zimmerman, J. B.; Wightman, R. M. J. Neurosci. Methods 1993, 48, 225-240. (2) Kristensen, E. W.; Kuhr, W. G.; Wightman, R. M. Anal. Chem. 1987, 59, 1752-1757. (3) Gerhardt, G. A.; Oke, A. F.; Nagy, G.; Moghaddam, B.; Adams, R. N. Brain Res. 1984, 290, 390-395.
2084 Analytical Chemistry, Vol. 68, No. 13, July 1, 1996
of the electrode by larger molecules. However, a limitation to this approach is that it lowers the response time of the sensor. This is a disadvantage since the concentration of neurotransmitters in the extracellular fluid can change on the subsecond time scale. Consequently, thin films must be employed so that the film does not create a diffusional barrier that creates an unacceptable temporal delay. Another way to increase electrode sensitivity is by electrochemical treatment of the electrode surface.4,5 Often the results of these procedures are short-lived,6,7 while it is necessary for the electrode response to be stable for several hours in the brain. Furthermore, these procedures may slow the electrode response time because they promote adsorption.5 The negatively charged perfluorinated ionomer, Nafion, has been the most commonly used polymer film for determination of the catecholamine neurotransmitters.3 Nafion enhances sensitivity to the neurotransmitters dopamine and norepinephrine. It also reduces responses to ascorbate, which is ubiquitous in brain tissue, and dihydroxyphenylacetic acid (DOPAC), the major metabolite of dopamine. Both interfering compounds are oxidized at approximately the same potential as dopamine.1 Furthermore, they are present at concentrations in the micromolar to millimolar range, whereas the concentration of dopamine is normally much lower. Thus, Nafion-coated electrodes greatly aid in these measurements. Brajter-Toth and co-workers have proposed overoxidized polypyrrole films as a substitute for Nafion films.8-11 In its oxidized form, polypyrrole is a positively charged conducting polymer.12 Upon overoxidation, it loses its conductivity and charge.13 Characterization of these films by XPS14,15 and FT-IR16 revealed that overoxidation results in addition of carbonyl and carboxylic groups. These groups attract dopamine cations and reject (4) Gonon, F.; Buda, M.; Cespuglio, R.; Jouvet, M.; Pujol, J. F. Nature 1980, 286, 902-904. (5) Kovach, P. M.; Deakin, M. R.; Wightman, R. M. J. Phys. Chem. 1986, 90, 4612-4617. (6) Gonon, F. G.; Buda, M.; Cespuglio, R.; Jouvet, M.; Pujol, J. F. Brain Res. 1981, 223, 69-80. (7) Hafizi, S.; Kruk, Z. L.; Stamford, J. A. J. Neurosci. Methods 1990, 33, 4149. (8) Witkowski, A.; Brajter-Toth, A. Anal. Chem. 1992, 64, 635-641. (9) Witkowski, A.; Freund, M. S.; Brajter-Toth, A. Anal. Chem. 1991, 63, 622626. (10) Hsueh, C.; Brajter-Toth, A. Anal. Chem. 1994, 66, 2458-2464. (11) Freund, M.; Bodalbhai, L.; Brajter-Toth, A. Talanta 1991, 38, 95-99. (12) Diaz, A. F.; Castillo, J. I. J. Chem. Soc., Chem. Commun. 1980, 397-398. (13) Beck, F.; Braun, P.; Oberst, M. Ber. Bunsenges. Phys. Chem. 1987, 91, 967974. (14) Ge, H.; Qi, G.; Kang, E.; Neoh, K. G. Polymer 1994, 35, 504-508. (15) Palmisano, F.; Malitesta, C.; Centonze, D.; Zambonin, P. G. Anal. Chem. 1995, 67, 2207-2211. (16) Christensen, P. A.; Hamnett, A. Electrochim. Acta 1991, 36, 1263-1286. S0003-2700(96)00153-9 CCC: $12.00
© 1996 American Chemical Society
ascorbate and DOPAC anions. Because the carbonyl groups create a high electron density that is also diffuse, they may not cause strong electrostatic interactions with dopamine and other cations such as those found with the negatively charged sulfonate groups in Nafion.8 While there have been several reports of voltammetric studies of overoxidized polypyrrole-coated electrodes at relatively slow scan rates,8-11,17,18 in vivo experiments require the time resolution of fast-scan cyclic voltammetry. In this work, carbon fiber microelectrodes are electrochemically coated with overoxidized polypyrrole for use in fast-scan cyclic voltammetry measurements in vivo. Cylindrical electrodes constructed from carbon fibers are relatively easy to fabricate and have shown sensitivity advantages in vivo.3,4,6 Because of their larger surface areas, these electrodes provide large currents for dopamine oxidation. The larger surface area is gained without an increase in the dimensions of the carbon fiber itself, which would result in more physical damage to the rat brain. However, it is difficult to dip-coat an even film of Nafion on cylindrical electrodes.19 It is much easier to obtain an even coat of overoxidized polypyrrole on a cylindrical electrode since this film is electrodeposited. The use of carbon fiber cylindrical electrodes with overoxidized polypyrrole is described in this work. EXPERIMENTAL SECTION Cyclic Voltammetry. Carbon fiber electrodes with elliptical surfaces were constructed as previously described1 from 10 µm diameter carbon fibers (Thornel P-55, Amoco Corp., Greenville, SC). The tips were polished at a 25° angle on a diamond dust embedded beveling wheel (K. T. Brown Type, Sutter Instrument Co., Novato, CA). After polishing, a 1 µm thickness of glass capillary remained surrounding the carbon fiber at the tip. Cylindrical electrodes were constructed from 7 µm diameter carbon fibers (Thornel T-300). About 50-100 µm of the carbon fiber extended beyond the glass insulation. These electrodes were not polished. All electrodes were calibrated with a flow injection apparatus before and after measurements were taken in the brain. Measured currents were then converted into concentrations on the basis of these calibrations. Cyclic voltammograms were obtained with an EI-400 potentiostat (Ensman Instrumentation, Bloomington, IN). Flow injection experiments employed a locally constructed sodium saturated calomel reference electrode (SSCE), and in vivo experiments employed a Ag/AgCl reference electrode (BAS, West Lafayette, IN). A triangle waveform, from -400 to +1400 to -400 mV, was applied to the electrode every 100 ms at a scan rate of 300 V/s, and voltammograms were repeated at 100 ms intervals. The current response was filtered at 2 kHz with a two-pole filter. The data were digitized with an analog-to-digital converter (Labmaster DMA, Scientific Solutions, Solon, OH) interfaced to an IBMcompatible personal computer with either locally written or commercially available software (CV6, Ensman Instrumentation). Cyclic voltammograms shown in this work have been background-subtracted from responses collected immediately before exposure to an electroactive substance. Current-time traces of dopamine oxidation were constructed from the average current collected in the 100 mV range around the peak oxidation potential for this wave. The response time of the electrode was measured (17) Gao, Z.; Ivaska, A. Anal. Chim. Acta 1993, 284, 393-404. (18) Gao, Z.; Chen, B.; Zi, M. Analyst 1994, 119, 459-464. (19) Brazen, M. P.; Kasser, R. J.; Renner, K. J.; Feng, J.; Moghaddam, B.; Adams, R. N. J. Neurosci. Methods 1987, 22, 167-172.
as the amount of time necessary to rise from 10% to 90% of the maximum response in the calibration curve. Film Preparation. Overoxidized polypyrrole films were prepared on the basis of the method described by Hsueh and Brajter-Toth.10 Polypyrrole was formed on the electrode surface by pyrrole oxidation in an aqueous 50 mM pyrrole/200 mM KCl solution. This was accomplished with a potential scan from 0 to 1000 mV and back to 0 mV at 10 V/s. The overoxidation was accomplished by placing the polypyrrole-coated electrode in 0.5 M NaOH, followed by application of the 10 V/s waveform, from 0 to 1000 to 0 mV, 10 times at 2 s intervals. To ensure complete coverage, the electrode was returned to the pyrrole solution, coated with polypyrrole as before, and overoxidized again. These steps were repeated six times. The electrodes were stored in a 0.5 M, pH 7.0 phosphate buffer until they were ready for use (usually within an hour). Electrode pretreatment in NaOH has shown sensitivity enhancement for some species.20 However, increased sensitivity in this work is not the result of electrochemical treatment of the electrode by NaOH during overoxidation because the scans in NaOH used here were not sufficient to increase sensitivity at uncoated electrodes. Nafion films were prepared by dip-coating in a solution of Nafion (Aldrich, Milwaukee, WI) that had been diluted to 2.5% with 2-propanol as previously described.1 After 15 min, the electrodes were removed from the solution and dried with a heat gun for 10 min. The electrodes were then stored overnight and used the following day. In Vivo Experiments. Animal care was in accordance with the Guide for the Care and Use of Laboratory Animals (NIH Publication No. 86-23, 1985, Bethesda, MD) and approved by the Institutional Animal Care and Use Committee. Male SpragueDawley rats (300-500 g, Charles River Laboratories, Raleigh, NC) were anesthetized, placed in a stereotaxic frame, and maintained at 37 °C as previously described.1 After holes were drilled in the skull, a stimulating electrode was lowered with a micromanipulator to the medial forebrain bundle. A carbon fiber microelectrode was similarly lowered into the caudate putamen, and its vertical position was adjusted to maximize the amount of dopamine observed during stimulation. The electrode was then left in the same position for 3 h and stimulated every 20 min with a computergenerated 50 Hz biphasic stimulus pulse that lasted 2 s. In other experiments, the stimulus frequency was varied to measure release under different conditions. Reagents and Solutions. Flow injection analysis by cyclic voltammetry was conducted in a 200 mM Tris-HCl (Sigma, St. Louis, MO) buffer solution at pH 7.4 with 150 mM NaCl. Pyrrole was obtained from Kodak and purified over alumina before use. All other reagents were obtained from Sigma. Solutions of compounds tested were prepared by dilution of 50 mM stock solutions in 0.1 N HClO4. HEPES and other buffers that contain tertiary amines are electroactive at 1400 mV and thus could not be used. Solutions were prepared in distilled deionized water (Mega Pure System MP-3A, Corning Glass Works, Corning, NY). RESULTS AND DISCUSSION Film Preparation. A cyclic voltammogram obtained during electropolymerization of pyrrole at a carbon fiber disk microelectrode is shown in Figure 1A. The integral of the current at positive (20) Anjo, D. M.; Kahr, M.; Khodabakhsh, M. M.; Nowinski, S.; Wanger, M. Anal. Chem. 1989, 61, 2603-2608.
Analytical Chemistry, Vol. 68, No. 13, July 1, 1996
2085
Figure 1. Polymerization and overoxidation of pyrrole on a carbon fiber elliptical microelectrode. (A) Electropolymerization in a solution of 50 mM pyrrole in KCl solution at 10 V/s. (B) Overoxidation of polypyrrole in 0.5 M NaOH at 10 V/s. The waveform was applied seven times. With each successive application of the voltage waveform, the current measured becomes smaller until it stabilizes after about six or seven applications.
potentials gives a charge of 3.5 nC. Figure 1B shows the overoxidation of the polypyrrole film in 0.5 M NaOH. With each successive voltage scan, the amplitude of the current in the cyclic voltammograms decreases, indicating that the film loses its conductivity. Overoxidation is apparently complete after several scans because the current reaches a stable value. This entire procedure was repeated six times to ensure complete coverage.10 The total amount of charge during six polypyrrole depositions onto this electrode was 20 mC/cm2, and this corresponds to a polypyrrole film thickness of 42 nm.21 With uncoated carbon fiber electrodes, cyclic voltammograms of dopamine recorded at 300 V/s typically employ potential limits of -450 to +950 mV vs SSCE.1 These limits allow both the oxidation of dopamine to its o-quinone and the reduction of o-quinone back to dopamine to be observed. However, when this waveform is applied to overoxidized polypyrrole-coated electrodes, the current for dopamine oxidation decreases with successive exposures to dopamine. It is unlikely that electrochemical reduction of the film causes this loss of sensitivity because overoxidation of polypyrrole has been reported to be an irreversible process.13 Instead, we believe film instability was caused by the waveform employed. This problem was overcome with scans to the more positive potential limit of 1400 mV. Scans to such a high potential improve sensitivity because oxidation of the carbon surface increases adsorption of some species to the electrode surface.7 Figure 2 compares cyclic voltammograms of dopamine, ascorbate, and DOPAC with conventional scan limits and extended scan limits at an uncoated electrode. With the extended scan limits, the dopamine signal increases by a factor of 4, while the ascorbate and DOPAC amplitudes remain nearly the same as with conventional scan limits. For all three of these species, the peak potentials for oxidation shift to more negative values when the scan limits are extended, an indication that the electrode has become more activated. Film Selectivity and Response Time. Figure 3A-C compares cyclic voltammograms collected at uncoated, overoxidized polypyrrole-coated, and Nafion-coated electrodes. The overoxidized polypyrrole-coated electrodes is the most sensitive to dopamine. Both the overoxidized polypyrrole-coated and the Nafion-coated electrode reject ascorbate and DOPAC molecules equally well. DOPAC rejection at the overoxidized polypyrrole (21) Holdcroft, S.; Funt, B. L. J. Electroanal. Chem. 1988, 240, 89-103.
2086 Analytical Chemistry, Vol. 68, No. 13, July 1, 1996
Figure 2. Comparison of cyclic voltammetry at 300 V/s with two different sets of scan limits at an uncoated electrode. (A) When the oxidative scan limit is extended from +950 to +1400 mV, the current for 10 µM dopamine becomes much larger, and the oxidative and reductive peaks shift inward. This indicates an increase in adsorption. (B) For 200 µM ascorbate, the current does not increase with the higher oxidative scan limit, but the oxidative wave shifts inward. (C) The effects of a change in scan limit for 50 µM DOPAC are similar to what is observed for ascorbate.
Figure 3. Comparison of cyclic voltammetry at uncoated electrodes, Nafion-coated electrodes, and overoxidized polypyrrole-coated electrodes. (A) The overoxidized polypyrrole-coated electrode is the most sensitive to 10 µM dopamine, and the uncoated electrode is the least sensitive. (B) Both the Nafion-coated and overoxidized polypyrrolecoated electrodes reject 200 µM ascorbate well relative to the uncoated electrode. (C) The uncoated electrode measures the largest signal for 50 µM DOPAC, and the overoxidized polypyrrole-coated and Nafion-coated electrodes reject DOPAC. (D) Electrode response time to a 10 µM dopamine injection in the flow stream. The uncoated electrode is the fastest, and the overoxidized polypyrrole-coated electrode is the slowest.
electrode is similar to that observed by Brajter-Toth and coworkers.8 A quantitative comparison of the oxidative currents observed at the uncoated and film-coated disk electrodes is shown in Table 1. Because the concentration of dopamine in the extracellular fluid of the brain can change very rapidly,1 it is important that diffusion of the analyte through the film does not distort the measured concentration changes. Current-time traces in Figure 3D show the responses of three different electrodes to a bolus of
Table 1. Comparison of Peak Oxidative Currents Observed at Nafion-Coated (naf) and Overoxidized Polypyrrole-Coated (opp) Electrodes to That Obtained at Uncoated Carbon Fiber Electrodesa
inaf/ibare iopp/ibare
dopamine
ascorbate
DOPAC
1.8 ( 0.4 2.7 ( 0.5
0.3 ( 0.2 0.3 ( 0.1
0.6 ( 0.4 0.6 ( 0.3
a The values are the average responses observed at six to eight electrodes.
Figure 4. Cyclic voltammograms of various electroactive species that exist in the brain obtained at overoxidized polypyrrole-coated and Nafion-coated electrodes in the flow stream. DA, 10 µM dopamine; NE, 20 µM norepinephrine; E, 20 µM epinephrine; 5-HT, 1 µM serotonin; HI, 20 µM histamine; L-DOPA, 50 µM L-DOPA; HVA, 100 µM homovanillic acid; 5-HIAA, 50 µM 5-hydroxyindole-3-acetic acid.
dopamine. The current at the uncoated electrode increases the fastest, with a response time of less than 100 ms. Response is slowest at the overoxidized polypyrrole-coated electrode, which has a response time of about 200 ms. The pyrrole polymerization procedure described above was optimum for a relatively fast response time while maintaining reasonable rejection of ascorbate and DOPAC. Note that, with thicker Nafion films and the conventional scan limits, greater selectivity for dopamine can be achieved but at the expense of temporal resolution. Figure 4 shows cyclic voltammograms for several electroactive species that are found in the brain. These results were obtained at both Nafion-coated and overoxidized polypyrrole-coated electrodes. Voltammograms of the catecholamines dopamine (DA), norepinephrine (NE), and epinephrine (E) are shown in Figure 4. Larger currents are found for dopamine and norepinephrine at the overoxidized polypyrrole-coated electrodes than at Nafion. The oxidation of the catechol portion of these molecules occurs at similar potentials with both films, and thus they are difficult to distinguish. However, the alkyl side chain of epinephrine differs from that of the other two catecholamines because it contains a secondary amine that can be oxidized at high potentials at uncoated electrodes.22 Oxidation of the secondary amine can also be observed at electrodes coated with overoxidized polypyrrole but not at Nafion-coated electrodes. It has been proposed that the secondary amine of epinephrine can only be oxidized if the (22) Pihel, K.; Schroeder, T. J.; Wightman, R. M. Anal. Chem. 1994, 66, 45324537.
molecule can reorient itself on the electrode surface after oxidation of the catechol.22 Thus, reorientation may be more difficult in Nafion because of strong electrostatic attraction to the protonated amine. In contrast, the diffuse electron density in overoxidized polypyrrole would allow more facile reorientation. Other differences in the voltammograms are also due to different interactions of the catecholamines with the films. The voltammograms of the three catecholamines have smaller separation of the anodic and cathodic peaks at the overoxidized polypyrrole films. Also, the ratio of the current at the reductive peak to that at the oxidative peak is larger at Nafion-coated than at overoxidized polypyrrole-coated electrodes. This suggests that the oxidized forms do not diffuse out of the Nafion film as readily as they diffuse out of the overoxidized polypyrrole film because of stronger electrostatic interactions in Nafion. Alternatively, the rate of the reductive reaction may be slower in the overoxidized polypyrrole film than in Nafion. Serotonin (5-hydroxytryptamine) (5-HT in Figure 4) is another neurotransmitter in the brain. It adsorbs to the carbon fiber electrode surface to a greater degree than other species shown in this figure, and this results in a much larger current for the same concentration.23 Histamine (HI in Figure 4), also a neurotransmitter in the brain, is oxidized near the oxidative scan limit with both films. L-DOPA, a precursor to dopamine formation in the brain, is a zwitterion that can penetrate these films. Homovanillic acid (HVA), like DOPAC, is a dopamine metabolite that can only be detected at high concentrations at the coated electrodes. 5-Hydroxyindoleacetic acid (5-HIAA), a serotonin metabolite, also has an attenuated signal at both types of coated electrodes. The cyclic voltammogram of histamine also supports the hypothesis that electrostatic interactions are stronger in the Nafion film than in the overoxidized polypyrrole film.8 Oxidation of histamine, which occurs at 1400 mV, results in a larger current at the overoxidized polypyrrole-coated electrode than at the Nafion-coated electrode. The Nafion-coated electrode shows an additional oxidation and reduction peak that is not observed at the overoxidized polypyrrole-coated electrode. This additional peak increases in amplitude with each repetitive scan, which shows that it arises from products of histamine oxidation. Presumably, they are observed only at the Nafion-coated electrode because the electrostatic interactions are strong enough to trap these species in the film for a longer time. Overoxidized Polypyrrole on Cylindrical Electrodes and Measurement Linearity. Cylindrical electrodes have a greater surface area and thus yield larger currents for the same concentration. In regions of the brain where dopamine nerve terminals are densely distributed, such electrodes are advantageous to elevate the signal above electronic sources of noise. Therefore, cylindrical electrodes were coated with overoxidized polypyrrole. Studies at carbon fiber elliptical electrodes have shown that the edge plane of the carbon contains many electrochemically active oxides,24,25 and the surface density of these oxides is increased when the electrode is mechanically polished.26 The sides of the (23) Jackson, B. P.; Dietz, S. M.; Wightman, R. M. Anal. Chem. 1995, 67, 11151120. (24) Kozlowksi, C.; Sherwood, P. M. A. J. Chem. Soc., Faraday Trans. 1985, 81, 2745-2756. (25) Engstrom, R. C.; Johnson, K. W.; DesJarlais, S. Anal. Chem. 1987, 59, 670673. (26) Kamau, G. N.; Willis, W. S.; Rusling, J. F. Anal. Chem. 1985, 57, 545-551.
Analytical Chemistry, Vol. 68, No. 13, July 1, 1996
2087
Table 2. Calibration of Overoxidized Polypyrrole-Coated Cylinder Electrodes after in Vivo Use
Figure 5. Comparison of cyclic voltammetry at a 50 µm long uncoated and overoxidized polypyrrole-coated carbon fiber cylindrical electrode. The electrode was first tested as an uncoated electrode, cleaned with 2-propanol, coated with overoxidized polypyrrole, and tested again. (A) The film-coated electrode is about twice as sensitive to 10 µM dopamine as the uncoated electrode. (B) The film coated electrode rejects 200 µM ascorbate compared to the uncoated electrode. (C) Response to 50 µM DOPAC is greater at the film-coated electrode compared to the uncoated electrode.
cylindrical electrode may have fewer of these oxides, and the sides of a carbon fiber cannot be polished. Because of these differences in the surface of carbon between cylindrical electrodes and polished elliptical electrodes, differences in relative sensitivity were anticipated. Indeed, while uncoated cylindrical electrodes have good sensitivity to dopamine, they have poor sensitivity to ascorbate and DOPAC based on their surface area.5 Because of this, rejection of ascorbate and DOPAC is less important at carbon cylindrical electrodes. The overoxidized polypyrrole coating doubled the dopamine sensitivity of an uncoated cylindrical electrode which was not subjected to any type of pretreatment other than a soak in 2-propanol (Figure 5A). Voltammograms of ascorbate and DOPAC are also shown in Figure 5B and C. The overoxidized polypyrrolecoated cylindrical electrode has a decreased oxidative peak for ascorbate compared to the uncoated electrode. However, the DOPAC current is larger at the film-coated electrode than at the uncoated electrode. This may occur because the overoxidized polypyrrole does not completely cover the electrode. In fact, at elliptical electrodes only partially coated with overoxidized polypyrrole, the signal for DOPAC is also found to increase compared with that at an uncoated elliptical electrode. Complete coverage with the film attenuates the response to DOPAC (Figure 3C). The pyrrole polymerization and overoxidation procedure thus appear to affect the surface state of the electrode. Table 2 shows calibration data obtained at five overoxidized polypyrrole-coated cylinder electrodes. The cylinder lengths ranged from 10 to 95 µm, and measured currents were normalized to the cylinder length to obtain a comparison of cylinders of different lengths. The calibration curve for dopamine is not linear between 1 and 10 µM. Therefore, the data collected with overoxidized polypyrrole-coated cylinder electrodes are less reliable in this concentration range. However, this would not be a problem for most measurements because dopamine measured in vivo is usually below 1 µM, where the calibration curve is linear. 2088 Analytical Chemistry, Vol. 68, No. 13, July 1, 1996
dopamine concn (µM)
i (nA)/length of cylinder (µm)
0.1 1 10
0.05 ( 0.01 0.5 ( 0.1 2.5 ( 0.6
In Vivo Measurements. To test the in vivo stability of the overoxidized polypyrrole-coated carbon fiber microelectrode, the electrode was used in the brain of an anesthetized rat. The specific location was the caudate nucleus, a region with a high density of dopamine nerve terminals. Electrical stimulation of dopamine neurons with a stainless steel electrode placed in a region of the brain where the axons are located causes dopamine efflux in the extracellular fluid that can be detected by cyclic voltammetry.1 Stimulation of the neurons with 50 Hz pulses for 2 s led to an increase in dopamine concentration as identified by cyclic voltammetry. When the stimulation was repeated 3 h later, a response of similar amplitude was obtained. In three different experiments, a bare cylinder and a polypyrrole-coated cylinder electrode were lowered together into the brain for 3 h, and the waveform with extended scan limits was applied to both electrodes. Over the 3 h period, the bare electrode response increased by an average of 12%, while the polypyrrole-coated electrode response increased an average of 13%. Since the neuronal response with this stimulus is stable,27,28 this shows that the electrode provides a stable response to large concentration changes. During this time, the cyclic voltammograms show no change in shape and were identical to those for dopamine in physiological buffer. At low stimulation frequencies, such as 10 and 20 Hz, a steadystate response is obtained at bare cylinder electrodes or Nafioncoated electrodes29 using normal scan limits, while at stimulation frequencies higher than 20 Hz, a non-steady-state response is obtained. At overoxidized polypyrrole-coated electrodes, however, a non-steady-state response is obtained at all frequencies when extended scan limits are used. We attribute this difference to greater adsorption of dopamine onto electrodes coated with overoxidized polypyrrole. Experiments with uncoated cylinders and the extended scan limits also show this behavior. Thus, the extended scan limits that are necessary for film stability make it impossible to extract kinetic information from in vivo data. The adsorption problem is also seen in the pre- and postcalibrations of electrodes used in vivo. Figure 6 shows the calibration of a typical overoxidized polypyrrole-coated electrode before and after in vivo use. After in vivo use, the electrode response time was much slower, and its sensitivity to dopamine increased. Similar trends were seen with bare electrodes when extended scan limits were used. With normal scan limits, electrodes are less sensitive after use in the brain. Figure 7 shows dopamine release induced by two stimulus pulses separated by 100 ms and measured with an overoxidized polypyrrole-coated cylinder electrode. The overoxidized polypyr(27) Ewing, A. G.; Bigelow, J. C.; Wightman, R. M. Science 1983, 221, 169171. (28) Wiedemann, D. J.; Garris, P. A.; Near, J. A.; Wightman, R. M. J. Pharmacol. Exp. Ther. 1992, 261, 574-579. (29) Garris, P. A.; Wightman, R. M. J. Neurosci. 1994, 14, 442-450.
Figure 6. Calibration of an 80 µm long overoxidized polypyrrolecoated carbon fiber cylinder with 1 µM dopamine before (solid line) and after (dashed line) in vivo use. The half-rise time of the precalibration is 300 ms, and the half-rise time of the postcalibration is 1500 ms.
role-coated electrode has much better sensitivity compared to a Nafion-coated elliptical electrode, as can be seen by comparison with our prior work.30 Experiments with minimal stimulation, such as that shown in Figure 7, are critical to evaluate dopamine dynamics in the brain. Coating cylinder electrodes with overoxidized polypyrrole provides a route to this goal in terms of sensitivity but at the expense of kinetic resolution. CONCLUSIONS Overoxidized polypyrrole-coated carbon fiber microelectrodes are more sensitive to dopamine than Nafion-coated electrodes when used for fast-scan cyclic voltammetry at 300 V/s. The overoxidized polypyrrole film was stable on a carbon fiber elliptical microelectrode for several hours in the rat brain when the electrode was scanned to relatively high potentials (1400 mV vs SSCE). While scans to these high potentials further increased the electrode sensitivity, the high potentials also caused problems with adsorption at the electrode surface. Like Nafion, the film (30) Garris, P. A.; Ciolkowski, E. L.; Pastore, P.; Wightman, R. M. J. Neurosci. 1994, 14, 6084-6093.
Figure 7. In vivo dopamine response to a two-pulse, 10 Hz stimulation. (A) Current-time response and (B) cyclic voltammogram at a 60 µm long overoxidized polypyrrole-coated cylinder electrode.
rejected ascorbate and DOPAC anions and slowed fouling of the electrode surface. Cyclic voltammograms of various neurotransmitters support the theory that interaction of these molecules with Nafion is different from that with overoxidized polypyrrole. Carbon fiber cylindrical electrodes were also successfully coated with overoxidized polypyrrole and were about twice as sensitive for dopamine as uncoated cylindrical electrodes. However, the use of a high oxidative scan limit caused a loss in temporal resolution. ACKNOWLEDGMENT This research was supported by a grant from NIH (NS 15841). Received for review February 16, 1996. Accepted April 4, 1996.X AC960153Y X
Abstract published in Advance ACS Abstracts, May 15, 1996.
Analytical Chemistry, Vol. 68, No. 13, July 1, 1996
2089
