Anal. Cham. 1991, 63, 85-88 A
0.23
L__,l
;
_
r
X
*o a>
over
r
>
"5,
The vibrating mirror device has a number of advantages previously published setups: because the device allows for time averaging, it has superior angular resolution compared to a rotation stage (1CF3 vs 10™2 deg); furthermore, the response is essentially independent of slow fluctuations in ambient light and/or laser intensity, as the detected parameter is a minimum in light intensity. Another useful feature is that the device has a relatively high dynamic response (~50 ms), combined with an angular sensitivity of ca. 0.01 deg. The response time is primarily determined by the vibration frequency of the mirror. The complete reflectance curve can thus be recorded within this time interval, which allows for the study of relatively fast surface processes. In this context, it can be considered as an additional advantage that a true minimum of the reflectance is measured, in contrast to turntable devices, where, in order to obtain sufficient sensitivity, the angle is kept fixed in the region of maximum slope (4). Finally, from a practical point of view it is not unimportant that the device is inexpensive. A disadvantage of the present setup is that during an angular scan the probed spot is not stationary and can shift 0.1-1 mm over the surface. For SPR experiments on surfaces that are inhomogeneous over this characteristic length, this can complicate the interpretation of the experimental results. An alternative setup is under development, where the spot position remains stationary. The use of a reference channel has been shown to considerably improve the overall angular stability. Apart from its application in compensation for unwanted silver responses (e.g., temperature, nonspecific adsorption), the reference channel is useful in minimizing other common channel effects such as variations in the composition of the analyte solution.
MEASUREMENT CHANNEL CH2
REFERENCE CHANNEL CH1
H
0-22
c
\
O' 0_
4k
B
!
0.21
200
100
c
•
300
400
time
(s)
500
600
700
DIFFERENCE
CH2-CH1 0.14
^
r
>
3
•
0·'3 0.12
** -
:
CZ Q_
0.11 ··
*
*·
00
o
0-10
°
0.09
.
·
·
··
· ¡
.¡A 0
100
200
¡
B
300
400
time
(s)
500
600
85
700
Figure 3. (a, top) SPR response of channel 1 and channel 2. both channels contain only buffer. During the A-B Interval, the solutions in both channels are replaced as Indicated in the text. From B, immunoreaction takes place (only channel 2). (b, bottom) SPR response of the difference signal (channel 2 channel 1). -
LITERATURE CITED indicative of a diffusion-controlled surface reaction (8). The net effect is an angle shift of about 4 X 10™3 deg (see Figure 3b), corresponding to a covered fraction of ~0.01. Furthermore, from the figure, the overall angular resolution can be estimated at 2 X 10"3 deg.
this
case
(1) Nylander, C. J. Phys. E: Scl. Instrum. 1985, 18, 736. (2) Kooyman, R. P. H.; Kolkman, H.; Gent, J. van; Greve, J. Anal. Chlm.
Acta 1988, 35, 213.
(3) Uedberg, B.; Nylander, C.; Lundstrom, I. Sens. Actuators 1983, 4,
299.
(4) Cullen, D. C.; Brown, R. G. W.; Lowe, C. R. Biosensors 1988, 3, 211. (5) Raether, H. In Physics of Thin Films·, Hass, G., Francombe, M., Hoffman, R„ Eds.; Academic Press: New York, 1977; Vol. 9. (6) Matsubara, K.; Kawata, S.; Mlnami, S. Appl. Opt. 1988, 27, 1160. (7) Kooyman, R. P. H.; de Bruijn, . E.; Eenlnk, R. G.; Greve, J. J. Mol.
CONCLUSIONS We have described a simple and inexpensive device to determine a reflectance minimum. Although its use was demonstrated in a SPR setup, it is not restricted to this technique.
Struct. 1990, 218. 345.
(8) Eddowes, M. J. Biosensors 1988,
3,1.
Received for review June 21,1990. Accepted October 4,1990.
Application of (3-Mercaptopropyl)trimethoxysllane as a Molecular Adhesive in the Fabrication of Vapor-Deposited Gold Electrodes on Glass Substrates Charles A. Goss, Deborah H. Charych, and Marcin Majda* Department of Chemistry, University of California, Berkeley, California 94720 Blodgett and self-assembling films of amphiphiles (3, 7,10). In these types of experiments, the availability of a large number of optically flat, disposable electrodes is a necessity. Vacuum deposition techniques have been also used in the fabrication of interdigitated microelectrode arrays (18,19) and electrochemical devices (20). In all these applications, ca. 5-nm transition-metal underlayers have to be vapor-deposited first to enhance the gold adhesion to the substrate (1-4,11-18,20). Otherwise, directly deposited gold films do not withstand chemical treatment such
INTRODUCTION Vapor deposition of 50-20-nm gold films on glass and Si02 substrates have been used as a convenient and consistently reproducible method of electrode fabrication (1-10) and in the preparations of standard gold substrates for self-assembly of alkylthiol monolayers and their structural studies (11-17). In electrochemistry, this method of electrode fabrication has been particularly successful in experiments involving electrode modifications by spin coating of polymer films (1, 6) and silanization reactions (2) and by formation of Langmuir0003-2700/91/0363-0085$02.50/0
©
1990 American Chemical Society
86
e
ANALYTICAL CHEMISTRY, VOL. 63, NO.
1,
JANUARY 1, 1991
as cleaning in strongly oxidizing solutions and are unstable during potential cycling in voltammetric experiments. Similar problems are common to Pt films on insulating substrates where sputtering of Ti and W have been also used to promote adhesion (19,21). However, the use of Cr and other metals as undercoating layers presents problems related to Cr (Ti, W) diffusion along grain boundaries to the nobble-metal surface (21-24). This phenomenon leads inevitably to altering of surface properties of gold (Pt) and may effect electron transfer and adsorption processes (21, 25). To combat the problem of interdiffusion, various schemes have been developed, such as the formation of diffusion barriers (26), chemical cleaning (24), and electrochemical pretreatment (25). An alternative approach is to replace the transition-metal underlayer with an insulating organic coupling agent, such as an organofunctional silane. For example, McGee (27) and Ponjee et al. (28) have reported that (3-mercaptopropyl)trimethoxysilane (MPS) can be used to improve the adhesion of Ag and Au evaporated films to glass and Si02 substrates. Aliara and Nuzzo used a cystamine derivative of (isocyanatopropyl)triethoxysilane for the same purpose (29,30). Most recently, Wasserman et al. showed that very adherent gold films were obtained when Au was vapor-deposited on Si02 pretreated with ordered monolayers of ll-(trichlorosilyl)undecyl thioacetate or the corresponding mercaptan (31). In all these instances, it is apparent that the silane moieties covalently bind to the oxide surface through siloxane bonds, while the thiol groups engage in strong binding to the evaporated gold coating. The most interesting facet of this approach is that it involves chemical modification of the glass surface often by a single molecular layer, rather than the application of a relatively thick metallic underlayer. This paper presents fabrication and characterization of durable thin-film gold electrodes prepared by Au vapor deposition on glass substrates pretreated with (3-mercaptopropyl)trimethoxysilane (MPS). The method we developed is similar to that reported by McGee (27) but produces gold films of superior adhesion. The advantage of our method over that presented by Wasserman et al. (31) is that it requires MPS, a reagent that is commercially available and inexpensive. We characterize the electrodes with regard to (1) the Au adhesion, (2) the voltammetry of Au oxidation, and (3) the self-assembly of octadecyl mercaptan monolayers, and with regard to the extent to which they passive electrode surfaces. The results demonstrate the superior characteristics of these electrodes in comparison to those produced with Cr underlayer.
EXPERIMENTAL SECTION Chemicals. House-distilled H20 passed through a Barnstead cm) was used Nanopure II purification train (R > 18 throughout. Reagent-grade concentrated H2S04, 30% H202, 2-propanol, toluene, 70% HC104, and Na2Cr207 were from Fisher. (3-Mercaptopropyl)trimethoxysilane was from Petrarch Systems Inc. and was found to give a satisfactory *H NMR spectrum. Octadecyl mercaptan (987o) was from Aldrich. All chemicals were used as received. Substrate Preparation. Glass microscope slides (Corning; 7.54 cm X 2.54 cm) were used as electrode substrates. Visible residue on their surfaces was removed by wiping with lens paper (Fisher). The slides were then chemically cleaned by immersion in 1:4 H202/concentrated H2S04 (Piranha) at 70 °C for about 20 min, rinsed with H20, stripped of visible surface water with a stream of Ar, and then placed in an oven at 105 °C for about 10 min. The substrates intended for glass/Cr/Au type electrode fabrication were transferred to the vacuum evaporator at this point. Caution: “Piranha solutions” react violently with many organic materials and should be handled with extreme care. MPS Application. MPS was applied from a boiling aqueous alcohol solution as follows. The silanization solution was prepared by adding 10 g of MPS to a solution of 10 g of H20 + 400 g of
2-propanol. After the solution was brought to reflux, the cleaned and dry substrates were immersed for 10 min and then carefully rinsed with 2-propanol, blown dry with a jet of Ar, and cured in a drying oven at 100-107 °C for 8 min. The curing step is crucial to the performance of the coupling agent (see Results and Discussion). The procedure of immersing in the MPS solution, rinsing, and curing was carried out additional two times. At the completion of these steps, the slides appeared clear and clean as before the MPS treatment. Au Evaporation. Gold vapor deposition was carried out in a Veeco Model 7700 bell jar system equipped with a resistive heating evaporation apparatus. The system operated at a base pressure of (1-2) X 10~7 Torr. Metal masks were used to define the electrode pattern. In cases where Cr (99.987c, Balzers) was used, approximately 5 nm of Cr was coated at 1 nm/s and a pressure of about 1 X 10~* Torr. The metal film thickness was measured with a quartz crystal thickness monitor (Slone Inc.). After the pressure was allowed to return to the initial value, 70-200 nm of Au (99.9997c, Lawrence Berkeley Laboratory) was deposited at about 1 nm/s and a pressure of ca. 5 X 10"7 Torr. The thickness of the metal layers was occasionally confirmed with an Alpha-Step Model 100 stylust profilometer (Tencor). The Cr/Au and MPS/Au electrodes were stored in polyethylene boxes for as long as several weeks and were used with no pretreatment unless described otherwise. Octadecyl Mercaptan Self-Assembly. Electrodes were immersed in a solution of 20-50 mM octadecyl mercaptan in toluene for 1-24 h, rinsed with toluene, and then dried with a stream of
Ar. Contact Angle Measurements. The advancing contact angle of water was determined on sessile drops with a Ramé-Hart Model 100 contact angle goniometer. On very hydrophobic surfaces, it was often difficult to separate the drop from the needle. This tended to produce positive errors in the contact angle. To eliminate this problem, the following procedure was used. First, a sessile water drop was applied to the surface and the needle was separated. Then, small quantities of water were added by bringing a drop suspended from the needle into contact with the drop on the surface. Upon contact, the suspended drop and the sessile drop coalesce, thus separating the hanging drop from the needle and causing the sessile drop to expand over the surface. The contact angle was measured after each of three such additions. Electrochemical Instrumentation. Electrochemical experiments were carried out in a single compartment cell by using the usual three-electrode configuration. A Pt wire counter electrode and a Radiometer Model K401 SCE reference electrode were used. The area of the working electrode was 0.28 cm2. In all experiments, solutions were purged and blanketed with Ar. Cyclic voltammetry experiments were done by using a PAR Model 173 potentiostat/galvanostat, Model 179 digital coulometer, and Model 175 Universal Programmer.
RESULTS AND DISCUSSION Gold Adhesion. A simple qualitative test was used to examine the adhesion of gold films. A piece of adhesive tape (Scotch, Magic Transparent) was firmly placed over the Au film and the surrounding glass substrate and was then removed. The fraction of the film that was transferred to the tape was then a relative measure of the film adhesion. We found that when Au was coated directly onto glass substrates, the gold was completely removed by the tape. In fact, the film was removed by just rinsing the slide with water. No transfer to the tape was observed when either a Cr or MPS underlayer was used. Wasserman et al. obtained similar results by using a mechanized version of the tape test in their study of Au films coated on monolayers of ll-(trichlorosilyl)undecyl thioacetate (31). From their data, we infer that the yield strength for the MPS/Au assembly is greater that 80 g/cm. The MPS/Au electrodes were subjected to a variety of commonly employed cleaning methods. Complete adhesion retained after (a) immersion in 60 °C chromic acid cleaning solution and/or 1:20 concentrated HF/H20 solution, (b) immersion in 0.1 M NaOH, (c) exposure to an air or 02
was
ANALYTICAL CHEMISTRY, VOL. 63, NO.
1,
JANUARY
1,
1991
·
87
Table I. Advancing Contact Angles and Extent of Passivation of Octadecyl Mercaptan Monolayers on Gold Electrodes electrode
MPS/Au/OM Cr/Authick/OM Cr/Authin/OM
,
deg
113.8 ± 0.7 113.9 ± 0.8 111.7 ± 0.9= to 108 ± 0.7d
1OOO06
1.8 ± 0.9 1.9 ± 1.0 12.4 ± 5.1
“Average of at least 12 electrodes, each measured 3 times. Erare 1 standard deviation. 6 Average of 10 electrodes. Errors = are 1 standard deviation. Obtained within 24 h following Au ded position. Self-assembly of OM layer and contact angle measurements were carried out 1 week after Au deposition. rors
integrating the oxide reduction wave instead of the oxidation wave, we obtained the average roughness factor (based on five electrodes) for MPS/Au electrodes of 1.27 ± 0.04. This value is consistent with our earlier estimates of the roughness factor for these types of surfaces (3). The dashed trace in Figure 1 shows a typical response of
Cr/Au electrode in the same solution. In this case, the Au oxidation wave is poorly defined because of the sharp increase in the anodic current at +1.4 V. This behavior is independent of the thickness of Au films in the range 70-200 nm. We attribute this to the oxidation of Cr present at the electrode/solution interface. In fact, visual inspection of the back side of the electrode after several cycles extended to +1.6 V revealed that a substantial fraction of Cr had been oxidized from underneath the Au. Following this experiment, the gold film failed the tape test in the area previously exposed to the electrolyte solution. These electrodes also exhibit an anomalous reduction wave at about +0.7 V, which appears to be associated with the reduction of Cr species since it is essentially absent if the oxidation scan is reversed at potentials less positive than the onset of the anodic current at +1.4 V. The Cr/Au electrodes used immediately after fabrication and those that were stored for several weeks showed similar response. This indicates that the diffusion of Cr through Au films is rapid and/or that the 70-200-nm-thick gold films are appreciably porous so that Cr electrooxidation can proceed rapidly. Deposition of much thicker gold films is neither a practical nor viable option since in one literature report Cr203 was detected on the surface of 3-6-jitm-thick Au films used in microcircuits (24). Octadecyl Mercaptan Monolayers. Several studies demonstrated that one can use the self-assembly technique to prepare organized monolayers of octadecyl mercaptan (OM) on Au surfaces (7-17). These monolayers were shown to inhibit Au electrooxidation (7,17) and electron transfer to redox species in solution (7-10). The presence of Cr or its oxide on the Au surface might alter the ability of OM to form a well-organized monolayer, particularly when the Au films are only a few hundred angstroms thick. To probe these differences, we investigated the contact angles of water and the extent of passivation of the OM-coated Cr/Au and MPS/Au surfaces. In the former case, “thin” (70-80-nm) and “thick” (160-200-nm) gold (Cr/Au) electrodes were prepared. Monolayers of OM were self-assembled from toluene solutions as described in the Experimental Section. The advancing contact angles ( ) are listed in Table I. We found no significant differences in the contact angles depending on the time of self-assembly from 1 to 24 h. The average values of obtained at MPS/Au/OM and Cr/Authick/OM surfaces compare favorably with those reported by others (7,11,17). The contact angles obtained on Cr/Authin/OM surfaces were lower by as much as 6 deg, particularly when the self-assembly of OM was carried out 1 week or longer after Au deposition. These data indicate that Cr diffusion to the gold surface is a factor determining the extent of organization of the OM a
1/ II
1-1-1-1_I_i_i_i_I_i_i_i_I_i_i_i_I +
1.6
+0.8
+1.2 E,
+0.4
0.0
Volts vs. SCE
Figure 1. Cyclic voltammograms of Cr/Au (dashed trace) and MPS/Au (solid trace) electrodes recorded In 1.0 M HCI04; v = 50 mV/s, A = 0.28 cm2.
plasma, and (d) voltammetric cycling through Au oxidation in 1 M HC104. We note that adhesion was lost when the electrodes were heated to 270 °C or immersed in Piranha solution. Electrodes tested after several months of storage produced identical results. The effectiveness of the MPS layer depends on the details of the application procedure. Addition of H20 to the MPS solution and refluxing, described in the Experimental Section, were essential to obtaining good adhesion. The procedure outlined by McGee (27) involves deposition from a roomtemperature 2-propanol solution containing 2% MPS (w/w based on 2-propanol), 6% H20, and 4% glacial CH3COOH. When this procedure was used in our laboratory, we found that the adhesion of gold was improved over that at bare glass to the extent that the Au films were not removed by exposure to solvents. However, the films were completely stripped during the tape test. If, using our procedure, the glass slides were immersed only once in the MPS solution, the adhesion was not always uniform over the glass surface. When the MPS-treated slides were cured at 125 °C rather than prescribed 100-107 °C, the Au films were completely removed in the tape test. If the electrodes were not subjected to curing, the gold films were adherent initially but the adhesion was lost after immersion in a chromic acid cleaning solution. This wide variation in the MPS coupling efficiency is not surprising as it is well documented that the properties of organofunctional silane layers depend critically on the method of application and subsequent curing procedures (32, 33). Gold Electrooxidation. To evaluate and compare the MPS/Au and Cr/Au electrodes, we performed cyclic voltammetry in aqueous 1 M HC104 solution. The solid trace in Figure 1 shows the response of a MPS/Au electrode over the* potential range 0.0 to +1.6 V. The voltammetric behavior is consistent with that of a clean polycrystalline Au surface (34-36). Following the method of Burshtein et al. (37) but
88
ANALYTICAL CHEMISTRY, VOL. 63, NO.
·
1,
JANUARY
1,
1991
LITERATURE CITED
Au/OM J
0.5
(1) Moran, K. D.; Majda, M. J. Electroanal. Chem. Interfacial Electro-
chem. 1988, 207, 73-86.
µ
(2) Wldrig, C. A.; Majda, M. Anal. Chem. 1987, 59, 759-760. (3) Wldrig, C. A.; Majda, M. Langmuir 1989, 5, 689-695. (4) Wehmeyer, K. R.; Deakln, M. R.; Wlghtman, R. M. Anal. Chem. 1985,
57, 1913-1916.
(5) Sanderson,
2388-2293.
D.
G.;
Anderson, L. B. Anal. Chem. 1985, 57,
J. Electroanal. Chem. Interfacial Electrochem. 1989, 269, 77-97. Finklea, . O.; Avery, S.; Lynch, M.; Furtsch, T. Langmuir 1987, 3, 409-413. Sabatani, E.; Rubinstein, I.; Maoz, R.; Saglv, J. J. Electroanal. Chem. Interfacial Electrochem. 1987, 219, 365-371. Sabatani, E.; Rubinstein, I. J. Phys. Chem. 1987, 91, 6663-6669. Rubinstein, I.; Steinberg, S.; Tor, V.; Shanzer, A.; Sagiv, J. Nature 1988, 332, 426-429.
(6) Oh, S. M.; Faulkner, L. R.
(7) (8) (9) (10)
1.....'I.......I +
1.4
+0.7
0.0
E, Volts vs. SCE Figure 2. Cyclic voltammograms of Cr/AutVOM (dashed trace) and MPS/Au/OM or / %**/0 (solid trace) electrodes recorded in 1.0 M HCI04; v= SO mV/s, A = 0.28 cm2.
layers on Cr/Au surfaces when Au films thickness is less than ca. 100 nm. The MPS/Au electrodes are free of this problems regardless of the thickness of the gold films and their storage
time. Cyclic voltammetry was used to further quantify the difference in OM assembly between these types of electrodes. Typical cyclic voltammograms obtained at OM-coated electrodes in 1 M HC104 are shown in Figure 2. The MPS/ Au/OM and Cr/Authick/OM electrodes are clearly more passivated than the Cr/Au^/OM electrodes, consistent with the water contact angles. The potential range was limited to +1.4 V because excursions to more positive potentials resulted in a large Au oxidation current and an increased capacitive current. This is presumably due to desorption of the OM monolayer (7). Upon continuous scanning over the potential range shown in Figure 2, the Au oxide reduction wave became progressively larger so that only first scan was used for analysis. The average fraction of electrode area not passivated by OM (0) was calculated as the ratio of the charge under the Au oxide reduction wave obtained at Au/OM and bare Au electrodes following an oxidation scan reversed at +1.4 V. The values of 0 are listed in Table I. The area of exposed Au at the Cr/Authm/OM electrodes is about 6 times larger than that at the MPS/Au/OM and Cr/Authick/OM electrodes. This is additional evidence that Cr may indeed interfere with the self-assembly of octadecyl mercaptan. In conclusion, we demonstrated that the Cr/Au electrodes appear to be contaminated by Cr species unless ca. 200-nmthick films of gold are vapor-deposited. The MPS/Au electrodes described in this report are free of this and related problems regardless of the thickness of gold films and the storage time. They consistently exhibit behavior characteristic of clean bulk gold surfaces. We have also relied on the procedure described here in the fabrication of interdigitated microelectrode array devices (38) and in the prepration of 50-nm-wide microband electrodes (39), where we observed significantly lower background currents compared with the procedure requiring a chromium underlayer.
(11) Bain, C. D.; Troughton, E. B.; Tao, Y.-T.; Evall, J.; Whitesides, G. M.; Nuzzo, R. G. J. Am. Chem. Soc. 1989, 111, 321-335. (12) Bain, C. D.; Whitesides, G. M. J. Am. Chem. Soc. 1988, 110,
3665-3666.
(13) Nuzzo, R. G.; Fusco, F. A.; Aliara, D. L. J. Am. Chem. Soc. 1987, 109, 2358-2368. (14) Nuzzo, R. G.; Zegarski, B. R.; Dubois, L. H. J. Am. Chem. Soc. 1987, 109, 733-740. (15) Bain, C. D.; Whitesides, G. M. J. Am. Chem. Soc. 1988, 110,
5897-5898.
(16) Porter, M. D.; Bright, T. B.; Aliara, D. L.; Chidsey, C. E. D. J. Am. Chem. Soc. 1987, 109, 3559-3568. (17) Troughton, E. B.; Bain, C. D.; Whitesides, G. M.; Nuzzo, R. G.; Aliara, D. L.; Porter, M. D. Langmuir 1988, 4, 365-385. (18) Kittleson, G. P.; White, H. S.; Wrlghton, M. S. J. Am. Chem. Soc. 1984, 106, 7389-7396. (19) Chidsey, C. E.; Feldman, B. J.; Lundgren, C.; Murray, R. W. Anal. Chem. 1986, 58, 601-607. (20) Morita, M.; Longmire, M. L.; Murray, R. W. Anal. Chem. 1988, 60,
2770-2775.
(21) Josowicz, M.; Janata, J.; Levy, M. J. Electrochem. Soc. 1988, 135,
112-115.
(22) Tisone, T. C.; Drovek, J. J. Vac. Sci. Technol. 1972, 9, 271-275. (23) Ashwell, G. W. B.; Heckingbottom, R. J. Electrochem. Soc. 1981, 128, 649-654. (24) Holloway, P. H. Gold Bull. 1978, 12, 99-106. (25) Cohen, R. M.; Janata, J. J. Electroanal. Chem. Interfacial Electrochem. 1983, 151, 33-39. (26) Murarka, S. P.; Levinstein, H. J.; Blech, I.; Sheng, T. T.; Read, . H. J. Electrochem. Soc. 1978, 125, 156-162. (27) McGee, J. B. U.S. Patent 4,315,970, 1982. (28) Ponjee, J. J.; Nelissen, J. W. A.; Verwijlen, C. J. A. Eur. Pat. Appl. EP 111,957, 1984. (29) Aliara, D. L.; Hebard, F. J.; Padden, F. J.; Nuzzo, R. G.; Falcone, D. R. J. Vac. Sci. Technol. A 1983, 1, 376-382. (30) Aliara, D. L; Nuzzo, R. G. U.S. Patent 4,690,715, 1983. (31) Wasserman, S. R.; Biebuyck, H.; Whitesides, G. M. J. Mater. Res. 1989, 3, 886-892. (32) Plueddemann, E. Silane Coupling Agents Plenum Press: New York, 1982. (33) Plueddemann, E. In Silane Surfaces and Interfaces·, Leyden, D. E„ Ed.; Gordon and Breach Science Publishers: New York, 1986. (34) Dickertmann, D.; Schultze, J. W.; Vetter, K. J. Electroanal. Chem. Interfacial Electrochem. 1974, 55, 429-443. (35) Chialvo, C.; Triaca, W. E.; Arvia, A. J. J. Electroanal. Chem. Interfacial Electrochem. 1984, 171, 303-316. (36) Brummer, S. B.; Makrides, A. C. J. Electrochem. Soc. 1984, 111, ·,
1122-1128.
(37) Michri, A. A.; Pshchenichnikov, A. G.; Burshtein, R. Kh. Elektrokhlmiya 1972, 8, 364-365. (38) Goss, C. A.; Majda, M. J. Electroanal. Chem. Interfacial Electrochem., in press. (39) Wldrig, C. A.; Miller, C. J.; Majda, M. J. Am. Chem. Soc. 1988, 110,
2009-2011.
Received for review May 10,1990. Accepted October 10,1990. We gratefully acknowledge the National Science Foundation for supporting this research under Grant CHE-8807846. C.G. acknowledges and thanks B.P. America for a graduate fellowship (1988-1989).
