Anal. Chem. W90, 62, 2740-2742
2740
(IO) Smampalam. 0.;Wilson, G. S. Anal. Chem. 1983. 55, 1608. (11) S a W , S. V.; Pierce, R. J.; Walk, R.; Yacynych, A. M. Anal. Chem. 1990. 62. 1 1 1 1 . (12)Foulds, N; C.: Lowe, C. R . J. Chem. Soc.. Faraday Trans. 1 1986, 8 2 , 1259. (13)Umana, M.; Wailer, J. Anal. Chem. 1988, 58, 2979. (14) Iwakura, C.: Kajlya. Y.: Yoneyama, H. J. Chem. Soc.. Chem. Commun. 1988, 1019. (15) Kotowski, J.; Janas, T.: Tien, H. Bioekctrochem. Bioenerg. 1988. 19,
277. (16)Pandey, P. C. J. Chem. SOC.,Faraday Trans. 1 1988, 84, 2259. (17) CaDon. A.: Parsons. R. J. Electroanal. Chem. Interfaclal flectrochem. 1975, 65, 285. (18) Malitesta, C.; Sabbatini, L.; Zambonin, P. G.; Peraklo Bicelli, L.; Maffi, S. J . Chem. SOC.,Faraday Trans. 11989, 85, 1685. (19) Ohnuki, Y.; Matsuda, H.; Ohsaka, T.; Oyama, N. J. Nectroanal. Chem. Interfaclal Electrochem. 1983, 158, 55. (20)Brlggs. D.;Riviere, J. C. I n Practical Surface Analysis; Briggs, D., Seah, M. P., Eds.; Wile?: New York, 1983;pp 133-135. (21) &ah. M. P. Swf. Interface Anal. 1986. 9 . 85. (22)Hendbook of Conducting Pofiers; Skotheih, T. A., Ed.; Marcel Dekker: New York, 1986. (23) Chiba, K,; Ohsake, T,; Ohnuki, y,; Oyama, N, J , E/ectroana/,Chem, Interfadel Electrochem. 1987, 219, 117. ’
(24) White, H. S.; Abruna, H. D.; Bard, A. J. J. Electrochem. SOC.1982, 129, 265. (25)Heineman, W. R.; Wieck, H. J.; Yacynych, A. M. Anal. Chem. 1980, 52, 345. (26) Saraceno, R. A,; Pack, J. G.; Ewing, A. G. J. Electroanal. Chem. Interfaclal Electrochem. 1988, 197, 295. Vithanage, R. S . J. Electroanal. Chem. Interfaclal (27)Finklea, H. 0.; Electrochem. 1984, 161, 283. (28) Bourdiilon, C.; Bourgeois, J.; Thomas, D. Biotechnol. Bioeng. 1979, 21. 1877. (29)&;ton, L. Anal. Chim. Acta 1985, 178, 247. (30) Mell, L. D.; Maloy, J. T. Anal. Chem. 1975, 47, 299. (31)Bartlett, P. N.;Whitaker, R. G. J. Electroanal. Chem. InterfacialElectrochem. 1987,224, 37. (32)Castner, J. F.; Wingard, L. B. Biochemistty 1984, 2 3 , 2203. (33) Yokoyama, K.; Tamiya, E.; Karube, I . J. Electroanal. Chem. Interfacial Electrochem. 1989, 273, 107.
RECEIVED for review June 12,1990. Accepted September 28, 1ggO. The work was carried out with financial assstance from “Ministero dell’ Universitl e della Ricerca Scientifica e Tecnologica” and ”Consiglio Nazionale delle Ricerche”.
TECHNICAL NOTES Cylindrical Thin-Layer Cell for Spectroelectrochemistry P a u l A. Flowers* a n d G a r y Nealy Department of Physical Science, Pembroke State University, Pembroke, North Carolina 28372 The use of thin-layer spectroelectrochemical (SEC) techniques for characterization of electroactive species is a wellestablished practice in many research laboratories ( I , 2). Widespread application in the ultraviolet-visible region has resulted in numerous SEC cell designs, ranging from variations on the popular “sandwich” cells (3)to more elaborate designs involving optical fibers (4). Another recent innovation that warrants mention is the development by Kuwana’s lab of long optical path thin-layer cells (LOF’TC), which may be used with planar working electrodes of any material and which possess relatively long path lengths, hence affording optically sensitive analyses ( 5 ) . Of these designs, the sandwich cells are by far the simplest to construct and use. Their utility is limited in many instances, however, by required adhesives that may be attacked by organic solvents. When the use of adhesives is avoided, somewhat complex, and hence often irreproducible, assembly procedures are involved. Described herein is a novel spectroelectrochemical cell designed to circumvent the problems mentioned above. The cell is of cylindrical geometry and permits double-pass transmittance sampling of the thin layer, resulting in an enhanced optical sensitivity relative to standard cells. It is constructed entirely from glass, platinum, and Teflon tape, requires minimal glassworking for initial construction, and may be very simply and reproducibly disassembled/reassembled between experiments. Additionally, the cell’s cylindrical geometry is beneficial in permitting its use with inexpensive fixed-wavelength spectrometers that accept cylindrical (“test-tube”) cuvettes. EXPERIMENTAL SECTION Reagents. o-Tolidine (practical grade) was purchased from Sigma Chemical Corp. and used without further purification. All other chemicals were of reagent grade and were used as received from Fisher Scientific Co. Cell Construction. An illustration of the SEC cell is given in Figure 1. The cell consists primarily of two parts: (1)a Pyrex 0003-2700/90/0362-2740$02.50/0
insert, which serves both as a mechanical support for the auxiliary and working electrodes and as a salt bridge for the reference electrode and (2) a 13 mm 0.d. cylindrical cuvette (Fisher Scientific). The insert was constructed by near-complete flame sealing of one end of an 11mm 0.d. Pyrex tube, leaving a pinhole opening to permit ionic contact between the salt bridge and sample solutions. Bulk mixing of these solutions is prevented by a small plug of KN03-containing agar. A central segment of the insert was drawn to a reduced diameter in order to provide a larger sample reservoir volume. Platinum gauze (52 mesh, Fisher Scientific) served as the optically transparent working electrode (OTE) and the auxiliary electrode. These electrodes were secured to the outer wall of the insert by tightly wrapping with Teflon tape, which also served to electrically insulate the 0.127-mm diameter Pt-wire leads from the sample solution and each other. A gel-filled Ag/AgCl electrode (Bioanalytical Systems, Inc.) was employed as a reference electrode, and all potentials are reported relative to this reference. Instrumentation. Voltammetry and chronocoulometry experiments were performed by usinga Bioanalytical Systems CV-27 voltammograph. Spectroscopic measurements were made with a Milton Roy Spectronic 20 fixed-wavelengthsingle-beam spectrometer. The outputs of the spectroscopic and electrochemical instruments were recorded with Houston Model 4510BF XT and Model 200 XY recorders. Procedure. The SEC cell was assembled by simply Filling the insert with 2.5 M KN08, adding ca. 1 mL of sample solution to the cuvette, sliding the insert into the cuvette, and placing the reference electrode in the upper opening of the insert. The assembled cell was then placed into the sample holder of the spectrometer and electrically connected to the voltammograph, and the desired experiment was performed. RESULTS A N D DISCUSSION Thin-Layer Dimensions. The optical path length of the cell was determined by measuring the absorbance of a standard chromate solution at 373 nm (c = 4813 M-’ cm-’) (6). Five replicate measurments were made, the cell being disassembled, cleaned, and reassembled between each measurement. This procedure yielded an average path length of 0 1990 American Chemical Society
ANALYTICAL CHEMISTRY. VOL. 62. NO. 24, DECEMBER 15. 1990
4
'
nn'
'
ibn
lime
6)
'
un
2741
Cn
Plots of absorbance at 438 nm (A) and charge passed (01 after a potential step from 0.5 to 0.8 V vs AgIAgCI (same condnions as in Figure 2). Flgure 3.
path
Flgure 1. Diagram 01 the SEC ceII: (a) auxiliary electrode lead (b) OTE lead (c) Teflon tape; (d) sample solution; (e) agar plug.
E C V VI AgIAgCl ) Figure 2. Cyclic voltammogram of 0.263 mM 0-tolidhe In 0.5 M acetic acidll.0 M perchloric acid solutlon obtalnsd In the SEC cell at a scan rate of 1 mV1s.
0.0527 cm with a standard deviation of 0.0005 cm (relative standard deviation of less than 1%).Since the optical path length results from essentially a double-pass through the sample, i.e., on the incident and exit sides of the cuvette, the sample layer thickness is half of this value, 0.0264 cm. The thin-layer volume was determined by chronocoulometric analysis of 4.09 mM K,Fe(CN), in 0.1 M KCI to he 56 rL. Comparison of this value to the geometrically calculated thin-layer volume of 220 rL indicates that the sample solution occupies roughly one-quarter of the available volume, the remaining three-quarters being occupied by the OTE. If desired, the sample volume (and the exhaustive electrolysis time as discussed below) could be reduced by using Pt gauze of a finer mesh. Spectroelectrochemical Characteristics. A cyclic voltammogram of 0.263 mM o-tolidine in 0.5 M acetic acidp.0 M perchloric acid solution obtained in the SEC cell is shown in Figure 2. A small edge effect is manifested in the failure of the measured current to return to zero after sweeping the OTE potential through the anodic wave. Furthermore, slight ohmic distortion is observed as the oxidation and reduction peaks are separated hy ca. 25 mV (scan rate of 1mV/s). These deviations from ideal thin-layer behavior (7) are relatively minor, however, and the electrochemical characteristics observed here compare admirably to those of similar cells. For a simple, reversible couple, the ratio of the equilibrium concentrations of oxidized to reduced forms at the electrode, [O]/[R], is determined by the applied potential, E, according to the Nernst equation
E = E"'
+ R T / n F In [O]/[R]
(1)
where Eo' is the formal potential, R is the molar gas constant, T i s the temperature, n is the number of electrons involved
in the reaction, and F is the Faraday constant. A plot of E versus In [O]/[R] yields a straight line whose slope and y intercept given n and E", respectively. Such a plot for the oxidation of o-tolidine was generated by monitoring the ahsorbance a t 438 nm (wavelength of maximum absorption for the oxidation product) as the OTE potential was stepped through the anodic feature. Ample time was allowed between potential steps and absorbance measurements to ensure equilibrium between the working electrode and solution (see below). The measured absorbance values were used to calculate the ratio [O]/[R] according to the equation [OI/[RI= A/&,, - A ) (2) where A is the absorbance at 438 nm at a given potential and A, is the absorbance at 438 nm when the potential is much greater than Eo' and only the oxidation product is present (o-tolidine does not ahsorh at 438 nm). The data obtained yielded a linear plot of applied potential versus log [Ol/[R] ( r = 0.9998) with a slope and intercept corresponding to n = 2.00 and Eo' = 0.612 V, in excellent agreement with previously reported values (8). The exhaustive electrolysis characteristics of the cell were examined by monitoring both the absorbance a t 438 nm and charge passed after stepping the OTE potential from 0.5 to 0.8 V. Plots of absorbance and charge versus time are given in Figure 3. The absorbance-time plot clearly shows that complete electrolysis of the thin layer is achieved in less than 1min. The minor edge effect noted above is again observed in the charge-time plot as a failure of the curve to reach a limiting value. In conclusion, the SEC cell described has been shown to permit accurate control of the working electrode potential with minimal ohmic drop and fairly short electrolysis times. Perhaps of more practical importance, the cell is constructed from relatively inert materials and may be quickly and reproducibly disassembled, cleaned, and reassembled between experiments. If glassblowing facilitiesare available, the design may be slightly modified by replacing the agar plug with a fritted-glass disk to yield a more durable junction. Also, a fritted-glass "ring- could be sealed to the outer perimeter of the insert just above the OTE to isolate the sample layer from species produced at the auxiliary electrode. Finally, the use of quartz tubing instead of glass would permit applications in the ultraviolet region.
LITERATURE CITED (1) Heineman. W. R.: Hawkridge. F. M.: Blarnl. H. N. In EeClrCXnab?kaI C%"bhy:Bard. A. J.. Ed.; Marcel Dekker: New Ywk. 1984 Val. 13. P 1. (2) Robinson. J. I n Elechochemisby: Specraibt Peradical Reports: Plelcher. D., Ed.; The Royal Society of Chemisby. Burlington HOUSE: London. 1984 VoI. 9. p 101. (3) Kuwana,T.; Winograd. N. EleclrCXmWCaI Chemlshy; Bard. A. J.. Ed.: Marcel DBkker: New Ywk. 1974 Val. 7. p 1. (4) Zhang, C.: Park. S.-M. Anal. Chem. 1988. 60. 1639. (5) Gui. Y.: Soper. S. A,: Kuwana. T. Anal. Chem. 1988. 6 0 , 1645. (6) Haupt. G. W. J . Res. &ti. Bur. Stand. ( U . S . ) 1952. 48. 414.
Anal. Chem. 1880, 62, 2742-2744
2742
(7) Bard, A. J.; FauLner, L. R. €kscb.ochemical Adethods: Fundamentals and Applicetlons; Wiley: New York, 1980; p 406. ( 8 ) DeAngells, T. P.;Heineman, W. R. J . &em. Educ. 1976, 53,594.
RECEIVED for review May 18,1990. Accepted September 14,
1990. We gratefully acknowledge financial support by the Research and Pembroke State University ment Program. This work was Dresented at the 104th North Carolina gectional ACS Confeience.
ControlkbPotential Electrolysis of Bulk Solutions at a Modified Electrode: Application to OxMations of Cysteine, Cystine, Methionine, and Thiocyanate James A. Cox* and Thomas J. Gray Department of Chemistry, Miami University, Oxford, Ohio 45056
INTRODUCTION Glassy-carbon electrodes modified by electrodeposition of a mixed-valence ruthenium oxide film stabilized by cyano cross-links (1, 2), mixed-valence ruthenium cyanide (mvRuCN), have been demonstrated to be highly stable (1, 3) and to promote the oxidation of several species that either are not electroactive or are oxidized only a t very positive potentials (1-5). The physical-chemical stability and freedom from fouling during use relative to bare electrodes suggest that these electrodes can find application to investigations that require bulk-scale electrolysis. Although some electrode surfaces such as RuOz or I r 0 2 on a titanium support, Pb02, and Ni02 are used for such purposes, systems that are generally considered as .modified electrodes have not been used for bulk-scale electrolysis (6). Our primary interest in applying the mvRuCN electrode to bulk electrolysis is to test the hypothesis that this surface modification changes the course of certain electrochemical reactions by promoting both electron and oxygen transfer. A successful test of such a hypothesis can lead to new routes for electrosynthesis. The test species in the present study were thiocyanate, cysteine, methionine, and cystine. The oxidation of thiocyanate at bare platinum and glassy-carbon electrodes caused passivation (7, 8). The reported products included trithiocyanide, (SCN)3-, which was stabilized in the presence of a large excess of SCN- in 0.1 M HC104 (9);CN- and S042- (8, 10); and NO2-, S042-, and C 0 2 (11). Using in situ infrared spectroscopy to monitor the process a t a platinum electrode, Pons et al. ( I I) found no evidence of CN-. The passivating film on a glassy-carbon electrode fit the formula C6N4S4(8). Parathiocyanogen, (SCN),, was also reported to form on bare electrodes upon oxidation of SCN- (10). Cysteine oxidation a t bare electrodes also yields passivating films (12-14). The formation of the disulfide dimer at a platinum electrode (12,14)with adsorbed thiyl radicals (12) is indicated when the oxidation is performed in acid solution. The only suggestion of formation of an oxide of sulfur during electrolysis of cysteine is in a study by Reynaud et al. (15). X-ray photoelectron spectroscopy indicates that the adsorbed film in part comprises cystine sulfoxide and an S,S-dioxide from a Kolbe-type decarboxylation; however, even after exhuastive oxidative electrolysis, no evidence of a sulfur oxide as a solution species is seen (15). Koryta and Pradac investigated the electrochemistry of cystine a t platinum and gold electrodes, a t which oxidation was observed at 1.3 and 1.45 V vs NHE, respectively (16,17). In each case, the mechanism was reported as cleavage of the sulfur-sulfur bond and chemisorption of the resulting thiyl radicals. In contrast to the above compounds, the electrochemical oxidation of methionine has not been reported a t either bare or modified electrodes except for one analytical study in the 0003-2700/90/0362-2742$02.50/0
concentration range 0.1-100 ng of methionine/L (18). In this case, it was oxidized a t 1.7 V vs Ag/AgCl a t an anodically pretreated glassy-carbon electrode in a cell used as a detector for high-performance liquid chromatography. The present study yields evidence that oxygen transfer is involved in the electrochemical process a t the mvRuCN electrode. Equally important, it demonstrates that modified electrodes can possess the stability needed for large-scale applications in addition to the analytical applications that generally cause electrolysis of less than 10-loequivlexperiment.
EXPERIMENTAL SECTION An IBM Instruments EC/225 voltammetric analyzer was used
without electronic filtering for all constant-potential measurements. Simultaneous output was made to a strip-chart recorder and an IBM XT computer (DASH 16 analog/digital interface, Metrabyte Corp.). Data acquisition and analyses were performed by using MYSTANT+ (Macmillan Software). The electrochemical cells and electrodes (three-electrode configuration) were prepared in-house. Glassy carbon (Tokai Carbon) was purchased from Aimcor (Pittsburgh, PA). All experimental potentials reported are vs SCE. Indicator half-cells were separated from auxiliary half-cellsby 4-8pm (pore size) fritted-glassdisks (8-mm diameter). For each of the oxidations reported, hydronium ion from lov2M HCl in 1.0 M KC1 was the cathodic depolarizer, and the auxiliary electrode was a pltainum coil. Unless otherwise noted, the chemicals were ACS reagent grade and were used without further purification. RuC13.3H20 was was obobtained from Pfaltz and Bauer, and K,RU(CN)~.~H~O tained from Alfa Products. L-Cysteine hydrochloride hydrate, L-cysteic acid monohydrate, and sodium thiocyanate were purchased from Aldrich. L-Cystine hydrochloride, L-methionine, L-methionine sulfoxide, and L-methionine sulfone were obtained from Sigma Chemical Co. Primary standard arsenic trioxide was used to calibrate the cell. Supporting electrolyte solutions were 1.0 M in KC1, adjusted to pH 2.0 by using HCl. The water used was house-distilled and further purified by passing through a Barnstead NANOpure II system. Thin-layer chromatographywas performed either with Macherey-Nagel Polygram Silica Gel-G plates or Macherey-Nagel300 cellulose acetate plates, each with the fluorescing indicator at 254 nm (Alltech Associates, Inc., Deerfield, IL). Glassy-carbon rods, 0.5-cm diameter X 2.0-cm length (geometric area = 3.3 cm2)for use with the half-cell with the 5.00-mL sample volumes and 0.3-cm diameter X 1.9-cm length (geometric area = 1.9 cm2) for use with the half-cell with the 3.50-mL sample volumes, were sealed in Pyrex glass tubing by using Torr-Seal epoxy (Varian Associates). Prior to surface modification, the glassy-carbon rods were polished successively by using 1-,0.3-, and 0.05-pm alumina (Mark V Laboratory, East Granbury, CT) on a metallographic polishing cloth (Buehler, Evanston, IL) with deionized water as the lubricant. The electrodes were thoroughly rinsed and sonicated with deionized water (sonication time, 10 min). The freshly prepared electrodes were immersed in 10.0 mL of plating solution containing 2 mM RuC13, 2 mM K,Ru(CN),, and 0.5 M KCl with the pH adjusted to 2.0 by using HCl. The glassy-carbon indicators 0 1990 American Chemical Society
