Permselectivity, Sensitivity, and Amperometric pH Sensing at Thioctic

Anal. Chem. 1996, 68, 4180-4185

Permselectivity, Sensitivity, and Amperometric pH Sensing at Thioctic Acid Monolayer Microelectrodes Quan Cheng and Anna Brajter-Toth*

Department of Chemistry, University of Florida, Gainesville, Florida 32611-7200

Gold fiber thioctic acid (TA) monolayer (thickness of ∼8 Å) ultramicroelectrodes (UMEs) have been fabricated and characterized. Gold fibers were acid etched in order to produce monolayers that can act as high-quality permselective membranes. The selectivity and sensitivity of the TA monolayer UMEs compares favorably to that of TA monolayers prepared on Au vacuum deposited on large (∼1 cm2 area) single-crystal silicon wafers. The results obtained here illustrate selectivity of the monolayer electrodes to electroinactive electrolyte anions, which participate indirectly in the electrode reaction, in addition to the previously established selectivity to electroactive cations and anions. The permselectivity of the monolayer to electrolyte anions can result in lower than expected steady-state currents at the monolayer UMEs, at the reaction potentials of fast-reacting analytes, and at high overpotentials for kinetically slow probes. Application of TA UMEs to amperometric pH sensing is also described in the determination of the pKa of TA at the monolayer surface. The method is compared with other methods of monolayer pKa measurements. Ultramicroelectrodes (UMEs) allow high spatial and temporal resolution of measurements;1 the resolution is of key importance in sensor design.2-4 In sensor applications, electrode selectivity, which is also needed, is typically acquired by modifying the surface of an UME with polymers, including biopolymers. However, polymer morphology5 is often difficult to control,6-8 even when ultrathin polymer films are obtained by controlled electropolymerization.9 Molecular self-assemblies,10-21 allow greater control over the architecture and response of the surface coatings.

Chart 1. Schematic Illustration of Thioctic Acid Self-Assembled Monolayer on Au

The fundamentals of design of an ultrathin thioctic acid (1,2dithiolane-3-pentanoic acid, TA) monolayer UME, with emphasis on selectivity and sensitivity of the monolayer, are the subject of this work. Chart 1 illustrates the self-assembled monolayer (∼8 Å thick) of TA on gold exploited in this study. The five-carbon alkane chain TA, a carboxyl-terminated alkane disulfide, assembles on gold via the disulfide forming the ultrathin monolayer, with the carboxyl groups facing the solution.20,21 The charge density at the TA monolayer/solution interface is controlled by the solution pH.20,21 The results shown here illustrate the importance of the quality of the Au substrate 21 in determining the quality of the monolayer UME. Because of the ability to easily achieve highly efficient solution mass transport to the UME,22 permselectivity, defined here as the ability to suppress the response of electroactive ions present in

(1) Heinze, J. Angew. Chem., Int. Ed. Engl. 1993, 32, 1268 and references therein. (2) (a) Wightman, R. M.; May, L. J.; Michael, A. C. Anal. Chem. 1988, 60, 769A. (b) Adams, R. N. Prog. Neurobiol. 1990, 35, 297. (3) (a) Grunewald, R. A. Brain Res. Rev. 1993, 18, 123. (b) Cammack, J.; Ghasemzadeh, B.; Adams, R. N. Brain Res. 1991, 565, 17. (c) Pierce, R. C.; Miller, D. W.; Reising, D. B.; Rebec, G. V. Brain Res. 1992, 597, 138. (4) (a) Andrieux, C. P.; Hapiot, P.; Saveant, J.-M. Chem. Rev. 1990, 90, 723. (b) Wipf, D. O.; Michael, A. C.; Wightman, R. M. J. Electroanal. Chem. 1989, 269, 15. (c) Hsueh, C. C.; Brajter-Toth, A. F. Anal. Chem. 1993, 65, 1570. (5) Whitesides, G. M.; Mathias, J. P.; Seto, C. T. Science 1991, 254, 1312. (6) (a) White, H. S.; Leddy, J.; Bard, A. J. J. Am. Chem. Soc. 1982, 104, 4811. (b) Martin, C. R.; Rubinstein, I.; Bard, A. J. J. Am. Chem. Soc. 1982, 104, 4817. (c) Buttry, D. A.; Anson, F. C. J. Am. Chem. Soc. 1983, 105, 686. (7) Samec, Z.; Trojanek, A.; Samcova, E. J. Phys. Chem. 1994, 98, 6352. (8) Witkowski, A.; Brajter-Toth, A. Anal. Chem. 1992, 64, 635. (9) Hsueh, C. C.; Brajter-Toth, A. Anal. Chem. 1994, 66, 2458. (10) (a) Nuzzo, R. G.; Allara, D. L. J. Am. Chem. Soc. 1983, 105, 4481. (b) Nuzzo, R. G.; Zegarski, B. R.; Dubois, L. H. J. Am. Chem. Soc. 1987, 109, 733. (c) Nuzzo, R. G.; Fusco, F. A.; Allara, D. L. J. Am. Chem. Soc. 1987, 109, 2358.

(11) Mallouk, T. E., Harrison, D. J., Eds. Interfacial Design and Chemical Sensing; ACS Symposium Series 561; American Chemical Society: Washington, DC, 1994. (12) Ulman, A. Introduction of Thin Organic Films: From Langmuir-Blodgett to Self-Assembly; Academic Press: Boston, MA, 1991. (13) (a) Troughton, E. B.; Bain, C. D.; Whitesides, G. M.; Nuzzo, R. G.; Allara, D. L.; Porter, M. D. Langmuir 1988, 4, 365. (b) Bain, C. D.; Whitesides, G. M. Langmuir 1989, 5, 1370. (c) Bain, C. D.; Troughton, E. B.; Tao, Y. T.; Evall, J.; Whitesides, G. M.; Nuzzo, R. G. J. Am. Chem. Soc. 1989, 111, 321. (14) Bard, A. J.; Abruna, H. D.; Chidsey, C. E.; Faulkner, L. R.; Feldberg, S. W.; Itaya, K.; Majda, M.; Melroy, O.; Murray, R. W.; Porter, M. D.; Soriaga, M. P.; White, H. S. J. Phys. Chem. 1993, 97, 7147. (15) Murray, R. W., Ed. Molecular Design of Electrode Surfaces; Techniques of Chemistry Series XXII; John Wiley & Sons, Inc.: New York, 1992. (16) Porter, M. D.; Bright, T. B.; Allara, D. L.; Chidsey, C. E. O. J. Am. Chem. Soc. 1987, 109, 3559. (17) Finklea, H. O.; Avery, S.; Lynch, M.; Furtsch, T. Langmuir 1987, 3, 409. (18) Miller, C.; Cuendet, P.; Gratzel, M. J. Phys. Chem. 1991, 95, 877. (19) (a) Sun, L.; Johnson, B.; Wade, T.; Crooks, R. M. J. Phys. Chem. 1990, 94, 8869. (b) Malem, F.; Mandler, D. Anal. Chem. 1993, 63, 67. (20) Cheng, Q.; Brajter-Toth, A. Anal. Chem. 1992, 64, 1998. (21) Cheng, Q.; Brajter-Toth, A. Anal. Chem. 1995, 67, 2767.

4180 Analytical Chemistry, Vol. 68, No. 23, December 1, 1996

S0003-2700(96)00329-0 CCC: $12.00

© 1996 American Chemical Society

solution, of a film coated on an UME can be easily investigated. When the films coated on UMEs are ultrathin, investigations of the processes controlled by the film are facilitated because of the highly sensitive current responses that can be achieved for electroactive probes.9,21 UMEs with ultrathin self-assembled films have been used in the investigation of the fundamentals of electron transfer.23 It has been shown previously, from the investigation of the current responses of electroactive ions present in solution, that the permselectivity of the TA monolayer electrodes can be controlled by controlling the solution pH, through control of the charge density at the monolayer/solution interface, and through control of the monolayer hydrophobicity, by coassembly of TA with an alkanethiol.21 The pH and the composition-controlled permselectivity result from the film effectively suppressing the response of electroactive anions or cations present in solution. The results obtained here illustrate, additionally, permselectivity of the TA monolayer UME to nonelectroactive ions. TA UMEs were also tested in amperometric pH sensing. Selfassembled monolayers have been used previously in the design of amperometric pH sensors.24,25 By immobilizing a ferrocenyl thiol and a quinone thiol on a microelectrode array, the quinone pH-sensitive half-wave potential, E1/2, was used to indicate the pH, while pH-insensitive E1/2 of the ferrocene center was used as a reference.24,25 In the application illustrated here, TA monolayer was used as a pH sensor in the determination of the pKa of TA bound to the surface. EXPERIMENTAL SECTION Reagents. Hexaammineruthenium(III) chloride [Ru(NH3)6Cl3] was purchased from Alfa Products. Potassium ferricyanide [K3Fe(CN)6] was obtained from Fisher Scientific. Thioctic acid was purchased from Aldrich. Tris [tris(hydroxymethyl)aminomethane hydrochloride] was purchased from Sigma. All chemicals were used as received. Aqueous solutions were freshly prepared from doubly distilled, deionized water. Prior to use, solutions were purged with nitrogen for at least 5 min. Electrode Preparation. A gold wire (d ) 12.7 µm, 99.9%, Johnson Matthey) was mounted on a copper wire with silver epoxy (Type 410E, Epoxy Technology). The gold wire and the connecting joint were sealed in a glass capillary with an inert epoxy (Epoxi-Patch, Dexter Corp., Hysol Division). The electrode was then baked in an oven at 50 °C for 24 h to complete the curing of epoxy. Before use, the electrode was first polished on 600-grit polishing paper (Mark V Laboratory, East Granby, CT) and then on a polishing cloth, using 0.1-µm γ-alumina slurry (Fisher Scientific). After polishing, the electrode was sonicated in distilled water for 5 min. The polished gold wire was then dipped in dilute aqua regia (HCl/HNO3/H2O 3:1:16) for 5 min, washed thoroughly, and sonicated for 5 min in deionized water. This etched Au was used as the bare electrode. For modification with TA, the etched (22) (a) Wightman, R. M.; Wipf, D. O. In Electroanalytical Chemistry; Bard, A. J., Ed.; Marcell Dekker: New York, 1989; Vol. 15, p 267. (b) Fleischmann, M.; Pons, S.; Rollison, D. R. In Ultramicroelectrodes; Schmidt, P. P., Ed.; Datatech Systems: Marganton, WV, 1987. (c) Bond, A. M.; Oldham, K. B.; Zoski, C. G. Anal. Chim. Acta 1989, 216, 177. (d) Zoski, C. G.; Bond, A. M.; Allinson, E. T.; Oldham, K. B. Anal. Chem. 1990, 62, 3724. (e) Carpo, R.; Coppola, S.; Fromter, E. Pflugers Archiv. 1994, 429, 193. (23) Forster, R. J.; Faulkner, L. R. J. Am. Chem. Soc. 1994, 116, 5453. (24) Rubinstein, I. Anal. Chem. 1984, 56, 1135. (25) Hickman, J. J.; Ofer, D.; Laibinis, P. E.; Whitesides, G. M.; Wrighton, M. R. Science 1991, 252, 688.

Figure 1. Cyclic voltammograms of 5 mM Fe(CN)63- in 0.1 M KCl and 10 mM tris (pH 7.4): bare Au UME (a), TA monolayer UME on mechanically polished (b), and aqua regia etched Au UME (c). Scan rate 25 mV/s; electrode diameter 12.7 µm.

electrode was immersed in 0.2% TA/ethanol solution for 30 min and then rinsed extensively with deionized water. Typical reproducibility is ∼1%, and the electrodes have been used for up to 50 measurements without showing any deterioration in the reproducibility. All measurements reported here have been carried out at least four times (see tables for details). The TA UMEs made using the above procedure typically show a characteristic sigmoidal response at subsecond time scales. However, after several scans in acidic solutions of Fe(CN)63-, peakshaped response was observed. When the electrode was subsequently transferred to Fe(CN)63--free, 0.1 M KCl solution (pH 7.4), Fe(CN)63- response was observed at first and then gradually disappeared. Since the response did not disappear when the electrode was transferred to Fe(CN)63--free, acidic 0.1 M HClO4 solution (pH 1.5), low pH must play an important role in the apparent adsorption of Fe(CN)63-. It is presumed that at low pH strong interactions between Fe(CN)63- and the electrode surface (possibly involving the adjacent epoxy, which can be protonated at low pH) may occur, while at higher pH the interactions do not take place. In low pH measurements with Fe(CN)63-, long exposure of the electrode to the solution was avoided; before each measurement the surface was polished in order to obtain reproducible results. Typical reproducubility was between 1 and 4% with 20 electrodes tested. Instrumentation. A Bioanalytical Systems electrochemical analyzer (BAS-100) with a home-made preamplifier26 was used in the electrochemical measurements. Data were transferred to a Northgate 386 personal computer for analysis. A two-electrode setup was employed with a self-assembled monolayer microelectrode as the working electrode, and the SCE in saturated KCl as both the auxiliary and the reference electrode. All potentials are reported vs SCE at room temperature unless specified. RESULTS AND DISCUSSION Effect of Au Fiber Treatment on Voltammetric Response of TA UME. Figure 1 shows the response of Fe(CN)63-, in 0.1 M KCl at pH 7.4, on the bare (Figure 1a) and on the TA-modified Au fiber UME (Figure 1b,c). Fe(CN)63- shows a sigmoidal steadystate response on the bare Au and attenuated response on the (26) (a) Huang, H. J.; He, P.; Faulkner, L. R. Anal. Chem. 1986, 58, 2889. (b) Hsueh, C. C.; Brajter-Toth, A. Anal. Chim. Acta 1996, 321, 209.

Analytical Chemistry, Vol. 68, No. 23, December 1, 1996

4181

TA monolayer. The results in Figure 1b and c confirm that the TA monolayer is permselective, suppressing the response of Fe(CN)63- present in solution at pH 7.4 when the monolayer carboxylate head groups are predominantly negatively charged.20,21 The results in Figure 1b were obtained on an Au fiber that was mechanically polished with alumina slurry while the results in Figure 1c were obtained on an Au fiber that was additionally etched in dilute aqua regia for 5 min, before the monolayer was assembled.27 The results show a significant improvement in permselectivity at the monolayer UME prepared on acid-etched Au. This confirms the previously reported correlation between the method of the Au surface treatment and the quality of the monolayer self-assembled on Au.21,27 The results also illustrate the validity of acid etch treatment as a method27 of fabrication of Au fiber UME substrates. Permselectivity of TA Monolayer UMEs. Figure 2a shows a well-developed response of Fe(CN)63- on the TA monolayer UME at pH 1.5, which is in contrast to the suppressed response observed at pH 7.4 (Figure 1c). Figure 2b illustrates the similar well-developed response of Ru(NH3)63+ on the TA monolayer UME at pH 7.4, which can be contrasted with the suppressed response of Ru(NH3)63+ at pH 1.5 (Figure 2c). The results in Figures 1 and 2 confirm the behavior previously observed at the TA monolayer macroelectrodes, where at high pH the current was suppressed at the Fe(CN)63- reaction potential, leading to apparent slow kinetics of Fe(CN)63-,20,21 and where there was a significant decrease in current at low pH at the Ru(NH3)63+ reaction potentials.28,29 It was suggested previously that low Ru(NH3)63+ current at low pH indicated transport problems at the monolayer during Ru(NH3)63+ reaction.20,21 Since, as shown in Figure 2c, Ru(NH3)63+ current is significantly more suppressed at the TA monolayer UME under conditions of efficient solution mass transport, the results confirm that the monolayer is the source of the low currents. The i-E curves, such as in Figures 1a, 2a, and 2b, were obtained in different electrolytes and were analyzed in order to characterize the kinetics at the monolayer UME. The results are summarized in Figure 3 and Table 1. The reciprocals of the slopes of the log (iL - i)/i vs E plots30,31 that were obtained, except for the slopes of Ru(NH3)63+ in F- at pH 7.4 (Table 1), which are discussed later, are all close to 59 mV, confirming fast kinetics. Electrolyte Effect on Permselectivity of TA UMEs. In previous studies a small effect of electrolyte composition on the monolayer electrode current was observed.21 This was unexpected for the predominantly diffusion-controlled processes and was taken as an indication of the influence of the monolayer on the reaction.20,21 The apparent high permselectivity of TA monolayer UMEs, which results in low currents of Ru(NH3)63+ at low pH (Figure 2c) and of Fe(CN)63- at high pH (Figure 1b,c), was previously attributed for Fe(CN)63- to slow electrochemical kinetics.21 In this work it was observed that Ru(NH3)63+ currents were not only (27) Creager, S. E.; Hockett, L. A.; Rowe, G. K. Langmuir 1992, 8, 854. (28) Becka, A. M.; Miller, C. J. J. Phys. Chem. 1993, 97, 6239. (29) Delahay, P. Double Layer and Electrode Kinetics; Wiley-Interscience: New York, 1965. (30) (a) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications; John Wiley & Sons: New York, 1980; p 160. (b) Howell, J. O.; Wightman, R. M. Anal. Chem. 1984, 56, 524. (31) Pena, M. J.; Fleischmann, M.; Garrard, N. J. Electroanal. Chem. 1987, 220, 31.


Figure 2. Cyclic voltammograms on the bare Au UME (solid line) and TA UME (dash line): 5 mM Fe(CN)63- in 0.1 M HClO4 (pH 1.5) (a), 5 mM Ru(NH3)63+ in 0.1 M KCl and 10 mM tris (pH 7.4) (b), and 5 mM Ru(NH3)63+ in 0.1 M KCl (pH 1.0) (c). Conditions as in Figure 1.

low at Ru(NH3)63+ reaction potentials but remained low at high overpotentials, not reaching steady-state values expected from the electrode radius (Figure 2c). Ru(NH3)63+ i-E curves were investigated in different electrolytes to identify the reasons for the low currents. Table 2 summarizes the steady-state current values at the monolayer, iml, and the bare Au, ibare, together with the current values calculated from the electrode radius, icalc, assuming a disk geometry for the UME. The current ratios iml/ibare are also summarized in Table 2. In phosphate, the ratio points to a ∼18% decrease in the steadystate current at the TA UME compared to the current at the bare Au (Figure 4a). In F- the decrease is ∼95%, and the steady-state current of Ru(NH3)63+ is only slightly higher than the background current (Figure 4b). The iml for Ru(NH3)63+ is the same in sodium

Table 1. Half-Wave Potentials and Nernstian Slopes for Fe(CN)63- and Ru(NH3)63+ in Different Electrolytesa,b probe

electrode

electrolyte

pH

E1/2 (V)

reciprocal of slope of log ((iL - i)/i)) vs E plotb

Fe(CN)63-

bare Au UME bare Au UME TA UME bare Au UME bare Au UME bare Au UME bare Au UME TA UME TA UME TA UME TA UME TA UME passivated in F-

0.1 M HClO4 0.1 M KCl and 10 mM tris 0.1 M HClO4 0.1 M KCl and 10 mM tris 0.1 M phosphate 0.1 M NaF and 10 mM tris 0.1 M NaF and 10 mM tris 0.1 M KCl and 10 mM tris 0.1 M phosphate 0.1 M NaF and 10 mM tris 0.1 M NaF and 10 mM tris 0.1 M KCl and 10 mM tris

1.5 7.4 1.5 7.4 7.4 7.4 9.6 7.4 7.4 7.4 9.6 7.4

0.36 0.18 0.37 -0.18 -0.22 -0.19 -0.18 -0.19 -0.23 -0.20 -0.19 -0.20

56 62 56 58 59 56 55 63 63 84 55 69

Ru(NH3)63+

a

Scan rate, 25 mV/s. b Number of samples, n ) 4.

Figure 3. Log ((iL - i)/i) vs E: (A) 5 mM Fe(CN)63- in 0.1 M HClO4 (pH 1.5) on the bare Au UME (a); TA UME (b) and in 0.1 M KCl and 10 mM tris (pH 7.4) on the bare Au UME (c). (B) 5 mM Ru(NH3)63+ in 0.1 M KCl and 10 mM tris (pH 7.4) on bare Au UME (a) and TA UME (b). Conditions as in Figure 1.

and potassium phosphate electrolytes, indicating that iml is determined by the anion. The results in Table 2 show that the calculated current, icalc, is lower than ibare in Cl- by an average of ∼13%. Greater deviations

are observed in phosphate. The differences between icalc and ibare are likely due to the experimental deviations from the microdisk electrode geometry assumed in the calculations. The log plot kinetic analysis30,31 of the i-E curves in different electrolytes (Figure 3) is summarized in Table 1. Ru(NH3)63+ kinetics remain fast when iml decreases in phosphate, as indicated by the ∼59 mV reciprocal slope of the log plot. In F-, however, the major decrease in iml leads to a significantly higher reciprocal slope of the same plot, indicating slower kinetics. Faster kinetics in F- (and currents approaching ibare) can be obtained by increasing the solution pH to 9.6 (Figure 5 and Table 1) or can be approached at pH 7.4 by switching the electrode to a solution of Cl- electrolyte (the steady-state current reaches 3 nA compared to 10.3 nA on the freshly made TA UME in Cl- at pH 7.4), but the response does not change after transfer to phosphate. For Fe(CN)63- at low pH (Table 2) iml values are equivalent to icalc in Cl- and ClO4-. A positive shift in E1/2 of Fe(CN)63-/4- is observed at low pH (Table 1) as expected due to the protonation of Fe(CN)63-.32 Amperometric pH Sensing with TA UME with Fe(CN)63as the Mediator. The results described above confirm the influence of the charge of the TA head group on the electrochemical response of Fe(CN)63- at the monolayer UME. At high pH, when the COOH groups are dissociated, Fe(CN)63- current is suppressed; as the solution pH is lowered, Fe(CN)63- current increases to a maximum value that can be predicted from the steady-state current equation.30,31 Since the magnitude of the steady-state current changes with solution pH, the current can be used as a measure of the solution pH. From an equivalent of an amperometric pH titration, the current can be used to determine the acid dissociation constant of the COOH head groups of TA in the monolayer. Figure 6 (closed circles) shows the results of such a titration monitored at the TA UME by recording Fe(CN)63- current. The solution pH was simultaneously monitored with a conventional glass electrode, and the results are shown in Figure 6 as closed squares. The pKa determined for the surface-bound TA is ∼7.0, significantly higher than the solution pKa ) 5.0;33 dissociation of the surface COOH groups occurs over ∼3 pH units, as shown by the titration. Fe(CN)63- current can be recorded instanta(32) Hanania, G. I. H.; Irvine, D. H.; Eaton, W. A.; George, P. J. Phys. Chem. 1967, 71, 2022. (33) Wagner, A. F.; Folkers, K. Vitamins and Coenzymes; Interscience: New York, 1964; p 247.


4183

Table 2. Voltammetric Results for Fe(CN)63- and Ru(NH3)63+ on the TA and Bare Au Microelectrodes probe

electrolyte

pH

icalc (nA)

iml (nA)

ibare (nA)

iml/ibare (%)

5 mM Fe(CN)63-

0.1 M HClO4 0.1 M KCl and 10 mM tris 0.1 M KCl and 10 mM tris 0.1 M KCl and 10 mM tris 0.1 M phosphate 0.1 M NaF and 10 mM tris 0.1 M NaF and 10 mM tris

1.5 7.4 1.0 7.4 7.4 7.4 9.6

9.3a 9.3a 9.3b 9.3b 6.8c d d

10.6 ( 0.1

11.0 ( 0.3 10.4 ( 0.1 9.6 ( 0.4 10.8 ( 0.1 9.5 ( 0.1 10.5 11.0

96

5 mM Ru(NH3)63+

10.3 ( 0.1 7.8 ( 0.2 0.5 9.8

a Calculated with D ) 7.6 × 10-6 cm2/s in 0.1 M KCl.42 b Calculated with D ) 7.6 × 10-6 cm2/s in 0.1 M KCl.43 0 0 × 10-6 cm2/s in 0.1 M phosphate.44 d Not calculated due to lack of diffusion coefficient data.

c

95 82 5 89

Calculated with D0 ) 5.5

Figure 5. Cyclic voltammogram of 5 mM Ru(NH3)63+ in 0.1 M NaF (pH 9.6) on the bare Au UME (solid line) and the TA UME (dash line). Other conditions as in Figure 1.

Figure 4. Cyclic voltammograms of 5 mM Ru(NH3)63+ on the bare Au UME (solid line) and TA UME (dash line) in 0.1 M phosphate (pH 7.4) (a); 0.1 M NaF and 10 mM tris (pH 7.4) (b). Other conditions as in Figure 1.

neously after the solution pH is changed, since the response of the UME is rapid. Ward and co-workers used a quartz crystal microbalance (QCM) to investigate the properties of carboxylate-terminated long-chain thiols, self-assembled at the monolayer/water interface.34 They reported that the pKa of HS(CH2)15COOH monolayer was ∼8.0 and that the transition region was ∼4 pH units. Compared to the solution pKa, which is typically between 4 and 5, this surface-bound carboxylic acid also showed higher pKa. Similar high pKa values have been reported from contact angle measurements by Bain and Whitesides for the carboxylateterminated thiol monolayers.13b Amperometric pH measurements described here provide an alternative route for the measurement of pKa of surface-bound acids and bases. Ward et al. observed that the QCM frequency did not change when the experiments were performed with (34) Wang, J.; Frostman, L. M.; Ward, M. D. J. Phys. Chem. 1992, 96, 5224.


Figure 6. Titration curve of 0.1 M HClO4 with 2 M NaOH: 9, monitored potentiometrically; b, monitored by amperometry with 5 mM Fe(CN)63-.

monolayers prepared from HSCH2CH2COOH, indicating that the measurements may be insensitive at monolayers of small thickness.34 Considering the electrode area and assuming ∼1 × 10-10 mol/ 2 cm as the monolayer surface coverage by the carboxylate groups, ∼1.27 × 10-16 mol of H+ is required to fully protonate this monolayer if the buffer effect is ignored. This value is equivalent to ∼7.6 × 107 H+ ions, which shows that in the present design TA UMEs (∼10 µm radius) are capable of sensing extremely small numbers of protons. UMEs with monolayers, such as the TA

monolayer may, therefore, find use in amperometric pH measurements at microlevels. For example, for cultured cells such as a fibroblasts, pumping rates of 108 protons/s are often obtained in steady state.35 CONCLUSIONS The results of this work confirm the superiority of acid-etched gold27 fiber for producing high-quality TA monolayer UMEs. The quality of the monolayers is verified by their excellent permselectivity, defined here as the ability to efficiently suppress the response of electroactive ions in solution, and their high sensitivity. High sensitivity of TA monolayer UMEs results from the ultrathin thickness of the monolayer (∼8 Å).9,21 In this study, the high quality and the high sensitivity of the TA monolayer UMEs allowed unequivocal detection of a strong electrolyte effect. This effect is well illustrated at pH 7.4 with Ru(NH3)63+ as a probe where the most sensitive response, corresponding to the steady-state current predicted from the electrode radius, was observed in the presence of the most hydrophobic electrolyte anion, Cl-;36 a smaller steady-state current was measured in the more hydrophilic phosphate, indicating that the magnitude of the steady-state current is related to the hydrophobicity of the electrolyte anion36 and to the monolayer hydrophobicity. Electrolytes involved in the hydrophobic interactions with the monolayer influenced the magnitude of the current but did not appear to significantly influence the electrochemical kinetics of the probe. However, with F- as the electrolyte, interactions with the monolayer, other than hydrophobic interactions, most likely involving hydrogen bonding with the partly protonated carboxylate head groups of the monolayer, were indicated from the results. At high pH (9.6), which favors complete dissociation of the carboxylate head groups at the monolayer, the effect of Finteractions disappeared; the effect was very small at low pH after the electrode was exposed to Cl- but did not disappear in phosphate. F- had an effect on the current as well as on the kinetics of the electroactive probe. The results indicate that hydrogen-bonding interactions, unlike hydrophobic interactions, can slow the electrode kinetics. Slow kinetic behavior, similar to that observed in F-, was observed in Cl- at low pH, showing presumably the effect of increased hydrogen bonding between the carboxylic acid head groups of the monolayer. The interactions may account for the attenuated response of Ru(NH3)63- at low pH in HCl as the electrolyte; the well-developed response of Fe(CN)63-/4- in the same electrolyte at the same low pH may result, in part, from proton bridging36 between Fe(CN)63-/4- and the monolayer, supported by the tenacity of Fe(CN)63-/4- toward adsorption at the monolayer at low pH, and from possible proton transport in the monolayer during Fe(CN)63-/4- reaction. (35) McConnell, H. M.; Owicki, J. C.; Pearce, J. W.; Miller, D. L.; Baxter, H. G.; Wada, H. G.; Pitchford, S. Science 1992, 257, 1906. (36) Marcus, Y. Biophys. Chem. 1994, 51, 111.

It can be argued that the apparent blocking of the electrode (or apparent decrease in the electrode radius), as well as the slower kinetics, in some electrolytes, may result from monolayer pinholelike behavior and a kinetic distribution over the monolayer surface. Additional verification of film microstructure, which is difficult for the ultrathin ∼8-Å-thick monolayers, would help verify the film model. A practical application of TA UMEs was illustrated in the determination of the pKa of the surface carboxyl groups. The method may provide complementary information about surface structure with techniques such as QCM. The ∼2 orders of magnitude higher pKa of ∼7 determined for the carboxylate groups in the monolayer points to the surface environment influencing the pKa. The result is consistent with previous reports; the high surface pKa may be due, as previously proposed, to hydrogen-bonding stabilization of the acid form of the COOH groups in the monolayer.34 Such hydrogen-bonding interactions are consistent with the permselectivity observed at low pH in Cland F-. In addition to the increase in the pKa of the carboxylate groups of TA immobilized in the monolayer, titration of the surface carboxylate groups indicated some cooperativity in dissociation of these groups, normally not attributed to weak acid dissociation. Similar behavior has recently been observed for surface functionalized polymers37-39 and has been related to polymer compactness37 and local hydrophobicity and charge effects,38 all consistent with the monolayer model discussed here. Finally, TA UMEs as ultrathin films with hydrophilic carboxylic acid head groups are anticipated to be useful in sensor design. Willner and Riklin have recently reported enzyme electrodes prepared by covalent immobilization of enzymes onto cystamine assembled on Au electrodes.40 Related strategies with TA monolayer electrodes can be envisioned.41 ACKNOWLEDGMENT This work has been supported in part by the U.S. Army and by the Division of Sponsored Research at the University of Florida. We thank Marc Porter for a preprint of his manuscript, and David Bergbreiter, Philippe Buhlmann, and Rudy Seitz for helpful discussions and for bringing refs 36 and 37 to our attention. We acknowledge Jeff Flowers for his help with the final version of the manuscript. Received for review April 2, 1996. Accepted September 20, 1996.X AC960329W (37) Thomas, J. L.; You, H.; Tirrell, D. A. J. Am. Chem. Soc. 1995, 117, 2949. (38) Bergbreiter, D.; Bandelia, A. J. Am. Chem. Soc. 1995, 117, 10589. (39) Seitz, R., personal communication, 1996. (40) Willner, I.; Riklin, A. Anal. Chem. 1994, 66, 1535. (41) Duan, C. M.; Meyerhoff, M. E. Anal. Chem. 1994, 66, 1369. (42) Stackelberg, M.; Pilgram, M.; Toome, V. Z. Electrochem. 1953, 57, 342. (43) Bard, A. J.; Crayston, J. A.; Kittelson, G. P.; Shea, T. V.; Wrighton, M. S. Anal. Chem. 1986, 58, 2321. (44) Baur, J. E.; Wightman, R. M. J. Electroanal. Chem. 1991, 305, 73. X Abstract published in Advance ACS Abstracts, November 1, 1996.


4185

Permselectivity, Sensitivity, and Amperometric pH Sensing at Thioctic

Recommend Documents