摘要:

ANALYST, AUGUST 1992, VOL. 117 1265 Gas and Vapour Detection With Poly(pyrro1e) Gas Sensors* Jonathan M. Slater and Esther J. Watt Analytical Science Group, Birkbeck College, University of London, 29 Gordon Square, London WCIH OPP, UK Neville J. Freeman, lain P. May and Donald J. Weir Long Range Reseach Laboratory, Hirst Research Centre, East Lane, Wembley HA9 7PP, UK The response mechanism of the conducting polymer poly(pyrro1e) to a selection of gases and vapours was investigated using two techniques: measurement of resistance change and mass changes using a piezoelectric quartz crystal microbalance with the objective of characterizing responses for incorporation in sensor arrays. Bromide-doped films were exposed to methanol, hexane, 2-2-dimethylbutane, ammonia and hydrogen sulfide.Polymers of different thicknesses were also exposed to methanol vapour and the response profiles were studied. The responses were all of a Fickian type except the piezoelectric signal, which exhibited an anomalous non-Fickian response to methanol. This suggests that the poly(pyrro1e) resistance changes frequently observed are partly due to one stage in the two-stage sorption perhaps involving the swelling of the polymer. It was concluded that the response mechanism of poly(pyrro1e) sensing of different gases and vapours is due to a mixed response involving electronic effects and physical effects. Keywords: Methanol detector; piezoelectric gas detector; gas sensor; poly(pyrrole) sensor Poly(pyrro1e) is a conducting polymer whose properties, preparation and uses have been described previously in many publications.1-6 One of its many potential uses is as a gas-sensing material as it exhibits changes in its conductivity in different gas environments.7-11 The ease of preparation of different poly(pyrro1e) layers by electrochemical deposition makes it potentially an ideal material for producing sensor arrays; however, its application is limited as it is not yet possible to predict the response of the material to vapours and prepare sufficiently reproducible devices.Many publications are available on such work but, despite this volume of research, the mechanism of the observed response remains unclear. The most widely accepted theory involves ascribing a p-type semiconductor nature to the polymer. The conduction mechanism along the polymer chains is via the movement of polarons and bipolarons while the between chain conduction depends on electron hopping.Electrophilic gases withdraw electrons from the polymer, thereby increasing the number of charge carriers and hence the conductivity, while nucleophilic gases have the opposite effect resulting in a conductivity decrease.12 While this theory is a satisfactory explanation of the response mechanism to electroactive species such as NH3 and NO,, it does not explain the magnitude of the response to organic vapours, many of which give large reversible resistance changes at room temperature. A second response mechanism appears to be involved for some vapours and this must be characterized if effective sensor arrays are to be designed.Previously, two methods of monitoring the poly(pyrro1e) gas sensor response to ammonia gas have been reported:13 resistance responses and mass change profiles of the polymer as it is exposed to the gas on a quartz-crystal microbalance. Such a piezoelectric gas detection system works on the principle that a gas adsorbed onto a crystal coating material will cause a mass change, which is reflected in the frequency of the crystal. 14.15 The relationship between frequency change and mass change is described by the Sauerbrey equation,l6 AF = -2.3 X lo6 F$ AmlA where Am is the change in mass of the crystal (g), AF the related frequency change (Hz), A the gas-sensitive area (cm2) * Presented at the meeting on Analytical Applications of Chemi- cally Modified Electrodes, Bristol, UK, January 7-8, 1992.and Fo the initial frequency of the quartz crystal (MHz). However, this equation assumes that a frequency change is solely dependent on the mass change at the crystal surface and takes no account of the physical properties of the coating material. A number of workers have expanded on Sauerbrey's original theory by including factors such as wave propagation,17 energy losses caused by changes in the crystal acoustic impedance18 and stress19 in the material. Although it is important to bear in mind such errors in the equation when interpreting results, their effect is minimal if the mass of crystal coating does not exceed 2% of the crystal mass. The objective of the work described in this paper is to use this technique in conjunction with measurements of change in resistance and thickness in order to interpret the response mechanism of poly(pyrro1e) films to vapours such as methanol.Experimental Apparatus The piezoelectric microbalances used were 10 MHz quartz crystals plated with gold electrodes (McKnight). The measurement electronics were as described previously,13 consisting of a double crystal oscillator (driving a blank, reference crystal and coated sensor crystal) powered by a 5 V d.c. power supply and giving an analogue output related to the difference in crystal frequencies. This could be displayed either on a chart recorder or transferred to an IBM PC via an analogue-to-digital interface, where the frequency profiles can be stored using specially written data logging software.The frequency of the sensor crystal could also be monitored directly on a frequency counter. Conductivity measurements were made with a Degussa dew-point sensor, a device consisting of a meander of interdigitated platinum electrodes with 10 pm spacing, which can be bridged by the poly(pyrro1e) during polymer growth thus allowing measurement of the polymer conductivity. The resistance, and thus the conductivity, was monitored using a simple Wheatstone bridge set-up to give an analogue output to a y-t chart recorder. Measurements of changes in polymer thickness were made with modified thermomechanical analysis (TMA) equipment. By using only the thickness-measuring facility, consisting of a quartz platform and counter-weighted probe that rests on the1266 ANALYST, AUGUST 1992, VOL.117 25 I 1 Sample gas Yitrogen (100%) !' v2 F2 I 1 Dilution F4 v3 Solenoid :1 Crystal cell 4!-!4Lu valves Fig. 1 needle valve Schematic diagram of a gas blender; F = flow meter and V = sample placed on the platform, a linear variable transformer converted the probe movements (caused by thickness changes) to analogue outputs, which were then monitored on a chart recorder. By placing one of the coated Degussa conductivity sensors in the apparatus, the correlation between the conductivity and profile changes in the polymer could be examined. The microstructure of the poly(pyrro1e) film was investi- gated using a Jeol JSM 35CF scanning electron microscope. Gas-Vapour Generation Apparatus The test gases and vapours were introduced into the sensors with the nitrogen carrier gas in a double impinger exposure cell, as described previously.13 However, the method of gas and vapour generation was different.A system (as shown in Fig. 1) was designed consisting of a series of needle valves controlling the gas flow rates, which could be monitored via a bank of flow meters. Target samples were produced in two ways. Organic vapours were generated by bubbling a stream of nitrogen carrier gas (via gas wash bottles) through the volatile liquids (hexane, cyclohexane, 2-2-dimethylbutane and methanol), thus producing a continuous flow of saturated vapour, the concentration of which depended on the vapour pressure of the liquid. Analyte samples that are normally in a gaseous state at room temperature were prepared by dilution of 99% pure ammonia and hydrogen sulfide (lecture bottles, Aldrich).The apparatus allows switching by solenoids between a continuous flow of purge and sample gases, generated by one of the above methods. Preparation of Poly(pyrro1e) Poly(pyrro1e) films were prepared by the electrochemical oxidation of the pyrrole monomer (freshly distilled, Aldrich) in a KBr (Fisons) solution using a Princeton Applied Research 174A polarographic analyser to perform cyclic voltammetry of a three-electrode potentiostatic cell. The cell consisted of a saturated calomel reference electrode, a glassy carbon auxiliary electrode and either the gold electrode of the piezoelectric crystal or the platinum surface of the Degussa sensor as the working electrode. Care was taken to ensure similar orientation of the working and auxiliary electrodes during each polymerization reaction.- 20 0 20 40 60 80 100 Film coating/kHz Fig. 2 Mass of poly(pyrro1e) film versus QCM response to a 5 min exposure to methanol The procedure for poly(pyrro1e) deposition and conditioning has been described in detail previously.20 The polymer film is formed on the working electrode as it is cycled between 0.15 and 0.9 V in an aqueous solution containing pyrrole (0.05 mol dm-3) and KBr (1.0 mol dm-3). The number of cycles in this stage determines the final mass of polymer deposited. Once prepared, the film is rinsed and conditioned by further cycling in a solution containing the electrolyte only (aqueous 1.0 mol dm-3 KBr), i.e., there is no further pyrrole available for polymerization.Cyclic voltammetry between -0.8 and 0.7 V causes the polymer to be switched successively between its reduced, non-conducting and oxidized, conducting form; doping ions flow in and out of the polymer film. It is anticipated that this procedure will lead to reproducible ion doping between polymer films. Finally the potential was held at 0.7 V for 5 min in this second solution as the current decays towards a steady state, thus giving polymers of maximum oxidation and doping concentration. Following this process, the polymer films were rinsed thoroughly in de-ionized water and dried in an oven at 100 "C for 1 h, after which they were stored in a desiccator until required. Results Dependence of Sensor Response on Mass of Coating Material The effect of poly(pyrro1e) thickness on the response to methanol vapour was examined by laying down polymer films of different masses onto the piezoelectric crystals. The mass was controlled by the number of cycles used in stage 1 of the preparation of poly(pyrro1e) and ranged from 4 kHz (0.5 cycles) to -85 kHz (10.5 cycles) frequency loadings.Five sensors of approximately the same thickness were prepared and exposed to pulses (5 min) of methanol alternated with purging (5 min) with clean dried nitrogen for each of the five different polymer thicknesses tested. The decrease in mass (i.e., the increase in frequency) observed in some of the thinnest films is attributed to corrosion of the gold electrode during the poly(pyrro1e) conditioning due to minimal coverage of the gold by the polymer.A plot of frequency change caused by exposure to methanol versus total frequency loading of poly(pyrro1e) is shown in Fig. 2. The plot is linear with a high degree of scatter. However, as shall be shown later, exposure times of 5 min do not allow the complete sorption reaction to take place; the rate will vary according to polymer-film thickness. The readings (Fig. 2) are taken after 5 min when the reaction has reached a different point in the two stage process for each film thickness. Thus a plateau-shaped plot, not a linear plot, would be expected as the thinner films are closer to their complete response. It is possible that a plateau-type plot would be obtained at greater thicknesses (this could not be achieved owing to crystal loading constraints).ANALYST, AUGUST 1992, VOL.117 1267 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0 8 0 1 2 3 0 0 4 0 8 1 2 1 6 2 0 I 2 4 N z . & 2.8 5 0.20 1 .oo 1.80 2.60 3.40 0 1 2 3 0 - 4 -8 -12 -16 - 20 - 24 0 MeOH ( d) I I I 1 2 3 Time/lO3 s Fig. 3 QCM response profiles of poly(pyrro1e) of different thick- nesses to 5 min pulses of methanol: (a) thinnest films, frequency loading = 422 Hz; (b) frequency loading = - 15 137 Hz; (c) frequency loading = -23917 Hz; and (d) frequency loading = -58472 Hz The pattern of response profiles also changes with different polymer thicknesses (Fig. 3). The thinnest films show an anomalous inflection in the absorption curve after about a 2 min exposure to methanol. This inflection changes in position with subsequent exposures (Fig. 4).Thicker films give plots with similar inflections exhibiting an increase in gradient as the exposure continues. This is not an expected Fickian-type -20 I I I I I I I I 0 2 4 6 8 10 12 14 16 tus Fig. 4 Plot of frequency change versus the square root of time for the exposure of poly(pyrro1e) to methanol vapour: A, first exposure; B, third exposure; C, fourth exposure; and D, fifth exposure adsorption which would show an ever decreasing gradient in the curve that would lead finally to a plateau at the saturation point of the polymer. More information can be gained by plotting these profiles against the square root of time (@). Scanning electron microscopy applied to the films used in these studies provides a useful comparison. Fig. 5 shows the morphology changes exhibited by films of different thick- nesses.A general progression can be seen from the thinnest to the thickest films. Once the film reaches a mass of approxi- mately 6.53 x 10-5 g cm-2 small nucleic structures become apparent which increase in size and number with thickness. These spherical stuctures are typical of a three-dimensional nucleation and growth reaction. The plates also show that the spheres are loosely packed on the surface and the polymer is fairly porous. The high porosity suggests that the surface area will increase with mass loading and thus the relationship between thickness, mass loading and surface area is not easily determined. Although it is not possible to relate the mass-response profiles to either polymer layer thickness or area, the thickest coatings [Fig.3(d)] clearly show an enhanced response. Response to a Range of Vapours Piezoelectric A series of experiments was designed to test both the selectivity and sensitivity of the polymer-coated sensor towards the target gases. Three fresh sensors were prepared for each of the gases or vapours and were exposed to five 5 min pulses of the gas alternated with the nitrogen purge gas. The results and an example of the response to cyclohexane vapour are shown in Fig. 6. The largest response was observed to methanol vapour and the smallest, also the least reproducible, response observed to hydrogen sulfide gas. Response profiles did not vary significantly between the hexane vapours, all resembling the profile depicted in Fig. 6(b). The anomalous profiles of the methanol absorptions were not observed for the equivalent profiles of the other gases and vapours tested.Conductivity A similar set of experiments was carried out for the conductivity responses. The results are shown in Table 1. The conductivity data give good agreement with the piezoelectric response. However, it is important to notice that the ratio of responses varies between vapours suggesting that the relationship between mass and conductivity change is not direct. The response profile, shown in Fig. 7, for the conductivity response to methanol does not exhibit any of the anomalous characteristics of methanol absorption observed for the piezoelectric results. Comparison of the profiles also shows that the saturation plateau is reached at a much earlier stage in the conductivity measurements than the piezoelectric measurements.1268 ANALYST, AUGUST 1992, VOL.117 Fig. 5 Scanning electron micrographs (X3500 magnification) of poly(pyrro1e) films on the gold electrode of a piezoelectric crystal with a frequency loading of (a) - 15 566 Hz, (b) -23 917 Hz, (c) -58 472 Hz and (d) -880329 Hz Mechanical analysis Other workers have shown the effect of differing preparation conditions on the mechanical properties of poly(pyrro1e) .21.22 Although there are a wide range of reported values for properties such as tensile strength and the Young's modulus, most indicate that the polymer has elastic properties. In order to investigate the effect of dimensional changes on the conductivity of the polymer, equipment that was designed to 1600 1200 800 400 N h I 250 100 r' Cyclohexane Ammonia L - m Methanol L Hexane 2,P-Dimethyl- Hydrogen butane sulfide 6 i - LI -800 - 1200 0 20 40 60 80 Time/min Fig.6 (a) QCM response to gases and vapours averaged over three freshly prepared sensors for each sample (unshaded area) and the weighted response which is the ratio of the absolute frequency response to the mass of poly(pyrro1e) coating, (calculated from the Sauerbrey equation) (shaded area). (b) Example of response profile to cyclohexane vapour Table 1 Summary of conductivity responses Analyte Mean response/ Target gas preparation sz Methanol Bubbler 1011.4 Cyclohexane Bubbler 19.8 Hydrogen sulfide Gas blender 350.7 0 1000 2000 3000 Timels Fig. 7 Conductivity response profile to methanol vapour measure the thermal expansion of materials was utilized. The probe of the TMA apparatus was placed on the poly(pyrro1e) sample with the applied force of 0.01 N (1 g mass).The probe indicated the sample was compressed by this application of force. The thickness was then monitored for a period of 20 min and a slow expansion recorded. Resistance changes were measured during this process and the signal used to calculate an approximate percentage change in thickness. The actual thickness of the polymer was not accurately known as the sample examination took place on the Degussa sensor. The thickness of samples prepared under similar conditions was ~ 0 . 0 7 mm. A plot of conductivity against time (Fig. 8) shows a decrease in conductivity as the polymer expands.ANALYST, AUGUST 1992, VOL.117 26.10 I I f , 1269 26.05 m v) b 25.95 cT 25.90 .- 25.85 2 0 . 2 - 1 / 25.80 6 I I I I 0 5 10 15 20 25 Time/min Fig. 8 Plot of conductivity change and thickness change versus time 0 5 10 15 20 t' Is Fig. 10 Conductivity response versus the square root of time for the exposure of poly(pyrro1e) to methanol vapour LL -5 6 8 10 12 14 16 0 2 4 tx3 Fig. 9 QCM sorption (A) and desorption (B) profiles of methanol vapour Discussion In order to help elucidate the response profiles obtained for poly(pyrro1e) exposure to gas, they were drawn using a Fickian axis, i.e., mass change versus the square root of time, Fig. 9. A normal Fickian plot would initially be linear having a decreasing slope that would plateau as the absorbent becomes saturated, the diffusion coefficient being solely dependent on concentration.23 This type of plot for the interactions between poly(pyrro1e) and the gases tested agrees well with this expected profile.However, an analogous plot for the frequency response to methanol vapour shows different characteristics. These profiles show some similarities of the anomalous absorption of some organic vapours by polymers in their glassy state. Such systems have the following characteristics. (i) The mass versus square root of time curves of the sorption have a point of inflection, the gradient increasing sharply before decreasing to a final equilibrium value. (ii) The initial desorption rate is higher than the initial sorption rate but decreases in the later stages so that the two lines intersect.(iii) A polymer containing an initial concentration of sorbent (i.e., if the material has not fully recovered after the first exposure) shows an initial gradient larger than for a fresh polymer. Such anomalous behaviour is also characterized by a change from the glassy to rubbery state of the polymer during absorption of the sorbent, i.e., the polymer swells.24-26 Swelling of the polymer starts as it is exposed to the vapour; however, such molecular rearrangement does not take place instantaneously but will occur at a finite rate. As the polymer morphology changes the diffusion coefficient, D, will also vary. However, as there is a time lag between the sorbent entering the polymer and the changes in morphology (Fig. 9, d = 6), D will not reach its equilibrium value for a particular concentration as long as the polymer continues to change.In the early stages of exposure, the concentration of sorbent will be increasing rapidly and D will lag behind its equilibrium value changing slowly. During the later stages of sorption (Fig. 9, @ = 12), when the concentration change has slowed and diffusion can more closely match its equilibrium value, D would be expected to increase with concentration and the gradient will be higher in these later stages. As the polymer is in its swollen state after vapour adsorption, the value of D is already at a maximum. The rate will decrease towards lower equilibrium values as the material shrinks during desorption resulting in the plateau profile observed in Fig. 9, i.e., a desorption profile with a higher initial gradient than the sorption profile, which decreases as desorption proceeds.It should be noted that the desorption profile does not display a point of inflection, which may be due to removal of the sorbent before the swollen polymer can relax. The anomalous behaviour seen in the poly(pyrro1e)- methanol system fits this scheme. Tentative evidence comes from responses at varying thicknesses. Methanol sorption at different thicknesses shows that the point of inflection occurs at a later stage in the thicker films than in the thin films. A linear relationship between the frequency change caused by a 5 min exposure to methanol vapour and the mass of film deposited, not a plateauing curve as would be expected, is shown in Fig.2. During the 5 min exposure time, the films increase in mass by approximately 12-20%, irrespective of thickness. The sorption profiles, Fig. 3, show that the thinner films are closest to completion of the sorption reaction. This suggests that the initial rate of adsorption is much faster in the thicker films, a feature of their more porous nature (Fig. 5), supported by Fig. 3(d) when compared with Fig. 3(a)-(c). The initial response to methanol shown in Fig. 3 is larger than subsequent responses. Detailed plots of one such profile show an increasing gradient from exposure to exposure caused by the initial concentration of methanol absorbed in the films (Fig. 4). The absorption of methanol into poly(pyrro1e) appears to be a two stage process involving firstly the penetration of vapour into the polymer accompanied by swelling and then diffusion at an increased rate into the swollen rubbery material.Full recovery is not attained after the first exposure leading to a second exposure with a modified polymer containing initial concentrations of methanol. As swelling is the rate-determining step of the sorption, subsequent expo- sures will take the polymer to the same point in the sorption curve as the first 5 min exposure. When interpreting these profiles, it is important to account for the inaccuracies in the Sauerbrey equation. The swelling of the polymer will have some effect on the frequency response of the quartz crystal. However, the errors caused by phen- omena such as an acoustic impedance change are only of significance when the crystal loading is more than a 2% frequency shift of the original crystal frequency .18 The maximum total frequency shift, including that caused by the poly(pyrro1e) deposition, was less than 1% of the 10 MHz fundamental frequency of the crystal.It is also important to note that the anomalous adsorption will cause a swelling of only a few per cent; thus, any changes in acoustic impedance, stress, etc., will be small in comparison with the over-all1270 ANALYST, AUGUST 1992, VOL. 117 mass-change effect. Thus any signals caused by the swelling alone are negligible when interpreting these sensor responses. Such a response profile is found only for the methanol- poly(pyrro1e) systems and not for the other vapour and gases tested. The magnitude of the methanol response is also greater (approximately ten times) than for other analytes.This could be due to the ease with which methanol will form hydrogen bonds with the solvation sheath surrounding the doping ions. In such a way methanol would act as a solvent, also causing the polymer to swell, where other organic vapours cannot. The conductivity response of poly(pyrro1e) to methanol was plotted against the square root of time (Fig. 10). This plot does not exhibit a two-stage behaviour. The curve shows a Fickian type response with the plateau appearing much earlier in the plot than for the piezoelectric responses. This suggests that the conductivity change is caused not by the second stage diffusion but by the initial swelling of the polymer, which occurs at a rate commensurate with polymer swelling.It has been observed in the thickness profile measurements that poly- (pyrrole) exhibits simultaneous conductivity and thickness changes. It is expected that during the swelling the conducting ‘cross-links’ between polymer chains, the conductivity limiting factor, are lengthened and thus the conductivity will decrease. The relatively high increase in resistance when poly(pyrro1e) is exposed to methanol in comparison with other organic vapours is evidence to support this theory. However, it is also important to note that the hydrogen sulfide conductivity change is larger in comparison with methanol vapour than is the frequency change, which was not biphasic. These data suggest that at least two phenomena are involved in conductiv- ity changes in poly(pyrro1e) on exposure to gases and vapours and does not exclude the widely held belief that conductivity changes in poly(pyrro1e) are caused by the electronic influence on the conducting polarons by strongly electrophilic gases.In conclusion it would seem that the poly(pyrro1e) conduc- tivity response to gases and vapours has a dual nature. Firstly, there is the widely accepted view of nucleophilic gases causing increases in resistance with electrophilic gases having the opposite effect. Secondly, it has been shown that certain vapours have a solvent type action on the polymer causing it to swell (and presumably if the conditions are right contract). Changes in dimension of the polymer are accompanied by a conductivity change; hence changes of apolar analytes also change the polymer conductivity, possibly affecting the activation energy for electron hopping between chains and thus the conductivity decreases.It would be expected that varying contributions from both mechanisms would normally be observed on exposure to gases and vapours. References 1 Advances in Electrochemical Science and Engineering, eds. Gerischer, H., and Tobias, W., VCH, Weinheim, 1990, pp. 2 Diaz, A. F., Castillo, J. I . , Logan, J. A., and Lee, W., J. Electroanal. Chem., 1981, 129, 115. 3 Asavapiriyanont, S., Chandler, G. K., and Gunawardena, G. A., J. Electroanal. Chem., 1984, 177, 229. 4 Diaz, A. F., Kanazawa, K. K., and Gardini, G. P., J. Chem. SOC., Chem. Commun., 1979, 635. 5 Diaz, A . F., and Castillo, J. I., J. Chem.SOC., Chem. Commun., 1980, 397. 6 Kanazawa, K. K., Diaz, A. F., Geiss, R. H., Gill, W. D., Kwak, J. F., Logan, J. A., Talbot, J. F., and Street, G. B., J . Chem. SOC., Chem. Commun., 1979, 854. 7 Nylander, C., Armgreth, M., and Lundstrom, I., Proceedings of the International Meeting on Chemical Sensors, Fukuoka, 1983, eds. Seiyama, T., Fukei, K., Shiokawa, J., and Suzuki, S., Elsevier, Amsterdam, 1983, pp. 203-207. 8 Miasik, J. J., Hooper, A., and Tofield, B. C., J. Chem. SOC., Faraday Trans. I , 1986,82, 1117. 9 Hanawa, T., Kunabata, S., and Yoneyama, H., J. Chem. SOC., Faraday Trans. I , 1988,84, 1587. 10 Hanawa, T., and Yoneyama, H., Bull. Chem. SOC. Jpn., 1989, 62, 1710. 11 Bartlett, P. N., and Ling-Chung, S. K., Sens. Actuators, 1989, 19, 141. 12 Blackwood, D., and Josowicz, M., J. Phys. Chem., 1991, 95, 493. 13 Slater, J. M., and Watt, E. J., Analyst, 1991, 116, 1125. 14 Alder, J. F., and McCallum, J. J . , Analyst, 1983, 108, 1169. 15 McCallum, J. J., Analyst, 1989, 114, 1173. 16 Chemical Sensors, ed. Edmonds, T. E., Blackie, Glasgow, 1988, 17 Mecea, V., and Bucur, R. V., Thin Solid Films, 1979,60,73. 18 Lu, C. and Lewis, O., J. Appl. Phys., 1972,43,4385. 19 Eernisse, E. P., J. Appl. Phys., 1972, 43, 1330. 20 Slater, J. M., and Watt, E. J., Anal. Proc., 1992, 29, 53. 21 Wynne, K. J., and Street, G. B., Macromolecules, 1985, 18, 2361. 22 Diaz, A. F., and Hall, B . , IBM J. Res. Develop., 1983,27,342. 23 Crank, J., Mathematics of Diffusion, Clarendon Press, Oxford, 1975. 24 Long, F. A., and Kokes, R. J., J. Am. Chem. SOC., 1953, 75, 2232. 25 Bagley, E., andLong,F. A., J. Am. Chem. SOC., 1955,77,2172. 26 Long, F. A., and Richman, D., J. Am. Chem. SOC., 1960, 82, 513. 1-74. pp. 295-317. The authors thank GEC/Marconi Hirst Research Centre and British Gas for supporting sensor research. Paper 2/00409G Received January 27, 1992 Accepted March 17, 1992

ISSN:0003-2654    

DOI:10.1039/AN9921701265

出版商:RSC    

年代:1992

数据来源: RSC

摘要:

ANALYST, AUGUST 1992, VOL. 117 1271 Amperometric Chemical Sensors Using Microheterogeneous Systems* Michael E. G. Lyons, Cormac H. Lyons and Athanase Michas Physical Chemistry Laboratory, University of Dublin, Trinity College, Dublin 2, Ireland Philip N. Bartlett School of Chemistry, University of Bath, Claverton Down, Bath BA2 7AY, UK A theoretical model describing heterogeneous redox catalysis at metal oxide-Nafion composite modified electrodes is presented. The metal oxide particles are assumed to be dispersed in a homogeneous manner within the polymeric Nafion matrix. The composite material is deposited on the surface of a support electrode in the form of a thin film. The surface of each oxide particle is assumed to be catalytically active. The interaction between substrate and the catalytically active oxymetal surface group is assumed to follow Michaelis-Menten kinetics.The transport and kinetics of the substrate within the polymer film is examined and various limiting expressions for the reaction flux are proposed. The theoretical predictions are compared with experimental results obtained for the electro-oxidation of catechol at Ru02-Nafion composite electrodes. Keywords: Sensors; amperometry; microheterogeneous system; theory Introduction The design, fabrication and application of novel ampero- metric chemical and biological sensors has been a topic of considerable interest in recent years. 1.2 This intense research activity has occurred in tandem with rapid developments in the area of polymer modified electrodes.3 In this paper aspects of our current work pertaining to the development of polymer modified electrodes for use as amperometric chemical sensors are presented.Our approach to sensor design has involved the development of mathematical models which would define the pertinent operational characteristics of the sensor probe when operated in the steady-state mode. This approach has the advantage that the models will enable us to predict the optimum conditions (layer thickness, active sensing element loading, etc.) for sensor operation. In a number of previous papers a description of the transport and kinetics in multicomponent microhetero- geneous catalytic systems has been presented.G The situation where catalytically active microparticles are dispersed uni- formly throughout a conducting polymer matrix is discussed, the matrix being deposited as a thin film on a supporting electrode surface .4,5 The analysis was subsequently extended to the more complicated situation where both catalytic microparticles and electron transfer mediators are dispersed in a polymeric matrix.5Jj Although the transport and kinetics of reactions in chemically modified electrodes have been analysed and approximate analytical solutions are avail- able,7-9 these treatments are not applicable directly to the situation of dispersed microparticles. This is because for microscopic catalytic particles it is necessary to include, explicitly, both the spherical diffusion to the particle surface within the polymer matrix and the electrode kinetics of the reactions at the particle surface. In the present paper a related problem is addressed, where catalytic particles (being the active sensing element) are dispersed in a thin polymeric film, where the substrate- product reaction occurs via Michaelis-Menten kinetics. The analysis discussed in this paper will also have direct relevance to the associated problem of immobilized enzyme cataly- sis,10.11 diffusion, and chemical reaction problems in chemical engineering.12 Our theoretical work is motivated by * Presented at the meeting on Analytical Applications of Chemi- cally Modified Electrodes, Bristol, UK, January 7-8, 1992. experimental studies involving the electro-oxidation of cate- chol at a multicomponent modified electrode consisting of Ru02 particles dispersed in a Nafion film.Conductive metallic oxide electrodes such as Ru02 coated titanium have found wide application in the area of electrocatalysis. The inherent catalytic activity results from the fact13 that the outermost region of the oxide surface is hydrated: the catalytically active centres are surface-bound oxymetal groups which act as binding centres for solution phase substrates. Nafion, an ionically conducting perfluorinated polymer, has been used as a matrix for the incorporation of various electroactive substances to form novel classes of chemically modified electrodes.14 Aspects of the theory developed in this paper have recently been described independently by Albery et a1.15 for enzyme electrodes. Theory The Model In this paper a homogeneous distribution of catalyst particles in a polymeric matrix is discussed.A specific example could be Ru02 particles in Nafion. The effect of concentration polar- ization in the solution is ignored and only the substrate diffusion and chemical reaction within the polymer matrix are considered. The latter is assumed to be deposited in thin film form on the surface of a support electrode. We therefore consider a bounded diffusion problem. The modified elec- trode is illustrated in Fig. 1. We also assume that steady-state conditions apply. The differential equation describing the transport and kinetics in the layer is given by where s denotes the concentration of substrate and Ds is the substrate diffusion coefficient in the layer. We assume that the substrate reacts with the catalyst via Michaelis-Menten (or in the parlance of surface chemistry, Langmuir-Hinshelwood) kinetics, according to the following scheme: Kml kc oxide particle S + C [SC] + P + C’ electrode C’+ k‘E c where C and C’ represent the catalytically active form of the oxymetal surface group and the pre-catalytic form, respec-1272 ANALYST, AUGUST 1992, VOL.117 Polymer matrix I / I Solution support electrode Catalytic particle X 0 L 0 1 ,X K,,, Particle: s + c i= [SCI 5 P + C' Electrode: C' 2 C Fig. 1 Schematic representation of the oxide-ionomer composite modified electrode tively. We note from this scheme that the catalytic process is cyclic and is an example of heterogeneous redox catalysis. The species [SC] denotes the substrate-catalyst complex. Hence k = k(s) = kcCz/(Km + S) (2) In this expression k, denotes the catalytic rate constant (s-l), cz represents the total catalyst concentration (mol cm-3) and K, is the Michaelis constant (mol cm-3).Hence, from eqns. (1) and (2) we obtain Ds d2,/dX2 - k,czs/(K, + S) = 0 (3) This is a non-linear differential equation. The non-linearity arises because of the Michaelis-Menten kinetic term. This equation is solved subject to the following boundary condi- tions. First, we assume that the substrate does not react at the support electrode, thus x = 0, ds/dx = 0 Also, given that the polymer layer is of thickness L, and neglecting concentration polarization in the solution, we write x = L, s = Ks" where sm represents the bulk concentration of substrate, and K represents the partition coefficient.(4) ( 5 ) Transformation to Non-dimensional Form In order to proceed further it is useful to transform the differential equation [eqn. (3)] and the boundary conditions [eqns. (4) and (5)] into non-dimensional form. We introduce the following normalized variables: u = s/Ksm, a = Ks"IK,, X = xIL (6) We also introduce a parameter termed the Thiele modulus @ which is given by $2 = kccXL2/K,Ds (7) It is instructive to note that the first-order rate constant for the chemical transformation of substrate to product k' is given by k' = k,c~IK, (8) Hence, we note that the Thiele modulus may be recast in the form +2 = k'L2/Ds = k'L/kPD = k'Lsm/(Dssm/L) = jR/jD (9) In this expression k'D represents the diffusional rate constant in the layer. We note from eqn. (9) that $2 is simply the ratio of the maximum chemical conversion rate in the layer (jR) to the maximum diffusional transport rate in the layer GD).We can write another useful relationship from eqn. (9): @ = L(k'/DS)''' = L/Xk (10) where we have defined the reaction layer thickness (Le., the distance into the layer the substrate travels before undergoing reaction with the catalyst) as x k = (Ds/k')'I2. Hence we note that the Thiele modulus is simply the ratio of the layer thickness to the reaction layer thickness. It is easily shown that the master equation describing transport and kinetics in the polymer matrix may be trans- formed to the following non-dimensional format: dWdX2 - @W(1 + &u) = 0 X = 0, du/dX = 0 (11) (12) X = l , u = l (13) (14) The boundary conditions transform to and We may introduce a dimensionless flux parameter y given by: y = a(du/dX),, 1 The dimensionless flux parameter y is related to the flux j (mol cm-2 s-1) via the relationship j = K,Dsy/L (15) The expression outlined in eqn.(11) is non-linear, and in order to facilitate discussion we will consider a number of approximate solutions to this equation. When a <<1 (s << K,) we have unsaturation and the system exhibits simple first- order kinetics. In contrast when u >> 1 (s >> K,) saturation conditions apply and the kinetics are zero order. Furthermore the finite dimension of the layer must be taken into account which is accomplished by examining the L/xk ratio. Conse- quently L/Xk << 1 represents the thin film situation where reaction takes place throughout the entire film.In this case no concentration polarization exists within the film. In contrast when L/xk >> 1 we have thick films and reaction occurs mostly in a first-order reaction layer near the polymer/solution interface. We note therefore that the system may be charac- terized by a case diagram which consists of a plot of log (L/Xk) versus log a, or alternatively, log @ versus log a. With no Concentration Polarization in the Layer First, the situation where the concentration of substrate .is uniform throughout the layer is examined. This will be valid for thin films. In this situation we may set u = 1 throughout the film. Hence, the master equation [eqn. ( l l ) ] transforms to d2uldX2 = @2/(1 + a) Integration of this equation from x = 0 to x = 1 results in the assignment du/dX = @W(1 + a) (16) (17) (18) The normalized flux is given by y = a(du/dX),= 1 = a@'/( 1 + a) Re-transforming back into dimensioned parameters we obtain the reaction flux j , given by j = k,czLKsm/(K, + Ks") (19) Eqn.(19),is the simple Michaelis-Menten equation that is well known in Fzyme kinetics. This equation describes the flux for the reaction of substrate at the oxide catalyst when there is no concentration polarization of substrate in the film. The equation is valid for all values of the substrate concentration.ANALYST, AUGUST 1992, VOL. 117 1273 Low Substrate Concentration Limit: Unsaturated Kinetics We now examine the situation of low substrate concentration where Ks" << K , (unsaturation). When a << 1 we have 1 + au = 1 and eqn.(11) reduces to d2ddX2 - $ 2 ~ = 0 (20) Hence the problem reduces to that of diffusionheaction with first-order kinetics. The solution to this equation is presented in Appendix A. The normalized flux y obtained from the analysis outlined in Appendix A is y = a(du/dX),=, = a$ tanh $ (21) The equation for the normalized flux will be valid for low substrate concentrations where a << 1, i.e., when sm << K,. It will also be valid over the entire range of layer thickness and the equation shows how the rate of mass transfer is enhanced by the chemical reaction of substrate at the dispersed catalyst. We also show in Appendix A that the substrate concentration profile is described by the following expression: u(X) = {cosh($X)/cosh($)} (22) This concentration profile is illustrated in Fig.2 as a function of the Thiele modulus. We note from examination of Fig. 2 that when $ is small the concentration profiles are fairly shallow. Hence the concentration throughout the layer is close to the value at X = 1. However, when @ is large the concentration of substrate falls rapidly with distance into the polymer film. The reaction is essentially complete within a thin region near the polymer/solution interface. This discussion may be amplified by examining eqn. (21) in more detail. When $ < 0.3 one may show that tanh $ = $. In this case assuming that $I << 1 ( i e . , L << Xk), then the expression for the flux outlined in eqn. (21) reduces to y = a$2 (23) Introducing the expression for $2 obtained from eqn.(7) into eqn. (23), noting that j = K,Dsy/L, we subsequently obtain (after simplification) that j = k,cxK,-lLKsw (24) Hence, provided that sm < K,, and we work with thin films, the flux will be linearly dependent on the substrate concentra- tion and the thickness of the layer. The flux will also depend linearly on the catalyst loading. The reaction zone encom- passes the entire layer and the rate of reaction at the catalyst surface will be much lower than the rate of substrate transport. In this sense when $ << 1 the rate is reaction controlled. Note also that for small $ cosh $I = 1, cosh (QX) = 1 and so from eqn. (22) we obtain that u(X) = 1, which indicates a uniform concentration distribution throughout the layer.1 .o 3 0.5 0 r 0.1 I 0.5 1 .o X Fig. 2 Plot of normalized substrate concentration versus normalized distance when sm < K,. The numbers next to the concentration profiles refer to values of the Thiele modulus LIXk Alternatively, for large @ ($I > 2) we may show that tanh$ = 1. This is the situation for thick films where L >> X,. In this case the flux expression [eqn. (21)] reduces to Y = a$ (25) j = (kcc2Ds/K,)1/2Ksm (26) which further simplifies to In the case where sm << K, and if the layer thickness is large, the flux will be independent of layer thickness and will be linearly dependent on substrate concentration. It will also vary as the square-root of the catalyst loading. In this case there is a thin reaction layer and the reaction kinetics will be very much more rapid than the diffusional transport of substrate.Therefore, the reaction will be diffusion controlled, and there will be considerable concentration polarization within the film. This is clearly indicated in Fig. 2. High Substrate Concentration Limit: Saturated Kinetics The situation at high substrate concentrations is now exam- ined, when a >> 1 (sm >> K,), in which the catalyst is saturated by substrate. This saturation region results in zero-order kinetics. When a >> 1, 1 + au = au; hence eqn. (11) reduces to d2UldX2 - $2/a = 0 (27) The latter expression may be readily integrated with respect to $ to yield du/dX = $2Wa Further integration of eqn. (28) yields: (28) (29) u(X) = 1 + $2(X2 - 1)/2a The latter expression for the concentration profile will be valid for $2 < 2a.The normalized flux y is readily obtained from eqn. (28): y = $2 (30) From eqn. (7) we introduce a relationship for $2 and subsequently, eqn. (30) reduces to the rather simple form j = k,cxL (31) Thus under conditions of reactant saturation the flux is independent of substrate concentration, but is first order with respect to layer thickness and catalyst loading. The rate- determining step involves breakdown of the catalyst-substrate complex (described by the parameter k,). The General Situation We now return to the master differential equation, eqn. (11). We require an analytical solution to this non-linear equation in order to derive an expression for the flux. Most of the previous work in this area10,11716 has employed a numerical approach to this problem, although Albery et ~ 1 .1 5 examined a similar type of problem in enzyme electrochemistry in parallel to our studies. In this section we consider an approximate analytical solution to eqn. (ll), bearing in mind some previously published comments by Albery et ~1.15 The following analysis will be valid for thick films where $ >> 1. In Appendix B the normalized flux is given by the following expression : y = {2$2[a - ln(1 + a)]}1'2 (32) This is a rather complex expression for the flux and is valid for thick films and for all values of the substrate concentration. When a << 1 then ln(1 + a) = a - a2/2 and eqn. (32) reduces to y = a$ as obtained previously [eqn. (25)]. For $ > 1 and a >> 1 we note that ln(1 + a) =: In a and the flux reduces to y = {2$2[a - In a]}1/2 = (2a)1/2@ (33)1274 ANALYST, AUGUST 1992, VOL.117 The expression outlined in eqn. (33) is valid for thin films when part of the layer is saturated and the other part unsaturated. Eqn. (33) will be valid up to $2 = 2a. Transforming eqn. (33) into dimensioned parameters we obtain that the flux j is j = (2Kk,cxDssm)1~2 (34) Hence the flux is half-order in substrate and catalyst concen- trations and independent of layer thickness. The Case Diagram We now present the result of our analysis in terms of a case diagram. We have utilized this approach in previous papers and have found that the case diagram presents a very useful summary of the kinetic behaviour of a system.First, we note that a natural set of axes to choose are log Q as ordinate and log a as abscissa. Hence movement along the ordinate takes us from thin to thick films and movement along the abscissa takes us from a condition of reactant unsaturation to one of saturation. The case diagram is illustrated in Fig. 3. We note from this diagram that four cases must be considered. The main approximate expressions for the flux are given by eqn. (24) (case I), eqn. (26) (case 11), eqn. (31) (case 111) and eqn. (34) (case IV). The expressions for the flux and the reaction orders with respect to substrate concentration, catalyst loading and layer thickness for each of the four cases, are shown in Table 1. Case I is defined by eqn. (24) and is located in the quadrant bounded by @ d 1 and a d 1.Hence from the viewpoint of amperometric chemical sensor applica- tions, case I defines a useful region as the flux and hence the amperometric response will be linearly related to the substrate concentration (via the a term). Furthermore as @ << 1 in this quadrant, the entire layer will be catalytically active. There will be no concentration polarization of substrate in the layer. II IV Thick films, unsaturated Thick films, saturated @ << 1, @ >> 1, ux, >> 1 a>> I,@<< 1 Ir y = a+tanh+ y = v‘% +tanh($2/2a)”2 ux, = 1 I I toga 1 (Y << 1, + << 1, LIX, << 1 (Y>>1,+<<1 sr << K, srn = K, sx >> K, I I Thin films, unsaturated Ill Thin films, saturated Fig. 3 Case diagram summarizing the transport and kinetics modified electrode From Table 1 we note that the current response will be first order in substrate concentration, layer thickness and catalyst loading.Case I1 is defined by the quadrant bounded by @ 2 1 and a d 1. In this thick-film region the flux is governed by eqn. (26). Again case I1 defines a useful region for sensor applications, in that the sensor response will vary linearly with substrate concentration, but will also depend inversely on the value of the Thiele modulus, and hence on concentration polarization effects in the layer. Only a portion of the layer will be utilized in this case. Another feature to be noted from Table 1 is that the current response in this case will be independent of layer thickness and half-order in catalyst loading.These are useful diagnostic points. Case I11 is located in the lower right-hand region of the case diagram and is defined by the boundary lines a = 1 and a = @2/2. In this case the flux expression is given by eqn. (31). The current will be first order in layer thickness and catalyst loading in this region. This situation or zero-order kinetics with respect to substrate concentration is not usually analytically useful as the ampero- metric response remains independent of substrate concentra- tion. However, in some situations one may wish to utilize a device which operates on the principle of inhibition of the catalyst by species. In such a situation this is a good case to aim for because the flux depends on catalyst activity in this region.The region defining cases I and I11 can be readily described in terms of the Michaelis-Menten equation outlined in eqn. (19). We transpose the latter expression into dimensional form to obtain l/j = (Km/k,LKcx) sm-1 + l/k,Lcx (35) Immediately we note that this is the form of the well known Lineweaver-Burk equation.17 This equation serves as a useful diagnostic plot to confirm the operation of Michaelis-Menten kinetics. Hence a plot of inverse flux versus inverse substrate concentration should be linear. The Michaelis constant K, and the catalytic rate constant k, may be readily obtained from analysis of the slope and intercept of this plot. Reference may be made to the work of Gorton and co-workers’s20 who have utilized such an analysis to examine the oxidation of NADH (dihydronicotinamide adenine dinucleotide) at Medola Blue and 1,2-benzophenoxazine modified electrodes.Albery et aZ.21 have also utilized this plot to examine NADH oxidation at conducting organic salt electrodes. Finally, we examine case IV which is defined in the upper right-hand region (where @ >> 1) which is bound in the case diagram by the lines a = 1 and a = @2/2. In this case the flux is governed by eqn. (34), and the current will be independent of layer thickness and half order in substrate concentration and catalyst loading. Again the flux is governed by the value of the Thiele modulus and we will operate under reaction layer conditions where only a fraction of the film is utilized. Thus the case diagram presents the kinetics in a very convenient manner.The diagnostic criteria illustrated in Table 1 enable the determination of any particular case for any combination of experimental parameters. We now apply our theoretical analysis and consider the electro-oxidation of catechol at Ru02 particles dispersed in a Nafion matrix. Table 1 Mechanistic indicators for microheterogeneous systems displaying Michaelis-Menten kinetics Reaction order with respect to in the Case Eqn. Flux expression cx L sm I 24 j = k,cxKm-’LKsm 1 1 1 I1 26 j = d(kcczDs/Km)Ksm 112 0 1 I11 31 j = kccz L 1 1 0 IV 34 j = d2KkcczDssm 112 0 112ANALYST, AUGUST 1992, VOL. 117 1275 Experimental A standard three-electrode electrochemical cell was used, containing working, reference and counter electrodes. The cell consisted of a Metrohm amber glass water-jacketed vessel of volume 20-90 cm3, with a nitrogen inlet/outlet tap.Water from a thermostated bath was pumped through the outer sheath of the cell to provide good temperature control (+ 0.l"C). All work was carried out at 25°C. A Metrohm working electrode was used, consisting of a glassy carbon or platinum disc (exposed geometric area, 0.071 cm2) encapsu- lated into a rotating disc assembly using a PTFE body and separate electrical contacts. A Metrohm Ag-AgC1 (KCI saturated) reference electrode (E = 0.197 V versus NHE) was used, and all potentials are quoted with respect to this electrode. A large surface area (about 3 cm2) platinum foil was used as the counter electrode. All solutions were prepared from ultrapure Milli-Q water, using AnalaR grade reagents where possible.The electrochemistry measurements were carried out in 0.2 rnol dm-3 HZSO4 supporting electrolyte, pH 0.8. Cyclic voltammetry and steady-state measurements were performed using a PAR EG & G Model 273 potentiostat/ galvanostat. In the steady-state current response measure- ments the sensing potential applied to the probe was 950 mV. Ruthenium oxide was prepared via the thermal decomposi- tion of the chloride, RuC13.xHz0, at 450°C for 4 h. Ruthenium oxide dispersions (3% by mass) were prepared by ultrasonication of the oxide powder in a Nafion solution (1-570 m/v in ethanol). Aliquots of these dispersions were dropped onto the working electrode disc surface using a micropipette and the solvent was allowed to evaporate slowly.This protocol resulted in the fabrication of a thin film oxide-Nafion composite sensor probe and afforded a typical oxide loading of about 1 mg cm-2 and a layer thickness of 1-2 Pm. Results and Discussion Redox Behaviour of the Oxide-Ionomer Composite Typical cyclic voltammograms recorded for the Ru02-Nafion composite material in 0.2 mol dm-3 HZSO4, 0.2 rnol dm-3 NaOH and phosphate buffer, pH 7.4, at a sweep rate of 10 mV s-1, are outlined in Fig. 4. It can be seen that in all cases, the charge involved in the cathodic and anodic sweeps is approximately equal. The broad nature of the current profile across the entire potential window examined suggests that changes in oxidation state for the oxide particles can occur over a wide potential range. Previous work13 has indicated that the peaks in the voltammogram involve the RU"*/RU'~ and RU'~/RU~' redox transitions.Owing to the rather ill defined peaks in the voltammogram it is difficult to estimate the standard redox potentials of the oxyruthenium species, but an approximate assignment in aqueous acid is 440 k 20 mV (RUVRU~~) and 820 k 20 mV (RU'~/RU~'). In basic solutions, the peaks located at 100 and 390 mV may be attributed to the RU'~/RU~' and RuV1/RuVII redox transformations, respectively. It should be noted that the voltammograms recorded in neutral solutions do not exhibit any fine structure. Hence the voltammetric peaks may be assigned to a series of surface redox transitions involving tightly bound oxyruthenium sur- face groups.13 These surface groups are assumed to have a low degree of bridging oxygen coordination to the bulk oxide particle.The redox chemistry may be described using the following equations (the bridging oxygens being represented as -0-): (-O-)2RuOH(OH2)3 = (-O-)2Ru(OH)2(OH2)2 + H+ + e- RU'"/RU'~- (36) (37) Ru'~/Ru''- (-O-)~RU(OH)~(OH~)~ = (-O-)2Ru(OH)4 + 2H+ + 2e- I I I 0 0.5 1 .o t ' Z -0.4 0 0.4 I I 1 I -0.4 0 0.4 0.8 EIV Fig. 4 (a) Typical cyclic voltammogram for an Ru02-Nafion composite electrode in 0.2 mol dm-3 H2SO4 (pH = 0.83). Sweep rate, 10 mV s-l. (b) Voltammetric response for the composite electrode recorded in 0.2 mol dm-3 NaOH. (c) Voltammetric profile recorded (sweep rate, 20 mV s-1) in phosphate buffer solution, pH 7.4 RU~'/RU~''- (-O-)2Ru(OH)4 = (-O-)2Ru(OH)3 + H+ + e- (38) Note that the oxyruthenium surface groups maintain an octahedral geometry in all cases. Confirmation that oxidation involves the injection of protons from the hydrated oxide layer into the electrolyte (with proton injection from the electrolyte on the subsequent return cathodic sweep) has been demon- strated22 using a suitable proton-sensitive dye, Bromocresol Blue.Thus, each oxide microparticle has an outer hydrated, catalytically active layer which consists of 'dangling' oxy- ruthenium surface groups. A graph of voltammetric charge (which is proportional to the oxyruthenium group coverage and thus catalytic activity) as a function of pH is plotted in Fig. 5. A minimum in voltammetric charge and hence in oxymetal group coverage is observed at pH 7.4. The e-/H+ stoichiometry implicit in the assignment of the oxyruthenium group redox behaviour presented in eqns.(36)-(38) suggests that the standard potential of each of the redox transitions should vary in a1276 ANALYST, AUGUST 1992, VOL. 117 Nernstian manner with changes in solution pH. The results of one such series of experiments are illustrated in Fig. 6. In this case the EO value for the RU*~/RU~' transition is found to shift cathodically in a linear manner by about 60 mV per unit change in pH. Catechol Oxidation at the Oxid-Nafion Composite Catechol undergoes a 2e-, 2H+ oxidation reaction as outlined in Scheme 1. The cyclic voltammetric response of a dilute catechol solution (1.0 mmol dm-3) in 0.2 mol dm-3 HZS04 at the oxide-Nafion composite electrode is illustrated in Fig.7. The redox chemistry is quasi-reversible, as implied by the large peak separation between the anodic and cathodic peaks (the latter being 325 mV at 20 mV s-1). An immediate point to note is that a good match exists between the stoichiometry exhibited by the transformation involving the oxyruthenium surface groups [eqn. (37)] and the catechol oxidation stoi- chiometry outlined in Scheme 1. It is clear from Fig. 7 that catechol oxidizes in the potential region where the Ru'~/Ru~' redox chemistry predominates. It is therefore reasonable to suppose that there is mediation of substrate oxidation via the surface bound oxyruthenium groups. The mediation occurs via participation of the RuV1 group as follows: RuV' + S = Rut" + P, where S and P denote substrate and product, respec- 1.4 0 6 o.8 0.2 4 8 12 PH Fig.5 Variation of integrated anodic charge, obtained from the voltammetric response recorded at low sweep rate, with solution pH 0.8 0.6 ? 0.4 0.2 0 2 6 10 14 PH Fig. 6 Variation of voltammetric peak potential corresponding to the RU'~/RU~' redox transition with solution pH. dE/dpH = 60 mV decade-' OH 0 0 +2H+ +2e- Scheme 1 electro-oxidation of catechol Stoichiometry of the redox processes describing the tively. This reaction is an example of heterogeneous redox catalysis. We assume that the mediation process may be described in terms of Michaelis-Menten kinetics, where a precursor and a successor complex are formed as part of the reaction sequence. We recall from the theory section that our analysis neglected the effects of concentration polarization in the solution next to the surface of the composite electrode.The most efficient way to effect a clear separation between solution phase diffusional effects and transport and kinetic effects associated with processes occurring in the polymer layer is to utilize a rotating disc electrode (RDE).23 In this technique the current/potential response is obtained at a series of different rotation speeds and current data are obtained corresponding to conditions of infinite rotation speed. The net result of this protocol is to obtain current data in which the solution phase diffusional transport of substrate (i. e., concen- tration polarization effects) may be neglected. A typical series of RDE voltammograms corresponding to the oxidation of a dilute catechol solution at the composite electrode are outlined in Fig.8. The over-all shape of the curves again reflects the quasi-reversible nature of the redox chemistry. Well defined mass transport limited currents are observed for potentials more positive than 850 mV. Concen- tration polarization effects may be separated by resorting to t ' Z 0 0.5 1 .o EIV Fig. 7 Voltammetric response for catechol (1.0 mmol dm-3) at the oxide-Nafion composite electrode in 0.2 mol dm-3 H2S04. Sweep rate. 20 mV s-1 t .Z + _i I I I 0 0.5 1 .o E N Fig. 8 Typical RDE voltammograms for the oxidation of catechol (1 mmol dm-3,0.2 mol dm-3 H2S04) at the oxide-ionomer composite electrode. The lowest curve corresponds to a rotation speed of 500 rev min-1, whereas the highest curve was obtained at 3000 rev min-l. The rotation speed increment was 500 rev min-1.Sweep rate, 2 mV s-lANALYST, AUGUST 1992, VOL. 117 1277 the Koutecky-Levich method of analysis23 where we note that lli = SKL ~ - 1 1 2 + I,, (39) Hence a plot of lli versus w-112 (where w represents the rotation speed, in Hz) should be linear with slope SKL given by SKL = 1.55nFAD2/3~-%~ [where n = number of electrons transferred, F = Faraday constant (96 500 C mol-1) and A = geometric area of electrode] and intercept given by IKL. We note that D represents the diffusion coefficient of catechol in the solution, v is the kinematic viscosity, and S” is the bulk concentration of catechol. The response of the steady-state current i, corresponding to the oxidation of a dilute catechol solution at a rotating composite electrode, recorded at a fixed potential of 950 mV, to a series of rotation speed steps, is outlined in Fig.9. The rotation speed was stepped in 500 rev min-1 intervals from an initial value of 500 rev min-1 to a final value of 3000 rev min-1. Good steady-state current responses were observed at all rotation speeds. The response remains invariant on stepping down again. In this manner a series of Koutecky-Levich plots may be rapidly obtained for a wide range of substrate concentrations. The results of such a series of experiments are outlined in Fig. 10. It is clear from these results that the Koutecky-Levich equation [eqn. (39)] is valid for all substrate concentrations examined. The most important point to note from these data is that the Koutecky- 50 s M f- Fig.9 Amperometric response of the composite electrode to a sequence of rotation speed steps. Sensing potential, 950 mV. This value of applied potential ensures that the catechol (1 mmol dm-3) is efficiently oxidized at the electrode surface. 1,500; 2,1000; 3,1500; 4, 2000; 5,2500; and 6,3000 rev min-1 A 6 t P I I I 0 4 10 16 ( ( d s ) - ” 2 x 102 Fig. 10 Typical Koutecky-Levich lots for catechol oxidation at the composite electrode in 0.2 mol dm-i: H2S04. Catechol concentration: A , 2; B, 4; C, 6; D, 8; E, 10; F, 12; and G, 14 mmol dm-3 Levich intercept varies inversely with substrate concentration. This effect is predicted from the model discussed at the beginning of this paper. We recall the Lineweaver-Burk expression from eqn.(35) and identify that JKL = (Km/nFAkcLKcx)sm-’ + l/nFAk,Lcx (40) Hence if the mathematical model is valid, we note that a plot of ZKL versus lls“ should be linear. Furthermore, the fundamental parameters K , and k, may be obtained from the slope and the intercept of the plot. The Lineweaver-Burk plot is illustrated in Fig. 11. A good straight line is obtained, thereby confirming the theoretical model. The slope SLB of the plot is given by 1.56 mol dm-3A-1, whereas the intercept ILB = 0.16 X 103 A-1. We note from eqn. (40) that the catalytic rate constant k, = l/nFAZLBT, where r = Lcz is the surface coverage of oxyruthenium species. In our experiments r = 2 X 10-7 mol cm-2. Noting that n = 2, A = 0.071 cm2 and F = 96 500 C mol-1 we obtain that k , = 2.3 s-1.Furthermore the Michaelis constant K , is given by Km = nFAkcTSLB, from which we obtain that K , = 9.75 mmol dm-3. Therefore, there is a linear relation between steady-state current and bulk substrate concentration up to concentrations of about 10 mmol dm-3. This prediction was confirmed, as illustrated in Fig. 12, when the variation of steady-state current is recorded as a function of bulk catechol concentration. A good linear response is obtained up to 10 mmol dm-3 substrate. A deviation from linearity is observed for concentrations greater than this value. The current response recorded for an aged electrode (1 month old) is also included in this diagram. It is clear that little degradation of the current response with time is exhibited by the composite sensor.The practical operation of the composite sensor in aqueous acid solution is shown in Fig. 13. In this sequence of experiments a constant potential of 950 mV was applied to the composite electrode and a series of aliquots of a catechol solution was injected into the electrochemical cell. The current response of the probe was determined after each injection. A number of features are of immediate interest. Note the low baseline current (about 5 PA) in the absence of injected substrate. This means that low levels of catechol may be determined amperometrically. The response of the compo- site material to an injection of catechol is also rather rapid, and a steady-state current response is obtained within about 5 s. The material exhibits virtually no degradation of the current response with time.Further experiments have shown that the oxide-Nafion composite is chemically stable provided the catechol concentration does not exceed 50 mmol dm-3. The composite undergoes dissolution in rather concentrated cate- chol solutions. The mechanism of this degradation process is currently under investigation. 1 I I 1 2 3 sx/103 dm3 mol-’ Fig. 11 Lineweaver-Burk analysis of the data presented in Fig. 101278 ANALYST, AUGUST 1992, VOL. 117 4 3 $ a E a 2 1 0 5 10 15 s'/mmol drn 3 Fig. 12 Plot of steady-state catechol oxidation current versus substrate concentration. Applied potential, 950 mV. Supporting electrolyte, 0.2 mol dm-3 H2S04. The linear first-order region corresponding to substrate unsaturation and the zero-order region corresponding to substrate saturation are clearly outlined.The current response for virgin and aged electrodes is compared. Little degradation in amperometric response is observed after 1 month. 0, 'Fresh' film; and 0, after 1 month i' t - Fig. 13 Amperometric response of the oxide-ionomer com osite electrode to a series of injections of a dilute catechol solution. cm3 aliquots of a 0.1 mol dm-3 stock solution.) Supporting electrolyte, 0.2 mol dm-3 H2S04; applied potential, 950 mV Experimental Verification of the Theoretical Model Results obtained for catechol oxidation in 0.2 rnol dm-3 H2S04 illustrating the utilization of the theoretical model described in the first part of this paper are now presented. This is done to show how the operational characteristics of the oxide-ionomer composite sensor may be evaluated quantita- tively.The variation of steady-state current for the oxidation of a dilute catechol solution (0.8 mmol dm-3), recorded at a high rotation speed in order to minimize concentration polarization effects in solution, with Nafion thickness L is illustrated in Fig. 14. In these experiments the catechol concentration is much less than the Michaelis constant (10 mmol dm-3), hence the system is unsaturated. A good linear response is obtained in the thin-film region (2-10 pm). Hence as predicted by case I, the current is first order in layer thickness, when the Nafion film is relatively thin. For thicker Nafion deposits (L > 10 pm), a tendency towards zero-order behaviour is observed, as predicted by the case I1 scenario.1.5 . N E 1.0 . a u E .> 0.5 . 0 5 10 15 Ll!trn Fig. 14 Variation of steady-state amperometric response (0.8 mmol dm-3 catechol, 0.2 mol dm-3 H2S04, E = 950 mV) with changes in Nafion thickness. The changeover from first order to zero order can be discerned 0 1 2 3 4 c:, (% m/m) Fig. 15 Variation of steady-state amperometric response with oxide loading. 0.8 mmol dm-3 catechol, Nafion thickness, 2 pm (case I) We now examine the reaction order with respect to catalyst loading. The variation of steady-state current with cx for a Nafion film 2 pm thick is outlined in Fig. 15. A goad linear plot is obtained over the concentration range 14% m/m Ru02. These experiments were conducted for [catechol] = 0.8 mmol dm-3. Hence we have a first order dependence on cz, corresponding to that theoretically predicted for case I.The situation corresponding to thicker Nafion layers ( L = 18 pm), and low substrate concentrations, is illustrated in Fig. 16. In this case the current depends linearly on ~ ~ 1 ' 2 . This is as predicted theoretically for a case I1 situation. The current is also linear in substrate concentration for thick Nafion layers (Fig. 17). Catechol Oxidation Kinetics as a Function of Solution pH Catechol is oxidized via a 2e-/2H+ process in aqueous solutions. This assertion is confirmed in Fig. 18 where the standard potential for catechol electro-oxidation varies by 60 mV per unit change in solution pH. A typical series of RDE voltammograms for catechol (1 mmol dm-3) oxidation in 0.2 mol dm-3 NaOH and in phosphate buffer solution are outlined in Fig.19. We note that the oxidation process is more sluggish kinetically in the neutral pH medium. A series of Koutecky-Levich plots obtained from the RDE experiments conducted in media of different pH values are illustrated in Fig. 20. The magnitude of the Koutecky-Levich intercept provides a convenient estimate of the heterogeneous electro- chemical rate constant k'me characterizing the degree of facility of the electro-oxidation process. If this approach is taken, then the Koutecky-Levich intercept is23 I,, = l/nFAs"k',, (41)ANALYST, AUGUST 1992, VOL. 117 1279 0 1 2 [cz: (% m/m)]l Fig. 16 Variation of steady-state amperometric response with the square root of oxide loading. 0.8 mmol dm-3 catechol, Nafion thickness, 18 pm (case 11) 0.2 0.2 0.6 Ib) 0 0.4 0.8 E/V Fig.19 T pica1 RDE voltammograms recorded for catechol (1 buffer, pH 7.4 (6). The lowest curve corresponds to a rotation speed of 500 rev min-*, whereas the highest curve corresponds to 3000 rev min-1. Rotation speed increment, 500 rev min-l. Sweep rate, 2 mV s-1 mmol dm- Y ) oxidation in 0.2 mol dm-3 NaOH (a) and phosphate I I I 2 4 6 sx/mmol dm-3 5 Fig. 17 Variation of amperometric response with substrate concen- tration for thick Nafion layers (case II), L = 18 pm c a E N -. $ 3 .--' \ .- 0.8 a 0.6 2 G.' 0.4 0.2 1 8 12 16 4 (w/s)-"2 x 102 0 Fig. 20 Typical Koutecky-Levich plots for catechol oxidation as a function of solution pH. A, pH 0.8; B, pH 2.7; C, pH 7.4; D, pH 11; and E, pH 13 0 2 6 10 14 PH Fig. 18 Variation of half-wave potential for catechol oxidation with solution pH.A Nernstian shift of 60 mV per unit pH change is obtained, corresponding to a 2e-/2H+ process high and low pH values, Fme is typically 2 x 10-2 cm s-1, whereas in neutral medium the kme value decreases to 0.6 x 10-2 cm s - * . A plot of kfme versus pH is shown in Fig. 21. A distinct minimum is observed at pH values close to 7. Hence the rate of oxidation is minimal at neutral pH. There is, therefore, a correlation between the kinetics of catechol oxidation and the extent of oxyruthenium surface group coverage, as the latter is also minimal near pH 7. This observation provides strong evidence for our assertion that the Comparing eqns. (40) and (41) we obtain For low values of the substrate concentration we have Vme = k,TIKm.Hence, even though the parameter kfme is a compo- site factor it still provides an estimate of the degree of reversibility of the mediated electron transfer reaction. At1280 2.0 7 1, 1.5 G N I 2 1.0 . E j , 0.5 0 I I I I 2 6 10 14 PH Fig. 21 Variation of the heterogeneous electrochemical rate con- stant kfme with solution pH. Data obtained from analysis of the Koutecky-Levich intercepts in Fig. 20 oxyruthenium surface groups act as mediators for the electro- oxidation of catechol. Conclusions A theoretical model quantifying the operational characteris- tics of a metal oxide-Nafion composite amperometric sensor device has been presented. The sensor operates via a heterogeneous redox catalytic mechanism and the kinetics of the substrate-sensor element interaction may be described in the context of the Michaelis-Menten formalism.Various limiting expressions for the current are derived, and the kinetic behaviour of the sensor is illustrated using a case diagram. The specific example of catechol oxidation at Ru02 particles is examined and the theoretical predictions of the model are compared with experimental data. Good agree- ment between theory and experiment is found. The authors acknowledge financial support for this research programme from the EC Science Programme, EOLAS and the British Council. Appendix A In this appendix the solution to eqn. (20) is presented. This equation is linear and readily admits the following solution: u(X) = Aexp[@X] + Bexp[-@XI duldX = @Aexp[@X] - @Bexp[-@XI (Al) (A21 also We recall that (duld,), = = 0 and so A = B.Furthermore, we note that (duldX),,l = 2@A sinh @ (A31 u(X) = 2A cosh(@X) (A4) A = 1/2 cosh@ (A51 and Also when X = 1, u = 1, and so 1 = 2A cosh @, which leads to From eqns. (A3), (A5) and (14) in the text (the expression for the normalized flux y ) eqn. (21) is readily obtained. Further- more, from eqns. (A4) and (AS), the substrate concentration profile is described by the following expression: u(X) = {cosh(@X)/cosh(@)} (A61 ANALYST, AUGUST 1992, VOL. 117 Appendix B In this appendix a derivation of the flux expression is presented, eqn. (32) in the text. To do this we again return to eqn. (ll), multiply both sides of the equation by duldX, and note the identity ddX{ duldX}2 = 2(du/dX)(d2u/dX2) (B1) to obtain (ddX)(duldX)*dX = {2@2u/(1 + cxu)}du (B2) Integrating this expression and noting that when u = 0, du/dX = 0, we obtain the equation: (duldX)2 = 2(@/a)2[au - ln(1 + au)] (B3) The normalized flux y is given by y = a(du/dX),, 1.From eqn. (B3) we obtain (noting that when X = 1 u = 1) y = {2+2[a - ln(1 + cx)]}”2 (B4) which is our required result. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 References Borman, S., Anal. Chem., 1987, 59, 1091A. J. Chem. SOC., Faraday Trans. I , 1986, 82, 1033. Special Issue, ‘New Electrochemical Sensors’. Hillman, A. R., in ‘Electrochemical Science and Technology of Polymers’, ed. Linford, R. G., Elsevier, London, 1987, vol. 1, pp. 103-291. Lyons, M. E. G., McCormack, D. E., and Bartlett, P. N., J. Electroanal. Chem., 1989, 261, 51. Lyons, M. E. G., McCormack, D. E., Smyth, O., andBartlett, P. N., Faraday Discuss. Chem. SOC., 1989, 88, 139. Lyons, M. E. G., and Bartlett, P. N., J . Electroanaf. Chem., 1991,316, 1. Andrieux, C. P., Dumas-Bouchiat, J. M., and Saveant, J. M., J. Electroanal. Chem., 1982, 131, 1. Adrieux, C. P., Dumas-Bouchiat, J. M., and Saveant, J. M., J. Electroanal. Chem., 1984, 169, 9. Albery, W. J., and Hillman, A. R., J. Electroanal. Chem., 1985, 170,27. Engasser, J. M., and Horvath, C., in Applied Biochemistry and Bioengineering: Immobilized Enzyme Principles, eds. Wingard, L., Katchalski-Katzir, E., and Goldstein, L., Academic Press, New York, 1976, vol. 1, pp. 127-221. Engasser, J. M., and Horvath, C., Biotechnol Bioeng., 1974,16, 909. Ark, R., The Mathematical Theory of Diffmion and Reaction in Permeable Catalysts, Clarendon Press, Oxford, 1975, vols. 1 and 2. Lyons, M. E. G., and Burke, L. D., J. Chem. SOC., Faraday Trans. I , 1987, 83, 299. Whiteley, L. D., and Martin, C. R., J. Phys. Chem., 1989,93, 4650. Albery, W. J., Cass, A. E. G., and Shu, Z. X., Biosens. Bioelectron., 1990, 5 , 367. Yokoyama, K., Tamiya, E., and Karube, I., J. Electroanal. Chem., 1989,273, 107. Laidler, K. J., Chemical Kinetics, Harper and Row, New York, 3rd edn., 1987,~. 403. Gorton, L., Torstenssen, A., Jaegfeldt, H., and Johansson, G., J. Electroanal. Chem., 1984, 161, 103. Gorton, L., Johansson, G., and Torstensson, A., J. Electroanaf. Chem., 1985, 196, 81. Gorton, L., J. Chem. SOC., Faraday Trans. I , 1986,82, 1245. Albery, W. J., Bartlett, P. N., and Svanberg, L. R., in Charge and Field Effects in Biosystems, eds. Allen, M. J., and Usherwood, P. N. R., Abacus Press, London, 1984, pp. 443472. Jang, G. W., Tsai, E. W., and Rajeshwar, K., J. Electrochem. Soc., 1987, 134,2377. Albery, W. J., Electrode Kinetics, Clarendon Press, Oxford, 1975, ch. 3, pp. 49-67. Paper 2/01 223 E Received March 6, 1992 Accepted May 11, 1992

ISSN:0003-2654    

DOI:10.1039/AN9921701271

出版商:RSC    

年代:1992

数据来源: RSC

摘要:

ANALYST, AUGUST 1992, VOL. 117 1281 Chemically Modified, Screen-printed Carbon Electrodes* Stephen A. Wring Glaxo Group Research Ltd., Division of Drug Metabolism, Park Road, Ware, Hertfordshire SG12 ODP, UK John P. Hart Department of Science, Bristol Polytechnic, Coldharbour Lane, Frenchay, Bristol BS 16 IQY, UK The design, fabrication and evaluation of chemically modified screen-printed carbon electrodes is described with particular emphasis being placed on the practical details for sensor construction. This work employed cyclic and differential-pulse voltammetry, together with amperometry in stirred solutions to investigate systematically electrodes containing phthalocyanine and ferrocene-based mediators for the determination of several important biomolecules. An overview of the development of an amperometric assay using the enzyme glutathione peroxidase, and some preliminary results, for the selective determination of reduced glutathione in human whole blood is also presented.Keywords: Screen-printed chemically modified carbon electrodes; electrochemical sensors; reduced glutathione; ascorbic acid; coenzyme A Screen-printing technology can be usefully applied to the mass production of inexpensive, reproducible and sensitive dispos- able electrochemical sensors for the determination of trace levels of important compounds in biological fluids. The selectivity of these devices can be enhanced by judicious modification of the electrode substrate with electron media- tors and enzymes. Success in the development of these devices has led to amperometric assays for several biomolecules including: glucose,lJ cholesterol,3 hydroxybutyrate4 and several drugs.5 In recent reports69 we have described a method for the fabrication of screen-printed carbon electrodes containing selected organometallic electrocatalysts and dis- cussed their application as amperometric sensors for several other molecules of biomedical interest.This paper attempts to summarize our investigations, with particular emphasis being placed on the practical details for sensor construction and their subsequent evaluation. Three areas of our work have been described. Firstly the production of the screen-printing stencil, and the design and fabrication of the chemically modified carbon electrodes. This is followed by an overview of the electrochemical behaviour of the devices modified with phthalocyanine and ferrocene-based mediators together with their application as sensors for the direct determination of reduced glutathione (GSH) , ascorbic acid and coenzyme A.Finally, we present a summary of the development of an amperometric assay using the enzyme glutathione peroxidase together with some preliminary results regarding the selective determination of reduced glutathione in a biological sample. Production of the Screen-printing Stencil, and the Design and Fabrication of the Chemically Modified Screen-printed Carbon Electrodes (SPCEs) The aim of this work was to develop a simple technique suitable for the mass production of disposable, chemically modified carbon-based electrodes printed on an inert and robust support material.Screen-printing technology is poten- tially ideal for this application as it does not require complicated or expensive equipment and the printing process is simple, rapid and is easy to automate. The SPCEs employed for our investigations consisted of a circular 3 mm working area with a 25 x 1 mm connecting strip * Presented at the meeting on Analytical Applications of Chemi- cally Modified Electrodes, Bristol, UK, January 7-8, 1992. (Fig. 1). They were printed in parallel groups of six electrodes using the method and template described previously.4 Briefly, this is a two-stage process involving the production and implementation of a screen stencil to allow a thin-film of chemically modified or unmodified graphite suspension to be deposited onto a poly(viny1 chloride) (PVC) support material.Production and Design of the Screen-printing Stencil Fig. 2 shows the construction of the screen-printing stencil that was used to fabricate the chemically modified SPCEs (all the materials used for stencil construction and the screen-printing process were obtained from Sericol, London). It consists of a layer of Capillex direct photostencil film deposited on top of a Saatilene Monofilament polyester 100 mesh fabric, pre-tensioned to =15 N cm-* over a 254 x 254 mm wooden frame. As indicated in Fig. 2, the working stencil containing four blocks of six electrode templates (Fig. 1) was produced by placing four photographic positives of the desired electrode shapes directly on top of the Capillex emulsion (the positives were kept in intimate contact with the emulsion by compres- sion under a glass plate). The emulsion was then exposed to strong ultraviolet (UV) light (in accordance with the manu- facturer’s instructions), which served to stabilize and fix the emulsion to the screen permanently blocking the mesh pores; the areas under the opaque areas of the positives remained ‘soft’ and were subsequently washed out with a jet of water.The stencil was then force dried with a hair drier. Once dry, the working screen-printing equipment was produced using two hinges, which were used to attach one side 3 mm U H 1 mm Fig. 1 Six-electrode strip produced by the screen-printing process1282 ANALYST, AUGUST 1992, VOL. 117 of the stencil to a piece of plywood coated with a laminated plastic sheet (e.g.Formica or Contiplas board). This assembly allowed the stencil to be raised and lowered as required during the printing process. Screen-printing the Unmodified and Chemically Modified Carbon Electrodes The graphite suspension An unmodified graphite suspension was produced by adding 1.1 g of a 1.5% m/m solution of cellulose acetate in a 1 + 1 v/v mixture of cyclohexanone and acetone to 0.5 g of Ultra ‘F’ Purity graphite (Ultra Carbon Corporation) in a small glass vial. These components were mixed to give an even suspen- sion which could then be printed through the screen onto an inert PVC support (Pentawhite; Adhesive and Display Products). Fig. 2 Design and components used for the production of the screen stencil Screen-printing the electrodes The screen-printing process involved 3 steps [Fig.3(a)-(c)]. Firstly, a suitably sized ( ~ 6 0 X 45 cm) piece of PVC support material was cleaned with ethanol and attached with insulating tape (RS Components), to the solid base of the screen-printing equipment. The mixed graphite suspension was deposited onto the underside of the screen immediately adjacent to a block of six electrode templates [Fig. 3(a)]. Next, the permeable regions of the screen, corresponding to the electrode templates, were loaded with graphite suspension by lightly drawing the squeegee across the screen’s surface forcing the graphite into the mesh [Fig. 3(6)]. During this stage the screen should be slightly raised from the surface of the PVC. Finally, for the printing stage, the screen was completely lowered onto the PVC and the graphite suspension was forced through the mesh onto the support [Fig.3(c)]. The screen was then raised enabling the printed electrode strip to be removed. Once printed, the electrodes were left in a fume cupboard overnight to allow the solvents to evaporate. After use, the stencil was cleaned in a commercial thinner solution (Sericol XG). The chemically modified electrodes were prepared by adding 5% m/m of the required electron mediator to the graphite prior to the addition of the cellulose acetate solution. Immediately before use individual electrodes were cut from the PVC sheet, and the connecting strip was trimmed to 15 mm and covered with electrical insulating tape (RS Com- ponents), leaving the 3 mm working area exposed, in addition to a 6 mm length at the opposite end to allow electrical contact with the space connector in the electrode holder, (shown in Fig.4). The SPCEs were then ready for use in the voltam- metric and amperometric investigations. Electrochemical Behaviour of the SPCEs Chemically Modified with the Phthalocyanine and Ferrocene-based Mediators We began our investigations using the SPCEs chemically modified with cobalt phthalocyanine (CoPC); this organo- metallic electron mediator had previously shown promise for reducing the overpotential necessary for the oxidation of several biologically important compounds at carbon elec- trodes. 10-12 The chemically modified SPCEs were evaluated by performing cyclic voltammetry on blank solutions of 0.05 mol dm-3 phosphate buffer (pH 5) and on similar solutions containing either 1.4 mmol dm-3 ascorbic acid, 0.74 Fig.3 Screen-printing process. (a) Deposit the graphite suspension onto the screen; (b) load the screen mesh with the graphite suspension; and (c) force the graphite suspension through the screen and onto the PVC support Fig. 4 Design of the temperature-controlled amperometric cell and the SPCE holderANALYST, AUGUST 1992, VOL. 117 1283 mmol dm-3 coenzyme A or 2.2 mmol dm-3 GSH.4 It was found that the magnitude of the overpotential (q) was reduced by 350 mV for ascorbic acid and at least 600 mV for GSH and coenzyme A. (In a separate study" we determined that the anodic peak potential (Epa) for GSH at an unmodified carbon-paste electrode is c= + 1.2 V versus a saturated calomel (reference) electrode (SCE); it was not practical to measure this value at the unmodified SPCEs owing to excessively high background currents.) The oxidation of these compounds at the modified electrodes follows the mechanisms proposed in earlier work139l4 (e.g.GSH presented in Fig. 5 ) , involving the appearance of a wave at =+O. 3 V versus SCE that is absent in the cyclic voltammograms recorded at the modified electrodes in plain buffer or at the unmodified electrodes in the test solutions of each biomolecule. The useful electrochemical window for the unmodified devices determined using cyclic voltammetry with a 1 pA background cut-off point was between -1.08 and +0.85 V; this suggested that for some applications the unmodified SPCEs can also be employed for quantitative analysis.Hydrodynamic voltammograms were constructed for each of the three biomolecules using amper- ometry in stirred solutions of each compound. This technique was performed by stirring the test solutions with a small magnetic stirring rod placed in the bottom of the ampero- metric cell and measuring the difference in the anodic current between 0 V and each of the potentials applied. In each instance, insignificant currents were recorded at the unmodi- fied electrodes in either test or plain buffer solutions; likewise, only small currents were recorded at the modified electrodes in plain buffer. Indeed, the peak currents recorded for each of the biomolecules at the modified SPCEs were approximately 7.2, 20 and 150 times the background currents for ascorbic acid [Epa, +0.31 V; anodic peak current (ipa), 7.2 PA], coenzyme A (Epa, +0.45 V; ipa, 0.3 PA) and GSH (Epa, + 0.5 V; ipa, 3 PA), respectively.The performance of the CoPC modified SPCEs for the quantitative analysis of ascorbic acid and GSH was investi- gated using amperometry in stirred solutions and differential- pulse voltammetry. Using the former technique the limit of detection (LOD) was 5 x 10-8 and 1 x 10-7 mol dm-3 for ascorbic acid and GSH, respectively, based on a signal-to- noise ratio of 3 : 1. The calibration graphs were both linear from the LODs to 2 mmol dm-3 concentrations and their slopes were 1 pA per 0.16mmoldm-3 and 1 pA per 1.94 mmol dm-3 for ascorbic acid and GSH, respectively. The differential-pulse voltammograms for ascorbic acid and GSH revealed good peak shape and resolution, which indicated that excellent electron transfer kinetics were attain- able with the modified SPCEs.The calibration graphs of ipa versus concentration were linear for both biomolecules over the range 0-2.22 mmol dm-3 (n = 6). The slopes were 1 yA per 0.41 mmol dm-3 and 100 nA per 0.37 mmol dm-3 for ascorbic acid and GSH, respectively. The imprecision of the differential-pulse voltammetric method was assessed by repeated measurements of 0.05 mol dm-3 phosphate buffer (pH 5) solutions containing either 1.49 mmol dm-3 ascorbic acid or 0.92 mmol dm-3 GSH, using a fresh SPCE for each determination. The coefficients of variation were 2.8 and 6.9% for ascorbic acid and GSH, respectively (n = 6 in each instance).Fig. 5 Electrode process for the electrocatalytic oxidation of GSH at the CoPC modified screen-printed carbon electrodes In a subsequent investigation,6 SPCEs were fabricated containing the mediators presented in Table 1. The electro- chemical behaviour of these devices and their performance in the determination of GSH was assessed using cyclic voltam- metry and amperometry in stirred solutions. Cyclic voltammograms were recorded, for each of the modified SPCEs, in 0.05 mol dm-3 phosphate buffer (pH 3 , 5 and 7) and in similar solutions containing 0.48 mmol dm-3 GSH. Typically, for the ferrocene group of mediators, the cyclic voltammograms obtained in plain buffer solutions showed a single quasi-reversible redox couple (Epa, = +0.3- +0.64 V; cathodic peak potential (Epc), =+0.14-+0.49 V; AEp, -0.08-0.330 V).The exception was the SPCEs modified with dimethylferrocenedicarboxylic acid which only gave a single anodic wave (Epa, =+0.8 V); hence, the oxidation appeared to be irreversible over the potential range studied (-0.5 V-+1.2 V versus SCE). The voltammetric behaviour of the mediators was independent of pH except for ferrocenecar- boxylic acid and ferrocenedicarboxylic acid; in each instance, a break in the E,, versus pH graph (not shown) occurred at pH 5, which indicated the presence of a pKa (or pK') for the carboxylic acid groups. Similar redox behaviour was observed for each of the mediators in the presence of 0.48 mmol dm-3 GSH; however, in solutions of pH 7 an extra anodic wave (Epa,+0.65 V) was seen for the dimethylferrocenedicarboxylic acid modified SPCEs.All of the SPCEs, except these modified with dimethylferrocene, afforded an enhanced anodic current response in the presence of GSH. Interestingly, the most enhanced current responses were recorded with the electrodes modified with either the ferrocenecarboxaldehyde or dimethylferrocenedicarboxylic acid mediators which had the most positive Epa values. Of the remaining organometallic mediators presented in Table 1 only CoPC and FePC afforded a stable enhanced anodic current in the presence of GSH. It is considered that the high solubility of the other species caused them to leach from the electrode surface. The cyclic voltammograms recor- ded for the FePC modified SPCEs in plain phosphate buffer solutions (pH 3) revealed two anodic waves (Epal, = + O .l V; E a2, =+0.42 V) and one broad cathodic wave (Epc, =-0.15 Vj. The position of the more positive anodic wave was found to be dependent on the pH of the supporting electrolyte. In the pH 5 and 7 buffer solutions the single cathodic wave was resolved into two smaller waves (pH 5: EPCI, -0.08 V; Epc2, -0.23 V and pH 7: E,,, -0.13 V; Epc2, -0.30 V). These results suggested that the electrode reactions for FePC are reversible. Similar voltammetric behaviour was observed for the modified SPCEs in the presence of GSH. Bearing in mind the intention of developing an amper- ometric assay for GSH employing the enzyme glutathione peroxidase to enhance selectivity it was decided to investigate the electrocatalytic behaviour of the modified SPCEs under hydrodynamic conditions by use of amperometry in stirred solutions of 0.05 mol dm-3 phosphate buffer (pH 7).Comparison of the hydrodynamic voltammograms construc- ted for each of the modified electrodes in plain buffer Table 1 Mediators incorporated into the screen-printed electrodes Ferrocene Ferrocenecarboxaldehyde Dimeth ylferrocene Ferrocenecarboxylic acid Ferrocenedicarboxylic acid Dimethylferrocenedicarboxylic acid Cobalt phthalocyanine Iron phthalocyanine Prussian Blue Potassium hexacyanoferrate(ii1)1284 ANALYST, AUGUST 1992, VOL. 117 Table 2 Optimized calibration response factors for the mediated amperometric determination of GSH at the modified SPCEs where: ALr)AMP = EAMP unmodified -EAMP modified Reduction Applied in over- potential/ Response potential mV versus factor/pA (AL~)AM~)/ Mediator SCE mmol-ldm3 mV Cobalt phthalocyanine +350 0.299 >lo00 Iron phthalocyanine + 70 0.100 >lo10 Ferrocenedicarboxylic acid +240 0.001 >lo00 Ferrocenecarboxaldeh y de +450 4.760 >880 Dimethylferrocene- +725 3.911 >670 dicarboxylic acid solutions and those containing 0.48 mmol dm-3 GSH revealed that under these conditions only ferrocenedicarboxylic acid, ferrocenecarboxaldehyde, dimethylferrocenedicarboxylic acid, FePC and CoPC afforded a significant and stable enhancement in the anodic current response.The maximum calibration response factors obtained using amperometry in stirred solution in conjunction with the SPCEs modified with these mediators are presented in Table 2; this table also presents the reduction in the measured overpotential attain- able with each group of modified device.This data indicated that the choice of the most suitable mediator for use in the enzyme assay would be a compromise between sensitivity and selectivity. The FePC modified devices were likely to be the most selective although their sensitivity was poor; conversely, the ferrocenecarboxaldehyde modified devices afforded the greatest sensitivity but their selectivity was inferior. There- fore, CoPC was selected as the mediator of choice for further investigation as we considered it afforded the most promising balance between the two criteria. Amperometric Assay Employing the CoPC Modified SPCEs and Glutathione Peroxidase for the Selective Determination of GSH In this study we set out to further enhance assay selectivity by use of the CoPC modified SPCEs in conjunction with bovine erythrocyte glutathione peroxidase (E.C.1.11.1.9). Our intention was to use the enzyme to selectively remove GSH from solution; the rate and extent of this reaction would be followed amperometrically using the CoPC modified SPCEs. Optimization and Characterization As part of these investigations we performed several experi- ments to characterize the enzyme’s behaviour with a view to optimizing the experimental conditions for its application to the amperometric determination of GSH in biological mat- rices such as human whole blood. Initial studies employed UV spectroscopy, cyclic voltam- metry and amperometry in stirred solutions (the cell used for this technique is presented diagrammatically in Fig.4) to compare the performance of several peroxides as potential electron acceptors for the enzymic oxidation of GSH. Several other workers have studied this reaction, the most promising results for our current work were described by Awasthi eta1.15 who used tert-butyl hydroperoxide, Zakowski and Tappel16 who employed cumene hydroperoxide, and Paglia and Valen- tine,17 Nakamura et al.18 and Hua et al.19 who studied the use of hydrogen peroxide. The enzymic oxidation reactions with these compounds can be represented by the following equations.20 Hydrogen peroxide: 2GSH + H202 + GSSG + 2H20 Cumene and tert-butyl hydroperoxide: 2GSH + ROOH GSSG + ROH + H2O For our amperometric assay, we were particularly interested in selecting a substrate that would not be electrochemically active under the reaction conditions employed and would allow the enzymic reaction to proceed rapidly whilst not causing any spontaneous oxidation of GSH.To investigate these selection criteria we initially used UV spectroscopy (A, 263 nm) to follow the formation of GSSG and the DTNB [5,5’-dithiobis(2-nitrobenzoic acid)] colorimetric reaction21 to determine GSH, after the addition of stoichio- metric quantities of each peroxide to a cuvette containing 3 cm3 of 1 mmol dm-3 GSH in 0.05 mol dm-3 phosphate buffer (pH 7.4) and similar solutions containing the enzyme. These experiments confirmed the proposed reaction mechanisms given above and earlier results presented by Flohk et ~1.22 and Abedinzadeh et al.23 who suggested that significant spon- taneous oxidation of GSH occurred in the presence of hydrogen peroxide.In addition, cyclic and hydrodynamic voltammograms (the latter were constructed from the current responses recorded using amperometry in stirred solutions [Eappljed, -0.5 to +0.80V versus SCE]) revealed that this substrate, and also cumene hydroperoxide, were unsuitable for further work as they gave rise to high background anodic currents at the CoPC modified SPCEs over the intended applied potential range (+0.2 to +0.4 V versus SCE). Indeed, only tert-butyl hydroperoxide satisfied the 3 requirements for a substrate in the amperometric assay. Subsequent investigations using tert-butyl hydroperoxide revealed that the optimized solution conditions for maximum glutathione peroxidase activity were: 0.05 mol dm-3 phos- phate buffer (pH 8.8) containing 1 mmol dm-3 ethylenedi- aminetetraacetic acid (EDTA) at 37°C.However, GSH is unstable in alkaline media and these conditions caused rapid spontaneous oxidation of a standard 1 mmol dm-3 GSH solution (0.16 pmol cm-3 min-1); indeed, in order to maintain GSH stability it was necessary to use pH 7 phosphate buffer and to reduce the temperature to 25°C. Under these conditions the apparent Michaelis-Menten constant (K,) was 11.9 mmol dm-3. This is an apparent K, for GSH as the enzyme does not saturate with respect to the thiol concentra- tion. 17724 The activation energy determined from an Arrhenius plot was 35.2 kJ mol-1, which is in good agreement with the value of 34.3 kJ mol-1 determined by Awasthi et al.13 for human glutathione peroxidase with tert-butyl hydroperoxide.The pKa corresponding to the enzyme’s active site was pH 6.4, which agrees with previously quoted values of pH 4.9-6.5.17 Calibration and Precision A typical anodic current response profile is shown in Fig. 6(a). As indicated, the current response increased in a step-wise manner following the addition of the GSH to the amper- ometric cell (final volume, 5 cm3) and the current rapidly reached a steady state after a short delay whilst the solutions mixed. After approximately 5 min the glutathione peroxide and tert-butyl hydroperoxide were added to the cell where- upon the magnitude of the anodic current response decreased as the GSH substrate was consumed by the enzyme.Calibra- tion graphs (not shown) of the initial rate of current change measured over the first 60 s (following the addition of the enzyme) versus GSH concentration were linear using standard solutions containing 0, 10, 20,30,40 and 50 ymol dm-3 GSH. Concentrations beyond these values were not studied as they would not have been appropriate to physiological levels after the envisaged sample preparation stage corresponding to a 20-fold dilution of whole blood in phosphate buffer.ANALYST, AUGUST 1992, VOL. 117 1285 ( a ) 0.025 cm3 of 0.01 mol dm-3 Po nA I GSHadded Enzyme and tert-butyl hydroperoxide added lk I I SPCE 1 Enzyme and tert-butyl hydroperoxide Time - Fig. 6 Amperometric current response profiles for ( a ) a 0.05 mol dm-3 phosphate buffer solution (pH 7) spiked at the indicated point to give 50 pmol dm-3 GSH and (b) a haemolysate sample prepared from human whole blood after pretreatment by ultrafiltra- tion The imprecision of the amperometric method for several buffer solutions containing 40 pmol dm-3 GSH was 7.28% (n = 5).Selectivity The selectivity of the amperometric assay was assessed using several important physiological and pharmaceutical com- pounds. In each instance, a small volume of a stock solution of the potential interferent was introduced into the amperome- tric cell to give a 50 pmol dm-3 final concentration, and the magnitude of any steady-state current measured. Subse- quently, the tert-butyl hydroperoxide and glutathione peroxi- dase were added and any corresponding decreases in the anodic current were measured.The former current measure- ment provided an index of any interference owing to direct electrochemical activity at the modified electrode surface; whilst the latter could be used to determine any cross- reactivity of the test compound with the enzyme (Table 3). The data presented in Table 3 indicated that of the naturally occurring species studied only ascorbic acid, cysteine and uric acid gave rise to a current response at the modified SPCEs; however, their cross-reactivity with the enzyme was deemed to be of no practical consequence for the determination of GSH in whole blood. For the selected group of drugs and metabolites that can be found in the circulation of patients undergoing treatment for rheumatoid arthritis (a possible indication for the quantitation of whole blood GSH), only ~-penicillamine and gentisic acid gave a current response at the SPCEs and neither species exhibited any cross-reactivity with glutathione peroxidase.Preliminary Studies towards the Determination of GSH in Whole Blood Whole blood haemolysate solutions were prepared by snap- freezing 250 mm3 aliquots of screened human blood samples in an acetone-solid carbon dioxide (dry ice) bath for 30 s; whereupon, 4.75 cm3 of the 0.05 mol dm-3 phosphate buffer (pH 7) was added and the suspension mixed. After centrifuga- Table 3 Amperometric assay selectivity: the extent of any direct electrochemical interference at the CoPC modified SPCEs and values for any enzyme cross-reactivity Compound studied D-Glucose Creatinine Pyruvic acid Lactic acid Ascorbic acid Uric acid Cysteine D-Penicillamine Azapropazone Naproxen Sodium aurothiomalate Diclofenac Sodium salicylate Gentisic acid Reduced glutathione (GSH) Anodic current responseInA 0 0 0 0 299 48 93 51 4 0 0 0 0 244 46 Enzyme activity compared with GSH (%) 0 0 0 0 0 1 13 0 0 0 0 0 0 0 100 * tion, the supernatant was removed from the pellet of cellular debris and transferred either directly to the amperometric cell or passed through an Amicon ultrafiltration unit equipped with a 30000 Da relative molecular mass cut-off filter.As expected, our initial attempts using the amperometric assay on samples of unfiltered haemolysate were unsuccessful owing to rapid fouling of the electrode surface. However, fouling could be prevented by subjecting the haemolysate sample to ultrafiltration prior to performing the amperometric assay.Fig. 6(6) presents a current response profile for a treated human haemolysate sample and clearly demonstrates the efficacy of the ultrafiltration step in preventing electrode contamination. Future work will involve the incorporation of membranes directly onto the SPCE surface to prevent fouling and hence, remove the need for a separate ultrafiltration stage. In conclusion, our work has successfully demonstrated that chemically modified screen-printed carbon electrodes may be produced easily and rapidly; furthermore they may be fabricated to contain different mediators as appropriate to the required electrochemical sensor application. In addition, we have shown encouraging results that indicate that these carbon-based sensors chemically modified with the electron mediator cobalt phthalocyanine might be used in conjunction with the enzyme glutathione peroxidase for the selective determination of reduced glutathione in whole blood samples.The authors are grateful to the National Advisory Board for financial support. They also thank Louis Bracey, Brian Birch, Glaxo Group Research and co-workers at Bristol Polytechnic for their interest in this work. References Frew, J. E., Bayliff, S. W., Gibbs, P. N. B., and Green, M. J., Anal. Chim. Acta, 1989, 224, 39. Scott, D., Lancet, 1988, 778. Hilditch, P. I., and Green, M. J., Analyst, 1991, 116, 1217. Batchelor, M., Green, M. J., and Sketch, C. L., Anal. Chim. Acta, 1989, 221, 289. Green, M. J., and Hilditch, P. I., Anal. Proc., 1991, 28, 374. Wring, S. A., Hart, J. P., Bracey, L., and Birch, B. J., Anal. Chim. Acta, 1990, 231, 203. Wring, S. A., Hart, J. P., Thompson, J. F., and Birch, B. J., Anal. Proc., 1990, 27, 209. Wring, S. A., Hart, J. P., and Birch, B. J., Analyst, 1991, 116, 123. Wring, S. A., Hart, J. P., and Birch, B. J., Electroanalysis, 1992,4,299.1286 ANALYST, AUGUST 1992, VOL. 117 10 11 12 13 14 15 16 17 Zagal, Z., Fierro, C., and Rozas, R., J. Electroanal. Chem., 1981, 119,403. Korfage, K. M., Ravichandran, K., and Baldwin, R. P., Anal. Chem., 1984,56, 1517. Halbert, M. K., and Baldwin, R. P., Anal. Chem., 1985, 57, 591. Wring, S. A. Hart, J. P., and Birch, B. J., Analyst, 1989, 114, 1563. Wring, S. A., Hart, J. P., and Birch, B. J., Anal. Chim. Acta, 1990,229,63. Awasthi, Y. C., Beutler, E., and Srivastava, S. K., J . Biol. Chem., 1975,250,5144. Zakowski, J. J., and Tappel, A. L., Anal. Biochem., 1978, 89, 430. Paglia, D. E., and Valentine, W. N., J. Lab. Clin. Med., 1967, 70, 158. 18 Nakamura, W., Hosoda, S., and Hayashi, K., Biochem. Biophys. Acta, 1974,358, 251. 19 Hua, C., Smyth, M. R., and O’Fagain, C., Analyst, 1991, 116, 929. 20 Wendel, A., in Enzymatic Basis Of Detoxification, Jakoby, W. B., Academic Press, New York, 1980, pp. 333-353. 21 Ellman, G. L., Arch. Biochem. Biophys., 1959, 82,70. 22 FlohC, L., Loschen, G., Gunzler, W. A., and Eichele, E., Hoppe-Seyler’s Z . Physiol. Chem., 1972, 353, 987. 23 Abedinzadeh, Z., Gardes-Albert, M., and Ferradini, C., Can. J. Chem., 1989, 67, 1247. 24 FlohC, L., Basic Life Sci., 1988, 49, 663. Paper 21001 1 4 0 Received January 10, 1992 Accepted April 29, 1992

ISSN:0003-2654    

DOI:10.1039/AN9921701281

出版商:RSC    

年代:1992

数据来源: RSC

摘要:

ANALYST, AUGUST 1992. VOL. 117 1287 Electrochemical Immobilization of Enzymes Part V.* Microelectrodes for the Detection of Glucose Based on Glucose Oxidase Immobilized in a Poly(pheno1) Filmt Philip N. Bartlett and Daren J. Caruana School of Chemistry, University of Bath, Bath BA2 7AY, UK The adsorption of glucose oxidase at a platinum electrode followed by immobilization in an electrochemically polymerized phenol film was found to be a reproducible method for the fabrication of enzyme microelectrodes responsive to glucose. The observed responses were comparable to those predicted on the basis of monolayer coverage of the enzyme. The electrodes were found to be stable for more than 40 d on storage at 4 "C. Keywords: Enzyme immobilization; glucose oxidase; D-aminO acid oxidase; poly(pheno1) film; glucose microsensor Enzyme immobilization technology has attracted much atten- tion in recent years in the area of electrochemical biosensors.Immobilization of enzymes onto solid surfaces is an important part of amperometric or potentiometric biosensor design. A variety of innovative immobilization techniques have been used such as adsorption, covalent binding, entrapment in a porous matrix and confinement behind a dialysis mem- brane.*.* Immobilization can lead to changes in enzyme structure and hence cause the kinetics, stability and specificity to differ from that of the enzyme in homogeneous solution.3 Electrochemical polymerization has been used successfully to immobilize enzymes at electrode surfaces4.5 and to reduce interferences and fouling in amperometric biosensors.6J This method has a number of significant advantages as an approach towards biosensor fabrication: first, the method is flexible and can be readily controlled; second, it is simple to carry out; and third, the polymer deposition is localized at the electrode surface so that the method is suited to the spatially localized deposition of enzymes onto microelectrode arrays.In previous papers it was shown that glucose oxidase was adsorbed onto the surface of a platinum electrode and that this thin layer of adsorbed enzyme retained its enzymic activity when held in place by an electropolymerized film of N-methylpyrrole*.9 or by electropolymerized films of phenol or phenol derivatives.10 The use of electropolymerized phenolic films is attractive because the pre-adsorption of the enzyme onto the electrode surface provides some degree of molecular self-assembly and because the growth of the polymer is self-limiting.In addition, the insulating phenolic film may provide some degree of selectivity against interferent species and fouling of the electrode surface. Recent studies have shown that poly(pheno1) films can be used to confer permselectivity at the electrode surface. 11-12 The immobilization of enzymes onto microelectrodes can be used to produce biosensors for in vivo applications and for use with small sample volumes. In addition, the response times for microelectrodes can be much faster than those for the corresponding macroelectrode. Shinohara et a f . 13 have described a microsensor for glucose based on the entrapment of glucose oxidase in an electropolymerized aniline film grown on a 50 ym diameter platinum wire.By using this system they were able to detect glucose by monitoring the consumption of oxygen at -0.6 V versus Ag-AgCI. They observed responses which saturated at about 5 mmol dm-3 glucose. Wang et al. 14 have described a modified 7 pm carbon fibre microsensor for * For Part IV of this series see ref. 9. 7 Presented at the meeting on Analytical Applications of Chemi- cally Modified Electrodes. Bristol, UK, 7-8 January, 1992. glucose based on the electrochemical co-deposition of glucose oxidase and platinum. In this instance a 4-5 mm length of the fibre was modified and exposed to the sample stream and the production of hydrogen peroxide at +0.8 V versus Ag-AgC1 was measured.They observed a 50% decrease in response over a 20 d period using the electrode for flow injection. The same group have extended this approach to co-deposit glucose oxidase and rhodium onto a 7 pm carbon fibre.15 In this instance, because rhodium is a good catalyst for the oxidation of hydrogen peroxide, they were able to detect hydrogen peroxide at +0.35 V. Unfortunately, rhodium also catalyses the oxidation of uric acid and ascorbate so that interference from these species remained a problem. Abe et af.16 have described the modification of a 2 pm diameter platinized carbon ring electrode for the in vivo measurement of glucose. In this instance the enzyme was immobilized by gluter- aldehyde cross-linking onto the porous platinum coated electrode. Glucose was monitored by detection of hydrogen peroxide at +0.6 V versus SSCE (sodium chloride saturated calomel electrode).In this paper, previous work on the immobilization of glucose oxidase in phenol filmslo is extended; the immobiliza- tion of glucose oxidase by pre-adsorption onto 125 and 25 pm diameter platinum microelectrodes followed by electrochem- ical deposition of a poly(pheno1) film is reported. &xperimental Reagents All solutions were freshly prepared using water purified by a Whatman WRSO RO/de-ionizing system followed by a What- man STILLplus carbon filter. Phenol (Fisons, SLR grade, sublimed) solution was freshly prepared before use in 0.150 mol dm-3 disodium hydrogen orthophosphate buffer, pH 7.0 (Fisons Analytical Reagent), containing 0.1 mol dm-3 tetraethylammonium tetrafluoroborate, TEATFB (Aldrich, recrystallized three times from AR methanol and dried under vacuum).Glucose oxidase [E.C. 1.2.4.4, Type VII, 203.8 U mg-1 (1 U = 16.67 nkat), purified from Aspergiffus niger] was a gift from MediSense UK, and was stored in buffer at 4 "C. Measurements of the electrode responses were made in 0.150 mol dm-* disodium hydrogen orthophosphate buffer, pH 7.0. Stock solutions of P-D-(+)-glucose (1 .O mol dm-3) were prepared from AnalaR D-glucose [BDH (now Merck)] and allowed to equilibrate at room temperature for 24 h before use. Uric acid (Aldrich), 4-acetoamidophenol (Sigma), 2-deoxy-~-glucose (Sigma), D-galactose (Sigma), D-alanine (Fluka) and D-amino acid oxidase (E.C.1.4.3.3, Sigma, porcine kidney Type I) were used as received.1288 ANALYST, AUGUST 1992, VOL. 117 Solutions of ascorbic acid (Aldrich) were used immediately after preparation to avoid problems of oxidative decomposi- tion. Instrumentation All measurements were made using a conventional three- electrode system consisting of a large area platinum gauze counter electrode, a saturated calomel reference electrode (SCE) and a platinum microelectrode, unless stated other- wise. All potentials are reported with respect to the SCE. The potentiostat was purpose-built and incorporated a Faraday cage to reduce noise. It was used in conjunction with a Bryans 60O00 series XYIt recorder and a Keithley digital voltmeter. All measurements were made at room temperature in an undivided cell.The solution was stirred using a purpose-built air-powered stirrer to facilitate mixing on addition of aliquots of glucose solution. Preliminary flow injection measurements were made using the same potentiostat but connected in a two-electrode configuration using a laboratory-built calomel electrode as the combined reference and counter electrode, and with the microelectrode as part of a purpose-built flow system based on a Pharmacia single-channel pump. Non- linear least-squares fitting of the experimental responses to theory was carried out using the SigmaPlot program (Jandel) on a 486 microcomputer. Microelectrode Fabrication Platinum wire (25 and 125 vm diameter) (Goodfellow Metals) was sealed in soft glass capillaries and the end sections were polished using 'wet and dry' paper followed by successively finer grades of diamond lapping compounds (Engis) down to 0.1 pm.All electrodes were stored in buffer solution at all times. Prior to any experiment the electrode was cycled in 2.0 rnol dm-3 sulfuric acid (Aldrich; AR grade) between -0.2 and +1.5 V for approximately 15 min. The electrodes were regularly checked by chronoamperometry in 1 mmol dm-3 aqueous ferrocenemonocarboxylic acid and any electrodes which showed signs of leaking or large capacitative currents were discarded. Enzyme Immobilization The enzyme was immobilized from a solution containing 3.15 mg cm-3 glucose oxidase, 0.1 mol dm-3 TEATFB and 0.05 mol dm-3 phenol in 0.15 mol dm-3 disodium hydrogen orthophosphate buffer at pH7.0. The electrode was im- mersed in the growth solution and allowed to stand for 15 min at open circuit prior to electropolymerization, which was carried out by connecting the electrode at 0.0 V, waiting for 20 s and then stepping the potential to +0.9 V and holding it at this value for 8 min.The electrode was then stepped back to 0.0 V, disconnected after 20 s and washed for 5 min in stirred background buffer solution. For the stability measurements the electrodes were kept in glucose-free buffer solution at 4 "C between measurements. Results Theory The steady-state current for an electrode coated with a thin film containing glucose oxidase is given by10 where n is the number of electrons transferred, F is the Faraday constant, A is the electrode area, iobs is the observed current for hydrogen peroxide detection, KM and kcat describe the reaction of the enzyme with glucose, k is the rate constant describing the re-oxidation of the mediator, 1 is the thickness of the immobilized layer, ez is the enzyme concentration in the layer, Ks and KA are the partition coefficients into the film and s, and a , are the bulk concentrations on the substrate and mediator, respectively.The parameter a describes the balance between the detection of the reduced mediator at the electrode surface and its loss to the bulk solution and is given (after correcting a typographical error in ref. 10) by where XD is the diffusion layer thickness, DB and KB are the diffusion and partition coefficients for the reduced mediator in the film, DB,soln is its diffusion coefficient in solution, and 0.5 < a: < 1.For a microdisc electrode XD is replaced by the electrode radius, r0.17 By using eqn. (l), our data from studies of the current for the enzyme-coated electrode can be fitted as a function of the concentration of substrate, the concentration of mediator and the enzyme loading and hence the kinetics for the reaction in the immobilized layer can be characterized. Responses to Glucose Fig. 1 shows responses to the addition of aliquots of stock glucose solution for a 25 pm diameter poly(pheno1)-glucose oxidase modified electrode in air-saturated buffer solution, held at +0.9 V to detect hydrogen peroxide. The response of the electrode is rapid and is determined by the mixing time in these experiments. The response is approximately linear up to 12 mmol dm-3 (Fig.2) with an estimated lower detection limit 500 I 400 300 Qn *% 200 100 I I I 0 100 200 300 Tim e/s Fig. 1 Current responses for a 25 pm diameter glucose oxidase microelectrode recorded at 0.9 V in background buffer at pH 7. Each step corresponds to a change in glucose concentration of 5 mmol dm-3 250 200 Qn 150 .P . .A 100 50 0 4 8 12 16 [ G lucose]/m mo I dm - 3 Fig. 2 Calibration graph for the measurement of glucose at a 25 pm diameter electrode in background buffer solution at pH 7. The electrode was held at 0.9 V for 15 min before the measurement to allow the background current to stabilize. The background current, ibg, was 40 PA. The solid line is the best fit to eqn. (1)ANALYST, AUGUST 1992, VOL. 117 1289 400 I 1 300 B . .p 200 100 0 20 40 60 80 [Glucosel/mmol dm-3 Fig.3 Response curves for three different 25 ym diameter glucose microelectrodes. Conditions as in Fig. 2. The background currents, ibg, were between 50 and 100 PA. The solid lines are the best fits to eqn. (1); the fitting parameters are given in Table 1 for glucose of 230 ymol dm-3. The response saturates at about 0.1 mol dm-3 glucose. The responses are reproducible from electrode to electrode and from day to day for freshly prepared electrodes. Fig. 3 shows a set of three different electrodes prepared on different days using different growth solutions of the same composition. Table 1 summarizes the kinetic data based on the analysis of the responses to glucose using eqn. (1) for a number of electrodes.Similar results were obtained when the concentra- tion of enzyme in the growth solution was reduced to 0.63 mg cm-3 (Table l), consistent with the suggestion that the enzyme is pre-adsorbed onto the electrode surface. The responses are proportional to the electrode area. According to eqn. (1) the response at high glucose concentration can be limited by either the reaction of the mediator with the enzyme or by the saturated enzyme- substrate kinetics. In order to investigate this, measurements were made in both air-saturated and oxygen-saturated buffer. Fig. 4 shows the responses to glucose at a microelectrode in the two solutions. The current clearly saturates at a lower level in the air-saturated buffer, indicating that the current at high glucose concentrations under these conditions is limited by the rate of re-oxidation of the enzyme by the oxygen.Kinetic parameters from the analysis of these data according to eqn. (1) are given in Table 1. Stability and Specificity The specificity of the microelectrodes for glucose was investi- gated (Fig. 5). Fig. 5(a) shows that without the enzyme in the film there is no measurable response to glucose. It was found that the specificity for glucose as against other sugars is unaffected by the immobilization process [Fig. 5(b)]. Glucose oxidase is known to oxidize 2-deoxy-~-glucose at a reduced rate, whereas it has no activity towards D-galactose oxida- tion,18 and this is also seen for our microelectrodes. The stability of the electrodes was also studied by recording glucose calibration graphs at intervals over a period of 40 d (Fig.6). The response falls initially but then remains stable on storage of the electrode. Analysis of the responses allows the corresponding kinetic parameters for the electrode to be calculated (Table 2). Preliminary studies show that the electrodes are also stable in a flowing system and can be used for flow injection. Interferences Some of the major potential interferent species encountered in in vivo measurements, or in whole blood, are ascorbate, uric acid and acetoamidophenol.7~19 In principle the phenol film should help to block the reaction of these interferent species at Table 1 Kinetic parameters from non-linear least-squares best fits of data to eqn. (1) OLKSkcated Nkcated KM KM (1 + kCat/kKAa,) Ks( 1 + kCat/kKAa,) /ems-' /mol cm-2s-1 /mol cm-3 Large Pt electrode (0.38 cm2):* Air saturated 5.8 x 10-6 1.1 x 19 Ozsaturated 6.2 x 2.5 X l0-lo 40 25 pm diameter electrode, 3.5 mg ~ r n - ~ glucose oxidase in growth solution : Air saturated: #76 19 x 10-6 4.5 x 10-10 24 #77a 22 x 10-6 5.0 x 10-10 23 #81 20 x 10-6 4.8 x 10-10 24 #78 18 x 10-6 4.5 x 10-10 25 02saturated 45 x 10-6 9.7 x 10-10 22 25 pm diameter electrode, 0.63 mg cm-3 glucose oxidase in growth solution: Airsaturated 20 x 4.3 X 22 125 ym diameter electrode: Air saturated 22 x 10-6 5.4 X 10-10 25 * Ref.10. I 1 I 600 .!. I I P /-- I 2oo F d I I I 0 20 40 60 80 [Glucose]/mmol dm-3 Fig. 4 Responses to glucose for a 25 pm diameter glucose micro- electrode in 0, air-saturated; and V , oxygen-saturated buffer solution.The solid lines are the best fits to eqn. (1); the fitting parameters are given in Table 1 the platinum electrode. Fig. 7 shows voltammograms for these three species, at maximum physiological concentrations, at a clean platinum microelectrode, at a poly(pheno1) coated platinum microelectrode and at a poly(pheno1)-glucose oxi- dase coated platinum microelectrode. It is clear that at the polymer coated electrodes the response to all three species is significantly reduced. The film containing the enzyme appears to be slightly less good in this respect when compared with the enzyme-free poly(pheno1) film. In practice the responses for the interferent species must be compared with those obtained for glucose (Table 3). Studies of the response to glucose in the presence of a fixed amount of ascorbate (5 mmol dm-3) showed that the currents due to ascorbate oxidation and glucose are directly additive, i.e., the presence of ascorbate has no effect on the kinetics of the enzyme reaction within the film but rather simply contributes a constant amount (in this instance 150 PA) to the background current.Immobilization of D-Amino Acid Oxidase This approach to the immobilization of enzymes on microelectrodes is reasonably flexible. A preliminary study of1290 ANALYST, AUGUST 1992, VOL. 117 a 0 I I e 200 100 A t t3t c t I 1 I I 1 0 50 100 150 Ti me/s Fig. 5 (a) Res onse to additions of aliquots of glucose solution for a 25 pm diameter platinum microelectrode coated with an enzyme-free poly(pheno1) fih. Each arrow corresponds to a 5 mmol dm-3 change in glucose concentration; the currents were recorded at 0.9 V in air-saturated background buffer at pH 7.(b) Response of a 25 pm diameter platinum electrode coated with a poly(pheno1)-glucose oxidase film to various sugars. Each arrow corresponds to a step change in concentration of 5 mmol dm-3: A, 2-deoxy-~-glucose; B, D-galactose; and C, glucose. Currents were recorded at +0.9 V in air-saturated buffer solution, pH 7 8t 0 10 20 30 40 50 [Glucose]/mmol dm-3 Fig. 6 Effect of storage on the response of a 125 pm diameter glucose sensor. The electrode was stored at 4°C in background buffer between measurements. Responses were measured at 0.9 V in air-saturated buffer, pH 7. The curves are the best fits to eqn. (1); the fitting parameters are given in Table 2.The background currents, ibg, were in the range 0.35-0.45 nA. 0, Day 1; V, day 7; 0, day 23; and 0, day 40 the immobilization of D-amino acid oxidase at a 25 ym platinum microelectrode was carried out. Again, it was possible to detect the substrate by following the production of hydrogen peroxide at +0.9 V. Fig. 8 shows a typical calibra- tion graph for the response of the electrode to the addition of aliquots of D-alanine solution. In this instance the response saturates at about 60 mmol dm-3. Analysis of the response according to eqn. (1) gives aKskcatexZ/KM = 4.1 X 10-7 cm s-1 and akca,exZ/(l + kcat/kKAa,) = 6 x 10-12 mol cm-2 s-1 and hence KM/Ks(l + kcat/kKAa,) = 2.4 mmol dm-3. Discussion The fabrication method for glucose microelectrodes described here is reproducible, as evidenced by the kinetic parameters Table 2 Kinetic parameters for a 125 pm electrode as a function of storage time. All measurements were made in air-saturated buffer Day 1 22 x 10-6 5.4 x 10-10 25 Day 7 14 x 10-6 4.7 x 10-10 34 Day 23 13 x 10-6 4.8 x 10-10 37 Day 40 14 x 10-6 4.5 x 10-10 32 for replicate electrodes presented in Table 1, and it produces stable electrodes.Examination of the stability data (Table 2) suggests that the decrease in the response between day 1 and day 7 is, in part, due to changes in the partition of glucose into the film, described by Ks, and in part due to loss of enzyme activity . By using this method of immobilization it is believed that there is a monolayer of adsorbed enzyme entrapped within a thin poly(pheno1) film.10 It is instructive to compare the current densities obtained for these electrodes with the value calculated on the basis of a monolayer coverage of enzyme at the electrode surface and assuming that rate constants for the enzyme-catalysed oxidation of glucose are unaltered from their homogeneous solution values.Fig. 9 shows the calcu- lated response assuming an enzyme coverage of 1.6 X 10-12 mol cm-2 (based on the studies of glucose oxidase adsorption at gold electrodes by Szucs et aL20), taking kcat and KM to be 800s-1 and 33mmoldm-3, respectively,21 and assuming that the re-oxidation of the enzyme is not rate limiting. Current densities calculated on this basis are very similar to those measured by Sun et aZ.22 for a monomolecular layer of glucose oxidase cross-linked in a Langmuir-Blodgett film and deposited at a platinum electrode.Fig. 9 also shows the observed responses to glucose for one of the electrodes prepared here. In the microelectrode experiments, loss of hydrogen peroxide by diffusion away from the electrode surface is expected to be efficient, so that in eqn. (1) (x =r 0.5. ItANALYST, AUGUST 1992, VOL. 117 ............................ .’: c----- --L ..... -.; / I larr.. /&*.T.? =._.-.- . - -.- - . . . . . . . . . . . . . SH ..... . .>--- ./ ....... 1291 1.0 0.5 0 - 1.5 1 .o 0.5 0 -0.5 - - 1.5 1 .o 0.5 0 -0.5 -1.0 I I I I I I -0.5 I I I I I I 1 -200 0 200 400 600 800 1000 EImV versus SCE Fig. 7 Cyclic voltammograms for: (a) 0.11 mmol dm-3 L-ascorbic acid; (b) 0.17 mmol dm-3 acetoamidophenol; and (c) 0.48 mmol dm-3 uric acid.All voltammograms were recorded at 20 mV s-1 in background buffer, pH 7, at 25 pm diameter microelectrodes. In each instance the three curves correspond to: clean platinum (solid curve); poly(pheno1) coated platinum (broken curve); and poly(pheno1)-glucose oxidase coated platinum (dotted curve) Table 3 Electrode response for various interferents at maximum physiological concentrations for a 25 pm diameter electrode at 0.9 V in pH 7 buffer Equivalent glucose Concentration/ Response/ concentration*/ Interferent mmol dm-3 PA mmol dm-3 Ascorbic acid 0.11 40 2.1 Uric acid 0.48 18 0.9 Ace toamidophenol 0.17 355 19 * Based on the unsaturated glucose response. 5 4 2 3 \ .s .L 2 1 0 I I I 20 40 60 [o-Alanine]/mmol dm-3 Fig.8 Calibration graph for a poly(phenol)-D-amino acid oxidase 25 pm diameter platinum microelectrode for D-alanine recorded at 0.9 V in background buffer, pH 7. The background current, ibg, was 40 PA I I I 1 0 50 100 150 200 [Glucosel/mmol dm 3 Fig. 9 Comparison of the response calculated for a 25 pm diameter electrode on the basis of a monolayer coverage of glucose oxidase at the electrode (solid line) and allowing for loss of hydrogen peroxide to the bulk solution, a = 0.5 (broken line). The points are typical experimental data for air-saturated buffer at pH 7 is clear that, when this factor of 0.5 is taken into account, the calcuIated and observed currents are of very similar magni- tude. This indicates that the adsorbed enzyme layer remains active and that the magnitude of the response is determined by the rate of enzyme turnover; this method of immobilization does not appear to decrease the activity of the enzyme significantly .The results of the studies of the interferent species are less satisfactory. Although it is clear that the poly(pheno1) film has a significant effect on the currents measured for uric acid, acetoamidophenol and ascorbate when compared with the clean platinum electrode it is also clear that the suppression of1292 ANALYST, AUGUST 1992, VOL. 117 the response is still not sufficient. The responses to these interferents at maximum physiological concentrations remain significant when compared with the measured glucose responses (Table 3). There are a number of possible ways to overcome this problem.First, the use of a poly(pheno1ic) film with negatively charged substituents should help to exclude the anionic interference species.23 Second, the reduction of the operating potential for the electrode by the use of a redox mediator24 or by modification of the enzyme25-29 may help. These possibilities are currently being investigated. Finally, it is instructive to consider the data for the D-amino acid oxidase electrode. First, it is encouraging that this approach can be used for other enzymes. Second, it is noticeable that the currents for the D-amino acid system are about 100 times smaller than those for the glucose electrode. In part this is because D-amino acid oxidase is a much less active enzyme than glucose oxidase30 (kcat = 55 s-1; KM = 1.8 mmol dm-3).The authors thank MediSense Inc. for the gift of glucose oxidase. This work was supported by an SERC Studentship to D. J. C. References Gray, D. N., Keys, M. H., and Watson, B., Anal. Chem., 1977, 49, 1067A. Immobilised Cells and Enzymes, ed. Woodward, J., IRL Press, Oxford, 1985. Engasser, J.-M., and Horvath, C., in Applied Biochemistry and Bioengineering. Vol. I , Immobilised Enzyme Principles, ed. Wingard, L. B., Katchalski-Katzir, E., and Goldstein, L., Academic Press, New York, 1976, pp. 127-220. Bartlett, P. N., Tebbutt, P., and Whitaker, R. G., Prog. React. Kinet., 1991, 16, 55. Bartlett, P. N., and Whitaker, R. G., Biosensors, 1987/88, 3, 359. Sasso, S. V., Pierce, R. J., Walla, R., and Yacynych, A. M., Anal. Chem., 1990, 62, 1111. Geise, R.J., Adams, J. M., Barone, N. J., and Yacynych, A. M., Biosens. Bioelectron., 1991, 6, 151. Bartlett, P. N., and Whitaker, R. G., J. Electroanal. Chem., 1987, 224, 37. Bartlett, P. N., Ali, Z., and Eastwick-Field, V., J. Chem. SOC., Faraday Trans., in the press. 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Bartlett, P. N., Tebbutt, P., and Tyrrell, C. H., Anal. Chem., 1992,64, 138. Wang, J., Chen, S.-P., and Lin, M.-S., J. Electroanal. Chem., 1989, 273,231. McClarley, R. L., Irene, E. A., and Murray, R. W., J. Phys. Chem., 1991,95,2492. Shinohara, H., Chiba, T., and Aizawa, M., Sens. Actuat., 1988, 13, 86. Wang, J., Li, R., and Lin, M.-S., Electroanalysis, 1989, 1, 151. Wang, J., and Angnes, L., Anal. Chem., 1992, 64, 456. Abe, T., Lau, Y. Y., and Ewing, A. G., J. Am. Chem. SOC., 1991, 113, 7421. Pons, S., and Fleischman, M., Anal. Chem., 1987,59, 1391a. Dixon, M., and Webb, E. C., The Enzymes, Longman, New York, 3rd edn., 1979, p. 243. Palleschi, G., Rahni, M., Lubrano, G. J., Ngwainbi, J. N., and Guilbault, G. G., Anal. Biochem., 1986, 159, 114. Szucs, A., Hitchens, G. D., and Bockris, J. O’M., J. Electro- chem. SOC., 1989,136, 3748. Swoboda, B. E. P., and Massey, V., J. Biol. Chem., 1965,240, 2209. Sun, S., Ho-Si, P.-H., and Harrison, D. J., Lungmuir, 1991,7, 727. Harrison, D. J., Turner, R. F. B., and Baltes, H. P., Anal. Chem., 1988,60, 2002. Cass, A. E. G., Francis, D. G., Hill, H. A. O., Aston, W. J., Higgins, 1. J., Plotkin, E. V., Scott, L. D. L., and Turner, A. P. F., Anal. Chem., 1984,56, 667. Degani, Y., and Heller, A., J. Phys. Chem., 1987, 91, 1285. Degani, Y., and Heller, A., J. Am. Chem. SOC., 1988, 110, 2616. Bartlett, P. N., Whitaker, R. G., Green, M. J . , and Frew, J. E., J. Chem. SOC., Chem. Commun., 1987, 1603. Bartlett, P. N., Bradford, V. Q., and Whitaker, R. G., Talantu, 1991, 38, 57. Bartlett, P. N., and Bradford, V. Q., J. Chem. SOC., Chem. Commun., 1990, 1135. Dixon, M., and Kleppe, K., Biochim. Biophys. Acta, 1965,96, 368. NOTE-Ref. 9 is to Part IV of this series. Paper 2/01 167K Received March 4, 1992 Accepted April 15, 1992

ISSN:0003-2654    

DOI:10.1039/AN9921701287

出版商:RSC    

年代:1992

数据来源: RSC

摘要:

ANALYST, AUGUST 1992, VOL. 117 1293 Stabilization of Analytical Enzymes Using a Novel Polymer-Carbohydrate System and the Production of a Stabilized, Single Reagent for Alcohol Analysis* Timothy D. Gibson Department of Biochemistry and Molecular Biology, University of Leeds, Leeds LS2 9JT, UK 1. John Higgins Cranfield Biotechnology Ltd., Newport Pagnall, Buckinghamshire MK16 9QS, UK John R. Woodward Yellow Springs Instruments Inc., Yellow Springs, OH 45387, USA A number of analytical enzymes including galactose oxidase, malate dehydrogenase and alcohol oxidase (from the methylotrophic yeast, Hansenula polymorpha) have been stabilized in a dry form by use of a novel, patented polymer+arbohydrate system. The enzymes were dried under vacuum at ambient temperature in the presence of a positively charged (cationic), soluble polymer such as diethylaminoethyl (DEAE)-dextran, and a carbohydrate sugar alcohol, lactitol.The dried enzymes retained high activity under conditions of thermal stress at 37 "C. Long term stability testing of alcohol oxidase indicated that 100% of the enzyme activity was retained for up to 2 months incubation at 37 "C in the presence of the stabilizers. In comparison, unstabilized enzyme, which was dried in phosphate buffer alone, retained only 26% activity after 7 d of incubation at 37°C. Stabilized alcohol oxidase preparations have been used to produce an alcohol assay reagent kit, having a shelf life of over 2 years when stored at 4 "C. Activity loss during the drying step was also reduced in the presence of the stabilizers. This type of stabilization system has application in the long term storage of active enzymes, enzyme products and in the area of enzyme-based analytical reagents and biosensors.Keywords: Stabilization; enzyme; dry storage stability; alcohol analysis Enzymes are used increasingly in a variety of analytical applications and offer the analytical chemist a powerful tool in the direct analysis of many compounds of interest. However, the major disadvantage of many enzymes is their lack of stability on storage, which limits their usefulness in any analytical method or biosensor device. There have been many reports of the experimental en- hancement of the stability of enzymes for use in analytical formats including entrapment in polymeric matrices or mem- branes,1-3 immobilization by covalent attachment to insoluble matrices,4-* covalent cross-linking of the protein structures9JO and addition of additives to the enzyme preparation.11,12 Dried enzymes tend to be more stable than those in solution; however, preserving biological activity during the drying step and the subsequent storage of the dry product usually requires the presence of some sort of stabilizer compound. Sugars, polyhydric alcohols, polymers and other proteins have all been used to stabilize enzymes.13-17 In an attempt to increase the stability of the enzyme alcohol oxidase, for use in the development of a diagnostic alcohol test, a large number of additives including sugars, sugar alcohols and polymers were tested to determine the effect on the storage stability of the enzyme.In the process of this work a stabilizing combination of a positively charged poly- electrolyte and a sugar alcohol was discovered. Further investigation into this unexpected result has indicated that this type of additive combination was able to stabilize the catalytic activity of a number of unrelated enzymes during storage at elevated temperatures, when compared with the results obtained from enzymes stored in the absence of the additives. This paper presents the data for eight enzymes that show increased thermal stability during storage in the presence of the stabilizer, and the use of the stabilizer combination to prepare a dry stabilized alcohol assay kit. * Presented at the meeting on Analytical Applications of Chemi- cally Modified Electrodes, Bristol, UK, January 7-8, 1992.Experimental Alcohol oxidase from Hansenula polymorpha was prepared 'in house' using the method of Gibson.** Galactose oxidase was a gift from Dr. P. Knowles, Leeds University. Horse- radish peroxidase (Type HRP-4) was obtained from Biozyme Laboratories. Choline oxidase, glycerol 3-phosphate oxidase, malate dehydrogenase, alkaline phosphatase and P-galactosi- dase were all obtained from Sigma. Lactitol was a gift from Cortecs and diethylaminoethyl (DEAE)-dextran was obtained from Pharmacia Diethanol- amine , sodium 3,5-dichloro-2-hydroxybenzenesulfonate, o-nitrophenyl-P-D-galactopyranoside, p-nitrophenyl phos- phate, DL-malic acid, 6-nicotinamide adenine dinucleotide (NAD) , glycerol 3-phosphate, choline chloride and D-galac- tose were obtained from Sigma.4-Aminoantipyrine was obtained from Aldrich. All other chemicals were obtained from BDH (Merck). Enzyme Drying and Stability Testing Fresh solutions of enzyme were prepared in suitable buffers and kept at 4 "C until used. Aliquots of enzyme solution were mixed with phosphate buffer and a combination of the stabilizer to give final concentrations of 5% m/v lactitol and 1% mfv DEAE-dextran in a volume of 1-2 cm3. Unstabilized enzyme preparations contained buffer and enzyme solution only. The composition of each enzyme drying system is shown in Table 1. Aliquots (100 mm3) were dispensed into the base of 4 cm3 disposable polystyrene cuvettes and dehydrated under vacuum (10 Pa) using a Gallenkamp vacuum oven set at a temperature of 30 "C.The chamber of the oven contained 500 g of fresh silica gel and the drying procedure was carried out for a total of 4 h. When drying was completed the dry enzyme preparations were stored in airtight plastic boxes containing silica gel. The preparation of enzyme reagent for alcohol test kits followed an identical procedure except that larger volumes of1294 ANALYST, AUGUST 1992, VOL. 117 Table 1 Enzyme drying compositions Enzyme Alcohol oxidase Choline oxidase Glycerol 3-phosphate oxidase Galactose oxidase Malate dehydrogenase Horseradish peroxidase P-Galactosidase Alkaline phosphatase Total activity/ Total protein/ cuvette cuvette 0.710 170.00 1 .Ooo 79.40 1 .Ooo 52.60 0.086 0.04 0.056 0.06 0.300 1.38 0.040 267.0 0.009 1.50 units per Crg Per Buffer Phosphate Tris-HC1 Phosphate Phosphate Phosphate Phosphate Phosphate Phosphate PH 7.0 8.0 7.0 7.0 8.0 7.0 7.0 7.0 Buffer concentration/ mmol dm-3 100 30 30 10 25 10 10 10 Table 2 Enzyme assay methods Enzyme Buffer* pH TemperaturePC Substrate Alcohol oxidase Phosphate 7.0 Choline oxidase Tris-HC1 8.0 Glycerol 3-phosphate oxidase Phosphate 7.0 Galactose oxidase Phosphate 7.0 Malate dehydrogenase Diethanolamine 9.2 Horseradish peroxidase Phosphate 7.0 P-Galactosidase Maleate 7.3 Alkaline phosphatase 2-Amino-2-methyl- 10.5 propan-1-01 * Buffer concentrations were standardized at 100 mmol dm-3.t Horseradish peroxidase. $4-Aminoantipyrine. 5 4-Hydroxybenzenesulfonic acid (Na salt). 7 3,5-Dichloro-2-hydroxybenzenesulfonic acid (Na salt). 25 37 37 25 37 30 37 30 Ethanol Choline chloride Glycerol 3-phosphate D-Galactose DL-Malic acid Hydrogen peroxide o-Nitrophen yl-D- galactopyranoside p-Nitrophen yl phosphate Detection method Coupled reaction with HRP,? 4-AAPS and PSA,$500 nm Coupled reaction with HRP, 4-AAP and DCHBSA,I 520 nm Coupled reaction with HRP, 4-AAP and DCHBSA, 520 nm Coupled reaction with HRP, 4-AAP and PSA, 500 nm NADH formation, 340 nm Dye formation, 4-AAP and PS A, 500 nm o-Nitrophenol release, 405 nm p-Nitrophenol release, 405 nm 140 I T i I 1 1 I 0 4 8 12 16 20 Incubation at 37 "Cld Fig.1 Stabilization of Hansenula polymorpha alcohol oxidase by a combination of DEAE-dextran and lactitol. In the absence of one or both stabilizers a much reduced stability is observed. Extended trials indicated 100% activity was retained for up to 2 months (60 d) incubation at 37 "C.Each point for the stabilized enzyme is an average of between 2 and 5 replicates. A, No stabilizers; B, DEAE-dextran; C, lactitol; and D, lactitol-DEAE-dextran. Error bars represent the scatter of individual results from each of the replicates enzyme and stabilizers were used and the drying procedure was carried out in 25 cm3 bottles rather than cuvettes. When drying was completed, 2.1 mg of 4-aminoantipyrine were added to the dry enzyme reagent. The stability of the dry enzyme preparations and the alcohol test kits was tested by stressing at elevated temperatures. In most instances a satisfactory rate of activity decay was observed at 37 "C for unstabilized preparations. P-Galactosidase was found to be fairly stable at this temperature; therefore, this enzyme was stressed at 50 "C.All of the incubations were carried out in the presence of silica gel as it was found that damp preparations of enzymes too 1 g 80 E" 8 20 1 0) .- .r 60 40 Y .- > .- Y 0 . -1 : 4 9 14 19 24 0 incubation at 37 "C/d Fig. 2 Stabilization of choline oxidase (Sigma C5896) by the DEAE-dextran and lactitol combination. The broken line indicates the loss of enzyme activity on drying. A, No stabilizers; and B , lactitol-DEAE-dextran were rapidly inactivated even with the stabilizer combinations present. Activity Determination The dry enzymes were reconstituted by adding buffer directly to the cuvettes in which they were dried. After standing for 2-3 min the cuvettes were gently inverted several times to ensure complete dissolution of the solids.Standard spectro- photometric assay techniques were always used to measure remaining activity using a thermostated mounting block in a Pye Unicam PU-8600 ultraviolethisible (UVNIS) spectro- photometer connected to a Philips PM8251 chart recorder. Measurement of NADH on the malate dehydrogenase reac-ANALYST, AUGUST 1992, VOL. 117 1295 tion was carried out using UV grade polystyrene cuvettes obtained from LIP Plastics. The details of all enzyme assays are shown in Table 2. Alcohol test kit assays were carried out by adding 25 cm3 of 100 mmol dm-3 phosphate buffer of pH 7.0 containing 25 mmol dm-3 4-hydroxybenzenesulfonic acid (Na salt) and gently inverting the bottle to dissolve the contents completely. To measure methanol 2.0 cm3 of reagent were used, whereas only 1.0 cm3 was needed for ethanol.The reagent was pre-incubated for 5 min at 37°C and 10 mm3 of either methanol or ethanol were added. The absorbance was measured at 500 nm after 20 min incubation at 37 "C, against a reagent blank. 120 I - _ I I I I I - 1 : 4 9 14 19 24 0 Incubation at 37 "C/d Fig. 3 Stabilization of glycerol 3-phosphate oxidase (Sigma G-9888) showing enhanced stability in the presence of the DEAE-dextran and lactitol combination. The broken line indicates enzyme activity after drying. A, No stabilizers; and B, lactitol-DEAE-dextran - = - B C A I 1 1 I -1 : 4 9 14 19 24 0 Incubation at 37 T / d Fig. 4 Stabilization of extracellular galactose oxidase from Dactiliurn dendruides (prepared 'in house') by a combination of DEAE-dextran and inositol.Lactitol could not be used in this instance as it was found to be a substrate of the enzyme. The broken line indicates enzyme activities after drying. A, No stabilizers; B, inositol-DEAE-dextran; and C, inositol .- 2 I \ I - . - 1 . 4 9 14 19 24 0 Incubation at 37 C/d Fig. 5 Stabilization of malate dehydrogenase. The malate dehy- drogenase (Sigma M9004) was partially stabilized with lactitol alone; however, inclusion of DEAE-dextran enhanced the stability conside- rably. The broken line indicates enzyme activity after drying. A, No stabilizers; B, lactitol; and C, lactitol-DEAE-dextran Results and Discussion The decrease in enzyme activity of thermally stressed enzyme preparations has been taken as an indication of the stability of the enzyme with respect to the length of time stored.The results are plotted in Figs. 1-8. For all except galactose oxidase inclusion of the combina- tion of the sugar alcohol, lactitol and the charged polymers (DEAE-dextran) enhanced the storage stability of the enzyme in the dry state when compared with the activity retention of 120, * 1 20 : A I I I I -1: 4 9 14 19 24 0 Incubation at 37 "C/d Fig. 6 Stabilization of horseradish eroxidase. The horseradish peroxidase (Sigma P-8250) was partial6 stabilized by lactitol alone; however, inclusion of DEAE-dextran enhanced the stability still further. This enzyme is often thought of as being very stable; however, the drying of dilute solutions of peroxidase indicates that this may not be so.The broken line indicates enzyme activities after drying. A, No stabilizers; B, lactitol; and C, lactitol-DEAE-dextran 170 100 - 0, C .- ,G 80 m /El 40 I I I 1 ' 9 14 19 24 29 0 incubation at 50 C/d Fig. 7 Stabilization of P-galactcsidase. The P-galactosidase (Sigma G-1875) was stabilized using DEAE-dextran and lactitol. The elevated temperature of 50 "C was necessary to emphasize enzyme degradation in unstabilized enzyme. Little activity was lost in unstabilized preparations incubated at 37 "C. The broken line indicates enzyme activities after drying. A, No stabilizers; and B, lactitol-DEAE-dextran i 2 0 1 . 1 6 0 ' . I 1 I 1 I 1 1 1 - 1 . 1 3 5 7 9 11 13 15 0 Incubation at 37 "C/d Fig. 8 Stabilization of alkaline phosphatase. The alkaline phos- phatase (Sigma P-7640) showed an increase in stability in the presence of DEAE-dextran and lactitol.The broken line indicates enzyme activities after drying. A, No stabilizers; and B, lactitol-DEAE- dextran1296 ANALYST, AUGUST 1992, VOL. 117 1.2 1 .o 0.8 0 0.6 0.4 0.2 0 200 400 600 800 Methanollmg dm-3 Fig. 9 Storage stability of alcohol test kits. The results shown were obtained by incubation of alcohol test kits for 83 d at 4 and 37 "C. Linearity was unaffected by the slight loss in absolute activity of the kit incubated at 37 "C. Duplicate samples were assayed in all instances. A, Freshly prepared kit; B, storage for 83 d at 37 "C; and C, storage for 83 d at 4°C 700 ?' 600 E II o, 500 E .- 400 . Y 8 8 I8 8 I 0 100 200 300 400 500 600 700 Ethanol (Sigma 332-UV dehydrogenase method)/mg dm 3 Fig.10 Regression analysis of the alcohol oxidase based test kit against an alcohol dehydrogenase based kit (Sigma 332-UV). The expression y = 1.017~ + 10.7 for n = 30 was obtained, r = 0.993 and the slope of the line of best fit was 0.994 Table 3 Reproducibility of alcohol test kit (values given are the absorbance values at 500 nm) AlcohoVmg dm-3 130.0 0.358 0.356 0.360 0.360 0.367 0.370 0.361 0.355 R 0.3609 u 0.0052 450.0 0.624 0.632 0.634 0.639 0.620 0.626 0.630 0.647 0.6315 0.0087 550.0 0.686 0.673 0.684 0.707 0.684 0.677 0.698 0.6880 0.0111 750.0 0.958 0.966 0.954 0.949 0.964 0.967 0.961 0.963 0.9600 0.0062 unstabilized preparations of enzyme. For galactose oxidase , lactitol was found to be a substrate of the enzyme and was replaced by inositol, which gave a similar result. The effect of drying alcohol oxidase in the presence of the sugar alcohol, lactitol, is shown in Fig.1. A partial stabiliza- tion is observed, which is also shown for malate dehydro- genase (Fig. 5), and horseradish peroxidase (Fig. 6 ) . Addition of DEAE-dextran to such mixtures had the unexpected effect of producing enhanced stabilization of the enzyme. The DEAE-dextran when used in the absence of lactitol has little stabilizing effect (Fig. 1). A total of six repeat experiments carried out using alcohol oxidase as the test enzyme gave a range of values for each test point. The scatter seen is indicated by the error bars shown in Fig. 1 for the lactitol and DEAE-dextran combination. It was thought that this scatter was due to slight inaccuracies in pipetting the rather viscous enzyme-stabilizer cocktail into the individual cuvettes.To test this theory, multiple samples were assayed using the stabilized alcohol reagent prepared in 25 cm3 bottles. The results presented in Table 3 show good reproduci- bility, indicating that the scatter seen in the figures was due to the individuality of the cuvettes produced rather than the assay method used. No large scale repeat experiments were carried out with any of the other enzymes reported. The results shown were obtained from a maximum of triplicate analyses which were averaged before being recorded. It was also noticed that in many instances a stimulation of enzyme activity occurs upon drying. This observation was not due to inaccuracies in the preparation of the cuvettes, which produced the scatter seen previously, as repeat assays gave the same over-all results. No satisfactory explanation for this Stabilization of the range of other enzymes reported using the same combination of reagents is unusual, as stabilizing conditions that promote activity enhancement for one enzyme on drying are often not as effective for others.For example the alcohol oxidase isolated from the yeast Pichia pasturis is reported to be stabilized by mannitol;13 however, this sugar alcohol does not stabilize alcohol oxidase from HansenuZa poZymurpha.18 In addition, testing of the stabilizer combina- tions with other enzymes indicated that activity loss during the drying step may be reduced. This is particularly noticeable in the results for choline oxidase (Fig.2) and malate dehydrogenase (Fig. 5). The mechanism of enzyme stabilization in the dry form is not completely understood even today. The effect of additives on the conformational structure of polypeptide chains and consequent stability of the protein molecule has been dis- cussed at length by various workers including, Back et aZ.,19 Ye et al. ,20 Monsan and Combes21 and Fujita et ~ 1 . 2 2 The over-all effect of including stabilizing combinations of additives such as sugars or sugar alcohols appears to be in their interaction with the water molecules surrounding the protein structure. This interaction has the effect of reducing the water activity by making the water more 'organized' and conse- quently intermolecular hydrophobic bonding of the poly- peptide chains of the protein molecule increases.Under these conditions, more energy is needed to disrupt the folded structure of the protein. This is seen experimentally as an increase in the thermal stability, which is measured in the enzyme systems described as the retention of activity under conditions of thermal stress. Electrostatic interactions between polyelectrolytes and enzymes have also been reported previously.23-25 The combi- nation of an electrostatic interaction between the enzyme and the polyelectrolyte, DEAE-dextran, in the presence of the sugar alcohol, lactitol, probably indicates a synergistic effect of the stabilizer on the enzyme structure. The electrostatic interaction of DEAE groups on the polymer surface with negatively charged groups on the protein surface is utilized in the purification procedure for alcohol oxidasel8 and in other enzymes.It is reasonable to assume that the soluble dextran molecule containing the same charged groups will interact with the protein and have some sort of structure stabilizing effect. Drying the electrostatically bound enzyme complex in the presence of lactitol has the effect of producing a highly stable preparation, whereas in the absence of lactitol little stabiliza- tion is observed. The DEAE-dextran in some way seems to enhance the effect of the sugar alcohol in stabilizing the enzyme. The relative effects of each of the components of the system are being investigated further. effect has yet been found. /ANALYST, AUGUST 1992, VOL.117 1297 The successful stabilization of alcohol oxidase from Han- senula polymorpha has an important application in the manufacture of a long lived test method for methanol and ethanol. The production of an alcohol test kit has been carried out using the stabilizer combination described. Results obtained for methanol analysis with a freshly prepared kit have been compared with those obtained from kits stored for 83 d at 4 and 37 "C; these are shown in Fig. 9. The small loss of stability seen at 37 "C might be attributable to the enzymes being incubated in the absence of desiccant. It has been estimated that even with this loss of absolute activity when compared with freshly prepared reagent, the shelf life of the kit should be in excess of 2 years if the kit is stored desiccated at 4 "C.Using the stabilized reagent kit for ethanol analysis pro- duced a linear relationship up to a concentration of 1000 mg dm-3 ethanol, with a sample volume of 10 mm3 and a reagent volume of 1.0 cm3. Regression analysis of the kit against a standard alcohol dehydrogenase method (Sigma 332-UV) gave the results shown in Fig. 10, where y = 1.017~ + 10.7 and the correlation coefficient, r = 0.993. Other applications of this technique to enzyme stabilization include the production of dry, stable enzyme-based biosensors and the production of dry chemistry formats (dip-stick devices) for analytical quantification of a variety of sub- stances. The results reported here form the basis of a patent application in the area of enzyme stabilization.26 References Guilbault, G.G., Analytical Uses of Immobilised Enzymes, Marcel Dekker, New York, 1984. Scheller. F., and Renneberg. R., Anal. Chim. Acta, 1983, 152, 265. Kirsteen, D., Scheller, F., Olsson, B., and Johansson, G., Anal. Chim. Acta, 1985, 171, 365. 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Gibson, T. D., and Woodward, J. R., Anal. Proc., 1986, 23, 360. Campbell, J., Hornby, W. E., and Morris, D. L., Biochim. Biophys. Acta, 1975, 384, 307. Bais, R., Poteznay, N., Edwards, J. B., Rofe, A. M., and Conyers, R. A. J., Anal. Chem., 1980, 52, 508. Gorton, L., and Hedland, A., Anal. Chim. Acta, 1988,213,91. Matsumoto, K., Hanada, O., Ukella, H., and Osajima, Y., Anal. Chem., 1986, 58, 2732. Taylor, P. J., Kmetec, E., and Johnson, J. M., Anal. Chem., 1975, 49, 789. Peterson, B. A., Anal. Chim. Acta., 1988, 209, 239. Kricka, L. T., and Carter, T. J. N., Clin. Chim. Acta, 1977,79, 141. Broughan, M. J., and Johnson, D. B., Enzyme Microb. Technol., 1981, 3, 225. Phillips, R. C., Eur. Pat. Appl., 0 133 481 A l , 1985. Tabb, D. L., and Tyhach, R. J., US Pat., 4 362 697, 1982. Roser, B. J., Int. Pat. Appl., Wo87/00196, 1986. Hopkins, T. R., US Pat., 4 729 956, 1988. Adams, E. C., US Pat., 4 786 596, 1988. Gibson, T. D., Ph.D. Thesis, 1991. Back, J. F., Oakenfull, D., and Smith, M. B., Biochemistry, 1979, 18, 5191. Ye, W. N., Combes, D., and Monsan, P., Enzyme Microb. Technol., 1988. 10, 498. Monsan, P., and Combes, D., Ann. N. Y. Acad. Sci., 1984,434, 48. Fujita, Y., Iwasa, Y., and Noda, Y., Bull. Chem. SOC. Jpn., 1982, 55, 1896. Katchalsky, A., Biophys. J., 1964, 4 (suppl.), 9. Elbein, A. D., Adv. Enzymol., 1976, 40, 29. Gianfrieda, L., Pirozzi, D., and Grecojr, G., Biorechnol. Bioeng., 1989,33, 1067. Gibson, T. D., and Woodward, J. R., PCT/GB91/00443 Enzyme Stabilisation, Publ. No. Wo91/14773, 1991. Paper 2/01 132 H Received March 3, 1992 Accepted April 28, 1992

ISSN:0003-2654    

DOI:10.1039/AN9921701293

出版商:RSC    

年代:1992

数据来源: RSC

摘要:

ANALYST, AUGUST 1992, VOL. 117 1299 Voltammetric and Amperometric Behaviour of Uric Acid at Bare and Surface-modified Screen-printed Electrodes: Studies Towards a Disposable Uric Acid Sensor* Markas A. T. Gilmartin and John P. Hartt Bristol Polytechnic, Faculty of Applied Sciences, Coldharbour Lane, Bristol BSI6 IQY, UK Brian Birch Unilever Research, Sensors Group, Colworth Laboratory, Colworth House, Sharnbrook, Bedford MK44 I l Q , UK Systematic voltammetric and amperometric studies have been undertaken to examine the electrochemical behaviour of uric acid at bare and surface-modified screen-printed electrodes. The precision of the electrode manufacture was determined by cyclic voltammetry with a 1 .O x 10-4 rnol dm-3 uric acid solution and was calculated to be 6.0% ( n = 5 ) .Several strategies were investigated in an attempt to eliminate interference from ascorbic acid. These involved coating the electrode surface with Nafion, or the enzyme L-ascorbic acid oxidase. The latter was immobilized using one of two methods: either by a simple adsorption process, or by cross-linking with bovine serum albumin and glutaraldehyde. The amperometric response at the surface- adsorbed enzyme electrode for uric acid was linear over the concentration range from 5.08 x 10-6 to 1.51 x 10-4 rnol dm-3; the limit of detection was 2.54 x 10-6 rnol dm-3 using a full-scale deflection of 0.5 pA. This modified electrode exhibited no response to ascorbic acid at levels up to 0.53 mmol dm-3. The electrode modified by cross-linking the enzyme to the surface showed no response to ascorbic acid concentrations of up to 0.093 mmol dm-3.Keywords: Cyclic and differential-pulse voltammetry; amperometry; uric acid; screen-printed electrode; 1-ascorbic acid oxidase Uric acid (2,6,8-trihydroxypurine) (1) is the primary end- product of purine metabolism. It is for this reason that its measurement remains the most important assessment of a range of disorders associated with altered purine metabolism, notably gout, hyperuricaemia and Lesch-Nyhan syndrome. 1 Additionally, as a reducing agent it ‘scavenges’ free oxygen radicals, preventing their destructive action towards tissues and cells. 0 I t . . - 1 Earlier chemical methods for the detection of uric acid exploit its reducing properties. These tests are susceptible to interferences from other reducing agents, are generally insensitive and also time-consuming. Enzymic procedures are potentially more selective, but inherently more expensive. They utilize uricase which catalyses the oxidation of uric acid to allantoin and the resulting change in absorbance is monitored at 293 nm.2 Other enzyme systems have been developed based on the amperometric measurement of the H202 produced from this reaction.3.4 However, procedures based on H202 are not always completely reliable, i.e., impurities in the enzymes (such as catalase) will degrade the peroxide resulting in a loss of the amperometric signal.Recently, we reported the development of screen-printed electrochemical sensors for the measurement of glutathione .5 These electrodes are inexpensive and simple to fabricate and therefore amenable to mass production; for these reasons they may be regarded as being disposable.As uric acid has been * Presented at the meeting on Analytical Applications of Chemi- -t To whom correspondence should be addressed. cally Modified Electrodes, Bristol, UK, January 7-8, 1992. shown to be electroactive at carbon-based electrodes,G9 it was considered that this screen-printing technology might be adapted to the development of a sensor for uric acid. This paper describes voltammetric and amperometric stud- ies towards the development of an electrochemical sensor based on the direct oxidation of uric acid at screen-printed electrodes (SPEs). Several strategies were attempted in order to improve the selectivity of the devices and these are also discussed.Experimental Chemicals and Reagents All chemicals were of analytical-reagent grade and obtained from BDH (now Merck) unless stated otherwise. Graphite (Ultra ‘F’ grade) was purchased from Ultra Carbon. Nafion (perfluorosulfonate ionomer), ascorbic acid and glutaral- dehyde were obtained from Aldrich. Uric acid, bovine serum albumin (BSA) and L-ascorbic acid oxidase (L-AAO) were obtained from Sigma. The enzyme was received in powdered form [1.6 mg, containing 250 U (1 U = 16.67 nkat) of catalytic activity] and was dissolved in 1 .O cm3 of 0.1 mol dm-3 sodium hydrogen carbonate buffer.9 Nafion (5% m/v) was diluted 1 + 19 with 90% ethanol prior to its use in surface modifications of SPEs. Cross-linking reagents consisted of BSA (100 mg cm-3 dissolved in phosphate buffer) and a stock glutaraldehyde solution (25% v/v diluted 1 + 4 with de-ionized water).Standard solutions were prepared daily and immediately wrapped in aluminium foil to prevent photo- and thermal biomolecule degradation. The supporting electrolyte used throughout was phosphate buffer, which was prepared from a 0.5 mol dm-3 stock solution of sodium dihydrogen ortho- phosphate and orthophosphoric acid. These were mixed to give a buffer of the required pH (a pH meter was used) and diluted with de-ionized water (if necessary) to yield the desired concentration. The dissolution of uric acid was effected in 50 cm3 of 0.05 rnol dm-3 sodium hydroxide. All solutions were prepared using water that had been de-ionized with a Purite R0200-Stillplus HP system.1300 ANALYST, AUGUST 1992, VOL.117 Apparatus Amperometry, cyclic and differential-pulse voltammetry were performed using a Metrohm E612 VA-scanner in conjunction with an E611 VA-detector; these were connected to a Linseis LY18100 x-y plotter to record the voltammograms. A three-electrode cell was employed incorporating an SPE, a saturated calomel reference electrode (Russel electrodes) and a platinum wire counter electrode. A small circular stirring disc (BDH) was used for stirring purposes during amper- ometry at constant potential. Electrode Construction and Modifications The SPEs were produced in a manner reported previously,5 and modified by drop-coating the electrode (bulb area = 0.071 cm2) with either 0.003 cm3 of a 0.25% Nafion solution or 0.020 cm3 of ascorbate oxidase giving a 7.5 pg Nafion coating or 5 U of catalytic activity, respectively.Alternatively, the enzyme was cross-linked to BSA via glutaraldehyde.10 In this instance, stock BSA was mixed on a vortex mixer with L-AAO and 0.050 cm3 of the resulting mixture was added to 0.025 cm3 of a 5% glutaraldehyde solution and mixed for 20 s. Next, 0.010 cm3 of the final solution was applied directly onto the electrode bulb and allowed to dry (10 min). Unbound enzyme and any impurities were removed by washing with a stream of de-ionized water for 30 s. Voltammetric Procedures Using SPEs A . Cyclic voltammetry Cyclic voltammograms were obtained for blank solutions of 0.25 rnol dm-3 phosphate buffer over the pH range 2-6 and then for similar solutions containing 0.1 mmol dm-3 uric acid, utilizing bare and Nafion-modified SPEs.Cyclic voltammo- grams were also recorded at pH 5.5 using SPEs coated with L-AAO by both surface adsorption and cross-linking, under the same conditions. Various concentrations of supporting electrolyte (0.01-0.5 mol dm-3) were used to study its effect on the anodic current for a 0.1 mmol dm-3 uric acid solution. Cyclic voltammograms were obtained in triplicate using three individual SPEs; the mean and relative standard deviation were calculated in each instance and these values were used to assign the upper and lower limits for the error bars shown in Fig. 4. Adsorption phenomena were investigated using the same uric acid standard in 0.25 rnol dm-3 phosphate buffer.Scan rates were varied from 20 to 250 mV s-1 and the resulting current was measured in the usual manner. The voltammetric conditions, unless stated otherwise, were as follows: initial potential, 0 V; scan rate, 20 m V s-1; and switching potential, 1.1 V. Optimization experiments were performed in triplicate, using a fresh electrode for each run. B. Differential-pulse voltammetry Differential-pulse voltammetric measurements were per- formed under the same solution conditions as described previously (section A above). In addition, voltammograms were recorded on mixtures of ascorbic and uric acids (0.5-2.5 mmol dm-3) at bare and coated SPEs. Voltammetric condi- tions were as follows: initial potential, 0 V; scan rate, 5 mV s-1; pulse repetition period, 0.4 s; pulse amplitude, 50 mV; and final potential, 1.2 V.C. Hydrodynamic voltammetry Hydrodynamic voltammograms were obtained for modified and unmodified SPEs by amperometry in stirred solutions of 0.25 rnol dm-3 phosphate buffer (25 cm3) and in similar buffer solutions containing 1.9 mmol dm-3 uric and ascorbic acids. Stepwise voltage increments were applied to the working electrodes; the resulting steady-state anodic current responses were measured for each plateau and plotted versus applied potential. Each experiment was performed in triplicate with fresh SPEs. D. Amperometry Constant-potential amperometry was carried out in stirred solutions of 0.25 rnol dm-3 phosphate buffer (pH 5.5); the magnitude of the anodic current responses following additions of small volumes of stock uric acid solution was recorded over the final concentration range from 5.08 x 10-6 to 1.51 x 10-4 rnol dm-3.In each instance, uric acid was added to 25 (33113 of plain phosphate buffer in the voltammetric cell. The effect of ascorbic acid on the initial amperometric response of surface- adsorbed and cross-linked SPEs was investigated for a 1.0 mmol dm-3 uric acid solution at an applied potential of 0.4 V. After attaining a steady-state current for uric acid, increasing volumes (up to 0.1 cm3) of ascorbic acid solution (10 mmol dm-3) were introduced into the voltammetric cell. The endogenous levels of uric acid in calf serum were determined amperometrically by the method of standard additions, using the two types of enzyme-modified SPE. A baseline was established with the optimized phosphate buffer; the cell was then exchanged for a cell containing a buffered serum sample (15 cm3 of pH 5.5 phosphate buffer and 10 cm3 of calf serum), and 0.1 cm3 aliquots of a 10 mmol dm-3 uric acid stock solution were added to the cell.The resulting steady-state current was used to calculate the uric acid content of the serum. Results and Discussion Cyclic and Differential-pulse Voltammetric Behaviour of Uric and Ascorbic Acids at Bare and Nafion-modified SPEs There have been no previous systematic studies on the electrochemical behaviour of uric acid at the SPEs developed by our group. In order to ascertain whether these SPEs could form the basis of a sensor, involving the direct oxidation of uric acid, cyclic and differential-pulse voltammetric studies were performed under a variety of conditions.The present investigation was begun with an uncoated (bare) SPE. The cyclic voltammetric behaviour of uric acid was studied with this electrode at a scan rate of 20 mV s-1 and one well defined anodic signal (Fig. 1) was observed over the pH range investigated. The peak potential ( E p ) for uric acid was found to be dependent on the pH of the solution; the equation relating E, and pH, over the pH range 2-6, was found to be: E, = +0.850 - 0.07 pH (V) Under these conditions, the cyclic voltammograms did not exhibit any cathodic peaks; therefore, the oxidation process was considered to be irreversible. I 10.25 I 0.4 0.6 0.8 1.0 Pote n tia IN Fig. 1 Cyclic voltammetry of 0.22 mmol dm-3 uric acid in 0.25 mol dm-3 phosphate buffer (pH 5.0) at a bare SPE.Initial potential, 0 V. Scan rate, 20 mV s-1ANALYST, AUGUST 1992, VOL. 117 1301 Further evidence for the irreversible nature of the electrode reaction was obtained by carrying out cyclic voltammetry at increasing scan rates. It was found that the E, values shifted from +0.42 V at 20 mV s-1 to i-0.78 V at 250 mV s-l. In order to ascertain whether uric acid was undergoing adsorption at the surface of the SPEs, a plot of current function (ip/cvh) versus vk (where i, is peak current in FA, c is the concentration in mol dm-3 and Y is the scan rate in mV s-1) was constructed. Fig. 2(a) shows that current function was independent of d ; as an increase is to be expected for adsorption processes it would appear that the peak current is diffusion controlled.The precision of electrode manufacture was determined by cyclic voltammetry with a 1 .O x 10-4 mol dm-3 solution of uric acid in phosphate buffer (pH 5.5); five individual unmodified SPEs were studied and the relative standard deviation was calculated to be 6.0%. These preliminary results indicated that our SPEs had potential as sensors for uric acid in physiological fluids. However, it is known that ascorbic acid is present in these matrices and, as it is electroactive, is a possible interferent. Therefore, mixtures containing uric and ascorbic acids were studied using cyclic and differential-pulse voltammetry. Initial cyclic voltammetric studies at bare SPEs showed that the oxidation peak for ascorbic acid occurred at less positive potentials than the anodic peak for uric acid, but the two voltammetric signals were poorly resolved.Similar results were obtained using differential-pulse voltammetry. Previous studies by other workers using different electrode materials and design have indicated that Nafion may be employed to prevent interference from ascorbic acid.11.12 Therefore, it was also decided to investigate this possibility with our SPEs; the effect of pH was studied over the range 2-6. Uric acid was found to give one well-defined voltammet- ric signal at bare SPEs (Fig. 1) but the signals were smaller and more drawn out at the Nafion-coated electrodes; values for the electron transfer coefficient (an,) were calculated ,13 where possible, for uric acid at both types of electrode and are given in Table 1. Clearly, the electron transfer process becomes much slower at the coated SPEs. This is also indicated by the decrease in the i,lcvk versus Y$ plot [Fig.2(b)]. However, uric acid still gives a measurable signal over the pH range studied. Ascorbic acid was also studied by cyclic voltammetry under 2500 1 1 si 0 4 6 8 10 12 14 16 v:/mV: s-: Fig. 2 Graph of idcvl versus v4 (see text for explanation of the symbols) for a 0.1 mmol dm-3 uric acid solution at A , a bare SPE; and B. a Nafion-coated SPE Table 1 an, values for a 0.1 mmol dm-3 uric acid solution at bare and Nafion-modified SPEs PH Bare SPE Nafion-modified SPE 2 0.60 0.13 4 0.59 0.16 6 0.48 0.17 the same conditions as used for uric acid at both bare and coated SPEs. At the pH values studied, the vitamin gave a smaller signal at the coated SPE than at the bare electrode; Table 2 summarizes the peak current data obtained for these two biomolecules.As can be seen from these results, the most efficient permselectivity for the measurement of uric acid over ascorbic acid occurs at pH 2. Clearly, at this pH both biomolecules are in their undissociated forms (Table 3). Wang et ~ 1 . 1 2 have reported that Nafion films exist in the negatively charged form at pH 2 and hence tend to exclude both neutral and anionic species. Our results indicate that the neutral form of ascorbic acid is excluded to a higher degree than the mono-anion. As our SPEs contain cellulose acetate this may be influencing the permselective properties of the film towards the vitamin.These results indicate that even at pH 2 only 71% of ascorbic acid is prevented from reaching the electrode surface and hence this was still regarded as being unsatisfactory. In order to obtain further information on the permselectiv- ity of Nafion-coated SPEs, differential-pulse voltammetric studies were performed on mixtures containing ascorbic and uric acids; the pH range investigated was 2-6. Interestingly, at the Nafion-modified electrode the ascorbic acid peak was now the smallest at pH 4 and appeared at a more positive potential than the uric acid peak. Unfortunately, this peak was still not resolved from the uric acid peak (Fig. 3). As the permselective properties of our Nafion-modified electrodes were insufficient to permit the voltammetric determination of uric acid in the presence of ascorbic acid it was considered that an alternative approach was necessary.Voltammetric and Amperometric Behaviour of Uric and Ascorbic Acids at Enzyme-modified SPEs In order to prevent ascorbic acid from reaching the SPE surface it was decided to explore the possibility of destroying Table 2 Peak current ratios for uric and ascorbic acids at bare (ib) and Nafion-coated (ic) SPEs Uric acid, 1, : l b 2 0.29 0.75 4 0.73 0.45 6 0.36 0.71 . . Ascorbic acid, . . PH 1, : lb Table 3 Literature pK, values for ascorbic and uric acidslj Biomolecule Ascorbic acid Uric acid PKa, PK,, 4.17 11.57 5.70 0.80 acid I I I I I 0 0.2 0.4 0.6 0.8 1.0 1.2 PotentialN versus saturated calomel reference electrode Fig. 3 Differential-pulse voltammogram of 0.5 mmol dm-3 and 2.5 mmol dm-3 ascorbic and uric acids, respectively, in 0.25 mol dm-3 phosphate buffer (pH 4.0) at a Nafion-modified SPE.Scan rate, 5 mV s-11302 ANALYST, AUGUST 1992, VOL. 117 the vitamin with an immobilized enzyme; initial studies were performed with L-AAO simply adsorbed onto the SPE. The optimum pH for immobilized L-AAO simply adsorbed onto the SPE. The optimum pH for immobilized L-ascorbic acid oxidase (L-AAO) activity has been reported to be 5.5;15 therefore, the cyclic voltammetric behaviour of the enzyme- modified SPEs was studied at this pH. The effect of ionic strength of phosphate buffer on the peak current (ip) of uric acid was investigated by recording cyclic voltammograms in triplicate utilizing three individual SPEs.Clearly, the largest response was obtained in 0.25 rnol dm-3 phosphate buffer (Fig. 4). Hydrodynamic voltammetry was carried out with the same SPE, under these latter conditions; the resulting voltammo- gram showed that a suitable signal was obtained at +0.4 V (see later and Fig. 7). Constant-potential amperometry was carried out at +0.4 V in a stirred solution, with a 1.0 mmol dm-3 uric acid standard in 0.25 mol dm-3 phosphate buffer (pH 5.5) with the L-AAO-SPE. Multiple standard additions (up to 0.1 cm3) of a 10 mmol dm-3 ascorbic acid solution were introduced into the voltammetric cell and the effect of this compound on the uric acid response was monitored. Fig. 5 shows that no response was obtained for ascorbic acid up to a concentration of 0.53 mmol dm-3; this level of elimination was considered sufficient to allow the subsequent determination of uric acid in biological fluids.Fig. 6 shows typical amperograms resulting from adding increasing concentrations of uric acid to a 25 cm3 aliquot of 0.25 mol dm-3 phosphate buffer (pH 5.5). The current was found to be linearly related to concentration over the range from 5.08 X 10-6 to 1.51 X rnol dm-3; the limit of detection was 2.54 x 10-6 rnol dm-3 using a full-scale deflection of 0.5 PA. 7.0 f g % 6.0 3 m P 0 ," 5.0 5 4.0 e a 3.0 0 0.1 0.2 0.3 0.4 0.5 Ionic strength of phosphate buffer/mol dm-3 Fig. 4 Variation in anodic current with increasing ionic strength of supporting electrolyte using L-AAO coated SPEs. Error bars were established by obtaining triplicate cyclic voltammograms using three separate SPEs.Scan rate, 20 mV s-1 I 1 11.0 ~ I A 100 s w t bmq Standard additions I SaAple Time - Fig. 5 Effect of ascorbic acid additions (up to 0.5 mmol dm-3) on the am erometric response for a 1 .O mmol dm-3 stirred uric acid solution witg L-AAO-coated SPEs. Eapp, + 0.4 V As discussed above, the response of the SPE formed by surface adsorption of L-AAO displayed suitable sensor characteristics (Figs. 5 and 6). However, in human urine and calf serum samples the endogenous uric acid levels were greater than expected and also showed great variation. At this stage it was considered that the enzyme layer was either desorbed or inactivated by the sample. Therefore, it was decided to investigate an alternative enzyme immobilization procedure.Preliminary studies were carried out using a method which involved cross-linking the enzyme to BSA with glutaral- dehyde. Fig. 7 shows the hydrodynamic voltammograms for uric and ascorbic acids at bare and enzyme-coated electrodes. It is evident that uric acid gave a measurable response at the latter SPE. At the high concentration of ascorbic acid studied (1.9 mmol dm-3) this substance also gives a response at the modified electrode; however, the signal is approximately 50% less than that for uric acid at an applied potential of +0.4 V. Therefore, the response of various concentrations of ascorbic acid was studied at an applied potential of +0.4 V. Therefore, the response of various concentrations of ascorbic acid was studied at an applied potential of +0.4 V.It was discovered that this enzyme-coated SPE excluded up to 9.3 x 10-5 rnol dm-3 of ascorbic acid; this should be sufficient to eliminate the levels of this compound present in the calf serum samples used in the present study. The determination of uric acid in the serum was carried out, by amperometry, using the method of standard additions. In this instance the level was found to be about 50% higher than the manufacturer's specifications (2.7 mg-Yo). From these results, and those obtained with the surface adsorbed L-AAO electrode, it would appear that some species other than ascorbic acid is causing an interference in the determination of t w S 0 Standard additions -k-L - A C T 0.1 yA 100 s I - Time - Fig. 6 Amperometric current response to standard additions of uric acid to 25 cm3 of 0.25 rnol dm-3 phosphate buffer (pH 5.5) yielding final concentrations of: A, 5.08 X 10-6; B, 1.68 X 10-5; and C, 3.36 x 10-5 rnol dm-3 using L-AAO modified SPEs 25 .-a E 15 E - S ; 10 $ 5 .- U 0 0.1 0.2 0.3 0.4 0.5 Applied potential (€,,,)A/ Fig.7 Hydrodynamic voltammetry of stirred 1.9 mmol dm-3 uric and ascorbic acids, at bare electrodes and electrodes coated with L-AAO by cross-linking to BSA via glutaraldehyde. A and B, ascorbic acid at coated and bare SPEs, respectively; C and D, uric acid at coated and bare electrodes, respectivelyANALYST, AUGUST 1992. VOL. 117 1303 uric acid. This conclusion seems likely, as the method involving cross-linking the enzyme to the electrode forms the basis of many biosensors;16 hence it is now considered unlikely that the L-AAO is deactivated or desorbed from the SPEs in the samples studied here. It can be postulated, therefore, that the L-AAO modified SPEs may be used as disposable amperometric sensors for the measurement of uric acid in the presence of certain concentra- tions of ascorbic acid (up to 0.53 mmol dm-3 for the surface adsorbed enzyme electrodes and 9.3 x 10-5 mol dm-3 for the cross-linked enzyme electrodes).In order to improve further the selectivity of the surface modified SPEs, so that reliable uric acid measurements in serum or urine may be made, several strategies are available. It may be possible to employ a simple ‘clean-up’ procedure involving a liquid-solid separation. Alternatively, it is feasible that a mediator may be incorporated in the base carbon electrode to reduce the overpotential for uric acid oxidation.This can lead to great improvements in selectivity and many reports on this approach have appeared.16.17 These approaches will form the basis of further studies on uric acid. The authors thank Bristol Polytechnic for financial support. 1 2 References Harper, H. A., Review of Physiological Chemistry, Lange Medical Publications, CA, 16th edn., 1977, p. 406. Dilena, B. A., Peake, M. J., Pardue, H. L., and Skorg, J. W., Clin. Chem. (Winston-Salem, N . C . ) , 1986, 32, 486. 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Tatsuma, T., and Watanabe, T., Anal. Chim. Acta, 1991, 242, 85. Keedy, F. H., and Vadgama, P., Biosens. Bioelectron., 1991,6, 491. Wring, S. A., Hart, J. P., and Birch, B. J., Analyst, 1991, 116, 123. Wang, J . , and Freiha, B. A., Bioelectrochem. Bioenerg., 1984, 12, 225. Gonzalez, E., Anal. Chim. Acta, 1991, 242, 267. Yao, T., Taniguchi, Y., Wasa, T., and Musha, S., Bull. Chem. SOC. Jpn., 1978, 51, 2937. Hart, J. P., Electroanalysis of Biologically Important Com- pounds, Ellis Horwood, Chichester, 1990, p. 51. Woodward, J., in Immobilised Cells and Enzymes, ed. Wood- ward, J., IRL Press, Oxford, 1985, p. 31. Nagy, G., Gerdhart, G. A., Oke, A. F., Rice, M. E., Adams, R. N.. Moore, R. B., and Szentirmary, M. N., J . Electroanal. Chem., 1985, 188, 85. Wang, J. Tuzhi, P., and Golden, T., Anal. Chim. Acta, 1987, 194, 129. Galus, Z., Fundamentals of Electrochemical Analysis, Ellis Horwood, Chichester, 1976, p. 237. West, E. S., Todd, W. R., Mason, H. S., and van Brugger, J. T.. Textbook of Biochemistry, Macmillan, New York, 4th edn., 1966. Greenway, G. M., and Ongomo, P., Analyst, 1990, 115, 1297. Hall, E. A. H., in Biosensors, Open University Press, Milton Keynes, 1990. Davies, G., in Biosensors, Fundamentals and Applications, eds. Turner, A. P. F., Karube, I., and Wilson, G., Oxford Univer- sity Press, Milton Keynes, 1990, p. 247. Paper 2/00893I Received February 20, I992 Accepted April 3, 1992

ISSN:0003-2654    

DOI:10.1039/AN9921701299

出版商:RSC    

年代:1992

数据来源: RSC

摘要:

ANALYST, AUGUST 1992, VOL. 117 1305 Characterization of Poly(pyrro1es) by Cyclic Voltammetry* D. J. Walton, C. E. Hall and A. Chyla Department of Applied Physical Sciences, Coventry Polytechnic, Priory Street, Coventry CVI 5FB, UK Redox voltammetry provides a means to distinguish between poly(pyrro1es) that appear similar in many other respects. Under a consistent set of experimental conditions, poly(pyrro1es) doped with various small mobile anions (exemplified by tetrafluoroborate, BF4-) gave broad reduction waves at facile potentials, whereas poly(pyrro1es) doped with several larger and less-mobile anions (exemplified by p-toluenesulfonate, PTS) gave strikingly sharp reduction waves at markedly more cathodic potentials. This distinction appears intrinsic to the different materials and alteration of polymer preparation conditions failed to convert BF4--type behaviour to PTS-type behaviour.In contrast, the reverse switch was achieved, but only under the extreme preparation conditions of low monomer concentration which yielded poor quality film morphology. However, benzenesulfonate-doped poly(pyrro1e) was more readily switched between the two types of redox behaviour depending on the polymerization solvent, this result offering opportunity for manipulation of redox properties, with implications for redox-based applications of these materials. Keywords: Conducting polymer; poly(pyrro1e); voltammetry; redox behaviour The redox behaviour of conducting polymers is relevant to a range of potential applications, including battery electrodes, electrochromic displays, microelectrochemical devices, ion gates and sensors of various types.This is because a change in polymer charge level alters not only the conductivity but also the morphology, porosity, ion mobility, wettability, bulk and surface properties and reactivity.1 Poly(pyrro1e) and related polymers of the ‘heteroaromatic’ class are produced by chemical or electrochemical oxidation directly in a conductive polycationic form, charge-compen- sated by counter anions from the oxidation medium. The polymer grows with incorporation of these anions (termed ‘dopants’), the nature of which is known to influence the physical, mechanical and electronic properties of the resulting polymer. 1v2 Poly( pyrrole) can conveniently be prepared by electro- oxidation as a thin-film coating on the anode.It is notably air-stable in its conducting oxidized form, but is amenable to reduction. Here, the polycationic chain tends towards neutral- ity and in principle a complementary ion movement is necessary to redress charge balance. A reduction wave is observed in the cyclic voltammogram, and scan reversal produces a re-oxidation wave on the return. The charge involved in the reduction process has been related to the doping level of the polymer.3 Neither wave has the classical shape of a diffusion-con- trolled process, representing instead the behaviour of an electroactive electrode coating that alters its conductivity, morphology and composition during the scan. The reduction wave is often reported to have a broad multiple outline, suggesting a number of overlapping processes thought to reflect polymer chain relaxations accompanying ion motion.4 However, the observed redox voltammetry depends on many factors, including not only polymer preparation and history but also solvent, electrolyte and other redox experimental conditions.A number of papers have discussed variously the significance in different situations of anion motion, cation motion in the opposite sense, and the relative importance of faradaic and capacitative contributions towards the current.5-7 This complexity in the literature is due to well known problems of irreproducibility in conducting polymer systems, such that ostensibly similar materials often show subtle distinctions. This reflects the number of parameters available * Presented at the meeting on Analytical Applications of Chemi- cally Modified Electrodes, Bristol, UK, January 7-8, 1992.for manipulation, deliberately or inadvertently. For example, even using a fixed pyrrole, solvent and electrolyte salt system for preparation it is possible to vary relative concentrations, the electrode material, cell configuration, or to use pulsed techniques, potentiodynamic cycling, or galvanostatic methods. Different materials ensue and characterization is difficult. Simple poly(pyrroles), i.e., those that are not functionalized or derivatized,2 are typically insoluble, infusible and intract- able, mitigating against conventional methods of polymer analysis, while interactions between the conductivity pathway and electromagnetic radiation interfere with spectroscopic analyses.Simple poly(pyrro1es) are usually matt black, and visually similar even when different preparation procedures and histories suggest otherwise. In this paper the redox properties of poly(pyrro1e) are examined as a function of the original dopant anion in the polymer as prepared. For consistency and reproducibility, standard conditions were employed whenever possible, and in particular all redox voltammetry was performed in the same solvent (acetonitrile) and electrolyte (0.1 mol dm-3 tetra- butylammonium tetrafluoroborate) , with the same electrodes and cell configuration. The dopant anions examined fall into two general cat- egories: (i) small univalent inorganic anions, viz. , tetrafluoro- borate (BF,-), perchlorate (C104-) and hexafluorophosphate (PF,-); and (ii) a range of organic sulfonates of different size, charge and complexity, viz., benzenesulfonate (BS) , p-tolu- enesulfonate (PTS) , ferrocenedisulfonate (DSFc) and poly- (styrenesulfonate) (PSS).The effect of varying the preparation parameters on the subsequent polymer redox behaviour was also examined for BF4-- and PTS-doped films. Here, it was endeavoured to alter a single parameter at a time, other factors remaining constant. Hence the electrolyte salt for the polymerization, its concen- tration, the electrode potential and the solvent were individu- ally varied, while cell geometry, electrode material, washing and post-treatment of the films, ageing and other time factors were kept constant. All experiments were performed at room temperature.Experimental Materials Pyrrole (Aldrich) was freshly distilled prior to use. Solvents were all of AnalaR or spectroscopic grade, and were redistilled as necessary. Tetrabutylammonium tetrafluorobor-1306 ANALYST, AUGUST 1992, VOL. 117 ate (Bu4NBF4) (Aldrich) and tetrabutylammonium perchlor- ate (Fluka) were recrystallized appropriately. Lithium tetra- fluoroborate (Aldrich), tetraethylammonium p-toluenesul- fonate (Aldrich) and poly(styrenesulfonate), sodium salt (Aldrich; relative molecular mass 70 OW), were used as received. The ditetrabutylammonium salt of ferrocene-1,l- disulfonic acid was prepared by sulfonation of ferroceneg and conversion into the required salt .9 Benzenesulfonic acid was used as its sodium salt (Hopkin and Williams) in aqueous media, whereas for non-aqueous media the tetrabutylam- monium salt was prepared by treatment of the sodium salt in the minimum of water with the stoichiometric amount of tetrabutylammonium hydrogen sulfate followed by extraction of the aqueous solution with dichloromethane, in which the desired salt is soluble, leading, after evaporation, to white crystals.9 Electropolymerization Procedure The aim of this work was to observe the influence of the preparation parameters on the subsequent polymer redox behaviour.Preparative variables are given appropriately in the results sections, but the following account is representative of the general procedure: 0.1 mol dm-3 pyrrole in a de-gassed solvent containing usually 0.1 mol dm-3 electrolyte salt was electro-oxidized at a fixed potential (usually +1.25 V) using a platinum wire electrode of 0.3 cm2 area, until 50 mC of charge had passed.This gave electrode coating films of calculated thickness between 0.5 and 1.5 mm.10 All films were prepared to a constant passage of charge. For comparison some films were prepared at a constant current of 1 mA cm-2. Aceto- nitrile (MeCN), propylene carbonate (PC) and water andor methanol were employed as solvents. Potentiostatic or gal- vanostatic control was provided by an EG & G Model 273 potentiostat/galvanostat, and a platinum foil counter elec- trode and a standard calomel reference electrode (SCE) were used throughout for all electrochemical measurements. The electrodes, visibly coated with a matt black film of poly(pyrrole), were removed from the polymerization medium and washed several times with the appropriate solvent(s) before being placed in the cyclic voltammetric cell.All films were examined within 1-2 h of being prepared and results are reproducible in this respect. Cyclic Voltammetry All redox scans were performed in a consistent manner, with the same cell and electrode configuration, and a fixed electrolyte medium, viz., 0.1 mol dm-3 Bu4NBF4 in MeCN, de-gassed with oxygen-free nitrogen for 30 min prior to the experiment. All scans were performed at a speed of 100 mV s-1, and were started from a moderately positive potential in an initially negative direction, using an EG & G Model 362 scanning potentiostat and a Rikadenki x-y recorder. A fresh polymer film was prepared for each scan.Results General Redox Behaviour Fig. 1 shows redox voltammograms for poly(pyrro1es) doped with six of the anions studied, C104- (a), BF4- (b), PF6- ( c ) , PTS (d), DSFc ( e ) and PSS (f). Benzenesulfonate is treated later. Exact preparation details for all materials listed in Fig. 1 are given in Table 1. Two types of redox voltammograms are evident, and there is a clear distinction between the two classes of anion. Hence poly(pyrro1es) doped with small inorganic species all show a very broad reduction wave at about +200 mV with a re-oxidation on the return scan at +300 mV. This characteristic shape is attributed to expulsion of the dopant anion during reduction, and is well reported in poly(pyrro1e) voltammetry.4 In contrast, the redox voltammograms of poly(pyrro1es) doped with large sulfonate species are very different in shape and redox potential.Hence for PTS [Fig. l(d)] a strikingly sharp reduction wave occurs at -500 mV with the associated re-oxidation at - 150 mV. This change in shape is a prominent feature, but most significant is the shift in redox potential, the whole redox process now occurring in the region some 500 mV more cathodic. It is noticeable that there is hardly any reduction current at about +200 mV, where there is substan- tial current in the curves shown in Fig. l(a)-(c). (The difference between the two types of behaviour is emphasized by superposition in Fig. 6, curves A and B, to assist a later comparison). A similar redox behaviour is observed for the more complex and larger DSFc anion [Fig.l(e)]. This curve is from a sample prepared from PC solution using a constant-current method, to show the wide range of preparation conditions available. A very similar curve is obtained from DSFc-poly(pyrrole) prepared potentiostatically in MeCN. The significance of redox potentials is shown by PSS [Fig. 1 0 1 . The broadness of the waves, the slight positive shift of the re-oxidation wave and the large envelope at the positive end of the cycle are characteristic of films produced from hydroxylic solvents [cf. Fig. 5 for equivalent PTS-poly- (pyrrole) behaviour]. Here, the PSS film was prepared from the sodium salt of the polyanion (relative molecular mass 70 OOO) in aqueous methanol. Similarity between PTS-poly- (pyrrole) and PSS-poly(pyrro1e) is evident in the curves shown in Fig.l(d) and 0, in spite of the size, mobility and charge difference between the monoanion and polyanion, and even though it has been shown that polyanions substantially reduce the conductivity of bulk free-standing films, apparently by localization of charge carriers on the chain." In this paper we show that visually similar films of poly(pyrro1e) can display two types of redox behaviour. For small inorganic anions, a de-intercalation-re-intercalation mechanism has been proposed, in which expulsion of the original counter anion occurs on the reduction scan.4 In the present work the electrolyte anion used throughout, BF4-, is the smallest in the series, which can readily re-intercalate to replace expelled C104- or PF6-.This anion-exchange process is supported by loss of the chlorine signal from energy dispersive X-ray analysis (EDAX) data for a C104--poly- (pyrrole) cycled in BF4- electrolyte. The ready mobility of small anions within the poly(pyrro1e) matrix is suggested by the very mild electrode potential required for reduction. However, such a facile de-intercalation process cannot oper- ate for PTS and the even larger and less mobile DSFc and PSS anions, on the timescale of cyclic voltammetry at 100 mV s-1 (although these constraints may not apply to PTS and DSFc exposed to prolonged reduction conditions). In addition, for DSFc another redox wave is evident in Fig. l(e) at 860 mV. This is due to the iron(rr)-iron(w) couple in the ferrocene, which is anodically shifted owing to interaction with the polycationic poly(pyrrole) chain.12 This second redox wave remains, and iron and sulfur are still visible by EDAX, confirming the continued presence of the counter anion even after the earlier polymer reduction sweep.There is also a weak but reproducible current in Fig. l(e) at 500 mV. This is close to the iron redox value at 490 mV for the free DSFc anion in solution in MeCN-Bu4NBF4,12 and suggests that some DSFc which is not involved in charge compensation with the chain is occluded into the poly(pyrro1e) matrix. This is further supported by the observation from Mossbauer spectra of species in the polymer with more than one iron atom in their environment, and also by a discrepancy between the dopant concentration calculated from polymerization and redox charge ratios versus the total dopant concentration obtained by microanalysis.12 Such an occluded salt has been proposed to take part in ionic movement during polymer redox.6 However, in the present work the large ferrocene species does not appear sufficiently mobile to be so involved.ANALYST, AUGUST 1992, VOL.117 I I I 1307 1 I 500 rnA (dl I I I I I l l I I -1.0 0 +0.8 -0.9 0 +1.1 50 CIA I I I -1.0 0 +0.8 I I I -1.0 0 +0.5 PotentiaiIV Fig. 1 Redox voltammograms of poly(pyrro1es) containing various dopant anions. Scans were performed in MeCN-O.l mol dm-3 Bu4NBF4 at a scan rate of 100 mV s-1 and were begun at the anodic extreme in an initially negative direction. (a) C104-; (b) BF4-; (c) PF6-; ( d ) PTS; (e) DSFc; and cf) PSS Discrimination between the two types of redox behaviour shown in Fig.l(a)-(c) and Fig. l(d)-(n may be explained qualitatively by the above anion-size argument. However, a feasible alternative, particularly for the curves in Fig. l(d)-(f) is to invoke cation motion in the opposite sense. The respective roles in poly(pyrro1e) redox of anion motion, cation motion and/or associated solvent transport have yet to be established unequivocally, in spite of recent studies using quartz crystal microbalance data in addition to redox voltam- metry.13.14 Throughout the present work a bulky cation, tetrabutylammonium (Bu4N), was used in the electrolyte to mitigate against facile cation inclusion. This cation is of appreciable size compared with PTS and inward migration at a rate indicated by the sharpness of the reduction wave in Fig.l ( d ) is unlikely under conditions where PTS cannot be expelled. However, in order to establish whether cation size has an influence on polymer redox behaviour under the conditions used here, this work is being extended to other cations. Preliminary results for PTS-poly(pyrro1e) show that proton addition to the normal redox electrolyte produces broadening of the voltammetric waves but without appre- ciably altering the characteristic peak potentials, whereas use of lithium tetrafluoroborate in MeCN produces a reduction voltammetric wave very similar in shape and potential to that shown in Fig. l ( d ) , showing here the insensitivity of the reduction process to the redox cation.15 Influence of Preparation Procedure Fig.1 shows that redox voltammetry provides a means for distinguishing between poly(pyrroles), but given the disparity in literature reports of redox behaviour it is necessary to establish if preparation procedures are the source of irrepro- ducibility. Accordingly, representative examples of both redox types, namely BF4--poly(pyrrole) and PTS-poly- (pyrrole), were prepared with alteration of one parameter at a time, other factors remaining constant where feasible. Hence polymerization potential, electrolyte concentration, pyrrole concentration and preparation solvent were varied, within limits. The following results were obtained. Tetrafluoro borate-doped poly (p yrrole) The broadness of the reduction and the shape of the redox envelope in Fig.l(b) mitigates against close comparisons and it was found that a polymerization potential between +0.8 and +2.0 V, an electrolyte concentration between 0.005 and 0.5 mol dm-3, a pyrrole concentration between 0.015 and 0.375 mol dm-3 and use of MeCN or aqueous preparative solvent media, while producing slight changes in the redox voltam- metry, all failed to bring about a switch to PTS-poly(pyrro1e) behaviour. At some extremes of conditions the poly- merization mechanism could not be supported and no satisfactory polymer film was obtained, but all the films that were obtained had the characteristic BF4--poly(pyrrole) redox profile. However, a C104--doped poly(pyrrole) has been reported to give a redox reaction in MeCN-Et4NC104 with peak potentials closer to PTS-type behaviour, although without the characteristic PTS shape and with a broad envelope in the +200 mV region.’ This result again emphasizes the signifi- cance of exact reproducibility of conditions and procedures.Here, it may be that film thickness, or ageing and time factors, provide the distinction. It should also be stressed that all data in this paper concern the first scan to which each sample is exposed, and a fresh film is prepared for each voltammogram. Toluenesulfonate-doped poly(pyrro1e) Here, the sharp reduction wave and broader re-oxidation wave at their respective potentials are maintained across films prepared under a variety of conditions, but not all. Polymerization potential Fig. 2 shows redox voltammograms of films produced at potentials ranging from +0.8 to +2.0 V.Both characteristic reduction and re-oxidation potentials are maintained through- out, whereas the strikingly sharp reduction wave shape is evident in all but the film prepared at +0.8 V. In this instance the polymerization is slow, requiring more than 15 min for the passage of 50 mC of charge, whereas at higher potentials less than 1 min is needed. It may be that the greater length of immersion during preparation is responsible for the broad- ness, as PTS films given prolonged immersion also display broadening of the waves. It is also interesting to note that the films produced at the most positive potentials are prepared in the region where in redox voltammetry a second oxidation wave is observed in poly(pyrro1es) and poly(thiophenes) (termed the ‘overoxidation’ process16).In spite of this, the reduction voltammetry appears unaffected. The redox reac- tion is of course carried out in a different medium to that used for the preparation procedure, and no pyrrole monomer is present. Electrolyte concentration Fig. 3 shows the redox behaviour of films prepared from Et4NPTS in MeCN at 0.005, 0.05 and 0.5 mol dm-3. At the two lower concentrations [Fig. 3(6) and (c)] there is little difference, but at the higher concentration [Fig. 3(a)] there is a1308 3 0 +- 2 t ANALYST, AUGUST 1992, VOL. 117 (a) ( b) ( C) 1 I500 pA I500 pA :: I500 PA I I I I I I I I 1 broadening and loss of shape, the re-oxidation is moved anodically and the redox envelope is more pronounced with a significant reduction current at +200 mV on the initial scan.The oxidation polymerization mechanism for pyrrole is still a matter for discussion, and may not involve the same process under all preparation conditions,but all suggested mechanisms at some point involve radical cations.lJ7J8 At high electrolyte concentrations a more intense anionic environment will prevail near the anode, with greater opportunities for ion pairing with cationic species. This will interfere with the activity of cations in the polymerization process, and it is notable that at an electrolyte concentration of 0.5 mol dm-3 there is much greater diffusion of dark-coloured material away from the anode, evidence of low relative molecular mass pyrrole-derived species that do not become incorporated in the growing polymer film.There is also an effect on the morphology of the final film. (4 1500 yA I I I Pyrrole concentration Fig. 4 shows the redox behaviour of films prepared at pyrrole concentrations between 0.015 and 0.375 mol dm-3. For the curves shown in Fig. 4(b)-(e) the characteristic PTS-type redox behaviour is observed, but for the lowest monomer concentration, the curve shown in Fig. 4(a) appears in both shape and potential to be more like the typical BF4- redox behaviour [cf. Fig. l(b)]. However, for this lowest monomer concentration the resulting film is patchy and brittle, unlike the normal run of PTS films which have particularly good physical and mechanical integrity.1 Here, a switch between PTS and BF4- redox behaviour has been produced by a change of preparation procedure.There is a difference between the polymerization for Fig. 4(a) and for Fig. 2(a). Both produce low-grade films under extremes of conditions, but in the former situation where the switch in redox behaviour is observed, low current reflects monomer starvation at the anode, with reactive species (4 J I500 PA I I I produced at high electrode potential (+ 1.25 V) appearing into a monomer-depleted environment; whereas in the latter situation low current reflects slow generation of species produced at a low potential, but appearing into a monomer- rich (0.1 mol dm-3) environment. This latter situation does not produce a switch in redox behaviour, and these results may have implications for the polymerization mechanism. Polymerization solvent Fig. 5 shows the redox behaviour of PTS-poly(pyrro1e) produced from MeCN, PC, water, methanol and water- methanol (3 + 1); Et4NPTS was the electrolyte salt in all preparations.Both organic solvents give the characteristic PTS-type behaviour in shape and potential, whereas the hydroxylic solvents produce broader peaks, an anodic shift of the re-oxidation wave with a greater proportion of current in this wave compared with the reduction wave, and a compara- tively wider envelope at the anodic end of the scan [cf. Fig. 1 01 for aqueous preparation of PSS-poly(pyrro1e)l . There is opportunity for nucleophilic attack of these solvents on intermediate (radical) cationic species during polymerization, and the curve from methanol [Fig. 5(d)] is particularly poorly defined, although such peaks that are evident are in the typical PTS positions.Relative solvation factors for electrolyte ions in free solution and for the ionomeric polymers produced are more significant in hydroxylic preparation media, but the redox voltammogram of the polymer prepared from water [Fig. 5(c)] shows a well defined shape in spite of these possible complications, and the reduction wave remains at the charac- teristic PTS potentials. Benzenesulfonate (BS) The above results show that the distinction between BF4-- poly(pyrro1e) and PTS-poly(pyrro1e) redox behaviour remains evident across a wide range of preparation conditions, t Y C 2 3 0 Fig. 2 Redox voltammograms of PTS-doped poly(pyrro1e) prepared at different potentials (versus SCE): ( a ) 0.8; ( b ) 1.0; (c) 1.25; ( d ) 1.5; and ( e ) 2.0 V.Conditions as in Fig. 11309 ANALYST, AUGUST 1992, VOL. 117 ( a ) 1 .o 0 t 0.5 b) -1.0 0 +0.5 [- I -1.0 0 +0.5 PotentialN Fig. 3 Redox voltammograms of PTS-doped poly(pyrro1e) prepared from MeCN-Et,NPTS at different electrolyte concentrations: ( a ) 0.5; (b) 0.05; and (c) 0.005 mol dm-3. Conditions as in Fig. 1 and that within the parameters studied, BF4--type behaviour could not be switched to PTS-type behaviour, whereas the reverse is feasible, but only at extremely low monomer concentration, producing a poor quality film. For the BS anion, however, a more facile switch between the two types of behaviour is found. This is demonstrated in Fig. 6, in which curves A and B are taken from Fig. l(d) and ( b ) , respectively, and are superimposed to emphasize the difference between PTS- and BF,--type behaviour (the potential scales are the same but the current scales are different), whereas Fig.6, curve C, shows a BS-poly(pyrro1e) film prepared by the standard procedure from the tetrabutyl- ammonium salt in MeCN, superimposed on curve D which shows a comparable film prepared from the sodium salt in water-methanol (3 + 1). Here, a complete switch in redox behaviour has been achieved by choice of polymerization solvent. Figs. 5 and 6 show graphically the propensity for different types of behaviour from various poly(pyrro1es) prepared under relatively similar conditions; this may explain the contradictory reports that have appeared in the literature. Conclusion In a study involving consistent procedures it has been shown that the redox voltammetric behaviour of poly(pyrro1es) t c S 2 I 1 I -1.0 0 +0.5 I b) I I 1 -1.0 0 +0.5 v -1.0 0 +0.5 -1.0 0 +0.5 PotentialN Fig.4 Redox voltammograms of poly(pyrro1e) doped with PTS prepared at different pyrrole monomer concentrations: ( a ) 0.015; (b) 0.075; (c) 0.1; and ( d ) 0.375 mol dm-3. Conditions as in Fig. 1 containing different dopant anions in MeCN containing Bu4NBF4 divides into two main types depending on the nature of the dopant. The small inorganic anions BF4-, C104- and PF6- give broad and poorly defined reduction waves at facile potentials of about +200 mV. This behaviour has previously been attributed to anion expulsion with associated polymer chain1310 ANALYST, AUGUST 1992, VOL. 117 1 .o t 4- 2 3 0 I I I -1.0 0 +0.5 1 t CI C 2 3 u 0 +0.5 -1.0 0 +0.5 : e) 1 -1.2 0 +0.5 Potent ia IN Fig.5 Redox voltammograms of PTS-doped oly(pyrro1e) prepared from different solvents: (a) MeCN; (b) PC; (c! water; (d) methanol; and (e) water-methanol (3 + 1). Conditions as in Fig. 1 relaxation processes. * A The explanation requires sufficient space within the polymer matrix and chain flexibility to provide channels for anion motion. It may be that the semi-rigid extensively conjugated poly(pyrrole) structure can provide suitable vacancies to allow motion of small, spherical anions, but this is insufficient for larger anions of complex shape. Hence for PTS-doped poly(pyrrole), and for the larger anions DSFc and PSS, a different redox behaviour is observed. There is very little reduction current at +200 mV, and instead a strong and strikingly sharp reduction wave occurs at about -500 mV.The whole reduction and re- oxidation occurs at a substantially cathodic potential shift. The sharpness of the reduction peak is not always evident, particularly with sample preparation from hydroxylic solvents, but the shift in potential is diagnostic. Bulk films of PTS- poly(pyrro1e) have relatively high conductivities allied to good physical and mechanical properties,' and this may reflect a 'good fit' between the PTS anion and vacancies in the polymer matrix, restricting anion motion but without stressing the polymer infrastructure. This quantitative size argument explains why a change in preparation conditions fails to switch BF,--type behaviour to PTS-type behaviour, and also why a switch in the reverse sense does occur, but only under extremes of preparation conditions where the normal FTS- poly(pyrro1e) morphology cannot be maintained and a less rigid polymer matrix is formed with greater opportunity for ion mobility.This proposal is further supported by the behaviour of BS, a smaller anion than PTS and without the three dimensionality in geometric profile due to the sp3 carbon-containing methyl group. Here, a switch between I I I -1.0 0 +0.8 PO0 pA I I I -1.0 0 +1.0 PotentialN Fig. 6 Superposition of redox voltammograms of poly(pyrro1es) containing various dopants. A, PTS; B, C104- (both prepared from MeCN); C, BS (ex MeCN); and D, BS (ex aqueous methanol). Conditions as in Fig. 1 Table 1 Preparation conditions for the poly(pyrro1es) shown in Fig.1 Electrode Monomer potential concentration/ Electrolyte or current Dopant Solvent mol dm-3 concentration density BF4- MeCN 0.1 0.1 moldm-3 +1.25V C104- MeCN 0.1 0.lmoldm-3 +1.25V PF6- MeCN 0.1 0.1 mol dm-3 + 1.25 V PTS MeCN 0.1 0.1 mol dm-3 + 1.25 V DSFc PC 0.1 0.05 mol dm-3 1 mA cm-2 PSS Water-MeOH 0.1 1% m/v +1.25V (3 + 1) redox behaviour follows from a change of polymerization solvent. The non-aqueous solvent MeCN produces a tight morphology that restricts anion motion, whereas aqueous media, where solvation complicates ionic behaviour during polymerization and in the subsequent polymer, produce a matrix that appears to allow anion mobility. A more extensive theory is needed to explain all the observed aspects of poly(pyrro1e) redox voltammetry.However, the procedures described here allow discrimination between apparently similar poly(pyrro1es) and provide a useful means for analysis of these materials. In addition, a rapid reduction process is of potential importance in many applications, and there have been recent reports concerning faster-than-normal redox behaviour in conducting poly-ANALYST, AUGUST 1992, VOL. 117 131 1 mers.19JO The redox discrimination presented in this paper may distinguish material suitability for such applications. This study is continuing into redox discrimination in the second oxidation wave of poly(pyrro1e) (‘overoxidation’), and further complex and functionalized materials are being examined. In addition, the roles of redox parameters are being confirmed by the comparison of other experimental con- ditions.We thank NAB and SERC (for funding to C. E. H. and A. C., respectively), and Drs. R. Mortimer (Loughborough Univer- sity), D. R. Rosseinsky (University of Exeter) and R. Lines (Courtalds Coatings plc) for helpful discussions. References 1 2 The Handbook of Conducting Polymers, ed. Skotheim, T. A., Marcel Dekker, New York. 1986, vols. 1 and 2. Walton, D. J . , in Electronic Materials-from Silicon to Organics eds. Miller, L. S., and Mullin, J. B., Plenum Press, New York, 1991. 3 Diaz, A. F., Chem. Scr., 1981, 17, 145. 4 Scrosati, B . , Prog. Solid State Chem., 1988, 18, 1. 5 Burgmayer, P., and Murray, R. W., J. Phys. Chem.. 1984,88, 2515. 6 Cai, Z., and Martin, C. R., J. Electroanal. Chem., 1991,300,35. 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Duffitt, G . L., and Pickup, P. G., J. Phys. Chem., 1991, 95, 9634. Weinmayer, V., J. Am. Chem. SOC., 1955, 77, 300. Hall, C. E., Ph.D. Thesis, CNAA, Coventry Polytechnic, 1992. Qian, R., Li, Y., and Yass, B., Synth. Met., 1989, 28, C51. Kuwabata, S . , Okamoto, K., Ikeda, O., and Yoneyama, H., Synth. Met., 1987, 18, 101. Walton, D. J., Hall, C. E., and Chyla, A., Synth. Met.. 1991,45, 363. Naoi, K., Lien, M., and Smyrl, W. H., J. Electrochem. SOC., 1991, 138,440. Hillman, A. R., Swann, M. J., and Bruckenstein, S., J. Phys. Chem., 1991,95, 3271. Walton, D . J., Hall, C. E., and Chyla, A., unpublished work. Krische, B., and Zagorska, M., Synth. Met., 1989,28, C257. Asavapiriyanont, S., Chandler, G. K., Gunwardena, G. A., and Pletcher, D . , J. Electroanal. Chem., 1984, 177, 229. Morse, N. J . , Mortimer, R. J., Rosseinsky, D. R., and Walton, D. J., J. Electroanal. Chem.. 1988, 255, 119. Kalaji, M., Peter, L. M., Abrantes, L. M., and Mesquita, J . C., J. Electroanal. Chem., 1989, 274,289. Diaz, A. F., Kanazawa, K. K., and Lacroix, J. C., J. Electrochem. Soc., 1989, 136, 1308. Paper 2/01 405J Received March 17, 1992 Accepted April 22, 1992

ISSN:0003-2654    

DOI:10.1039/AN9921701305

出版商:RSC    

年代:1992

数据来源: RSC

摘要:

ANALYST, AUGUST 1992, VOL. 117 1313 Functionalized a-Cyclodextrins as Potentiometric Chiral Sensors Ritu Kataky, Paul S. Bates and David Parker* Department of Chemistry, University of Durham, South Road, Durham DHI 3LE, UK Octylated cyclodextrins have been synthesized and characterized by mass spectrometry (+ fast atom bombardment, + field desorption) and 500 MHz 1H nuclear magnetic resonance spectroscopy. These highly lipophilic, enantiomerically pure molecules have been incorporated into solvent polymeric membranes and investigated as electrochemical sensors for chiral molecules incorporating aryl rings. Bis( 1 -butylpentyl) adipate (BBPA) and ortho-nitrophenyl octyl ether (0-NPOE) were used as plasticizers. Electrodes using BBPA as the plasticizer were stable and well defined with a limit of detection for ephedrine of -log[c] = 6.5.Interference from serum levels of Na+, K+ and Ca*+ is minimal; the best value obtained for -log kpot (the over-all selectivity coefficient) was 3.9 with BBPA as plasticizer and 1 x 10-3 mot dm-3 NH&I as inner filling solution. The electrodes were highly enantioselective in binding ephedrine (enantioselectivity k:$-,, 2.7). The o-NPOE-based electrodes, although enantioselective with minimal interference from serum levels of Na+, K+ and Ca2+, behaved in a time-dependent manner. Keywords: Cyclodextrin; sensor; potentiometry; chiral; enantioselective Cyclodextrins (CDs) are optically active oligosaccharides consisting of 6-12 D-glucose units with an ~ ~ ( 1 - 4 ) linkage. They form inclusion complexes in aqueous solution and in the solid state with various aromatic molecules and are toroidal in shape with each of the chiral glucose residues possessing a rigid 4C1 chair conformation (Fig.1). Complex formation requires that both the host and the guest molecules are complementary ( i e . , they possess a favourable match of the aryl moiety with the CD cavity). The interior of the CD cavity is highly hydrophobic and non-polar, and the hydroxy groups, which are directed away from the molecular cavity, are readily accessible for chemical modification. Cyclodextrins have been used for the resolution of race- mates by stereoselective complex formation. 1-3 They have also been used in aqueous liquid membranes for the enrich- ment of racemic mixtures.4 More recently, they have been used as chiral stationary phases in gas chromatographic (GC) and high-performance liquid chromatographic (HPLC) analy- ses.5.6 The peroctylation of a- and 6-CDs renders them highly lipophilic and suitable for incorporation into solvent poly- meric membranes.The primary aim of this work was to investigate the feasibility of using peroctylated a-CD l a as a sensing ionophore in potentiometric ion-selective electrodes for monosubstituted arylammonium ions. The enantiopure host was envisaged as forming diastereoisomeric complexes with chiral arylammonium analytes, allowing the possibility of selective detection of one particular enantiomer. Experimental Reagents and Chemicals The synthesis and characterization of the octylated a-CDs l a and l b used in the membrane preparation has been reported elsewhere .7 Ephedrine hydrochloride (Eph-HCI), norephedrine hydrochloride (norEpheHC1) and pseudo (9) ephedrine hydrochloride (VEph-HCl) were obtained from Sigma (Poole, Dorset, UK).High relative molecular mass poly(viny1 chloride) (PVC) , ortho-nitrophenyl octyl ether (0-NPOE), bis( 1-butylpenty1)adipate (BBPA) and potassium tetrakisk- chloropheny1)borate were obtained from Fluka (Buchs, Swit- zerland). Chloride salts of sodium, potassium and ammonia were obtained from BDH (Poole, Dorset, UK) and were of * To whom correspondence should be addressed. 3x0. l a R = R’ = C8HI7; n = 6 Peroctyl-a-cyclodextrin b R = CBH17, R’ = H +n Analytes Me Ph (-)-( 1 R,2S)-Ephedrine Me OH (+)-(I S,2R)-Ephedrine H H I (+ 141 S,2S)-Pseudoephedrine (-)-(I R,2R)-Pseudoephedrine Me Me H I Ho+il Ph * (-)-( 1 R,2S)-Norephedrine (+)-( 1 S,2R)-Norephedrine Plasticizers yo2 1 4 , L k”- Bu BBPA Fig.1 Structures of the compounds discussed in the text1314 ANALYST, AUGUST 1992, VOL. 117 AnalaR grade. A 1.0 rnol dm-3 calcium chloride solution (BDH AnalaR) was used. All standard solutions were prepared in de-ionized water (Milli-Q; Millipore-Waters, Milford, MA, USA) and their cation concentrations were checked by atomic absorption spectometry. Some of the membrane components are shown in Fig. 1 giving the structures of the peroctylated CD, the two plasticizers used, o-NPOE and BPPA, and of the lipophilic anion. Membrane Preparation The membrane composition for the o-NPOE-based mem- branes was 1.2% ionophore, 65.6% o-NPOE, 32.8% PVC and 0.4% potassium tetrakisb-chloropheny1)borate in 6 cm3 of tetrahydrofuran (THF).For the BBPA-based membranes, the composition was 2.0% ionophore, 65.6% BBPA, 32.0% PVC and 0.4% potassium tetrakisb-chloropheny1)borate in 10 cm3 of THF. The membranes were cast by a controlled evaporation method according to the published procedure.8 Unless other- wise stated compound l a was the ionophore used in these studies . Calibration and Selectivity Measurements A Philips IS (561) electrode body (Philips Analytical, Eind- hoven, The Netherlands) was used to mount the electroactive membranes. The reference electrode was a Philips double junction REYDJ electrode. The electrochemical cells were set up using two different inner filling solutions for the ion- selective electrode. (1) Ag,AgCI 1 0.01 rnol dm-3 Eph.HC1 1 PVC membrane I Analyte 11 0.1 mol dm-3 Li acetate (salt bridge) I KCl(salt) I (2) Ag,AgCI 1 1.0 mmol dm-3 NH4CI I PVC membrane I Analyte 11 0.1 rnol dm-3 Li acetate (salt bridge) 1 KCl(salt) 1 A constant dilution technique was used for calibration and selectivity measurements as described previously.9 The selec- tivity measurements were performed with a background of 150.0 mmol dm-3 NaCl, 4.3 mmol dm-3 KCI and 1.26 mmol dm-3 CaC12.All e.m.f. measurements were made at 25 "C (+ 0.1 "C). Hg2C12(s); Hg Hg2Cl2(4; Hg Bias Potential Measurements The bias potential between two peroctylated a-CD-BBPA electrodes, both containing 1.0 mmol dm-3 NH4CI as the inner filling solution, one conditioned in (+)-Eph-HCI and the other in (-)-Eph-HCI, was measured in the cell shown in Fig.2. One arm was filled with (+)-EpH-HCI and the other with (-)-Eph.HCI (0.1 rnol dm-3). The tap was carefully rotated to allow the solution of the (-)-enantiomer to move half way up the capillary tube. The second arm was filled with the (+)-enantiomer and this solution was forced down the capillary with the aid of a syringe fitted with a flattened needle. Care was taken to prevent the two solutions from mixing. The electrodes were immersed in the appropriate solutions and the potential difference was monitored over 4 h. Behaviour of the Electrodes in Solutions of Varying Enantiomeric Excess A range of solutions was prepared containing 0-100% of the (+)- and (-)-enantiomers of EphaHCI, respectively.The behaviour of the electrochemical cell containing 1 .O mmol dm-3 NH&I inner filling solution and with the - w Fig. 2 Cell used for measurement of bias potential I _ > La-- € > y: E 1" ,,j -50 2" 3 - 9 Fig. 3 Calibration graphs for peroctylated a-CD as the sensin ionophore with o-NPOE as plasticizer and a 1 X 10-2 mol dm- Eph.HC1 inner filling solution. l*, (-)-Eph.HCl; 2*, (+)-Eph.HCl; 3, (+)-E h.HC1; 4, (-)-Eph.HCl; and 5 , (+)-Eph.HCl. * Back- ground oFserum levels of Na+, K+ and Ca2+ (see Table 1) ion-selective electrode mounted with an a-CD-BBPA elec- troactive membrane conditioned either in (-)-Eph-HCl or (+)-Eph.HCI, was observed. Results Calibration of Electrodes Using o-NPOE as plasticizer The first set of electrodes to be tested had o-NPOE as plasticizer with l a as the sensing ionophore and used a 0.01 mol dm-3 solution of either (+)-Eph-HCI, (-)-Eph-HCI or (+)-Eph-HCl as the inner filling solution.The electrodes were conditioned in 0.01 mol dm-3 solutions of the appropriate enantiomer and were calibrated by continuous dilution. The (+)-enantiomer showed a normal Nernstian response with a detection limit -log[c] = 4.8. The (-)-enantiomer and the racemic mixture behaved in a Nernstian manner down to a concentration of 1 x 10-3 rnol dm-3. On further dilution an unusual hyper-Nernstian behaviour was observed. In a background of serum levels of Na+, K+, and Ca2+ this behaviour was not observed (Fig. 3 and Table 1). Both electrodes functioned satisfactorily with over-all selectivity coefficients of -1ogkpof = 3.82, and 3.68 for the (+)- and (-)-enantiomers, respectively.In the absence of added inorganic cations, the initial difference in measured electrodeANALYST, AUGUST 1992, VOL. 117 1315 Table 1 Behaviour of electrodes with l a using o-NPOE as plasticizer and 1 x 10-2 rnol dm-3 Eph-HCI inner filling solution Limit of Slope/mV detection, Selectivity, Sensor decade- -log[c] -log kpot (+)-Eph*HCI* 60.0 4.64 3.82 (-)-Eph*HCI* 60.0 4.54 3.68 (-)-Eph*HCI NQt NQt (-t)-Eph*HCI NQt NQt (+)-Eph*HCI 59.0 4.80 - - - * Background of serum levels of Na+, K+, Ca2+. t NQ = The slope and limit of detection have not been quoted because of the unusual behaviour of the electrode. Po 21 20 Timelh Fig. 4 Behaviour of electrode based on peroctylated a-CD (la), o-NPOE, 0.01 rnol dm-3 (-)-Eph.HCI inner filling solution, condi- tioned in 0.01 rnol dm-3 (-)-Eph.HCl.(a) Measurement of discrete solutions in the vicinity of 1 x 10-3 rnol dm-3. Further dilution revealed no hyper-Nernstian behaviour. (b) Electrode potential measured over 24 h in 0.1 rnol dm-3 (-)-Eph.HCI solution. The time-dependent behaviour is evident after 20 h potentials between the (+)-electrode immersed in 0.1 mol dm-3 (+)-Eph.HCl and the (-)-electrode immersed in 0.1 mol dm-3 (-)-Eph-HCl was 50 mV. This value was not constant with time and had reduced to 20 mV after further conditioning for 18 h in the same solutions (Fig. 3 and Table 1). Thereafter it remained constant. In discrete solutions, calibrations in which the electrode was transferred into solutions of decreasing concentration within 60 s, the hyper-Nernstian behaviour was not observed (Fig.4). However, on transferring the (-)-electrode from the condi- tioning solution into a 0.1 mol dm-3 solution of (-)-Eph-HCl, an abrupt and reversible drop in potential was observed after about 20 h (Fig. 4). Using BB PA as plasticizer The plasticizer was changed to BBPA. The concentration- dependent hyper-Nernstian response that was evident when o-NPOE was used as plasticizer was no longer observed (Table 2). The (-)-Eph.HCl electrode showed a slope 13 mV decade-' less than the (+)-Eph-HCI sensor and the difference in electrode potentials, AE[ (+) - (-)I was 14 mV. In a background of serum levels of Na+, K+ and Ca2+ the over-all selectivity coefficient, -1ogkW was 2.73 for the (+)-enantiomer.The next experiment was performed in an attempt to minimize the standard electrode potential differences. Elec- trodes were chosen that had 1.0 mmol dm-3 NH4Cl as the inner filling solution instead of the appropriate Eph-HC1 solution. The o-NPOE-based electroactive membrane behaved unusually again with (-)-Eph-HCl as analyte, after dilution beyond a concentration of 10-2.8 mol dm-3. The (+)-Eph-HCl electrode functioned satisfactorily, showing Table 2 Behaviour of electrodes with l a using BBPA as plasticizer and 1 x 10-2 mol dm-3 Eph-HCI as inner filling solution Limit of Over-all Slope/mV detection, selectivity, Sensor decade- -Io~[c] -log kpot (+)-Eph.HCI 59.0 5.05 - (-)-Eph*HCI 46.0 5.40 - (+)-Eph.HCl* 56.0 3.55 2.73 * Background of serum levels of Na+ , K+, Ca2+.Table 3 Behaviour of electrodes with l a using o-NPOE as plasticizer and 1.0 mmol dm-3 NH4CI inner filling solution Limit of Over-all Slope/mV detection, selectivity, decade- -log[c] -log kpot Sensor - (+)-Eph*HCl 56.0 5.25 (+)-Eph.HCl* 58.0 4.70 3.91 (-)-Eph.HCl NQt NQt (-)-Eph.HCI* 52.0 5.27 4.45 - * Background of serum levels of Na+ , K+, Ca2+. t NQ = The slope and limit of detection have not been quoted because of the unusual behaviour of the electrode. 290 > 190 E E i > y: 90 -10 0.5 2.5 4.5 6.5 -Loglcl Fig. 5 Calibration graphs for electrodes with a-CD-BBPA mem- branes with NH&I inner filling solutions. 1*, (+)-Eph.HCl; 2*, (-)-Eph-HCl; 3, (-)-Eph-HCl; and 4, (+)-Eph.HCl. * Background of serum levels of Na+, K+ and Ca2+ (see Table 4) the expected Nernstian response.In a background of serum levels of Na+, K+ and Ca*+ it showed very little interference, -log kpot = 3.9 (Table 3). The more well defined electrode had BBPA as plasticizer and 1.0 mmol dm-3 NH4Cl as inner filling solution. The unexpected behaviour observed with o-NPOE was again not evident. The slope of the (-)-Eph-HCl sensor was 10 mV decade-' lower than that of the (+)-Eph-HCI sensor. The difference in electrode potentials between enantiomeric elec- trodes AE[(+) - (-)I in the appropriate 0.1 mol dm-3 solutions was 26.0 mV in aqueous solutions and 21.0 mV in a background of serum levels of Na+ , K+ and Ca2+ correspond- ing to -log kG\,(-) of 2.7 and 2.3, respectively { -log kfy\,(-) = [E(+) - E,-)]/S, where S is the electrode slope}. The over-all selectivity coefficients were -log kpot 3.9 and 3.5 for the (+)- and (-)-enantiomen, respectively (Fig.5, Table 4). The only difference between the (+)-Eph-HCl and the (-)-Eph-HCl sensors used in this experiment was that the electrodes had been conditioned separately in a 0.1 mol dm-3 solution of the appropriate enantiomer. The 'bias' potential between these two electrodes was measured in the cell shown in Fig. 2. This cell was used in order to eliminate any errors that may arise due to liquid junction potentials. The measured potential was observed to be constant over 4 h at ambient temperature: AEbias = E(+) - I?(-) = 24.5 k 0.5 mV1316 ANALYST, AUGUST 1992, VOL. 117 Table 4 Behaviour of electrode with l a using BBPA and 1.0 mmol dm-3 NH4Cl inner filling solution Limit of Over-all Slope/mV detection, selectivity, Sensor decade-' -log[c] -log kpot (+)-Eph*HCl 60.0 6.3 - (-)-Eph*HCl 50.0 6.6 - (+)-Eph.HCl* 59.0 4.7 3.9 (-)-Eph.HCl* 59.0 4.4 3.5 * Background of serum levels of Na+, K+, Ca2+.-100 -50 0 50 100 (1 s,~R) Enantiomeric purity (%) (1 ~,2s) Fig. 6 Behaviour of electrodes in solutions of varying enantiomeric excess. Electroactive membrane: a-CD-BBPA; inner filling solution: 1.0 mmol dm-3 NH,Cl; conditioned in 10 mmol dm-3 (+)- or (-)-Eph-HC1. 1, (-)-Eph-HCI-BBPA; and 2, (+)-Eph-HCl-BBPA This corresponds to a free energy difference between the two diastereoisomeric complexes of 2.4 (0.05) kJ mol-1. Enantioselective Sensor The performance of the a-CD-BBPA electrodes containing 1.0 mmol dm-3 NH4CI as inner filling solution was assessed in solutions containing ephedrine of varying enantiomeric pur- ity.The electrodes were conditioned overnight in the appro- priate 0.01 mol dm-3 Eph-HCI solution. The (-)-Eph-HCl electrode appears to be the more sensitive of the two (Fig. 6), exhibiting a near linear e.m.f. response with varying enan- tiomeric purity. The stability of the BBPA-based ionophore was monitored over 5 weeks in 0.1 and 0.01 mol dm-3 solutions. The e.m.f. readings were reproducible to within k0.5 mV. Calibration and selectivity measurements were also per- formed with enantiomers of norEph-HC1 and qEph.HC1 using BBPA as plasticizer and the appropriate enantiomer as the inner filling solution (Table 5). The (-)-norEph-HCI- BBPA electrode showed a slope 12 mV decade-' less than that of the (+)-enantiomer.The other electrodes, although satisfactory in terms of slope, limit of detection and selectivity, did not show significant enantioselective behaviour. When a racemic mixture of vEph.HC1 was used as inner filling solution, the electrodes functioned satisfactorily in terms of slope, selectivity and limit of detection (Table 5). However, they did not show enantioselective behaviour. The less highly octylated a-CD l b (two octyl groups and one free hydroxy group per glucose residue) was not a good sensor for these P-hydroxyarylammonium salts. With BBPA as plasticizer and the appropriate enantiomer [i.e., (+)-enan- tiomer when testing (+)-enantiomer as the analyte] as the inner filling solution, the (+)-electrode gave a slope of 24 mV decade-*.The slope of the (-)-electrode was 59 mV decade-' at a 1 x 10-2 mmol dm-3 dilution; however, the Table 5 Behaviour of electrode with l a using BBPA as NorEph.HC1 and qJEph.HC1 sensor inner filling solution: 0.01 mmol dm-3 analyte Limit of Over-all Slope/mV detection, selectivity, Sensor decade-' -log[c] -1ogkpo' (+)-norEphVHC1 58.0 5.05 - (-)-norEphaHC1 46.0 3.80 - (+)-norEph-HC1 58.0 2.90 2.1 (+)-vEph.HCI 56.0 4.70 - (-)-qJEph*HCl 59.0 5.10 - (+)-qJEph*HCI 59.0 5.20 - * Background of serum levels of Na+, K+ , Ca2+. limit of detection was -log[c] = 2.9, much reduced compared with its peroctylated analogue la. Discussion Electrodes that are based on peroctylated a-CD appear to show a highly pronounced enantioselective behaviour pro- vided that the appropriate inner filling solution is used.With BBPA as plasticizer, this is evident both in the measured AE values and as a lower slope for the (-)-enantiomer. Based on the bias potential measurements, and assuming that the slopes of the electrodes are equivalent: log kY?j/(-) = 2.6; {log kf'?y(-) = [E(+) - E(--)]/S, where S = slope} The (+)-enantiomer thus appears to form a more stable complex with the CD than the (-)-enantiomer. Yasaka et aE.,10 and Bussmann et al.," have previously reported -log kf'$,(-) , using chiral 18-crown-6 based-macrocyclic polyethers, as 1.5 and 2.6, respectively, for the a-phenylethyl- ammonium ion determined by the separate solutions method. These sensors are limited by their sensitivity to Na+ and K+ and cannot be used in a clinical background of serum cations.With o-NPOE as the plasticizer, the electrodes behaved in an unexpected manner. While detection of the (+)-enan- tiomer was normal with a Nernstian response in both the presence and absence of serum cations at clinical concentra- tions, the detection of the (-)-enantiomer was concentration- and time-dependent and was also sensitive to the absence or presence of added cations. This intriguing behaviour could, in principle, be related to competititve binding of the o-NPOE by the peroctylated CD. It is known, for example, that o-nitrophenol forms a 1 : 1 complex with a-CD in aqueous solution in which the aryl nitro group enters the 'cavity' first.12-14 That such behaviour is not observed at all with the (+)-enantiomer (under any conditions of ephedrine concen- tration) suggests that this is unlikely particularly in the light of the apparent small difference in the free-energy of binding (about 2.4 kJ mol-1 at 298 K) of the two enantiomers.A more likely explanation may involve a concentration- and ionic strength-dependent aggregation phenomenon involving both the peroctylated CD and the plasticizer. When a charged arylammonium ion is bound by the peroctylated CD the complex may be regarded as amphiphilic. Enantioselective aggregation may occur beyond a critical concentration which is inhibited in the presence of added cations (i.e., at higher ionic strength). Thus the modest enantioselectivity observed at the molecular level may be amplified in the chiral aggregate.Preliminary *H NMR spectroscopic investigations with peroctylated a-CD and the trifluoroacetate salts of (+)- and (-)- Eph in CDCI3 (298 K, 1 : 1 stoichiometry, 0.05 mol dm-3) show that the chemical shift of certain of the CD resonances (3-H, 5-H, and C-6, CH20) is dependent on the nature and concentration of the enantiomer included, indicative of enantioselective binding. Further NMR and circular dichro-ANALYST, AUGUST 1992, VOL. 117 1317 ism experiments are in progress in order to define the structure and relative stability of the diastereoisomeric com- plexes, and will be reported subsequently. Conclusions The peroctylated a-CD-BBPA electrode using 1 .O mmol dm-3 NH4CI as the inner filling solution is suitable as a chiral sensor.It has an excellent sensitivity (60 mV decade-' at 25 "C), limit of detection (-log[c] -6.5), selectivity over serum levels of cations (-log kpot = 3.9) and enantioselectivity (-log kqyO:_ = 2.7). A calibrated electrode has been construc- ted that allows the enantiomeric purity of (-)-ephedrine (the pharmacologically active enantiomer) to be measured, even in the presence of its diastereoisomers, ( R , R)- and (S,S)-pseudo- ephedrine. We thank Professor Arthur K. Covington (Department of Chemistry, University of Newcastle-upon-Tyne, UK) for his helpful comments and for supplying the cell (Fig. 2). We thank the SERC for financial support and Professor M. Shankar, University of Durban, Westville, South Africa, who has confirmed our results by independent measurements at the Electrochemistry Research Laboratories, University of New- castle-upon-Tyne, UK. We thank him for his interest and help. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 References Cramer, F., and Dietshem, W., Chem. Ber., 1959,92,378. Benschop, H. P., and van den Borg, G. R., J. Chem. SOC., Chem. Commun., 1970, 1481. Mikolajczyk, M., and Drabouricz, J., J. Am. Chem. SOC., 1978, 100,2510. Armstrong, D. W., and Jin, H. L., Anal. Chem., 1987,539,2237. Konig, W. A., Carbohydr. Res., 1989, 192, 51. Konig, W. A., Lutz, S., and Wenz, G., Angew. Chem., Int. Ed. Engl., 1989,27, 979. Bates, P. S., Kataky, R., and Parker, D., J. Chem. SOC., Chem. Commun., 1992, 153. Craggs, A., Moody, G. J., and Thomas, J. D. R., J. Chem. Educ., 1974, 51,541. Kataky, R., Nicholson, P. E., Parker, D., and Covington, A. K., Analyst, 1991, 116, 135. Yasaka, Y., Yamamoto, T., Kimura, K., and Shono, T., Chem. Lett., 1986, 769. Bussmann, W., Lehn, J.-M., Oesch, U., Plumene, P., and Simon, W., Helv. Chim. Acta, 1981,64, 657. Thomas, A. P., Helv. Chim. Acta, 1979, 62, 2303. Bender, M. L., and Komiyama, M., Cyclodextrin Chemistry, Springer, Berlin, 1978, pp. 10-27. Saenger, W., Angew. Chem., Int. Ed. Engl., 1980, 19, 334. Paper 1104725F Received September 11, 1991 Accepted February 12, 1992

ISSN:0003-2654    

DOI:10.1039/AN9921701313

出版商:RSC    

年代:1992

数据来源: RSC

摘要:

ANALYST, AUGUST 1992, VOL. 117 1319 Primaquine-selective Electrodes Based on Macrocyclic Crown Ethers* B. B. Saad, 2. A. Zahid, S. A. Rahman and M. N. Ahmad School of Chemical Sciences, Universiti Sains Malaysia, I 1800 Penang, Malaysia A. H. Husin School of Pharmaceutical Sciences, Universiti Sains Malaysia, I 1800 Penang, Malaysia Electrodes based on poly(viny1 chloride) matrix membranes containing one of three macrocyclic crown ethers with dioctyl phenyl phosphonate solvent mediator and, in some instances, anion excluder have been studied for their potentiometric response to the primaquinium cation. Primaquine-selective electrodes based on dibenzo( 18-crown-6) gave su b-Nernstian responses while those based on dibenzo(24-crown-8) (DB24C8) and dibenzo(30-crown-I 0) (DB30ClO) exhibited good electrochemical characteristics such as Nernstian responses, fast dynamic response times (=30 s), a wide range of working pH (pH = 4-10) and good selectivity over many metal cations, chloroquine and sulfonamide drugs.The addition of 50% mole ratio anion excluder, relative to the sensor, not only led to an improvement in the e.m.f. stability, but also produced, in most instances, improved selectivity characteristics for both the DB24C8- and DB30C10-based electrodes. Determination of primaquine diphosphate (4.5-453.0 pg cm-3) using the standard additions method resulted in a mean recovery and relative standard deviation of 107.0 and 8.0%, respectively. Determination of primaquine in pharmaceutical preparations is also described. Keywords : Ion-selective electrode; crown ether ion sensor; primaquine electrode; antimalarial The high selectivity characteristics of many neutral carrier ligands are of much interest to analytical chemists and have been exploited in the development of high-performance ion-selective electrodes (ISEs) .Highly selective neutral car- rier-based ISEs, notably for K+, Na+, Ca2+ and to a lesser extent Li+, NH4+, Mg*+, Pb*+, Cd2+ and U022+, etc., are already well established.'-3 However, there is a need for studies to be undertaken on a wider range of organic ion sensors,4 especially those of pharmaceutical and clinical interest. Among the few studies reported to date on the utilization of neutral carrier ligands for organic ions are the potentiometric sensing of guanidinium ,5-8 diquat ,49899 paraquat ,899 4,4'-dipyr- idinium,8,9 adenosinium, 10 amphetaminiumll and alkyl-12 and benzylammoniuml2 ions.In these complexes, interaction between host and guest is mainly achieved by hydrogen bonding and dipole-induced forces. 13 In this paper, the study of three macrocyclic crown ethers, namely dibenzo( 18-crown-6) (DB 18C6), dibenzo(24-crown- 8) (DB24C8) and dibenzo(30-crown-10) (DB30ClO), as pos- sible potentiometric primaquinium ion sensors is reported. Interest in primaquine 1 [8-(4-amino-l-methylbutylamino)-6- methoxyquinoline] arises from the problem of the drug resistance of malarial parasites to the conventional anti- malarial drug, chloroquine. 14-17 Primaquine is active against both the pre- and exo-erythrocytic forms of the plasmodium parasite in the liver and spleen in addition to gametocytes of falciparum infections.16 It is particularly effective for the radical cure of relapsing malaria.y 3 + + H 2 N-CH (CH 2 13N H 3 (1) According to the US and British Pharmacopoeias, prima- quine in pharmaceutical preparations may be determined using titrimetry.18,19 However, this approach is subject to * Presented, in part, at the 11th Australian Symposium on Analytical Chemistry, Hobart, Australia, July 8-12, 1991. interference from other sulfonamide drugs. Other methods for primaquine determination are high-performance liquid chromatography (HPLC) ,*4 spectrophotometry*4~20-22 and spectrofluorimetry .I4 These techniques, while giving the required sensitivity, involve extensive sample pre-treatment.In contrast, the ISE approach is rapid and the sensitivity and selectivity can often rival those of the routine methods used pre~ently.2~ Experimental Reagents The reagents and materials used were obtained from the following sources: dioctyl phenyl phosphonate (DOPP) (Lan- caster Synthesis, Lancaster, UK); trioctyl phosphate (TOP) (Fluka, Buchs, Switzerland); poly(viny1 chloride) (PVC), DB18C6, DB24C8, DB30C10, potassium tetrakis(4-chloro- pheny1)borate (KTPB), diphosphates of primaquine and chloroquine, hydrochlorides of quinine , cinchonine and qui- nacrine (Sigma, St. Louis, MO, USA); sodium sulfacetamide (B.V. Drugpharm, Amsterdam, The Netherlands); and sodium sulfadimide (Halewood Chemicals, Middlesex, UK). Primaquine tablets (7.5 mg base per tablet) were purchased from a local drugstore.Chlorides of metal ions were all of analytical-reagent grade and distilled, de-ionized water was used throughout. Electrode Preparation and Measurement Electrode master membranes were prepared from PVC (170 mg) and crown ether (10 mg) with DOPP solvent mediator (360 mg). In some instances 50% mole ratio KTPB, relative to the sensor, was added to the cocktail. All other details relating to electrode fabrication were as described previously.24 The assembled electrodes were conditioned for 24 h by exposure to a 0.1 mol dm-3 primaquine solution. A similar concentration of primaquine diphosphate was used as the inner filling solution. All electrode potentials were measured with an Orion digital pWmillivoltmeter (Model 701A) versus an Orion 90-02 Ag-AgC1 double-junction reference electrode (Orion, Cambridge, MA, USA).The selectivity coefficients were determined using the separate solution method at 1 X 10-2 mol dm-3 concentration as described p r e v i o ~ s l y . ~ ~ ~1320 ANALYST, AUGUST 1992, VOL. 117 Standard Additions Procedure Standard primaquine diphosphate solutions and primaquine tablets were analysed using the standard additions method.25 Three batches of 15 primaquine tablets were each finely ground and mixed thoroughly, dissolved, and filtered into a calibrated flask (50 cm3) and diluted to the mark. The standard additions method25 was performed by spiking 150 mm3 of standard 1 x 10-2 mol dm-3 primaquine diphosphate solution into 15 cm3 of the test solution and measuring the change in potential.Results and Discussion The key electrochemical characteristics of the primaquine electrodes are summarized in Table 1. The data reveal near-Nernstian responses to primaquinium cation for the DB24C8- and DB30C10-based membranes (Electrodes 2 and 3). The response times are reasonably fast, being about 30 s to achieve a 95% steady-state dynamic response time for prima- quine diphosphate concentrations >1 x 10-5 rnol dm-3. However, primaquine electrodes based on DB 18C6 (Elec- trode 1) gave not only sub-Nernstian responses but also yielded unstable potential readings and are thus of little use to analytical chemists. This suggests that DB18CCprimaquine complex formation, based on the cavity size-ionic radius relationship, is unfavourable. The bulky cation of primaquine is difficult to accommodate within the coordination cavity of the DBl8C6 but crowns with larger cavity sizes such as DB24C8 and DB30C10 are more suitable for this purpose.X-ray and nuclear magnetic resonance (NMR) studies have confirmed that where the host cavity is too small to accommo- date the guest in the normal 1:l complex, there is the possibility of the formation either of (i) 2 : 1 host-guest sandwich complexes, or (ii) 1 : 1 complexes with flexible hosts having large cavities in which the host adopts a U-shaped conformation, allowing it to encapsulate the guest. Host- guest complexes of type (ii) have been observed for DB30C10 -K+4,9926 and DB30ClWiquat2+ dication,4~9 whereas (15- crown-5)2-K+ and (12-~rown-4)~-Na+ systems of type (i) are well-known.26.27 The DB18CCamphetaminium complex for- mation has been proposed11 as being due to the ability of the NH3+ cation (radius 1.43 A) of amphetamine to anchor into the cavity of the DB18C6 (radius 1.3-1.6 A) host.The response characteristics of electrodes based on DB24C8 and DB30C10 were tested towards a range of metal cations, sugars and other antimalarial and sulfonamide drugs. Interference from metal cations, glucose and fructose is generally negligible (Table 2). However, with the exception of chloroquine and the sulfonamides, other antimalarial drugs such as quinine, cinchonine and quinacrine cause significant interference. A mean recovery of 108.0% (n = 4) obtained from the determination of 1 x 10-2 rnol dm-3 primaquine diphosphate in the presence of 1 X 10-3 and 1 X 10-2 rnol dm-3 chloroquine diphosphate solutions is further Table 1 Electrochemical response characteristics of primaquine-selec- tive electrodes* (solvent mediator : DOPP) Correlation Detection Electrode Membrane Slope/ coefficient, limit/10-6 No.composition mV decade- 1 r mol dm-3 1 DB18C6 24.5 0.993 48.0 2 DB24C8 30.0 0.997 7.9 3 DB30ClO 31.6 0.998 7.9 4 DB24C8+ 34.1 0.997 1 .o 5 DB30C10 + 33.4 0.995 8.9 KTPB KTPB * All data are average of three determinations. evidence of the negligible interference of chloroquine in the electrode response. The response characteristics of electrodes based on DB24C8 and DB30C10 were further evaluated on addition of KTPB to their membranes. Addition of this membrane additive, as expected, resulted in electrodes with better e.m.f.stability due to the decreased membrane resistance and in most instances to improved selectivity characteristics (Table The effect of pH on the electrode was also studied for the DB24C8- and DB30ClO-based membranes by varying the pH of 1 X 10-2 mol dm-3 primaquine diphosphate solution by addition of small volumes of sodium hydroxide and/or hydrochloric acid. It was found that both electrodes have a working pH range of about 3.5-10. The electrodes could be used for about 30 d. After this period had elapsed, the electrode response became sluggish and noisier, presumably due to leaching of membrane components. Potentiometric determination of primaquine diphosphate using the known additions technique utilizing the primaquine- selective electrode based on DB24C8 and KTPB (Electrode 4) yielded an average recovery and relative standard deviation of 107.0 and 8.0%, respectively (Table 3).Such results are expected in potentiometric determinations where an error of 1 mV for a divalent ion will result in an 8% error in concentration. Variation in pH of the samples (pH range 2.56-5.34) could contribute to the rather high recoveries, especially at lower concentrations. Analysis of three batches of primaquine tablets yielded an average and mean relative standard deviation of 7.7 and 7.0% d m , respectively, compared with the quoted value of 8.0% m/m. 2). Table 2 Selectivity coefficients of primaquine electrodes (separate solution method, measured at [cation] = 1 x rnol dm-3) Interferent, i Electrode 2 Electrode 3 Electrode 4 Electrode 5 K+ Na+ Li + Ca2+ Mg2+ cu2+ Pb2+ Zn2+ Glucose Fructose Quinine Cinchonine Quinacrine Chloroquine Sulfacetamide Sulfadimide NH4+ -2.41 -2.65 -2.72 -2.01 -3.48 -2.14 -1.88 -3.03 -3.03 -1.84 -2.92 -0.54 - 1.01 +0.11 -2.45 -3.05 -2.60 -2.10 - 1.55 -1.42 -0.86 -1.52 -2.83 -1.59 -3.00 -1.00 -2.31 -2.03 -0.44 -0.59 +1.31 -3.09 -3.30 -3.52 -2.63 -2.81 -2.11 -2.78 -3.80 -2.13 -2.26 -3.31 -3.74 -2.33 -3.00 -0.12 -0.27 -0.52 -1.86 - 1.76 -1.96 -2.00 -2.16 -2.70 -2.18 -1.64 -3.18 -2.25 -3.49 -2.18 -2.00 - 1.80 +0.06 -0.47 -1.11 -1.64 -2.43 -3.05 Table 3 Determination of primaquine using the known additions technique* utilizing a primaquine-selective electrode based on Elec- trode 4 (DB24C8 + KTPB) Primaquine Primaquine diphosphate diphosphate concentration/ concentration Recovery pg cm-3 foundlpg cm-3 ("/.I 4.53 5.0 k 0.467 110.4 (9.2)$ 45.3 49.6 k 1.61 109.5 (3.2) 226.7 245.2 & 26.6 108.2 (10.8) 453.0 452.2 k 40.0 99.8 (8.8) * Mean of four determinations.7 Standard deviation (pg cm-3). $ Relative standard deviation (%).ANALYST, AUGUST 1992, VOL. 117 1321 Conclusions Primaquinium ion sensors of high selectivity and sensitivity can be realized using the DB24C8 and DB30C10 systems, the former possessing better selectivity characteristics. These sensors provide a simple and convenient method for the determination of primaquine in pharmaceutical preparations. Interferences from eight metal cations, glucose and starch are generally insignificant. The most interesting selectivity feature of these electrodes is that chloroquine, the most widely used antimalarial drug, causes minimum interference. These elec- trodes offer prospects for the determination of primaquine, even in the presence of chloroquine as these drugs are often administered together. 14-17 However, the effect of their metabolites14-17 on the electrode response needs to be carefully evaluated, if they are to be used for the analysis of body fluids.The authors thank the Federal Government of Malaysia for sponsoring this work through its National R & D programme. References Ammann, D., Morf, W. E., Anker, P., Meier, P. C., Pretsch, E., and Simon, W., Ion-Sel. Electrode Rev., 1983, 5 . 3. Morf, W. E., The Principles of ISEs and of Membrane Transport, Elsevier, Amsterdam. 1981, p.274. Simon, W., in Ion-Selective Electrodes in Analytical Chemistry, ed., Freiser, H., Plenum Press, New York, 1978, vol. 1, p. 211. Moody, G. J., Owusu, R. K., and Thomas, J . D. R., Analyst, 1987, 112, 121. Bochenska, M., and Biernat, J . , Anal. Chim. Acta, 1984, 162, 369. Assubaie, F. N., Moody, G . J . , and Thomas, J. D. R., Analyst, 1988, 113, 61. Assubaie, F. N., Moody, G. J., and Thomas, J. D. R., Analyst, 1989, 114, 1545. Assubaie, F. N., Moody, G. J., Owusu, R. K., and Thomas, J. D. R., Port. Electrochim. Acta, 1987, 5 , 103. 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Moody, G. J., Owusu, R. K., and Thomas, J. D. R., Analyst, 1988, 113, 65. Umezawa, Y., Kataoka, M., Takami, W., Kimura, E., Koide, T., and Nada, H., Anal. Chem., 1988, 60,2392.Hassan, S. S. M., and Elnemma, E. M., Anal. Chem., 1989,61, 2189. Bussman, W., Lehn, J. M., Oesch, U., Plumere, P., andSimon, W., Helv. Chim. Acta, 1987, 70, 657. Hilgenfeld, R., and Saenger, W., in Host Guest Complex Chemistry I I , ed., Vogtle, F., Springer-Verlag, New York, 1982, p. 5. Bergqvist, Y., and Churchill, F. C., J . Chromatogr. Biomed. Appl., 1988, 434, 1. Kondrashin, A. V., Southeast Asian J . Trop. Med. Public Health, 1986, 17, 682. Report of WHO Scientific Group, WHO Technical Report Series, 1984, 711, 84. McChesney, E. W., and Fitch, C. D., in Antimalarial Drugs I t , eds., Peters, W., and Richards, W. H. G., Springer-Verlag, New York, 1984, pp. 1-49. United States Pharmacopeia, X X Revision, US Pharmacopeial Convention, Rockville, MD, 1980, p. 925. British Pharmacopoeia 1980, HM Stationery Office, London, 1980, vol. 1, p. 462. El-Kommos, M. E., and Emara, K. M., Alexandria J . Pharm. Sci., 1988, 2, 171. Talwar. N., Gogoi, P. J., Vyas, S. P., and Jain, N. K., Indian Drugs, 1990, 28, 156. Ibrahim, F. A., El-Brashy, A., and Belal, F., Mikrochim. Acta, Part I , 1989, 321. Moody, G. J., Owusu, R. K., and Thomas, J. D. R., Analyst, 1987, 112, 1347. Craggs, A., Moody, G. J., and Thomas, J. D. R., J . Chem. Educ., 1974,51, 541. Koryta, J . , and Stulik, K., Ion-Selective Electrodes, Cambridge University Press, Cambridge, 2nd edn., 1983, p. 100. Moss, R. E., and Sutherland, I . O., Anal. Proc., 1988,25,272. Mallinson, P. R., and Truter, M. R., J. Chem. SOC., Perkin Trans. 2, 1972, 1818. Paper I I02 742 E Received June 10, 1991 Accepted February 13, 1992

ISSN:0003-2654    

DOI:10.1039/AN9921701319

出版商:RSC    

年代:1992

数据来源: RSC

摘要:

ANALYST, AUGUST 1992, VOL. 117 1323 Influence of Plasticizer, and Metallic and Graphite Conducting Substrate on the Characteristics of a Poly(viny1 chloride) Membrane Coated Wire Calcium Ion-selective Electrode Ruzica Matesic-Puac and Mihajlo Dimitrijevic Department of Analytical Chemistry, Faculty of Food Technolog y, University of Osijek, 54000 Osijek, Croatia Milan Sak-Bosnar Denit AG Chemicals, 8050 Zijrich, Switzerland Poly(viny1 chloride) membrane-based calcium ion-selective electrodes have been constructed, with calcium bis(2-ethylhexyl) phosphate as a sensing material and tetrahydrofuran as solvent. The influence of several plasticizers and conducting substrates on the analytical characteristics (slope, interval response, lifetime and response time) of the electrodes was studied.The best characteristics were obtained with a graphite- poly(tetrafluoroethy1ene) conducting substrate of the R8iiCka type and with tributoxyethyl phosphate and o-nitrophenyl octyl ether as plasticizers. Chemical interference from several inorganic cations was also investigated. The analytical performance of the electrodes investigated was strongly dependent on a suitable combination of plasticizer and conducting substrate. Keywords: Calcium ion-selective electrode; poly( vin yl chloride) membrane; plasticizer; metallic and graphite conducting substrate In general, and especially in terms of selectivity, the analytical characteristics of all poly(viny1 chloride) (PVC) membrane- based ion-selective electrodes (ISEs) depend both on the sensor and the plasticizer, and also on their relative propor- tions in a membrane mixture, as has been shown in a recent review.' As the analytical performance of calcium ISEs (and all other ISEs based on organic-type sensors) depends on the chemical structure of the sensor and the plasticizer, the variation of these two has been the subject of several fundamental studies.2-5 The nature of the plasticizer has a marked influence on the response, slope, curve linearity and selectivity of PVC membrane electrodes6 A plasticizer can be used with a polymer or polymerizable material, which help it to acquire a more homogeneous mixture or dispersion.It is better, however, to select the plasticizer so that it is compatible with both the polymer and the sensing material, thereby reducing the electrical resistance of the polymeric bead of film.Depending on the type of polymer or polymerizable material, suitable plasticizers would be cyclohexanone, dioctyl phenyl phosphonate, tributyl phosphate, dipentyl phthalate, and dioctyl and diphenyl phthalates.7.8 The other group of plasticizers includes the compounds containing a nitro group: o-nitrophenyl octyl ether (o-NPOE) and o-nitrophenyl phenyl ether (o-NPPE).9-12 Jaber et al. 12 described the preparation of the tetraphenyl- borate of the calcium complex with poly(propy1ene glycol) and its use for a PVC matrix-membrane calcium ISE over a wide pH range. Some workers have expressed the opinion that dibutyl and dioctyl phthalates in PVC matrices show a selective response to calcium and uranyl ions.13 The plasticizer dioctyl phenyl phosphonate14~1S and its nitrated analogues16717 were sepa- rately synthesized.The electrodes containing dioctyl phenyl phosphonate and tributyl phosphate showed the best results whereas dioctyl p-nitrophenyl phosphonate reduced the selec- tivity of the electrodes. The analytical characteristics of PVC membrane-based calcium ISEs were studied extensively by Craggs et al. 18 They investigated the influence of several primary aliphatic alcohols and certain plasticizing solvents as solvent mediators in conjunction with calcium bis(didecy1 phosphate) and calcium bis(di-2-ethylhexyl phosphate) as sensors. The coated-wire electrodes (CWEs) have the same components as the conven- tional ISEs; however, they are prepared without internal filling solution.Instead, a conducting substrate is directly coated with a membrane (usually PVC based), that is responsive to the ion measured. This conductor, commonly of metallic or graphite type, can be of any convenient geometric shape.19 Graphite has been successfully used in the prepara- tion of ISEs coated with different chemically modified polymer films .20 In the selection of conducting substrates, the main criterion is the possibility of electrically connecting them to a measuring device. It is practical for the substrate to take the form of a self-supporting sheet or film, or for it to be coated onto a non-conducting base.21 This type of relatively inexpensive ISE can be constructed easily and, as such, they are the subject of increasing interest to industry.22.23 In the studies described here, several potentiometric sensors for calcium, based on the calcium salt of bis(2- ethylhexy1)phosphoric acid, have been developed and investi- gated.The sensor alone was dissolved in a mixture of different plasticizers, organic solvent and PVC. A graphite-poly- (tetrafluoroethylene) (PTFE) electrode of the R8iiEka type, a graphite-polyethylene electrode, and platinum and silver electrodes have all been coated with such a mixture. The response of these CWEs was examined in Ca2+ solution in order to select the optimum plasticizer and conducting substrate. Experimental Reagents The reagents and materials used were obtained from the following sources: bis(2-ethylhexyl) phosphate (HDEHP) for synthesis from Merck-Schuchardt (Darmstadt, Germany); o-NPOE, PVC of high relative molecular mass and tetra- hydrofuran (THF) from Fluka (Buchs, Switzerland); tricresyl phosphate (TCP) and dioctyl phthalate (DOP) from Bayer (Leverkusen, Germany); tributoxyethyl phosphate (TBEP) from Lehman & Vossel (Hamburg, Germany); and 2-ethyl- hexyl diphenyl phosphate (Santicizer 141) (EHDPP) , isodecyl diphenyl phosphate (Santicizer 148) (IDDP) and octyl diphenyl phosphate (Santicizer 1439) (ODPP) from Monsanto1324 ANALYST, AUGUST 1992, VOL.117 (Brussels, Belgium). Analytical-reagent grade chlorides of lithium, sodium, potassium, ammonium, magnesium, barium, calcium, zinc and cobalt were supplied by BDH (now Merck) (Poole, Dorset, UK). Apparatus Electrode potential measurements were performed with a digital pH millivoltmeter (Iskra MA 5740) (Iskra, Kranj, Slovenia), with an accuracy to within k O .l mV, together with an Iskra TZ 4200 recorder. The electrodes used for investigations were as follows: graphite-PTFE Universal R8iiCka Selectrode F 3012 (RS), Radiometer (Copenhagen, Denmark) ; graphi te-pol ye th ylene electrode GPE 201.801 (GPE), Faculty of Technology and Metallurgy, Belgrade, Y~goslavia;2~ platinum metallic elec- trode HEP 0201 (Pt), Iskra; silver metallic electrode P 4011 (Ag), Radiometer; and reference calomel electrode HEK 0301, Iskra. PVC Membrane-coated Electrodes Synthesis of sensor A potentiometric sensor for calcium, based on the calcium salt of bis(2-ethylhexy1)phosphoric acid, was synthesized by dis- solving about 10 g of bis(2-ethylhexy1)phosphoric acid ester in 100 cm3 of 90% ethanol, heating, with stirring, and adding an excess of solid calcium carbonate when the temperature reached 75 "C.Membrane preparation The electrode coating mixture (0.6 g) was prepared by dissolving 30.6% PVC, 63.4% plasticizer (TBEP, o-NPOE, DOP and TCP were used) and 6.0% sensor as active electrode material in approximately 6 cm3 of THF. The membrane mixtures (0.6 g) were prepared by using a plasticizer of the Santicizer type (IDDP, ODPP or EHDPP) containing 62.2% PVC, 31.6% plasticizer and 6.2% sensor, dissolved in approximately 8 cm3 of THF. Electrode preparation All the electrodes used as conducting substrates (RS, GPE, Pt and Ag) were dipped in the membrane coating mixture and, after evaporation of the solvent, the procedure was repeated once.Conditioning of the electrode The electrode was initially conditioned by soaking in a 0.1 mol dm-3 calcium chloride solution for 20 min or in a 1 X 10-5 rnol dm-3 calcium chloride solution for 1 h. Between measurements, the electrode was stored in air and recondi- tioned immediately before use by soaking for 20 min in a 0.1 rnol dm-3 calcium chloride solution. Analytical Characterization of the Electrodes Calibration The electrode potential of the electrodes was measured at 25 k 0.1 "C in calcium chloride solutions covering the range 1 x 10-1-1 x 10-6 mol dm-3. The solutions of concentrations lower than 1 X 10-2 rnol dm-3 were always freshly prepared, immediately before measurement, and were stirred during measurement. Selectivity coefficient measurement The selectivity of the electrodes was investigated in solutions of several inorganic cations (Li+, Na+, K+, NH4+, Mg2+, Ba2+, Sr2+, Cu2+, Co2+ and Zn2+) by using the separate solution method described elsewhere.25 Response time measurement In the investigations described, the response time was measured by the injection method,26 based on inducing a rapid change in an intensively stirred 1 X 10-4 rnol dm-3 calcium chloride solution by injecting a more concentrated calcium solution (1 x 10-3 mol dm-3).The influence of the stirring effect in the solutions and the possible formation of liquid- junction potentials between solutions, in addition to some theoretical considerations resulting therefrom, were not considered. Results and Discussion Influence of the Nature of the Plasticizer and Conducting Substrate on the Electrode Response Four electrodes used as conducting substrates (RS, GPE, Pt and Ag) and seven plasticizers (TBEP, o-NPOE, DOP, TCP, IDDP, ODPP and EHDPP) were varied in order to investi- gate the influence of both conductor and plasticizer type on the response characteristics of the electrodes.The slope values (S), correlation coefficients ( r ) and intercepts ( y ) were calculated for the linear region of the calibration graphs, using five series of measurements, by means of linear regression analysis. GraphitePTFE Electrode of the R&iii.ka Type Results of the influence of different plasticizers on the electrode response are shown in Table 1, and the calibration graphs in Fig.1. Membranes containing TCP, DOP, TBEP and o-NPOE as plasticizers (plasticizer : PVC ratio 63 : 31) were transparent and plastic and had lifetimes of 3-4 months. It should be emphasized that membranes with Santicizer-type plasticizer (ODPP, IDPP and EHDPP) had a plasticizer : PVC ratio of approximately 30 : 60, because the opposite propor- tion gave very soft, easily stripped off membranes, which soon became white. Table 1 Characteristics of RSiiEka-type graphite-PTFE electrodes Parameter Membrane composition Interval of Plasticizer (%) Sensor (%) PVC (%) decade-' coefficient ( Y ) Intercept (y) mol dm-3 TCP, 62.2 6.2 31.6 28.65 0.9988 530.60 1 x 10-1-1 x 10-4 TBEP, 63.2 6.3 30.5 29.88 0.9999 496.70 1 x 10-1-1 x 10-5 ODPP, 31.6 6.2 62.2 29.05 0.9999 477.70 1 x 10-3-1 x 10-5 U-NPOE, 63.2 6.3 30.5 30.03 0.9999 445.10 1 x 10-1-1 x 10-4 DOP, 63.3 6.3 30.4 31.50 0.9998 436.00 1 x 10-2-1 x 10-5 Slope/mV Correlation linear response/ANALYST, AUGUST 1992, VOL.117 1325 From this type of plasticizer, only ODPP gave a satisfactory, stable Nernstian response, but in a restricted calibration range (1 x 10-3-1 X 10-5 mol dm-3). The other two branched-chain Santicizers (IDPP and EHDPP) were eliminated because of extremely low and erratic response (slopes of less than 10 mV decade-'). A high PVC content in the polymeric membrane probably caused the loss of plasticizer into the sample solution and hence the marked increase of electrical membrane resistance. Among the plasticizers investigated for the graphite-PTFE conducting substrate and with calcium bis(2-ethylhexyl) phosphate as sensor, the best analytical performances were shown by the electrode containing TBEP, an aliphatic phosphoric acid ester.The most common PVC plasticizers, such as aromatic esters of phthalic and phosphoric acids, afforded much poorer and even unacceptable results. GraphitePolyethylene Electrode (GPE) Only three (TCP, DOP and TBEP) of the seven plasticizers investigated exhibited a linear response to Ca2+. The other plasticizers exhibited no useful analytical data. Membrane composition and statistical data are shown in Table 2, and the corresponding calibration graphs in Fig. 2. 500 450 w 0 v, v) $ 400 > E G 350 300 -/./< I I I I 6 5 4 3 2 1 -Log([Ca2+]/rnol drn -3) Fig. 1 Influence of plasticizer nature on the response characteristics of Rfiiitka-type graphite-PTFE electrodes: A, TCP; B, TBEP; C, ODPP; D, o-NPOE; and E, DOP (values shown on graphs are in mV) It can be seen that the use of TBEP as a plasticizer was unacceptable €or the GPE (slope too low).Of the two remaining plasticizers (TCP and DOP), DOP is to be preferred (higher correlation coefficient and wider range of linear response) in spite of a slightly sub-Nernstian response, although TCP indicated a nearly theoretical response. The membrane compositions were the same as those used with the GPE, but the different analytical characteristics of the electrode could only be assigned to the different natures of the conducting substrates. Platinum Electrode Of all the plasticizers investigated, the best response to Ca2+ was exhibited by a PVC membrane-coated Pt electrode containing ODPP.The other plasticizers exhibited poorer analytical performances (erratic and sub-Nernstian response, lower correlation coefficients and narrowed interval of linear response). Membrane composition and statistical data are shown in Table 3, while the corresponding calibration graphs are shown in Fig. 3. Platinum, the most frequently used conducting substrate for CWEs, appeared inferior compared with the graphite-PTFE conducting substrate. It is known that the substrate nature could affect the standard potential, but the analytical behav- iour of the electrode remained unchanged.27 This has been proved for metallic substrates, but not for graphite-polymer substrates. 500 450 0 m v, 2 b > 6 400 / 128.70 / B I / c 1 / /' 6 5 -~og([Ca2+]/rnol 4 3 d w 3 ) 2 1 Fig.2 Influence of plasticizer nature on the response characteristics of graphite-PE electrodes of the RfiiiEka-type: A, TCP; B, DOP; and C, TBEP (values shown on graphs are in mV) Table 2 Characteristics of graphite-polyethylene electrodes of the Rfiiitka type Parameter Membrane composition Interval of Plasticizer (YO) Sensor (YO) PVC (%) decade-' coefficient ( r ) Intercept (y) mol dm-3 Slope/mV Correlation linear response/ TCP, 62.2 6.2 31.6 28.70 0.9968 573.20 1 x 10-2-1 x 10-5 DOP, 63.3 6.3 30.4 26.00 0.9992 495.40 1 x 10-1-1 x 10-5 TBEP, 63.2 6.3 30.5 9.06 0.9993 354.20 1 x 10-'-1 x 10-61326 ANALYST, AUGUST 1992, VOL. 117 Table 3 Characteristics of the platinum electrodes .-.-.Parameter Membrane composition Plasticizer (%) Sensor (%') PVC (%) ODPP, 31.6 6.2 62.2 DOP, 63.2 6.3 30.5 IDDP, 31.7 6.2 62.1 TBEP, 63.2 6.3 30.5 U-NPOE, 63.2 6.3 30.5 EHDPP, 31.7 6.2 62.1 Slope/mV decade- 1 29.35 34.20 11.80 13.55 18.84 20.40 Correlation coefficient ( r ) 0.9999 0.9975 0.9962 0.9622 0.9979 0.9753 Interval of linear response/ Intercept (y) rnol dm-3 631.80 575.20 396.90 402.60 1 x 10-2-1 x 10-5 1 x 10-2-1 x 10-5 1 x 10-1-1 x 10-5 1 x 10-1-1 x 10-4 1 x 10-2-1 x 10-4 306.90 288.70 1 x 10-1-1 x 10-6 500 450 W 0 3 ' 350 > E ci 400 s t LLI a 6 5 4 3 2 1 - Log( [Ca* + l/mol d m -3) Fig. 3 Influence of plasticizer nature on the response characteristics of Pt-CW calcium ISEs: A, ODPP; B, DOP; C, IDDP; D, TBEP; E, o-NPOE; and F, EHDPP (values shown on graphs are in mV) Table 4 Selectivity coefficient (log kEt, M) data for two different plasticizer-based calcium ISEs Plasticizer Interferent, M t Li + Na+ K+ NH4+ Mg2+ Ba2+ Sr2+ cu2+ co2+ Zn2+ Y-NPOE -3.80 -4.10 -3.50 -2.90 -2.10 -1.50 - 1.60 +0.09 -0.88 - 1.60 TBEP -2.80 -2.80 -3.10 -2.20 -1.60 -0.56 -1.70 -0.43 -0.63 -1.40 Silver Electrode The PVC membrane-coated Ag electrode indicated a negative undefined slope, caused probably by the C1- originating from calcium chloride solutions.Determination of Selectivity Coefficients The influence of some interfering ions (Li+ , Na+, K+, NH4+, Mg2+, Ba2+, Sr2+, Cu2+, Co2+ and Zn2+) was studied for graphite-PWE electrodes of the RfiiiCka type (RS), with use of two different plasticizers (TBEP and o-NPOE). The 10-4 - 10-3 rnol dm-3 -----c mol dm-3 A .-.-.-.-.-.- Est €95 Est B €95 I./ t95 t - Fig. 4 Dynamic response time of calcium ISEs based on two different plasticizers: A, o-NPOE; and B, TBEP. E,, = stable electrode potential value (after attaining equilibria); Eg5 = 95% value of electrode potential change; and t95 = time required to attain Egg value calculated selectivity coefficients, expressed as log kP!:M, are presented in Table 4. The results obtained showed only slight interference caused by Li+, Na+, K+ and NH4+. The selectivity coefficient for Mg2+ was satisfactory in relation to most of the calcium ISEs described in the literature,2J* where Mg2+ was defined as a serious interferent . Therefore, the disturbances from Mg2+ need not be taken into consideration in measurements on physiological solutions, where the Mg2+ content is usually considerably lower than that of Ca2+.In solutions where the Mg2+ concentration is considerably higher than that of Ca2+, the former ions should be removed or masked by suitable agents. Copper and cobalt ions seriously interfered with the determination and must be removed or masked before measurements. It can also be concluded that the use of o-NPOE as plasticizer improved the selectivity of the elec- trode. Response Time Measurement The dynamic response time was measured for a PVC membrane-coated graphite-PTFE electrode and two plasticiz- ers: o-NPOE and TBEP. For Ca2+ concentration changes from 1 x 10-4 to 1 X 10-3 mol dm-3, the results were as follows: t95 = 0.8 s for the o-NPOE-based membrane, and t95 = 0.5 s for the TBEP-based membrane (Fig.4). Conclusions The influence of seven plasticizers (o-NPOE, TCP, TBEP, DOP, EHDPP, IDDP and ODPP) and four conductingANALYST, AUGUST 1992, VOL. 117 1327 substrates (RS, GPE, Pt and Ag) on the analytical characteris- tics of PVC membrane coated wire calcium ISEs was investigated. Calcium bis(2-ethylhexyl) phosphate served as a sensing material and THF was used as a solvent. The best performances were obtained with a graphite-PTFE electrode of the RGiiEka type as the conducting substrate and with TBEP and o-NPOE as plasticizers. The type and amount of plasticizer were significant for the membrane properties, especially electrical membrane resistance, and therefore for the electrode response. The best analytical characteristics were exhibited by a membrane of the following composition: 30-33% of PVC, 5.9-6.9% of sensor and 60.1-64.1% of plasticizer.The electrode based on this membrane responded linearly to Ca2+ concentrations in the range 1 x 10-1-1 x 10-5 mol dm-3, and had the theoretical slope value. A stable electrode response (f0.5 mV) is concentration dependent, and was achieved within 4-5 min for the lowest concentrations. The electrode lifetime was 3-4 months. The chemical interference from Li+, Na+, K+ and NH4+ was negligible. Magnesium ions interfered only slightly, which is a significant advantage of the electrode described. Copper and cobalt ions interfered seriously and should be removed before measurement. The dynamic response time of the electrode was 0.5 s.The investigation indicated that the analytical performance of the electrodes studied was strongly dependent on the selection of an adequate plasticizer-conducting substrate combination. References Solsky, R. L., Anal. Chem., 1990, 62, 23R. Moody, G. J., Nassory, N. S., and Thomas, J. D. R., Analyst, 1978, 103, 68. Lakshminarayanaiah, N., J. Membr. Sci., 1981,8,255. Didina, S. E., Grekovich, A. L., Materova, E. A., and Bychkov, A. S., Zh. Anal. Khim., 1984, 11,2031. Cao, G., Lee, H., Lynch, W. M., and Mallouk, T. E., Inorg. Chem., 1988, 27, 2781. Attivat, A. S., Christian, G. D., Hallman, J. L., and Bartsch, R. A., Talanta, 1988, 35, 789. 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Tanaka, T., Hiiro, K., and Kawahara, A., Anal. Lett., 1974, 7, 173.Sugawara, M., Nagasava, S., and Ohashi, N., J. Electroanal. Chem. Interfacial Electrochem., 1984, 176, 183. Okada, T., Sugihara, H., and Hiratani, K., Anal. Chim. Acta, 1986, 186, 307. Ammann, D., Guggi, M., Pretsch, E., and Simon, W., Anal. Lett., 1975,8,709. Wuhrmann, P., Thoma, A. P., and Simon, W., Chimia, 1973, 27. 637. Jaber, A. M. Y., Moody, G. J., and Thomas, J. D. R., Analyst, 1977, 102, 943. Osipov, V. V., Mikhailov, V. A., Sergeeva, V. A., and Novikova, I. L., Izv. Sib. Otd. Akad. Nauk SSSR, Ser. Khim. Nuuk, 1982, 5 , 46. Khalil, S. A. H., Moody, G. J., and Thomas, J. D. R., Analyst, 1985, 110, 353. Hobby, P. C., Moody, G. J., and Thomas, J . D. R., Analyst, 1983, 108, 581. Craggs, A., Delduca, P. G., Keil, L., Moody, G. J., and Thomas, J. D. R., J. Inorg. Nucl. Chem., 1978, 40, 1943. Keil, L., Moody, G. J., and Thomas, J. D. R., Anal. Chim. Acta, 1978, 96, 171. Craggs, A., Keil, L., Moody, G. J., and Thomas, J. D. R., Talanta, 1975, 22, 907. Cunningham, L., and Freiser, H., Anal. Chim. Acta, 1986,180, 271. Lawton, R. S., and Yacynych, A. M., Anal. Chim. Acta, 1984, 160, 149. James, H., Carmack, G., and Freiser, H., Anal. Chem., 1972, 44, 856. Dobson, J. V., Br. Pat., 1481510, 1977. Freiser, H . , Br. Pat., 1375785, 1974. Jovanovic, V. M., and Jovanovic, M. S., Presented at the 7th Yugoslav Congress on Chemistry and Chemical Technology, Novi Sad, 1983, Paper V-33. Hoke, S. H., Collins, A. G., and Reynolds, C. A., Anal. Chem., 1979,51, 859. Toth, K . , Cavaler, J., and Pungor, E., Anal. Chim. Acta, 1971, 57, 131. Freiser, H . , Pure Appl. Chem., 1987,59, 539. Paper 1 l03237B Received June 28, 1991 Accepted March 25, 1992

ISSN:0003-2654    

DOI:10.1039/AN9921701323

出版商:RSC    

年代:1992

数据来源: RSC