ISSN:0003-2654    

DOI:10.1039/AN98813FX001

出版商:RSC    

年代:1988

数据来源: RSC

摘要:

ANALYST, JANUARY 1988, VOL. 113 3 Polarographic Adsorption Analysis and Tensammetry: Toys or Tools For Day-to-day Routine Analysis?* Pierre M. Bersier Central Analytical Laboratory, Ciba-Geigy Ltd., CH-4000 Bade, Switzerland Jacques Bersier Central Research Laboratories, Ciba-Geig y Ltd., CH-4000 Bade, Switzerland Introduction. Usefulness of Polarographic Adsorption Analysis and Tensammetry This review covers mainly the more recent literature on the application of polarographic adsorption analysis (suppression of polarographic maxima) and on tensammetry in its various forms from 1974 to 1987. A comprehensive list of substances studied and assayed by tensammetry before 1974, including short notes on the techniques used and analytical results obtained, has been given by Jehring.1 Perusal of the literature reveals highly contradictory opin- ions on the practical use of tensammetry.For example, Bagdasarov et aZ.2 stated in 1986: “In the last years it has been convincingly demonstrated that tensammetry is a simple and rapid methodl~3~4 that can be used to monitor the production and the content of surfactants in waste water. Tensammetry has been successfully used for the analysis of mixtures of such substances on the basis of their differential adsorptive behaviour.5” However, Bond and Jones6 in 1983 arrived at a different conclusion: “Whilst tensammetric response has been advanced for analytical purposes, it has been noted that mutual interference effects exist when more than one adsorb- ing species is present in solution. This is commonly the case, so that practical application of tensammetry has been very limited. Despite some advances tending to minimise such effects,’ interference effects are still the main drawback hindering development of tensammetry as a viable analytical technique.” The problems inherent in tensammetry are, according to these authors, best illustrated by the determina- tion by Cheney et aZ.8 of ethanol in spirit drinks.For vodka, the results are excellent, for gin the values are low and for blended whisky the method fails (Fig. 1). It is apparent that the results deteriorate as the matrix becomes more complex. Finally, there is the opinion of non-specialists, looking for an analytical method to determine the concentration of dispersants in sea water contaminated with oil, both in field and laboratory situations.Tensammetry was one of the methods investigated. The conclusions of Haynes et aZ.9 were: “Our attempt at developing such a detection method was unsuccessful because of (1) limited information on the use of tensammetry, (2) the complexity of the equipment and the sophistication of the technique and (3) the detection limits we experienced. Tensammetry may be used in limited analysis of surfactants, but it is too complex and sophisticated for routine use. ” This statement, if true, would certainly discourage a newcomer from attempting to use tensammetry in day-to-day routine work. The purpose of this review is to formulate answers to the statements by Haynes et aZ.9 and, hence, to the more general question: polarographic adsorption analysis and tensammetry: toys or tools for day-to-day routine analysis? * Presented at the International Symposium on Electroanalysis and Sensors in Biomedical, Environmental and Industrial Sciences, Cardiff, UK, 6-9 April, 1987.Principles and Features of Polarographic Adsorption, Analysis and Tensammetry Use of Tensammetry “Limited information on the use of tensammetry” is the first reason advanced by Haynes et aZ.9 for their failure with tensammetry. The use of polarography to determine non- reducible, non-oxidisable, surface-active organic molecules, based on measurement of adsorption phenomena at the mercury electrode, is well established. Such analysis can be based on the suppression of polarographic maxima of the first and second kind (later called polarographic adsorption analysis), effects on limiting currents and half-wave potentials of other electroactive species present and, finally, depression of the residual current in alternating current (ax.), square- wave and pulse tensammetry, etc.Table 1 summarises important landmarks of polarographic adsorption analysis and tensammetry. Gouy 10 first studied the adsorption of surface-active organic compounds at the mercury - electrode interface by the method of measuring the interfacial tension as a function of the applied potential. Polarography, on the other hand, began with the investigation by Heyrovsky of the anomalous electrocapillary phenomena described by Kucera.11912J8J9 Kucera observed that the interfacial tension at the mercury electrode in aqueous Pure ethanol Whisky I c -0.5 -1.0 -1.5 Unblended whisky b -0.5 -1.0 -1.5 E N vs.SCE Fig. 1. A.c. tensammograms of pure ethanol and spirit drinks. Broken line is base current (1 M KCl). Tensammograms recorded in the authors’ laboratory, according to method given in reference 84 ANALYST, JANUARY 1988, VOL. 113 solutions open to air, measured by the drop-weight method, was anomalously high on the anodic branch of the electro- capillary curve (Fig. 2) then dropped to the normal curve. In fact, the increase is only a few per cent. Heyrovsky and Simunekl3 found a similar anomalous flow of current in this region (see Fig. 2) owing to what we call the oxygen maximum.29 The indirect determination of surfactants by making use of their ability to suppress unwanted maxima in conventional direct current polarographic methods was first applied and introduced as adsorption analysis by Gosman and Hey- rovskyl4 in a study of the analysis of petroleum and its distillates by means of the dropping-mercury electrode, listed as Project 39 of the American Petroleum Institute Research (cf., Fig.3). The difference in height of the maxima gives a measure of surfactant concentration [in percentage terms (cf., reference 30)] in solution. Polarographic maxima of the second kind were discovered in 1940,16 and their practical use was first described by Kryukova ,17343 In 1952 Breyer and Hacobianlg-20 and Doss and co-work- em21922 found, independently, that the alternating current polarograms of surface-active substances [Fig. 4(b)] resem- bled the differential capacitance curves [Fig. 4(a)] obtained by Proskurnin and Frumkin in 1935 using the method of comparison with a standard capacitance.15 Variations in surfactant concentration affect the depth of the depression in addition to the height and the position of the peaks.Both effects may be exploited as analytical signals for the determi- nation of surfactants in solution. Breyer and Hacobian proposed for this new technique the term tensammetry, from the words surface “tension” and Table 1. Important steps in the development of polarographic adsorption analysis and tensammetry Year Event Reference 1903 Studies on adsorption of organic compounds at the mercury - electrolyte interface phenomenon of the first kind 1929 Description of the Kucera 1929 Description of polarographic maxima 1931 First analytical application of suppression of the oxygen maximum (polarographic adsorption analysis) 1935 Measurement of the differential capacity potential curve of octyl alcohol at the dropping-mercury electrode 1940 Detection of the maximum of the 1952 Detection of the a.c.tensammetric second kind technique 1955 Detection of the square-wave tensammetric technique 1966 Analytical application of square-wave tensammetry 1973 Introduction of Kalousek commutator technique in analysis of surfactants 1979 Analytical application of differential- pulse tensammetry Gouylo Kucerall-12 Heyrovsky and Simunek13 Gosman and Heyrovsky14 Proskurnin and Frumkin15 Kryukova1618 Breyer and Doss and Hacobian;lgJ” co-workers2l.Z Barker and Gardner23 Jehring and co-workers24.~ Cosovic and Branica26 Canterford and Taylor27 “ammetry.” The term “tensammetry,” which, although orig- inally a sinusoidal ax.polarographic technique, is not limited to that technique but has been applied to other transient techniques, such as Kalousek-type wave form, square-wave, pulse and chronopotentiometry , has been criticised by Frum- kin and Damaskin32 and others. Frumkin and Damaskin32 proposed the term “curves of nonfaradaic admittance,” whereas the maxima on these curves at the desorption potentials should be called adsorption - desorption peaks or maxima; see the remarks by Breyer and Hacobian quoted in reference 32, p. 45. At about the same time Barker and Gardnerm were studying adsorption - desorption tensammetric processes of octyl alcohol using square-wave polarography.At the end of the 1950s Barker also used radiofrequency polarography.33 The use of double-layer capacity measurements for the evaluation of purity of solutions was first proposed by Barker .33 In 1965 Jehring and co-workers24J4 first applied the square-wave tensammetric response for characterisation of the approximate average molecular masses of polymers. The poIarographic method of discontinuously changed potential using the Kalousek switch (Kalousek commutator) for the determination of surfactants in aqueous solutions was introduced by Cosovic and Branica26 in 1973. Canterford and Taylorn were the first to apply pulse tensammetry for the 0- 0.2 0.4 0.6 0.8 1.0 1.2 E.m.f.N Polarising e.m.f.N Fig. 2.(a) Electrocapillary curves and (b) polaro raphic curves according to Heyrovsky and Simunek.13 Curves (A? measured in 0.01~KCl in the absence of dissolved oxygen (“in hydrogen atmosphere”). Curves (B) measured in O.O~MKCI in presence of dissolved oxygen (solution “open to air”) I 1 1 T T T T I Potential - Fig. 3. Oxygen maxima measured in pure 0.001 M KCl (curve 1) and after addition of (2) 0.1 ml, (3) 0.6 ml, (4) 2.5 ml and (5) 5.0ml of a 0.001 M KC1 solution, which were obtained after 50 ml of 0.001 M KCl were shaken with 5.0ml of commercial gasoline of Russian origin. After Gosman and Heyrovsky. 14 Reproduced from reference 14 with permission. Reference 14 was originally presented at the 1931 Spring Meeting of The Electrochemical Society Inc., held in Birmingham, AlabamaANALYST, JANUARY 1988, VOL.113 5 80 - (Y 70- 60- 3 5 0 - 40 - 30 - 20 10 - LL m c u - O L 0:5 If0 115 PotentialN vs. SCE ~~~~ Potential-, Fig. 4. Comparison of (a) the differential capacity of the electrolytic double layer between mercury and a solution of 1 N Na2S04 (curve 1) and 1 ”a2S04 saturated with octyl alcohol (curve 2) (according to reference 15), and (b) with a.c. tensammetric waves of 1 N Na2S04 (curve 1) and 1 N Na2S04 saturated with octyl alcohol (curve 2). Working electrode, dropping mercury electrode; drop time, 0.8 s; applied alternating voltage, 15 mV (r.m.s.) at 75 Hz. Curves in (b) measured in the authors’ laboratory determination of surfactants in 1979. In differential-pulse polarography the main portion of the capacitive contribution to the current is removed but, according to Myers and Osteryoung35 the differential-pulse voltammogram in the absence of the faradaic reaction is essentially a differential capacity curve.The effects of adsorption of organic compounds on metal electrodes (surfaces) have been studied extensively by numer- ous workers including Damaskin et al.36 The theory of polarographic maxima, so called streaming maxima, has been discussed in depth, e.g., by Bauer,29 and in various electroanalytical te~tbooks.28~31937 Polarographic max- ima are always related to the appearance and subsequent disappearance of tangential motions (streams) on the surface maxima. A survey of the older literature on practical applications of polarographic adsorption analysis is to be found in reference 28; the more recent applications are discussed under Practical Applications of Polarographic Adsorption Analysis and Tensammetry.The theoretical aspects and applications of tensammetry were discussed in detail by Breyer and Bauer38 in 1963 and more recently in textbooks on electroanalytical chem- istry.aJ7J9-41 The only really comprehensive book on tensammetry is that by Jehring1(1974), in German; hence the readership of this excellent book is necessarily restricted. Jehring’s work is extremely valuable to the analytical chemist because the detailed characteristics of the responses of tensammetric wave~1.24J5~42 have been reported under a variety of conditions for both fundamental and second harm0nica.c. polarography . The square-wave response is also covered in Jehring’s work.Jehring has also shown that the technique is suitable for the quantitative theoretical interpret- ation of adsorption phenomena at electrodes with sufficient accuracy in most instances. The method is particularly favourable in view of the small requirements for apparatus and time.1143 Examples are given in older reviews,44-46 and by Venkatesan,47 but no modern review article has been pub- lished so far. I of the liquid mercury electrode, hence the name streaming Complexity of Techniques “Complexity of equipment and sophisticated techniques” is the second reason advanced by Haynes et a1.9 for their failure with tensammetry. Table 2. Techniques proposed and applied to the determination of surface-active compounds Technique Reference Electrocapillarycurvemethod . .. . . . 48-52 D .c. polarography (polarographic adsorption analysis) Suppression of the maximum of the fist kind Suppression of the maximum of the second (02, NF, HgII, CdI, CoTII) . . . . . . See text kind . . . . . . . . . . . . . . 17 , 18,3 1 Oscillo?olarography . . . . . . . . . . 53,54 Linear sweep voltammetry . . . . . . . . 55 Cyclic voltammetry . . . . . . . . . . 56,57 Tensammetry A.c. tensammetry First harmonic . . . . . . . . . . See text Second harmonic . . . . . . . . . . 24,34,42,58,59,66 Single drop . . . . . . . . . . . . 60 Pulse tensammetry . . . . . . . . . . 64-69 Kalousek commutator technique . . . . . . See text Derivativechronopotentiometry . . . . . . 7G72 Tensammetric titration . . . . . . . . . . 73-79 Combination of HPLC and tensammetric Square-wave tensammetry .. . . . . . . 2>25,34,61-63 detection . . . . . . . . . . . . . . See text Polarography and tensammetry require a fair amount of training. As opposed to some other commonly used physico-chemical - instrumental methods, neither technique can be applied without a fair knowledge and understanding of what is happening and what is being measured. This is especially true in view of the proliferation of methods proposed over the past years in polarographic adsorption analysis and in tensammetry, as illustrated by Table 2. How can an unskilled, or even a skilled, user make a judicious choice as to which technique should be used under given circumstances? A comparative study of the behaviour of humic and fulvic acids (HA, FA), by using various polarographic techniques, e.g., d.c., ax.and differential-pulse polarography, the Kalousek technique and the decrease of the 0 2 maximum has been made by Cominoli et al.64 in order to find the best method which might be used for the qualitative and quantita- tive analysis of these substances. In this instance a.c. tensammetry and polarographic adsorption analysis (suppres- sion of the O2 maximum) proved to be the best techniques for the measurement and characterisation of FA. Both methods are sensitive; the limit of detection of a.c. tensammetry is 2mgl-1, whereas for 0 2 suppression it is 0.5mg1-1 in air-saturated solutions and 1 mg 1-1 in oxygen-saturated solu- tions. In both instances an adsorption parameter may be defined which is characteristic of the adsorbing compound. As the different techniques make use of different time scales, they are sensitive to different phenomena and can thus help to obtain a better insight into the behaviour of the surface-active substances at the mercury interface.Adam et ~ 1 . 8 0 have studied the adsorption characteristics of 1-phenylpyrazolidin-3-one on metal electrodes using linear sweep voltammetry, electrocapllary measurements, capaci- tance measurements and chronocoulometric techniques. Characteristics for Evaluating an Andytkd Technique Characteristics used to evaluate an analytical technique include applicability, sensitivity, selectivity and specificity, and these are discussed in the following sections. Applicability Many organic molecules and ions are accessible to determina- tion by polarographic adsomon analysis and tensammetry,6 ANALYST, JANUARY 1988, VOL.113 including alcohols, amines, carboxylic acids, detergents, dyestuffs, esters, ethers, enzymes, hererocyclics, hydrocar- bons, insecticides, ketones, organic polymers, phenols, poly- nucleotides, proteins, sugars and terpenes.81 Despite the large number of species known to adsorb on a mercury electrode36 and thus give tensammetric peaks,82 tensammetry is little used. Out of about 7000 polarographic papers reviewed by Chemical Abstracts from 1965 to 1986, less than 200 are tensammetric. In the most recent issue of the Surfactant Science Series, Volume 18 (1987),83 out of 34 pages dedicated to physico-chemical analytical methods, only three- quarters of a page is devoted to polarographic adsorption analysis and tensammetric methods, and only eight out of the 149 quoted references are polarographic or tensammetric.Perhaps one reason why tensammetry is not as widely used as might be expected may be a misconception arising from early published work. The idea prevailed that, for example, for a mixture of two surfactants, only that substance which was more strongly adsorbed produced a tensammetric peak,lY so less strongly adsorbed components of a mixture could not be determined. However, as illustrated by Jehring,25 gradual replacement of one surfactant by another may be observed with tensammetry. That is, under certain conditions both constituents of a binary mixture give peaks on the same tensammogram.The problem of the determination of mix- tures will be discussed in more detail under Selectivity and specificity. However, the reader who is interested in the potential applications of tensammetric techniques should be familiar with their variability towards medium effects before undertak- ing too many experiments.84 Further, tensammetric waves frequently behave differently to faradaic processes. For instance, peak heights are unlikely to be a linear function of concentration, and the peak potential (position) is usually a function of concentration. Electrolyte type and concentration may also significantly influence the wave shape and position far more than for faradaic processes. As stressed by Bond,84 extreme care and common sense must therefore be employed in using tensammetric waves in analytical work.Sensitivity Haynes et a1.9 also claimed “high, inefficient, detection limits” in tensammetry. Crisps5 noted that a method will be regarded as capable of trace analysis if it can detect concentrations of non-ionic surfactants of less than 10 mg 1-1 of water. Ranges of practical usefulness of polarographic adsorption analysis and tensammetric techniques are summarised in Fig. 5 . A distinc- tion must be made between sensitivities provided by the different available techniques (instrumental) and the increased sensitivities provided by a combination of accumula- tion procedures in situ with the direct current (d.c.), cyclic voltammetry (CV) ,sb oscillopolarographic,”4.”7 a.c. ,42388 Kalousek26 and pulse27>65 techniques, using slowly dropping mercury electrodes or a hanging mercury drop electrode (HMDE).As little as 1 p.p.m. of surfactant can be determined using slow dropping capillaries (>40 s) and the sensitivity could be improved to 0.1 p.p.m. The pre-concentration of surfactants at a hanging mercury drop electrode was usually carried out at a potential close to the potential of maximum adsorption ,7,4*36JX3‘)?90 yet the adsorption of a substance at the electrode depends strongly on the electrode potential. Batycka and Lukaszewskiy’ were the first to study in detail the influence of the pre-concentration potential on the cathodic tensammetric peak of polyethylene glycol (PEG) 200, 600, 1000, 1500, 4000, 6000, 9000 and 20000. The behaviour of 6-14, 10-107 10-14,18-14,18-10 and 18-6 alcohols (first number = number of C atoms, second number = average number of oxyethylene sub-units) was studied using this technique.The relationships between the peak height and the pre-concentration potential and the concentration of surfactants were examined.92 An additional advantage of tensammetry with accumulation on the HMDE in comparison with the classical technique is its ability to use the differentiating action of the pre-concentra- tion potential for the determination of components of mixtures.92-94 As a rule, pre-concentration of a surfactant at an HMDE proceeds under conditions that are far from those of an adsorption equilibrium state because of the slow transport of low concentrations of surfactants to the electrode surface.The achievement of adsorption equilibrium by that system actually means the end of surfactant pre-concentration at the HMDE.9’ A comparison made by Benarkiewicz and Kublikhh of the potential of a.c. fundamental, a.c. second harmonic and differential-pulse tensammetry in conjunction with pre-con- centration of the surfactant at an HMDE did not allow a choice of a best technique or best signal for the surfactants studied. Each technique has its advantages and disadvantages. Differential pulse tensammetry permitted the determination of as little as 2 pg dm-3 of PEG 4000 but it was useless in determining sodium dodecyl sulphate.66 The capacity current measurements at an uncharged and negatively charged electrode, performed either by the Kal- Suppression of maximum of 2nd kind Suppression on Hg” maximum Instrumental sensitivities Tens ti trat I I I I -I 6 10-5 10 4 10-3 10 2 10 ’ Concentratiodg I- Fig.5. Concentration ranges of polarographic adsorption analysis and tensammetric techniquesANALYST, JANUARY 1988, VOL. 113 7 ousek commutator technique or by ax. tensammetry, are more sensitive to hydrophobic and highly soluble lipid material than measurements by the polarographic maximum method. On the other hand, the polarographic maximum method is more sensitive to polysaccharides, such as algi- nate.95-96 As practical analysts we are well aware that performances as presented in the literature, and consequently in our list (Fig. 5), rarely indicate what to expect in real life (cf. reference 97). Selectivity and specificity A serious shortcoming in all electrochemical techniques in general is their poor selectivity when compared with separa- tion-based techniques such as chromatography.Adsorption analysis. The disadvantage of these indirect methods is that they are non-specific, as all surface-active agents have qualitatively similar effects on maxima. The limited selectivity of the suppression of the oxygen maximum thus makes effective separation a prerequisite for determina- tions in real situations. According to Linhart,30 the separation of anionic and cationic from non-ionic surfactants using an ion-exchange resin as a batch method has advantages over the more usual column operation. Unlike the sharp peak of the oxygen maximum, the polarographic maximum of HgII extends over a wide potential range (Fig.6), thus allowing rough characterisation of the type of surfactant. HgII maxima are also more easily reproducible owing to accurate addition of mercury(I1) salt. Tensammetry. The difficulties of determining surfactants in mixtures and interference problems are certainly among the main drawbacks hindering the development of tensammetry as a viable analytical technique. f C E c 5: T 2.0 4 I I 1 0 - 500 - 1000 - 1500 EImV vs. SCE Fig. 6. Typical pofarograms of model sea water [DIN (Deutsche Industrie Normen) 50900]. (A) Pure model sea water and (B) oil loadedpolluted model sea water, using (a) polar0 raphic maximum of mercury(I1) (1 x 10-3~HgCl~) added and fb) polarographic oxygen maximum (oxygen saturated solution). Polarograms measured in the authors’ laboratory Earlier adsorption studies 6f mixtures of surfactants on a mercury electrode19,25,36,98-102 were mostly on a fundamental basis.Analytical implications have been explored only more recently,3~91J03-~~5 for instance, by Canterford.106J07 Canter- ford laid special emphasis on the feasibility of using the method for determining a less strongly adsorbed compound in a mixture, so greatly widening the areas of application of the technique. The simultaneous determination of mixtures of anionic and non-ionic detergents is feasible using the Kalousek commuta- tor technique.103 Thus the charging-current potential curve for a mixture of 0.3p.p.m. sodium lauryl sulphate and 0.2 p.p.m. Triton X-114 consists of two desorption peaks. Improving selectivity.The selectivity can be improved by the following three methods. (i) Choosing an adequate drop time.1JO7J** Two sur- factants present simultaneously in a mixture can be determined singly by using different drop times. The condition required is that the two surfactants reach diffusion equilibrjum at sufficiently different rates. If the drop time is high, the surfactant prevalently adsorbed will be that which establishes diffusion equilibrium quickly and the tensammetric curve will be mainly due to its adsorption-desorption. Con- versely, at slow drop rates and assuming that the slower surfactant is more strongly adsorbed, which is generally so, the slower surfactant will be able to reach the surface of the electrode and will displace the faster one, so that the tensammetric process will be governed mainly by the slow surfactant. Exam- ples of the so-called “separation tensammetry” are shown in Fig. 7.(ii) Choice of the pre-concentration potential. This factor differentiates the behaviour of the com- ponents in the mixt~re.93~94J09 (iii) Separating chromatographically.110 More recently a combination of HPLC and tensammetry has been proposed.111-114 Whereas tensammetric detection techniques relying on changes of the base line (residual current) are unselective,~~O-114 the peak- finding procedure developed by Bond et al.6J15 can provide both quantitative and diagnostic informa- tion, based on the position of the peak. In order to use the tensammetric peak as an analytical signal which is dependent on concentration, it is necessary to scan a range of potentials continuously and to apply a computer-based search routine to locate the peak current.To discriminate between tensammetric and background faradaic processes in normal-pulse tensammetry, two current -I ElrnV vs. SCE Fig. 7. Influence of drop time on selectivity of the tensammetric technique. (a) A.c. tensammograms of 2.2 x -10-1 M butan-1-01 and (b) a.c. tensammograms of 2.2 X 10-1 M butan-1-01 in the presence of 2 X 10-6w 01 ethylene glycol 20000, measured at a drop time of (1) 4 s; (2) 2 s; 6) Js; and (4) 0.5 s in 0.1 N H2S04. A paratus: Polarecord 506NA 633 (Metrohm). Applied alternating voiage, 15 mV (r.m.s.) at 75%; scan rate, 2mVs-1. Curves measured in the authors’ laboratoryANALYST, JANUARY 1988, VOL. 113 measurements are made during the pulse, and the difference response is used as the analytical signal.116 Excellent results for the polyethylene and phenol ethoxylate systems have been reported.6 The use of normal-phase and non-aqueous molecular exclusion chromatography is essentially precluded, as organic solvents are commonly adsorbed themselves and produce tensammetric responses in some instances. Thus, the optimum solvent should have the minimum possible organic component concentration consistent with adequate separation of com- pounds. Gupta and Holleckll7 and Sharma118 showed that organic compounds producing tensammetric waves in an aqueous medium provide desorption peaks of decreased magnitude in non-aqueous media. Adsorption - desorption processes of ionic and non-ionic surfactants in methanolic medium have been discussed.118 A combination of HPLC and tensammetry, especially with microprocessor determination of species which exhibit a tensammetric response, provides new prospects for the determination of species which exhibit a tensammetric response, provided that adequate instrumentation is commer- cially available. The lack of selectivity can, on the other hand, also be of advantage, for instance, when the sum of surface-active pollutants is of interest. The determination of the total content of surface-active substances in water (i. e., its total interface activity) is extremely important in view of their toxic influence on both lower and higher organisms living in fresh and sea water. Thus, petroleum components dissolve in the cell membranes and may destroy them (cf., reference 119).Organic compounds with surface-active properties, both natural and pollutants, are concentrated by adsorption processes at phase boundaries of water with the atmosphere, solid particles and sediments and biota. Surface-active sub- stances modify the structure of the interboundary layers at natural interfaces and influence adsorption processes and scavenging mechanisms of other microconstituents (see refer- ence 120 and references cited therein). Tensammetry has also been recommended for continuous monitoring of surfactant concentrations from 1 to 20 mg 1-1.121 The feasibility of flow injection analysis in conjunction with tensammetric detection at a solid working electrode was assessed by Bos ef al.122 As the majority of surfactant types possess no electroactive group, conventional direct polarography is of little use. One exception is amine oxide type surfactants, exhibiting well formed reduction and oxidation waves. 123 For the determina- tion of non-reducible or non-oxidisable surfactants numerous indirect polarographic methods using functionalisation, com- plexation, etc., have been proposed (cf., reference 123). Practical Applications of Polarographic Adsorption Analysis and Tensammetry According to Haynes et a1.9 tensammetry is much too sophisticated for routine work. Conditions Required for the Practical Use of an Analytical Technique in the Industrial or Technical Laboratory The use of polarographic adsorption analysis and tensam- metry in the industrial or technical laboratory obviously depends on the instruments available, the availability of knowledgeable personnel and the applicability of the methods to real problems. The availability of inexpensive, very reliable polarographs offering the pulse mode explains the renewal of interest in polarography and voltammetry over the last 10-15 years.The rather limited use of tensammetry, on the other hand, is due partly to the lack of selectivity, and partly to the laclcof trained people (cf., reference 97) and adequate commercially avail- able instruments. The instruments most used at present which provide a.c. fundamental and second fundamental a.c. modes include those manufactured by Metrohm (Polarecord 506), PAR (174A and 174/50 ax. Polarographic Analyzer) and the Tacussel PRG 3. A commercial instrument permitting Kal- ousek tensammetry is the Metrohm Polarecord 506.The Metrohm 647, PAR 384B and AMEC 433 have a square-wave mode. Modifications of the sampling period and delay time of the pulse mode to obtain optimum differential-pulse tensam- metric sensitivity and re~olution27~65 are not directly available on the aforementioned instruments, although these modifica- tions are easily carried out on the PAR 174A Analyzer and probably also on any commercial instrument. Table 3 summarises some recent practical applications of adsorption analysis and tensammetry. Water Analysis Contamination of the aqueous environment by surfactants originating from the degradation and metabolism of organ- isms and man-made pollutants, such as detergents and mineral oils, is widespread.123 The determination of surfactants in aqueous solutions (which represent a large part of dissolved organic matter in natural water) is of great importance both in production control and in the monitoring of, for example, sewage water.The most widely used method is the two-phase titration procedure; another technique is direct photometric determi- nation in the organic phase. Because these methods are time consuming, much attention has been given to the development of other methods. Important contributions in this field are infrared spectrophotometry, chromatographic techniques such as HPLC,124 potentiometry , particularly with ion-selec- tive electrodes, and polarographic and tensammetric tech- niques.Recently, electrocapillary measurements combined with convective adsorption accumulation at the surface of a growing mercury drop were applied to the determination of petroleum components in aqueous solutions. The detection limit of the electrocapillary method was 10-20 pg 1-1 for diesel oil and up to five times higher for petroleum samples.125 Dextran fractions (molecular mass 10 000), tetrabutyl- ammonium perchlorate and analogous substances can be determined in water at concentrations of the order of 10-7 M. 126 Conventional electrocapillary measurements with Lippmann's electrometer and drop-time measurements are unsuitable for low concentration ranges because of the long time necessary to reach adsorption equilibrium. In the convective adsorption accumulation method the surfactant is transported to the electrode by convective, rather than by free diffusion, so that the time necessary to attain equilibrium coverage is substantially shortened.Thus, the sensitivity of electrocapillary measurements is enhanced by up to two orders of magnitude.49-50 Polarographic adsorption analysis and tensammetry have found their widest application in the determination of surfactants in natural waters, sea water, waste waters, effluents, etc. Some practical examples are shown in Fig. 8. The methods respond to biogenic surfactants present in natural, unpolluted and polluted sea water and to pollutants such as dissolved petroleum fractions and detergents. The methods are simple and direct, as they usually only include the addition of electrolyte and, if necessary, appro- priate dilution or accumulation to reach a measurable concentration range.Polarographic adsorption analysis (suppression of the O2 or Hg" maximum) proved to be the method of choice to study the nature, natural levels and evolution of organic matter in sea water95J19J27-130 and to monitor pollution with surface-active pollutants, such as petroleum131 and surfactants. 103 HunterANALYST, JANUARY 1988, VOL. 113 9 Table 3. Applications of tensammetry and polarographic adsorption analysis Applications in aqueous environments and media- Determination of surface-active substances in distilled water, Degradability tests of detergents Determination of production of surfactants by phytoplankton Determination of critical micelle concentration Assay of colour developing agents of the p-phenylenediamine type Assay of PEGS in processing solutions Analysis of mixtures Assay of novocaine in injection ampoules (0, maximum) Assay of indolemelanin (0, maximum) Determination of vitamin B1 (tensammetry) Determination of polynucleotides, amino acids (tensammetry) HPLC with tensammetric detection of bile acids, cardiac glycosides Elucidation of membrane mechanisms (tensammetry) Tensammetric assay of adjuvenants in plastics, antistatic additives, anionic cationic and non-ionic surfactants, surfactants in poly- ethylene foils and plates Tensammetric determination of degree of polymerisation and molecular mass of unfractionated poly(viny1 alcoho1)s natural waters, sea water, effluents, electrolytes Applications in the photographic field- Applications in the pharmaceutical and biologicallbiochemical fields- Applications to polymers and plastics- and Liss,132 using the suppression of the HgII maximum, found a correlation between surfactant activity values of estuarine water samples and dissolved organic carbon (DOC) concen- trations.They proposed to use such values as a simple measurement of DOC. Linhartso successfully applied the O2 maximum suppression to the study of the biodegradation of numerous surfactants. A.c. tensammetry has been applied to the determination of surface-active substances in sea water,%J28 river, ground and tap water,l2oJ30 and also to distilled water, potable water and untreated natural waters and supporting electrolytes, directly or combined with accumulation on the HMDE.88 The method permits the efficacy of different purification procedures to be compared quantitatively.The direct determination of poly- oxyethylenated alcohols with low oxyethylene content at p.p.m. concentrations is feasible.4 Cominoli et aZ.64 showed that a.c. tensammetry may be useful for the determination of humic and fu4vic ads--(HA, FA) in natural waters, provided that well controlled condi- tions are chosen. According to Quentel and co-workersl33-135 the disadvan- tage of commonly used tensammetric”-% and polarographic adsorption techniques64 is their lack of sensitivity for the HA and FA assay as compared with adsorption stripping tech- niques. Quentel et al. 133 give detection limits of 5 pg 1-1 of C with a 3-min collection time for different humic acids.Application of this technique to determine the depth profile of an oceanic sea water was also described. At pH 2 the cathodic stripping of sea water at the HMDE displays peaks due to humic and fulvic acids (EP = -0.520 V vs. Ag - AgCl, 0.7 M C1-). A peak at -0.260V is attributed to trace amounts of organic compounds (10-7 M) with the thiocarboxylic acid functional group (-GO-SH) .I34 Locatelli et aZ.67 used a.c. and differential pulse tensam- metry for the determination down to a few p.p.m. of dibutylamine (DNBA) in aqueous solutions and for the study of the adsorption of DNBA on activated carbon at pH 13.67968 Only neutral DNBA yields tensammetric peaks and hence selective determination of DNBA and DNBNA (dibutyl-N- nitrosoarnine) is possible by choosing suitable pH conditions.Recent results136 indicate that water pollution with petroleum can be determined by differential-pulse tensammetry, with a sensitivity depending on the quality of the impurity and on the composition of the supporting electrolyte. Good calibration 0 -600 -1000 EImV vs. SCE Fig. 8. A.c. tensammetric curves obtained for various water samples acidified with HC104 to a concentration of 0.1 moll-’ after accumula- tion. Curves measured in the authors’ laboratory using procedure proposed by Bednarkiewicz et d.a Ap aratus: Polarecord 506, in combination with multielectrode VA 63f (Metrohm) . Accumulation time, 30 s; vertical line defines accumulation potential (-0.600 V vs. SCE) and starting oint of recordm . (1) Doubly distilled water; (2) de-ionised water; 5) tap water; (47 rocess water; (5) snow water (snow collected in the plant area); ( ornamental fish pond water.Applied alternating voltage = 15 mV 4 r.m.s.) at 75 Hz graphs were obtained with diesel oil at concentrations from 30-50 ~11-1 to 0.5 mg 1-1 and with Saratov petroleum from 0.1-0.2 to 3.3 mg 1-1. The Kalousek commutator technique has been extensively used to evaluate the surfactant activity of polluted sea water,11g3137J% to determine anionic and non-ionic surfactants in effluents, e.g., in waste waters from hospitals,103 and surface-active substances in sea water surface microlayers.gs The study of petroleum fractions has shown that the technique responds only to higher levels of dissolved petroleum fracdons (0.02-100mgl-1) found only at pollution sources such as harbours and refinery effluents.131 Very good agreement was obtained90 between this method and the standard methylene blue active substances in biodegradability tests on linear alkylbenzene sulphonate and tetrapropylene alkylbenzene sulphonate.Unlike many others, the proposed method is applicable to the determination of different types of non-ionic surfactants, regardless of the number of ethylene oxide groups, including polyethylene glycols and biodegradation intermediates with a low degree of ethoxylation. 139 The Kalousek technique was also used to study surfactant production by marine phytoplankton.137 Determination of the Critical Micelle Concentration Although theoretically a weak concept,l@ the critical micelle concentration (CMCjthe threshold bulk concentration at which a micelle first appears-is of practical use in all areas of chemistry where micelles and adsorption play a role, e.g., in the detergent industry.Determination of the CMC is often a lengthy procedure.141 Doubts were recently expressed140 about the CMC values provided by the suppression of polarographic maxima by surfactantsl42-144 (where CMC is taken as the concentration at which the maximum is completely suppressed). According to Malik and nand144 the method has, nevertheless, its own significance, as it is one of the few methods that can provide CMC data in the presence of various electrolytes. Another method which can be used for this purpose is the electrocapil- lary curve method.144J45 The tensammetric determination of the CMC has been dealt with in numerous papers.1JaJG155 According to Muller andANALYST, JANUARY 1988, VOL.113 Doerflerl55 measurements under equilibrium conditions (measured at an HMDE) are a prerequisite for the quantita- tive evaluation of the CMC, a fact not adequately considered in some earlier papers. CMC values obtained at the HMDE are said to be in good agreement with literature values, as opposed to non-equilibrium values. 155 Thus, values obtained at the dropping mercury electrode are said to refer to non-equilibrium values, which is indicated by the formation of three peaks at more positive potentials on the tensammogram, whereas tensammograms (Cd - E curves) measured at the HMDE display only two peaks.155 The plots of the first capacity maximum potentials against the logarithm of the surfactant concentration under equilibrium conditions show two straight lines intersecting at the critical micelle concentration.Tensammograms of octaethylene glycol mono- dodecyl ether measured at a dropping and at a hanging mercury drop electrode are shown in Fig. 9. A semi-automated computer-aided staircase-tensammetric titration for accurate measurements of CMC has been described, 140 Applications in the Photographic Field A.c. and differential-pulse tensammetric methods are in routine use in Kodak (Australasia) processing laboratories for the determination of colour developing agents of the p- phenylenediamine type3 and polyethylene glycols (PEG 1540 and 4000)106.156J57 in photographic processing solutions.Although some of the species are present at greater molar concentrations than PEG, they do not interfere, as they are less strongly adsorbed on the mercury electrode. Indirect interference by species leached out from film materials could be overcome by simply diluting the assayed developer.3J57 As the processing solutions under consideration contain a num- ber of surface-active agents, nonspecific indirect methods based on suppression of polarographic maxima could not be used for the determination of PEG. By means of tensammetric techniques the adsorption isotherms of hydroquinone, some N-substituted p-phenylene- diamines, l-phenylpyrazolidin-3-one, sulphite and iodide were measured on a stationary gold electrode. The results are in agreement with observations of the catalytic effects of some additives in photographic development.158 Pharmaceutical and Biochemical Applications The suppression of the polarographic oxygen maximum was employed for the determination of novocainamide in injection EImV vs SCE Fig.9. Comparison of a.c. tensammetnc curves of octaethylene glycol monododecyl ether (2 X 10-4 M) in 0.5 M KCl, measured (A) at the dropping mercury electrode, dro time = 0.5 s, and (B) at the hangin mercur drop electrode. Appeed alternating voltage, 10 mV (r.m.s.7 at 7 5 d z . A paratus, Polarecord 506 in combination with Multielectrode VA 6!3 (Metrohm). Curves measured in the authors’ laboratory ampoules159 and for the determination of 0.4-4p.p.m. of indolemelanin, an important biopolymer for living organisms for which no other analytical method has been developed.160 A method based on the suppression of the C03+ maximum in saline (0.9% NaCl) was proposed for the determination of lung surfactants (LS).161 The minute amounts of LS needed for the measurements and the possibility of converting the values obtained electrochemically into dyne cm-1 or mN m-1 are advantages of the proposed method.162 The practical application of a.c. tensammetry is illustrated by the stability control of thiamine hydrochloride (vitamin Bl) during the sterilisation process.163 Kemula and Kutnerllz and Kutner et aZ.164 proposed tensammetric detection for the determination of the polaro- graphically non-reducible, non-oxidisable, surface-active bile acids after HPLC separation. The detection limit of conju- gated bile acids present in the HPLC eluate was reported to be 5 n M per 5 pl injected sample, which makes the technique competitive with refractometric detection.164 Cardiac glycosides behave as strongly surface-active com- pounds at the dropping-mercury electrode.De Jong et al. 113 have tested the potential of the tensammetric detection of digoxin, digitoxin, gitoxin and lanatoside A, B and C for the analysis of pharmaceutical formulations after HPLC separa- tion. As the detection limit of 1 X M (16 ng per injection) is comparable to UV detection at 220 nm, no advantage would be obtained by the tensammetric mode. Vetterl et al.69 have shown that differences in the behaviour of faradaic and tensammetric currents of polynucleotides can be exploited as a diagnostic criterion of differential-pulse polarographic peaks of nucleic acids.The tensammetric behaviour of nucleic acids has been discussed in a number of papers (see references 165-170 and references cited therein). Jehring et aZ.171 used tensammetry in the elucidation of membrane mechanisms. Possible analogies between electro- sorption layers and signal-transmitting biomembranes were discussed. The behaviour of neuronal switches based on changes in membrane structure and permeability caused by chemical and electrical inputs can be simulated in simplified models by electrosorption layers of organic compounds at the mercury/electrolyte phase boundary. Released electrical phenomena can be measured by ax. tensammetry and simple conventional polarography.171 Polymers and Plastics Except for biopolymers,45J65 only a few modern applications of tensammetry in the polymer and plastic field could be located.A list of macromolecules studied by tensammetry is to be found in references 1 and 25. A review of additives and a critical approach to tensammetry in the determination of additives and textile additives used in the plastics industry date from 1969.172 More recently square-wave tensammetry was used for the determination of surfactants with antistatic properties ,61 for anionic, cationic active and a polymeric non-ionic surfactant T-4-2403 (Dow-Corning)62 and for the determination of ethoxylated alkylarene derivatives63 Characterisation of the surfactant or the simultaneous deter- mination of different surfactants is not feasible.63 The method is suitable for the fast, simple determination of a known surfactant in polyethylene film or plates in the concentration range of about 0.1% after extraction of the surface-active compounds from the powdered polyethylene sample with hot water.An interesting application of a.c.1J73J74 (cf., Fig. 10) and square-wave tensammetry, is that developed by Jeh~ing24~25 for measuring the degree of polymerisation of polyethylene glycols. Polyethylene glycols belong to the group of chain oligomers. Their tensammetric behaviour has been exten- sively studied. Jehring was able to show that with increasing molecular mass (1000, 3000, 5000,20000) the peak potentialANALYST, JANUARY 1988, VOL. 113 11 I I - 1500 - 1700 -1900 EImV vs. SCE Fig.10. A.c. tensammogram showing a curve of a mixture of polyethylene lycols with average molecular mass of 600,1540,6000 and 20 000, ink 1 M LiCI. The concentrations of the samples are 30,15, 15 and 23.5 mg 1-1, respectively. Curves measured in the authors’ laboratory moves progressively towards more negative values. The shift is a linear function of the reciprocal average molecular mass. A resolved square-wave tensammetric peak is obtained for each of these average molecular masses, the peak width decreases with increasing molecular mass. The average molecular mass of unfractionated poly(viny1 alcohol) was determined by a.c. tensammetry in KCl supporting elec- trolyte.175 Borchers et al. 176 applied a.c. tensammetry to pulping problems, e.g., determination of surfactants used as dispersing agents in the pulp process.Corrosion Electrocapillary data,177 the suppression of the 0 2 maxi- mum178 and tensammetry have been applied to corrosion ~tudies.179~180 Depression of the base current produced in 0.1 M KCl by different amines was studied as a means of correlating adsorbability with protection against corrosion. The potential of tensammetric measurement for the rapid screening of potential anti-corrosion agents is currently being studied in our laboratory. Interference of Tensammetric Peaks in Conventional Polar- ography; Electrochemical Masking Tensammetric waves can be a great nuisance in conventional a.c.182 and pulse polarography.ls3-185 The determination of drugs in tablets, creams and ointments can be performed by differential-pulse polarography without any previous separa- tion, provided that the drug is more strongly adsorbed at the surface of the electrode than the surfactant present in the formulation. If the formulation contains strongly adsorbed surfactants, severe distortion, overlapping or complete ob- scuring of the faradaic waves by the tensammetric wave occurs, thwarting the analysis.183-185 In certain instances, the presence of surfactants can also be of advantage.lmJ86J87 Adsorption peaks can also shift or obscure stripping peaks in inorganic stripping analysis.187J88 Such tensammetric peaks have occasionally been mistaken for metal ion peaks, resulting in erroneously high results for metals.l= The influence of surfactants on the measurement of copper and cadmium speciation in model sea water by differential anodic stripping voltammetry was studied by Krznanic.189 Electrochemical masking is based on electrochemical effects of adsorption of an electrochemically non-reducible, non-oxidisable substance on to the electrode surface.Thus the determination of T1I in the presence of a 1000-fold excess of CuII, of FeIII in the presence of a 2000-fold excess of CuII and of Cur1 with a 1000-fold excess of Bi”1 is possible. Electrochemical masking is insufficiently used in practical polarographic analysis, although a number of papers dealing with the inhibition of electrochemical activity have appeared in the field of d.c.283190 and ax. polarography. Depression and possible shift of waves by various surface-active substances have also been exploited to improve resolution in stripping measurements of metal pairs of similar peak potentials,l81J91 e .g , determination of TlI and PbII in Cd salts by anodic stripping with addition of surfactants to suppress the Cd peak.191 Conclusions It is reasonable to state that both techniques, polarographic adsorption analysis and tensammetry, generally need more attention and care than other instrumental techniques. The development of new tensammetric procedures requires a specialist. Experience shows that non-specialists with ana- lytical laboratory experience can use adsorption analysis and tensammetry for some selected real problems in routine work, provided that rigorous analytical methods are available. It is hoped that the practical examples presented here illustrate that polarographic adsorption analysis and tensammetry can, when properly used, be valuable tools, and therefore labora- tories working in the organic trace (surfactants) analysis field should consider the benefits that polarographic adsorption analysis and, especially, advanced tensammetic techniques can provide.No analytical technique can automatically be considered to be the best, good or useless for all problems. Therefore, only objective comparison of the experimental results of other techniques available in a given laboratory can decide the ultimate choice of the best technique for a given situation or problem. The technical assistance of Mr. H. G. Wenzel is gratefully acknowledged. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.15. 16. 17. 18. 19. 20. References Jehring, H., “Elektrosorptionsanalyse mit der Wechselstrom- polarographie,” Akademie Verlag, Berlin, 1974. Bagdasarov, K. N., Lokshina, G. A., Sadimenko, L. P., and Sokolov, V. P., Zh. Anal. Khim., 1986,41, 171. Canterford, D. R., Photogr. Sci. Eng., 1977,21,215. Rosen, M. J., Hua, X.-Y., Bratin, P., and Cohen, A. W., Anal. Chem., 1981,53,232. Canterford, D. R., J. Electroanal. Chem., 1980,111,269. Bond, A. M., and Jones, R. D., Anal. Chim. Acta, 1983,152, 13. CosoVic, B., Batina, N., and Kozarac, Z., J. Electroanal. Chem., 1980, 113,239. Cheney, M. C., Curran, D. J., and Fletcher, K. S., 111, Anal. Chim. Acta, 1981,126,213. Haynes, D. L., Kelly, D. G., Smith, J. H., and Fernandes, E. L., “Developing Methods for Analyzing Oil Dispersants in Sea Water,” US Department of Commerce, NTIS, EPA-600/2- Gouy, G., Ann.Chim. Phys., 1903,7,145; 1906,.8,291; 1906, 9,75. Kucera, G., Bull. Int. Acad. Sci. Boh2mie, 1903; cited in references 13 and 28. Kucera, G., Ann. Phys., 1903, IV, ( l l ) , 529 and 698; cited in reference 13. Heyrovsky, J., and Simunek, R., Phil. Mag., 1929,7,951. Gosman, B. , and Heyrovsky, J., Trans. Electrochem. SOC., 1931,59,249. Proskurnin, M . , and Frumkin, A. N., Trans. Faraday SOC., 1935,31, 110. Kryukova, T. A., Zavod. Lab., 1940,9,699. Kryukova, T. A., Zavod. Lab., 1948,14,511 and 767. Kryukova, T. A., Zavod. Lab., 1950,16, 134. Breyer, B., and Hacobian, S., Aust. J. Sci. Res. Ser. A, 1952,5, 500. Breyer, B. and Hacobian, S., Aust. J. Chem., 1953,6, 186. 84-144, August 1984.12 ANALYST, JANUARY 1988, VOL.113 21. DOSS, K. S. G., and Kalyanasundaram, A., Proc. Indian Acad. Sci., Sect. A , 1952,35,27. 22. Doss, K. S. G., and Gupta, S. L., Proc. Zndian Acad. Sci., Sect. A, 1952, 36,493; Chem. Abstr., l953,47,9181a. 23. Barker,G.C.,andGardner,A.W.,A.E.R.E.Rep.,A.E.R.E. C/R 1606, 1955. 24. Jehring, H., Horn, E., Reklat, A., and Stolle, W., Collect. Czech. Chem. Commun., 1968,33,1038. 25. Jehring, H., J. Electroanal. Chem., 1969,21,77. 26. Cosovic, B., and Branica, M., J. Electroanal. Chem., 1973,46, 63. 27. Canterford, D. R., and Taylor, R. J., J. Electroanal. Chem., 1979, 98,25. 28. Heyrovsky, J., and Kuta, J., “Principles of Polarography,” Czechoslovakian Academy of Science Publishing House, Prague, 1965. 29. Bauer, H. H., in Bard, A. J., Editor, “Electroanalytical Chemistry,” Volume 8, Marcel Dekker, New York, 1975, p.169. 30 Linhart, K., Tenside Deterg., 1972, 9, 28 and 241. 3 1. Levich, V., “Physicochemical Hydrodynamics,” Prentice Hall, New York, 1962. 32. Frumkin, A. N., and Damaskin, B. B., J. Electroanal. Chem., 1962,3,36. 33. Barker, G. C., in Jeager, E., Editor, “Transactions of the Symposium on Electrode Processes,” Wiley, New York, 1961 , p. 325. 34. Jehring, H., paper presented at the 4th International Congress on Polarography, Prague, 1966; cited in reference 108. 35. Myers, D. J . , and Osteryoung, J., Anal. Chem., 1974,46,356. 36. Damaskin, B. B., Petrii, 0. A., and Batrakov, V. V., “Adsorption organischer Verbindungen an Elektroden,” Akademie Verlag, Berlin, 1975. 37.Kalvoda, R., tn Kalvoda, R., Editor, “Electroanalytical Methods in Chemical and Environmental Analysis,” Plenum Press, New York, 1987, Chapter 4, p. 30. 38. Breyer, B., and Bauer, H. H., “Alternating Current Polar- ography and Tensammetry,” Interscience, New York, 1963. 39. Henze, G., and Neeb, R., “Elektrochemische Analytik,” Springer-Verlag, Berlin, 1986 40. Bond, A. M., “Modern Polarographic Methods in Analytical Chemistry,” Marcel Dekker, New York, 1980. 41. Dahmen, E. A. M. F., “Electroanalysis, Theory and Applica- tions in Aqueous and Non-aqueous Media and in Automated Chemical Control,” Elsevier, Amsterdam, New York, 1986. 42. Jehring, H., and Stolle, W., Collect. Czech. Chem. Commun., 1968,33, 1670. 43. Jehring, H., J. Electroanal. Chem., 1969,20,33. 44.Smith, D. E., in Bard, A. J., Editor, “Electroanalytical Chemistry,” Volume 1, Marcel Dekker, New York, 1966, 45. Smith, D. E., Crit. Rev. Anal. Chem., 1971, 2,247. 46. Rosen, M. J., and Goldsmith, H. A,, “Systematic Analysis of Surface-active Agents,” Wiley, New York, 1972, and refer- ences cited therein. 47. Ventkatesan, V. K., in White, R. E., Bockris, J. O’M., Conway, B. E. , and Yeager, E., Editors, “Comprehensive Treatise of Electrochemistry, Volume 8, Experimental Methods in Electrochemistry,” Plenum Press, New York, 1984, pp. 495-607. 48. Novotny, L., and Kalvoda, R., Collect. Czech. Chem. Com- mun., 1986, 51, 1595. 49. Novotny, L., Smoler, I., and Kuta, J , Collect. Czech. Chem. Commun., 1983,48,964. 50. Novotny, L., and Smoler, I., J . Electroanal. Chem., 1983,146, 183.51. Hillson, P. J., J. Photogr. Sci., 1961, 9, 234, 52. Bembi, R., Goyal, R. N., and Malik. W . U., Talanta, 1976,23, 667. 53. Kopanica, M., and Dolezil, M , Chemist Analyst, 1965,54,44. 54. Kalvoda, R., “Techniques of Oscillopolarography,” Second Edition, Elsevier, Amsterdam, 1965. 55. Loveland, J. W , and Elving, P. J., J. Phys. Chem., 1952,56, 250,935,941 and 945. 56. Nangniot, P., in Duculot, J., Editor, “La Polarographie en Agronomie et en Biologie,” Gembloux, Belgium, 1970, p. 285. 57. Spini, G., Profumo, A., and Soldi, T., Anal. Chzm. Acta, 1985, 176,291. 58. Shallal, A. K., and Bauer, H. H., Anal. Lett., 1971,4(4), 205. pp. 1-155 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85.86. 87. 88. 89. 90. 91. 92. 93 I 94. 95. 96. 97. 98. 99. 100. 101, 102. Banica, F. G., Sadoveanu, A., and Patroescu, C., Anal. Lett., 1985,18(A4), 467. Britz, D., Anal. Chim. Acta, 1980, 115, 327 Porubska, M., and Naplava, A , Czech. Put , 192854, 1981; Chem. Abstr., 1982, 96,219685t. Porubska, M., Chem. Prum., 1979,29,38; Chem. Abstr., 1979, 90, 13933911. Porubska, M., Tenside Deterg., 1978,15,241. Cominoli, A., Buffle, J., and Haerdi, W., J. Electroanal. Chem., 1980, 110,259. Canterford, D. R., and Brown, R. W., J. Electroanal. Chem., 1981, 119,355. Bednarkiewicz, E., and Kublik, Z., Anal. Chim. Acta, 1985, 176, 133. Locatelli, C., Borghesani, G., and Pulidori, F.. Ann. Chim. (Rome), 1985,75, 33. Borghesani, G., and Locatelli, C . , Ann. Chim. (Rome), 1983, 73, 137.Vetterl, V., Jelen, F., and Palecek. E., Stud. Bzophys., 1986, 114,59. Holmqvist, P., J. Electroanal. Chem., 1976, 68, 31. Holmqvist, P., Anal. Chim. Acta, 1977, 89, 315. Holmqvist, P., Anal. Chim. Acta, 1977, 90, 35. Breyer, K., Fuginaga, T. , and Sawamoto, H., Bunseki Kagaku, 1966, 15,487; cited in reference 76. Sharma, S. K., and Prasad, G. P., Nuova Chzm. (Milano), 1974,50,83; Chem. Abstr., 1974,81, 96887r. Sharma, S. K., Acta Chim. Acad. Sci. Hung., 1974, 83, 195. Nakagawa, G., and Nomura, T., Anal. Lett., 1972,5(10), 723. Nomura, T., and Nakagawa, G., Bunsekz Kagaku, 1975, 24, 673; Chem. Abstr., 1979, 85,28153f. Nomura, T., and Nakagawa, G., J. Electroanal. Chem., 1980, 111,319. Bos, M., Anal. Chim. Acta, 1982,135, 249. Adam, H. H., Joslin, T. A., and Thomas, B., J.Photogr. Sci., 1979,27,66. Curran, D. J., Znt. Lab., 1981, 11, 28. Jehring, H. , “Elektrosorptionsanalyse mit der Wechselstrom- polarographie,” Akademie Verlag, Berlin, 1974, pp. 314-416. Crisp, P. T., in Swisher, R. D., Editor, “Surfactant Biodegra- dation,” Second Edition, Surfactant Science Series, Volume 18, Marcel Dekker, New York, 1987, pp. 61 and 108. Bond, A. M., “Modern Polarographic Methods in Analytical Chemistry,” Marcel Dekker, New York, 1980, p. 367. Crisp, P. T., in Cross, J., Edrtor, “Nonionic Surfactants, Chemical Analysis,” Surfactant Science Series, Volume 19, Marcel Dekker, New York, 1987, Chapter 3, p. 78. Kemula, W., Kublik, Z. , and Axt, A., Roczn. Chem., 1961,35, 1009. Kalvoda, R , and Budnikov, G., Collect. Czech. Chem. Commun., 1963,28,838.Bednarkiewicz, E., Denton, M., and Kublik, Z., J. Electro- anal. Chem., 1981, 127,241 Kalvoda, R., Anal. Chsm. Acta, 1982, 138, 11. Cosovic, B., and Hrsak, D., Tenside Deterg., 1979, 16,262. Batycka, H. , and Lukaszewski, Z . , Anal. Chim. Acta, 1984, 162,207. Lukaszewski, Z., and Pawlak, M. K., in Smyth, M. R., and Vos, J. G., Editors, “Electrochemistry, Sensors, Analysis,” Elsevier, New York, 1986, p. 119. Batycka, H., and Lukaszewski, Z., Anal. Chim. Acta, 1984, 162,215. Lukaszewski, Z., Batycka, H., and Zembrzuski, W., Anal. Chim. Acta, 1985, 175,55. Cosovic, B., Zutic, V., and Kozarac, Z . , Croat. Chem. Acta, 1977,50,229. Cosovic, B., and Vojvodic, V., Limnol. Oceanogr., 1982,27, 361. Bersier, P. M., Anal. Proc., 1987,24,44. Gupta, S . L., and Sharma, S.K., Electrochim. Acta, 1965,10, 151. Tedoradse, G . A., Arakeljan, R. A., and Belokolos, E. D., Elektrokhimzya, 1966, 2, 563. Jehring, H., 2. Phys. Chem. (Leipzig), 1971, 246, 1. Dobren’kov, G. A., and Ryabinina, N. I., Elektrokhimiya, 1972,8, 1873. van Tilborg, W. J. M., Recl. Trav. Chim. Pays-Bas, 1977, 96, 213.ANALYST, JANUARY 1988, VOL. 113 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141. 142. 143. Kozarac, Z., Zutic, V., and Cosovic, B., Tenside Deterg., 1976, 13,260. Sharma, S. K., and Rema, R., J. Chin. Chem. SOC. (Taiwan), 1974,21,167. Pawlack, M. K., and Lukaszewski, Z., Chem. Anal. (Warsaw), 1985,30,377.Canterford, D. R., J. Electroanal. Chem., 1980, 111,269. Canterford, D. R., J. Electroanal. Chem., 1981, 118, 395. Breyer, B., Pure Appl. Chem., 1967, 15,313. Kalvoda, R., in Ryan, T. H., Editor, “Electrochemical Detectors, Fundamental Aspects and Analytical Applica- tions,” Plenum Press, New York, 1984, p. 133. Kemula, W., Behr, B., Borkowska, Z., and Dojlido, J., Collect. Czech. Chem. Commun., 1965,30,4050. Lankelma, J., and Poppe, H., J . Chromtogr. Sci., 1976, 14, 310. Kemula, W., and Kutner, W., J. Chromatogr., 1981,204,131. De Jong, H. G., Voogt, H. W., Bos, P., and Frei, R. W . , J. Liq. Chromatogr., 1983,6, 1745. De Jong, H. G., Kok, W.Th., and Bos, P., Anal. Chim. Acta, 1983,155,37. Anderson, J. E., Bond, A. M., Heritage, I. D., Jones, R.D., and Wallace, G. G., Anal. Chem., 1982,54,1702. Sokol, W. F., and Evans, D. H., Anal. Chem., 1981,53,578. Gupta, S . L., and Holleck, L., 2. Naturforsch., Teil B, 1972, 27,634. Sharma, S. K., Acta. Chim. Acad. Sci. Hung., 1976, 91,27. Kozarac, Z., Zvonaric, T., Zutic, V., and Cosovic, B., Thalass. Jugoslav., 1977, 13, (1/2), 109. Cosovic, B., Vojvodic, V., and Plese, T., Water Res., 1985,19, 175. Jehring, H., Lohse, H., and Horn, E., 29th Meeting ISE, Budapest, August 28-September 2,1978, Extended Abstracts, Part I, p. 188. Bos, M., Van Willigen, J. H. H. G., and Van der Linden, W. E., Anal. Chim. Acta, 1984, 156, 71. Birch, B. J., and Hart, J. P., in Smyth, W. F., Editor, “Polarography of Molecules of Biological Significance ,” Academic Press, London, 1979, p.205. Kunkel, E., Vom Wasser, 1983,60,49. Kalvoda, R., and Novotny, L., Vodni Hospod. Part B, 1984, 34(11), 291; Chem. Abstr., 1985, 102,100528. Novotny, L. and Smoler, I., “Proceedings of the Second J. Heyrovsky Memorial Congress on Polarography , Prague, Zvonaric, T., Zutic, V., and Branica, M., Thalass. Jugoslav., 1973, 9, (1/2), 65. Cosovic, B., Zutic, V., Vojvodic, V., and Plese, T., Mar. Chem., 1985, 17, 127. Zutic, V., Novakovic, T., and Zvonaric, T., “Proceedings of the Second Heyrovsky Memorial Congress on Polarography, Prague, August 25-29, 1980,” p. 197. Cosovic, B., Chem. Processes Lakes, 1985, 55; Chem. Abstr. 1985,103,58825j. Zutic, V., Cosovic, B., and Kozarac, Z., J . Electroanal. Chem., 1977, 78, 113, and references cited therein. Hunter, K. A., and Liss, P.S., Water Res., 1981, 15,203. Quentel, F., Madec, C., Le Bihan, A., and Courtot-Coupez, J., Anal. Lett., 1986, 19 (3-4), 325. Quentel, F., Madec, C., and Courtot-Coupez, J., Anal. Lett., 1985,18 (A12), 1493. Quentel, F., Madec, C., and Courtot-Coupez, J ,Anal. Lett., 1987,20 (l), 47. Kalvoda, R., and Novotny, L., Collect. Czech. Chem. Com- mun., 1986,51,1587. Zutic, V., Cosovic, B., Marcenko, E., Bihari, N., and Krsinic, F., Mar. Chem., 1981, 10,505. Kozarac, Z., Cosovic, B., and Branica, M., J. Electroanal. Chem., 1976,68,75. Kozarac, Z . , Hrsak, D., Cosovic, B., and Vrzina, J., Environ. Sci. Technol., 1983, 17,268. Britz, D., and Mortensen, J., J, Electroanal. Chem., 1981,127, 231. Mukerjee, P., and Mysels, K. J., NSRDS-NBS 36, National Bureau of Standards, Washington, DC, 1971; cited in reference 140.Colichman, E. L., J. Am. Chem. SOC., 1950,72,4036. Malik, W . U., and Chand, P., J. Am. Oil Chem. SOC., 1966,43, 448. August 25-29, 1980,” p. 128. 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166. 167. 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 178. 179. 180. 181. 182. 183. 184. 13 Malik, W. U., and Jhand, 0. P., J. Am. Oil Chem. Soc., 1972, 49, 170. Miiller, E., and Dorfler, H.-D., Tenside Deterg., 1977,14,75. Al’bota, L. A., and Zaverach, M. M., Kolloidn. Zh., 1974,36, 1022. Jehring, H., and Weiss, A., Tenside Deterg., 1969,6,251. Dietrich, P., Hager, G., Jehring, H., and Horn, E., Tenside Deterg., 1973, 10, 173. Vollhardt, D., Tenside Deterg., 1975, 12,255. Vollhardt, D., 2.Chem., 1976,16,159. Vollhardt, D., Colloid Polym. Sci., 1976, 254, 64. Palyi, G., KolloidZ. 2. Polym., 1971,249, 1158. Miiller, E., and Dorfler, H.-D., Abh. Akad. Wiss. DDR, 1976, 581; Electroanal. Abstr., 1978,16,4381. Miiller, E., and Dorfler, H.-D., 2. Chem., 1981,21,28. Miiller, E., and Dorfler, H.-D., J. Colloid Interface Sci., 1981, 83,485. Canterford, D. R., J. Electroanal. Chem., 1976,73,247. Canterford, D. R., Anal. Chim. Acta, 1977,94, 377. Jaenicke, W., and Kobayashi, H., Electrochim. Acta, 1983,28, 245. Kiseleva, L. A., and Orlov, Yu.E., Pharmatsiya (Moscow), 1977,26,38. Przegalinski, M., and Matysik, J., Bioelectrochem. Bioenerg., 1982,9,761. Ladanyi, E., Zugravu, E., and Tomoaia, M., Int. Arch. Arbeitsmed., 1974,33, 245. Ladanyi, E., and Stalder, K., J. Electroanal. Chem., 1979,99, 321, Kala, H., and Fahr, F., Pharmazie, 1974, 29, 10. Kutner, A., Jaworska, R., Kutner, W., and Grzeszkiewicz, A., in Gorogis, S., Editor, “Advances in Steroid Analysis,” Akademiai Kiadb, Budapesto, 1982, p. 333, Dryhurst, G., “Electrochemistry of Biological Molecules,” Academic Press, New York, 1977. Palacek, E., in Smyth, W. F., Editor, ‘Electroanalysis in Hygiene, Environmental, Clinical and Pharmaceutical Chem- istry,” Elsevier, New York, 1980, p. 79, and references cited therein. Temerk, Y. M., Valenta, P., and Niirnberg, H. W., in Smyth, W. F., Editor, Electroanalysis in Hygiene, Environmental, Clinical, and Pharmaceutical Chemistry, Elsevier, Amsterdam, New York, 1980, p. 113, and references cited therein. Barker, G. C., J. Electroanal. Chem., 1986,214,373. Barker, G. C., J. Electroanal. Chem., in the press. Temerk, Y. M., Kamal, M. M., Ahmed, M. E., and Ahmed, Z. A,, Bioelectrochem. Bioenerg., 1986, 16, 497, and refer- ences cited therein Jehring, H., Nguyen Viet Huyen, Trinh Xuan Gian, Horn, E., and Hirche, Ch., J. Electroanal. Chem., 1979, 100, 13. Helmstedt, M., Hagen, E., and Schroder, E., Plasde Kautsch., 1969, 16(3), 165, and references cited therein. Jehring, H., Habilitationsschrifi, Technical University of Dres- den, 1965; cited in reference 108. Jehring, H., Proc. Con$ Appl. Phys.-Chem. Methods Chem. Anal., Budapest, 1966,1,109; Chem. Abstr., l968,68,45553a. Sauer, W . , Kimmer, W., and Geyer, R., Plaste Kautsch., 1974, 21,518, and references cited therein. Borchers, B., Jehring, H. and Horn, E. , ZeZlstoff Papier, 1979, 257. Antropov, L. I., Trans. Novotcherk. Polytech. Inst., 1954, 25/39,5; Proceedings of the 5th Corrosion Congress, Moscow, 1955, cited in reference 180. Gatos, H. C., Trans. Electrochem. SOC., 1954,101,433, citedin reference 180. Rajagopalan, K. S., Proc. IndianNat. Sci. Acad., PartA, 1982, 48, 364. Rajagopalan, K. S., Venkatesan, V. K., and Balakrishnan, K., Proc. Indian Acad. Sci., Part A, 1959, SO, 272. Porubszky, I., Gyory-Szebenyi, E., and Gagyi Pallfy, E., Proc. Conf. Appl. Phys. Chem., 2nd, 1971, 1, 515; Chem. Abstr., 1972,76, 101534b. Spahr, L. J., and Knevel, A. M., J. Pharm. Sci., 1966,55,1020. Jacobson, E., in Smyth, W . F., Editor, “Electroanalysis in Hygiene, Environmental, Clinical and Pharmaceutical Chem- istry,” Elsevier, Amsterdam, New York, 1980, p. 227, and references cited therein. Opheim, L.-N., Anal. Chim. Acta, 1977,89,225.14 185. Jacobson, E . , and Korvald, B., Anal. Chzm. Acta, 1978, 99, 255. 186. Bersier, P. M., and Bersier, J., Crif. Rev. Anal. Chem., 1985, 16,81. 187. Ciszewski, A., and Lukaszewski, Z . , Anal. Chin Acta, 1983, 146,51. 188. Wang, J., “Stripping Analysis, Principles, Instrumentation and Applications,” Verlag Chemie, Weinheim, 1985, p. 104. ANALYST, JANUARY 1988, VOL. 113 189. Krznaric, D., Mar. Chem., 1984, 15, 117. 190. Lukaszewski, Z., Tafanta, 1977,24, 603. 191. Lukaszewski, Z., Pawlak, M. K., and Ciszewski, A . , Talanta, 1980, 27, 181, and references cited therein. Paper A71242 Received June 15th, 1987 Accepted June 19th, 1987

ISSN:0003-2654    

DOI:10.1039/AN9881300003

出版商:RSC    

年代:1988

数据来源: RSC

摘要:

ANALYST, JANUARY 1988, VOL. 113 15 Application of a Controlled-growth Mercury Drop Electrode to Polarograp hy* Zygmunt Kowalski School of Mining and Metallurgy, Institute of Materials Science, 30-059 Krakdw, al. Mickiewicza 30, Poland ~ ~~ ~~ ~ ~ In the proposed device for producing the controlled-growth mercury drop electrode (CGMDE), the mercury outflow is controlled by a fast response valve. Because of this the drop growth mode, rate and size can be modified by the pulse sequence and programmed in advance. The CGMDE retains many features of the DME and SMDE and also has other unique capabilities which are interesting from the point of view of polarography, voltammetry and automatic analysis; recali bration of the drop size may be achieved automatically. The results presented in this paper provide an illustration of these features, particularly those which are of interest in polarography.Keywords: Cuntrulled-growth mercury drop electrode; polarography The basic features of the conventional dropping-mercury electrode (DME) when used as a polarographic detector are the periodically renewed mercury surface and the complete regeneration of the solution in contact with the electrode. These features guarantee reproducible current - voltage curves (polarograms). A conventional DME assembly requires great care and careful manual operation but this cannot be easily achieved in analytical laboratory practice. The source of inconvenience here is the very small internal diameter of the capillary; however, the main electrochemical parameter of the DME also depends on the internal diameter of the capillary.The charging current arising from the continuous growth of the area is the main source of noise, the extent of which depends on the rate of formation of the drop. The continuous and slow expansion of the mercury drop involves relationships which offer information on the elec- trode process. The shapes of the current - time curves during the lifetime of a single drop allow some kinetic currents to be distinguished and they also demonstrate the presence of adsorption phenomena. A different problem in the generation of the mercury drop arises with the stationary mercury drop electrode (SMDE).lJ Here the electrode assembly incorporates a valve which allows the mercury flow to be stopped at selected time intervals to produce a stationary rather than a growing drop.In this instance the current - voltage curves represent the measurement data sampled during the stationary period of the drop lifetime. The SMDE retains some of the advantages of the conventional DME, e.g., it is a renewable electrode of constant size. The mode of operating the SMDE eliminates the charging current arising from the growth of the area. However, the data which can be obtained using an electrode with a continuously changing area are not available with the SMDE.3-5 Recently, a pressurised mercury electrode6 has been reported which can deliver reproducible drops in the hanging mercury drop electrode ( W E ) mode. The approach offered by the controlled-growth mercury drop electrode7 (CGMDE) combines the advantages of both the conventional DME and the SMDE.In this electrode the mercury outflow (drop growth) is controlled by a fast response valve, actuated by a pulse sequence generated by a pulse sequencer or a computer, which causes the drop size to increase in a step by step mode. The drop size can be described operationally by specifying the number of pulses at the given pulse width.8 Hence the velocity of the growth of the drop can be controlled by the time intervals between the pulses operating the valve. * Presented at the International Symposium on Electroanalyais and Sensors in Biomedical, Environmental and Industrial Sciences, Cardiff, UK, 6 9 April, 1987. In Fig. 1 the operation of the DME, SMDE and CGMDE are compared. As can be seen, the CGMDE retains the features of the DME and SMDE and also has some additional features which are interesting from the polarographic point of view.The drop growth period cafl be extended in time by the application of long time intervals between the pulses operating the valve. Fig. 2 shows the CGMDE assembly. The operation of the electrode consists in the flow of mercury out from the reservoir through a capillary of large internal diameter. The capillary has a metallic tube in its upper part. This tube, a flat Time - Time - I period , Time - 1. Comparison of the o erations of the DME, SMDE and of- vs. time curves for (a) DME; (b) %kDE. Schematic dia SMDE; and (c) C G M D r li 0.15-0.20 rnm Pys. 2. Diagram of the CGMDE assembly. A Ca metallic tube; (C) rubber gasket; (D) flat spring; {El s i ; 8 solenoid; (G) spacer; and 0 support16 ANALYST, JANUARY 1988, VOL.113 spring and a washer comprise a fast operating valve. The valve is activated by means of a pulling rod and solenoid. The dynamics ot the step by step growth of the drop depends on the time intervals between the pulses operating the valve. Hence, in an extreme example, when the valve is energiwd by one long pulse the drop is generated in one step as with the SMDE described by other workers.1.2 There are several advantages expected from the principles on which the device is based. (i) The drop mode of growth, rate and drop size can be modified by the pulse sequence and programmed in advance; (ii) sampling of the current during the off-pulse time allows the current - time curve to be recorded free of the capacity component; and (iii) recalibration of the drop size should be easy and could be achieved automatically by generating the drop with a pulse sequence until it reaches a gravity-controlled size.Some of the above features make the polarographic detector an integral part of the measuring system. The limitations include the mechanical cutting-off of the mercury outflow and the steplike movement of the drop which may produce some vibration of the drop. However, these possible limitations can be reduced if the technical problems are solved. Experimental and Results All chemicals used were of analytical-reagent grade. All solutions were de-aerated with purified argon before use. A PP-04 polarograph (Telpod, Krak6w) and a Radiometer PO-4 polarograph were used.The pulse sequence for activating the valve was generated either by a computer or by a Digitimer D 4030 pulse sequencer. The laboratory-made CGMDE used has been described8 and is shown in Fig. 2. Capillaries of various internal diameters (0.20 and 0.25 mm) were used. Some of the measurements performed in this work to verify the expected features of the CGMDE as a polarographic A sensing system were recorded in the conventional d.c. mode, whereas others were recorded using the sampling method. Hence, the results are more significant because the recorded profiles are presented in the conventional manner. The current - time profiles for a 0.1 mM cadmium solution are presented in Fig. 3 (A). For all of the profiles the maximum drop size is the same, but the dynamics of the drop growth are different and depend on the intervals between the pulses operating the valve.The drop growth is expressed here by the three different intervals between the pulses (0.6,l and 2 s) from which we obtained three current - time curves, differing from each other in the maximum current value. Only the stationary current remains almost the same as it depends on the actual drop size which is constant. In Fig. 3 (B) the current - time curves recorded for 0.1~ KNO3 represent the charging current on the growing mercury drop. The conditions of drop growth were the same as in Fig. 3 These three curves demonstrate a fact which is already known from DME studies, i.e., that the charging current attains its maximum value for the fastest growing drop.In Fig. 4 conventional d.c. polarograms for Ni are shown. Both curves were recorded using a dropping mercury elec- trode of the same maximum drop size. The first curve was recorded on a fast growing drop mercury electrode (the pulse intervals were 10 ms), and the mean lifetime of the drop was estimated as 0.5 s. The next polarogram was recorded on slow growing drops (160-ms pulse intervals) and the lifetime of the drop was estimated as 4.6 s. The differences between these two polarograms become evident if their limiting currents and half-wave potentials are compared, the currents differing %fold, with a 60-mV difference between the half-wave potentials. The recording of two similar curves by the conventional method with the DME would require the use of two different capillaries.Using the CGMDE, which allows the drop time to be (A). B Time - Fig. 3. (A) Faradaic current during and after dro rowth. Drop size, 20 pulses; pulse wtdth, 10 ms; CdlI, 0.1 mM; KN03, 0.5 M; and potential, -0.6 V vs. SCE. Pulse intervals: a) 0.8 fb) 1.0; and (c) 2 0 s. (B) Chargmg current during drop growth. Drop parameters as in A (a)-(c); KN03 0.5 M; and potential, -6.6 V vs. SCEANALYST, JANUARY 1988, VOL. 113 17 Time - 1 I I I 0.9 1.0 1.1 1.2 1.3 Potentia IN Time - Fig. 6. Effect of a surface-active substance on the shape of the current time m e for (A) an exponentidly growing surface of the mercury drop and @) a linearly growing surface. D sizes are the same in both mstances. me ulse sequence generatinze drop is 8s in Fi .5(6) for (A) and 5(c) k r B). a) 0.2 m~ CdII in 0.1 M KNO,. Oker lines same as (a) +th.(b) 0.602, (c) 0.004, (d) 0.008 and (e) 0.016% glue. Broken h e indxates current measured at the constant size period of the drop life time without glue illustrates the Faradaic current, and the lower row of curves shows the charging current. Fig. 5(b) represents the current - time profile of the drop generated by pulses of the same duration, and gives current - time profiles very close to that observed when the drop grows freely, as with the DME. Figs. 5 (a) and (c) show current - time relationships representing extreme instances, In the first instance [Fig. 5 (a)] the rate of growth of the drop decreases with time. As a result the Faradaic current has a distinct plateau, the charging current decreases rapidly and the ratio of the Faradaic to the charging current at the end of the drop life is low.This could be considered advantageous from the point of view of polarographic measurement. In Fig. 5 (c) the current - time relationship is almost linear both for the Faradaic and the charging current and is a result of the method used for generating the drop; here the drop grows more quickly with time. Experiments were also carried out to show how the step by step growfh of the mercury drop could be applied to the study of adsorption phenomena. The effects of the surface-active substances on the shapes of the current - time curves for individual drops growing continuously are well known and are described in many polarography handbooks.However, special equipment is needed to record a current -time curve. It was considered desirable to ascertain whether the intervals between the pulses controlling the dynamics of the drop growth would disturb the non-Farad 'c phenomena, i.e., applying the CGMDE in a similar manner to the DME. It is possible to demonstrate the effect of surfaceactive substances using the CGMDE by means of an extension of the drop growth period and by using a conventional polamgraph. Line (a) in Fig. 6 (A), for 0.2 m~ Cd, has an easily recognised exponential shape. This shape changes considerably with increasing concentrations of the surface-active substances [lines (b)-(e) in Fig. 6 (A)], demonstrating the presence of adsorption phenomena. In this work natural glue was used as the surface-active substance.A practical application of this effect is that the shape of the graph can supply information about possible electrochemical complications due to the surface-active substances which decrease the currents measured in the constant size period of the drop lifetime. Fig. 6 (B) shows the results of an experiment in which the same solutions as in Fig. 6 (A) were used. However, here the whether it would be possible to recor 3 similar profiles by I I I I I 0.3 1.0 1.1 1.2 Poten tia IN Fig. 4. Conventional and ulse width, 10 ms. (6) fa ms (rmm = 4.6 f 1.0 3 0.5 Time - f 0.05 9 lime- selected over a wide range, we can choose the best possible conditions for the determination of the half-wave potential and the kinetic parameters, which is essential in electrochemical studies.By controlling the process of drop growth it is possible to produce a drop, the surface of which (and hence the current) will not increase exponentially with time. It is not easy to estimate the usefulness of such a diverse method of generating the mercury drop in practice. Some experiments are presented here to illustrate the advantages of the method. Fig. 5 shows a comparison between three different methods of producing a mercury electrode. The upper row of curvesANALYST, JANUARY 1988, VOL. 113 18 t + e s u Time - Fig. 7. Effect of a surface-active substance on the shape of the current - time curve for the exponential1 growing surface of the mercury drop. Drop size and solution as in Jig. 6. Curves recorded by sampling the current I A I I t I Time - 6"' Fig.8. Effect of the deposition conditions on the anodic scan: stirred solution; (b unstirred solution; and c) stirring controlled y the growin drop. drop sizes equal for (a), (& and (c); 0.1 mM C d I I in 0.1 M m$ drop was generated in such a way that its growth period had a linear profile. In this instance the shape of the current - time profile is less dependent on the increasing concentration of the surface-active substance, but the maximum current undergoes a significant change. Therefore, this mode of drop generation may be more useful for the quantitative evaluation of the adsorption phenomena than the previously described mode. Fig. 7 shows a current - time profile, recorded by samplingg the current.The measurement was carried out 100 ms after the valve was closed. The shape of the curve does not differ from those recorded by the conventional method but these curves are distinguished by a lack of charging component. As was mentioned previously, the change in the area of this electrode may cause vibration of the drop, and this is shown by the small ringing effect observed on the non-Faradaic profiles. The question is to what degree does this change in area and the vibration cause convective disruption of the diffusion layer. To determine the experimental limitations due to convection when the mercury drop is growing in the step by step mode, special stripping experiments were carried out. The drop size and deposition time were constant in these experiments, but the conditions under which the deposition processes were carned out varied.In the first experiment the deposition was carried out with very gentle stirring of the solution, whereas in the second experiment the solution was not stirred. In the last experiment the stirring was controlled by the step by step growth of the mercury drop. As can be seen from Fig. 8, no significant difference is observed between the currents measured at the anodic scan in the second and third experiments, but the current at the anodic scan in the first experiment (with very gentle stirring) is five times greater. From these experiments we conclude that the effect of convection due to the step by step growth mode is negligible even when the drop is growing in 20-ms steps as in the experiment described above.To illustrate how simple and effective the recalibration of this device is, one of the possible variants of recalibration is briefly described. In the first stage of the experiment the mercury level in the CGMDE assembly was fixed at 195 mm. At this pressure the gravity-controlled mercury drop size (maximum drop size) was generated by 59 pulses (On pulse time was 4 ms). The applied drop size for the 0.2 mM cadmium voltammetric curve was recorded at 26 pulses. The measured current at this drop size was 95.1 pA. The measurement was repeated with the mercury level lowered to 150 mm. At the first stage of recalibration the gravity-controlled mercury drop was generated as before and it was found that the maximum gravity-controlled drop size was generated by 71 pulses.The number of pulses required for generating the same drop size as for the first experimental stage, when the mercury level was 195 mm, was obtained by simple mathematical calculations. In this instance the number of pulses found was 34. After this recalibration the current measured for the same solution was 95.6 pA. The range for these two values (95.1 and 95.6 pA) was 0.5%. This recalibration procedure was checked several times and satisfactory results were obtained each time. Conclusions The controlled-growth mercury drop electrode, used as a polarographic detector as described in this paper, should be widely applicable to conventional instruments in highly developed measuring systems and in automatic analysers. Automatic analysers can operate reproducibl y over long periods of time and hence recalibration is an important factor.The use of the CGMDE makes recalibration relatively easy. Polarography has often been used in teaching instrumental methods as the information supplied by the polarographic wave can be registered using very simple instruments. This aspect of polarography is even more evident if we use the CGMDE with a pulse sequencer. The choice of the para- meters over an extremely wide range for the dropping mercury electrode can now be made by switching instead of manipula- tions of the mercury reservoir and the capillaries. The author expresses his thanks to Dr J. Migdalski for his assistance in carrying out some of the experiments and to Mrs M. Jakubowska who wrote the program for the computerised data acquisition system. References 1. Peterson, M. W., Am. Lab., 1979,ll. 69.ANALYST, JANUARY 1988, VOL. 113 19 2. Novotny, L., “Proceedings of the J. Heyrowsky Memorial Congress on Polarography, Prague, August S 2 9 , 1980,” Volume 11, 1980, p. 129. 3. Bond, A. M., and Jones, R. D., A d . Chim. Acta, 1980,121, 1. 4. Lovric, M., Osteryoung, R. A., O’Dea, J. J., and Osteryoung, J., unpublished work. 5. Andemon, J. E., Bond, A. M., and Jones, R. D., J. Eledroanal. Chem., 1981,130,113. 6. Foresti, M. L., and Guidelli, R., J. Ek-1. Chem., 1986, 197,159. 7. KO-, Z., and Osteryoung, J. G., Invention No. R-343 “Hanging and Dropping Mercury Electrode,” 1985, (registered at the Resear& Foundation of the State University of New York, P.O. Box 9, Albany, NY 12201 OOO9). young, J., A d . Chem., 1987,59,2216. 8. KO-, Z., W O ~ , K. H., OSteryOUng, R. A,, a d Oster- Paper A71235 Received June loth, 1987 Accepted August 24&, 1987

ISSN:0003-2654    

DOI:10.1039/AN9881300015

出版商:RSC    

年代:1988

数据来源: RSC

摘要:

ANALYST, JANUARY 1988, VOL. 113 21 Study of the Polarographic Behaviour of Antibiotics Part II* Li-Zhen Cai and Mae-Lian Wu Department of Chemistry, Shanxi University, Taiyuan, China The pola rog rap hic be haviour of 7 - [ ~ ( -)-ar-(4-et hyl-2,3-dioxo-l -pi perazi neca rboxa mi do)-a-( 4-hydroxy- phenyl)acetarnido]-3-[(1-methyt-l H-tetrazol-5-yi)thiomethyl]-3-cephem-4-carboxylic acid was studied. The sample yields a clear single-sweep oscillopolarographic wave at about -1.1 V in KH2P04 - Na2HP04 buffer solution. The relationship between peak current and concentration is linear in two different concentration ranges. The results prove that the wave has adsorption characteristics. A mechanism is proposed for the reduction of the sample. Keywords : Antibiotics; 7-[u (-)-a- (4-e th yi-2,3-dioxo- I -piperazin eca rboxamid0)-a- (4- h ydrox yp hen y1)ace t- am ido]-3-[( 1 -me th yl- I H - te trazol-5-yl) th iom e th yl1-3-cep h em -4-ca rbox ylic acid; pola rog ra p h y Polarography and other voltammetric methods are becoming increasingly important in the determination af compounds of pharmaceutical interest.l-4 Squella et aZ.1 and Fogg et aZ.2 have reported methods for the determination of cephalexin.Forsman and Karlsson3 have reported the differential-pulse and direct-current polarographic behaviour of benzyl- penicilloic acid, which is the major degradation product of benzylpenicillin . 7 - { ~ ( -)-a-( 4-Ethyl-2,3-dioxo- 1- piperazinecarboxamido)- a- (4- h y droxyphenyl) acet amido] -3- [ (1-methyl-lH-tetrazol-5-yl) thiomethyl]-3-cephem-4- carboxylic acid (T-1551 acid) is a synthetic antibiotic.3 This paper reports the polarographic behaviour of this antibiotic in KH2P04 - Na2HP04 buffer solution.Experimental Apparatus A Model SJP-1A single-sweep oscillopolarograph (Lishui Radio Factory, China) was used in the three-electrode mode. The working electrode was a dropping-mercury electrode and the counter electrode was a platinum wire. A Model 217 saturated calomel electrode was used as the reference electrode. A Model XJP-821(B) neopolarograph (Jiansu Electroanalytical Instrument Factory, China) was used for cyclic voltammetry. A Model 883 polarograph (Shanghai Analytical Instrument Factory, China) was used to record the d.c. polarographs. A Model 25 pH meter (Leici Instrument Factory, Shanghai, China) was also used.Reagents All the chemicals used were of analytical-reagent grade except for T-1551 acid, which was provided by Taiyuan Pharmaceut- ical Factory. All solutions were prepared with de-ionised, distilled water. The stock solution of 10-3 M T-1551 acid was prepared by dissolving appropriate amounts of T-1551 acid in 50 ml of pH 7 phosphate buffer solution. Procedure Appropriate amounts of the T-1551 acid stock solution were added to the pH 7 phosphate buffer solution to yield the desired concentration. Potentials were scanned in a negative direction. All measurements were carried out at room temperature verszis SCE. Results and Discussion Choice of Supporting Electrolyte T-1551 acid solution (10-4 M) was examined in the following supporting electrolytes6: Britton - Robinson buffer, potas- sium dihydrogen phosphate - disodium hydrogen phosphate buffer, acetic acid - sodium acetate buffer, sodium chloride (0.1 M) and potassium chloride (0.1 M).When KH2P04 - Na2HP04 buffer solution of pH 7 was used, a clear single-sweep oscillopolarographic wave at about - 1.1 V was obtained, which was stable with time and easily measurable. Accordingly, this buffer solution was used as the supporting electrolyte in subsequent experiments. Effect of Initial Potential of the Cathodic Scan With a positive shift in the initial potential, the peak current of the solution increases. Therefore, it is necessary to use the same initial potential in all experiments. The initial potential was fixed at -0.75 V in subsequent experiments.Effects of Standing Time A 10-4 M solution of T-1551 acid was tested and the peak current remained unchanged for 4 d. Peak Current as a Function of Concentration The dependence of the peak current on concentration is shown in Fig. 1. A linear relationship holds between the peak current and the concentration of the T-1551 acid solution in ~ 1 0 - 5 M 1 2 3 4 * Presented at the International Symposium on Electroanalysis and Sensors in Biomedical, Environmental and Industrial Sciences, Cardiff, UK, 6-9 April, 1987. For Part 1 of this series, see reference 4. 0 3 5 7 ~110-4 M Fig. 1. Peak current as a function of the concentration of T-1551 acid22 ANALYST, JANUARY 1988, VOL. 113 the range 5 x 10-6-4 X 10-5 M and 3 X 10-4-8 X 10-4 M. An increase in concentration has little effect on the peak potential (a small drift towards negative potentials).Study of the Properties of the Polarographic Wave The electrocapillary curve is shown in Fig. 2. As can be seen, when 10-4 M T-1551 acid was present the surface tension of the mercury drop decreased in the potential range 0-1.3 V and a visible indentation appeared at about -0.9 V. The dependence of the peak current on temperature was examined in the range 4-40 "C. As the temperature rises, the peak current increases below 16 "C but decreayabove 21 "C and does not change between 16 and 21 "C: The effect of surfactants on the peak current was investi- gated for a 10-4 M T-1551 acid solution. When the first drop (ca. 0.04 ml) of surfactant (1% gelatin) is added to the sample solution, the peak current decreases and continues to decrease with the gradual addition of gelatin.The effect of mercury pressure on the d.c. wave was investigated for a 10-4 M T-1551 acid solution. The wave height is proportional to the height of the mercury column. The above experimental results indicate that the polaro- graphic wave has adsorption characteristics. To elucidate further the electrode reaction of T-1551 acid, a cyclic voltammogram was recorded at the hanging mercury drop electrode. In the cyclic voltammogram (Fig. 3), the reduction peak is visible on the cathodic branch but the oxidation peak is not found on the anodic branch. This result indicates that the electrode reaction is irreversible. A plot of ln[I/(Id - l)] vs. -E, where I d is the limiting current and E is the potential, from a d.c.polarogram of 10-4 M T-1551 acid solution was linear, and an an value(a = transfer coefficient, n = no. of electrons) of 0.49 was calculated from the slope of the line. The result of logarithmic analysis indicates an irreversible reaction. Study of Mechanism of Reduction D( -)-(4-Ethyl-2,3-dioxo-l-piperazinecarboxamido)phenyl- acetic acid (I) and 7-amin~-3-acetoxymethyl-3-cephem-4-car- boxylic acid (11) are intermediates in a synthetic process. Single-sweep oscillopolarography of compounds I, I1 and aminobenzylpenicillin (111) was carried out under the same conditions as those used for T-1551 acid. Only compound I yielded a reduction peak at about - 1.4 V. From this result and by comparing the molecular structures of I, 11, III and T-1551 acid, the following reduction mechanism for T-1551 acid can be proposed: 0.4 0.8 1.2 1.6 2.0 - E N VS. SCE Fig. 2. Electrocapillary curves. (1) pH phosphate buffer and (2) 10-4 M T-1551 acid \ P 0.5 0.7 0.9 1 1 7.3 1 5 -EN VS. SCE Fig. 3. Cyclic voltammogram of 10-4 M T-1551 acid solution References 1. Squella, J. A., Nunez-Vergara, L. J., and Gonzales, E. M., J. Phann. Sci., 1978, 67, 1466. 2. Fogg, A. G., Fayad, N. M., and Goyal, R. N., J. Pharm. Pharmacol., 1980, 32,302. 3. Forsman, U., and Karlsson, A., Anal. Chim. Acta, 1981, 128, 135. 4. Cai, L.-Z., and Wu M.-L., J . Shanxi Univ. (Nar. Sci. Ed.), 1986,2, 59. 5. Saikawa, I., Ya suda, T., Taki, H., Watanabe, Y., Tai, M., Matsubara, N., Takahata, M., and Fukuoka, Y., Yakugaku Zasshi, 1979,99, 1073. 6. Jiang, Y., and Zhou, T.-H., Acfa Pharm. Sin., 1984, 19, 195. NOTE-Reference 4 is to Part 1 of this series. Paper A 71213 Received May 27th, I987 Accepted July 20th, 1987 J-J N' I + 2H+ + 28 OH O H

ISSN:0003-2654    

DOI:10.1039/AN9881300021

出版商:RSC    

年代:1988

数据来源: RSC

摘要:

ANALYST, JANUARY 1988, VOL. 113 23 Elect roc hem ical Reduction of Cef su lodi n * Eulogia Muiioz, Luis Camacho, Jose Luis Avilat and Francisco Garcia-Blanco Department of Physical Chemistry, Faculty of Sciences, University of Cordoba, 14004 Cordoba, Spain ~~ ~ A study has been carried out on the electrochemical behaviour of cefsulodin in buffered aqueous solutions using differential-pulse polarography (DPP), direct-current polarography (DCP) and cyclic voltammetry. Both the DPP and DCP techniques allow the determination of this antibiotic in the concentration range 1-200 p.p.m. The main wave observed is attributed to the reduction of the isonicotinamide substituent. The electrochem- ical behaviour has been found to change at a concentration of about 60 p.p.m. Keywords : Cefsulodin determination; differential-pulse polarograph y; direct-current polarograph y; cyclic voltammetry; antibiotics The use of electrochemical techniques for the determination of compounds of pharmaceutical interest is continually gaining in importance.The inherent sensitivity and high selectivity of the techniques allow very simple determinations, both in commercial samples and in body fluids in the presence of metabolites or impurities such as precursors used in the synthesis of these compounds. Among the p-lactam antibiotics, cephalosporins are gener- ally electroactive, either because of the presence of oxidisable or reducible substituents or because of the double bond in the dihydrothiazine ring common to all p-lactams. The reduction of such a double bond reportedly involves cleavage of the molecule1 resulting in the substituent borne by the carbon atom at position 3 in the ring being detached.In fact, electrochemical techniques are applicable to this type of compound only when they contain a leaving group at position 3.2 It has been shown that cephalosporins containing only a methyl group at position 3 cannot be electrochemically reduced.3 Polarographic techniques in general and differential-pulse polarography in particular are considered to be useful tools for the determination of cephalosporins. A number of methods for the direct determination of electroactive antibiotics68 and for the indirect determination of the basic hydrolysis products of non-electroactive antibiotics have been proposed.9-13 This paper reports a study of the electrochemical behaviour of cefsulodin (I), a cephalosporin particularly active against pseudomonas and hence widely used in hospital chemo- therapeutics.S The pH was adjusted with solid sodium hydroxide and the ionic strength was adjusted to 0.7 M with sodium nitrate. The, polarographic recordings were run on a Metrohm 626 polarograph furnished with a pulse generator controlling the hammer. The capillary features of the Amel 490 electrode used, were as follows: rn = 2 mg s-1 and t = 4.5 s at pH 1.6 and h = 40 cm. The DPP technique was applied at a controlled drop time of 0.5 s, with a scan rate of 2 mV s-1 and a pulse height of -50 mV. The cyclic voltammetric study was carried out with a programmable HQ Instruments 305 function generator, an HQ Instruments 105 potentiostat, a digital-memory Prowler N 400 oscilloscope and a Houston Instruments 2000 recorder.The working electrode was a Metrohm EA-290/1 hanging mercury drop electrode. The auxiliary and reference elec- trodes were platinum and saturated calomel electrodes, respectively. All measurements were made in a nitrogen atmosphere at 25.0 k 0.1 "C. Results Differential-pulse Polarography In acidic media, cefsulodin yields three reduction peaks using differential-pulse polarography (Fig. 1), the least cathodic of which (A) is also the largest and best defined. The other two peaks (B and C) disappear at very low pH values and are not I coo- l Although this antibiotic has already been determined by microbiological and enzymic assays,14-16 and by HPLC with UV detection,l7-19 there are no reports of its reduction at electrodes or the possible application of electrochemical techniques to its determination.In this work we used differential-pulse polarography, direct-current polarography and cyclic voltammetry to determine cefsulodin. Experimental Cefsulodin, sodium salt, was supplied by Abell6. All other reagents used were of analytical-reagent grade from Merck. The background electrolyte used was a buffer solution consisting of 0.1 M each of acetic, phosphoric and boric acids. * Presented at the International Symposium on Electroanalysis and Sensors in Biomedical, Environmental and Industrial Sciences, Cardiff, UK, 6-9 April, 1987. t To whom correspondence should be addressed. -1200 -1000 -800 -600 -1400 -1200 -1000 -800 EImW F'ig. 1.Differential-pulse rlarograms for cefsuldin (15 p.p.m.) in a 0.1 M Britton - Robinson uffer of pH (a) 2.97 and (b) 7.05. Ionic strength 0.7 M, temperature 25 "C24 ANALYST, JANUARY 1988, VOL. 113 0.6 0.2 0.1 0 . 0 1 1 1 ” ” I J 1 2 3 4 5 6 7 PH Fig. 2. Variation of the mean current of peak A in Fig. 1 as a function of pH. For conditions, see Fig. 1 I .A- 0.00 - 0 1 2 3 4 5 6 Timelh Fig. 3. Variation of the peak current as a function of time for (A) cefsulodin, E = -1150 mV, and (B) for its hydrolysis product isonicotinamiie, E = - 1300 mV, in a 0.1 M Britton - Robinson buffer of pH 11.05. Inihil cefsulodin concentration, 15 p.p.m.; ionic strength, 0.7 M; temperature, 25 “C observed in basic media owing to their overlap with the background electrolyte discharge. The current remains virtu- ally constant throughout the pH range in which they are observed.The current of peak A is markedly dependent on pH (Fig. 2) and its maximum corresponds to approximately pH 3. The decrease in current above this pH is concomitant with an increase in the peak width (Fig. l), hence its area remains virtually constant. Another peak (D) appears above pH 6 at a potential about 150 mV more negative than that of peak A. Its appearance is concurrent with a decrease in the width of the main peak (A). This peak is due to a degradation product of cefsulodin and its intensity increases, at any pH, in cefsulodin from vials that have been opened for a few days. Above pH 9 the current of all the peaks is dependent on time.Peak A decreases gradually over time as peak D increases (Fig. 3). The effect of concentration on the peak current was studied in an acidic medium, where the antibiotic is stable. The results obtained show that the variation of the current with the concentration is linear only up to 60 p.p.m. [Fig. 4(a)], above which the graph curves, reaches a maximum. and then decreases and becomes almost constant above 150 p.p.m. The non-linearity of the variation of the peak current with the concentration above 60 p.p.m. is probably due to adsorption phenomena as it can be corrected by merely adding a surfactant to the medium. In fact, the addition of 5% Triton X-100 results in a linear variation of the peak current with the concentration as can be seen from Fig.4(b), which shows the results obtained at pH 3. Concentration, p.p.m. Fig. 4. (a) Variation of the current of peak A in Fig. 1 with cefsulodin concentration; (b) after addition of 5% Triton X-100; and (c) variation of the limiting current of the first d.c. wave with concentration. Conditions: 0.1 M Britton - Robinson buffer of pH 2.97; ionic strength, 0.7 M; and temperature, 25 “C 1 ’ I --__ -1200 -1000 -800 -600 ElmV Fig. 5. Direct-current polarograms of cefsulodin at a concentration of (a) 25 p.p.m. and (b) 75 p.p.m. in a 0.1 M Britton - Robinson buffer of pH 2.97. Ionic strength, 0.7 M; temperature, 25 “C Direct-current Polarography Using DCP, cefsulodin yields three reduction waves in acidic media. The first wave, i.e., the least cathodic, is the best defined of all three (Fig.5) and shows a polarographic maximum at concentrations below 60 p.p.m.ANALYST, JANUARY 1988, VOL. 113 25 0.2 30-0.1 0 1 2 3 4 5 6 7 8 9 10 PH Fig. 6. Variation of the limiting current of the first d.c. wave of cefsulodin (15 p.p,m.) in a 0.1 M Britton - Robinson buffer as a function of the pH. Ionic strength, 0.7 M, temperature, 25 "C -680 - > -690 - -710 - I I I I I 0.0 0.5 1 .o 1.5 2.0 Log (concentration, p.p.m.1 Fig. 7. Variation of the half-wave potential of the first d.c. wave of cefsulodin as a function of its concentration in a 0.1 M Britton - Robinson buffer of pH 2.97. Ionic strength, 0.7 M, temperature 25 "C The other two cathodic waves correspond to peaks B and C obtained by DPP. Both tend to disappear in basic or very acidic media through overlap with the background electrolyte discharge.It should also be noted that the sum of the heights of these two waves is roughly equal to the height of the first wave, We shall only consider the first wave, corresponding to peak A obtained by DPP. The limiting current of this wave remains virtually constant below pH 6 (Fig. 6), whereas it decreases above this value. Above pH 8, its height becomes constant and its current is approximately half that observed in an acidic medium (Fig. 6). In acidic media the half-wave potential shifts to more cathodic values as the pH increases. The variation of Ei as a function of pH is linear with a slope of -70 mV per pH. The diffusion current vanes linearly with the concentration throughout the range investigated (1-200 p.p.m.) [Fig.4(c)]. The effect of the concentration on the half-wave potential is shown in Fig. 7. As can be seen, there is a considerable change in the polarographic behaviour at 60 p.p.m. Below this level E4 shifts to less cathodic values as the concentration increases; the effect is the opposite at high concentrations. The variation of the logarithm of the limiting current with that of the dropping time at pH 3 is linear, with a slope of 0.22, clearly indicative of a diffusion-controlled process. The fact that such a slope is slightly steeper than the theoretically expected slope (0.17) can be attributed to the influence of the depolariser adsorption on the electrode. The fact that this first wave is controlled by diffusion allows the number of electrons involved (four per molecule) to be determined by microcoulometry at pH 3 and also the potential (-850 mV) corresponding to the diffusion plateau of this wave.Although the number of electrons involved was not determined for the other two waves observed at more cathodic potentials, their heights indicate the possibility of two electrons being exchanged in each. Cyclic Voltammetry The voltammetric reduction of cefsulodin is markedly influen- ced by its adsorption on the hanging mercury drop electrode. Hence, its peak current is not proportional to the concentra- -0.8 -1.0 -1.2 -1.4 EN Fig. 8. Cyclic voltammogram of cefsulodin (15 p.p.m.) in a 0.1 M Britton - Robinson buffer of pH 4.02. Ionic strength, 0.7 M, temperature, 25 "C. Only the fist two scans at 60 V s-* are shown.The dotted line represents the charge (.$ idt) corresponding to the first scan tion, and its behaviour can only be accounted for on the basis of adsorption isotherms. The reduction peaks obtained are completely irreversible, even at scan rates of the order of 500 Vs-1. The potentials at which they appear are more cathodic than those in the other techniques used. Fig. 8 shows the first two scans of the cyclic voltammogram, run at pH 4 for a cefsulodin concentration of 15 p.p.m. and at a scan rate of 60 V s-1. The recordings allow two interesting conclusions to be drawn. Firstly, only those cefsulodin molecules adsorbed on the electrode are in fact reduced, as can be inferred from the constant signal obtained in the second scan of the voltammo- gram.Secondly, the first reduction process, seen as a four-electron process using DCP, appears as two two-electron processes when using cyclic voltammetry. The integration of the current (Fig. 8) therefore yields four waves of roughly the same height corresponding to the exchange of two electrons each. The two more cathodic signals correspond to peaks B and C obtained by DPP. Discussion The variation of the current of peaks A and D with time, obtained in basic media, could be attributed to a hydrolysis reaction typical of this type of antibiotic, which results in the opening of the p-lactam ring, cleavage of the molecule and, consequently, detachment of the side-chain linked to the carbon atom at position 3. Accordingly, peak D, which increases with time, could be attributed to the reduction of the isonicotinamide formed in the hydrolysis.In fact, we have found that the addition of isonicotinamide to the medium increases the peak height in every instance. It seems reasonable to attribute the main peak (A) yielded by the antibiotic to the reductim of the isonicotinamide substituent in the c e f s u l o d i d c u l e ' because the reduction - - , potential ofcefsulodin (peak A) and that of free isonicotin- - - amide (peak D) are very close, and the DCP reduction w-- yielded by the antibiotic is a four-electron wave in acidic media. The behaviour of this main wave is consistent with that of isonicotinamide.20 Both reduction processes are probably similar, although they are clearwfferent in-d!a owing to the difficulty of protonating the>t- antibiotic molecule.26 ANALYST, JANUARY 1988, VOL.113 Isonicotinamide yields a four-electron wave between pH 4 and 6 owing to the reduction of the amide group (Scheme I). Q C O/'H 1 C OH HO' I \H H 0 / The change in the electrochemical behaviour of cefsulodin above 60 p.p.m. is indicated by the disappearance of the polarographic maximum, a shift of the half-wave potential to more negative values and in a non-linear variation of the peak current with the concentration. This behaviour suggests that the dropping-mercury elec- trode could be fully covered with antibiotic molecules at concentrations above 60 p.p.m. This also occurs with the hanging mercury drop electrode, even at much lower concen- trations, when sufficient time is allowed for the adsorption equilibrium to be reached. All these data indicate that the isotherm describing the behaviour of the antibiotic is more complex than a straightforward Langmuir isotherm.Finally, it should be noted that the two two-electron peaks clearly observed using DPP and cyclic voltammetry (B and C), appear at potentials close to the reduction potential of 7-aminocephalosporanic acid, indicating that one of the peaks is probably due to the reduction of the double bond in the dihydrothiazine ring. Scheme I However, below pH 4 the isonicotinamide four-electron wave decreases owing to the protonation of the hydrated form of the aldehyde, the reduction of which does not take place at acidic pH. At acidic pH, the wave yielded is a two-electron wave. This effect is also observed in basic media, although in this instance it is the result of the dissociation of the hydrated form; at these pHs the dissociation equilibrium is strongly displaced to the right.The effect is also seen in the reduction of cefsulodin in basic media. However, a four-electron wave is still observed below pH 6, even at very acidic pH, which can be accounted for by the difficulty of protonating the nitrogen atom in cefsulodin as it forms a link between the isonicotinamide group and the rest of the molecule. This allows the reduction of the aldehyde, a process which is hindered by the protonation of the nitrogen atom in free isonicotinamide. This assumption is supported by the voltammetric behavi- our of isonicotinamide. The cyclic voltammogram recorded at pH 4 shows two two-electron peaks of a highly irreversible character, as also occurs with cefsulodin.These peaks correspond to the sole four-electron wave obtained for isonicotinamide by DCP. The determination of the antibiotic is possible, at least over the concentration range investigated (namely between 1 and 200 p.p.m.), by DCP at any pH < 7. However, at very low concentrations, the precision of the measurements is affected by the appearance of a polarographic maximum below 60 p.p.m. Hence, DPP should be preferentially used for very low concentrations of antibiotic. The determination should ideally be carried out at pH 3, at which the main peak reaches its maximum current and is optimally defined and stable. The antibiotic can be determined in the range 1-60 p.p.m., which can be further extended to 200 p.p.m. by adding a surfactant such as Triton X-100 (5%) to the medium.1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. The authors express their gratitude to Laboratorios Abelld S.A., who kindly supplied the cefsulodin used in this work, and to the Junta de Andalucia for financial support received through the grant 07/CLM/MDM, 85-87. References Ochiai, M., Aki, O., Morimoto, A., Okada, T., Shinozaki, K., and Asahi, Y., J. Chem. SOC. Perkin Trans., 1, 1974,258. Hall, D. A., J. Pharm. Sci., 1973, 62, 980. Fogg, A. G., Fayad, N. M., Burgess, C., and McGlynn, A., Anal. Chim. Acta, 1979, 108,205. Jones, I. F., Page, J. E., and Rhodes, C. T., J . Pharm. Pharmacol., 1968,20,455. Siegerman, H., Methods Enzymol., 1975, 43, 373. Rickard, E. C., and Cooke, G. G., J. Pharm. Sci., 1977, 66, 379. Ivaska, A.. and Nordstrom, F., Anal. Chim. Acta, 1983, 146, 87. Camacho, L., Avila, J. L., Heras, A. M., and Garcia-Blanco, F., Analyst, 1984,109, 1507. Fogg, A. G., Fayad, N. M., and Burgess, C., Anal. Chim. Acta, 1979, 110, 107. Fogg, A. G., and Fayad, N. M., Anal. Chim. Acta, 1980,113, 91. Fogg, A. G , Fayad, N. M., and Goyal, R. N., J. Pharm. Pharmacol., 1980,32,302. Fogg, A. G., Fayad, N. M., and Martin, M. J., Anal. Proc., 1981, 18,211. Fogg, A. G., and Martin, M. J., Analyst, 1981, 106, 1213. Nomura, H., Fugono, T., Hitaka, T., Minami, I., Azuma, T., Morimoto, S., and Masuda, T., J . Med. Chem., 1974,17,1312. Nomura, H., Minami, I., Hitaka, T., and Fugono, Y., J. Antibiot., 1976, 29, 428. Cullmann, W., Dick, W., and Edelmann, M., J. Clin. Chem. Clin. Biochem., 1985,23, 151. Lecaillon, J. B., Rouan, M. C., Souppart, C., Febvre, N., and Juge, F., J. Chromatogr., 1982,228,257. Elrod, L., Jr., White, L. B., Wimer, D. C., and Cox, R. D., J. Chromatogr., 1982,237, 515. Das Gupta, V., and Stewart, K. R., J. Clin. Hosp. Pharm., 1984,9,21. Lund, H., Acta. Chim. Scand., 1963, 17,2325. Paper A71140 Received April 8th, 1987 Accepted June 26th, 1987

ISSN:0003-2654    

DOI:10.1039/AN9881300023

出版商:RSC    

年代:1988

数据来源: RSC

摘要:

ANALYST, JANUARY 1988, VOL. 113 21 Differential-pulse Adsorptive Stripping Voltammetry of the Psychotropic Drugs Triazolam and Clotiazepam" Rosa M. Alonso and Rosa M. Jimenez Departamento de Quimica, Facultad de Ciencias, Universidad del Pais Vasco, 48080 Bilbao, Spain and Arnold G. Fogg Chemistry Department, Lo ug h borough University of Technology, Loug h borough, Leicestershire LEI1 3TU, UK The surface-active properties of the psychotropic drugs triazolam and clotiazepam at a hanging mercury drop electrode allow their sensitive determination by differential-pulse adsorptive stripping voltammetry. The adsorptive stripping response has been evaluated with respect to accumulation time, accumulation potential, concentration dependence, electrolyte snd pH. Determination limits are 6 x 10-10 M for triazolam and clotiazepam with accumu!ation times of 4 and 10 min, respectively.Coefficients of variation at the 1 x 10-8 M levels (six determinations) were typically < I .5%. Keywords : Differential-pulse adsorptive stripping voltammetry; triazolam; clotiazepam The benzodiazepines are one of the most important groups of psychotropic drugs,l and many new drugs in this group have been developed by the pharmaceutical companies. Triazolam and clotiazepam belong to the triazolobenzodiazepine and thienodiazepine families, respectively, have properties typical of the benzodiazepines and are used as tranquillisers.2-4 The increasing need to determine low concentrations of these drugs has led us to evaluate their determination by adsorptive stripping voltammetry.There has been an increased interest in the use of adsorptive stripping voltammetry in the last few years5Jj owing to the availability of modern hanging mercury drop electrode (HMDE) devices that are easy to use. The technique is applicable to those compounds that are adsorbed strongly on mercury, the first step in the technique being the adsorptive accumulation of the compound on the surface of the HMDE. The adsorbed material is then determined voltammetrically. The technique has been shown to be applicable to a wide range of pharmaceutical compounds. Experimental Adsorptive and voltammetric experiments were performed using a Metrohm 646 VA processor in conjunction with a VA 647 stand which incorporates a multi-mode electrode and has the facility of easy, rapid and reproducible production of a hanging mercury drop electrode as required in the voltammet- ric cell.The three-electrode system was completed by a silver - silver chloride reference electrode and a glassy carbon auxiliary electrode. Stripping was carried out in the differen- tial-pulse mode using a 50 mV pulse amplitude and a 10 mV s-1 sweep rate (steps of 10 mV at 1-s intervals). Unless otherwise indicated the medium-sized drop (0.40 mm2) and the medium stirrer speed (1920 rev min-1) were used. Initial de-oxygenation was carried out for 12 min: subsequent de-oxygenations of the same solution between adsorptive cycles were carried out for only 10 s. Reagents Samples of triazolam and clotiazepam were kindly supplied by Upjohn Farmoquimica and Esteve Laboratories, both in * Presented at the International Symposium on Electroanalysis and Sensors in Biomedical, Environmental and Industrial Sciences, Cardiff, UK, 6-9 April, 1987.Spain. Stock solutions of triazolam (2.92 x 10-3 M) and clotiazepam (3.10 x 10-3 M) were prepared in methanol and stored refrigerated in the dark. Buffer solutions of pH 4,7 and 10 were prepared from analytical-reagent grade sodium acetate - acetic acid, disodium hydrogen phosphate - sodium dihydrogen phosphate and sodium carbonate - hydrochloric acid. A 0.1 M sulphuric acid stock solution was prepared and a 10-2 M sulphuric acid solution was used as an electrolyte. Results Differential-pulse adsorptive stripping voltammograms for triazolam and clotiazepam in different media, viz 10-2 M sulphuric acid, acetate, phosphate and carbonate buffers, obtained with no accumulation time and with 2 min of accumulation time at an accumulation potential of -400 mV, are shown in Figs.1 and 2. The adsorptive accumulation results in significant peak current enhancements for both drugs compared with the response without accumulation. As can be seen in Figs. 1 and 2, the adsorption behaviour is strongly affected by the pH and the electrolyte used. Both compounds showed the highest signal in neutral and alkaline media. Such behaviour may be attributed in part to the acid equilibria of these drugs738 (pKa triazolam = 2.20, PKa clotiazepam = 4.25). The neutral form exhibits strong -25 E 00 00 Q) I - 25 0 -1.000 0 -1.400 0 - 1.400 PotentialN PotentialN Potent la IN Fig.1. Differential-pulse adsorptive strip ing voltammograms of triazolam (5.8 x 10-8 M) in different media. pa) 10-2 M acetate buffer; (b) M hos hate buffer; (c) 10-2 M carbonate buffer. Accumula- tion time: &) 8 and (B) 2 minANALYST, JANUARY 1988, VOL. 113 28 -7 -6 -5 -4 P -3 -2 -1 0 - 1 .ooo PotentialN 0 - 1.200 Potential/V -17.5 r ( C) > -75 F I 8 40 - -125 - -10 - 2 P a - 7 5 - 1 1 1 0 -1.400 0 - 1.600 PotentiallV PotentialIV Fig. 2. Differential-pulse adsorptive stripping voltammograms of clotiaze am (6.2 X 10-8 M) in different media. ( a ) 10-2 M sulphuric acid; (IIP~O-~ M acetate buffer; (c) 10-2 M phosphate buffer, pH 7; (d) M carbonate buffer, pH 10. Accumulation time (A) 0; and (B) 2 min L -400 -500 -600 -700 -800 Accumulation potential/mV Fig.3. Effect of the accumulation potential on the differential-pulse adsorptive stripping peak height for a 5.8 x 10-8 M triazolam concentration in 10-2 M carbonate buffer, pH 10. Accumulation time 2 min. Peak potential, -1 109 V adsorption around the potential of zero charge. The hydrolysis reaction of triazolam at pH values lower than three did not allow a study of its accumulation behaviour in sulphuric acid. Carbonate buffer of pH 10 was chosen as the most suitable electrolyte to carry out the adsorptive3ri ping stuay;. gs m e d i d . awrptive behaviour of these psyclxxtrapic drugs-wawrried -1 but. The/dependence of the adsorptive stripp-peak current T clotiazeparn, in particular, showed a higher a : sormm in>tlps - A study of the different variables whicli'Influence The ' 30 I - I O 1 -boo -so0 -8Loo -1doA Accumulation potentiahv Fig.4. Effect of the accumulation otential on the differential-pulse adsorptive stripping peak height 8 r a 6.2 x 10-8 M clotiazepam concentration in 10-2 M carbonate buffer, pH 10. Accumulation time, 2 min. Peak potential, -1.294 V 30 0 30 90 750 210 270 Accu m u latron t i me/s Fig. 5. Effect of accumulation time on the differential-pulse adsorptive stripping peak height for triazolam solutions in carbonate buffer, pH 10. Accumulation otential, -700 mV Triazolam concentration: (A) 5.8 X M, &) 2.3 X M; (c) 1.2 X M 50 - 0 - 2 4 6 8 Accumulation time/mrn Fig. 6. Effect of the accumulation time on the differential-pulse adsorptive stripping peak height of clotiazepam solutions in carbonate buffer, pH 10.Accumulation potential, -800 mV. Clotiazepam concentration: (A) 6.2 X 10-8 M; (B) 3.1 X 10-8 M; (C) 1 2 X 10-8 M on the accumulation potential for triazolam and clotiazepam in carbonate buffer is shown in Figs. 3 and 4. The use of an accumulation potential that is close to the peak potential but still gives the full signal is convenient as it reduces the analysis time; accumulation potentials of -700 and -800 mV were chosen for the determination of triazolam and clotiazepam, respectively. The effect of accumulation time on the peak currents obtained at three different concentrations of both compounds is shown in Figs. 5 and 6 . For the highest concentration, the current increased linearly up to 150 s (triazolam) and 6 min (clotiazepam) with a levelling off of the signal at higher aecuxritdation times.For the lower concentration a rectilinear relationship between peak current and accumulation time was obtained. The different adsorptive behaviour of the two complmncbwith respect to the accumulation time is signifi- cant. For the higher concentration, triazolam reached aANALYST, JANUARY 1988, VOL. 113 29 0 10 20 Concentration/l0-8 M Fig. 7. Calibration graphs for (A) triazolam and (B) clotiazepam in pH 10 buffer. Accumulation potentials: triazolam, -700; and clotiazepam, -800 mV. Accumulation time, 120 s -25 - -20 . -15 % k - 10 -5 0 - 7.400 Potentia IN -8 B PotentialN Fig. 10. Differential-pulse adsorptive stripping voltammograms for triazolam (5.8 x 10-10 M) in carbonate buffer, pH 10.Accumulation time: (A) 0; and (B) 4 min Fig. 8. Differential-pulse adsorptive stripping voltammograms for different triazolam concentrations. Accumulation time, 120 s. Accu- mulation potential, -700 mV. Triazolam concentration: (A) 0; (B) 1.2 x 10--x M ; (C) 3.5 x 10 8 M; and (D) 5.8 X 10 -* M 0 -0.8 - 1.600 Potent i a I N -30 -25 - 20 2 & -15 -10 -5 - 30 - 25 -20 2 a -15 -10 -5 0 -1.600 0 - 1.600 Potent i a l/V Potenti a l/V Fig. 9. Differential-pulse adsorptive stripping voltammograms for different clotiazepam concentrations. Accumulation time, 120 s. Accumulation potential, -800 mV. Clotiazepam concentration: (A) 0; (B) 1.2 x 10 H M; (C) 2.5 x 1 0 - 8 M; (D) 3.7 X M; (E) 4.9.5 X 10-8 M ; (F) 6.2 x 10 M M; and (G) 7.4 x Fig. 11. Differential-pulse adsorptive stripping voltammograms for clotiazepam 6.2 x 10-10 M) in carbonate buffer, pH 10.Accumula- tion time: (A\ 0; (B) 5; (C> 10; and (D) 1.5 min significant situation close to full surface coverage in 150 s whereas clotiazepam needed at least 7 min. Peak currents were plotted against triazolam and clotiaze- pam concentration for an accumulation time of 120 s and at an accumulation potential of -700 (triazolam) or -800 mV (clotiazepam). Rectilinearity of the signal was obtained up to at least a concentration of 2.2 x 10-7 M for both compounds (see Fig. 7). Typical adsorptive stripping voltammograms for different concentrations of the benzodiazepine derivatives studied are shown in Figs. 8 and 9. Five determinations at the 1.34 x 10-8 M level of triazolam gave a mean peak height of 9.04 nA with a coefficient of variation of 1.4% and for a 1.86 X 10-8 M clotiazepam Concentration a mean peak height of 4.30 nA with a coefficient of variation of 1.4%.The effective pre-concentration associated with the adsorp- tion process results in extremely low determination limits, a clearly distinguishable peak for 5.8 x 10-10 M triazolam concentration being obtained with an accumulation time of 4 min (Fig. 10). An accumulation time of 10 min is necessary to determine clotiazepam at a level of 6.2 x 10-10 M (Fig. 11).30 ANALYST, JANUARY 1988, VOL. 113 Discussion From the studies made it is evident that voltammetric measurements of psychotropic drugs can be improved by accumulation at the hanging mercury drop electrode, allowing their determination at the nanomolar level with short accumu- lation times.In carbonate buffer, the compounds studied showed a difference between their peak potentials of about 200 mV, so that both drugs could be determined simultaneously, taking into account their different adsorptive behaviour with respect to accumulation time. The adsorption rate of triazolam was higher than that of clotiazepam, but their sensitivity was fairly similar using an accumulation time of 2 min. The adsorptive behaviour of these benzodiazepine deriva- tives is in agreement with the results obtained by Kalvoda9 for diazepam and nitrazepam. The applications of the method developed by adsorptive stripping voltammetry to clinical samples would require the study of the different interferences present in the samples, especially those interferences which show surface-active properties. Separation or clean-up proce- dures may be necessary. 1. 2. 3. 4. 5. 6. 7. 8. 9. References Shutz, H., “Benzodiazepines: A Handbook,” Springer-Verlag, Berlin, 1982. Nakinishi, M., Tsumagari, T., Takigawa, Y., Shuto, S . , Kenjo, T., and Fukada, T., Arzneim-Forsch. Drug Res., 1972, 22, 1905. Gall, M., Kamdar, B. V., and Collins, R. J., I . Med. Chem., 1978, 21, 1290. Wang, R. I. H., and Stockdale, S. L., J . Int. Med. Res., 1973,1, 600. Wang, J . , “Stripping Analysis: Principles, Instrumentation and Applications,” Verlag Chemie, Deerfield Beach, FL, 1985. Wang, J . , Int. Lab., 1985, 68. Jimenez, R. M., PhD Thesis, University of Bilbao, 1985. Alonso, R. M. ~ Fernandez-Arciniega, M., and Hernandez, L.. Quim. Anal., 1987, 6, 218. Kalvoda, R., Anal. Chim. Acta, 1984, 162, 197. Paper A71161 Received April 27th, 1987 Accepted August 5th, 1987

ISSN:0003-2654    

DOI:10.1039/AN9881300027

出版商:RSC    

年代:1988

数据来源: RSC

摘要:

ANALYST, JANUARY 1988, VOL. 113 31 Application of Adsorptive Stripping Voltammetry for a Study of the Reaction of Mouse IgG with Anti-mouse IgG* Malcolm R. Smytht and Eileen Buckley School of Chemical Sciences, NIHE Dublin, Glasnevin, Dublin 9, Ireland Juana Rodriguez Flores Department of Analytical Chemistry and Electrochemistry, University of Exfremadura, Badajoz, Spain and Richard O'Kennedy School of Biological Sciences, NIHE Dublin, Glasnevin, Dublin 9, Ireland The adsorptive stripping voltammetric behaviour of mouse immunoglobulin G (IgG) and anti-mouse IgG has been optimised with respect to accumulation potential, accumulation time and scan-rate. It has been possible to monitor the reaction of these two immunoglobulins directly in solution using this technique. Keywords: Adsorptive stripping voltammetry; immunochemical reaction; mouse lgG Adsorptive stripping voltammetry (AdSV) has received much attention in recent years for the determination of a variety of inorganic and organic species which can undergo adsorptive accumulation at electrode surfaces, either directly or follow- ing a suitable complexation reaction.192 This technique has also found application in studies involving biological macro- molecules such as DNA.3 In a recent paper, Rodriguez Flores and Smyth4 reported the results of an investigation of the use of AdSV for a study of the interaction of human serum albumin (HSA) with anti- HSA.This study showed the possibility of using this technique to monitor the interaction of an antigen with its antibody directly in solution.In this paper we report our results on a system which involves the reaction of two immunoglobulin (IgG) species, one of which has been raised as an antibody to the other. Experimental Reagents All compounds used were of analytical-reagent grade and solutions were prepared in water obtained by passing distilled water through a Milli-Q water purification system. A 0.1~ phosphate buffer solution (pH 7.4) was prepared using sodium dihydrogen orthophosphate and disodium hydrogen ortho- phosphate. Mouse IgG was obtained from Sigma. The antibody to this immunoglobulin was raised in a donkey according to the method of Dwyer et af.5 The anti-IgG was then purified by repeated precipitation with saturated ammonium sulphate followed by affinity separation on a CNBr-activated Sephar- ose column which contained bound mouse IgG.The bound antibody was eluted from the column using 0.1 M glycine - HC1 (pH 2.5). The presence of protein in each fraction (1 ml) was detected using a modified micro-Bradford assay. The absor- bance of each fraction was determined at 600 nm using a Bio-tek EL-307 ELISA reader.5 The purified anti-mouse IgG was then characterised by sodium dodecyl sulphate poly- * Presented at the International Symposium on Electroanalysis and Sensors in Biomedical, Environmental and Industrial Sciences, Cardiff, UK, 6-9 April, 1987. t To whom correspondence should be addressed. acrylamide gel electrophoresis (SDS-PAGE) and high-perfor- mance liquid chromatography (HPLC). Apparatus A Waters Model 6000A HPLC pump was operated in conjunction with a Waters Protein Pak 300 SW gel filtration column for HPLC separations. Detection was carried out using a Waters Model 440 absorbance detector. Adsorptive stripping voltammograms were obtained on a Princeton Applied Research Corp.(PARC) Model 264A Polarographic Analyser combined with a PARC Model 175 Universal Programmer, a PARC Model 303 mercury electrode system, a a PARC Model 305 magnetic stirrer and an Omnigraphic Model 2000 X- Y recorder. A pulse amplitude of 50 mV was used throughout. Procedures Before each voltammetric investigation, the buffer was purged with oxygen-free nitrogen for 8 min before adding the required amount of immunoglobulin to the cell and purging for a further 2 min. Care was taken to ensure that the purging did not cause significant frothing of the protein solution.Following the purging step, the electrode potential was set at the required accumulation potential (E,) for the required accumulation time (tact) with stirring, before the stirrer was switched off automatically and the voltammetric experiment carried out. Results and Discussion Preparation of Anti-mouse IgG The anti-serum used in this study was raised in a donkey against mouse IgG.5 This anti-serum was then purified by salt precipitation and affinity chromatography on a column of CNBr-activated sepharose to which mouse IgG had been immobilised. The use of the affinity chromatography purifi- cation ensures that the donkey antibody fraction is specific for mouse IgG. A typical affinity chromatogram is shown in Fig.l(a). The purity of the anti-mouse IgG was checked by SDS electrophoresis and HPLC. As shown in Fig. l(b), the purified material consisted of only one peak with a similar retention time to pure IgG. Therefore, other non-specific immunoglo- bulins and proteins had been removed to give a pure, specific antibody fraction,32 ANALYST, JANUARY 1988, VOL 113 12 - Z l O - 8 8 - - 5 6 - e 4 - n a 2 - 0 10 20 30 40 50 60 70 80 90 100 Fraction No. 0 4.0 8 0 12.0 16.0 20.0 Ti me/m I n Fig. 1. (a) PIot of absorbance vs. fraction num-er following purification of antiserum by affinity chromatography. (b) HPLC profile of (A) pure anti-mouse IgG and (B) affinity-purified anti- serum. Flow-rate, 0.5 ml min-1 Adsorptive Stripping Voltammetry of IgG The adsorptive stripping voltammetric behaviour of mouse IgG is shown in Fig.2(a). Using an accumulation potential of 0.0 V (vs. Ag- AgC1) and a scan-rate-of 10 mV s-1, a 2.0 mg 1-1 solution of the immunoglobulin gave rise to two peaks at (A) -0.25 V and (B) -0.56 V which increased in size with increasing concentration. The effects of accumulation poten- tial, accumulation time and scan-rate on the differential-pulse peak currents and peak potentials of these two peaks are collated in Table 1. This behaviour is similar to that reported for the antibody to human serum albumin.4 From this it can be seen that the optimum conditions for the analysis of mouse IgG are EaCc 0.05 V and scan-rate 10 mV s-1. Greater sensitivity is obviously achieved by using a longer accumula- tion time, but this in itself leads to a longer analysis time.Using an accumulation time of 400 s and the optimum conditions cited above, a linear calibration graph was obtained for mouse IgG between 1.0 and 10.0 mg 1-1 with a slope of 0.17 FA 1 mg-1 (B). Adsorptive Stripping Voltammetry of Anti-IgG The adsorptive stripping voltammetric behaviour of anti-IgG is shown in Fig. 2(b). Using an accumulation potential of 0.05 V and a scan-rate of 10 mV s-1, a 1.0 mg 1-1 solution of the immunoglobulin gave rise to two peaks at (A’) -0.23 V and (B’) -0.53 V which increased in size with increasing concen- tration. This behaviour is similar to that of mouse IgG and of anti-HSA.4 The optimum conditions regarding E,,, and scan-rate for the analysis of this immunoglobulin were the same as for mouse IgG.In this instance however, the peak currents for peaks A’ and B’ approached a limiting value when the accumulation time was 500 s or longer [Fig. 2 ( b ) ] . Using an accumulation time of 400 s, a linear calibration graph was again obtained between 1.0 and 10.0 mg 1-1 with a slope of 0.16 FA 1 mg-1 for this immunoglobulin. Reaction of Mouse IgG with Anti-mouse IgG The reaction of mouse IgG with anti-mouse IgG was monitored directly in solution by adsorptive stripping voltam- I I I J 0 0.3 0 6 0 9 -EN vs Ag - AgCl 0.0 0.2 0.4 0.6 0 8 -EN VS. Ag - AgCl Fig. 2. Effect of accumulation time (s) on the adsorptive stripping voltammetric behaviour of (a) mouse IgG and (b) anti-mouse IgG. In both instances the lower curve represents the supporting electrolyte Table 1.Influence of accumulation potential, accumulation time and scan-rate on peak potentials and peak currents of peaks A and B of mouse IgG Peak A Peak B t,,J EJ v/ id -Epl 14 -Epl S V mVs-1 p A V PA v 100 0.00 10 0.14 0.23 200 0.00 10 0.17 0.25 300 0.00 10 0.26 0.25 400 0.00 10 0.38 0.25 500 0.00 10 0.41 0.25 900 0.00 10 0.62 0.25 400 0.10 10 0.29 0.24 400 0.05 10 0.38 0.25 400 -0.05 10 0.23 0 25 400 -0.10 10 0.08 0,26 400 0.00 2 0.19 0.20 400 0.00 5 0.27 0.25 400 0.00 20 0.23 0.27 0.09 0.56 0.14 0.56 0.23 0.56 0.27 0.56 0.38 0.56 0.74 0.56 0.29 0.56 0.29 0.56 0.16 0.57 0.08 0 60 0.04 0.57 0.12 0.60 0.24 0.56 cp I 0.20 0 15 [Anti-lgG]:mg I Fig. 3. Graph of decrease in peak current of peak A and peak B of mouse IgG with increasing concentration of anti-mouse IgGANALYST, JANUARY 1988, VOL.113 33 metry using the optimum conditions of Eacc and scan-rate described previously. An accumulation time of 50s was employed in this study in order to decrease the time of the experiment and to minimise the possible denaturation of the mouse IgG - anti-mouse IgG complex under the effect of the Earn chosen. When anti-IgG was added at concentrations of 1.0-6.0 mg 1-1 to a solution of 8.0 mg 1-1 mouse IgG, the peak currents of peaks A and B both decreased. This is illustrated in Fig. 3, where it can be seen that peak A decreases in a two-stage process, whereas peak B decreases linearly with respect to concentration. This is not completely analogous to the results obtained for the reaction of HSA with anti-HSA,4 where a decrease was only observed in one of the peaks due to HSA at -0.68 V.However, it should be remembered that the structures of HSA and mouse IgG are not similar, HSA being a globular protein and mouse IgG a typical Y-shaped immunoglobulin with two heavy chains and two light chains. When the study was repeated adding anti-HSA to mouse IgG instead of the specific anti-mouse IgG, increases were obtained in the peak currents around -0.25 V and -0.55 V, as would be expected if no interaction was occurring between the two proteins in solution. Therefore, it would appear that this approach involving the use of adsorptive stripping voltammetry could be a useful method for studying immunochemical reactions directly in solution. Work is in progress to investigate more fully the nature of the electrochemical reactions giving rise to the voltammetric behltviour reported here. References 1. Wang, J., “Stripping AnaIysis: Principles, Instrumentation and Analysis,” Verlag Chemie, Deerfield Beach, FL, 1985. 2. Franklin Smyth, W., in Smyth, M. R., and Vos, J. G., Editors, “Electrochemistry, Sensors and Analysis,” Elsevier, Amster- dam, 1986, pp. 29-36. 3. Sequaris, J. M., and Valenta, P., in Franklin Smyth, W., Editor, “Electroanalysis in Hygiene, Environmental, Clinical and Pharmaceutical Chemistry,” Elsevier, Amsterdam, 1980, 4. Rodriguez Flores, J., and Smyth, M. R., J. Electroanal. Chern., in the press. 5. Dwyer, G., Gilvary, U., Carty, P., O’Kennedy, R., and Thornes, R. D., Biochem. Sac. Trans., 1987,15,293. Paper A7121 7 Received May 29th, I987 Accepted July 14th, 1987 pp. 9e112.

ISSN:0003-2654    

DOI:10.1039/AN9881300031

出版商:RSC    

年代:1988

数据来源: RSC

摘要:

ANALYST, JANUARY 1988, VOL. 113 35 Evaluation of Iridium Oxide Electrodes Formed by Potential Cycling as pH Probes* Michael L. Hitchmant and Subramaniam Ramanathan Department of Pure and Applied Chemistry, University of Strathclyde, 295 Cathedral Street, Glasgow GI 1x1, UK Iridium oxide electrodes formed by potential cycling in dilute acids were investigated for their use as pH sensors. It was shown that both the short- and long-term stabilities of such sensors were adequate for applications where an accuracy of k0.2 pH is acceptable. It was also found that thicker oxide films which give responses of up to about 80 mV pH-’ are to be preferred in order to ensure reproducible behaviour. As potentiometric probes either for direct monitoring of pH or for following typical acid - base titrations, these electrodes give comparable results to those obtained with glass electrodes.Several basic concepts relating to the practical management of potential-cycled iridium oxide electrodes are discussed. Keywords: pH measurement; pH sensors; iridium oxide electrodes pH is one of the most important electrochemical parameters in solution chemistry. The glass electrode is by far the most commonly used sensor to indicate degrees of acidity or alkalinity in aqueous electrolytes. In recent years there has been an increasing interest in utilising metal - metal oxide electrodes192 as pH probes for industrial process control and biomedical monitoring. The advantages of such probes include greater probe rigidity and increased scope for mini- aturisation.For a metal - metal oxide electrode to be useful as a pH sensor, it should satisfy several important criteria.3 For example, the metal must be noble enough to resist attack by all solutions over a useful pH range, and its oxide must be electrically conducting and be able to come into equilibrium with the solution without appreciable dissolution. As these basic requirements are somewhat difficult to achieve, most of the metal - metal oxide electrodes used for pH indication represent a trade-off between the nobility of the metal and the stability of the oxide. Among metal - metal oxide electrodes, the antimony electrode is by far the most widely used pH sensor,4-6 but another oxide electrode which has been investigated recently is that of iridium.7 Iridium oxide electrodes for pH measure- ments have been prepared thermally,s by sputtering9 or electrochemically. 10 The electrochemical technique is pro- bably the simplest with scope for readily regulating the rate of growth and thickness of the oxide film. de Rooij and Bergveldlo were the first to report on the use of iridium oxide electrodes formed by potential cycling in dilute acids as pH probes.A linear response was observed in their chosen pH range 6.5-8.0. The electrode also had a low impedance and low temperature coefficient. Later, Burke and co-workers11J2 reported on electrochemical investigations of thicker oxide films formed on iridium by cycling in various media and also highlighted their pH sensing properties and their anion sensitivities. A recent paper by Kinoshita et al.,13 focuses on a comparison be tween mono- and polycrystalline iridium oxide electrodes formed by potential cycling. A linear response is observed in the pH range 2.5-8.5 and an oxidised monocrys- talline electrode is claimed to have an improved response time over an oxidised polycrystalline electrode. In this paper further results are presented on the use of cycled iridium oxide electrodes as pH sensors. In particular it * Presented at the International Symposium on Electroanalysis and Sensors in Biomedical, Environmental and Industrial Sciences, Cardiff, UK, 6-9 April 1987. t To whom correspondence should be addressed. is shown that for the optimum behaviour of iridium oxide electrodes prepared by potential cycling, films with high charge storage capacities should be used. The drift charac- teristics of the electrodes, which would be an important consideration in determining the precision attained over operating pH ranges, have also been assessed systematically and a comparative study of the performance of the sensors with a glass electrode has been made.These investigations, together with the earlier results of de Rooij and Bergveld,lo Burke11 and Kinoshital3 should be useful in formulating a practical basis for using iridium oxide electrodes as pH sensors. Experimental A conventional electrochemical cell was used for voltammet- ric studies. A PAR Model 363 potentiostat - galvanostat controlled by a PAR Model 175 universal programmer was used to perturb the electrochemical interface with a triangular voltage waveform.The current and potential were recorded on a Houston Omnigraphic 2000 X - Y recorder. The working electrode was a length of iridium wire of 99.9% purity (Goodfellow Metals), 1.0 mm diameter and 8.5 mm exposed length. It was spot-welded to a length of platinum wire for ohmic contact to the external circuit. The junction was encapsulated with heat-shrunk PTFE tubing (Farnell Electronics). It has been our experience that PTFE insulation is superior to epoxy resin insulation. Preliminary experiments conducted with an iridium wire sealed with conductive cement in an aluminium holder and then insulated with epoxy resin showed excessive drifting behaviour under open circuit conditions, although the voltammetric characteristics appeared unaffected. A probable explanation for this increased drift is that the integrity of the seal was disturbed as a result of the strain caused by polishing of the wire between experiments, thereby exposing the substrate support to action by the electrolyte through minute crevices at the iridium - resin interface, even though visually the seal appeared to be strain-free. The PTFE-coated electrode was soaked in hot 5 M H2S04 for a few hours in order to remove any residual trace of hydrous oxide formed during previous experiments.14 It was then polished using standard metallographic techniques with a series of diamond pastes and finishing with a 1-pm paste.After polishing the wire was degreased in acetone and washed with copious amounts of doubly distilled water before introducing it into the electrochemical cell.The counter electrode was also made of iridium in order to minimise contamination of the test solution by any foreign material.36 A double-junction calomel electrode (Ingold) containing saturated KC1 as the inner electrolyte and 0.5 M H2S04 as the outer electrolyte was used to complete the three-electrode configuration. A double junction reference electrode was used because of possible corrosion of the iridium oxide by chloride ions leaking into the cell electrolyte. However, as all potential measurements were relative and not absolute, then, provided that the double liquid junctions were always the same, the use of a double junction did not cause any problems. A Luggin probe with the tip placed at about 1 mm from the surface of the working electrode was used to minimise ohmic drop effects in the electrolyte.The potential of the iridium electrode was cycled in 0.5 M H2S04 between -0.25 and +1.25 V vs. SCE at 3 V s-1 generally for 8000 cycles (see discussion below). For voltam- metric studies the number of cycles was varied. The chosen limits correspond to the onset of hydrogen and oxygen evolution, respectively. Electrochromism~~ was distinctly observed for low thicknesses of the oxide. Prior to pH measurement, the potential of the iridium oxide electrode was scanned at a slow rate of 30 mV s-1 to the potential maxima of the main charge storage peak, whereupon the scan was terminated. 11 The pH measuring instrumentation consisted of a Corning Model 160 pH/ion meter. It had an impedence of 1012 and the current drawn was typically 1 PA.Provision also existed for automatic temperature compensation. The level of resolu- tion achieved in the potentiometric mode of operation was k O . 1 mV. A voltage output from the pH meter could be sampled and stored automatically by a Keithley 175 digital multimeter. This procedure was frequently used for long-term stability tests of the sensor in various buffer solutions. A five-channel electrode selector (Corning), when coupled to the pH meter, allowed the manual selection of any one of up to five sensing electrodes with a common reference electrode. During potentiometric titration the glass electrode was monitored on one channel while the iridium oxide electrode was monitored on a second channel. Thus the pH at any point of a titration was effectively indicated by the glass electrode simultaneously to that monitored by an iridium oxide elec- trode. Buffers were prepared according to standard procedures.16 Their pH values were checked with a glass electrode previ- ously calibrated in precision buffers of pH 4.00 and 7.00 (Corning). To evaluate the stirring sensitivity, the potential of the cycled electrode was allowed to attain a steady-state value. Then nitrogen was bubbled through a porous f i t into the solution and the variation in potential was observed. All reagents used were of AnalaR grade and the water used was doubly distilled. Potential values quoted are referenced against the SCE. Measurements were all made at 25.0 k 0.1 "C. Results and Discussion Representative current - potential profiles as iridium is progressively cycled to a thicker oxide layer are shown in Fig.1. Interpretation of the voltammograms is fairly well established.*7-20 The anodic peak A2 and cathodic peak C2 could be attributed to an Ir"1 - IrIv redox transition17 associated with the reaction 21r02 + 2H+ + 2e E Ir203 + H20 . . (1) where E (mV vs. SCE) = 681.0 - 59.1 pH. For a completely reversible system with unit proton activity and with both oxidised and reduced forms adsorbed, the peak potentials would correspond to P . 2 1 In fact, at low oxide thicknesses the peak potentials for both A2 and C2 do correspond reasonably to the literature value22 for the potential for the redox equilibrium [reaction (l)], although at higher charge storage capacities the oxidation peak is shifted anodically and the reduction peak cathodically .During the anodic process ANALYST, JANUARY 1988, VOL. 113 5 1 II -0.2 0.2 0.6 1 .o EImV vs. SCE Fig. 1. Typical voltammograms as recorded at different stages of hydrous oxide growth on cycling the potential of an iridium electrode in 0.5 M H2S04; scan rate, 30 mV s-1; growth conditions, -0.25 to + 1.25 V vs. SCE at 3 V s- 1 ; number of cycles indicated on each curve 24 t L 20 16 b a a 12 a 8 4 I I I I I I 1 I 1 I 0 20 40 60 80 100 120 140 160 180 200 Volume of NaOH/ml Fig. 2. Differential potentiometric titration curves for titration of H3P04 with NaOH for, A, glass and B, iridium oxide electrodes electrons are transferred from the oxide to the metal substrate and simultaneously an equivalent amount of negative charge is injected into the oxide across the oxide - electrolyte interface in order to minimise the development of space charge effects inside the oxide.Conversely, during the cathodic process electrons are injected into the oxide from the metal while an equivalent amount of negative charge is ejected across the oxide - electrolyte interface. It is known that the higher oxide is a good electronic conductor23 whereas the lower oxide is poorly conducting.19 The build-up of an insulating oxide on the surface would account for the shift in peak potentials of A2 and C2 and the apparent loss of reversibility at higher charge storage values. However, as discussed below, the situation is rather more complicated than suggested by the equilibrium of reaction (1).The anodic peak Al which is more apparent at lower charge storage capacities and which occurs as a shoulder on the major anodic peak A2 is not evident at higher thicknesses of the oxide film; there are indications that the species giving rise to this transition are complexes involving the anion of the acid. 12 The cathodic peak C3 which occurs at about 1.13 V, and whose anodic counterpart can frequently be seen as a shoulder in the oxygen evolution region, may correspond to the electroreduction of a higher valent oxide.12 Although complications owing to the contribution of higherANALYST, JANUARY 1988, VOL. 113 37 valent oxides can, in principle, be minimised by cycling to a lower anodic limit, it is found in practice that growth of an oxide film is inhibited below a certain critical anodic limit.24 The performance of an electrode cycled 8000 times and used for monitoring a H3P04 - NaOH titration parallels that of the glass electrode except that at the equivalence point the oxidised iridium electrode gives a slightly larger potential break on account of its increased sensitivity (Fig.2). The corresponding Nernst plots are shown in Fig. 3. From reaction (l), the slope is expected to be 59 mV pH-1, that is, a ratio of protons to electrons (p/e) transferred of 1 : 1. Least-squares correlation of 28 data points for an iridium oxide electrode gives a value of 1.38 : 1, which clearly cannot be explained by the equilibrium (1). The least-squares value of 801.6 4 10.0 mV for the apparent Eo obtained from Fig.3 for the cycled electrode is also somewhat different from the EOvalue of 681.0 mV predicted by Pourbaixz for reaction (1). In order to rationalise these observations a hydrated equivalent of reac- tion (1) can be formulated11 2[Ir02(OH)2-,.(2 + x)H20]Q-~)- + (3 - 2x)H+ + 2e [Ir203(OH)3.3Hz0]3- + 3H20 . . . . . . . . (2) The oxidised and reduced forms are now both hydrated oxyhydroxides of IrlII and IrIV. For the results in Fig. 3, a value of x of 0.12 is required. A detailed analysis of reaction (2) is being presented elsewhere.25 What is important to note here for the evaluation of iridium oxide electrodes as pH sensors is that the slope of the Nernst plots for electrodes subjected to an increasing number of growth cycles reaches a limiting value of about 81 mV pH-1.Fig. 4 shows Nernst slopes (or p/e) plotted as a function of charge storage capacity of the oxide films; charge storage capacities, which can be considered as a measure of oxide thickness, were obtained from the cyclic voltammograms by integrating the cathodic profiles between the limits 0.15 and 1.25 V.12 Apart from the fact that high charge storage capacity films give slightly improved electrode sensitivity compared with a glass electrode, clearly it will be better to use such films as their Nernstian behaviour is not critically dependent on the number of growth cycles used in their preparation. It is interesting to note in this context that the lower Nernst slopes of 65 mV pH-1 obtained by de Rooij and Bergveldlo and 62-68 mV pH-1 by Kinoshita ef aZ.13 for polycrystalline and monocrystalline iridium were all obtamed with only a few hundred potential growth cycles.Another advantage of using electrodes which have been oxidised with a 41 600 400 W u cn 9 200 > E Q C -2w 0 4 8 12 PH Fig. 3. Variation in potential of A, glass and B, iridium oxide electrodes with pH during titration of 1.0 M H3P04 with 1.0 M NaOH large number of potential cycles is that as charge storage capacity reflects oxide thickness these electrodes will have thick oxide films and so some measure of film corrosion and dissolution can occur without a significant change of sensitiv- ity; certainly thin oxide films in prolonged contact with alkaline solutions have been found to show significant decreases in the Nernstian slope with time." It is clear that high storage capacity films will be expected to show more stable and reproducible electrochemical characteristics than low capacity films and so the evaluation discussed here of cycled iridium electrodes for pH sensing has been restricted to such high capacity films.The extent to which a pH sensor is serviceable depends to a large extent on the rapidity with which it develops a potential proportional to the hydrogen ion activity in solution and maintains this level for long periods of time. Any drift of potential should be within limits of the degree of precision of pH required. Fig. 5 shows the stability plots with various buffer solutions obtained for an electrode freshly cycled 8000 times in 0.5 M H2S04. Except in more alkaiine solutions, the drift of the electrode is not unreasonable.For example, after a conditioning period of 10 min the drift for the rest of the hour was found to be usually no greater than 2 mV for pH in the range 1.96-8.30. For pH 9.98 the drift amounts to about 6 mV whereas for pH 12.01 it is about 47 mV. The behaviour of the same electrode aged in distilled water for a month is shown in Fig. 6. For pH 1.96 to 8.30 there has been no significant change in the drift characteristics, but at higher pH values the drift has decreased. For pH 9.98 the drift after an initial conditioning period of 10 min is 0.5 mV for the 1.40 0 .- c. 1.30 2 CI 1.20 $ 1.10 p t 4- 0 0.4 0 8 1.2 1.6 2.0 2.4 Log(O/mC cm-*) Fig. 4. Dependence of Nernst slope and proton : electron ratio on charge storage capacity 61 0 ptf 4.02 pH 6.02 4 4 0 20 40 60 Timehin Fig. 5.Time dependence of electrode potential in various buffer solutions for an electrode freshly cycled 8000 times in 0.5 M H2S0438 580 ANALYST, JANUARY 1988, VOL. 113 pH 1.97 - t 4 5 0 t pH 4.02 pH 6.01 0 v) pH 8.31 pH 9.90 - 20 t -150? -170 0 20 40 60 Tirnelmin Fig. 6. Time dependence of electrode otential in various buffer solutions for an electrode aged in distillegwater for 1 month 0 20 40 60 80 1 00 Time/h Fig. 7. Long-term stability performance of the iridium oxide electrode in pH 6.02 buffer at 25 "C rest of the hour whereas for pH 11.99 it is about 24 mV-an almost two-fold reduction over the rate of drift before ageing. The general trend of improved drift characteristics after ageing in distilled water is probably associated with the degree of solvation attained by the oxide particles becoming more uniform which will necessarily affect the energetics of any equilibria; X-ray analysis27 has shown that the water content of a freshly cycled electrode decreases on passing from the oxide - electrolyte interface to the metal - oxide interface.The drift of open circuit potential of a freshly cycled electrode will therefore be due to changes in the film stoicheiometry [cf. equation (2)]. As already mentioned the greater drift which occurs at high pH values is probably due to the solubility of the hydrous oxide film.26 The longer-term stability of the sensor in buffer solution is adequate for many purposes. For example, Fig. 7 shows that the drift of electrode potential in pH 6.02 buffer from the immersion value is a maximum of 8 mV in over 100 h of operation.For applications where an accuracy of no better than kO.1 pH is required this would be acceptable. The Nernst plots for both the freshly cycled and aged electrodes are shown in Fig. 8. A slight reduction in E" from 781.8 +_ 28.1mVto745.3 k 38.1mVandinslopefrom76.3 k 4.1 mV pH-1 to 75.5 k 5.4 mV pH-1 are the two noticeable features arising after ageing in distilled water. In particular, the change in EO highlights the importance of obtaining a calibration plot prior to a measuring sequence. However, the good linearity of the plots means that a two-point calibration, as typically made with a glass electrode, would be sufficient. It is interesting to note that the value of the Nerstian slopes are lower than the slope of 81.0 mV pH-1 obtained from the titration of H3P04 with NaOH.One possible explanation for 600 - 400 - tn > 200 - E ia 0 - -200 4 8 12 PH Fig. 8. Variation in potential of iridium oxide electrode with pH of buffer slution for A , freshly cycled electrode and B, same electrode aged in distilled water for 1 month 4 8 12 Glass electrode pH Fig. 9. Comparison of pH responses of glass and iridium oxide electrodes this is that the nature of the anion in the buffering electrolyte has an effect. Burke" has suggested that the anion sensitivity may be due to some incorporation of anions in oxy-anion coordination sites at the peripheral regions of the hydrous material. Such incorporation evidently will affect the resulting open circuit potential of the electrode and therefore the values of EO and of the slope may vary slightly.The reproducibility of electrodes prepared by potential cycling was found to be reasonable. For example, for two identically prepared electrodes the Nernst plots, constructed on the basis of open circuit potential measurements in various buffer solutions, gave E" values for the freshly cycled electrodes which did not differ by more than 5 mV (781.8 and 777.3 mV) and slopes which only showed a discrepancy of about 1 mV pH-1 (76.3 and 77.5 mV pH-1). After a month's ageing in distilled water, the EO values of the two electrodes still showed good agreement (745.3 and 749.2 mV) as did the slopes (75.5 and 75.6 mV pH-1). The potential drift during the first hour for both freshly cycled and aged electrodes was practically identical with the results already presented in Figs.5 and 6. A comparison of the performance of a cycled electrode with that of a glass electrode was made. For this experiment both the glass and cycled electrodes were calibrated in Corning precision buffers at pH 4.00 and 7.00. The electrodes were then used to sense the pH of a series of buffer solutions in the range 2-12. In both the calibration and test solutions the pH for both electrodes was read after 10 min conditioning. Fig. 9 shows the correlation. The correlation coefficient obtained forANALYST, JANUARY 1988, VOL. 113 39 I I I 0 40 80 Volume of redox titranvml Fig. 10. Variation in potential of A. Dlatinum and B. iridium oxide ektrodes during iedox titratiin of FeS0,.7H20 with Ce(SO&.4H2O the pH range 2.01-12.10 is 0.9989. These results indicate again that a cycled electrode can readily measure the pH of a solution to within 0.1 pH or better.The effect of stirring on the potential of a cycled electrode by bubbling nitrogen into the test solution was found to be a rapid displacement of about 1 mV from the equilibrium value of the potential, essentially in agreement with Burke’s resultsll where an invariant displacement was recorded. This very small dependence of the electrode potential on both stirring and oxygen partial pressure suggests that there is no significant oxidative corrosion of the electrode and this is a desirable feature for a potentiometric electrode. However, a stronger sensitivity to other redox couples is found.Fig. 10, for example, shows a comparison between a cycled electrode and a platinum electrode in monitoring the course of a redox titration. Although the platinum electrode is the more sensitive indicator electrode, the fact remains that the almost parallel performance of the cycled electrode precludes its use as a pH sensor in electrolytes containing redox couples. This is in agreement with the observations of other workers.l3*2* Conclusions From the results presented in this paper it can be seen that a potential cycled iridium electrode can be satisfactorily used as a pH sensor. A sensitivity of about 80 mV pH-1 is obtained, but nevertheless performances parallel those of a glass electrode in direct pH monitoring of solutions and following the course of typical acid - base titrations.Both short-term and long-term stabilities of the sensor in various buffer solutions are adequate for many applications, particularly where an accuracy of better than ItO.1 pH is not required. For optimum behaviour in all applications thick oxide films are desirable to ensure reproducible behaviour. In this context it is interesting to note that while, in principle, an oxidised electrode could be prepared by cycling its potential in a solution of any pH, in practice it is found that only a limited range of solutions are possible, e.g., H2S04, NaOH and Na2C03. Dissolution of the electrochemically formed oxide predominates if the acid electrolyte is too concentrated (e.g., 5 M HZS04) so that oxide growth is rather slow.17.26 The morphology of the hydrous films26 also depends on the electrolyte used for cycling. For example, the degree of dispersion of the oxide is considerably greater in the example of the acid-grown material as compared with that grown in alkali, but the base-grown films are less adherent than their acid-formed analogues.On balance, a pH electrode is probably best formed by cycling in dilute acid. It is also interesting to note that from the work of Edwall4-6 on solid state pH electrodes based on single crystal antimony, one might expect a monocrystalline iridium electrode to exhibit improved behaviour. However, extensive cycling of the potential of a monocrystalline electrode between the limits corresponding to the onset of hydrogen and oxygen evolution will disturb the integrity of the lattice to a significant-extent so that the electrode will become increasingly polycrystalline.This probably explains why Kinoshita ef al. 13 were not able to observe any differences under a scanning electron microscope between cycled electrodes prepared from polycrystalline and monocrystalline iridium. The major disadvantage of a poten- tial cycled iridium oxide electrode for pH monitoring as compared with a glass electrode would appear to be the high sensitivity to redox couples, although oxygen does not have a significant effect on response, probably as a result of the very low exchange current normally associated with oxygen couples. The support of an ORS award for S. R. is gratefully acknowledged.We are also pleased to acknowledge financial support for this work from Ingold Messtechnik AG, and many useful discussions with Drs. R. Bucher and H. Buehler of that company. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. References Niedrach, L. W., J. Electrochem. SOC., 1980,127,2122. Ask, P., Edwall, G., Johanssen, K. E., and Tibbling, L., Med. Biol. Eng. Comput., 1982,20,383. Ives, D. J. G., and Janz, G. J., “Reference Electrodes,” Academic Press, New York, 1969, Chapter 7. Edwall, G., Electrochim. Acta, 1979,24,595. Edwall, G., Electrochim. Acta, 1979, 24, 605. Edwall, G., Electrochim. Acta, 1979,24,613. Papeschi, G., Bordi, S., Beni, C., and Ventura, L., Biochim. Biophys. Acta, 1976,453, 192. Macur, R.A., US Pat. 1348912, 1974. Yuen, M. F., Lauks, I., and Dautremont-Smith, W. C., Solid State Ionics, 1983,11, 19. de Rooij, N. F., and Bergveld, P., “Proceedings of the International Conference on Monitoring of Vital Parameters During Extracorporeal Circulation, Nijmegen 1980,” Karger, Basle, 1981, p. 156. Burke, L. D., Mulcahy, J. K., and Whelan, D. P., J. Electroanal. Chem., 1984, 163,117. Burke, L. D., and Whelan, D. P., J. Electroanal. Chem., 1984, 162,121. Kinoshita, E., Ingman, F., Edwall, G., Thulin, S., and Glab, S . , Talanta, 1986,33,125. Gottesfeld, S . , and McIntyre, J. D., J. Electrochem. SOC., 1979, 126,742. Gottesfeld, S . , McIntyre, J. D. E., Beni, G., and Shay, J. L., Appl. Phys. Lett., 1978, 33, 208. Weast, R. C., Editor, “CRC Handbook of Chemistry and Physics,” 59th Edition, CRC Press, Boca Raton, FL, 1979, D-186. Rand, D. A. J., and Woods, R., J. Electroanal. Chem., 1974, 55,375. Burke, L. D., and Whelan, D. P., J. Electroanal. Chem., 1980, 124, 1467. Glarum, S . H., and Marshall, J. H., J. Electrochem. SOC., 1980, 127,1467. Conway, B. E., and Mozota, J. M., Electrochim. Acta, 1983, 28, 9. Southampton Electrochemistry Group, “Instrumental Methods in Electrochemistry,” Ellis Horwood, New York, 1985, Chapter 6. Pourbaix, M., “Atlas of Electrochemical Equilibria,” National Association of Corrosion Engineers, Houston, TX, 1974, Ryden, W. D., Lawson, A. W., and Sartain, C. C., Phys. Lett. A, 1968,26,209. Hall, H. Y., and Sherwood, P. M. A., J. Chem. SOC., Faraday Trans. I . , 1984,80, 135. Hitchman, M. L., and Ramanathan, S. R., submitted for publication. Burke, L. D., and Scannel, R. S . , J. Electroanal. Chem., 1984, 174,119. McIntyre, J. D. E., Basu, S . , Peck, W. F., Brown, W. L., and Augustyniak, W. M., Phys. Rev B, 1982,25,7242. Fogg, A., and Buck, R. P., Sensors Activators, 1984,5,137. Paper A71262 Received June 25th, 1987 Accepted August 24th, 1987 p. 397.

ISSN:0003-2654    

DOI:10.1039/AN9881300035

出版商:RSC    

年代:1988

数据来源: RSC

摘要:

ANALYST, JANUARY 1988, VOL. 113 41 Determination of Copper(l1) with a Carbon Paste Electrode Modified with an Ion-exchange Resin* Lucas Hernandez,t Pedro Hernandez, Manuel H. Blanco and Maria Sanchez Departamento de Qulinica, Universidad Aut6noma de Madrid, 28049-Madrid, Spain The determination of copper(l1) was studied using a carbon paste electrode modified with an ion-exchange resin (Dowex SOW-X8) using differential-pulse voltammetry. The conditions necessary for the pre- concentration of copper were determined in an open circuit by ion exchange; copper levels down to 18 nM were determined. The method was applied to the determination of copper in pharmaceutical products. Keywords : Copper(//) determination; modified electrode; carbon paste electrode; ion-exchange resin; differentia f -pulse vo f tammetry In electroanalysis the most commonly used technique in the determination of elements at trace levels is undoubtedly the stripping method, as it is the most sensitive.However, numerous parameters influence its use and these are sometimes difficult to control in order to achieve excellent reproducibility of the working conditions; this directly influ- ences the measurements performed. With a view to minimising these physical factors, interest is growing in replacement of the mercury drop electrode or mercury film by electrodes whose active surface is more reproducible or, if not, is not critical. At the same time one also aims at avoiding the problems of diffusion towards the inside of the electrode and of the formation of metallic complexes.Further, electrodes are sought that can be used for elements that cannot be determined with mercury electrodes, either because no amalgam is formed because the substance is not adsorbed on its surface in the pre-concentration or accumulation phase, or because it is not electroactke in the range of its use. These aims can be partially achieved by using carbon electrodes, in particular the glassy carbon variety, or using electrodes composed of modified graphite. With this kind of electrode it is possible to conserve the advantages of great sensitivity and selectivity, owing to a selective pre-concentra- tion, and to eliminate to a large extent the drawbacks mentioned above. Pre-concentration on the modified electrode takes place through a non-electrochemical process, after which the analyte accumulated on the surface is measured by simple voltammetry.These non-electrochemical mechanisms of pre- concentration include complexing,1-5 ion exchange637 and covalent bonding.8 In this paper we describe the behaviour of a carbon paste electrode modified with an ion-exchange resin used for the determination of Cu2+ ions; the method was chosen owing to the simplicity of the construction of the electrode, with direct incorporation of suitable amounts of an ion-exchange resin in the carbon paste. This incorporation offers a beneficial alternative for the accumulation of ions because one is dealing with a cationic resin in the acid formed and the pre-concentra- tion reaction may be expressed as nRS03 + Mn+ c-) (RS03),M + nH+ according to the efficiency of the pre-concentration phase of the distribution coefficient. The measurement phase is performed by application of a potential to the electrode, in this instance producing reduc- tion of the ion retained: (RSO&M + 2e- c--) 2R2- + MO 2RS03- + 2H+ (or 2K+) c, 2RHS03 * Presented at the International Symposium on Electroanalysis and Sensors in Biomedical, Environmental and Industrial Sciences, Cardiff, UK, 6-9 April, 1987.t To whom correspondence should be addressed. Experimental Copper(II) solution. Prepared from electrolytic copper. Spectroscopic graphite. Particle size less than 42 pi. Dowex 5OW-X8 ion-exchange resin in the protonated form. Water de-ionised using a Milli-Q and Milli-R system Reagents and Materials (Millipore) was employed throughout. Apparatus An Amel 448 oscillopolarograph with a Hewlett-Packard Model 862 X - Y recorder and a Metrohm Model E-506 Polarecord polarograph were used.Throughout the study two cells of 50 ml were used, one for carrying out the pre-concentration and the other for measur- ing; the latter was equipped with a platinum counter electrode and a saturated calomel reference electrode (SCE) . As the working electrode a mixture of carbon paste and Dowex SOW-XS ion-exchange resin was employed, using as the support a polyethylene tube with a geometric surface area of 1.62 mm2. Contact with the carbon paste was made with a platinum wire. A Metrohm Model 654 pH meter was used. Preparation of Electrode To a weighed amount of spectroscopic graphite activated electrothermally is added vaseline at a graphite to vaseline ratio of 1 : 1 m/V; this forms the carbon paste. To this product is added moist, ground ion-exchange resin and the mixture is homogenised, then introduced into the polyethylene support- ing tube, achieving direct contact with the platinum wire.Determination of Copper A 25.0-ml volume is taken from the solution containing copper, adjusted to the conditions appropriate for each study and with constant stirring, and into it is introduced the electrode at previously chosen times under open-circuit conditions. The electrode is then placed in the measuring cell, which contains the supporting electrolyte, and the intensity - poten- tial curve is recorded between +0.2 and -0.9 V against SCE when a linear potential scan is made and between +0.2 and -0.4 V in differential-pulse voltammetry.Having performed the measurement, the electrode is regenerated by successive potential sweeps in both techniques until it reaches the initial residual current that it has in the absence of copper. The regeneration of the electrode in differential-pulse voltammetry can be achieved with the same effectiveness by maintaining the electrode at a potential of -0.8 V or at the peak potential for 5 min. The carbon paste is changed daily and the measurement solution after every ten determinations.42 ANALYST, JANUARY 1988, VOL. 113 0.2 0.0 -0.2 -0 4 -0.6 -0.8 EN Fig. 1. Voltammograms obtained at different scan rates: A, 250; B, 150; C, 100; D, 60; and E, 40 mV s-1 Results and Discussion The studies conducted to establish the procedure for the determination of copper(I1) with a carbon paste electrode modified with an ion-exchange resin consisted of two steps, one directed towards establishing the pre-concentration con- ditions and the other towards carrying out the measurement itself.In the former step linear scanning voltammetry was used with the aim of choosing the most suitable conditions for the retention of the ion because, in spite of knowing these conditions,g such characteristics are modified as the grain size of the resin decreases and because it is mixed with the carbon paste. In order to be able to follow the retention process it is essential to establish the optimum response conditions in the linear scanning technique; to do so a study was made of its stripping variables in a non-complexing support electrolyte consisting of 1 M KN03 at pH 7.4.The reduction of the retained copper ions converts them into metallic copper with an Ep close to -0.2 V; reversibly, the variation in Ep with the scan rate is in agreement with what is expected for this kind of system. Fig. 1 shows the modification of the voltammograms at different scan rates, with an increase occurring in the peak as the scan rate increases and showing a linear variation with V, the residual current increasing slightly with the scan rate. For rates greater than 200 mV s-l, distortion of the waves takes place, which hinders measurements; its use is therefore not recommended. Throughout the process of ion exchange the pH of the solution to be determined plays an important role because, as is well known, the proton concentration of the solution affects the resin - solution balance and pH is therefore an important variable.It was found (Fig. 2) that for acidic solutions of pH <1 no copper ions are retained on the electrode; maximum retention is observed at pH 2.0; the retention decreases rapidly with increase in pH and shows an absence of waves at pH higher than 10. The conditions for maximum retention differ from those found for the same resin when used in a column'(); this is logical in view of the fact that the particle size had been altered. Another variable that must be established is the time during which the electrode should be kept in contact with the solution 1 I 1 I I 0.2 0.0 -0.2 -0.4 -0.6 -0.8 EN Fig.2. Effect of sample pH in pre-concentration cell: (A) 2.0; (B) 4.0; (C) 6.0; and (D) 10.0 1 1 I I I I 0 1 2 3 4 5 Time/min Fig. 3. Effect of pre-concentration period on peak current from 1 vg ml-1 CU*+ solution Table 1. Influence of proportion of resin on peak intensity Resin, % i&A MFrA i, - iR/pA 2.0 13.2 3.2 10.0 3.9 33.0 9.0 24.0 7.8 53.0 12.0 41 .O 10.7 53.0 40.0 13.0 16.7 - 80.0 - to be determined. Fig. 3 shows the variation in peak intensity with contact time; an increase can be seen in peak intensity up to 120 s, the time at which exchange equilibrium is reached. Variations in the stirring rate do not influence the process; as long as stirring is maintained, no control is necessary. An increase in the proportion of resin in the carbon paste will endow the electrode with a greater capacity for exchange but, owing to the greater conductivity of the resin, this increases the resistance of the electrode, which will lead to an increase in the residual current. With a view to studying this variable, electrodes were prepared with varying proportions of resin and pre-concentration of copper(II), and measurement by linear potential scanning under the optimum conditions found earlier were performed.The results are shown in Table 1, from which it may be inferred that the proportion of resin that provides the highest peak intensity is 7.8%.ANALYST, JANUARY 1988, VOL. 113 - - -0 2 43 50 30 e 3 10 ; PH 0 4 8 12 Fi .4. Dependence of Ep and ip on pH in measuring cell: (A) ip; and ( 4 Ep 0 20 40 60 80 Concentration/ng ml-1 F ~~ 1.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 E N Fig.5. Differential-pulse voltammo rams obtained with increasin copper concentration: (A) 1; (B) 5; &) 10; (D) 30; (E) 50; and (Ff 80 ng ml-1 Regarding the measurement step, it is first essential to establish the optimum conditions for the measurement solu- tion, choosing the supporting electrolyte and pH of the solution. The best electrolyte found was 1 M KN03, as it gave the best defined waves; no modifications were observed when KC104 or Na2HP04 was used at the same concentration. Fig. 4 shows the variation in potential and peak intensity found on modifying the pH of 0 . 5 ~ KN03 solution. A small variation can be seen in Ep at pH 11.0, leading to the reduction of copper(I1) retained in the resin at more negative potentials and with a decrease in ip for pH values greater than 9.0, this being the pH at which the maximum peak intensity is found.These observed modifications may be accounted for in terms of the influence that the concentration of OH- ions may exert on the mechanism of the reduction process, as the substance is converted into metallic copper and an increase in OH- ions produces a competition between the ions and the exchange groups that affects the reduction process in the electrode - solution interphase. The following are the most probable mechanisms: - R2Cu + 2e- + 2K+ c--) 2RK + Cuo (for pH c9.0) RZCU + 40H- + 2K+ 2RK + (OH)&U*- t C U O +JiEH- (for pH >9.1) This would account for the reduction at more negative potentials, owing to losses in the interphase produced by the diffusion of copper ions from the resin towards the solution because of the formation of hydroxy complexes.Differential-pulse voltammetry was chosen for the measure- ment step, performing the retention of 80 ng ml-1 CuII solutions at pH 2.0 in solutions stirred for 120 s, the measurement cell containing 0.5 M KN03 (pH 9.1). The best results were obtained with a scan rate of 8 mV s-1 and a pulse amplitude of -60 mV. Increases in pulse amplitude, although they do produce greater intensities, also increase the residual current and the width of the peak, and peaks are deformed at AE = -100mV. Having chosen the optimum conditions, measurements were performed for different concentrations of copper( 11). Fig. 5 shows the results obtained for different solutions of copper(II), with a linear behaviour for concentrations between 1.0 and 80 ng ml-1, which after treatment by least-squares yields ip (pA) = 0.42 + 0.026 ng ml-1 with a correlation coefficient r = 0.995.From the statistical study it is concluded that the results obtained have an error of 1% and a relative standard deviation of 6%. The method was applied to the determination of copper in pharmaceutical products containing as major species camphor, undecylenic acid, boric acid, resorcinol, salicylic acid and zinc sulphate. A good correlation was found between the results obtained with the proposed method and those obtained using atomic absorption spectrometry. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. References Cheek, G. T., and Nelson, R. F., Anal. Lett., 1978,11,393. Cox, J. A., and Majda, M., Anal. Chem., 1980,52, 861. Lubert, K. H., Schnurrbusch, M., and Thomas, A., Anal. Chim. Acta, 1982,144, 123. Izutsu, K., Nakamura, T., Takizawa, R., and Hanawa, H., Anal. Chim. Acta, 1983,149, 147. Guadalupe, A. R., and Abruma, H. D., Anal. Chem., 1985, 57,142. Cox, J. A., and Kulesza, P. J., Anal. Chim. Acta, 1983,154,71. Wang, J., Greene, B., and Morgan, C., Anal. Chim. Acta, 1984,158,15. Price, J. F., and Baldwin, R. P., Anal. Chem., 1980,52,1940. Minczewewski, J., Chwastowska, J., and Dybczynski, R. , “Separation and Preconcentration Methods in Inorganic Trace Analysis,” Ellis Horwood, Chichester, 1982. Baldwin, R. P., Christensen, J. K., and Kryger, L., Anal. Chern., 1986,58,1790. Paper A7lI33 Received April 6th, 1987 Accepted September 15tk, I987

ISSN:0003-2654    

DOI:10.1039/AN9881300041

出版商:RSC    

年代:1988

数据来源: RSC

摘要:

ANALYST, JANUARY 1988, VOL. 113 45 Copper( 11)-selective Electrodes Based on Macrocyclic Polythiaethers* Satsuo Kamata,t Kuzutomo Yamasaki, Morihide Higo, Ajay Bhale and Yumi Fukunaga Department of Applied Chemistry, Faculty of Engineering, Kagoshima University, Korimoto, Kagoshima 890, Japan PVC membrane electrodes that are selective and sensitive to copper(l1) have been developed. The best electrode, based on 13,14-benzo-I ,5-tetrathiacyclopentadecane and 2-nitrophenyl octyl ether solvent mediator with a potassium tetrakis(p-chloropheny1)borate anion excluder, exhibits a linear response in the range 10-1-10-6 M over the pH range 2.9-6.7 and has a Nernstian slope of 28-29 mV per decade at 25 "C. The response time is 10 s. Keywords: Macroc yclic pol ythiaethers ion sensor; neutral carrier; copper(l1)-selective electrodes Much interest has been shown in the co-ordination chemistry of macrocyclic polythiaether ligands .1-3 Polythiaethers exhibit a remarkable ability to form complexes with transition metal cations.In this paper, we describe the use of 13,14-benzo-1,5- tetrathiacyclopentadecane (BTTCP) and 7,8-benzo-l,S- dithiacyclononane (BDTCN) as neutral carriers in ion-selec- tive electrodes. As reported previously,3 14- and 15-mem- bered ring structures form complexes with copper ions. BTTCP has a 15-membered ring cavity and k ~ e gne molecule would be expected to form a complex with copper, whereas BDTCN which has a nine-membered ring, requires two molecules to form a complex with copper(I1). We therefore investigated the use of these compounds as neutral carriers in ion-selective electrodes.Experimental Reagents Propane-l,3-dithiol, 2-chloroethanol, thiourea, a,a'-di- bromo-o-xylene, dioctyl phthalate (DOP) and poly(viny1 chloride) (PVC) were obtained from Nakarai Chemicals, Japan. 2-Nitrophenyl octyl ether (NPOE) and potassium tetrakis(pchloropheny1)borate (KTCPB) were obtained from Dojin Kagaku, Japan. All the compounds were of analytical- reagent grade, including the metal chlorides of copper, cadmium, cobalt and iron. Preparation of Compounds BTTCP and BDTCN (Fig. 1) were synthesised using the methods of Rosen and Busch4 and Allen et al. ,5 respectively. The structures of these compounds were confirmed by elemental analysis, IR and NMR spectroscopy. The proper- ties of the compounds are as follows.BTTCP: m.p. 87-88 "C; yield 60%; found C 54.38, H 7.01, S 38.53%; calculated C 54.49, H 6.71, S 38.80%; TR 2925(-CH2-), 1490(-C=C-), BTTCP Fig. 1. Macrocyclic polythiaethers BDTCN * Presented at the International Symposium on Electroanalysis and Sensors in Biomedical, Environmental and Industrial Sciences, Cardiff, UK, 6-9 April, 1987. t 'To whom correspondence should be addressed. 705(-C-S-); NMR <C;3Clj) 6 ?.h4(s:4ArH), 5.9X(s,4H), 7.18(s,t,12H), 8.08(p,2H). BDTCN: m.p. 88-90 "C; yieiu 40%; found C 62.59, I1 6.68, S 30.58'%,; calculated C 62.90, €3 NMR (CDC13) T 2.90(s,4ArH), 6.20(s,4H), 7.44(t,4EI), 8.2O( m ,2H). During the preparation of BDTCN, we also obtained 2,3,6,9,11,12,15,18-octahydrodibenzo[ g , p ] [ 1 , S , 10,147 tetra- thiacyclooctadecene (yield s%), a dirner o f BDTCN.We also investigated this compou~iif ;is ;ii; icE-sclective membrane electrode, but it did not show ii good affinity for the PVC membrane. 6.71, S 30.49% ; IR 2945(-CH,-), 1493(-C=C-), 700(-C-S-); Preparation of Electrode The membrane electrode was prepared as rcportcd previ- ously.6 The sensor material, with DOP or NPOE as a plasticiser, and PVC were dissolved in tctrahydrofur-an (THF). In some iiist;inccs, KTCPB was also added as an ion exchanger. The THF solution was poured into a glass ring of 35 mm diameter on a glass plate and stored for 2 d at 30 "C. A membrane disc of 6 mm diamcter was cut using a master membrane and was fixed to the PVC tubing. The electrode was usually conditioned before use by soaking for 24 h in a 10-3 M solution of the respective metal chloride.The composition of the electrode membranes is surnmarised in Table 1. / A 1 d 1 2 0 r n ~ / P Fig. 2. Calibration graphs for different metal ions measured by electrode No, 1 based on B'ITCP and DOP, which is soaked in the respective metal chloride solutions (A2+). The slopes (mV decade-') and response times (s) were found to be as follows: for A, Cu2+, 29 mV,20s;B, Cd2+,27mV,50s;C,Co*+,27mV,60s;andD,Fe2+,23 mV, 70 s46 ANALYST, JANUARY 1988, VOL. 113 Table 1. Composition of ion-selective membrane electrodes. Values in parentheses are mass ratio in per cent Electrode Component 1 2 3 4 Sensor . . . . . . BTTCP(7) BTT'CP( 6) B'lTCP(6) BDTCN(9) Plasticiser . . . . DOP(31) NPOE(47) NPOE(46) NPOE (44.5) - - KTCPB(2) KTCPB (2) Exchanger .. . . PVC . . * . * . (62) (47) (46) (44.5) Table 2. Properties of Cu"-selective membrane electrodes Electrode Property 1 2 3 4 Detectionlimith . . . , 2.5 x 10-5 3.2 X 10-6 3.2 x 10-7 5.0 X lo-' Slope/mV per decade . . 28 29 28 30 Responsetime/s . . . . 20 10 10 25 Lifetime/months . . . . 2 2 3 0.7 pHrange . . . . . . 3.5-6.5 4.4-6.6 2.9-6.7 3.4-6.7 A B 7 6 5 4 3 2 1 -Ls~:;Zc"2+FM] - - Fig. 3. Calibration graph for Cu2+. Composition of membrane electrodes: A, B n C P and DOP (No. 1); B, B'ITCP and NPOE (No. 2); C, BlTCP, NPOE and KTCPB (No. 3); and D, BDTCN, NPOE and KTCPB (No. 4) Construction and Calibration of Electrodes The electrochemical cell was as follows: Ag - AgClIlO-3 M metal chloride solution1 sensor membrane Isample solution reference electrode.The e.m.f.s were measured relative to Ag - AgCl (Denki Kagaku Keiki, Japan) with an Orion 901 ionalyser. The performance of the electrode was investigated by measuring the e.m.f.s of metal chloride solutions prepared in the concentration range 10-1-10-7 M by serial dilution. The solutions were stirred and the potential readings recorded when they became stable. These were then plotted as a function of the logarithm of the metal cation activity. All the measurements were carried out at 25 "C. All the metal chloride solutions were freshly prepared by accurate dilution from their stock standard solutions, 10-1 M, with distilled, de-ionised water. Results and Discussion The B n C P membrane electrode (electrode 1) was found to be responsive to CuII, CdII, CoII and Fen ions (Fig.2). The electrode has a better response time and Nernstian linearity for Cuu than for the other metal ions. We therefore studied in detail the properties of the membrane for the copper(I1) ion. The linearity and detection limit improved from electrode 1 to electrode 2 by changing the plasticiser from DOP to NPOE, and even better results were observed by adding KTCPB to I 120rnV 'D I Mg. 4. Effect of pH. Composition of membrane electrodes: A, B'ITCP and DOP (No. 1); B, BTTCP and NPOE (No. 2); C, BTTCP, NPOE and KTCPB (No. 3); and D, BDTCN, NPOE and KTCPB (No. 4) the PVC membrane electrode (electrode 3), as shown in Fig. 3. Electrode 4, based on BDTCN, also showed a similar response to the copperjiij ion. The influence of pH on the response of the PVC membrane electrode to various concentrations of copper(I1) is shown in Fig.4. It was observed that above pH 6.5 the potential of all the membrane electrodes decreases owing to the formation of copper hydroxide in solution. At low pH, the potentials increase, indicating that the electrodes respond to hydrogen ions. The response times of the electrodes were tested by measuring the time required to achieve a steady potential (within 1 mV fluctuation) in a 10-3-10-2 M solution by a rapid 10-fold increase in the copper concentration (dynamic response). Electrodes 1, 2 and 3 yielded a steady potential within 20, 10 and 10 s, respectively, but electrode 4 had a response time of 25 s. The properties of the electrodes are summarised in Table 2.The selectivity coefficients for foreign ions were determined by the mixed solution method using the relationship 2.303RT ZAF E = constant + - lOg[aA 4- KKb(a&A''B] . (1) where E is the cell potential of the measuring cell, aA is the activity of the ion to be measured, aB is the activity of the interfering ion, Kgb is the selectivity coefficient of the electrode referred to interfering ion B, ZA is the charge of theANALYST, JANUARY 1988, VOL. 113 47 ~ ~~~ No. 1 No. 2 No. 3 No. 4 K Na cu - cu - cu - cu - - -4 t I \ Ca Mn Fig. 5. Com~riSoii of seiectivity coefficients with respect to the &*+ ion for kfferent metal cations. Electrode construction: No. 1, B'TTCP and DOP; No. 2, BTTCP and NPOE; No. 3, B'ITCP, NPOE and KTCPB; and No. 4, BDTCN, NPOE and KTCPB ion to be determined and ZB is the charge of the interfering ion.The Nernstian slope is 29.58 mV at 25 "C when zA = 2. The selectivity coefficient is therefore The activity coefficients y, of ions in aqueous sn!?rtic~ were &uiaied using equation (3) -logy = 0.511~ 2 vp -0.2p) . . . . (3) A (1 + 1.5G where p is the ionic strength. A comparison of the selectivity coefficients for different membrane electrodes is given in Fig. 5. The selectivity coefficient of electrode 2 for various metal ions is better than that of electrode 1, except for Cd2+ ions, because of the change in the plasticiser used from DOP to NPOE owing to the higher dielectric constant. Electrode 3, which contained NPOE and a KTCPB anion excluder, has better selectivity coefficients than the other electrodes except for Zn2+, Na+ and K+ ions.Electrode 4, based on BDTCN, has a good calibration graph for CuII (Fig. 3), but the selectivity coefficients for various other metal ions are not good. We concluded that each metal ion formed a complex with BDTCN non-selectively. The selectivity of BDTCN may be improved by the bis-type structure of this compound. Conclusion Neutral carrier CuII-selective electrodes have been prepared based on B'ITCP and BDTCN. It was observed that the electrodes based on BTTCP, NPOE and KTCPB have a better selectivity for Cuu than the BDTCN electrode containing the same solvent mediator and an anion excluder. 1. 2. 3. 4. 5. 6. References Desimone, R. E., and Glik, M. D., J. Am. Chem. SOC., 1976, 98,762. Diaddaris, L. L., Zimmer, L. L., Jones, T. E., Sokol, L. S. W. L., Cruz, R. B., Yee, E. L., Ochryrnowycz, L. A., and Rorabacher, D. B., J. Am. Chem. Soc., 1979,101,3511. Jones, T. E., Zhmer, L. L., Diaddaris, L. L., Rorabacher, D. B., and Ochrymowycz, L. A., J. Am. Chern. SOC., 1975,97, 7163, Rosen, W.: EE! BiiSch, D. H., J. Am. Chem. SOC., 1969, 91, 4694. Allen, D. W., Braunton, P. N., Millar, I. T., and Tebby, J. C., J. Chem. SOC. C, 1971,3454. Kamata, S., Higo, M., Kamibeppu, T. , and Tanaka, I., Chem. Lett., 1982,287. Paper A7/196 Received May 18th, I987 Accepted July 2nd, I987

ISSN:0003-2654    

DOI:10.1039/AN9881300045

出版商:RSC    

年代:1988

数据来源: RSC