摘要:

ANALYST, SEPTEMBER 1991, VOL. 116 919 Simultaneous Determination of Toxic Metabolites by Linear Combination Derivative Spectrophotometry Lin Liming Weifang Drug Control Institute, Shandong, People's Republic of China Zhao Naixin" Weifang Medical College, Shandong, People's Republic of China A linear combination derivative spectrophotometric method is described. The method overcomes the problem of overlapping in derivative spectrophotometry and allows the maximum use of quantitative information. In addition, the method can be used t o increase the selectivity, sensitivity and accuracy of the simultaneous analysis of multicomponent mixtures. The application o f the method t o the simultaneous determination of bongkrekic acid and toxoflavin, the toxic metabolites produced by Pseudomonas farinofermentans, is described.Keywords: Linear combination derivative spectrophotorn etry; pseudo m o n a s fa r i n ofe r m e n t a n s ; bongkrekic acid; toxoflavin Derivative spectrophotometry is zero-order spectropho- tometry transformed by differential calculus. It has the ability to separate the signal of the constituent to be determined from the signals of interfering constituents by mathematical processes and to enhance the sensitivity of the determination;' however, derivative spectrophotometry has some limitations for the determination of a constituent, the spectrum of which severely overlaps those of interfering constituents. In order to obtain better selectivity, sensitivity and accuracy, a linear combination derivative spectrophotometric method has been developed.The method overcomes the problem of over- lapping in derivative spectrophotometry and permits the maximum use of quantitative information. Bongkrekic acid and toxoflavin, the toxic metabolites produced by Pseudomonas farinofermentuns, play an import- ant role in food poisoning caused by fermented corn meal and white fungus (Tremellu fuciformis) contaminated with bac- teria.2.3 It has also been found that an unknown yellowish substance is present in the metabolites. To date, there have been no reports on the simultaneous determination of these two toxins. After a simple extraction procedure, the two toxins can be determined satisfactorily by the method de- scribed here without the need for separation. Theory In measurements using a derivative spectrum, suppose that there is no linear relationship between two constituents each of which accords with Beer's law.If the measured wavelength combination is taken as hi and the reference wavelength combination as hi, and if the derivative spectra of the two constituents, the unknown component a and the interfering component b, in solution, overlap, the linear combination derivative values will be i = l i = I j = 1 / = 1 Considering the different wavelengths, the measured linear combination derivative values of a mixture are the sum of the linear combination derivative values of all the constituents, i.e., rt1 m n n I: Di,, .C D&, .I: Dj, and , C Djb, respectively. m ni i = 1 i = 1 Dcj= C Di, + C Dib (1) n n Dc, = j = 2 1 Dja + ;= C 1 Djb (2) * To whom correspondence should be addressed.where L, denotes the derivative value of the absorbance, the subscript i denotes the ith measurement wavelength, the subscript j denotes the jth reference wavelength ('j can be 1 or 2 points) and c is the concentration of the mixture. On the same derivative curve, the ratio of the linear combination derivative at hi and hi is equal to K, and can be written as: m n C Dib - K C Dj, = 0 (3) (4) ( 5 ) i = 1 ;= 1 m n i.e., K = z = I . C Dib/, / = 1 2 01, Let ADc = Dci - KDc; from eqns. (1) and (2) we obtain 1 = 1 Dj, ;= 1 Djb) (6) from eqns. (3) and (6) we obtain If then m n ADci = , C Di, - K C Dj, I = 1 ;= 1 dEia dh j Di, = - c, where dddh is the derivative value of the absorption coeffi- cient. Hence ADc depends only on the concentration of component a; however, it accords with Beer's law as before, i.e., the interference from component b is eliminated.Suppose that the derivative spectra of two components are as shown in Fig. 1, in which curves a and b overlap and that the selected wavelengths are h,, h2, h3. If b is taken as the interfering component, then: i. e., ADb = Dbh, + Dbh3 - KDbh2 = 0 Dbh, + Dbh3 Dbh, K = where the linear combination value of the component to be determined (a) is ADa =(Dak, + Dah3)- KDah, = 1 Dak, 1 + 1 Dab, 1 + 1 KDah, I920 ANALYST, SEPTEMBER 1991, VOL. 116 x U $ 0 I I I A 1 h2 h3 Wavelength Fig. 1 a and b) Ideal model of derivative spectra (see text for details of Similar results are obtained by using a as the interfering component and b as the component to be determined (all the derivative values were substituted by experimentally deter- mined positive or negative values).Hence, by selecting an appropriate wavelength combina- tion, the value of K can be found and then by linear combination the interference from one component can be eliminated such that the absolute value of the derivative of the other component can be determined. This will allow the simultaneous determination of the two components. If the measured wavelength combinations are chosen at the zero-crossing point of the derivative spectrum of the interfering component, the calculation of K can be avoided and the component of interest can be determined directly based on the linear combination derivative values. If a number of wavelength points (say 2-5 points), at which the value of Di,lDib is large, are found for the measured wavelength combination hi and a number of wavelength points (say 1-2points), atwhich thevalueofDj,lDjbissmall, are found for the reference wavelength combination hi, and the wavelength points are chosen at points on the spectral curves where the slopes are as flat as possible, i.e., at peaks and valleys, it is known from eqn.(10) that the total derivative absorption coefficient is many times greater than that of a single wavelength; hence the sensitivity will be increased. From error transmission, the accuracy of A Dc is indicated by the relative standard deviation (RSD) where 6i and 6j denote the measurement error of each measured point expressed as the standard deviation.From eqn. (11) it can be seen that the total standard deviation, i.e., the numerator in eqn. ( l l ) , is less than the sum of the linear standard deviations, and that the value of ADc, i.e., the denominator in eqn. (11) , is a single linear combination. When the number of measurement points is increased, the increase in the denominator will exceed that of the numerator; hence the accuracy will be improved. Experimental Instrumentation and Reagents A Shimadzu UV-210A spectrophotometer with a DES-2 derivative attachment was used. Standard toxoflavin was prepared according to the procedure of van Damme et al.4 The product was recrystallized from propan-1-01 , filtered and dried in vacuo below 40 "C to constant mass. The identity and purity of the product were confirmed by its melting-point, by chromatography, and from ultraviolet and infrared absorption ~~ Table 1 Conditions for the determination of bongkrekic acid in toxoflavin Scan Scan Slit- Sample derivative nm min--l nm nm Order of speed/ rangel width/ Bongkrekic acid First derivative 100 360-270 1 Toxoflavin First derivative 100 330-240 1 2 ' t - 3 1 I 270 300 330 Wavelengthhm Fig.2 First-derivative spectra of bongkrekic acid (solid line) and toxoflavin (broken line) spectra. Bongkrekic acid was kindly provided by the Bagian Medisch Chenmie Fakultas Kedokteran Vrije Universitas (V.U.), Amsterdam, The Netherlands. All other reagents were of analytical-reagent grade. Preliminary Extraction of the Culture of Pseudomonas farinofermentans A 50 ml volume of the culture, which had been incubated for 48 h, was taken and sufficient solid ammonium sulphate was added in order to obtain a saturated solution.The solution was filtered and 5 g of sodium chloride were added to the filtrate. Exactly 20 ml of the filtrate were taken and the pH was adjusted to 7-8 with sodium hydroxide (1 mol dm-3.5 The filtrate was extracted twice with half its volume of light petroleum (b.p. range, 4MO"C) after which the pH of the aqueous phase was adjusted to 2-3 with hydrochloric acid (1 moll-'). The aqueous phase was then extracted with three 20 ml portions of chloroform. The chloroform extracts were combined and the volume was made up to exactly 100 ml with chloroform. Selection of Experimental Conditions Solvent At pH 2, bongkrekic acid and toxoflavin can be extracted simultaneously into chloroform.The extract was analysed by derivative spectrophotometry. The experimental stability and linearity range can meet the requirement of the determina- tion. The conditions used for the determination of bongkrekic acid and toxoflavin are given in Table 1. Determination of bongkrekic acid The first-derivative spectra of bongkrekic acid and toxoflavin are shown in Fig. 2. Based on the derivative spectrum of bongkrekic acid, two wavelengths, the measured wavelength combination hi and the reference wavelength combination hj, were chosen at the peak and valley, respectively, of the derivative spectrum of the interfering constituent, toxoflavin. Different concentrations of toxoflavin were scanned in order to obtain its first-derivative spectrum.The measured wavelength combination hi, 290 and 300 nm, and the referenceANALYST, SEPTEMBER 1991, VOL. 116 921 21 I 1 0 Table 3 Recovery of bongkrekic acid and toxoflavin Bongkrekic acid Toxoflavin Added Found/ Recovery Added Found/ Recovery No. pgml-1 pgml-1 (%) pgml-1 pgml-l (%) x -1 -0 s -2 -3 -4 -5 1 I I I 240 270 300 Wavelengthlnm Fig. 3 krekic acid (broken lines) First-derivative spectra of toxoflavin (solid line) and bong- Table 2 Linearity ranges and minimum measuring concentrations of bongkrekic acid and toxoflavin Bongkrekic acid* Toxoflavin Concentration/ pgml-1 2.65 3.98 4.77 5.30 6.63 7.95 9.28 ADc 1.66 2.73 3.37 3.82 4.85 5.92 6.99 Concentration/ pgml-1 2.24 4.48 6.72 7.84 8.96 11.2 13.4 ADc 2.50 5.38 8.27 9.73 11.2 14.1 16.9 Linearity range (pg ml- 1) : Minimum measuring concentration 2.65-9.28 2.24-13.4 (pg ml-I): 2.65 2.24 Regression equation: c(pgml-l)= 1.245 x ADc+ * K = 1.65. c(pgml-1)=0.7735 x ADc+ 0.574 ( r = 0.9999) 0.311 ( r = 0.9999) wavelength combination hi, 320 nm, were chosen at the zero- crossing point and at the flat region of the derivative curves where the derivative curve of bongkrekic acid exhibits a maximum.Based on the derivative values for different concentrations of toxoflavin at the wavelengths stated above, the value of K was calculated according to eqn. (4); K = DT&&,j(D$gx), the average value being K = 1.65 (n = 5 ) (RSD = 0.32%). The linear combination derivative value of bongkrekic acid, ADcBON = Dfg!& -1.65D:gN, was utilized to determine bongkrekic acid.Determination of toxoflavin As shown in Fig. 3, the measured wavelength hi and the reference wavelength hi were chosen at the zero-crossing points, 252 and 272 nm, respectively, of the derivative curve of bongkrekic acid, the interfering constituent. Calculation could be carried out directly according to the linear combination derivative values of the two wavelengths of toxoflavin. The eqn. ADc = DToX 252 - D;gx was utilized to determine toxoflavin. [Note: thevalueof DT$’Xwasnegative.] Linearity ranges and minimum measuringconcentrations are shown in Table 2. Recovery Study and Sample Analysis A synthetic mixture of bongkrekic acid and toxoflavin was prepared. The ratio of bongkrekic acid to toxoflavin was fixed 1 2 3 4 5 6 7 8 9 10 3.56 3.55 3.56 3.53 3.56 3.57 4.87 4.87 4.87 4.91 4.87 4.81 5.20 5.21 5.20 5.26 6.43 6.36 6.43 6.49 99.58 99.21 100.3 100.0 100.9 98.7 100.2 101.1 100.9 98.97 Mean 99.99 k 0.85% (RSD) 4.48 4.52 100.9 4.48 4.43 98.87 5.60 5.66 101.1 5.60 5.59 99.87 7.91 7.80 98.56 7.91 7.89 99.75 7.91 8.01 101.3 10.32 10.28 99.58 10.32 10.49 101.6 10.32 10.34 100.2 100.17 k 1.03% (RSD) Table 4 Simulated data for the linear combination derivative spectrophotometric method Di Dhl Dh2 Dh3 Dib 1 .o 1 .o 1.0 Di, 2.5 -2.5 2.5 so that it corresponded to the ratio found in the metabolites produced by Pseudomonas farinofermentans.After treating the solution as described under Preliminary Extraction of the Culture of Pseudomonas farinofermentans, an appropriate amount of the chloroform solution was taken and made up to exactly 100 ml with chloroform.Recoveries were calculated according to the regression equation. The results are pre- sented in Table 3. Sample analysis A 20 ml volume of Pseudomonas farinofermentans culture was taken and treated as described under Preliminary Extraction of the Culture of Pseudomonas farinofermentans. The remainder of the procedure was the same as that described under Recovery Study. As the concentration of toxoflavin was relatively low, the standard additions method was adopted. The content of bongkrekic acid was 60-80 yg ml-1 and that of toxoflavin 16-18 pg ml-1. Six different culture samples were analysed. Discussion Based on previous work,6,7 an analytical calculation method for analysing the quantitative information on a mixed system by utilizing the combination of derivative spectrophotometry and mathematical treatment has been developed here.By using this method, binary mixtures in which the spectra of the constituents interfere with each other to such an extent that the direct use of derivative spectrophotometry is not possible, can be analysed. In addition, the resolution and the adaptabil- ity of the spectra are enhanced. In comparison with an ultraviolet spectrum, a derivative spectrum has advantages such as the distinct increase of specific absorption peaks, the presence of positive and negative values above and below the baseline, the formation of a zero-crossing point where the derivative curve crosses the baseline at a certain wavelength, the formation of a straight line by the elimination of sample turbidity or of the flat region of the spectral curve, etc.If all these advantages are used and combined with mathematical analysis, it is possible not only to retain the advantages of a derivative spectrum, but also to maximize the use of quantitative information and to increase the selectivity and the measurement accuracy. By inspection of an ideal model utilizing eqn. ( l l ) , the relationship between the improvement of the accuracy and the number of determination points can be studied.922 ANALYST, SEPTEMBER 1991, VOL. 116 Suppose that: ( i ) the accuracyof the determination of Di, 6i = 0.01 ( i = 1, 2, 3); and (ii) the derivative spectra of two components a and b are as shown in Fig. 1; their derivative values and data for various determination points are presented in Table 4 .Single-point method. Component a is determined alone [i.e., component b is absent or the contribution of component b is zero (zero-crossing point)]. Take an arbitrary point in the range Dhl-Dh3: 0 01 2.5 RSD = - X 100% = 0.40% Two-point combination method. Take an arbitrary point in Dhl and Dh3 with Dh2, and proceed according to the combination method described above: K = 1.0/1.0 = 1.0 ADia=2.5- (-2.5)=5 J Three-point combination method. Take Dhl , Dh3 and Dh2: K = (1.0 + l.O)/l.O = 2.0 ADia = 2.5 + 2.5 - 2.0 X (-2.5) = 10 (0.012 + 0.012 + 2.02 x 0.012): RSD = X 100% = 0.24% 10 Hence from an inspection of the ideal model it can be seen that the accuracy determined by the three-point combination method is greater than that obtained by either the two-point combination or the single-point method; however, as the number of points used is increased, the extent to which the accuracy can be improved further gradually diminishes.The situation described above exists only under ideal conditions; under normal conditions, provided the wavelength combina- tion was suitable, relatively ideal conditions could also be obtained. The critical factor in the successful application of the proposed method is the choice of the wavelength combina- tion. As a derivative value can have a positive or negative sign, then, by choosing the appropriate wavelength at which to combine the plus or minus values, it is possible not only to eliminate the interference but also to increase the sensitivity. It has been shown that the proposed method can be applied successfully to the determination of bongkrekic acid and toxoflavin in the culture of Pseudomonas farinofermentans. The method might also be used to determine these two toxins in body fluids or in contaminated foodstuffs such as fermented corn meal and white fungus contaminated with bacteria. References Chen, G., Huang, X., Liu, W. Z., and Wang, Z., UItra-vioIet and Visible Spectroscopy, Atomic Energy Press, Beijing, 1980, pp. 207-216 (in Chinese). Research Group for Pathogenesis of Fermented Corn Flour Poisoning, Zhongguo Yixue Kexueyuan Xuebao, 1980,2, 77. Zhao, N., Wang, C., Li, Z., Wang, F., Lin, L., and Yu, X., Zhongguo Gonggong Weishong Zazki, 1987,6,65 (in Chinese). van Damme, P. A., Jahannes, A. G., Cox, H. C., and Berends, W., Red. Trav. Chim. Pays-Bas Belg., 1960, 79,255. Hu, W., Chen, X., Wang, P., Tian, C., Du, C., Meng, H., and Meng, Z., Weishong Yanjiu, 1984, 13, 34 (in Chinese). Lin, L., Yaoxue Xuebao, 1988, 23, 53. Lin, L., Fenxi Huaxue, 1990. 18, 773. Paper 0/026541 Received June 13th, 1990 Accepted April 22nd, 1991

ISSN:0003-2654    

DOI:10.1039/AN9911600919

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

ANALYST, SEPTEMBER 1991, VOL. 116 923 Polymer-based Cation-selective Electrodes Modified With Naphthalenesulphonates Tatsuhiro Okada, Hidenori Hayashi," Kazuhisa Hiratani, Hideki Sugihara and Naoto Koshizaki Industrial Products Research Institute, M.I. T.I., Yatabe, Tsukuba, lbaraki 305, Japan Polymer-based cation-selective electrodes based on cation exchangers incorporated in electrochemically polymerized pyrrole films deposited on solid electrodes are described. Pyrrole was polymerized anodically on Pt or pyrolytic graphite electrodes from CH3CN-H20 mixtures of various proportions together with naphthalenesulphonate compounds as dopants in the polymerized film. The pH was adjusted t o 1-2 with HCIO4 and polymerization was accomplished by galvanostatic methods. Hence electrodes selective t o monovalent cations were prepared.The composition of the CH3CN-H20 solvent was found t o be an important factor governing the e.m.f. response of the electrodes. X-ray photoelectron spectroscopic analysis revealed that the amount of naphthalenesulphonate relative t o polypyrrole varied with the composition of the CH3CN-H20 solvent used for the electrochemical polymerization. A sulphur t o nitrogen ratio >0.20 was required for a Nernstian response t o cationic activities. The monovalentldivalent selectivity was between -2.8 and -3.4, and was most pronounced for the naphthalenetrisulphonate compound. Although there were small differences in the selectivities of the electrodes towards monovalent cations, it might be possible t o obtain specific ion selectivities by modification of the polymer films.The possibility of constructing miniaturized cation-selective electrodes by means of an electrochemical procedure was also demonstrated. Keywords : Pol y (p yrrole) film; cation -selective electrode; electrochemical polymerization; naphthalene- sulphonate There have been numerous applications of liquid membrane type ion-selective electrodes based on neutral carriers in the analytical and clinical fields, because of the ease of achieving high ion selectivity and sensitivity by the use of specially synthesized ion-ligating compounds.l.2 In this instance ion- selective membranes, composed of, for example, a poly(viny1 chloride) matrix containing a solvent mediator (plasticizer), ion-selective agents and lipophilic additives, are utilized for detecting specific ions in solution.An electromotive force (e.m.f.) arises across the membrane which separates two solutions with different ionic activities; from this e.m.f. value the activities of the primary ion can be measured even in the presence of other interfering ions in the solution. However, there are limitations with this type of electrode because of the number of components required for its construction. For example, the electrode assembly typically has the following composition: inner reference electrode ( e . g . , Ag-AgCI) I inner solution of constant ionic activity (1 ion-selective membrane 11 test solution I outer reference electrode (e.g., KCI solution-AgCI-Ag). If the first part of this assembly could be replaced by a solid material so that the composition was: solid electrode I ion-selective layer I test solution I outer reference electrode, then this would be a novel type of construction. In fact, such electrodes were constructed by using solid-state ion conductors for the sensing materials, prior to the development of the neutral carrier based ion-selective electrodes.' Owing to the increasing demands for the application of ion-selective electrodes in the fields of clinical and micro- analytical measurements of ions, the miniaturization of electrodes is attracting considerable interest .3 In order to accomplish this, coated wire electrodes and ion-selective field effect transistor electrodes are being developed.4 Another possibility is to form an ion sensing layer on a small tip of material (electron conductor) by chemical methods.This type of electrode has been fabricated for pH sensing electrodes,s chloride and perchlorate electrodes ,697 glucose sensors, etc. ,8 but very few electrodes of this type have been reported for sensing cations of common interest. In this work, electrodes * Present address: Mitsubishi Denki Builtechno Service Company, Arakawa 7-19-1, Tokyo 116, Japan. modified with poly(pyrro1e) films containing ion-selective compounds were investigated in order to establish a method for preparing polymer-based cation-selective electrodes. Experimental Solutions for Electrochemical Polymerization of Pyrrole Solutions of pyrrole (0.06-0.15 mol dm-3) and (C2H5)4NBF4 (0.01 mol dm-3) in CH3CN as solvent were used for the determination of the optimum conditions for the electro- chemical polymerization of the pyrrole film on Pt sheets.The temperature was 25 "C and the pH was adjusted to 1, 3 or 5 with the use of H2SO4, CF3C02H or HC104, respectively. Solutions (0.01 mol dm-3) of sodium naphthalene-1-sulpho- nate (1-NSNa) , sodium naphthalene-2-sulphonate (2-NSNa), disodium naphthalene-l,5-disulphonate (1,5-NDSNa), diso- dium naphthalene-2,6-disulphonate (2,6-NDSNa), disodium naphthalene-2,7-disulphonate (2,7-NDSNa) and trisodium naphthalene-l,3,6-trisulphonate (1,3,6-NTSNa) were used instead of (C2Hs)4NBF4 as dopants in the poly(pyrro1e) film. The solvent was a mixture of CH3CN and H20 in various proportions. The pH was adjusted to 1-2 with HC104. All the solutions were de-aerated by bubbling nitrogen through them, and were stirred with a magnetic stirrer both before and during the electrolysis, except when the cyclic voltammograms were being recorded.Electrodes and Electrochemical Polymerization The electrodes used were Pt, Au (sheets), pyrolytic graphite (basal plane), Ta, Mo, W, Ir, Ni, Sn, Ag and In (bars or wires). The area of the electrodes exposed to the electrolytes was less than 0.5 cm2. The electrodes were polarized by a potentiostat (Nikko Keisoku DPGS-l), together with a voltage scanner (Hokuto Denko HB-103) and a coulometer (Hokuto Denko HF-201). Cyclic voltammograms were obtained at a scan rate of 0.1 V s-1. For the anodic polymerization of the pyrrole film, the current was applied galvanostatically for a fixed number of coulombs to the electrodes.The potentials of the electrodes were measured against an Ag-AgC1 reference electrode in the same solution.024 ANALYST, SEPTEMBER 1991. VOL. 116 Characteristics of the Polymer Modified Electrodes Electrodes coated with poly(pyrro1e) films (amount of charge: 2.88 C cm-2 at 8 x 10-4 A cm-2) were conditioned for 2 h in 1 x 10-2 rnol dm-3 LiCl solution. The response of the electrodes to solutions of alkali and alkaline earth metal chlorides was measured using an automated testing ap- paratus.9 The electrochemical system used for the e.m.f. measurements was assembled as follows: Pt I poly(pyrro1e) film I test solution I 0.1 rnol dm-3 NH4N03 I saturated The composition of the poly(pyrro1e) films deposited on Pt sheets by passing a charge of 2.88 C cm-2 at 8 x 10-4 A cm-2 was analysed with an X-ray photoelectron spectroscopic analyser [Physical Electronics; target, Mg( Ka) ; acceleration voltage, 10 kV; current, 40 mA].Also, for the determination of the bulk composition of the films, the surfaces were sputtered with Ar+ ions and then analysed by X-ray photo- electron spectroscopy (XPS). In experiment 1, Ar+ sputtering was conducted at a current density of 1 X 10-7 A cm-2 for 10 min (total dose 4 x 10'4 ions cm-2) and in experiment 2, at a current density of 1 X 10-6 A cm-2 for 10 min (total dose 4 x 1015 ions cm-*), following the sputtering in experiment 1. KCI-AgCI-Ag . Results Cyclic Voltammograms of Pyrrole Polymerization The cyclic voltammograms of Pt in pyrrole-CH3CN solutions with the anionic dopant (C2H5)4NBF4 are shown in Figs.1 and 2 over potential ranges of 0.04.6, 0.04.8 and 0.0-1.0 V versus Ag-AgC1; The poly(pyrro1e) film started to be depo- sited at potentials greater than 0.8 V versus Ag-AgC1, with a marked increase in the current in the voltammograms. The pH of the solution had a marked effect on the characteristics of the film; the adherence of the film and the extent of deposition were more pronounced at lower pH values. The concentration of pyrrole monomer in the solution was also an important factor in the extent of polymerization as indicated by an increase in the current; however, above 0.1 rnol dm-3 the effect was not significant in the tested range 0.06-0.15 rnol dm-3 pyrrole. Of the acids (H2SO4, CF3C02H and HC104) used for pH adjustment, HC104 proved to be the most suitable for deposition of a thick film and for good e.m.f.response of the electrodes. Hence, a 0.1 rnol dm-3 pyrrole solution adjusted to pH 1-2 with HC104 was used as the standard solution for electrochemical polymerization. t >. 0) C a, TJ I-' .- I-' E 3 u 1 I I I I 1 0 0.2 0.4 0.6 0.8 1 I I I I 0 0.2 0.4 0.6 0.8 E N versus Ag-AgCI Fig. 1 Cyclic voltammograms of Pt in pyrrole-CH3CN solutions. The 0.0-0.6,0.0-0.8 and 0.0-1 .O V scans are superimposed. Composi- tion of the solution: (C2H5)4NBF4, 0.01 mol dm--7; pyrrole: ( a ) 0.06 and (6) 0.10 mol dm-', pH adjusted to 1 with HC104. The arrow shows the increase in the current during the potential sweep The electrode material also affected the characteristics of the deposited film.Of the materials tested, Pt, pyrolytic graphite and Ta were found to be the most suitable for deposition of the poly(pyrro1e) film and also the e.m.f. response, as shown below. The other materials, except for Au, did not exhibit good adherence to the film, and corrosion of the substrates was observed in acidic pyrrole solutions. It appears that the substrate needs to be chemically stable and also adhesive to the deposited films. Cyclic voltammograms of Pt over the potential range 0.0-0.8 V versus Ag-AgC1 in 0.01 mol dm-3 1-NSNa and 0.1 rnol dm-3 pyrrole solutions at pH 1 are shown in Fig. 3 for t > rn e, U w .- I 1 I 1 1 I I I 1 I I I 1 0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8 EN versus Ag-AgCI Fig. 2 Cyclic voltammograms of Pt in pyrrole-CH3CN solutions. The 0.0-0.6,0.0-0.8 and 0.0-1.0 V scans are superimposed.Composi- tion of the solution: (C2H5)4NBF3r 0.01.mol dm-3; pyrrole, 0.15 mol dm-3, pH adjusted to (a) 3 and ( b ) 1 with HCIOj I I I 1 0 0.2 0.4 0.6 I I I 0 0.2 0.4 0.6 E N versus Ag-AgCI Fig. 3 Cyclic voltammograms of Pt at 0-0.8 V versus Ag-AgC1 in 0.01 mol dm-3 1-NSNa and 0.10 rnol dm--7 pyrrole solutions with CH3CN-H20 as the solvent. CH3CN-H20: (a) 100 + 0; (h) 90 + 10; ( c ) 70 + 30; and (d) 50 + 50. The pH was adjusted to 1 with HC104ANALYST, SEPTEMBER 1991, VOL. 116 925 0.30 0.25 0.20 0.15 0.10 0.35 0.30 + 0.25 0.20 0.15 W 0.30 Pel 0.25 0.20 0.15 0.10 - -5 -3 -1 0.30 0.25 0.20 0.15 0.10 0.25 0.20 0.15 0.10 0.05 u -5 -3 -1 Log a Fig. 4 E.m.f. response of electrodes composed of poly(pyrro1e) films deposited on Pt from 0.01 rnol dm-3 1-NSNa and 0.10 rnol dm-3 pyrrole solutions with various proportions of CH3CN-H20 as the solvent.CH3CN-H20: (a) 100 + 0; (6) 90 + 10; ( c ) 70 + 30; (d) 50 + JO; ( e ) 30 + 70; and cf) 0 + 100. The triangle [in (b)] shows the slope of Nernstian response for monovalent cations. Log a is the log of the activity of metal ions. 0, H+; A , Li+; V Na+; 0, K+; A, Mg2+; and V, Ca2+ CH3CN-H20 solvent mixtures of various proportions. It appears that 1-NSNa also acts as a dopant in the poly(pyrro1e) film. Deposition of the film started to occur at potentials greater than 0.8 V for all the solutions tested, and the current increased slightly as the proportion of H20 in the CH3CN- H 2 0 solvent mixture increased. However, the e.m.f.charac- teristics of the electrodes are not merely determined by the amount of film deposited, as will be shown below. E.m.f. Characteristics of the Electrodes Pyrrole was polymerized by a galvanostatic method on Pt sheets from various types of solutions containing pyrrole and a naphthalenesulphonate. In most instances the current density was 8.0 x 10-4 A cm-2 and the amount of charge passed was 2.88 C cm-2. The current density was fixed to give an electrode potential more positive than 0.8 V, as described above. The pH was adjusted to 1-2 with HC104. Smooth, black films were deposited on the surface of Pt electrodes. The thickness of the films under these conditions was greater than 30 pm. The e.m.f. response of cells containing Pt electrodes coated with poly(pyrro1e) films deposited from 0.01 mole dm-3 1-NSNa and 0.1 rnol dm-3 pyrrole solutions with various proportions of CH3CN-H20 as the solvent is shown in Fig.4. A Nernstian response to the activities of cations was obtained only over a very limited range of CH3CN-H20 solvent compositions, the optimum being CH3CN-H20 (90 + d 0.30 la: 0.25 0.20 0.15 0.10 3 4 0.30 0.25 0.20 0.15 -5 -3 -1 -5 -3 -1 Log a Fig. 5 E.m.f. response of electrodes composed of poly(pyrro1e) films deposited on a basal plane of pyrolytic graphite from 0.01 rnol dm-3 1-NSNa and 0.10 mol dm-3 pyrrole solutions with various proportions of CH3CN-H20 as the solvent. CH3CN-H20: (a) 100 + 0; (6) 90 + 10; (c) 80 + 20; and (d) 70 + 30. Log a is the log of the activity of metal ions. Symbols as identified in Fig.4 10). In this instance the monovalent/divalent cation selectivity (log kPPalMg) was -2.8. For other proportions of CH3CN-H20, the responses were sub-Nernstian, except for the H+ ion. In some instances, the electrodes showed responses to chloride ions with negative slopes in the e.m.f. versus log a curves (see also reference 6). The response time of the electrodes to the change in ionic concentration was 10-20 s; the e.m.f. values attained were stable for more than 10 min. The electrodes showed good reproducibility for repeated e .m .f. measure- ments. The drift in the electrode potentials was about 5 mV d-1. The current density and the total charge for electrochemical polymerization had very little influence on the e.m.f. characteristics of the electrodes, as far as the black film was concerned.A similar result was found when pyrolytic graphite was used as the substrate for the polymerization of pyrrole, as shown in Fig. 5. A bare Pt sheet, bare pyrolytic graphite and a poly(pyrro1e) film deposited on these materials using (C2H5)4NBF4 as a dopant did not exhibit a Nernstian reponse. The compound 2-NSNa also showed the same e.m.f. characteristics as 1-NSNa. When a disodium naphthalenedisulphonate was used instead of 1-NSNa, the range of CH3CN-H20 solvent compositions over which a Nernstian response was obtained was wider than that for 1-NSNa. This is shown for 2,7-NDSNa in Fig. 6. The monovalent/divalent cation selectivity (log kE'&) was -2.8 for the electrodes prepared using CH3CN- H 2 0 (90 + 10).The e.m.f. response characteristics of 1,5-NDSNa and 2,6-NDSNa were similar to those of 2,7- NDSNa. The naphthalenetrisulphonate derivative, 1,3,6-NTSNa, showed the best e.m.f. response compared with the mono- and disulphonate derivatives. This is shown in Fig. 7. The monovalent/divalent cation selectivity (log kk\)Mg) was -3.4 for the electrodes prepared using CH3CN-H20 (90 + 10). The Nernstian response increased in the order monosulphonate < disulphonate < trisulphonate derivatives of naphthalene. Composition of the Polypyrrole Film The poly(pyrro1e) films deposited on Pt sheets from a 0.01 rnol dm-3 1-NSNa and 0.1 rnol dm-3 pyrrole solution and a926 ANALYST, SEPTEMBER 1991, VOL. 116 1,3,6-NTSNa and 0.1 mol dm-3 pyrrole solution in three CH&N-H20 solvent mixtures of different composition were analysed by XPS.A typical spectrum is shown in Fig. 8. From a window analysis of the X-ray photoelectron spectroscopic peaks, the atomic concentration of C, 0, N, S in the pyrrole films was calculated. The results are shown in Table 1. Also, the atomic concentration in the bulk of the two different films containing 1-NSNa or 1,3,6-NTSNa is shown in Fig. 9. Compared with the composition at the outermost surface of the poly(pyrro1e) films, the atomic concentration of C was higher in the bulk of the films, whereas that of the other elements was lower. However, the sulphur to nitrogen (S : N) ratio was almost constant or somewhat higher in the bulk of the films compared with that at the surface. Discussion By comparing the results in Table 1 and Fig.9 with the e.m.f. response of the electrodes, the following conclusions can be drawn. (1) The amount of S in the poly(pyrro1e) film is greater 0.30 0‘35 t P ‘ 0.30 0.25 0.20 0.15 0.10 0.30 0.25 0.20 0.15 0.10 0.25 0.20 0.15 I I 1 0.30 m 0.25 0.20 0.15 0.10 0.05 0.25 0.20 E 0.15 0.10 0.05 W 0.25 2 0.20 .c: 0.15 0.10 W 0.25 0.20 0.15 0.10 0.05 L 0.15 m 0.25 0.25 0.20 0.20 0.15 0.15 0.10 0.10 0.05 0.10 0.05 0.00 - -5 -3 - 1 u -5 -3 - 1 u -5 -3 -1 -5 -3 -1 Log a Fig. 6 E.m.f. response of electrodes composed of poly(pyrro1e) films deposited on Pt from 0.01 mol dm-3 2.7-NDSNa and 0.10 rnol dm-3 pyrrole solutions with various proportions of CH3CN-H20 as the solvent. CH3CN-H20: (a) 100 + 0; ( b ) 90 + 10; (c) 70 + 30; ( d ) 50 + 50; ( e ) 30 + 70; and m 0 + 100.Log a is the log of the activity of the metal ions. Symbols as identified in Fig. 4 Log a Fig. 7 E.m.f. response of electrodes composed of poly(pyrro1e films) deposited on Pt from 0.01 rnol dm-3 1,3,6-NTSNa and 0.10 rnol dm-3 pyrrole solutions with various proportions of CH3CN-H20 as the solvent. CH3CN-H20: (a) 100 + 0; (b) 90 + 10; ( c ) 70 + 30; ( d ) 50 + 50; ( e ) 30 + 70; and 0 + 100. Log a is the log of the activity of the metal ions. Symbols as identified in Fig. 4 Table 1 Atomic concentrations of C, 0, N and S in poly(pyrro1e) films deposited on Pt. Analysed by XPS peak height sensitivities. Values in parentheses were determined by peak area sensitivities obtained using XPS Atomic concentration (%) Atomic ratio, 0 N S S:N 15.1 7.7 2.1 0.27 (18.0) (5.4) (1.3) (0.24) 0.14 13.8 5.9 0.8 (16.7) (4.6) (0.8) (0.17) 11.2 8.8 1.6 0.18 (14.5) (5.9) (1.4) (0.24) 13.0 8.5 2.0 0.24 14.3 10.7 2.3 0.21 (16.5) (6 * 9) (1.7) (0.25) 0.31 18.1 10.1 3.1 (18.3) (8.1) (2.5) (0.31) (15.8) (6.0) (1.3) (0.22) Dopant CH3CN-HZO C 1-NSNa 90+ 10 75.1 50 + 50 79.4 0 + 100 78.4 1,3,6-NTSNa 90 + 10 76.5 50 + 50 72.7 o + 100 68.7 (75.3) (77.9) (78.2) (76.8) (74.9) (71.1)ANALYST, SEPTEMBER 1991, VOL.116 927 51 z 4 1 .- 3 - 1000 - 800 - 600 - 400 -200 0 Fig. 8 X-ray photoelectron spectroscopic spectrum of a poly- (pyrrole) film deposited from 0.01 rnol dm--7 1,3,6-NTSNa and 0.10 rnol dm-3 aqueous pyrrole solutions. pH, 1-2 (HCIO,); CHxCN- H20. 0 + 100; electrode, Pt; and charge passed, 2.88 C cm-2 Binding ene rg yleV 0.5 1 1 0 .2 t I 1 0.2 Before Ar+ Sputter Sputter sputtering 1 2 Fig. 9 Atomic concentrations and S : N ratios for poly(pyrro1e) films before and after sputtering with Ar+ (see under Experimental). A, C; B, 0; C, N; D, S; and E, S/N. ( a ) Poly(pyrro1e) film deposited on Pt from 0.01 rnol dm-' 1-NSNa and 0.10 rnol dm-3 pyrrole aqueous solutions. (b) Poly(pyrro1e) film deposited on Pt from 0.01 rnol dm--7 1,3,6-NTSNa and 0.10 rnol dm--7 pyrrole aqueous solutions for the 1,3,6-NTSNa doped film than for the 1-NSNa doped film. (2) For the 1-NSNa doped film, the highest S : N ratio is obtained when CH3CN-H20 (90 + 10) is used as the solvent, whereas for the 1,3,6-NTSNa doped film the highest S : N ratio is obtained for films deposited from H20. (3) In all instances, the electrodes show a Nernstian response when the S : N ratio on the surface of the poly(pyrro1e) film is >0.20.It appears that the density of the sulphonic acid groups in the film plays an important role with respect to the e.m.f. characteristics of the electrodes. The sulphonic acid groups in the naphthalene compounds could act as anionic counter ions and compensate for the positively charged imino groups in the poly(pyrrole) network. Also, the remaining sulphonic acid groups could form cation-exchange sites for positive ions in the solution. Hence the charge density of the sulphonic acid groups would be a major factor governing the Nernstian response of the electrodes. The compositon of the solvent plays a significant role in determining the inclusion of the naphthalene derivatives in the poly(pyrro1e) film during the electrochemical polymerization from electrolyte solutions.In this work it has been demonstrated that polymer-based cation-selective electrodes can be constructed by electrochem- ical polymerization methods using solid materials such as Pt and pyrolytic graphite as substrates. Although poly(pyrro1e)- based anion-selective electrodes have been prepared success- fully by doping chloride or perchlorate ions during electro- chemical polymerization,6~7 very few electrodes of this type selective to cations have been reported. The proposed method offers the possibility of preparing a novel type of polymer- based cation-selective electrode. The method could also be applied to other types of polymers, such as poly(thiophene), poly(aniline), and their derivatives.The salient features of the proposed method for constructing cation-selective electrodes are as follows. (1) Adhesive, flawless films of uniform thickness can be deposited on the surface of solid substrates by electrochemical polymerization. (2) The amount of the ion-selective layer deposited can easily be controlled by controlling the conditions of the electrolysis. (3) The response times of the electrodes are very short in comparison with those of coated wire electrodes, which are typically more than 30 s. (4) Miniaturized electrodes could easily be constructed by using tips or fibres as substrate materials. Specific cation selectivities could be studied further by modifying the polymer films with dopants which have specific ligating abilities towards such cations. Further investigations in this area are currently in progress. The authors thank the Tomoe Kogyo Co. Ltd., Tokyo, for supplying the pyrolytic graphite discs. References Ion-Selective Electrodes in Analytical Chemistry. ed. Freiser, H., Plenum Press, New York, 1978, vol. 1. Ammann, D., Morf, W. E., Anker, P., Meier, P.C., Pretsch, E., and Simon, W., Ion-Sel. Electrode Rev., 1983, 5, 3. Ion Measurements in Physiology and Medicine, eds. Kessler, M., Harrison, D. K., and Hoper, J., Springer-Verlag, Berlin, 1985. Physical Methods of Chemistry. Volume II: Electrochemical Methods, eds. Rossiter. B. W., and Hamilton, J. F., Wiley, New York, 1986, ch. 2. Ohnuki, Y., Matsuda, H., Ohsaka, T., and Oyama, N., J. Electroanal. Chem. Interfacial Electrochem., 1983, 158, 55. Dong, S., Sun, Z., and Lu, Z., Analyst, 1988, 113, 1525. Lu, Z., Sun, Z., and Dong, S., Electroanalysis, 1989, 1, 271. Couves, L. D., and Porter, S. J., Synth. Met., 1989, 28. C761. Okada, T., Hiratani, K., and Sugihara. H., Analyst, 1987, 112, 587. Paper 1/00873 K Received February 22nd, 1991 Accepted May 3rd, 1991

ISSN:0003-2654    

DOI:10.1039/AN9911600923

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

ANALYST, SEPTEMBER 1991, VOL. 116 929 Determination of Glutathione at Enzyme-modified and Unmodified Glassy Carbon Electrodes Chi Hua and Malcolm R. Smyth* School of Chemical Sciences, Dublin City University, Dublin 9, Ireland Ciaran O'Fagain School of Biological Sciences, Dublin City University, Dublin 9, Ireland The oxidation of the reduced form of glutathione (GSH) was found t o occur at a decreased overpotential at a glassy carbon electrode when 1.2 mot dm-3 of dipotassium hydrogen phosphate were used as the supporting electrolyte. The resulting current for the oxidation peak of GSH varied linearly over the concentration range 0.16-2.4 mmol dm-3 GSH. The reaction between GSH and hydrogen peroxide catalysed by the enzyme GSH-peroxidase (GSH-PX) could be monitored using a glassy carbon electrode with GSH-PX immobilized under a layer of Nafion film.Keywords: Reduced glutathione; amperometric biosensor; glutathione peroxidase Glutathione exists in the human body in both the reduced (GSH) and oxidized (GSSG) forms and is essential for many metabolic processes. 1 In particular it affords protection against metabolic stresses. Firstly, it can reduce membrane substances such as peroxides and free radicals non-enzymic- ally. Secondly, through the action of glutathione-S-trans- ferases, it can be conjugated to xenobiotic or electrophilic compounds thus aiding in their detoxification.2 Defects in glutathione synthesis and metabolism are associated with disease states in humans, and levels of glutathione and its metabolizing enzymes may be significant in cancer.3 Gluta- thione peroxidase (GSH-PX) (glutathione: hydrogen perox- ide oxidoreductase, E.C.1.11.1.9) is found in many animal tissues and is regarded as a major protective system against endogenously and exogenously induced hydrogen and lipid peroxides.4 The enzymic activity of GSH-PX is the major pathway for the elimination of hydrogen peroxide in red blood cells,5 which are extremely sensitive to the accumulation of peroxide.2 It catalyses the reaction between reduced gluta- thione and hydrogen peroxide (or organic peroxides) resulting in the reduction of the peroxide as follows: 2GSH + H202 -+ GSSG + 2H20 The enzyme, GSH-PX, possesses an unusual feature, an essential selenocysteine residue. It is an analogue of the amino acid cysteine but contains selenium in place of sulphur.3Jj The enzyme obtained from beef red blood cells has a relative molecular mass of 85000 and appears to consist of four identical sub-units, with one atom of selenium in each.7 It is important to have suitable analytical methods to follow the reaction process of GSH with hydrogen and lipid peroxides.A variety of electroanalytical methods have been developed for the detection of GSH.4-8-11 Methods based on mercury electrodes" may be undesirable because of their possible toxicity. Reduced glutathione can also be oxidized at bare carbon electrodes; however, inconveniently large work- ing potentials are needed.8 Modified carbon paste electrodes have therefore been studied in order to reduce the over- potential for the determination of GSH and have shown promise.g.12 Recently, Wring et aZ.9 have developed a carbon electrode chemically modified with cobalt phthalocyanine.The modified electrode was found to reduce greatly the overpotential necessary for the oxidation of GSH at the carbon electrode surface. * To whom correspondence should be addressed. The overpotential required for the oxidation of GSH can also be reduced by the use of high concentrations of potassium phosphate as the supporting electrolyte, as described in this paper. The reaction of GSH with hydrogen peroxide on a glassy carbon electrode surface immobilized with GSH-PX is also reported. Experimental Chemicals and Reagents All chemicals were of analytical-reagent grade unless stated otherwise. The GSH and GSH-PX were purchased from Sigma, while Nafion (in aliphatic alcohols + 10% water) was purchased from Aldrich.The supporting electrolytes used were prepared with de-ionized water obtained by passing distilled water through a Millipore Milli-Q water purification system. Apparatus Differential potential voltammetry (DPV) was performed using an EG&G Princeton Applied Research Model 174A polarographic analyser. A three-electrode cell was employed incorporating either a glassy carbon or a modified glassy carbon electrode plus a saturated calomel (SCE) reference electrode and a platinum wire counter electrode. Construction of the Electrode The enzyme-modified electrode was prepared by placing 2 p1 of enzyme solution containing GSH-PX at a concentration of 5 mg ml-1 onto the surface (3 mm in diameter) of a glassy carbon electrode.After the solution had dried on the surface, 2 pl of a solution containing a suitable concentration of Nafion were applied to the electrode. After the Nafion solution had evaporated to dryness, the electrode was washed with de-ionized water before use. Voltammetric Procedure Differential potential voltammetry was first performed on suitable blank solutions, then on solutions containing GSH. The voltammetric conditions were in most instances as follows: initial potential, +0.20 V; scan rate, 10 mV s-1; pulse height, 10 mV; and final potential, 0.95 V.930 Results and Discussion Effect of the Composition of the Electrolyte The effect of the concentration of phosphate buffers was first investigated in order to optimize the response for GSH at a glassy carbon electrode.It was found that GSH yielded an oxidation peak at +0.51 V when 1.2 mol dm-3 dipotassium hydrogen phosphate was used as the electrolyte. A similar response was obtained with saturated disodium hydrogen phosphate as the electrolyte. The response of GSH decreased by 70% when the concentration of dipotassium hydrogen phosphate was decreased to 0.12 mol dm-3 and no response was observed when 0.012 mol dm-3 disodium hydrogen phosphate was used as the electrolyte. This result indicated that it was essential to have a high concentration of phosphate in the electrolyte in order to obtain the oxidation peak for GSH. Reduced glutathione has been reported previously to be oxidized at glassy carbon electrodes at high working electrode potentials in a medium consisting of 0.1 mol dm-3 mono- chloroacetate (pH 3) and methanol in the proportion of 95 + 5 (% vh), but the oxidation current was concealed by the background current." The oxidation of GSH at glassy carbon electrodes in concentrated phosphate buffer has not been reported.This system provided a convenient method for the detection of GSH with glassy carbon electrodes. However, a high concentration of phosphate is not desirable for detection following high-performance liquid chromatography because it would be likely to cause high pressure in the column. Effect of pH on the Response of GSH at the Bare Glassy Carbon Electrode The effect of pH on the response of GSH at the bare glassy carbon electrode was studied by varying the pH of solutions containing 1.2 mol dm-3 dipotassium hydrogen phosphate and 1.6 mmol dm-3 GSH.As shown in Fig. 1, the peak current increased when the pH was increased from 6 to 10. Reduced glutathione has been reported to be unstable in alkaline solution as it is easily oxidized.8 Similarly , the electrochemical oxidation of GSH was also facilitated in high pH medium at glassy carbon electrodes. This is consistent with the results obtained using a carbon electrode modified with cobalt phthalocyanine, as reported by Wring et aZ.8 Linearity of Response The linearity of the response for GSH at the glassy carbon electrode was studied by varying the concentration of GSH in a buffer solution containing 1.2 mol dm-3 dipotassium hydrogen phosphate adjusted to pH 9.0.The resulting I I ANALYST, SEPTEMBER 1991, VOL. 116 I I I +0.92 +0.56 +0.20 EN versus SCE Fig. 2 Voltammograms of GSH obtained by analysing different concentrations of GSH in 1.2 rnol dm-3 dipotassium hydrogen phosphate adjusted to pH 9.0. 1, 0.158; 2, 0.474; 3, 1.106; 4, 1.738; and 5, 2.370 mmol dm-3 GSH in the electrolyte t U c 2 3 u Y m L Fig. 3 I I +0.92 +0.56 +0.20 \ ( b I I +0.92 +0.56 +0.20 EN versus SCE Reaction of GSH with hydrogen peroxide monitored with a glassy carbon electrode modified with either (a) a Nafion film or (b) GSH-PX and Nafion. Curve 1, obtained by analysing a sample solution containing 1.2 rnol dm-3 dipotassium hydrogen phosphate and 2.6 mmol dm-3 GSH adjusted to pH 7.0; curves 2-7, successive additions of 4.4 rnol dm-3 hydrogen peroxide to the sample solution 6 7 8 9 10 PH Fig.1 Effect of pH on the response of GSH at a glassy carbon electrode obtained by analysing solutions containin 1.2 mol dm-3 to different pH values dipotassium hydrogen phosphate and 1.6 rnmol dm- F GSH, adjusted voltammograms are shown in Fig. 2. A linear response was obtained over the GSH concentration range 0.16-2.4 mmol dm-3 with a correlation coefficient of 0.9986, a slope of 33.6 pA mol-1 and an intercept on the current axis of -0.00321 PA.ANALYST, SEPTEMBER 1991, VOL. 116 93 1 Modification of a Glassy Carbon Electrode With Glutathione Peroxidase Different methods of immobilizing GSH-PX onto a glassy carbon electrode were investigated. The GSH-PX solution was first placed on the glassy carbon electrode surface and left to dry.This method was not successful as the GSH-PX on the electrode surface quickly dissolved in the electrolyte solution. A dialysis film was then placed over the enzyme layer on the electrode surface. In this instance, however, no response could be seen for GSH, because GSH could not penetrate through the dialysis film to reach the electrode surface. A film of Nafion was then used to cover the enzyme on the electrode by placing 2 p1 of Nafion solution over the enzyme on the electrode surface. After the Nafion solution had dried, the electrode was tested for the determination of GSH. It was found that if the concentrated Nafion solution (as obtained from the supplier) was placed on the electrode surface, no response for GSH appeared. Diluting the Nafion solution 10-fold with water before modification also did not result in a response to GSH.If the Nafion solution was diluted 100-fold with water before being used in the modification process, the oxidation peak for GSH appeared. A thin and even film could be seen on the electrode surface. The Nafion solution was also diluted 300-fold with water before being applied to the electrode surface. In that instance, however, the resulting film was poorly distributed on the electrode surface and contained many large pores. It was found that the Nafion solution diluted 100-fold was the best for the modification. This result indicated that the size of the pores in the Nafion film could be conveniently controlled by suitable dilution of the Nafion solution with water before making the film, so that desired compounds could penetrate the film to reach the electrode surface.This method of modification provides a general way of immobilizing enzymes or some other modifying materials onto an electrode surface, thus providing the system with both size- and charge-selective properties. Monitoring of Reaction of GSH With Hydrogen Peroxide at Unmodified and Enzyme-modified Electrodes The reaction of GSH with hydrogen peroxide was investigated by analysing solutions containing 1.2 mol dm-3 dipotassium hydrogen phosphate and 2.6 mmol dm-3 GSH, to which various concentrations of hydrogen peroxide were added at specified pH values. Modified and unmodified electrodes were both used to monitor changes in the concentration of GSH. At pH 8 and 9, a bare glassy carbon electrode was used to follow the reaction.It was found that on addition of hydrogen peroxide to the solution containing GSH, the peak height of GSH decreased and finally disappeared when more hydrogen peroxide was added, indicating that hydrogen peroxide oxidized GSH chemically at these pH values. At pH 7, when a bare glassy carbon electrode or a glassy carbon electrode covered solely with Nafion film was used, the peak due to GSH remained in spite of the addition of hydrogen peroxide to the GSH solution. When more hydrogen peroxide was added, a peak corresponding to hydrogen peroxide appeared at +0.88 V, while the GSH peak remained constant. If a GSH-PX modified electrode was used at this pH value, the GSH peak rapidly decreased in size on addition of hydrogen peroxide to the GSH solution.After further addition of hydrogen peroxide, the GSH peak disappeared and the peak for hydrogen peroxide appeared (Fig. 3). This result showed that the reaction between GSH and hydrogen peroxide, catalysed by the immobilized GSH-PX at the electrode surface, could be followed with this system. Conclusion A much lower overpotential for the oxidation of GSH at a glassy carbon electrode was obtained using a high concentra- tion of dipotassium hydrogen phosphate as the supporting electrolyte. The modification of a glassy carbon electrode with GSH-PX provided a means of monitoring the enzymic reaction of GSH with hydrogen peroxide in aqueous solution. This work was funded through a grant (SC/90/118) from EOLAS (The Irish Science and Technology Agency) under the Scientific Research Programme. 1 2 3 4 5 6 7 8 9 10 11 12 References Meister, A., 1. Biol. Chem., 1988, 263, 17205. Matthews, C. K., and Van Holde, K. E., Biochemistry. Benjamin-Cummings, California, 1990, p. 71 1. Meister, A., and Anderson, M. E., Ann. Rev. Biochem., 1983, 62, 711. Wendel, A., Methods Enzymol.. 1981, 77, 325. Cohen, G., and Hochstein, P.. Biochemistry, 1963, 2, 1420. Forstrom, J. W., Zakowski, J. J., and Tappel, A. L., Biochemistry, 1978, 17, 2639. Flohe, L., Ciba Found. Symp., 1979, 65, 95. Wring, S. A., Hart, J. P., and Birch, B. J., Analyst, 1989, 114, 1563. Wring. S. A.. Hart, J. P., and Birch, B. J., Analyst. 1989, 114, 1571. Halbert, M. K., and Baldwin, R. P., Anal. Chem., 1985, 57, 591. Allison, L. A., and Shoup, R. D., Anal. Chem., 1983.55, 8 . Halbert, K. H., and Baldwin, R. P., J . Chromatogr., 1985,345, 43. Paper 1101049B Received March 16th, 1991 Accepted May loth, 1991

ISSN:0003-2654    

DOI:10.1039/AN9911600929

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

ANALYST, SEPTEMBER 1991, VOL. 116 Amperometric Monitoring of Bacteria-induced Milk Platinum Disc Microelectrode M. Antonietta Baldo, Salvatore Daniele" and Gian A. Mazzocchin Department of Physical Chemistry, University of Venice, Calle Larga S. Marta, Marco Donati Direzione Scientifica, Parmalat SPA, Parma, ltal y 933 Acidity Using a 2137, 30123 Venice, Italy The acidity induced by the action of bacteria in milk samples was monitored amperometrically by using a platinum microelectrode. The measurements were performed directly on commercial packs of milk, stored at 32 "C, and were continued for 9-10 d after inoculation. The data were compared with those obtained by measuring the pH of the samples and the results are discussed on the basis of the metabolism of each bacterial species.The effects of the following bacteria were examined: Staphylococcus aureus, Bacillus cereus, Streptococcus faecalis, Bacillus subtilis, Aeromonas, and Corynebacterium. Keywords: Microelectrode; milk; acidity; bacteria; amperometry Numerous micro-organisms, predominantly bacteria, can proliferate in milk, thus changing its properties. 1 In particular, bacterial fermentation can cause the production of some acids, such as lactic and acetic acids. Bacterial growth and action in milk can be either deliber- ately induced under controlled conditions, as in the manufac- ture of fermented products, or undesired, and the production of acids can cause sourness. The acidity induced by the microbic contamination of milk supplies coming from farms is normally determined by measuring the pH of the milk, and a parallel microbiological test is also performed in order to determine the bacterial population.The buffering action of the numerous acidic and basic groups present in milk often leads to a negligible pH change, when, for instance, the amount of acids produced is low. Hence the determination of the amount of acids formed, instead of the pH, should, in principle, be a more sensitive method for measuring the change in the acidity of the medium. A useful parameter for this purpose is the titrable acidity of milk.2 It has recently been shown3 that it is possible to measure the acidity of milk samples amperometrically by using a platinum microelectrode. In the present paper this method was used to monitor the change in the titrable acidity of milk caused by the presence of bacteria in the milk samples; the data obtained were compared with those obtained by pH measurements.The effect of the following bacterial species on the change in acidity was examined: Staphylococcus aureus, Bacillus cereus, Streptococcus faecalis, Bacillus subtilis, Aeromonas and Corynebacterium .4 Experimental Electrodes and Instrumentation In order to prepare the microelectrode, a platinum wire of diameter 25 pm was sealed directly in glass as reported previously.5 Prior to each measurement, the electrode was polished with graded alumina powder (down to 0.05 pm) on a polishing microcloth. The reference electrode used was a saturated calomel electrode. The experiments with the platinum microelectrode were carried out in a two-electrode cell configuration maintained in a Faraday cage made of sheets of aluminium to avoid external noise.Linear sweep and cyclic voltammetric waves were * To whom correspondence should be addressed. generated by a Princeton Applied Research 175 function generator; a Keithley Model 485 picoammeter served as a current-measuring device, and data were plotted with a Hewlett-Packard 7045 B x-y recorder. A Metrohm 605 pH-meter was employed for pH measure- ments. The samples were kept at 32 "C in a Memmert thermostatic oven until required for measurement. Samples and Procedure Bovine milk samples were kindly provided by Parmalat (Parma, Italy). The analyses were performed directly on the pack containing the milk; the electrodes were inserted into the pack through an appropriate hole made immediately before the analysis.Contaminated samples were prepared by inoculation, under aseptic conditions, of a number of different mesophyll facultatively anaerobic bacterial species (Staphylococcus aureus, Bacillus cereus, Streptococcus faecalis, Bacillus subti- lis, Aeromonas and Corynebacterium) which can contaminate milk supplies.6 Cell concentrations for each micro-organism were adjusted to a target range of 10-100 colony-forming units (cfu) per 0.5 1 of milk. The inoculated milk samples were then stored for 9-10 d at 32 "C, which are typical favourable conditions for bacterial growth. For each experimental datum at least three replicates both for current and pH measurements were performed on two samples of the same stock of milk.Results and Discussion Principle of the Amperometric Method The amperometric method is based on the determination of the current of peak (1) in Fig. 1, which can be recorded in milk samples with a platinum disc microelectrode at scan rates lower than 10 mV s-1. The reduction mechanism occurring at this peak has been studied in detail elsewhere;3 it involves a sequence in which both the preceding and following chemical reactions are associated with the transfer of electrons. The mechanism proposed is as follows: H,A-M C H+ + H,, - 1) A-M H2P04- H+ + HP042- H+ + e- $H2 H2P04- + Ca'+ e CaHP04(s) 2HP042- + 3Ca*+ S Ca3(P04)z(s) + 2H+934 ANALYST, SEPTEMBER 1991, VOL. 116 where H,A is the generic acid interacting with the micellar aggregate M present in the milk.The sparingly soluble CaHP04 and Ca3(P0&, formed in competitive equilibria, precipitate onto the electrode surface, thus causing inhibition of the electrode process and giving rise to the peak-shaped wave instead of the expected steady-state limiting current plateau.' Moreover, the addition of bases or acids leads to a decrease or increase in peak (l), respectively,3 as a conse- quence of acid-base equilibria which take place in the bulk solution, leading ultimately to the formation or dissociation of H2P04- .3 Consequently, by monitoring the peak (1) height, the variation of the acid concentration in the medium can be measured. Applications to Contaminated Samples The acidity induced by bacterial fermentation of ultra-high temperature treated (UHT) sterilized milk samples contami- nated deliberately was followed amperometrically for 220 h; the corresponding behaviour of uncontaminated milk samples was taken as a reference.The results obtained at different times for the peak (1) height and the corresponding pH value, from a series of measurements on reference samples and the contaminated samples, are presented in Table 1. For ease of interpretation u 0 -0.5 -1.0 PotentialN Fig. 1 Cyclic voltammogram recorded at a 25 km platinum microelec- trode on an untreated fresh whole bovine milk sample. Scan rate, 5 mV s- l . For an explanation of peak 1, see text of the data, Figs. 2 and 3 show the variation (in YO) of the current height of peak (1) and pH with respect to the blanks under the same experimental conditions.From these data, it can be seen that the sensitivity of the amperometric method is greater than that of the potentio- metric method. For instance, data concerning the reference samples' reported in Table 1 indicate that after storage for 220 h at 32 "C a moderate increase in the peak height of about 10% is observed, whereas the variation in pH is small. The increment in the acidity of the uncontaminated samples is probably due to changes in composition during storage of the milk at 32 "C, caused by, for instance, the liberation of phosphoric acids from their esters by enzymic action, or the production of free fatty acids by lipolysis, which is known to occur on increasing the temperature of milk.1 Inoculated samples produced an increase in acidity in all instances, but to an extent that depended on the products formed from the metabolism of each bacterial species.$ 50 I a - 30 c E 3 10 S a, cn .- 2 10 0 5 72 120 168 216 24 24 72 720 168 216 Time/h Fig. 2 Variation of A, peak current and B, pH with time for bovine milk samples inoculated with ( a ) Bacillus cereus and ( b ) Staphyiococ- cus aureus at 32 "C Table 1 Peak currents (i k 0.1 nA)* and pH values (kO.01) obtained for bovine milk samples contaminated with different bacterial species under aseptic conditions Samples (1) contaminated with: Samples (2) contaminated with: Reference samples (1) Bacillus cereus Time1 h 8 26 31 50 74 98 147 170 194 220 il nA pH 16.1 6.67 17.0 6.67 17.1 6.65 17.2 6.65 17.5 6.64 17.5 6.65 17.6 6.63 17.6 6.62 17.5 6.64 - - il nA pH 16.3 6.67 27.3 6.41 27.7 6.39 29.5 6.32 30.5 6.31 32.8 6.29 34.0 6.27 33.0 6.28 33.0 6.27 - - Staphylococcus aureus il nA pH 16.3 6.66 17.7 6.66 18.0 6.60 18.3 6.58 18.6 6.57 18.8 6.58 19.3 6.58 19.5 6.53 19.6 6.53 - - Reference samples (2) il nA pH 15.1 6.72 16.1 6.66 16.7 6.65 17.7 6.64 17.8 6.63 - - - - - - - - - - Streptococcus faecaiis it - nA pH 15.0 6.72 17.0 6.62 19.0 6.00 22.5 5.60 28.1 5.48 - - - - - - - - - - Bacillus subtilis Aeromonas Coryne- bacterium il nA pH 15.4 6.69 16.8 6.63 17.6 6.63 20.3 6.59 21.5 6.55 - - - - - - - - - - ~~ il nA pH 15.2 6.72 26.3 6.20 28.0 6.16 30.3 6.14 37.0 5.97 - - - - - - - - il nA pH 15.0 6.72 16.2 6.64 - - - - - - 17.8 6.62 19.7 6.62 - - - - * Average values obtained from five replicates (relative standard deviation <1.5%).ANALYST, SEPTEMBER 1991, VOL.116 935 50 25 24 72 120 168 21 6 24 72 120 168 216 24 72 120 168 216 II" + I m B K I ' I I I 24 72 120 168 21 6 Time/h Fig. 3 Variation of A, peak current and B, pH with time for bovine milk samples inoculated with ( a ) Streptococcus faecalis; ( b ) Bacillus subrilis; ( c ) Aeromonas; and ( d ) Corynebacrerium at 32 "C Bacillus cereus is known to produce, among other species, phospholipase C, which hydrolyses phosphogl ycerides leading to the formation of phosphoric acids.s.9 Streptococcus faecalis degrades carbohydrates, particularly D-gluclose, through homolactic fermentation, yielding lactic acid as the sole end product.X.9 These characteristics explain the large increase in the peak current (1) and the corresponding decrease in pH observed in the presence of these two types of micro- Table 2 Analysis of UHT bovine milk samples 216 h after inoculation with different bacterial species, kept at 4 "C and equilibrated at 32 "C immediately before the measurements i (kO.l)/ Sample nA pH(k0.01) Reference 15.4 6.74 Streptococcus faecalis 15.1 6.72 Bacillus subtilis 15.0 6.72 Aeromonas 14.9 6.70 Cory nebacterium 15.0 6.72 organism, as shown in Table 1 and Figs.2(a) and 3(a). Data obtained for Bacillus cereus indicate a higher activity with respect to Streptococcus faecalis in relation to the amount of acids produced, whereas the decrease in the pH does not correspond to the rate at which the acids are formed. A possible explanation for this behaviour is that the mixture of phosphoric acids, such as, for instance, H2P04- and HP042-, produced by Bacillus cereus, might act as a buffer system.This would lead to a smaller pH change in these milk samples compared with those contaminated with Streptococcus faecalis which yields lactic acid as the sole product. On the other hand, as shown previ~usly,~ the increment of the peak current (1) is dependent on the amount of H2P04- or lactic acid formed regardless of their strength, and is related only to the number of hydrogen ions released in their dissociation, which is one for both species at the pH of milk. More interesting results were obtained when the samples were contaminated with Staphylococcus aureus and Bacillus subtilis, both of which are proteolytic bacteria and are able to ferment carbohydrates. lo These characteristics can lead either to the formation of amino acids, nitrogen bases and ammonia, or to the formation of acids such as lactic and butyric acids.These products have different effects on the pH, and the expected result is a small variation in this parameter with respect to the blank. As the production of acids is the predominant process,s in agreement with the experimental data obtained here, the peak current (1) increases progress- ively with time [see Table 1 and Figs. 2(b) and 3(6)]; however, a significant variation in the current response (about 10-15%) is evident only some 144-192 h after the inoculation. Similarly, Aeromonas are also able to ferment carbo- hydrates yielding acids," but the simultaneous production of indole from their metabolism attenuates their effect on the pH.This verifies the experimental findings reported here, in that a large increase in the peak current corresponds to a moderate decrease in the pH [see Table 1 and Fig. 3(c)]. Corynebacterium is a lipolytic micro-organism10 exhibiting clearly defined nutritional needs and characterized, under normal conditions, by a very slow growth rate.8 From the data obtained in the presence of this species [see Table 1 and Fig. 3(d)], it can be stated that only the amperometric method provided evidence for the alteration of the sample during the period of analysis (220 h), with an increase ill the current of about 11% with respect to the blank, whereas the pH response remained virtually unchanged, with a variation of only 0.1-0.2% (which is within the experimental error).The possibility of having some type of warning sign in this instance is very important from a sanitary point of view, as Corynebac- terium is recognized as an agent of some of the diseases transmitted by milk, for instance, diphteritis.8.10 Influence of the Temperature and the Interaction With the Atmosphere The growth rate of the bacterial population is governed by various factors, such as the temperature and the oxygen level. Moreover, if the samples come into contact with the environ- ment, other bacteria, apart from those studied here, might contaminate the milk.936 ANALYST, SEPTEMBER 1991, VOL. 116 Table 3 Peak currents (i k 0.1 nA),* pH values (kO.01) and their variation (in YO) obtained for bovine milk samples contaminated with different bacterial species, kept at 32 "C, under aerobic conditions Samples contaminated with: Reference Streptococcus Bacillus samples faecalis su btilis Timet it Ai ApH it Ai ApH it Ai APH h nA (YO)? pH (YO)$ nA (%)? pH (%)$ nA (YO)? pH (YO)$ 8 15.1 0.00 6.72 0.00 15.0 0.00 6.71 0.00 15.4 0.00 6.69 0.00 26 16.6 9.94 6.62 1.50 17.0 13.3 6.52 2.81 16.6 7.80 6.62 1.10 56 18.3 21.2 6.56 2.44 25.0 66.7 5.55 17.3 70.0 354 4.44 33.6 75 32.3 114 5.56 17.3 44.0 193 4.91 26.9 89.5 481 4.32 35.4 146 102 579 4.41 34.3 70.0 367 4.54 32.4 91.0 491 4.39 34.4 194 102 579 4.42 34.2 72.0 380 4.49 33.1 91.5 494 4.38 34.5 * Average values obtained from three replicates (relative standard deviation <1 SYo).? Variation of the peak current with respect to the initial value.$ Variation of the pH value with respect to the initial value. Aeromonas Corynebacterium it Ai APH nA (Y)? pH (YO)$ 14.8 0.00 6.72 0.00 26.0 75.7 6.28 6.46 39.3 165 5.57 17.1 86.5 484 4.48 33.2 85.0 474 4.51 33.0 - - - - it Ai nA (YO)? 15.0 0.00 16.4 9.30 23.8 58.7 26.2 74.7 82.5 450 83.5 457 APH pH (Yo)$ 6.71 0.00 6.64 1.10 6.28 6.38 6.17 8.00 4.53 32.5 4.52 32.6 The current (i) and pH values obtained for a number of milk samples treated with some of the micro-organisms listed in Table 1, but stored at 4 "C for 216 h, are given in Table 2. At this temperature the inoculated bacteria do not grow to any significant extent and their activity is negligible ,1 in agreement with the experimental results obtained here, the responses found being very close to the initial responses.At this temperature the methods studied are unable to provide evidence for bacterial contamination. Data for a series of measurements performed on samples inoculated with several different bacteria, kept at 32 "C, are presented in Table 3. A hole was made in the packs in order to allow interaction between the milk and the atmosphere. A large enhancement of the peak current and a corresponding large decrease in the pH was observed for both the reference and the contaminated samples. These values greatly exceed those obtained under aseptic conditions and the variations are not easily predictable. These findings demonstrate that the data reported in Table 1 are not affected by external influences. In conclusion, an amperometric method based on the use of a microelectrode has been shown to be useful for monitoring the variations in the acidity of milk, caused by microbic contamination; the method appears to be more sensitive than the potentiometric method usually applied for this purpose.Moreover, as the use of microelectrodes generally leads to a simplification of the instrumentation and requires small volumes for the analysis,' miniaturization of the apparatus should be possible. The authors thank D. Rudello for skilful experimental assistance. Financial aid from the Italian National Research Council (CNR) is also gratefully acknowledged. 1 2 3 4 5 6 7 8 9 10 11 References Walstra, P., and Jenness, R., Dairy Chemistry Physics, Wiley, New York, 1984. Egan, H., Kirk, R. S., and Sawyer, R., Pearson's Chemical Analysis of Foods, Churchill Livingstone, New York, 1981. Daniele, S., Baldo, M. A., Ugo, P., and Mazzocchin, G. A., Anal. Chim. Acta, 1990, 238, 357. Banwart, G. J . , Basic Food Microbiology, Van Nostrand Reinhold, New York, 1981. Fleischmann, M., Lasserre, F., Robinson, J., and Swan, D., J. Electroanal. Chem. Interfacial Electrochem., 1984, 177, 97. Silliker, J. H., Elliott, R. P., Baird-Parker, A. C., Bryan, F. L., Christian, J. H. B., Clark, D. S., Olson. J . C., and Roberts, T. A., Microbial Ecology of Foods, Academic Press, New York, 1980. Fleischmann, M., Pons, S., Rolison. D. R., and Schmidt, P. P.. Ultramicroelectrodes, Datatech, Morganton, 1987. Tiecco, G., Microbiologia Degli Alimenti di Origine Animale, Edagricole, Bologna, 1984. Lehninger, A. L.. Biochemistry, Worth, New York, 1975, p. 291. Vitagliano, M., Zndustrie Agrarie, UTET, Torino, 1976, p. 413. Alais, C., Scienza del Latte, Tecniche Nuove, Milan, 1988. Paper 1 I00368 B Received January 25th, 1991 Accepted May 3rd, 1991

ISSN:0003-2654    

DOI:10.1039/AN9911600933

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

ANALYST, SEPTEMBER 1991, VOL. 116 937 Study of Complexation Equilibria Using Polarized Metallic Electrodes V. F. Vetere and R. Romagnoli" ClDEPlN T, Research and Development Center for Paint Technology, CIC-CONICET, Calle 52 Entre 727 y 722, (1900) La Plata, Argentina The response of a metallic electrode depends on its chemical nature and on the characteristics of the environment. The response time and the repeatability of the measurements depend on the previous history of the electrode and are determined by its superficial structure. In a very general sense, it can be stated that a metallic electrode seldom acquires the potential predicted by the Nernst equation, because the equilibrium MO e Mz+ + ze- is not easily established at the electrode surface. Here, the response of an anodically polarized electrode was studied in relation to the response time and the repeatability o f the measurements; the electrode was then employed t o determine stability constants by a procedure described in the literature. In order t o be suitable for analytical purposes, an electrode must give fast and reproducible measurements.By polarizing a metallic electrode with low anodic current intensities a fast response time and good repeatability were achieved. The response time was less than 10 s. This procedure enables base metal electrodes t o be employed in situations where platinum, mercury or gold electrodes are preferred. In order t o determine stability constants, a metallic electrode was chosen (according t o the cation of which the complexes were t o be studied) and was polarized anodically with different current intensities. The potential of the electrode was recorded for increasing ligand concentrations. This particular technique has some advantages with respect t o the most common methodologies currently known.The problem of the dependence of the shape and the position of the polarization curves on the stability constants of the complex ions and the influence of the electrode surface in voltammetric techniques is solved by working at low anodic current intensities and by taking measurements when the electrochemical equilibrium has been reached. The equilibrium MO * ME+ + ze- is readily achieved at the electrode surface when the electrode is polarized; this represents a considerable advance with respect t o potentiometric techniques in general.The total cation concentration does not change during the determination, and depends on the magnitude of the polarizing current. The free ligand concentration virtually coincides with the total ligand concentration because the concentration of cation generated at the electrode/solution interface is about 1 x 10-5 mol dm-3. Keywords: Complexation equilibria; polarized metallic electrode; stability constant; potentiometry The purpose of the work described here was to develop a potentiometric method employing polarized electrodes which could be applied in the field of electroanalysis. In this particular instance it was used to determine the stability constants of complex ions. Firstly, the response of a metallic electrode polarized with low anodic current densities was studied.In the second stage, the polarized metallic electrode was employed in the above-mentioned potentiometric tech- nique, which, in turn, was applied to the determination of the stability constants of complex ions. The use of polarized metallic electrodes enables the calculation of the formation constants to be simplified. The response of a metallic electrode depends on its chemical nature and on the nature of its environment. The response time and the repeatability of the measurements depend on the previous history of the electrode and are determined by its superficial structure. For this reason, a platinum electrode is preferred to other indicator electrodes in potentiometry. A platinum electrode is inert and does not take part in most chemical reactions.Silver, copper, zinc, cadmium and mercury electrodes can be used in this way but with some limitations. Most metals are not suitable for use as indicator electrodes in potentiometry because the equilibrium MO e Mr+ + ze- is not easily established at the electrode surface. 1 These electrodes, in almost all instances, under practical conditions do not acquire the electrode potential predicted by the Nernst equation.2 These factors seriously affect the measurements, particularly the potentiometric measurements; however, these electrodes are used for this type of determination because only potential changes related to concentration changes are of interest. 1,2 The difficulties increase as the complexes become weaker and the metal used for the electrode becomes less noble.* To whom correspondence should be addressed. Several different treatments can be used to overcome the difficulties mentioned above. Each treatment attempts to generate a definite and reproducible superficial configura- tion.3-5 Some are not sufficiently effective whereas others are too complex and laborious to be employed in the field of analytical chemistry. Voltammetric techniques are often used to study metal- ligand equilibria in solution. The electrode processes must be reversible and the chemical equilibrium fast. The ligand concentration, the stability constants and the nature and state of the electrode surface all affect the shape and position of the peaks in the voltammograms.4~6-~~ The displacement of the waves depends on the ligand concentration in the bulk solution and on the stability constants of the complexes. The ligand concentration must be higher than the metal ion concentration. 11-14 Garnppls has developed a technique to calculate stability constants for complex ions at low metal to ligand ratios.This procedure involves application of the least-squares and simulation methods to data obtained from cyclic or linear-sweep voltammetry. Most workers employ a mercury or platinum electrode as these offer advantages with respect to other metallic electrodes.4.lh In the work described here, the response of an anodically polarized base metal electrode was studied in relation to the response time and the repeatability of the measurements. Low anodic current densities were employed.In order to be suitable for analytical purposes an electrode must give fast and repeatable measurements. The treatment applied was com- pared with other treatments, which are reported to yield reproducible surfaces. Hence, the metallic electrode was polarized in solutions containing different ligand concentrations and the electrode potential recorded for each concentration. The total metal concentration was constant for each polarization current. This procedure is advantageous for ions which are susceptible to938 ANALYST, SEPTEMBER 1991, VOL. 116 atmospheric oxidation, because the manipulation of such ions is avoided by generating them at the electrode-solution interface. The equilibrium constants were calculated by taking into account conventional calculation procedures.17-19 The values obtained using this technique were compared with those reported in the literature.20 Experimental Four copper disc and four lead disc electrodes, made from the corresponding metallic rods of spectrochemical purity, were used. The electrodes were 0.5 cm in diameter and were supported in Teflon bars, polymeric acrylic matrices or Araldite resin bodies.For potentiometric measurements, with no polarization of the electrode, the exposed area was increased to approximately 4 cm2. Each electrode was abraded with No. 600 emery paper and de-greased with alkaline substances such as sodium carbonate or calcium hydroxide. The working electrode is shown in Fig. 1. The reference electrode was an Orion Model 90-01 satu- rated calomel electrode (SCE).The electrolytic cell is illustrated in Fig. 2. The potential measurements were made with an Orion Model 701 A voltmeter. The electrodes were dipped into copper(I1) or lead perchlor- ate solutions, of varying concentration from 1 x 10-4 to 1 x 10-1 mol dm-3, including solutions in which the cation concentration was equal to zero. The ionic strength was adjusted with a sodium perchlorate solution of higher concen- tration to match the value of a 1 rnol dm-3 sodium perchlorate solution. U Fig. 1 Working electrode: (a) side view; and (b) top view U Fig. 2 Electrolytic cell: ET, working electrode; CE, counter electrode; ER, reference electrode; P, sintered glass; S, agar bridge; A, stirrer; B, nitrogen bubbler tube; and 0 and D, holes In the first stage, the electrodes were polarized anodically with different currents viz., 100, 50 and 25 PA, in order to study their behaviour. In all instances, the response time and the final value attained by the electrode potential were recorded.The current was then switched off and the time required for the potential to reach a final stable value was recorded again. This procedure was repeated four times. The polarization circuit is shown in Fig. 3. The results obtained are presented in Tables 1 and 2. The electrodes were subjected to other surface treatments (as described in the literature314) such as: mechanical polish- ing, chemical etching and electroplating. In this instance, the stabilization of the electrode potential took 2-4 h and differences in the measured values ranged from 20 to 50 mV.When the electrodes were subjected to anodic polarization the response time was less than 10 s and an insignificant scatter in the measurements was observed. In the second stage, in order to determine the stability constants of complex ions, a metallic electrode was chosen (according to the cation of which the complexes were to be studied) and was polarized anodically with different current intensities. The potential of the electrode was recorded for increasing ligand concentrations. The electrode potential in the absence of polarization was also recorded. The systems copper acetate, lead acetate and lead oxalate were studied. Hence, it was necessary to prepare a set of acetate solutions with concentrations ranging from 1 X to 2 rnol dm-3 and a set of oxalate solutions with concentrations varying from 1 x 10-4 to 0.25 rnol dm-3.The ionic strength was kept constant at the above-mentioned level. The results obtained are shown in Fig. 4. +- Fig. 3 Polarizing circuit: ET, working electrode; CE, counter electrode; ER, reference electrode; C, electrolytic cell; V, voltmeter; A, ammeter; and R, variable resistance Table 1 Copper electrode potential E (mV) versus SCE for several molar concentrations of added cation and for different anodic polarizations ilpA [Cation]/ rnol dm-3 100 50 25 0 0 -27.5 -41.0 -53.9 -87.0 1.0 x 10-4 - 10.8 -13.3 - 14.4 - 14.9 1.0 x 10-3 +15.5 +15.2 +15.1 +15.0 1.0 x 10-2 +45.1 +45.0 +44.8 +45.2 1.0 x lo-' +75.0 +75.2 +74.9 +75.1 Table 2 Lead electrode potential E(mV) versus SCE for several molar concentrations of added cation and for different anodic polarizations i/pA [Cation]/ rnol dm-3 100 50 25 0 0 -515.6 -524.6 -531.0 -575.0 1.0 x 10-4 -495.6 -497.1 -497.8 -499.0 1.0 x lo-' -467.6 -467.8 -467.9 -468.1 1.0 x 10-2 -437.0 -437.0 -436.8 -437.1 1.0 x 10-1 -406.0 -406.1 -406.2 -406.0ANALYST, SEPTEMBER 1991, VOL.116 939 In all instances, the solutions were previously de-gassed with purified nitrogen and stirred at 1800 rev min-1. The working temperature was 20 & 0.2 “C. Results and Discussion Study of the Electrode Response This study will be restricted to the anodically polarized electrodes because the other treatments applied to the electrodes did not yield reproducible responses within a relatively short time.A simplified model was developed in order to understand the response of an anodically polarized electrode. When the electrochemical equilibrium is reached, the electrode poten- tial is related to the concentration by a relationship described by the Nernst equation. Two cases must be considered. (1) No complexing agent is added to the system under study: E() = K + a log(cM0 + CMa) E = K + a log(CMo + cMa + ki) + q(e) (1) (2) (with no current flowing through the cell) (current flowing through the cell) (2) A complexing agent is added to the system: E’0 = K + a log(CMo + cMa) f (3) (with no current flowing through the cell) E’ = K + a + log (cM0 + cMa + ki) f + q’(e) (4) (current flowing through the cell) where E = electrode potential versus SCE, Eo = electrode potential when no current is flowing in the cell, E’ = electrode potential in the presence of a complexing agent, K = electrode constant (this resembles the parameter E in the Nernst equation), a = Nernstian slope [a = 2.3RT/zF ( R = gas constant, T = temperature, z = charge and F = Faraday constant)], cM0 = cation activity generated spontaneously at the electrode-solution interface, cMa = added cation activity, ki = cation activity generated because of the current flow at the electrode-solution interface, f = fraction of uncomplexed cation (= free cation activity/total cation concentration), and q(e) and q’(e) = polarizations at the electrode-solution interface, different from concentration polarization.It has been shown that the slope of the plot of E versus log cMa is a/2.3.11 From the data in Tables 1 and 2, the values of a and K were calculated to be 30 mV decade-’ and +I05 mV versus SCE, respectively, for the copper electrode, and 31 mV decade-’ and -375 mV versus SCE, respectively, for the lead electrode.160 140 120 u v) $100 $ 80 > E I& 60 40 20 0 -1 -0.000122 1 -4 -3 -2 Log D-I Fig. 4 Average electrode potential difference (m) as a function of log [L] for A, copper acetate; B, lead acetate: and C, lead oxalate From eqn. (1) cMo was calculated to be 4.0 X 10-7 mol dm-3 for the copper electrode and 3.5 x 10-7 mol dm-3 for the lead electrode. From eqn. ( 2 ) , the cation concentration at the electrode- solution interface for a given anodic current intensity can be calculated in the usual way.11 It ranges from 4.0 x 10-7 to 3.8 x 10-5 mol dm-3 for the copper electrode and from 3.5 X 10-7 to 2.9 x 10-5 mol dm-3 for the lead electrode when the current intensity is varied from 0 to 100 FA.As the value of E’ - E is similar to Eo’ - Eo (see Tables 1 and 2 and Fig. 4), it can be concluded that the difference q’(e) - q(e) approaches zero; hence the calculation of f is unaffected by the polarizations occurring at the electrode surface. Calculation of the Stability Constants The material balance equations for a multi-stage equilibrium involving complex ions are: CM = [MI + [ML] + [ML2] + [ML3] + . . . CL = [L] + [ML] + 2[ML2] + 3[ML3] + . . . ( 5 ) (6) where M = metal and L = ligand. In order to solve these equations it is necessary to know CM and CL and to determine [MI or [L] experimentally.Several calculation procedures have been described in the literature.17-*9 When employing polarized electrodes, the cation concentration generated at the interface by polarization is about 1 X 10-5 rnol dm-3; hence it can be stated that [L] = cL. The value of the quotient cM/[M], which is the inverse off, must be known instead of [MI. This value can be obtained from the plot of E - E’ versus log [L] (Fig. 4). Once this value has been obtained, the successive formation constants ( k ) can easily be obtained by regression analysis using eqn. (7) : 1 - = CM/[M] = 1 f ki[L] + klk2[L] + k1k2k3[Ll3 -!- . . . (7) f In this instance a regression analysis computer program was employed21 based on the procedure described by Mar- quardt.22 The results are presented in Table 3 and compared with those reported in the literature.20 Good agreement was found.Conclusions By polarizing a metallic electrode with low anodic current intensities, a fast response time and good repeatability were achieved. The response time is less than 10 s. This procedure proved to be superior to coating the electrode with the corresponding metal at constant current intensities. The other surface treatments applied did not yield repeatable responses and good response times unless the electrodes were subse- quently polarized with low anodic current intensities. When working at low anodic current intensities, the electrode slopes appeared to be Nernstian. The cation concentration generated at the electrode/solution interface does not compete with the concentration of the metal because the current intensities are low.When low anodic current intensities are employed, the polarizations occurring at the electrode/solution interface cancel each other out; hence the calculations carried out from the potential measurements are not subject to any serious error from the polarizations. The electrode polarized in this way could be used to determine free metal concentrations in the same way as a primary electrode, provided that [M2+] >> cM0, but with two advantages: better repeatability and a faster response time. When determining equilibrium constants, the problem of the dependence of the shape and position of the polarization curves on the stability constants of the complex ions is avoided940 ANALYST, SEPTEMBER 1991, VOL.116 Table 3 Comparison of the experimental values of the stability constants (k) with those obtained from the literature for media of ionic strength 1 mol dm-3 Literature values* Experimental values Temperature/ Temperature/ Ligand Metal Equilibrium "C k "C k * Reference 20. t This value corresponds to an ionic strength of zero. 51.3 10.0 2.45 0.63 7.58 2.63 0.20t 126 2089 151 20 49.2 10.6 2.35 0.66 8.33 2.30 0.30 20 123 20 2100 150 by polarizing the electrode with low anodic current intensities and taking measurements when the electrochemical equi- librium has been reached. The equilibrium MO e Mz+ + ze- is readily achieved at the electrode surface when the electrode is polarized. This represents a considerable advance with respect to potentiometric techniques in general.The total cation concentration does not change during the determination, and depends on the magnitude of the polariz- ing current. It is not necessary to know the true value of the cation concentration at the electrode/solution interface; instead, the fraction of uncomplexed cation U, is calculated from potential measurements. The free ligand concentration virtually coincides with the total ligand concentration over the whole range of concentrations because the concentration of cation generated at the electrode/solution interface is about 1 x 10-5 mol dm-3. This methodology is not restricted to mercury or noble metal electrodes; base metal electrodes can be employed provided they are polarized with low anodic current intensities and provided the metal does not become passive in the chosen medium. Unstable cations can be studied by using this procedure because it is not necessary to handle solutions of the cation, which is generated electrolytically at the electrode/ solution interface.We thank CIC (Comisi6n de Investigaciones Cientificas) and CONICET (Consejo Nacional de Investigaciones Cientificas y Tkcnicas) for their sponsorship of this research. References Harris, D. C., Quantitative Chemical Analysis, Freeman, New York, 1987. Delahay, P., New Instrumental Methods in Electrochemistry, Interscience, New York, 1954. Encyclopedia of Electrochemistry of the Elements, ed. Bard, A. J., Marcel Dekker, New York, 1978. 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Charlot, G., Badoz-Lambling, J . , and TrCmillon, B., Electro- chemical Reactions, Elsevier, Amsterdam, 1962. Cervifio, R. M., Triacca, W. E.. and Arvia, A. J., J. Electroanal. Chem., 1985, 182, 51. Killa, H. M., Mercer, E. E., and Philp, R. H., Anal. Chem., 1984, 56, 2401. Killa, H. M., and Philp, R. H., J. Electroanal. Chem., 1984, 175,223. SavCant, J. M., and Xu, F., J. Electroanal. Chem.. 1986, 208, 197. Spell, J . E., and Philp, R. H., J. Electroanal. Chem., 1980,112, 281. Castleberry, A. A . , Mercer, E. E., and Philp, R. H., J. Electroanal. Chem., 1987, 216, 1. Bard, A. J . , and Faulkner, L. R., Electrochemical Methods, Wiley, New York, 1980. Lingane, J. J.. Chem. Rev., 1941, 29, 1. De Ford, D. D., and Hume, D. N., J. Am. Chem. SOC., 1951, 73,5321. Schaap, W. B., and MacMasters, D. L., J. Am. Chem. SOC., 1961, 83, 4699. Gampp, H., Anal. Chem., 1987, 59, 2456. Kolthoff, 1. M., and Lingane, J. J., Polarography, Interscience, New York, 1952, vol. 1. Yatsimirskii, K. B., and Vasil'ev, V. P., Instability Constants of Complex Compounds, Consultant Bureau Enterprises, New York, 1960. Sullivan, J. C., and Hindman, J. C., J. Am. Chem. SOC., 1952, 74,6091. Rossotti. F. J . C., and Rossotti, K. S . , Acta Chem. Scand., 1955, 9 , 1166. Martell, A. E., and Smith, R. M., Critical Stability Constants. Volume 3. Other Organic Ligands, Plenum Press, New York, 1979. Marquardt, D. W., Share Program No. 3094, 1963. Marquardt, D. W., SIAM J. Appl. Math., 1963, 11, 431. Paper 0101529F Received April 4th, 1990 Accepted April 22nd, 1991

ISSN:0003-2654    

DOI:10.1039/AN9911600937

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

ANALYST, SEPTEMBER 1991, VOL. 116 941 Differential-pulse Polarographic Behaviour of Selenium in the Presence of Copper, Cadmium and Lead Hasan Aydin Gazi Universitesi, Fen-Edebiyat Fakultesi, Kim ya Bolumu, Ankara, Turkey G. H. Tan* Department of Chemistry, University of Malaya, 59 100 Kuala Lumpur, Malaysia A differential-pulse polarographic method for the determination of Se is described. The effect of Cu, Pb and Cd on the Se peaks was studied. In order to find a suitable medium, HC104 (pH I), 0.1 rnol dm-3 KN03 (pH 4), a buffer solution of pH 4 (CH3COOH-H3B03-H3P04), a buffer solution of pH 9 (NH3-NH4CI) and 10% Na2S03 (pH 9.5) were studied. In NH3-NH4CI and 10% Na2S03 media the expected peak for Se032- was not observed; however, the SeS032- ion gave a peak at -800 and -600 mV, respectively.In addition, in the CH3COOH-H3B03-H3P04 buffer solution of pH 4, Se032- gave two peaks at -610 and -1310 mV. The detection limit for Se in a buffer solution (pH = 4) of CH3COOH-H3B03-H3P04 was 1 x 10-8 rnol dm-3. The method was applied t o some environmental samples. Keywords: Selenium; lead; cadmium; differential-pulse polarograph y; environmental samples The importance of Se as an essential trace element together with Cu, Cd and Pb in animal nutrition, natural waters and waste waters has stimulated the development of many sensitive analytical methods for its determination. In order to measure the amount of Se in different media, atomic absorption spectrometry,lJ neutron-activation analysis,3 chromatography,'$ polarography,s-9 differential-pulse strip- ping voltammetry,'() cathodic stripping voltammetry," anodic stripping voltammetry12-14 and titrimetry's have been used.However, some of these methods require elaborate instrumentation and precisely controlled experimental con- ditions. The determination of Se by pulse polarographic methods is well-known, having been studied by many workers. The main difficulty encountered in the past has been the influence of heavy metal ions on the determination of Se by electrochemical or other methods. However, in order to eliminate these interferences, many workers have used Chelex-100 resin,8 ion-exchange separation16 and other methods. This paper describes a simple pulse polarographic method for the determination of Se, Cu, Cd and Pb in various environmental samples. The proposed method permits the determination of Se in the presence of heavy metal ions.In addition, the effect of the metal ion concentrations on the Se peak, the effect of the medium on the peak potentials of Se and the application of the method to some environmental samples are also described. Experimental Apparatus Polarograms were obtained with a Princeton Applied Research PAR 174 A polarographic analyser and a Hewlett- Packard x-y type recorder. A two-electrode system was used, with a dropping mercury electrode (DME) as the working electrode and a saturated calomel electrode (SCE) as the reference electrode. Reagents All reagents used were of analytical-reagent grade. Distilled, de-ionized water was used for preparing all the solutions and at all stages of analysis. * To whom correspondence should be addressed.Buffer solution (pH 4). Prepared by mixing equal parts of 0.04 rnol dm-3 CH3COOH, 0.04 rnol dm-3 H3P04 and 0.04 rnol dm-3 H3B03. The pH of the resulting solution was adjusted to 4.0 by the addition of about 25 ml of 0.2 rnol dm-3 NaOH solution to 100 ml of this stock solution. Ammonia-ammonium chloride buffer solution. Prepared by dissolving 100 g of NH4CI in 380 ml of NH3 solution and diluting to 500 ml with de-ionized water. Stock standard Se solution, 1 X 10-2 rnol dm-3. Prepared by dissolving 0.1729 g of Na2Se03 in 100 ml of de-ionized water. Working standard Se solutions, 1 x 10-4 and 1 x 10-5 rnol dm-3. Prepared fresh each day by diluting the 1 x 10-2 rnol dm-3 stock standard solution. Working standard Cu" solutions, 1 x 10-4 and 1 x 10-5 rnol dm-3.Prepared by diluting a 1 x 10-2 rnol dm-3 stock solution of Cu(N03)2. Working standard Pb" solutions, 1 x 10-4 and 1 x 10-5 mol dm-3. Prepared by diluting a 1 x 10-2 rnol dm-3 stock solution of Pb(N03)2. Working standard Cd" solutions, 1 x 10-4 and 1 X 10-5 rnol dm-3. Prepared by diluting a 1 x 10-2 rnol dm-3 stock solution of Cd(N03)2. Procedure Synthetic samples for analysis were prepared by adding aliquots of a concentrated supporting electrolyte solution to the analyte solution and the pH was measured with a pH meter. Thorough de-aeration of the solution to be analysed was required to avoid interference from reduction of dissolved oxygen. Pre-purified nitrogen was used for de-aeration after pre-saturating the gas with water by bubbling it through freshly prepared de-ionized water.Nitrogen was passed through the solutions for about 30 min before each exper- iment. In order to study the interference of heavy metal ions on the Se peak and to determine the peak potentials of the heavy metal ions and Se, various experiments were carried out by adding the interferents to the supporting electrolyte both before and after the addition of Se. The influence of the potential scan rate, mercury drop time and supporting electrolyte solutions on the peak height was studied. In each experiment, one of the parameters was varied while the others were kept constant. Water samples from the Klang River (see Fig. 1 for map) were used for application of the method. A 5 ml volume of (1 + 1) or HCI-HN03 (3 + 1) was added to 100 ml of the water HN03, HCI04-HN03-H2S04 (24 + 24 + 2), HC104-HN03942 ANALYST, SEPTEMBER 1991, VOL.116 Fig. 1 Klang River basin: sampling station (A = sampling station) ~~ Table 1 Peak potential of Cu, Pb, Cd and Se in various media Peak potential/mV versus SCE Supporting electrolyte PH c u Pb Cd Se SeS03*- HC104 1.0 0.00 -415 -615 -510 - 0.1 mol dm-3 KN03 4.0 - 60 - 370 - 480 - 650 Buffer solution* 4.0 -0.00 -390 - 680 -610 - - 1310 - 800 Buffer solution? 9.5 -270 - 490 - 780 10% Na2S03 9.3 - - -610 - -600 - - * CH3COOH-H3B03-H3P04. ? NH3-NH4Cl. sample and the volume was then reduced to 10 ml by evaporation. After this procedure, the volume was adjusted to 100 ml with distilled water. The samples that had been acidified only with HN03 were not heated.The pH of the samples was adjusted to 4 with NaOH solution. After pH adjustment, 5 ml of 0.1 mol dm-3 KN03 (pH 4) or buffer solution of pH 4 were added to 20 ml of the sample solution. In order to obtain the SeS032- ion, 1 .O g of Na2S03 was added to the acidified sample solution. The pH of the solutions was adjusted to 7-8 by using NaOH solution and 10 ml of 10% Na2S03 solution were then added to each sample. In order to accelerate the formation of SeS032-, the solutions were heated. A 5 ml volume of 10% Na2S03 or NH3-NH4Cl buffer solution was added to 20 ml of these solutions before analysis. Results and Discussion In differential-pulse polarography the peak potential of metal ions depends on the supporting electrolyte and on the pH of the electrolyte solution.The electrolytes mentioned above were used for the determination of the peak potential of Cu, Pb, Cd and Se. The results of these experiments are given in Table 1. It was found that the use of a buffer solution containing CH3COOH, H3B03 and H3P04 to determine the amounts of Cu, Pb, Cd, and Se gave two peaks for Se at -610 and -1310 mV (see Fig. 2). The influence of Cu, Pb and Cd ions on these peaks was studied and it was found that the second peak for Se 3 0 500 1000 1500 - EImV versus SCE Fig. 2 Differential-pulse polarographic behaviour of SelV. 1, Blank solution; 2, 63.9 pg 1-1 of SelV; and 3, 127.8 pg 1-1 of SeIv. Measurement of the peak height is indicated by x at - 1310 mV could be used for the determination of Se in the presence of Cu, Pb and Cd.The peak at -1310 mV was still observed when the Se concentration was 1 x 10-7 mol dm-3. When Cu, Pb and Cd solutions were added to the Se solution,ANALYST, SEPTEMBER 1991, VOL. 116 943 0 500 1000 1500 - NmV versus SCE Fig. 3 Effect of Pb" and Cd" on the determination of Se". 1, 127.8 pg 1-1 of Se'"; 2.38.3 pg 1-1 of Pb" + 127.8 pg 1-1 of Selv; 3,78.6 pg I-' of Pb" + 127.8 pg 1-1 of Se'"; 4,24.4 yg 1-1 of Cd" +78.6 pg 1-1 of Pb" + 127.8 pg 1-1 of SdV; and 5, 48.8 pg 1-1 of Cd" + 78.6 pg I-' of Pb" +127.6 pg 1-1 of Selv. Measurement of the peak height is indicated by x Table 2 Effect of Pb and Cd on the determination of Se in buffer solution (pH 4) ConcentratiodlO-6 mol dm-3 Peak height (arbitrary units) Se032- Pb2+ Cd2+ Se Se* Pb Cd 25f 1 .o 501- 2.0 2.0 0.26 - 427 2.0 0.52 - 347 2.0 0.52 0.22 llj- 2.0 0.52 0.44 - 2.0 0.52 1.10 1% 2.0 0.52 2.50 71§ 2.7 0.52 2.50 549 3.4 0.52 2.50 370 - - - - * At -1300 mV.7 At -610 mV. $ At -560 mV. 0 At -580 mV. 7 14 14 14 14 14 14 14 21 28 - - - - - 12 24 24 24 24 24 5 24 5 24 5 - - - - ~~ ~ ~ Table 3 Effect of Se on the determination of Cu, Pb and Cd in buffer solution (pH 4) ConcentratiodlO-6 mol dm-3 Peak height (arbitrary units) Se Cu Pb Cd Se Se* Cu Pb Cd - 0.72 - - - - 5 - - - 0.72 - 0.78 - - 5 - 3 - 2.10 - 1.60 - - 16 - 7 - 2.10 0.87 1.60 - - 16 4 7 - 2.10 1.70 3.20 - - 16 8 14 0.76 2.10 1.70 3.20 - 2 16 8 10 2.30 2.10 1.70 3.20 - 6 14 8 6 3.80 2.10 1.70 3.20 - 10 11 8 2 5.30 2.10 1.70 3.20 - 14 10 8 0.5 9.10 2.10 1.70 3.20 - 2 4 7 8 - * At -1300 mV these ions were found to affect the Se peak at -610 mV (Table 2).However, these ions did not interfere with the other Se peak at -1310 mV (see Fig. 3). Hence it was possible to determine 1 x 10-7 mol dm-3 Se in a solution containing Cu, Pb and Cd ions with a relative,standard deviation of 1.5%. In addition, it was found that the influence of Cd was greater than that of Cu and Pb. For instance, when Se was added to a I I I I I I 1 Potential scan rate/mV s-' 0 2 4 6 8 1 0 1 2 Fig. 4 Effect of sweep rate on the peak height of SeIv at different mercury drop times in buffer solution (pH = 4). Drop time: A, 0.5; B, 1.0; C, 2.0; and D, 5.0 s 40 I 0 2 4 6 8 1 0 1 2 Potential scan rate/mV s-1 Fig. 5 Effect of sweep rate on the peak height of SeIV at different mercury drop times in 10% Na2S03 solution.Drop time: A, 0.5; B, 1.0; C, 2.0; and D, 5.0 s 2o t v) C 4- .- 15 - 2 E + -!E 10 E .I- .- - (3, 0) -c .- 0 2 4 6 8 1 0 1 2 Potential scan rate/mV s-l Fig. 6 Effect of sweep rate on the peak height of SelV at different mercury drop times in NH3-NH4CI buffer solution. Drop time: A, 0.5; B, 1.0; C, 2.0; and D, 5.0s buffer solution (pH 4) containing Cu, Pb and Cd ions, the Cd peak began to decrease and eventually became a small peak, after which the Cu peak also began to decrease. In another experiment with a buffer solution (pH 4) containing Se032-, Pb and Cd ions were added respectively. In this experiment, after observing the influence of Pb on the Se peak, Cd ion was added. The effect of Pb and Cd ions on the Se peak is shown in Tables 2 and 3.In this experiment the peak potential of Se changed from -610 to -560 mV after the addition of Cd. When Se was added to the same solution, the same peak was shifted to -580 mV. In addition, the unknown peak at -420944 v) Y '5 10 2 2 2 8 - C .- 6 - C r u, .- 2 4 - 2 2 - Y (D 12 I 1 - I I I 1 I I 0 1 2 3 4 5 6 Mercury drop time/s Fig. 7 Effect of mercury drop time on the peak height of Sel" at different scan rates in the buffer solution (pH = 4; peak potential = -540 mV). Scan rate: A, 1; B, 2; C, 5; and D, 10 mV s-1 0 1 2 3 4 5 6 Mercury drop ti me/s Fig. 8 Effect of mercury drop time on the peak height of Sel" at different scan rates in the buffer solution (pH = 4; peak potential = -1310 mV). Scan rate: A, 1 ; B, 2; C, 5; and D, 10 mV s-1 0 1 2 3 4 5 6 Mercury drop t im e/s Fig.9 Effect of mercury drop time on the peak height of Set" at different scan rates in NH3-NH4CI buffer. Scan rate: A, 1; B, 2; C, 5 ; and D, 10 mV s-1 mV versus SCE was not observed until the addition of Cd after which the height of this peak increased with the addition of Cd2+ and Se032- solutions, respectively. The height of the unknown peak was proportional to the concentration of Cd and Se added to the buffer solution, i.e., the addition of Cu, Pb and Cd did not affect the Se peak at -1310 mV. In 10% Na2S03 or NH3-NH4Cl buffer solution, the Se peak was not observed between 0.00 and -1600 mV. However, when SeS032- was added to the 10% Na2S03 o r NH3-NH4CI buffer solution the Se peak was observed at -600 and -800 mV, respectively.Dunhu et al. 7 observed the Se peak at -570 mV in the HCIO4-Na2SO3-NH3-NH4CI-NH20H-HCI-KIO3 system (pH 10). They attributed this peak to the SeS03*- ion. ANALYST, SEPTEMBER 1991, 40 I VOL. 116 0 1 2 3 4 5 6 Mercury drop time/s Fig. 10 Effect of mercury drop time on the peak height of Set" at different scan rates in 10% Na2S03 solution. Scan rate: A, 1; B, 2; C, 5; and D, 10 mV s-1 Table 4 Determination of Cu, Pb, Cd and Se in water samples from the Klang River in Malaysia Monitor- ing Sample station No. No.* 1 624 2 610 3 629 4 628 5 605 6 626 7 627 8 621 9 604 Concentration of elements in sample (ppb) Cu Pb Cd Se - 62 22 - - 37 8 - 33 - - 32 5 - - 39 3 - - 42 3 - 41 12 - - 50 5 - 43 - - - - - - Se added to sample 35 16 16 25 25 25 35 16 16 (PPb) * See Fig.1 for map showing location of each station No. Se found 33 16 15 26 23 23 33 14 15 (PPb) In order to obtain the SeS032- ion, Na2S03 solution was added to an acidic Se032- solution. The amorphous Se formed was filtered off and then dissolved in 10% Na2S03 solution. In order to accelerate the formation of SeS03*- the solution was heated. This species undergoes electrochemical reduction on the DME and forms a distinct pulse polaro- graphic wave. The mechanism of this reaction is: SeS032- + 2e- -+ Se2- + SO3'- The SeS032- ion is initially decomposed into Se and SO3?--. Selenium is adsorbed on the electrode surface and reduced to Se2- at a potential of -600 mV in 10% Na2S03 and -800 mV in NH3-NH4C1 buffer solution. In order to obtain the maximum peak height, the effect of sweep rate and drop time of the DME on the peak height was studied at different scan rates in buffer (pH 4), 10% Na2S03 and NH3-NH4CI solutions.The results are shown in Figs. 4-10. In each experiment, one parameter was varied while the others were kept constant. The results showed that a higher scan speed resulted in a decrease in resolution and peak height. In addition, the effect of the potential scan rate on the peak height and mercury drop time varied according to the pH, peak potential and electrolyte solutions. For instance, the effect of the mercury drop time on the peak height of Se'" at different scan rates in a buffer solution of pH 4 varied according to the peak potential of Set". In order to obtain better resolution, scan rates of 1 and 2 mV s-1 were used.From the results presented in Figs. 7-10, the appropriate mercury drop time was selected. The wet digestion methods described above were applied to the river water samples. However, no significant differencesANALYST, SEPTEMBER 1991, VOL. 116 945 between the methods were found. Therefore, the digestion method involving the use of HN03 without heating was employed for the determination of Cu, Pb, Cd and Se. In addition, when a buffer solution of pH 4 was added to the samples, no detectable amounts of Cu and Se were found (Table 4). However, when Se was added to the samples before wet digestion, Se peaks were observed at -610 and -1310 mV. The results of these experiments obtained by using the peak at -1310 mV are shown in Table 4. References 1 Sturgeon, R. E., Willie, S. N., and Berman, S. S., Anal. Chem.. 1985. 57. 6. 2 Vijan, P. N., and Wood, G. R., Talanra, 1976, 23, 1. 3 Kronborg. 0. J., and Steinnes. E.. Analyst, 1975. 100, 835. 4 Shimoishi, Y.. Analyst, 1976, 101, 298. 5 Christian, G. D.. and Knoblock, E. C., Anal. Chem., 1963,35, 1128. 6 7 8 9 10 11 12 13 14 15 16 Hasdemir, E., and Somer. G., Analyst, 1990, 115, 297. Dunhu, W., Diyang, Z., and Xiaoming, L., Analyst, 1989,114, 793. Batley, G. E., Anal. Chim. A d a , 1986, 187. 109. Bound, G. P., and Forbes, S., Analysr, 1978, 103, 176. Vos. L., Komy, Z., Reggers, G., Roekens, E., and van Grieken, R.. Anal. Chim. Acta, 1986, 184, 271. Dennis, B. L., Moyers, J . L., and Wilson, G. S., Anal. Chem., 1976, 48, 1611. Posey, R. S., and Andrews, R. W., Anal. Chim. Acta, 1981, 124, 107. Aydin, H., and Somer, G., Anal. Sci., 1989, 5 , 89., Aydin, H., unpublished results. Aydin, H., and Somer, G., Talanta, 1989,36, 723. Adeloju, S. B., Bond, A. M., Briggs, M. H., and Hughes, H. C., Anal. Chem., 1983, 55, 2076. Paper I I001 67A Received January I4th, 1991 Accepted May 7th, 1991

ISSN:0003-2654    

DOI:10.1039/AN9911600941

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

ANALYST, SEPTEMBER 1991, VOL. 116 947 Thermodynamic and Kinetic Implications Involved in the Titration of Polyfunctional Acids by Catalytic Thermometric Titrimetry Oswaldo E. S. Godinho, Helena S. Nakatani, Ivo M. Raimundo Jr., Luiz M. Aleixo and Graciliano de Oliveira Net0 lnstituto de Quimica, Universidade Estadual de Campinas, C. P. 6154, 13081 Campinas, Siio Paulo, Brazil The influence of the concentration of reactants on the results of titrations of salicylic acid in acetone with potassium and tetramethylammonium hydroxides in propan-2-01 were investigated. The titrations were performed by catalytic thermometric titrimetry using acetone as the end-point indicator and by potentiome- tric titrimetry. These and other results are discussed in terms of the equilibrium and kinetic aspects involved in the titrations of polyfunctional acids capable of forming intramolecular hydrogen bonds.Keywords: Catalytic titrimetry, thermometric titrimetry, hydrogen bond, salicylic acid Vaughan and Swithenbankl introduced acetone as a thermo- metric end-point indicator in the titration of a variety of acidic substances with strong bases. In their method, the rise in temperature, caused by the exothermic reaction of dimeriza- tion of acetone catalysed by the excess of strong bases, is employed to locate the end-point of the titration. Greenhow and co-workers24 performed an extensive and systematic study of the titration of polyfunctional acids by catalytic thermometric titrimetry. They have shown the possibility of obtaining selectivity by appropriate control of the stoichiometry at the end-point of the titration.According to these workers, the variation of stoichiometry can be attained by choosing the appropriate end-point indicator, sample solvent, titrant or in some instances the concentration of titrant. They have investigated more specifically the influence of titrant and sample solvent on the stoichiometry attained in the titration of some hydrogen bonded polyfunctional acids.5 The influence of the sample solvent is related to the extent of dissociation of the titrant ion-pair with potassium hydroxide as titrant whereas tetrabutylammonium hydroxide is considered to be completely dissociated. A mechanism , which considered the formation of a four-centre intermediary between the half-neutralized difunctional acid and the potassium hydrox- ide ion-pair, was presented to explain the results. The neutralization and indicative reactions are considered to be competitive processes.In this paper, a study of the influence of the sample and titrant concentrations on the titration of salicylic acid (in acetone j with potassium hydroxide (in propan-2-01) is presen- ted. The results are compared with those obtained by using tetramethylammonium hydroxide as the titrant. These and other results, in the literature, are explained by considering the thermodynamic and kinetic aspects involved in the titrations of polyfunctional acids capable of forming intramolecular hydrogen bonds. In this context, the influence of the nature of the titrant and concentrations of both titrant and sample on the extension and rate of neutralization reactions are discussed.The mechanism employed to relate the rate of neutralization reaction of polyfunctional acids, capable of forming intramolecular hydrogen bonds, with the concentration of sample and titrant is based on recent investigations by Hibbert and Spiers.' These workers investi- gated the kinetics of the removal of protons from substituted salicylates, by hydroxyl ions and buffers. Finally, the compari- son of the rates of neutralization and catalysed reactions is employed to discuss the stoichiometries obtained at the end-poin t. Experimental Reagents Salicylic acid and potassium hydroxide were of analytical- reagent grade. Acetone and propan-2-01 , of laboratory- reagent grade, were dried with 3 A molecular sieves before use.Solutions of 0.5 and 1.0 mol dm-3 potassium and tetra- methylammonium hydroxide in propan-2-01 were prepared and standardized with benzoic acid in ethanol using phenol- phthalein as indicator. Other solutions were prepared by appropriate dilution of the 0.5 and 1.0 mol dm-3 solutions in propan-2-01. Apparatus A motor driven micrometer syringe, as described by Green- how and Spencer,g was employed to introduce the titrant at a constant delivery rate in both the potentiometric and thermo- metric titrations. In the thermometric titrations the tempera- ture changes were detected by means of a thermistor placed in one arm of a Wheatstone bridge, and were recorded on a strip-chart recorder as described elsewhere.9 A Micronal B 375 pH meter, a platinum electrode and a nichrome wire inserted into the titrant solution were employed in the potentiometric titrations.Procedure In the thermometric titrations, the desired volume of salicylic acid solution was pipetted into a 25 ml unsilvered Dewar flask and diluted with acetone to 10 ml. In both the potentiometric and thermometric titrations the titrant was added at a constant delivery rate of 0.13 ml min-1. During the titrations the solutions were stirred with a magnetic stirrer. The amount of salicylic acid was adjusted in order that the volume of titrant delivered at the end-point of titration was between 0.25 and 0.5 ml. The concentration of propan-2-01 was kept between 2.5 and 5.0%. Results and Discussion The results of the titrations of salicylic acid in acetone with potassium and tetramethylammonium hydroxides in propan- 2-01 by catalytic thermometric titrimetry are presented in Fig.1. For potassium hydroxide, it was observed that the stoichiometry attained at the end-point increases from -1 to948 =2 when the concentration of the titrant is increased from 0.01 to 0.5 rnol dm-3. It was also observed that when the titrant is tetramethylammonium hydroxide, only one group is titrated and the stoichiometry is independent of the concentration of the reactants. The results obtained by thermometric titrimetry with tetramethylammonium hydroxide as titrant are explained by considering the small second dissociation constant of salicylic acid. Consequently, the concentration of hydroxyl ions necessary to start the catalysed reaction is attained before the second equivalence point.In fact, as can be seen in Fig. 2, the dissociation constant is not sufficiently large to produce the second inflection in the potentiometric titration curve of salicylic acid with 0.1 and 0.8 rnol dm-3 tetramethylam- monium hydroxide solutions. However, the inflection corre- sponding to the titration of the phenolic group (absent when 0.1 rnol dm-3 potassium hydroxide is used) is observed when 1.0 rnol dm-3 potassium hydroxide is used as the titrant. Another point to be considered is that when potassium hydroxide is used as the titrant at concentrations above 0.1 rnol dm-3, the formation of a white precipitate of potassium salicylate is observed. Therefore it is necessary to consider, in addition to the neutralization reaction, the precipitation of potassium salicylate.HSal- + OH- Sa12- + H20 (1) (2) Sa12- + 2K+ G K2Sal(s) The precipitation of potassium salicylate causes a displace- ment to the right of the equilibrium represented in eqn. (1). 2.0 2. E c .- L .- 5 1.5 v) C 0 (0 [r .- - m €lB Y I 1 I I 0 0.2 0.4 0.6 0.8 1.0 Tirant concentration/mol 1-1 Fig. 1 Effect of concentration on the titration of salicylic acid in acetone by catalytic thermometric titrimetry. Titrant: A, KOH in propan-2-01; and B, Me4NOH in propan-2-01. Volume of acetone, 10 ml. Concentration of salicylic acid = 1/40 of the concentration of the titrant ANALYST, SEPTEMBER 1991, VOL. 116 Therefore, if potassium hydroxide is used in place of tetramethylammonium hydroxide, the reaction in eqn.(1) will occur, to a large extent, before the concentration of hydroxyl ions reaches the value necessary to start the catalysed reaction. Furthermore, in catalytic titrimetry it is necessary to consider the competition between the determinative and the indicative reactions.5.6 These implications are particularly important in the titration of hydrogen bonded acids, as with the second acidic group of salicylic acid, where the rates of the neutralization reaction are not diffusion controlled as for normal acids. 1 0 According to Hibbert and Spiers,7 the removal of the second proton from substituted salicylates involves two steps; breaking of the hydrogen bond and proton transfer. For salicylic acid, these two steps are represented by the following equations: (3) These workers have shown in their studies using dimethyl sulphoxide-water mixtures that, for concentrations of hydroxyl ions above 0.02 rnol dm-3, the breaking of the hydrogen bond is the rate-limiting step and that below this value the proton transfer is the rate-limiting step. In the range of concentration of hydroxyl ions below 0.02 rnol dm-3, the rate of the reaction increases linearly with the increase of concentration of hydroxyl ions.Above this value the rate of reaction varies little and tends to be independent of the concentration of hydroxyl ions. In the application of this mechanism to these results, it is assumed that when using potassium hydroxide as the titrant the concentration of hydroxyl ions is in the concentration range where proton transfer is the rate-limiting step.This assumption is reasonable as, in this medium, potassium hydroxide is present mainly as ion pairs.5-6 Consequently, the rate of the neutralization reaction increases to a larger extent than that of the catalysed reaction with the increase of t P 0 r P 7 D - / tt , Titrant volume (1 division = 0.6 ml) Fig. 2 Potentiometric and thermometric titration curves of salicylic acid in acetone. Concentration of salicylic acid = 1/40 of the concentration of the titrant. Volume of acetone, 10 ml. (a) KOH 0.1 rnol dm-3 in propan-2-01; (6) Me4NOH 0.1 mol dm-3 in propan-2-01; (c) KOH 1.0 rnol dm-3 in propan-2-01; and ( d ) Me4NOH 0.8 mol dm-3 in propan-2-01. P, Potentiometric titration curve; D, first derivative of potentiometric titration curve; and T, thermometric titration curve.The arrows indicate the start of the titrationsANALYST, SEPTEMBER 1991. VOL. 116 949 concentration of potassium hydroxide. This mechanism is consistent with the fact that the stoichiometry at the end-point increases with an increase in the concentration of potassium hydroxide and salicylic acid. The fact that the phenolic group of salicylic acid is not titrated at the end-point with tetra- methylammonium hydroxide has already been explained by considering the value of the dissociation constant of the acid. However, Greenhow and Shafis have compared the titra- tion of a variety of polyfunctional acids with tetrabutylam- monium and potassium hydroxide by employing acrylonitrile as the end-point indicator.Among the acids investigated, the benzenecarboxylic acids are of particular interest to the kinetic considerations studied in this paper. They have shown that, while all of the acidic groups are titrated with 0.5 mol dm-3 potassium hydroxide as the titrant, only some of these groups are titrated with 0.1 mol dm-3 tetrabutylammo- nium hydroxide. This behaviour is observed in benzenecar- boxylic acids, with the exception of terephthalic acid, capable of forming intramolecular hydrogen bonds. These differences in stoichiometry obtained with the two different titrants are not observed when using benzenecarboxylic acids, which do not form intramolecular hydrogen bonds. It is not possible to explain the low stoichiometry obtained with tetrabutylammonium hydroxide on the basis of equilib- rium considerations.In fact, the pK, values for these acids in water vary from 1.40 to 6.96. As tetrabutylammonium hydroxide is completely dissociated ,536 it is supposed that the concentration of hydroxyl ions is in the concentration range where the breaking of the hydrogen bond is the rate-limiting step. Therefore, when the rate of production of the ‘opened’ form of the acid [eqn. (3)], capable of reacting with hydroxyl ions, is not sufficiently high, an excess of hydroxyl ions sufficient to start the catalysed reaction is available. When potassium hydroxide is used as the titrant, ion pairs are mainly present and the concentration of hydroxyl ions is low. Therefore, it is reasonable to suppose that the system is in the concentration range where proton transfer is the rate-limiting step.Consequently, the rate of consumption of hydroxyl ions is sufficient to prevent the start of the catalysed reaction. Finally, in order to acertain if a given acidic group of a polyfunctional acid capable of forming hydrogen bonds can be titrated at the end-point, it is necessary to consider both the thermodynamic and the kinetic implications involved in the titration. Firstly, it is necessary to consider the value of the dissociation constant of the acidic groups and also the possibility of displacement of the equilibrium by virtue of the precipitation of the anion formed. Next it is necessary to take into account that the removal of protons from such acidic groups by hydroxyl ions is not diffusion controlled as for normal acids. Either the breaking of the hydrogen bond or proton transfer is the rate-limiting step, depending on the conditions under which the titration is performed. By the appropriate choice of the conditions and by considering the competition between the neutralization and catalysed re- actions, it is possible to obtain selectivity in the titrations. The authors thank the Funda@o de Amparo a Pesquisa do Estado de Sgo Paulo for financial support and Dr. F. Y. Fujiwara for reviewing the English text. References 1 Vaughan, G. A., and Swithenbank, J. J . , Analyst, 1965,90,594. 2 Greenhow, E. J., and Spencer, L. E., Analysf, 1973, 98, 90. 3 Greenhow, E. J., and Hargitt, R., Proc. SOC. Anal. Chem., 1973, 10, 276. 4 Greenhow. E. J.. and Shafi. A. A., Proc. Anal. Div. Chem. SOC., 1975, 12, 286. 5 Greenhow, E. J., and Shafi, A. A., Analyst, 1976, 101, 421. 6 Greenhow, E. J., Chem. Rev., 1977, 77,835. 7 Hibbert, F.. and Spiers, K. J . , J. Chem. SOC., Perkin Trans. 2, 1989, 2, 67. 8 Greenhow, E. J . , and Spencer, L. E., Analyst, 1973, 98, 98. 9 Chagas, A. P., Godinho, 0. E. S., and Costa, J . L. M., Talanta, 1977. 24, 593. 10 Eigen, M., Angew. Chem., Int. Ed. Engl.. 1964.3, 1. Paper 1 /00038A Received January 3rd, 1991 Accepted March 19th, 1991

ISSN:0003-2654    

DOI:10.1039/AN9911600947

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

ANALYST, SEPTEMBER 1991, VOL. 116 95 1 lodimetric Method for the Determination of Mono- and Disaccharides With Vanadium(v) in Perchloric Acid Amalendu Banerjee, Banasri Hazra, Anuva Putatunda, Dinabandhu Mandal, Gopal Chandra Banerjee and Sachchidananda Dutt Department of Chemistry, Jadavpur University, Calcutta-700 032, India Sodium (or ammonium) metavanadate in dilute perchloric acid was found t o be a suitable reagent for the determination of fructose, mannose, galactose, arabinose, ribose, xylose, sucrose and maltose. A mixture containing the saccharide and an excess of the reagent was refluxed over a small flame for 40 min. The residual metavanadate was determined iodimetrically. The amount of saccharide was then determined by standardizing the metavanadate solution against either sodium thiosulphate solution or the respective saccharide solution of known strength; the latter procedure gave consistently better results. Keywords: Mono- and disaccharide determination; metavanadate-dilute perchloric acid; iodimetric method The development of a simple and precise method' for the determination of glucose led us to investigate its application to the determination of mono- and disaccharides.In addition to the methods cited previously,' both spectrophotometric24 and spectrofluorimetrics methods for the determination of various sugars have been reported. The determination of glucose and other sugars by different colorimetric methods has been discussed by several workers.610 Chromatographic"-13 and polarographic14~15 methods for the determination of sugars have also been reported.Bark et ~ 1 . 1 6 developed a thermometric method for this purpose. An automated assay for the direct determination of D-fructose in a mixture with D-glucose has been developed.17 The simple and rapid determination of ketoses by circular dichroism measurements has recently been reported.18 Experimental The saccharides (Merck) were stored over concentrated sulphuric acid in a desiccator for at least 1 week before use. It is known that fructose is hygroscopic and becomes moist on prolonged exposure to the atmosphere. Hence it was dried at 50 "C under vacuum for 16 h and then analysed. (Found: C, 40.37; H, 6.70. Calc. for C6H1206: C, 40.00; H, 6.67%). Sodium (or ammonium) metavanadate, perchloric acid (70%), potassium iodide, potassium dichromate (all of guaranteed-reagent grade, Merck), sodium thiosulphate (Merck) and hydrochloric acid (analytical-reagent grade, Biosol) were used. A freshly prepared starch solution was used as the indicator.Doubly distilled water (prepared by refluxing over KMn04) was used throughout. Preparation and Standardization of Sodium Thiosulphate Solution Sodium thiosulphate pentahydrate (25 g) was dissolved in water (400 ml) and allowed to stand at room temperature for 48 h with occasional stirring, during which time a small amount of a colourless precipitate was formed. The solution was filtered through cotton wool and the clear filtrate diluted to 1000 ml. The sodium thiosulphate solution thus obtained was standardized iodimetrically using standard potassium dichromate solution.19 This solution can be diluted further whenever required. Preparation and Iodimetric Standardization of the Vanadium(v) Solution The clear yellow solution obtained by dissolving sodium (or ammonium) metavanadate (0.05 mol) in hot water (200 ml) was allowed to cool to room temperature. It was poured slowly into a cold solution of perchloric acid (70% ; 170 ml) in water (200 ml) with stirring, whereupon a transparent yellow solution of the metavanadate (approximately 0.1 mol 1-1, caused by the reduction of VV to VIV) was obtained. This solution was standardized iodimetrically by the method described previously. 1 Standardization of the Metavanadate Solution With Standard Fructose Solution: Determination of Fructose in Solutions of Unknown Strength A solution of fructose (0.9244 g dl-1) was prepared.A mixture of this fructose solution (2 ml) and the standardized metavana- date solution (30 ml, added from a burette) was placed in a standard-jointed round-bottomed flask (100 ml) fitted with a condenser (8 in) and was heated under reflux over a small flame for 40 min. The resulting green-blue solution was cooled to room temperature. The condenser was washed with water (10 ml) and the washings were added to the reaction mixture. The residual metavanadate was determined iodi- metrically. * The end-point was indicated by a sharp and stable colour change from purple to deep blue. Blank experiments were carried out in the above procedure, which was adopted for the determination of fructose in solutions of unknown strength (Table 1).The titre values remained unchanged on refluxing the reaction mixture for different periods of time ranging from 30 Table 1 Determination of fructose in solutions of unknown strength with a vanadium(v) solution, which was standardized against a fructose solution of known strength Solution No. 1 2 3 4 5 6 7 8 9 10 11 12 Fructose concentratiodg dl-1 Actual Found* 0.1084 0.1072 k 0.0015 0.3299 0.3321 k 0.0002 0.4561 0.4575 k 0.0026 0.6867 0.6822 k 0.0025 0.7709 0.7743 k 0.0020 0.9048 0.9051 k 0.0065 0.9770 0.9765 k 0.0016 1.3976 1.3913 k 0.0015 2.8956 2.9148 k 0.0028 5.5740 5.5037 k 0.0035 6.9672 7.0453 k 0.0030 10.9192 10.6147 -L 0.0053 * Mean of six determinations * standard deviation.952 ANALYST, SEPTEMBER 1991, VOL.116 to 120 min. This was found to be true for all of the sugars studied. Therefore, refluxing for 40 min is recommended. Determination of Fructose in a Solution With Vanadium(v) Standardized Against a Glucose Solution of Known Strength The solutions of glucose (1.0564 g dl- 1 ) and fructose (1.2668 g dl-1) were taken as the known and the unknown solutions, respectively. The metavanadate solution was standardized against the glucose solution (2 ml for each experiment) and then fructose (2 ml for each experiment) was determined using the standardized metavanadate solution as described above (Table 2). Determination of Fructose Taken in the Solid State Accurately weighed samples of fructose (30-150 mg) were placed in a standard-jointed round-bottomed flask (100-250 ml) and treated with a known excess of the metavanadate solution (40-160 ml depending on the amount of fructose taken).Each reaction mixture was then refluxed for 40 min and the excess of metavanadate was determined in the usual way (Table 3). Calculations Equal volumes of known and unknown sugar solutions were taken for each determination. If W is the concentration of sugar (g dl-1) present in a solution of known strength, z is the titre value (ml) of thiosulphate solution required for a blank titration of the vanadium(v) solution taken, and x and y are the titre values (ml) of the same thiosulphate solution required Table 2 Determination of fructose in solutions of unknown strength with a vanadium(v) solution, which was standardized against a glucose solution of known strength Fructose concentratiodg dl- Solution No.Actual Found* 1 1.2668 1.2748 2 0.9048 0.9054 3 0.7709 0.7742 * Mean of six determinations. Table 3 Determination of fructose, by directly weighing solid fructose into the reaction flask, with a vanadium(v) solution, which was standardized against either a fructose or glucose solution Experiment No. 1 2 3 4 5 6 Amount of fructose/g Actual Found 0.0365 0.0361 0.0510 0.0506 0.0680 0.0677 0.0877 0.0874 0.0911 0.0917 0.1218 0.1222 Table 4 Determination of mannose in solutions of unknown strength with a vanadium(v) solution, which was standardized against a mannose solution of known strength Mannose concentratiodg dl- Solution No. Actual Found* 1 1.0310 1.0260 2 0.3093 0.3110 3 0.5155 0.5124 4 0.7217 0.7158 for the determination of the residual vanadium(v) after the oxidation of sugar solutions of known and unknown strengths, respectively, then the concentration of sugar S (g dl-1) present in the solution of unknown strength is given by: W ( z - Y ) ( z - x) S= Determination of Other Monosaccharides A number of other monosaccharides, viz., mannose, galac- tose, arabinose, ribose and xylose, were determined using the procedure described above for the determination of fructose.In each instance the respective monosaccharide solution was standardized against the same monosaccharide solution of known strength, and also against a standard glucose solution. The results are presented in Tables 4-14. Determination of Disaccharides Sucrose and maltose were also determined by the procedure described above.In this instance the respective disaccharides were standardized against the same sugar and also against each other. The results are presented in Tables 15-18. Table 5 Determination of mannose in solutions of unknown strength with a vanadiurn(v) solution, which was standardized against a glucose solution of known strength Glucose Mannose concentratiodg dl-1 Solution concentration/ Found* 1 1.0962 1.0310 1.0257 2 1.0962 0.7418 0.7437 3 1.0962 0.3093 0.3115 No. gdl-l Actual * Mean of six determinations. Table 6 Determination of galactose in solutions of unknown strength with a vanadium(v) solution, which was standardized against a galactose solution of known strength Galactose concentratiodg dl- 1 Solution No. Actual Found* 1 0.1600 0.1614 2 0.3440 0.3431 3 0.6032 0.6022 * Mean of six determinations.Table 7 Determination of galactose in solutions of unknown strength with a vanadium(v) solution, which was standardized against a glucose solution of known strength Glucose Galactose concentratiodg dl-1 Solution concentration/ No. g dl-1 Actual Found* 1 0.2587 0.8051 0.8030 2 0.2587 0.6032 0.6050 3 0.2587 0.1600 0.1620 *Mean of six determinations. Table 8 Determination of arabinose in solutions of unknown strength with a vanadium(v) solution, which was standardized against an arabinose solution of known strength Arabinose concentratiodg dl-* Solution No. Actual Found* 1 0.2724 0.2787 2 0.4086 0.4086 0.5404 3 0.5448 * Mean of six determinations. * Mean of six determinations.ANALYST. SEPTEMBER 1991, VOL.116 953 Table 9 Determination of arabinose in solutions of unknown strength with a vanadium(v) solution, which was standardized against a glucose solution of known strength Glucose Arabinose concentration/g dl-l Solution concentration/ No. g dl-1 Actual Found* 1 1.0962 0.6811 0.6813 2 1.0962 0.9572 0.9580 3 1.0962 1.0852 1.0876 *Mean of six determinations. Table 10 Determination of ribose in solutions of unknown strength with a vanadium(v) solution, which was standardized against a ribose solution of known strength Ribose concentratiodg dl-l Solution No. Actual Found* 1 0.9335 0.9325 2 0.8090 0.8069 3 0.5601 0.5617 4 0.3734 0.3790 * Mean of six determinations. Table 11 Determination of ribose in solutions of unknown strength with a vanadium(v) solution, which was standardized against a glucose solution of known strength Glucose Ribose concentratiodg dl-1 Solution concentration/ No.g dl-1 Actual Found" 1 0.9587 0.9335 0.9276 2 0.9587 0.8090 0.8122 3 0.9587 0.6265 0,6295 * Mean of six determinations. Table 12 Determination of ribose, by directly weighing solid ribose into the reaction flask, with a vanadium(v) solution, which was standardized against either a ribose or glucose solution Amount of ribose/g No. Actual Found 1 0.0672 0.0667 2 0.0759 0.0751 3 0.0882 0.0887 4 1.2628 1.2648 Experiment Table 13 Determination of xylose in solutions of unknown strength with a vanadium(v) solution, which was standardized against a xylose solution of known strength Xylose concentration/g dl-1 Solution No. Actual Found* 1 0.9950 0.9934 2 0.8182 0.8192 3 0.7620 0.7649 4 0.3640 0.3627 * Mean of six determinations.Table 14 Determination of xylose in solutions of unknown strength with a vanadium(v) solution, which was standardized against a glucose solution of known strength Glucose Xylose concentratiodg dl-' Solution concentration/ Found* 1 0.2587 0.8182 0.8195 2 0.2587 0.4142 0.4152 3 0.2587 0.3640 0.3680 No. g dl-1 Actual *Mean of six determinations. Table 15 Determination of sucrose in solutions of unknown strength with a vanadium(v) solution, which was standardized against a sucrose solution of known strength Sucrose concentratiodg dl- Solution No. Actual Found" 0.1212 0.3049 0.4279 0.5000 0.7660 0.9080 1.5820 3.2418 5.4720 0.1218 k 0.0004 0.3059 f 0.0038 0.4258 f 0.0047 0.5013 f 0.0067 0.7704 f 0.0086 0.9057 f 0.0054 1.5714 f 0.0045 3.2615 k 0.0091 5.5134 k 0.0048 * Mean of six determinations f standard deviation.Table 16 Determination of sucrose, by directly weighing solid sucrose into the reaction flask, with a vanadium(v) solution, which was standardized against a sucrose solution of known strength Amount of sucrose/g Experiment No. 1 2 3 4 5 6 7 8 9 Actual 0.0520 0.0616 0.0671 0.0941 0.0996 0.1035 0.1409 0.1577 0.1680 Found 0.0524 0.0615 0.0671 0.0930 0.1004 0.1020 0.1402 0.1594 0.1690 Table 17 Determination of maltose in solutions of unknown strength with a vanadium(v) solution, which was standardized against a maltose solution of known strength Maltose concentratiordg dl-1 Solution No. Actual Found* 1 4.6468 4.6906 2 0.9777 0.9635 3 0.7708 0.7771 4 0.5950 0.5979 0.3891 5 0.3854 * Mean of six determinations.Interference Studies It was observed' that toluene, acenaphthene, styrene, phenol, 2-naphthol, benzaldehyde, cyclohexanol, nitrobenzene, ani- line, mandelic acid, salicylic acid, formic acid, N,N-dimethyl- formamide and chloroform were serious interferents, whereas benzene, naphthalene, ethanol, tert-butanol, benzophenone, glycine and dimethyl sulphoxide interfered to a lesser extent. Acetone and acetic acid did not interfere. Depending on the properties of the interferents, some of them can be removed by appropriate techniques such as evaporation, steam distilla- tion and solvent extraction. Discussion As reported previously,l doubly distilled water must be used throughout in order to avoid inconsistent results.The approxi- mately 0.1 mol 1-1 metavanadate solution has to be prepared very carefully. Solutions of sodium (or ammonium) metavana- date (0.05 mol) in hot water (200 ml) and perchloric acid (70% ; 170 ml) in water (200 ml) must be cooled separately to room temperature and the solutions mixed very slowly with constant stirring. Rapid addition of the metavanadate solution to the perchloric acid leads to a rise in temperature, and the formation of a red precipitate, which renders the solution954 ANALYST, SEPTEMBER 1991, VOL. 116 Table 18 Determination of maltose in solutions of unknown strength with a vanadium(v) solution, which was standardized against a sucrose solution of known strength Sucrose Maltose concentratiodg dl- 1 Solution concentration/ No.g dl-1 Actual Found* 1 1.2960 1.2765 1.2700 2 1.2960 0.9635 0.9530 3 1.2960 0.3854 0.3882 * Mean of six determinations. unsuitable for use. It should be noted that the strength of the reagent is critical, as a more concentrated metavanadate solution invariably caused the red precipitate to appear during refluxing, whereas with very dilute metavanadate solutions the consumption of the reagent by the sugar was incomplete even on prolonged refluxing . The recommended procedure is to reflux a known volume of a saccharide solution for 40 min with a known volume of the oxidant solution composed of metavanadate (1 g) and a mixture of perchloric acid (70% ; 30 ml) and water (72 ml). Refluxing for less than 30 rnin resulted in the incomplete oxidation of all the mono- and disac- charides; beyond 30 rnin no change of titre values was noted on prolonging the time of reflux up to 120 min.Perchloric acid having this specified dilution does not liberate iodine from potassium iodide under the experimental conditions. The vanadium(v) solution was standardized1 by titration against a thiosulphate solution of known strength.19 Hence, a known volume of the metavanadate solution (taken from a burette) was heated under reflux for 40 min, cooled and then determined iodimetrically. Identical results were obtained when this titration was conducted at room temperature, thereby confirming that the metavanadate solution is not affected by refluxing; it also remained unchanged on storage, even after 6 months at room temperature.The volume of water required for the washings in each titration was restricted to 10 ml in order to ensure that a stable end-point was obtained. Moreover, in order to ensure that the reaction conditions are identical in each instance, the standardization of vanadium(v) through refluxing is recommended. The lower and upper limits of the accurate determination of fructose were established by using the metavanadate solution standardized against a fructose solution of known strength (approximately 1%). The results are given in Table 1, which shows that accurate results were obtained when the concentra- tion of the fructose solution ranged from 0.3 to 3.0 g dl-1; however, with either very dilute or very concentrated solu- tions the results were inaccurate. Similarly, the metavanadate solution standardized against a sucrose solution of known strength (approximately 1%) was used to ascertain the upper and lower limits of the accuracy of the determination of sucrose in solution (Table 15).The results show that the accuracy of the method is good; however, the precision is moderate for the determination of sucrose. It should be noted that accurately weighed samples of mono- and disaccharides, placed directly in the reaction flask, also afforded good results (Tables 3, 12 and 16). Moreover, it was observed that metavanadate solutions standardized against a monosac- charide did not afford good results for the determination of the disaccharides, and vice verso. Further studies in this area, particularly on the determination of sugar alcohols, saccharic acids and other classes of sugars, and also on product analysis, are currently in progress.Conclusion A method for the determination of mono- and disaccharides has been developed, which is less hazardous and tedious than Fehling’s method (see Table 19), although the latter is suitable Table 19 Comparison of the standard Fehling’s method with the proposed metavanadate method Fehling’s method Metavanadate method 1 Sugar determined in solution 2 Titration carried out under boiling conditions; hence, difficult and hazardous 3 The end-point is unstable and difficult to ascertain even with Methylene Blue as indicator, particularly under conditions of insufficient illumination Sugar determined in solution and also in the solid state Iodimetric titration carried out at room temperature; hence the method is much easier The end-point is stable, sensitive and very sharp (purple to deep blue) and can be ascertained without any difficulty 4 Comparatively large amounts Relatively small amounts (decigrams) of sugars are required determined 5 Not very accurate; error is occasionally within 55% within k 1% 6 Time required for each Refluxing followed by titrimetric titration is 10-15 min analysis requires 40 min 7 Only one experiment can be Simultaneous determinations carried out at a time can be carried out 8 Applicable to reducing sugars Applicable to all types of sugars, only; hence, mixtures including sugar alcohols, containing mono- and di- saccharic acids and related saccharides can be compounds.Mixtures con- determined taining mono- and di- saccharides cannot be determined (milligrams) of sugar can be Very accurate; error is generally 9 Less susceptible to inter- Various organic and inorganic ference from different materials interfere classes of organic com- pounds for the determination of mixtures of reducing monosac- charides (e.g., glucose) and non-reducing disaccharides (e.g., sucrose).Glucose or sucrose solutions of known strengths can be employed for the standardization of the metavanadate solution for the determination of mono- and disaccharides, respectively. Better results are obtained if a metavanadate solution standardized against a saccharide solution of known strength is employed for the determination of the same saccharide. The advantages of the proposed method are: (i) inexpensive and non-toxic chemicals are used; (ii) simul- taneous determinations can be carried out in any laboratory; and (iii) the end-point is stable and sharp.The authors thank S. P. Bag, Head of the Department of Chemistry, Jadavpur University, and G . Raymahasay, M / S . Life Pharmaceuticals, Calcutta, for generous support of this work. A. P. thanks B. Bose, Head of the Department of Chemistry, Lady Brabourne College, for helpful suggestions and B. H. thanks the University Grants Commission, New Delhi, for a Research sponsorship. References Banerjee, A., Hazra, B., Chatterjee, H . , Banerjee, S . , and Dutt, S., Analyst, 1989, 114, 1151. Zipf, R. E., and Waldo, A. L., J. Lab. Clin. Med., 1952, 39, 497. Ando, M., and Kiuchi, R., Nippon Nogei Kagaku Kaishi, 1964, 38, 428; Chem.Abstr., 1965, 63, 4501b. Rogers, C. J., Chambers, C. W., and Clarke, N. A., Anal. Chem., 1966,38, 1851. Lokar, L. C., Univ. Studi Trieste, Fac. Econ. Commer., Ist. Merceol., (Pubbl.), 1965, 25, 24; Chem. Abstr., 1967, 66, 34 656w.ANALYST, SEPTEMBER 1991, VOL. 116 955 6 7 8 Nanba, A., and Matsuo, Y., Hakko Kyokaishi, 1963, 21, 105; Chem. Abstr., 1965, 62, 15 432c. Avigad, G., Carbohydr. Res., 1968, 7, 94. Kachalova, M. F., Kulikov, Yu. M., Kozlov, V. V., and Khrustaleva. V. N., Izv. Vyssh. Uchebn. Zaved., Pishch. Tekhnol., 1971, 6, 154; Chem. Abstr., 1972,76, 142 114x. Namigohar, F., Makhani. M., and Kamal, F., Q. J. Crude Drug Res., 1981, 19, 169; Chem. Abstr., 1982, 96, 129 873y. Tsutsui, K., Tanaka, T., and Oda, T., Anal. Biochem., 1977, 79, 349. Barbiroli, G . , Rass. Chim., 1965, 17, 62; Chem. Abstr., 1965, 63, 11 843d. Bracher. C., and Bauly, L. E., Food Manuf.. 1965, 40, 38; Chem. Abstr., 1965, 63, 13 549f. Sweely, C. C., Wells, W. W., and Bentley, R., Methods Enzymol.. 1966, 8, 95; Chem. Abstr., 1968, 69, 16 709y. 14 Nanba, A., and Matsuo, Y., Hakko Kogaku Zasshi, 1964,42, 216; Chem. Abstr., 1966,65, 36e. 15 Corlett, R. D., Breck, W. C., and Hay, G. W., Can. J. Chem., 1970, 48, 2474. 16 Bark, L. S., Edwards, D., and Prachuabpaibul, P., Proc. SOC. Anal. Chem., 1974, 11, 170. 17 Kennedy, J. F., and Chaplin, M. F., Carbohydr. Res.. 1975,40, 227. 18 Kimura, A., Chiba, S., and Yoneyama, M., Carbohydr. Res., 1988, 175, 17. 19 Vogel, A. I., A Text-Book of Quantitative Inorganic Analysis, Longman, London, 4th edn., 1978, p. 376. Paper 0/05627H Received December 14th, 1990 Accepted May 9th, 1991

ISSN:0003-2654    

DOI:10.1039/AN9911600951

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

ANALYST, SEPTEMBER 1991, VOL. 116 957 Application of a Microwave Oven for Drying and Nitric Acid Extraction of Mercury and Selenium From Fish Tissue Suei Y. LamLeung, Vincent K. W. Cheng and Yuet W. Lam Department of Chemistry, Hong Kong Baptist College, 224 Waterloo Road, Kowloon, Hong Kong The drying and nitric acid digestion of fish tissue using a microwave oven for the determination of Hg and Se were studied. For the drying process, the method was compared with the freeze-drying method. The water content found in the muscle of tilapia, Oreochromis mossanicus, by using a freeze dryer at -40 "C and a pressure of ~ 1 0 pmHg for 24 h, and a microwave oven (convection mode) at 70 "C for 3 h was 78.16 and 78.09% m/m, respectively. The Hg and Se levels found in fish tissues dried by the two methods were consistent with a 95% confidence level.The efficiency of dissolution of Hg and Se from dried muscle of tilapia, using concentrated HN03 in the closed vessel microwave digestion method was found to be much greater than that of traditional open vessel digestion. The microwave heating digestion method was tested satisfactorily using two certified reference materials. The recovery of added standards of Hg and Se was found to be 84.8 and 96.6%, respectively. Keywords: Microwave heating and drying; fish tissue; nitric acid ashing; selenium determination; mercury determination The process of drying and dissolution of biological materials is a lengthy task. The application of a microwave oven for wet ashing in a closed vessel containing geological and biological materials has been found to be an efficient method for the preparation of sample solutions for metal determination.1-3 A study of the application of a commercial microwave oven in drying and wet ashing of fish tissue for the determination of Hg and Se is reported here. Only HN03 was used in this ashing process, as it is a strong oxidizing agent and also the nitrates of most metals are soluble in water. Low microwave power was used for safety reasons. A sample of muscles of small individuals of tilapia of length <19 cm was used for the investigation of the microwave and freeze-drying processes. The muscles of large individuals of tilapia of length 26-31 cm were used for the study of the wet ashing processes in a closed vessel by microwave heating and in an open vessel by thermal heating.Recovery of added standards using the microwave method of digestion in a closed vessel was also studied. The microwave digestion method was tested by its application to two certified reference materials. Experimental Apparatus A commercial microwave oven (Sharp, Model R-9H10) with a frequency of 2450 MHz (power output of 750 W for microwave mode and 1500 W for convection heating) and a turntable was used for digestion and drying without modifica- tion. Sample digestion was carried out in capped 60 ml polytetrafluoroethylene (PTFE) vessels (Savillex, Model 561R2). A block digestor (Techne, Model DG-1) with a temperature controller (Techne, Model TC-D1 A) was used for open vessel digestion of samples. For the wet ashing process, a wide-mouthed earthenware vessel with a lid was used as it protects the microwave oven from the corrosive vapour released or from any explosion of the digestion vessel.All containers were treated successively with detergent (washing), tap water (rinsings), 70% m/m HN03 for glassware (rinsing) or 0.5% m/v HN03 for plastics (soaked for at least 24 h). An atomic absorption spectrometer (Varian, Model SpectrAA-10) equipped with a hydride vapour generation accessory (Perkin-Elmer, Model MHS- 10) was used for the determination of Hg and Se. A freeze dryer (Labconco 75034) was used for drying the fresh muscle of tilapia collected from a local river. Reagents All chemicals used were of analytical-reagent grade. A stock solution of 1000 pg ml-1 of Hg was prepared by dissolving 1.080 g of HgO in the minimum volume of HC1 (1 + l ) , and diluting to 1 1 with 1.5% m/v HCI.A stock solution of 1000 pg ml-1 of Se was prepared by dissolving 1.000 g of Se in approximately 5 ml of 70% m/m HN03, and diluting to 1 1 with Working standard solutions were prepared immediately before use by serial dilution of the stock solutions. The working ranges were as follows: Hg, 2-16 ppb for sample analysis and 50-200 ppb for recovery investigation; and Se, 2-12 ppb for sample analysis and 50-120 ppb for recovery investigation. The reducing solution was prepared fresh daily by dissolving 6 g of NaBH4 and 2 g of NaOH in 200 ml of distilled water. 5% m/v H2S04. Samples The axial muscle of tilapia, Oreochromis mossanicus (Peters), collected from a local river, and National Research Council Canada (NRCC) certified reference materials (DORM-1 , Dogfish Muscle and DOLT-1, Dogfish Liver) were studied.Analytical Procedure During wet digestion in a closed vessel by microwave heating the following precautions must be considered: the mass of sample used for digestion should not be greater than 0.2 g; the total volume of reactants should not exceed one tenth of the volume of the container;4 the heating times at high power should not be longer than 5 min; the digestion vessel must be cooled before opening; and a digestion vessel with a pressure releasing cap should be used. Drying of sample A collection of axial muscles of small individuals (<19 cm in length) of tilapia caught from a local river was homogenized in a stainless-steel blender. About half of the sample, 50 g, was spread thinly on several clean watch-glasses and placed in a microwave oven.The samples were dried (convection mode) at 70 "C for 2-3 h until constant mass was obtained. The958 ANALYST, SEPTEMBER 1991, VOL. 116 samples were removed from the oven and turned over every 20 min to ensure thorough drying. The remaining portion of the sample was dried in a freeze dryer at -40 "C and <lo pmHg for 24 h. The axial muscles of large individuals (2631 cm) of tilapia were treated in the same manner using a microwave oven. Dissolution of sample The microwave digestion of tilapia tissues or certified refer- ence material was carried out using four replicates of about 0.2 g of the dried muscle weighed into a dried, cleaned PTFE digestion vessel. A 4 ml aliquot of 70% m/m HN03 was added to each digestion vessel.Four tightly capped digestion vessels and one small beaker filled with 30 ml of water were placed in a wide-mouthed earthenware vessel. The lidded vessel was placed in the microwave oven. The samples were treated using three heating stages: low power (10%) for 8 min, medium-low power (30%) for 8 min and medium power (50%) for 4 min.5 The earthenware vessel was removed from the microwave oven and then uncovered and cooled to room temperature. The cap of the digestion vessel was opened slightly to release any pressure before being removed. The digest was quantitat- ively transferred into a 100 ml beaker and evaporated to <1 ml by heating to decompose any unused HN03.The solution was filtered through Whatman No. 42 filter-paper and quantita- tively transferred into a 25 ml calibrated flask. It was diluted to volume with 0.1% m/v HN03. The solution was then stored in a polyethylene bottle. For open vessel thermal digestion, four replicates of a sample (about 2 g) of dried muscle of tilapia were weighed into four 50 ml boiling tubes. The sample was pre-digested at 20 "C for 24 h with 10 ml of 70% m/m HN03. The sample was then heated at 100 "C for.3 h. Three additional portions of 5 ml of 70% m/m HN03 were introduced into the boiling tube successively at intervals of 30 min. When a clear solution was obtained, the temperature was raised to 130 "C to decompose all the unused HN03. After cooling, the digest was filtered through Whatman No.42 filter-paper then quantitatively transferred into a 25 ml calibrated flask and diluted to volume with 0.1% m/v HN03. The sample was stored in a poly- ethylene bottle for elemental determination. Recovery of added standard A known mass of the analyte, 100,200,300 and 400 ng of Hg and 100,140,200 and 240 ng of Se, together with 4 ml of 70% m/m HN03 were introduced into the Teflon digestion vessel. The digestion process in the microwave oven was the same as for the fish sample. Pa. The absorbance at 253.7 nm was measured for the determination of Hg and at 196.0 nm for Se.6 An acid blank solution was used in the determination of the analytes. Results and Discussion The data given in Table 1 indicate that the quality of the microwave-dried tissue was satisfactory.The colour of the microwave-dried tissue was similar to that of sun-dried tissue. The water content found in the species studied was 78.09% m/m (by microwave drying) and 78.16% (by freeze-drying). There was no significant difference (P > 0.05) in the values of Hg and Se found in specimens dried by the two methods. These results were consistent with results reported by K0h.7 The recovery of the added Hg and Se standards digested by microwave heating in a closed vessel was also found to be satisfactory (Table 2). The recovery of added standard was Table 1 Water content and concentration of Hg and Se in muscle of small species of tilapia dried by a freeze dryer and microwave oven Analyte Microwave oven Freeze dryer Water content (% d m ) 78.09 78.16 Hg/pg g- dry mass 3.777 f 0.198* 3.846 f 0.127* Setpg g- dry mass 2.224 k 0.021* 2.176 k 0.022* nations.* Mean values with standard deviations for four replicate determi- Table 2 Recovery of added standard after microwave heating under pressure in the presence of 70% m/m HN03 Element Addedhg 100 200 300 400 Se 100 140 200 240 Hg Found*/ng Recovery (YO) Mean (YO) 84.5 f. 0.0 84.5 t- 0.0 177.9 f. 0.9 89.0 f 0.5 252.8 k 1.4 84.3 k 0.5 325.0 f 2.4 81.3 f 0.6 84.8 f. 3.2 98.1 f 0.0 98.1 f 0.0 139.5 k 4.4 99.6 f 3.1 197.4 f 3.4 98.7 f 1.7 215.6 _+ 2.7 89.8 k 1.1 96.6 _+ 4.5 * Mean values with standard deviations for triplicate deterrnina- tions. Table 3 Comparison of the dissolution of dried muscle of large species of tilapia by using closed vessel microwave digestion and open vessel digestion in the presence of 70% m/m HN03, shown by recovery of added analytes Recovery/pg g-1 Mercury and selenium determination In each trial, 10 ml of the sample solution or working standard solution were used.A 1 ml volume of 30% HCI and 2 ml of reducing agent (a mixture of 3% m/v NaBH4 and 1% m/v NaOH) were added to the vapour generator. The solution was continuously purged with a stream of N2 at a pressure of 250 Microwave digestion* Open vessel 3.91 t- 0.29 1.24 k 0.02 1.71 f 0.08 1.25 f. 0.00 Element (closed vessel) digestion Hg Se * Mean values with standard deviations for four replicate determi- nations. Table4 Comparison of the found and certified values of Hg and Se in the NRCC certified reference materials DORM-1 and DOLT-1 (reference 9) Hg Found*/ Certified Sample g- value/pg g-1 DORM-1 Trial 1 0.759 k 0.030 Trial 2 0.754 t 0.019 Mean 0.757 f 0.025 0.798 f 0.074 DOLT-1 Trial 1 0.185 * 0.026 Trial 2 0.194 f.0.012 Mean 0.190 f 0.019 0.225 f 0.037 Se ~~~~~ Found?/ Certified Pg g- value/pg g- 1 1.44 f 0.03 1.43 k 0.01 1.44 f 0.02 6.64 t 0.16 6.60 f 0.15 6.62 f 0.16 1.62 k 0.12 7.34 f 0.42 * Mean value with standard deviation for four replicate determinations using cold vapour atomic absorption spectrometry. t Mean value with standard deviation for four replicate determinations using hydride generation atomic absorption spectrometry.ANALYST, SEPTEMBER 1991, VOL. 116 959 found to be 84.8 & 3.2 and 96.6 & 4.5% for Hg (100-400 ng) and Se (100-240 ng), respectively.The dissolution of the more volatile Hg and Se from fish tissue was superior to the traditional open vessel digestion (Table 3). The recovery of Hg and Se from samples of the large tilapia by open vessel digestion was only 35 and 68%, respectively, of the recovery by microwave heating under pressure. The results show that the closed digestion vessel can contain the samples tightly and the sample tissue is digested more completely at an elevated temperature and under pressure by microwave heating8 The microwave method was tested by studying two NRCC certified reference materials. A comparison of the values of Hg and Se found in DORM-1 (Dogfish Muscle) and DOLT-1 (Dogfish Liver) with the certified values is given in Table 4.9 The mean concentration of Hg found was within one standard deviation of the certified value in both the Dogfish Muscle and Dogfish Liver. For Se, the mean concentration was within two standard deviations of the certified value.The recovery of Hg from Dogfish Muscle and Dogfish Liver was 94.9 and 84.4%, respectively. For Se, the recovery was 88.9 and 90.2%, respectively. In conclusion, a method of microwave heating has been applied for the effective determination of the more volatile elements (Hg and Se) in biological samples. Microwave heating can be used for drying and wet ashing of fish tissue more efficiently than thermal heating in an open vessel in terms of a shorter time and smaller amounts of chemicals consumed. References Fischer, L. B . , Anal. Chem., 1986, 58, 261. Bettinelli, M., Baroni, U., and Pastorelli, N . , Anal. Chim. Acta, 1989, 225, 159. Aysola, P., Anderson, P., and Langford, C. H . , Anal. Chem., 1987, 59, 1582. Gedye, R., Smith, F., and Westaway, K., Educ. Chem., 1988, 25, 55. Vermeir, G., Vandecasteele, C., and Dams, R., Anal. Chirn. Acta, 1989, 220, 257. Tsalex, D. L., Atomic Absorption Spectrometry in Occupational and Environmental Health Practice, CRC Press, Boca Raton, FL, 1986, vol. 2, pp. 127-145 and 167-178. Koh, T. S . , Anal. Chem., 1980.52, 1978. Kingston, H. M., and Jassie, L. B . , Anal. Chem., 1986, 58, 2534. Marine Analytical Chemistry Standards Program, Dogfish Muscle and Liver Reference Materials for Trace Metals, National Research Council Canada, Ottawa, 1986. Paper 1101249E Received March 15th, 1991 Accepted May 20th, 1991

ISSN:0003-2654    

DOI:10.1039/AN9911600957

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

ANALYST, SEPTEMBER 1991. VOL. 116 961 Separation of Niobium From Chloride Media by Solvent Extraction With Dicyclohexyl-18-crown-6 N. V. Deorkar and S. M. Khopkar" Department of Chemistry, Indian Institute of Technolog y, Powai, Bomba y-400 076, India Niobium was quantitatively extracted from 7-10 rnol dm-3 hydrochloric acid with 2.0 x 10-2 rnol dm-3 dicyclohexyl-I 8-crown-6 in dichloromethane. The metal was stripped from the organic phase with 0.1 rnol dm-3 sulphuric acid and determined spectrophotometrically at 540 nm as its complex with 4-(2-thiazolylazo)resorcinol. This method permits the sequential separation of niobium from vanadium, tantalum, zirconium, hafnium and iron. Keywords: Niobium; solvent extraction; crown ether; separation; spectrophotometry Solvent extraction is a useful method for the separation of niobium and tantalum.Diethyl ether,' 8-hydroxyquinoline2 and N-benzoyl-N-phenylhydroxylamine3 have been used for the separation of these two elements as they do not extract tantalum. Isobutyl methyl ketone4 and tributyl phosphate (TBP)s have been applied as extractants for tantalum. The separation of niobium from tantalum has been carried out with Amberlite LA-2 from malonate media;6 tantalum was not extracted and was therefore separated from niobium. In extraction-chromatography, bis-(2-ethylhexyl)phosphoric acid has also been used to separate niobium from tantalum.' Dibenzo-18-crown-6 has been used for the extractive spectro- photometric determination of niobium.*,9 However, no systematic investigations into solvent extraction separation with crown ethers have been carried out so far; such studies are described in this paper.Experimental Apparatus and Reagents An Orion Model 901 ion analyser with a combined glass and calomel electrode, an Electronic Corporation of India (ECIL) Model GC866C spectrophotometer with 10 mm matched Corex glass cuvettes, and a wrist-action flask shaker were used. A stock solution of niobium (1 mg ml-1) was prepared by first fusing 0.370 g of niobium pentoxide with potassium hydrogen sulphate. The fused mass was then dissolved in 20 ml of 20% tartaric acid solution and made up to 250 ml with deionized water. The solution was standardized gravimetric- ally.10 A solution containing 25pg ml-1 of niobium was prepared by appropriate dilution of the stock solution. 15-Crown-5 (15C5), 18-crown-6 (18C6), dibenzo-18-crown- 6 (DB18C6), dicyclohexyl-18-crown-6 (DC18C6) and di- cyclohexyl-24-crown-8 (DC24C8) (Merck) were used without further purification.General Procedure To an aliquot of a solution containing 25pg of niobium, hydrochloric acid was added in order to adjust the concentra- tion of the latter to 7 rnol dm-3 in a total volume of 10 ml. The solution was transferred into a separating funnel. Then, 10 ml of a 2.0 x 10-2 rnol dm-3 solution of the appropriate crown ether in dichloromethane were added. The solution was shaken on the wrist-action flask shaker for 10 min. The two phases were allowed to settle and separate. Niobium was stripped from the organic phase with 10 ml of 0.1 rnol dm-3 sulphuric acid and determined spectrophotometrically at 540 nm as its complex with 4-(2-thiazolylazo)resorcinol (TAR).1 1 Results and Discussion Effect of Hydrochloric Acid Concentration Of all the acids examined in the extraction, viz., nitric, sulphuric, perchloric and hydrochloric, only the last named was found to be effective, the ideal concentration for a phase ratio of 1 : 1 for the quantitative extraction of niobium with 2.0 x 10-2 rnol dm-3 DC18C6 being 7.0-10.0 rnol dm-3 hydro- I 2 3 4 5 6 7 8 9 10 Hydrochloric acid/mol dm-3 Fig. 1 concentration. A, DC18C6; B, DC24C8; C. 18C6; and D, DB18C6 Extraction of niobium as a function of the hydrochloric acid Table 1 Effect of the DC18C6 concentration on the, extraction of niobium [DC18C6]/ Extraction 10-3 mol dm-3 (Yo 1 D 0.5 1 .o 1.5 1.7-2.0 2.5 2.7 3.0 3.5 3.7-4.0 4.5 5.0 5.5 6.0 10.0 15.0 20-50 22.2 35.5 44.4 47.3 48.8 54.5 58.3 62.9 66.6 68.7 70.6 71.4 72.9 81.5 93.6 100.0 0.285 0.550 0.80 0.90 0.953 1.2 1.4 1.7 2.0 2.2 2.4 2.5 2.6 4.4 14.4 m * To whom correspondence should be addressed.962 ANALYST, SEPTEMBER 1991, VOL.116 ~ ~~ ~~~~ Table 2 Effect of stripping agents Niobium stripped (%) Stripping 0.01 0.05 0.1 0.5 1 .0 2.0-4.0 5.0 6.0 agent rnol dm-3 rnol dm-3 rnol dm-3 mol dm-3 mol dm-3 mol dm-3 mo] dm-3 mol dm-3 HCl 45.0 62.5 100 100 100 100 61.7 20.0 HN03 31.0 56.9 78.0 100 100 - 100 100 96.1 H2S04 52.1 87.0 100 100 100 100 100 100 HC104 38.6 61.0 73.1 100 100 100 100 100 CH3COOH 13.3 33.3 46.6 68.3 87.6 100 100 100 Table 3 Separation of niobium from binary mixtures.Amount of Nb taken, 25 pg Foreign ion Lil Na' K' Be" Call Mg" Sr" Ball AP1 Sb"l SC"' Ylll Ti1" ZrIV HP" V'" Ta" Cr"l Mn" Fell1 Nil1 Zn" Cd" U"' Thl" La"' Ce"' Nd"' NO3 - S042- c104- Ci trate Acetate Tartrate Bill1 Added as LiCl NaCl KCI BeCI2 CaCI2*6H20 S I - ( N O ~ ) ~ - ~ H ~ O Ba( N03)2*4H20 Bi(N03)3*5H20 MgS04.7H20 Al(N03)3*3H20 SbCl3 SC(N03)3 Y(N03)3 Ti(S04)2 Hf( S04)2.4H20 VOS04sH20 Ta205 Zr(N03)4*4H20 Cr(N03)3 MnSO4-4H20 FeCI3 NiS04*6H20 ZnS04.7H20 3CdS04.8H20 U02(N03)2.6H20 Th( NO3)4.4H20 La(N03)3-6H20 Ce(N03)3-6H20 Nd(N03)3*6HzO HN03 H2S04 HC104 Citric acid Acetic acid Tartaric acid Ratio 1 : 200 1 : 120 1 : 100 1 : 100 1 : 100 1 : 110 1:80 1:80 1:50 1:40 1 : 4 1: 120 1 : 100 1:30 1:20 1:30 1 : 40 1:40 1:50 1:40 1 : 4 1:40 1:80 1:80 1:2 1:80 1 : 50 1 : 80 1:40 1:200 1:200 1:200 1 : 120 1:200 1:40 Tolerance limit/mg 5.0 3.0 2.5 2.5 2.5 2.75 2.0 2.0 1.25 1 .0 0.10 3.0 2.5 0.75 0.50 0.75 1 .o 1.0 1.25 1 .o 0.10 1.0 2.0 2.0 0.05 2.0 1.25 2.0 1 .o 5 .O 5 .O 5.0 3.0 5.0 1 .o Effect of Diluent Of the solvents examined, viz., benzene, toluene, xylene, carbon tetrachloride, chloroform, dichloromethane, 1,2-di- chloroethane and nitrobenzene, the last three gave 100% extraction.Dichloromethane was preferred as the diluent because it gave better phase separation. Effect of Stripping Agents Niobium was stripped after extraction with various concentra- tions of acids (0.014 rnol dm-3). Hydrochloric (0.14 rnol dm-3), nitric (0.5-5 rnol dm-3), sulphuric ( 0 . 1 4 rnol dm-3), perchloric (0.5-6 rnol dm-3) and acetic (2-6 rnol dm-3) acids were effective in stripping the niobium completely.Sulphuric acid was preferred as the stripping agent (Table 2) as it could be used over a wide range of concentrations and facilitated spectrophotometric determin- ation. Effect of Period of Shaking A study of the extraction as a function of time indicated that equilibration for 10 min was adequate for the quantitative extraction of niobium. Nature of Extracted Species The nature of the extracted species was ascertained by plotting log D ( D = distribution ratio) versus log [DC18C6] at 7 rnol dm-3 hydrochloric acid and log D versus log [hydrochloric acid] at 2.0 x 10-2 rnol dm-3 DC18C6. The corresponding slopes were 0.81 and 6.2, respectively. The probable compo- sition of the extracted species is [(NbC16)- (H30-DC18C6)+].These findings are in agreement with those reported pre- viously by other workers.9.12.13 chloric acid (Fig. 1). The extraction with the other crown ethers was incomplete but 95% extraction was obtained over a narrow acidity range. Effect of Crown Ether Concentration The optimum concentration of crown ether for the quantitat- ive extraction of niobium was ascertained by extracting the latter with various concentrations of crown ether (in the range from 0.5 x 10-3 to 5.0 x 10-2 rnol dm-3) (Table 1). The extraction was 71% with 5 X 10-3 rnol dm-3 DC18C6 and quantitative with 2.0 X 10-2 rnol dm-3 DC18C6. Effect of Counter Ions A study of the extraction of niobium in the presence of various counter ions and from mineral acids showed that quantitative extraction was obtained only from hydrochloric acid.Separation of Niobium From Binary Mixtures Niobium was extracted in the presence of a number of foreign ions (Table 3). The tolerance limit was defined as the amount of foreign ion causing an error of k2.0% in the recovery of niobium. Alkali metals, alkaline earth metals, scandium and yttrium were tolerated at a ratio of 100: 1, while bismuth, titanium, vanadium, tantalum, chromium, manganese, nickel and neodymium were tolerated at a ratio of 40 : 1. Antimony, iron(ir1) and uranium(v1) were tolerated at a lower ratio of 4 : 1. The anions studied were tolerated at a ratio of 200 : 1. Separation of Niobium and Tantalum Niobium and tantalum were separated by first extracting niobium from 7 rnol dm-3 hydrochloric acid with 2.0 x 10-2 rnol dm-3 DC18C6, when tantalum was not extracted. Separation of Niobium From Multicomponent Mixtures Niobium was separated from several multicomponent mix- tures.In those separations involving 3 or 4 metals, tantalum, vanadium and chromium were not extracted, while zirconium,ANALYST, SEPTEMBER 1991, VOL. 116 963 Table 4 Separation of niobium from multicomponent mixtures Mixture Stripping No. Components Extractant Counter ion agent 1 Nb DC18C6.0.02 mol dm-3 7 mol dm-' HCI 0.1 mol dm-3 H2S04 Hf DC18C6.0.07 mol dm-3 9 mol dm-3 HCI 0.1 mol dm-3 HC104 Ta Not extracted - Aqueous phase Nb DC18C6,0.02 mol dm-3 7 mol dm-3 HCI 0.1 rnol dm-3 H2S04 Ta Not extracted - Aqueous phase 3 Th 18C6,0.065 rnol dm-' 0.04 rnol dm-3 0.5 mol dm-3 HN03 2 Zr TBP, 0.75 mol dm-3 8 rnol dm-3 HN03 1 mol dm-3 H2S04 picric acid (PH 2.0) Nb DC18C6,0.02 mol dm-3 7 mol dm HCI 0.1 mol dm-3 H2S04 Ta Not extracted - Aqueous phase Nb DC18C6,0.02 mol dm-3 7 mol dm-3 HCl 0.1 mol dm-3 H2S04 Cr'Wa Not extracted - Aqueous phase 4 Fe"' DB18C6,O.Ol mol dm-3 7 mol dm-3 HCl 0.05 rnol dm-3 H2S04 5 Pb" DBlSC6,0.02 rnol dm-3 0.01 rnol dm-3 1 mol dm-3 HC104 picric acid (PH 3.0) Nb DClSC6,0.02 mol dm-3 7 mol dm-3 HCI 0.1 mol dm-3 H2S04 V'" Not extracted - Aqueous phase 6 Nb DC18C6,0.02 mol dm-3 7 mol dm-3 HCl 0.1 mol dm-3 H2S04 Ta DC18C6,0.05 mol dm-3 3.5 mol dm-3 HCI + 0.1 mol dm-3 HC104 V'V Not extracted - Aqueous phase 7 MeV' DB18C6,O.Ol rnol dm-' 6 mol dm-3 HCI 0.1 mol dm-3 NaOH Nb DClSC6,0.02 mol dm-3 7 mol dm-3 HCI 0.1 rnol dm-3 H2S04 Mn" Not extracted - Aqueous phase 0.2 rnol dm--7 KF * Values in parentheses are the wavelengths at which the determinations were carried out.t Determination carried out by atomic absorption spectrometry. Reagent for determination TAR (540)* Xylenol Orange (540) Brilliant Green (650) Arsenazo I11 (6.55) TAR (540) Brilliant Green (650) Arsenazo 111 (650) TAR (540) Brilliant Green (650) 1 ,lo-Phenanthroline (510) TAR (540) AASt (319)/Brilliant Green AASt (217.0) (650) TAR (540) AAST (318.4) TAR (540) Brilliant Green (650) AASt (318.4) Tiron (390) TAR (540) Brilliant Green (650) thorium , iron and molybdenum were invariably extracted before the extraction of niobium; hafnium was extracted after the extraction of niobium. The results of these separations are given in Table 4. The proposed method is simple, rapid and selective.The separation of niobium from vanadium, tantalum, zirconium, hafnium and iron is significant as these elements are associated with each other in ores and minerals. The authors thank the Council of Scientific and Industrial Research (India) for sponsoring this project and for awarding a Senior Research Fellowship to one of them (N. V. D.). 4 S 6 7 8 9 10 11 12 13 References 1 Lauw-Zecha, A. B. H., Lord, S. S., Jr., and Hume, D. N., Anal. Chem., 1952, 24, 1169. 2 Luke, C. L., Anal. Chim. Acta, 1966, 34, 165. 3 Lyle, S. J., and Shendrikar, A. D., Tulanta, 1965, 12, 573. Theodore, M. L., Anal. Chem., 1958,30, 465. Morris, D. F. C., Scargil, D., and Olya, A., Tulanta, 1960, 4, 194. Rao, R. R., and Khopkar, S. M., Anal. Lett., 1984, 17, 523. Vin, Y. Y., and Khopkar, S. M., Talanta, 1991,38, in the press. Hubert-Pfalzgraf, L. G., and Tsunoda, M., Znorg. Chim. Acta, 1980, 38, 43. Blanco-Gomis, D., Arribas Jimeno, S., and Sanz-Medel, A., Talanta, 1982, 29, 761. Mjumdar, A. K., and Mukherjee, A. K., Anal. Chim. Acta, 1958, 19, 23. Patrovsky, V., Taluntu, 1965, 12, 971. Caletka, R., Hausbeck, R., and Krivan, V., Talanta, 1986,33, 219. Davey. D. E., Catral, R. W., Cardwell, T. J., and Magee, R. J., J . Inorg. Nucl. Chem., 1978, 40, 1135. Paper 1 I00453 K Received January 31st, 1991 Accepted April 29th, 1991

ISSN:0003-2654    

DOI:10.1039/AN9911600961

出版商:RSC    

年代:1991

数据来源: RSC