886 Analyst, December, 1973, Vol. 98, fip. 886-894 Voltammetric Determination of Tocopherols by Use of a Newly Developed Carbon Paste Electrode BY SAMUEL S. ATUMA AND JORGEN LINDQUIST (Department of Analytical Chemistry, University of Uppsala. Box 531, S-751 21, Uppsala 1, Sweden) A voltammetric method is described for determining tocopherols in vegetable oils, foods and pharmaceuticals by a newly developed carbon paste electrode. The samples are saponified and the unsaponifiable fraction is extracted and determined voltammetrically. No elaborate purification method is necessary as the substances that interfere with photometric procedures are electrochemically inactive in the potential range of operation. Detailed procedures for the preparation and the working of the electrode, and results for the precision of the method, are presented.THE importance of vitamin E calls for specific and simple quantitative analytical techniques for determining both the total tocopherols and the various individual forms in natural and enriched products. Among the most common naturally occurring tocopherols are CC-, P-, y- and 6-tocopherol, of which a-tocopherol exhibits the highest biological potency. The parent compound, from which they are derived, is tocol [2-methy1-2-(4’,8’,12’-trimethyltridecyl)- chroman-6-01] and a-t ocopherol is 5,7,8- trimet hylt ocol, P- t ocopherol 5,8-dimet h ylt ocol, y-tocopherol 7,8-dimethyltocol and 6-tocopherol 8-methyltocol. The purpose of this study was to develop a quantitative voltammetric method for the determination of tocopherol involving the use of a newly developed carbon paste electrode. Most of the existing methods are either too tedious, incomplete or non-specific. Until recently the best known method for the quantitative determination of tocopherol in biological materials and pharmaceutical preparations has been the direct or modified application of the Emmerie-Engel colorimetric method.l-15 This method depends on the reduction of the iron(II1) ion to iron(I1) by the tocopherols and subsequent reaction with 2,2’-bipyridyl to form the red iron(I1) complex, the absorbance of which is measured at 520nm in a spectrophotometer. Prior to measurement the tocopherols are purified and separated into the individual forms.The several techniques that have been used in the separation procedure involve the use of column chromatography on different materials,*-11J3 paper chromat~graphy~s~*~~J~-~~ or thin-layer ~hromatography.~J~J~~~~~~1-~5 The best separa- tion seems to have been obtained by S t ~ w e , ~ ~ who used a five-component solvent system for chromatographing tocopherols on thin layers of silica gel and was able to separate 18- from y-tocopherol.The application of these separation techniques does not necessarily preclude the general preliminary treatment of tocopherol samples by saponification and adsorption chromato- graphy on to floridin earth. The Emmerie-Engel reaction is, however, susceptible to many interferences. Booth16 has pointed out that tocopherol simulators from fat solvents and adsorbents could give spurious results with the Emmerie-Engel method, thus making the method highly non-specific and unreliable. The “nitroso” method of Quaife26 and the “dianisidine coupling” method of Weisler, Robeson and B a ~ t e r ~ ~ for determining individual tocopherols are tedious and suffer from certain Recently gas - liquid techniques for separating and determining quantitatively the major tocopherol forms have been described and a ~ p l i e d .~ l - ~ ~ Some of the work reported is con- centrated on pharmaceutical^^^-^^ in which the tocopherol content is high and interferences are limited, and in many instances only cc-tocopherol is present. In applying gas - liquid chro- matography to naturally occurring mixtures, initial purification is of the utmost importance as sterols and cholesterols appear in the region of the tocopherols in gas chromatograms.40s41 The polarographic method for determining tocopherols was introduced by Smith, Spillane and Kolthoff .42 They found that a-tocopherol was oxidised at the dropping-mercury anode.@ SAC and the authors.ATUMA AND LINDQUIST 887 The anodic waves of p- and y-tocopherol did not give a diffusion current because the oxidation occurred at too positive potentials. The same authors43 also showed that a-tocopherol could be determined at the dropping-mercury cathode by first oxidising a-tocopherol to a-tocopheryl quinone with iron(II1) chloride. This polarographic technique has been-used, with modifica- tions, to determine tocopherols in vegetable oils, fats and pharmaceutical^.^^-^^ In these instances, the tocopherols were oxidised with cerium(1V) sulphate and the resulting tocopheryl quinones analysed polarographically at a mercury electrode.Tocopherols have also been determined by amperometric titration with gold( 111) chloride by using a dropping-mercury electrode as the indicator ele~trode.~’ Cospito, Raspi and Lucarini48 titrated the tocopherols with cerium( IV) sulphate and used a bubbling platinum electrodegg as the indicator electrode. Amperometric titrations are carried out only with samples containing a high tocopherol concentration and do not distinguish between the various forms of tocopherol. The application of a voltammetric method has been reported by Lucarini, Cospito and Raspi.50 By using a platinum electrode, with “periodic surface renewal ,” as indicator elec- trode51 they were able to determine the individual tocopherols in fats, oils, pharmaceuticals and foods.The electrode system in this method seems, however, to be rather complicated, and the poisoning effect of the electrode surface is not eliminated. The use of a wax-impregnated graphite electrode in the polarography of organic com- pounds was introduced by Gaylor, Conrad and Landerl,52 and Nash, Skauen and Purdy53 applied it to the study of the polarographic behaviour of certain antioxidants, including a-tocopherol. The need for an easily renewable stationary electrode for organic compounds in the anodic region has been greatly felt because surface film formation and adsorption of reactants or intermediates are known to influence the reproducibility of peak current measurements. Platinum has been the most widely used material for solid electrodes in anodic voltammetry, but it is far from being ideal for quantitative measurements. Its pre-treatment is of the utmost importance and it is very difficult to reproduce the surface accurately.For this reason wax-impregnated graphite52 and carbon paste electrodes54 seem to be the most practical solid electrodes for analytical work. The surface of the paste electrode is easily reproducible within 1 per cent.55956 A new carbon paste, impregnated with ceresin wax, has, however, been developed in this Department for use in all of the common solvents used in electrochemistry. In the present study of the voltammetric determination of tocopherols, this electrode was used as the working electrode.EXPERIMENTAL INSTRUMENTATION- The voltammetric measurements were made with a three-electrode system polarograph for linear-sweep voltammetry, built at this Department. It is a solid-state device with an analog section based on operational amplifiers, and a control logic section with transistor gates, reed relays and manual switches. The intention with this system has been to facilitate easy and safe operation with some consideration of possible automation. A full description of the polarograph and its electronics can be obtained on request. Any commercial three- electrode polarograph could, however, be used (even a two-electrode polarograph would suffice if corrections were made for eventual I R drop). The voltammograms were recorded with a Philips recorder (PM 8100) and a Houston Instrument XY-recorder (Model 2000).CELL AND ELECTRODES- The electrolyte consisted of a 0.2 M solution of sulphuric acid in 75 per cent. ethanol. The reference electrode was a calomel electrode with a salt bridge containing a saturated, aqueous solution of lithium chloride (Radiometer). A platinum wire served as the counter electrode. The working electrode was the newly developed carbon paste electrode prepared in the following way. An amount (0-5 g) of ceresin wax was dissolved in 20 ml of warm n-hexane (40 to 50 “C) in a beaker placed on a water-bath and 9.5 g of graphite powder were added with stirring. The dry graphite powder, now containing 5 per cent. m/m of ceresin wax, was carefully mixed with silicone oil MS 510 in the proportions 5:3 m/m to give a homogeneous paste.In any case, interference can be very trouble~orne.~~ The stirring was continued until all of the n-hexane had evaporated.888 ATUMA AND LINDQUIST : VOLTAMMETRIC DETERMINATION OF [Analyst, VOl. 98 The paste was then tamped into an electrode holder,55 care being taken to ensure that the paste was not too firmly packed. The electrode surface, with an area of 0-31 cm2, was smoothed by cutting off the excess of paste with a stretched length of a 0.2 mm piano wire while the electrode was being rotated. After each voltammetric measurement the electrode surface was easily renewed by pressing out about 1 mm of the paste and repeating the cutting procedure. In Fig. 1 are shown the anodic ranges of the electrode in 85 per cent.ethanol, acetonitrile, acetic acid and dimethyl sulphoxide. The supporting electrolytes were sodium perchlorate, ammonium acetate, sodium acetate and lithium chloride. Potential versus S.C.E./V + 1.5 +- 1.0 D I 2 14 3 > Background currents for the carbon paste electrode in different solvents: a, 85 per cent. V / V ethanol; b, acetonitrile; c, acetic acid ; and d, dimethyl sulphoxide Fig. 1. REAGENTS- Pure tocopherols (Hofman L a Roche Inc.). Tocol (Koch-Light Laboratories Ltd.). Floridin earth X S . Graphite. Ceresin wax, white. Silicoae oil MS 510. Pyrogallol solution-A 5 per cent. m/V solution in ethanol was prepared daily. Sodium ascorbate solution-A 5-g amount of sodium ascorbate was dissolved in 10 ml of distilled water. Potassium hydroxide solution-A 160-g amount of potassium hydroxide was dissolved in 100ml of distilled water.All the electrolytes used were of analytical-reagent grade. CALIBRATION GRAPH- A standard solution was made by dissolving pure cc-tocopherol in absolute ethanol. From this solution a series of solutions was prepared, by serial dilution, in the concentration range of a-tocopherol from 3 X lov6 to 7 X loA4 M, with a 0.2 M solution of sulphuric acid in 75 per cent. ethanol as solvent. Each solution was transferred into a polarographic cell (Metrohm, 20 ml) and thermostatically controlled at 20 & 0.1 "C. At least three voltammo- grams were recorded for each solution, with the electrode surface renewed after each run. Fig. 2 shows a typical voltammogram for a-tocopherol. The voltage scan was 0.5 V min-1.The peak current, i,, was measured for each solution and the blank subtracted. The height of the peak was a linear function of the concentration of tocopherol in the solution (Fig. 3). All samples were saponified in the presence of ethanolic pyrogallol or aqueous sodium ascorbate with potassium hydroxide solution for 15 minutes and the unsaponified material was extracted and determined voltammetrically. The method of saponification used was a modified version of that recommended by the Analytical Methods Committee.17 A small amount of the sample (0.5 to 3 g, depending on the tocopherol content of the sample) was weighed into a round-bottomed flask. Next, 1 ml of sodium ascorbate or 4 ml of pyrogallol solution and 20 ml of methanol were added, and the mixture was heated on a water-bath under reflux.Then 1 ml of potassium hydroxide solution (2 to 3 ml if 2 to 3 g of oil sample were saponified) was added and the refluxing continued for 15 minutes with occasional swirling. After cooling, 20 ml of distilled water were added and the unsaponified material was extracted with 75 ml of diethyl ether, about 25 ml being ALKALINE HYDROLYSIS AND ANALYSIS OF SAMPLES-December, 19731 TOCOPHEROLS BY USE OF A CARBON PASTE ELECTRODE Potential versus S.C.E./V 0.7 0.6 0.5 0.4 .Oh3 889 Fig. 2. A typical anodic voltammogram for the oxidation of a-tocopherol in 0.2 M sul- phuric acid solution in 75 per cent. ethanol. used for each extraction. The combined ether extracts were washed with 25-ml portions of water until neutral to phenolphthalein; the ether was then removed by evaporation under reduced pressure, while warming the solution on a water-bath.The dry residue was dissolved in 10 ml of 80 per cent. ethanol (20 ml if the tocopherol content was high), and 0-7 ml (or 1.4 ml) of 3 M sulphuric acid added to give a solution in 75 per cent. ethanol that was 0.2 M in sulphuric acid. This solution was then transferred into a polarographic cell and subjected to voltam- metry. ACID HYDROLYSIS- 12 ml of 2.5 M sulphuric acid in 55 per cent. ethanol were added. for 3 hours. 150-ml calibrated flask. tocopherol determined directly. The sample was dissolved in 10ml of absolute ethanol in a round-bottomed flask and This solution was refluxed The flask was cooled and its contents were transferred quantitatively into a The final volume was adjusted with 75 per cent. ethanol and the 1 10 100 Concentration/pMol Calibration graph for a-tocopherol Fig.3. RESULTS AND DISCUSSION The voltammetric behaviour of cc-tocopherol in a 0-2 M solution of sulphuric acid in 75 per cent. ethanol is shown in Fig. 2. De-aeration of solutions prior to voltammetric recording was found to be unnecessary as the results obtained with de-aerated and un- de-aerated solutions were essentially the same.890 ATUMA AND LINDQUIST : VOLTAMMETRIC DETERMINATION OF [Analyst, VOl. 98 On the basis of their structural relationship the various forms of tocopherol could be assumed to have approximately the same diffusion coefficient. A calibration graph for the a-form could therefore suffice for the determination of the other forms.Solutions to be analysed should contain 1.5 to 240 pg ml-l of tocopherol. Table I contains the peak and half-peak potentials for the four forms of tocopherol that were considered in this study. The voltammograms from which these potentials were calculated were recorded with a Houston Instruments XY-recorder (Model 2000), with a calomel in saturated potassium chloride electrode as reference electrode. TABLE I PEAK (E,) AND HALF-PEAK POTENTIALS OF THE DIFFERENT TOCOPHEROL FORMS E , versus S.C.E./V EP12 veisus S.C.E./V a-Tocopherol .. .. 0.473 0.445 fi-Tocopherol . . .. 0.550 0.520 y-Tocopherol .. .. 0-555 0.525 &Tocopherol . . .. 0-623 0.59 1 Tocol . . .. . . .. 0.646 0.614 Hedenburg and Freiser5’ have shown that an alkyl substituent in the meta position in a phenol ring has little effect on the ease of oxidation of phenol, but the inductive effects of alkyl groups in the ortho and para positions render substituted phenolic compounds more easily oxidisable than phenol itself.The work of Suatoni, Snyder and Clark58 on voltammetric studies of phenol and aniline ring substitution is in agreement with Hedenburg and Freiser’s conclusions. This phenomenon explains the order in which the tocopherols are oxidised at the carbon paste electrode. The small difference between the oxidation potentials of 18- and y-tocopherols can then be understood as they both have a methyl group in one ortho position and another methyl group in the meta position of the same ring.No doubt, the oxidation potential of the parent compound, tocol, lies very close to (possibly a little more positive than) that of &-tocopherol because the methyl group in the meta position of &tocopherol contributes very little to the ease of its oxidation. This belief was eventually confirmed (Table I). PURE TOCOPHEROLS- Pure tocopherol was treated as described under Alkaline hydrolysis and analysis of samples and the voltammograms were recorded after thermostatic control of the solution temperature in a polarographic cell. The results shown in Table I1 indicate that no significant losses of tocopherol were encountered during the saponification and extraction processes. TABLE I1 RECOVERY OF PURE a-TOCOPHEROL AFTER SAPONIFICATION AND EXTRACTION 0.108 0.106 0.001 0.115 0-115 0.001 0-615 0.616 0.001 1.184 1.181 0.003 1-648 1.649 0.002 2.283 2.280 0.003 Taken/mg Foundlmg Standard deviation TOCOPHERYL ACETATE- The tocopheryl acetate did not give a peak if it was not first hydrolysed.Acid hydrolysis of the ester gave as good a result as alkaline hydrolysis, the only disadvantage with the acid hydrolysis procedure being the length of time required for complete hydrolysis. PHARMACEUTICALS- All the tablets analysed in this work contained a rather high concentration of cc-toco- pherol; none of them contained any other form of tocopherol. The tablets were thoroughly ground and mixed, then about 0.5 g was saponified directly and the unsaponifiable matter extracted and analysed. Pre-extraction before the saponification procedure was found to be unnecessary.The results obtained with some tablets are shown in Table 111.December, 19731 TOCOPHEROLS BY USE OF A CARBON PASTE ELECTRODE TABLE I11 COMPARISON OF VOLTAMMETRIC AND EMMERIE-ENGEL ASSAYS FOR a-TOCOPHERYL ACETATE I N MULTIVITAMIN TABLETS. 891 Voltammetric assay Emmerie-Engel assay Declared & ----7 potency/ Found/ Standard Found/ Standard Sample mg per tablet mg per tablet deviation mg per tablet deviation 1 5 to 6 5.10 0-04 5.02 0-24 2 5 to 6 5.58 0.01 5-61 0.1 1 3 5 to 6 5.76 0.09 5.83 0.24 4 5 to 6 5.58 0.03 5-33 0.3 1 5 5 to 6 5.23 0.06 5.21 0.09 6 5 to 6 5-55 0.07 5-27 0-28 7 10 11.24 0.06 10-27 0.33 OILS- All the oils examined had to undergo saponification in order to remove as much of the oil base as possible because the presence of oil in the final solution could affect the peak current of the tocopherol present.The results obtained when oils of corn germ, sunflower, linseed and wheat germ were examined are shown in Table IV. TABLE IV VOLTAMMETRIC ASSAY OF TOCOPHEROLS I N SOME NATURAL PRODUCTS Product Corn germ oil Sunflower oil Linseed oil Micromilk Wheat germ Wheat germ oil Mixed tocopherols Declared tocopherol content/ mg g-l 1.0 0.9 0.25 0.15 to 0.3 2.5 About 500 - U-TOCO- pherol Trace 0.900 - - 0.145 2.09 314 Tocopherol content found/mg g-1 p- and/or y- S-Toco- tocopherol pherol Total 0.968 0.083 1.050 Trace - 0.900 0.781 - 0.781 0.160 0.073 0.233 0.073 - 0.2 18 0-30 - 2.39 157 41 512 A 1 Standard deviation 0,018 0.018 0.021 0.002 0.003 0.025 10 TOCOTRIENOLS AND OTHER REDUCING SUBSTANCES- The behaviour of tocotrienols, which are structurally related to the tocopherols, has not yet been studied.It is presumed that they may be electrochemically active in the same potential region as the tocopherols. Studies were carried out on vitamins A and D, /3-carotene and cholesterol, but these substances did not give rise to any waves in the region of operation. TOCOPHEROL MIXTURE- The determination of the different forms of tocopherol in a mixture presented some difficulties as their peak potentials are rather close to one another (Table I and Fig. 4). The second peak lies on the slope of the first and the third peak on the slope of the second, and this gives an incorrect interpretation of the heights of the subsequent peaks. In addition, these subsequent peaks are affected by the poisoning effect of the electrode surface, resulting from the recording of the first peak or peaks. A method has been developed for determining two tocopherol forms from a single voltam- mogram.In very rare cases three forms may be present, but the third is usually either present as a trace amount or so little is present that its approximate concentration can be obtained by graphical extrapolation. Fig. 5 shows voltammograms of wand P-tocopherol, both singly and in a mixture. The first peak current, AB, is measured without difficulty. The true value of i p for the second peak corresponds to CE and not to CD. By plotting a graph of EF, is, v m u s the con- centrations of a-tocopherol corresponding to the various heights of AB, a linear relationship is obtained between is and the a-tocopherol concentration over the range 3 x lo4 to 7 x 1 0 - 4 ~ (Fig.6). It is very important that i, values should be measured at the peak Usually only one or two tocopherol forms occur in any one natural product.892 -\ -1 _----I-- \ '\ \ ATUMA AND LINDQUIST : VOLTAMMETRIC DETERMINATION OF [A%dySt, VOl. 98 - 4 2, - 6 :: 5- Potential versus S.C.E./V 0.8 0.7 0.6 0.5 0.4 0.3 - - -I 3 - 8 2 z -10 3, 2 D 12 14 16 Fig. 4. A voltammogram of a-, /?-, and 8-tocopherol in a mixture. Electrolyte : 0.2 M sulphuric acid solution in 75 per cent. ethanol. 1, a-Tocopherol ; 2, /?-tocopherol; and 3, 6- tocopherol Potential . C Fig. 5. Voltammo- grams of a- and /3- tocopherol when present separately (dotted line) and in a mixture (solid line) potential of /3-tocopherol.When AB is known, EF is obtained from the calibration graph. In any mixture of cx- and @-tocopherol, therefore, the true value of i p is obtained by subtracting EF from CF. The value of i p can be expressed mathematically in the following way: where k = i,/iM. The mixtures of cc- and &tocopherol and @- or y- and 8-tocopherol are treated in a similar way. p- and y-tocopherol have approximately the same oxidation poten- tials and it was impossible to separate them by use of this method. It is very rare, however, for these two tocopherol forms to appear in the same natural product. i, = Z'CF - k x ~ A B 10 Y Q, a Concentration/pMot A, peak current of a-tocopherol as a function of concentration; B, limiting current of a-tocopherol a t the peak potential of /3-tocopherol as a function of concentration Fig.6. In Table V the results obtained when a mixture of pure or-tocopherol and linseed oil was analysed by using the method outlined above are shown. The value of &)-tocopherol in linseed oil is in agreement with the value obtained without the addition of a-tocopherol. These results indicate the high precision of the method. A series of samples of differentDecember, 1873_ TOCOPHEROLS BY USE OF A CARBON PASTE ELECTRODE 893 tocopherol contents was analysed in triplicate and the mean results are shown in Table IV. All of the oils analysed were bought in a local shop. TABLE V RECOVERY OF WTOCOPHEROL ADDED TO LINSEED OIL a-Tocopherol a-Tocopherol Standard p( y)-Tocopherol Standard Sample added/mg recovered/mg deviation found/mg g-l deviation 0.021 1 2 0.171 0.173 0.002 0-780 0.019 3 0.560 0.559 0.001 0-779 0.015 4 1.950 1.953 0.010 0.782 0-015 - 0.781 - - For comparison purposes some pharmaceutical tablets were also analysed by the Emmerie-Engel reaction method.With this method, all of the blank solutions, as well as the sample solutions, were run through the column of floridin earth to correct for any toco- pherol simulatorslS from the adsorbents. Table I11 is a summary of the results obtained with the different methods. CONCLUSION I t is worthy of note that this voltammetric method coinpletely obviates the need for column, thin-layer or paper chromatography as a necessary clean-up procedure for the samples prior to analysis.Carotenoids, vitamin A, steroids and other reducing substances, which interfere in other methods, are completely inactive electrochemically in the potential range of operation. The method is sensitive, simple and specific and its high precision in comparison with the widely used Emmerie-Engel method can be seen from Table 111. The insolubility of the electrode paste in all the common organic solvents and the very low background currents that are obtained over a very wide range of potentials make this electrode unique among the existing solid electrodes used in the study of organic compounds dissolved in organic solvents. It would appear that the great reliability and simplicity of the electrode should make this method a most suitable tool for the determination of toco- pherols. The authors thank Professor Bengt Nygard and Professor Folke Nydahl for their kind interest in this work.Grants from The Swedish Association for the Pharmaceutical Prepar- ation Industry and the Faculty of Mathematics and Natural Sciences, University of Uppsala, are greatly appreciated. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. REFERENCES Emmerie, .I., and Engel, C., Nature, 1938, 142, 873. Tsen, C. C., Analyt. Chem., 1961, 33, 849. Green, J., and Marcinkiewicz, S., Analyst, 1959, 84, 297. Marcinkiewicz, S . , and Green, J., Ibid., 1959, 84, 304. Edisbury, J . I<., Gillow, J., and Taylor, R. J., Ibid., 1954, 79, 617. Tosic, J., and Moore, T., Biochem.J., 1945, 39, 498. Lambertsen, G., and Braekkan, 0. R., Analyst, 1959, 84, 706. Lambertsen, G., Myklestad, H., and Braekkan, 0. R., Acta Agric. Scand., 1967, 17, 13. Bro-Rasmussen, F., and Hjarde, W., Acta Chem. Scand., 1957, 11, 34. -- , Ibid., 1957, 11, 44. Lambertsen, G., Myklestad, H., and Braekkan, 0. R., J . Fd Sci., 1964, 29, 164. Whittle, K. J., and Pennock, J. F., Analyst, 1967, 92, 423. Eggitt, P. W. R., and Norris, F. W., J . Sci. Fd Agric., 1955, 6 , 689. Sturm, P. A., Parkhurst, R. M., and Skinner, W. S., Analyt. Chem., 1966, 38, 1244. Crawford, R. K., Naramore, D. C., and Esmerian, 0. K., J . Pharm. Sci., 1968, 57, 1716. Booth, V. H., Analyt. Chem., 1961, 33, 1224. Analytical Methods Committee, Analyst, 1959, 84, 356. Brown, F., Biochem. J.. 1952, 52, 523.Eggitt, P. W. I<., and Ward, L. D., J . Sci. Fd Agric., 1953, 4, 569. Green, J., Marcinkiewicz, S., and Watt, P. R., Ibid., 1955, 6, 274. Dilley, K. X., Analyt. Biochevut., 1964, 7, 240. Lambertsen, G., Myklestad, H., and Braekkan, 0. R., J . Sci. Fd Agric., 1962, 13, 617. Schmandke, H., J . Chromat., 1964, 14, 123. Stowe, H. D., Archs Biochem. Biophys.. 1963, 103, 42. Herting, D. C., and Drury, E. E., J . Chromat., 1967, 30, 502. Quaife, M. L., J . Biol. Chem., 1948, 175, 605. Weisler, L., Kobeson, C. D., and Baxter, J. G., Analyt. Chem., 1947, 19, 906. Lehman, R. W., Meth. Biochem. Analysis, 1955, 2, 153.894 ATUMA AND LINDQUIST Polister, B. H., Analyt. Chem., 1954, 26, 407. Eggitt, P. W. R., and Norris, F. W., J . Sci. Fd Agric., 1956, 7, 493.Wilson, P. W., Kodicek, E., and Booth, V. H., Biochem. J., 1962, 84, 524. Libby, D. A., and Sheppard. A. J., J . Ass. 08. Agric. Chem., 1964, 47, 371.. Slover, H. T., Lehman. J., and Valis, R. J., J . Amer. Oil Chem. SUC., 1969, 46, 417. Nelson, J. P., and Milun, A. J., Ibid., 1968, 45, 848. Pillsbury, H. C., Sheppard, A. J., and Libby, D. A., J . Ass. Off. Analyt. Chenz., 1967, 50, 809. Kovensky, B. J., and Day, R. I., J . Chrumat. SGZ‘., 1971, 9, 442. Bowman, P. B., and West, W. E., J . Pharm. Sci., 1968, 57, 470. Sheppard, A. J., Hubbard, W. D., and Prosser, A. R., J . Ass. Off. Analyt. Chenz., 1969, 52, 442 Mahn, F. P., Viswanathan, V., Plinton, C., Menyharth, A., and Senkowski, B. Z., J . Pharm. Sci. Nair, P. P., and Turner, D. A., J . Amer. Oil Chenz. SOC., 1963, 40, 353. Carroll, K. K., and Herting, D. C., Ibid., 1964, 41, 473. Smith, L. I., Spillane, L. J., and Kolthoff, I. M., J . Amer. Chem. SOC., 1942, 64, 447. --- , Ibid.. 1942, 64, 644. Knohoch,’E., Macha, F., and Mnoucek, K., Chemicke’ Listy, 1952, 46, 718. Niederstebruch, A., and Hinsch, I., Fette Seifen Anstr-Mittel, 1967, 69, 559. Wisser, K., Heimann, W., and Fitsche, C., 2. analyt. Chem., 1967, 230, 189. Smith, L. I., Spillane, L. J., and Kolthoff, I. M., J . Amer. Chem. Soc., 1942, 64, 646. Cospito, M., Raspi, G., and Lucarini, L., Analytica Chim. Acta, 1969, 47, 388. Cozzi, D., Raspi, G., and Nucci, L., J . Electroanalyt. Chem., 1963, 6, 275. Lucarini, L., Cospito, M., and Raspi, G., Farrnacu Edizione Practica, 1969, 25, 39. Cozzi, D., Raspi, G., and Nucci, L., J . Electroanalyt. Chem., 1966, 12, 36. Gaylor, V. F., Conrad, A. L., and Landerl, J. H., Analyt. Chem., 1957, 29, 224. Nash, R. A., Skauen, D. M., and Purdy, W. C., J . Amer. Pharm. Ass., 1958, 47, 433. Adams, R. N., “Electrochemistry at Solid Electrodes,” Marcel Dekker Inc., New York, 1969. Lindquist, J., J. Electroanalyt. Chtem., 1968, 18, 204. Marcoux, L. S., Prater, K. B., Prater, B. G., and hdams, R. N., Analyt. Chem., 1965, 37, 1447. Hedenburg, J. F., and Freiser, H., Ibid., 1953, 25, 1355. Suatoni, J. C., Snyder, R. E., and Clark, R. O., Ibid., 1961, 33, 1894. 1968, 57, 2149. Received March 29th, 1953 Accepted July 4th. 1973 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58.