498 Analyst, August, 1968, Vol. 93, fifi. 498-506 The Application of Polarography and Related Electroanalytical Techniques to the Determination of Sodium Diethyldithiocarbamate BY M. J. D. BRAND AND B. FLEET (Chemistry Defiartment, Imperial College of Science and Technology, London, S. W. 7 ) Cathodic stripping analysis has been shown to be an extremely sensitive technique for the determination of compounds that form insoluble mercury salts. The experimental parameters governing the application of this tech- nique have been studied and applied to the determination of sodium diethyl- dithiocarbamate. The results obtained were compared with those from con- ventional d.c. polarography, a.c. polarography and linear potential-sweep chrono-amperometry . DERIVATIVES of dithiocarbamic acid are widely used as fungicides.The compounds that have been used can be divided into three main typesl- simple substituted dithiocarbamic acid salts R S \ II N-C-S- M’+/Tz R .;’ the thiuram disulphides R S S R \ I1 I / N-c-s-s-LN R / ‘R where R = CH,, etc. R’ = H, CH,, etc. M = Na+, Fe3+, Zn2+, etc. where R= CH,, etc. and salts of ethylenebisdithiocarbamic acid S II CHZ-NH-GS- CH,-NH--C--S- i 2Mn+/.n where M = Na+, Zn2+, Mne+, etc. II S The existing analytical methods for this class of compounds are based on conversion of the sulphur into carbon disulphide, which is then determined colorirnetrically, e.g., after reaction with Viles’ reagent.2 The sensitivity of this somewhat lengthy procedure is limited. More- over, the method is only semi-specific in that it cannot distinguish between dithiocarbamates and thiuram disulphides.Polarography affords a selective and sensitive technique for the determination of a wide range of organosulphur compounds, but has so far found limited application in pesticide analysis: particularly in regard to the dithiocarbamic acid derivatives. As the dithio- carbamate ion is the fungitoxic species and is also relatively unstable, methods of analysis for this class of compound are more meaningful if the anodic wave of the dithiocarbamate anion, rather than the cathodic wave of the metal ion, is measured. Nangniot has used this approach in the determination of residues of zinc4 and irons dimethyldithiocarbamates in plants . 0 SAC and the authors.BRAND AND FLEET 499 This paper describes the study of simple substituted dithiocarbamic acid salts, sodium diethyldithiocarbamate being chosen as the model substance.Simple mono- and dialkyl- substituted dithiocarbamates show anodic waves caused by the formation of insoluble mercury salts.6 The formation of insoluble films on liquid electrodes has been studied in detail by potentiostatic and a.c. impedance measurements.' Most of the detailed polarographic and related studies have been concerned with inorganic anions, in particular the halide^.^^^ Although many organic compounds form insoluble mercury derivatives only a few of these have been studied.10 This reflects the difficulties associated with elucidation of the electrode process, e.g., determination of n, and the isolation of the metastable products of the electrode reaction.At present only a qualitative description of the electrode process for this latter group of compounds is available; this is sufficient basis, however, for the development of an analytical method. EXPERIMENTAL REAGENTS- All reagents used were of analytical-reagent grade. Sodium diethyldithiocarbamate. B u f e r solutiout--This was made from 1 M ammonia solution and 1 M ammonium nitrate. Sodium diethyldithiocarbamate stock solutions , to M, aqueous. An aliquot of the appropriate stock solution was added to 1 ml of buffer solution and diluted to 10ml. For a.c. measurements undiluted buffer solution was used as supporting electrolyte. Solutions were de-oxygenated by passage of nitrogen for 3 minutes. APPARATUS- Direct current polarograms were recorded on a polarograph, type OH-102 (Metrimpex, Hungary).A Kalousek cell with a separated saturated calomel electrode (S.C.E.) was used. The characteristics of the dropping-mercury electrode, measured at 0.0 volt versus S.C.E. in 0.1 M potassium chloride solution were t = 3-82 s, m = 2-32 mg s-1 at h 65 cm. Alternating current polarograms were measured with a Univector and general-purpose polarograph (Cambridge Limited, London), with the Metrimpex polarograph being used as the current-output recorder. The connections to the positive cell and negative meter on the general-purpose polarograph were removed and the two terminals short-circuited with a wire. This modification removed the 6-K ohm sensitivity shunt from the output, at the same time eliminating the zero and damping controls.The sensitivity, zero, and damping controls present on the recorder were subsequently used. A Kalousek cell was used in which a platinum-wire counter electrode was placed near the dropping-mercury electrode. The reference and counter electrodes were connected to a 5000-pF capacitor. Linear potential-sweep chrono-amperometric measurements with a dropping-mercury electrode were made with a KlOOO cathode-ray polarograph (Southern Analytical Limited, Surrey). Measurements were also made with a Davis differential cathode-ray polarograph (Southern Analytical Limited, Surrey). In each of these a Kalousek cell was used, as the cells supplied with the instrument have a mercury-pool anode that is unsuitable for studying compounds that react with mercury.The Davis differential polarograph was used in the single cell mode as it was impossible to use two Kalousek cells without major modification of the electrode stand. Cyclic voltammograms were obtained with the Metrimpex polarograph with a hanging mercury drop electrode in a Kalousek cell. The hanging mercury drop electrode was con- structed by cementing a glass capillary tube (0-2-mm bore) into the barrel of an Agla micro- meter syringe burette (Burroughs Wellcome Limited, London). Electrical connection was via a platinum wire sealed into the syringe barrel, which was thermally insulated with expanded polystyrene. For cathodic stripping the Metrimpex polarograph was used to supply both the applied potential for the pre-electrolysis and the voltage sweep for the stripping process.Two types of electrode system were used: a hanging mercury drop electrode and a mercury-coated platinum electrode. The mercury-coated platinum electrode consisted of a platinum wire (10 x 0-6 mm) sealed into a glass tube and coated with mercury by using the method of Joyce and Westcott.ll The reference electrode was a large current capacity S.C.E. con- nected by a salt bridge to a micro liquid junction tube (E.I.L. Limited, Surrey). The cell500 BRAND AND FLEET: APPLICATION OF POLAROGRAPHY TO THE [AndYSi!, VOl. 93 used for the hanging mercury drop electrode was a 25-ml tall-form beaker closed with a rubber bung through which electrodes and gas tubes were passed. For the mercury-coated platinum electrode the cell was made from a B45 socket reduced on to a tube of 3-cm diameter and closed with a flat end.A Perspex cover was machined to fit the ground-glass socket. Solutions were stirred magnetically, a glass-covered rotor being driven by a small magnet attached to a motor. The mercury-coated platinum electrode was mounted in a PTFE stirrer gland (Quickfit and Quartz Limited) and rotated at about 600 r.p.m. by a belt drive from a squirrel cage motor. RESULTS AND DISCUSSION DIRECT CURRENT POLAROGRAPHY- The results of a preliminary survey of the d.c. polarographic behaviour of sodium diethyldithiocarbamate were in agreement with earlier findings6 and followed the general pattern of insoluble mercury salt formation. In alkaline solution two anodic waves are observed. The wave at more negative potentials (I) is concentration independent above M, and shows a linear relationship between current and reservoir height.The wave at more positive potentials (11) begins more or less discontinuously and is only present at concentrations above ~ O - * M when it is con- centration dependent. The first wave (I) has been classified as a Brdicka absorption wave, and arises from the formation of an insoluble monolayer on the drop surface. Further oxidation is then hindered until a more positive potential is reached when the barrier to the charge-transfer reaction is overcome, resulting in a discontinuity in the current - voltage graph. ALTERNATING CURRENT POLAROGRAPHY- In alkaline solution sodium diethyldithiocarbamate showed two waves in the concen- tration range 10-3 to l o - 4 ~ (Fig.1) at potentials corresponding to the d.c. waves. The concentration dependence of these two peaks followed the pattern previously observed with d.c. polarography. Peak I was concentration independent over the range to 1 0 4 ~ , but linearly dependent in the range to 1 0 4 ~ . Peak I1 showed a rectilinear peak height - concentration relationship over the range M but disappeared below lo-* M. The use of these two peaks gave a rectilinear ip - concentration dependence over the range 10-6 to 103 M (Tables I and 11). Thus the wide concentration range attainable, together with to -0.2 4 . 3 -0.4 -0.5 -0.6 -0.7 -0.8 -0.9 Volts versus S.C.E. Fig. 1. Concentration dependence of a.c. waves of sodium diethyldithio- carbamate (I) 2 x M, (11) 4 X lo-* M, (111) 6 X M, (Iv) 8 X M, (v) 10-3 MAugust, 19681 DETERMINATION OF SODIUM DIETHYLDITHIOCARB4MATE TABLE I COMPARISON OF POLAROGRAPHIC METHODS FOR THE DETERMINATION OF SODIUM DIETHYLDITHIOCARBAMATE Currents in pA A I .50 1 Concentration of sodium diethyl- dithiocarbamate, M 2 x 10-5 4 x 10-5 8 x 10-5 1 x 10-4 2 x 10-4 4 x 10-4 8 x 10-4 1 x 10-3 6 x 6 x Direct current Alternating current polarography polarography +7 +--7 I I1 I I1 - - 0.54 - - - 1.08 - - - 1.60 - - - 2.13 - 0-29 0 2-64 0 0.29 0.34 4.54 0.64 0-29 1-42 6.30 2-05 0.29 2-50 6.43 3-46 0.29 4.08 6.40 4.84 0-29 - 6-40 6-24 * Cathodic voltage sweep. I Pre-wave. I1 Diffusion controlled wave. Linear potentiallsweep chrono-amperometry * & I I1 0.61 - 1-15 - 1-74 - 2-30 - 2-67 0 3-05 4.9 3-05 17.6 3-05 28-8 - - - - the advantages associated with the derivative type signal obtained, makes this technique ideally suited to the analysis of dithiocarbamate formulations.A concentration of 10-6 M must be taken as the lower limit attainable with the apparatus used, as at this level the instrument noise became comparable with the signal. About 1 per cent. a.c. ripple was observed on the rectified output of the Univector unit at twice the applied frequency of about 35 Hz. The recorder that was used (Metrimpex polarograph) incorporated parallel T-filters tuned to frequencies of 1.6, 0.8, 0.8 and 0.4 Hz, which are effective in filtering oscillations caused by drop growth, but do not significantly reduce the 70-H~ ripple. The limit of detection could be extended by the use of an instrument based on solid-state electronics12 or by incorporating a filter tuned to higher frequencies.At high concentrations, e.g., low3 M, when both waves are well developed it is apparent that the width of wave I at half-height, 144 mV, is considerably greater than the corresponding TABLE I1 COMPARISON OF CATHODIC STRIPPING VOLTAMMETRY WITH POLAROGRAPHIC TECHNIQUES FOR THE DETERMINATION OF SODIUM DIETHYLDITHIOCARBAMATE Currents in pA A I Concentration of sodium diethyl- dithiocarbamate, 2 x 10-8 4 x 10-8 6 x 8 x 10-8 M 1 x 10-7 2 x 10-7 4 x 10-7 s x 10-7 6 x 1 x 10-8 2 x 10-6 4 x 10-6 6 x 8 x 10-6 1 x 10-5 Linear Alternating potential- current sweep chrono- polarography amperometry" - - - - - - - - - - - - - - - - - - - - 0-03 0.05 0.07 0.11 0.10 0.16 0.14 0-22 0.17 0.27 * Cathodic voltage sweep.t 5 minutes' electrolysis. $ 2 minutes' electrolysis. 5 1 minute's electrolysis. Cathodic stripping Stirred R o t a t ; n g solution electrode - 0.377 - 0.567 - 0.787 0-77t 0*56$ 2-53? 147$ 2.78t 2*38$ 3-78? 2-88: 0.943 - 1.948 - 2-78s - 4.063 - 4.735 I - 0.207 - 0.987 1-47t l.lS$502 BRAND AND FLEET: APPLICATION OF POLAROGRAPHY TO THE [Analyst, Vol. 93 value for wave 11, 88 mV (Fig. 2). Theory predicts that the width of an a.c. wave at half- height is 90*5/n mV for low frequencies.13 However, with the apparatus used only the com- ponent of the a.c. current in phase with the applied a.c. voltage is rectified and recorded as the a.c. wave. While this results in an increase in sensitivity by eliminating purely capacita- tive currents, some distortion of the wave shape is inevitable because most faradaic processes have phase angles between 0" and 45".I t is therefore not possible to derive n values from this evidence. It is clear, however, that the two processes corresponding to waves I and I1 both show large in-phase components of the a.c. current, hence both must involve charge- transfer reactions. Potential Fig. 2. Cathode-ray' polarogram, 6 x M sodium diethyldithiocarbamate. Ano- Sensitivity 7.5 pA f.s.d. dic voltage sweep. Starting voltage -0.7 volt versus S.C.E. LINEAR POTENTIAL-SWEEP CHRONO-AMPEROMETRY AT A DROPPING-MERCURY ELECTRODE- Anodic voltage sweeps (from negative to positive potentials) showed two peaks in the concentration range 10-3 to 10-4 M (Fig.2). The height of the more negative peak (I) was concentration independent, while peak I1 showed a rectilinear concentration dependence. Thus peak I clearly corresponds to the d.c. polarographic pre-wave and peak I1 to thed.c. diffusion wave. Potential Fig. 3. Cathode-ray polarogram, 5 x 10" M sodium diethyldithiocarbamate. Cathodic voItage sweep. Sen- sitivity 30 p A f.s.d. Starting voltage -0.3 volt versus S.C.E.August, 19681 DETERMINATION OF SODIUM DIETHYLDITHIOCARBAMATE 503 Cathodic voltage sweeps (from positive to negative potentials) also showed the same two peaks, but the peak currents obtained were several times greater (Fig. 3). At concen- trations above 6 x 10" M, peak I (the concentration-independent pre-wave) was obscured by the larger peak 11, which required a reduction in the sensitivity control of the instrument to accommodate it.This increase in sensitivity was caused by pre-concentration of the insoluble mercury salt on the drop surface during the 5-second delay period. At concentrations below lo-* M peak I1 disappeared and peak I became concentration dependent (Table I) over the range lo4 to M. The consecutive use of the two peaks obtained by cathodic voltage sweeps enabled a rectilinear i, - concentration dependence to be established over the con- centration range 1 x 10-6 to 6 x lo-* M. The potential-sweep synchronisation circuit of the polarograph requires the use of a dropping-mercury electrode with a drop time of between 6.5 and 7.0 seconds. In the presence of sodium diethyldithiocarbamate the capillary was found to drop erratically, presumably because of the deposition of insoluble material in and around the capillary orifice.This problem was overcome by the use of forced-drop detachment. With the Davis differential cathode-ray polarograph the potential sweep is synchronised with mechanical detachment of the drop. At concentrations below M peak heights were not reproducible. However, modification of the instrument to allow the use of twin cells with external reference electrodes would result in a decrease in the lower concentration limit. CYCLIC VOLTAMMETRY- A cyclic voltammogram of sodium diethyldithiocarbamate in alkaline solution is shown in Fig. 4. A single oxidation- reduction peak is observed at the sweep rates used. The enhancement in peak current during the cathodic sweep compared with the anodic sweep is clearly illustrated.Whenever the electrode potential is more positive than that of the cathodic peak (0.4 volt 'uwsus S.C.E.), the insoluble mercury - diethyldithiocarbamate com- pound is formed at the electrode surface. This suggested that cathodic stripping analysis is applicable to the problem. In this technique the electrode is held for a fixed time at a potential a t which the insoluble mercury salt is formed, and then a cathodic voltage sweep is applied; the resulting current is recorded as a stripping peak, the height of which is proportional to concentration. Cyclic voltammetry, which allows the direct comparison of the oxidation and reduction processes, was used to determine the potential range available for pre-electrolysis and the optimum sweep rate.Volts versus S.C.E. Fig. 4. Cyclic voltammogram, 1 x lo-* M sodium diethyldithiocarbamate. Single cycle, initial scan anodic. Scan rate 25 mVs-1504 BRAND AND FLEET: APPLICATION OF POLAROGRAPHY TO THE [Auta&St, VOl. 93 The observed separation of anodic and cathodic peak potentials (l40mV) is greater than the expected value of about 60 mV for a reversible one-electron transfer at comparatively slow sweep rates.14 About 20 to 30mV of this separation arises through backlash in the polarograph gears. The displacement is still significantly different from the ideal case, and suggests that an activation energy is associated with the stripping process. CATHODIC STRIPPING ANALYSIS- There has been considerable work on the technique of anodic stripping15~leJ7 but relatively little work has been done on the converse technique of cathodic stripping.It is not generally realised that cathodic stripping does not suffer from many of the complicating factors encountered in the analogous anodic technique. As the electrolysis is carried out at relatively positive potentials, fewer impurities are plated out. Moreover, the cathodic stripping of an insoluble film from a mercury surface gives an improved peak definition when compared with anodic stripping in which slow diffusion from the bulk of the electrode causes tailing. Two types of electrode system were studied. Firstly a hanging mercury drop electrode was investigated and subsequently a mercury-coated platinum electrode.HANGING MERCURY DROP ELECTRODE- The stripping curve, (a), obtained with this type of electrode after pre-electrolysis in unstirred solution is shown in Fig. 5 . Peak I corresponds to the pre-wave and peak I1 to the diffusion controlled wave in Fig. 1. A random dependence of peak current versus elec- trolysis time for peak I1 was found, while the height of peak I remained constant. When the solution was stirred during the pre-electrolysis an increase in sensitivity was obtained and a rectilinear peak current - electrolysis time relationship was found. Under these conditions the pre-wave was not clearly defined, suggesting that no discrete monolayer was formed as shown in (b) in Fig. 5. The magnitude of the stripping peak showed a rectilinear concen- tration dependence over the range 5 x lo-' to 10" M.Volts versus S.C.E. Fig. 5. Stripping peaks at hanging mercury drop electrode, 2 x 1 0 - 6 ~ sodium diethyldithiocarbamate; 4 minutes' electrolysis at 0.0 volt vemysus S.C.E. Scan rate 25 mVs-1: (a) unstirred solution; (b) stirred solution MERCURY-COATED PLATINUM ELECTRODE- The two major disadvantages of the hanging mercury drop electrode, i.e., its extreme sensitivity to vibration and the difficulties associated with reproducibility of the drop size are eliminated by using a mercury-coated platinum electrode. Many of the practical difficulties experienced by early workers in the use of this electrode have recently been overcome by Joyce and Westcott.11 In the present work initial experiments were performed with a stationary electrode in a stirred solution.Results indicated that for a given depolariser concentration, in order to minimise the electrolysis time it was necessary to increase theAugust , 19681 DETERMINATION OF SODIUM DIETHYLDITHIOCARBAMATE 505 rate of stirring until just before turbulence occurred. Under these conditions well developed stripping peaks, (a) in Fig. 6, were obtained after 2 minutes’ electrolysis for concentrations down to 10-6 M. At concentrations below this level it was necessary to extend the electrolysis period to 5 minutes to allow the use of a sufficiently low current sensitivity to eliminate instrument noise; under these conditions there was a slight decrease in precision. This problem was overcome by rotating the electrode, at constant speed, during the pre-electrolysis instead of stirring the solution.The stripping peaks, (b), obtained after 2 minutes’ pre- electrolysis at a rotating electrode were better defined, as shown in Fig. 6(b), and allowed the lower concentration level to be extended to lo-’ M without loss of precision. The deter- mination of lower concentrations was possible but it was preferable to increase the electrolysis time rather than the instrument sensitivity. 2 4 t V c 2 (3 I 0 -0.2 -0.4 -0.6 0 -0.2 -0.4 -0.6 Volts versus S.C.E. Fig. 6. Stripping peaks at mercury-coated platinum electrode, 8 x lo-’ M sodium diethyl- dithiocarbamate; 2 minutes’ electrolysis at 0.0 volt versus S.C.E. Scan rate 100 mVs-l: (a) stirred solution; (b) rotated electrode CONCLUSIONS The results of the present investigation have shown that of the techniques examined, cathodic stripping proved the most suitable for the determination of trace amounts of sodium diethyldithiocarbamate.Alternating current polarography and potential-sweep chrono- amperometry were both applicable to a wide range of concentrations but in both cases the lower limit was 10-6 M. The sensitivity of cathodic stripping was limited only by instrument noise, and it is evident that with improved instrumentation by using electronic potential sweeps and oscilloscopic read-out a much lower detection limit could be achieved. A rotating mercury-coated platinum electrode was found to be the most suitable for cathodic stripping. Its sensitivity and reproducibility were far superior to the hanging mercury drop electrode; in addition it was not affected either by the rate of stirring of the solution or by vibration.Although the present work has been confined to mercury electrodes, it should be noted that certain types of solid electrode, e.g., pyrolytic graphite and carbon paste, might offer some advantages in that the electrode process will involve a true oxidation, rather than mercury salt formation. We wish to thank the Agricultural Research Council for the provision of a research assistantship to one of us (M. J. D. B.). We would also like to thank Morganite Research and Development Limited for the loan of the KlOOO cathode-ray polarograph, and Dr. D. A. Pantony, Metallurgy Department, Imperial College, for the use of the Davis differential cat hode-ray polarograph.506 BRAND AND FLEET REFERENCES Thorn, G.D., and Ludwig, R. A., “The Dithiocarbamates and Related Compounds,” Elsevier Publishing Company, Amsterdam, 1962, Chapter 10. Lowen, W. K., and Pease, H. L., in Zweig, G., Editor, “Analytical Methods for Pesticides, Plant Growth Regulators and Food Additives,” Academic Press, New York and London, Volume 111, 1963, p. 69. Gajan, R. J., in Gunther, F. A., Editor, “Residue Reviews,” Springer-Verlag, Berlin, Gottingen and Heidelberg, Volume V, 1964, p. 80; Volume VI, 1964, p. 77. Nangniot. P., Bull. Inst. Agron. Stns Rech. Gembloux, 1960, 28, 365. Halls, D. J., Townshend, A., and Zuman, P., Analyst, 1968, 93, 219. Armstrong, R. D., and Fleischman, M., J . Polurogr. Soc., 1965, 11, 31. Vlcek, A. A., Colln Czech. Chem. Commun., Engl. Edn, 1954, 19, 221. Biegler, T., J . Electroanal. Chem., 1964, 6, 357, 365 and 373. Brezina, M., and Zuman, P., “Polarography in Medicine, Biochemistry and Pharmacy,” Inter- science Publishers Inc., New York and London, 1958, p. 460. Joyce, R. J., and Westcott, C. C., “Symposium on Trace Characterisation,” National Bureau of Standards, Washington, 1966. Smith, D. E., in Bard, A. J., Editor, “Electroanalytical Chemistry,’’ Arnold Ltd., London, Volume I, 1966, p. 102. Breyer, B., and Bauer, H. H., “Alternating Current Polarography and Tensammetry,” Interscience Publishers Inc., New York, 1963, p. 50. Matsuda, H., and Ayabe, Y., 2. Electrochem., 1955, 59, 494. Shain, L., in Kolthoff, I. M., and Elving, P. J., Editors, “Treatise on Analytical Chemistry,” Interscience Publishers Inc., New York, 1963, Part I, Volume 4, p. 2533. Kemula, W., and Kublik, Z., in Reilley, C. N., Editor, “Advances in Analytical Chemistry and Instrumentation,” Interscience Publishers Inc., London, New York and Melbourne, Volume 111, 1963, p. 123. Barendrecht, E., in Bard, A. J ., Editor, “Electroanalytical Chemistry,’’ Arnold Ltd., London, Volume 11, 1967, p. 53. Received December 8th, 1967 -, Ibid., 1960,28, 313. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.