- Email: [email protected]
Polarography of some sulphur-containing compounds
Cltfmlca Ada Elswfer PubliihlngCompany,Amrtcdam Rwtytica
459
Prhtcd (nTttc Ncthcrkmds POLAROGRAPHY PART THEIR
XXI’. ANODIC REACTIONS
OF
SOME
SULPHUR-CONTAINING
COiMPOUNDS
WAVES OF MONOALKYLDl[THIOCARBAMATES WITH HEAVY Ml2TAL.S
AND
Analytical polarographic methods are commonly based on waves corresponding to reduction or oxidation of the clectroactive particle, but anodic waves which correspond to the formation of a mercury compound at the surface of the dropping mercury electrode are also useful l-3. In connection with csrlier work on the anodic waves of sulphur compounds4 as well as studies on monothiocarbamates and dithiocarbamatesb and with respect to the application of this group of compounds as behaviour of some monoalkyldithiocarbamates was pest icidesa, the polarographic studied. As in other practical applications of polarograplly, it is essential to elucidate the course of the electrode process, at least in principle, because only then is it possible to anticipate the role of other components of the supporting electrolyte. A fundamental study was very important in the present case, because the effects of adsorption and the composition of the compounds formed depend on the reaction medium, The types of process governing the various waves and the chemical reactions involved were therefore studied, As the anodic wave corresponds to a chemical reaction between a particular form of the dithiocarbamatc and mercury ions, it was of interest to compare information on the structure of the mercury compound from polarugraphic data with that obtained for reactions with heavy metals, Reports on the polarography of monoalkyldithiocarbamates are rather limited in number and scope; only dithiocarbamates derived from amino acids have been studied74 At high PH values, the compounds give a single wave, which is split into two waves below PH IO. Based only on logarithmic analysis of these waves it was assumed’*8 that the anodic wave corresponds to a one-electron process. Some of the features of the polarographic waves of monoalkyldithiocarbamates and the possibility of using these waves for analytical purposes have been in.dicated in a preliminary report? EXPERIMENTAL
Polarographic *
Part
Xf:
Co&xztion
curves Czech.
Chenr.
were Commwt.,
recorded 22
using
a Pofariter
PO4
(Radiometer,
(1957) QZQ,
On leave from J. Hcyrovuky Xnstitutc of Polarography, Czcchoslovak Academy of Scicnccs, Prihgue. **
AmIf. Claim. ACta, 40 (1968) 459-472
D. J. HALLS, A. TOWIVSHEXD, P. ZWMAN
460
Copenhagen) in conjunction with a Kalougek vessel with a saturated calomel electrode. The dropping electrode used had the following characteristics: out-flow velocity 3;fz=x.59 mg SCC-1, drop-time 11=3.6 see at the potential of the saturated calomel electrode and at a mercury pressure It =68 cm in 0.1 IM potassium cliloride. Controlled-potential electrolysis wlas carried out with a dropping mercury electrode in o+-2.0 ml of solution stirred by the falling drops. A vessel designed by MANOU&K*” with a reference electrode separated by sintcred glass and an agar bridge saturated with potassium nitrate and with a device to keep the level of the fallen mercury constant was used, The potential during electrolysis was kept constant manually, by means of a potentiometer. A Wnicum SP 800 spcctrophotometcr was used to record the spectra. A PyeDynacap pH meter was used with a glass electrode to measure PH. The pH values of the solutions were checked after the polarogrnphic electrolysis had been carried out.
Monoalkyldithiocarbamates I-III (R-NHCSS-1 Na+; I, R=C&It-; IX, R= XII, R =C~HU-) Were prepared by the following procedure. To 0.1 mole of the appropriate amine was adder1 6 E: of sodium hydroxide (o.15 mole) dissolved in the minimum amount of water. The reaction mixture was cooled and 7 ml of carbon disulphide (0. I:I mole) Was acldcd slowly. The solution was heatccl under vacuum to remove most of the water. Acetone was added to extract the dithiocarbamate. To the filtered acetone extract, ether was added to crystallize out the clithiocarbarnate, which was then filtered off, redissolved in acetone or ethanol and again precipitated with ether, These compounds decompose slowly on standing and re-crystallisation was necessary once a fortnight. The water of crystallisation was determined by loss of weight at 78* z’n 21&~2t0.Each compound was a &hydrate. Melting points were: I, gg”; IX, 88”; III, 64-&Y (dccomp.), uncorrected. Aqueous 0.01: M stock solutions of sodium monoalkyldithiocarbamates made up in distilled water, borate buffer (PH 9.3) or 0.01: M sodium hydroxide showed about 10% decomposition after storage at z” for 3 days, This reaction proceeded especially quickly in 0.01 IM sodium hydroxide solution where a white precipitate was formed, probably a disulphide resulting from autosidstion. Hence, small quantities of stock solution in distilled water were prepared daily. Buffer solutions and all other supporting electrolytes were’ prepared from AnalaR-grade chemicals, CHa-;
In polarographic studies, IO ml of the buffer or other supporting electrolyte was cleoxygcnated by a stream of nitrogen, then the appropriate volume (usually 0.5 ml) of the stock solution was added, and, after a brief flush with nitrogen, the current-voltage curve was recorded. For logarithmic analysis, polaropaphic curves were recorded automatically at n slow rate of scanning (xoo mV/min) to reduce the effects of hysteresis of the recording instrument, To study the reaction of dithiocarbamates with metal ions, various amounts of a stock solution (0.004 M) of the metal salt were successively added to buffered n9d.
Chit&
ACta,
40
(ICJfi8)
4SI)-d(72
POLAROGRAPHY
OF MONOALKYLDITHIOCARBA~~AT~S
46x
4.85 010-4 M dithiocarbamate solutions and the i-E curves wsre recorded after each addition. For the millicoulometric determination of the number of electrons transferred (n), o,s-2.0 ml of the solution of the examined compound in a suitable supporting electrolyte was transferred to the cell10 and the capillary was positioned so that its tip was just under the surface of the solution. The solution was deoxygenatcd, the i-E curve recorded and a potential (usually corresponding to the limiting current) selected. The solution was discarded and replaced by the same quantity of a fresh solution. A voltage corresponding to the chosen potential was taken off.‘the potentiometer and the change of current with time at this potential was recorded at a slow chart speed. After IO-ZO~/~ conversion had been achieved by electrolysis (which required 30-60 min), the electrolysed solution was replaced by the supporting electrolyte alone, in which the condenser current was recorded in order to correct the measured limiting current. The logarithm of the current was plotted as a function of time and from the slope (d logildl) of the resulting straight line the value of n was calculated from: tt-
(4 . 5 *IO’@ (ia)o}/{c~
(d log i/at) )
and z, the volume where (id)0 is the limiting current at t--0, c the molar concentration (in ml), The method was tested by determining s for thallium(I) and cadmium(I1). The average reproducibility was better than =f=5. For the identification of the electrolysis products the same basic procedure was followed, but instead of the continuous i-l record, polarographic i-E curves were recorded at intervals. After electrolysis for 24 h, polarographic curves and U.V. spectra were recorcled. RESULTS
Because the polarographic behaviour of afl straight-chain monoalkyldithiocarbamates studied was found to be similar, it was considered appropriate to choose monoethyldithiocarbamate (I) as a representative compound for a more detailed study. Effect of fin. The ethyl derivative (I) shows in hydroxide solutions one welldeveloped anodic wave (i) (Fig. I). At concentrations above 6 -1~0-4M this is accompanied by an adsorption pre-wave in (Fig. z) at a potentia.l only slightly different from wave i. Below PH xx the wave i is accompanied by another adsorption pre-wave ia which is observed at concentrations greater than I *IO-* M, Above PH IX this prewave merges with the wave i. Below phi: 8, the main wave (i) is split into two waves ([it +&J and &), the ratio of which remains constant and is practically I: : x, The sum of the waves (&f-iz) is constant over the entire PH range studied and is equal to the height of the wave i in hydroxide solutions. In aciclic solutions the more positive wave & is indistinct because of its closeness to the current corresponding to the dissolution of the mercury in the supporting electrolyte. Wave i2 is also complicated by adsorption effects. The separation of waves ix and ia: is best understood from the ps-dependence of the half-wave potentials (Fig. 3). The half-wave potential of wave Z’Zshifts with Anal.
Claim, Ada,
40 (1968)
459-472
D. J.
HALLS,
A. TOWNSHEND,
P. ZUMAN
* 1 ’ c . J. A!‘._; ‘9L a ,u. ‘; U” L- . _ . 3: Pig. I. 2%~ cffcct of ptt on the WUVCY of yodi~m.cLthyi~ithiocnrbnmatc. 5. m-4 M YOCKU~ cihj;clithiocarb(rmuto, f3ritton-Robinson buffers (prt shown on polarogrrrm). Curves rccorclcd from - t .o V towrrrd~ more positive potcntialu, Ir - 08 cm. .Y
I
Fig. 2. The effect of concentration on the WQVCSof sodium c?thyIditfriocarbam~tc, o, T M NnOH concentration of sodium cthylclithiocclrbatnatc: (I) o; (2) 2.0; (3) 3.9; (4) 5.8; (3) 9.9; (6) x1.5 ( 010-4 M). Curves rccorcled from - x.0 V towards more positive potcntirrls, k = 50 cm,
increasing PH towards more negative potentials with a slope of 0.067 mV/pH, whereas that of wave z”lremains practically constant, When the half-wave potential of wave ia approaches -0.3 V at which wave $1appears, both waves merge into wave i, the half-wave potential of which is shifted by 0.057 mV/pH to more negative potentials in a similar manner to wave ia. The half-wave potential of wave i. changes with PH only below PH 3 (Fig. 3), and at higher PH values is practically pw-independent, This causes waves iu and i to merge above pi XL A difference in shifts of halfwave potential can also explain why the wave in is not separated from the total wave i below pw 13. Anal. C~liWLACta, 30 (19683)
459-472
POLAROGRAPHY
463
OF n-IOXOALlKYLlXTHlCOCARBAMATES
Whereas the height of the adsorption pre-wave i. increases with increasing concentration only up to about I *IO -4 M and then becomes constant, the sum of (ia+&) ancl of (i,+il+&) are both linearly proportional to concentration (Fig. 4a) -4 M; this was proved in acetate buffer, pi 4.7, and up to a concentration at 8.~0 phosphate buffer, PH 6.9.
Fig. 3. The p;r clepuntlcncc of the half-wave potentials mate. 0 = Britton-Robinson buffers (see Fig. I). ()
of the W~VCY of sodium f-i&C& solutions.
cthytdithiocarba-
=
a
0
Fig, 4. The dcpcndcncc of the height of the wavc~ (a) Phosphate buffer ptr G.9, (b) 0.1 M N;aOW,
2
4
6
of cthytdithiocarbamatc
8
10
on concentration.
The height of wave in limits at about 6.10 -4 M (Fig. 4b) whereas the total height of waves i~+i in 0.x M sodium hydroxide solution is a linear function of concentration of ethyl derivative I up to x.2 x0-3 M. The half-wave potential of the wave i is shifted to more positive vatues with increasing concentration of the dithiocarbamate I. A similar direction of shifts was observed for other compounds3 forming slightly soluble mercury compounds. With increasing concentration of dithiocarbamate a decrease of the anodic current due to reaction of hydroxyl ions with mercury was observed. In OJ MT sodium hydroxide solution, 5.8 10 -4 M in dithiocarbamate I (curve 4, Fig. z), of the hydroxyl wave is similar in height to wave ir ; at still higher concentrations l
l
Anal.
Cirim. Acta,
40 (xgG8) 459-472
D. J. HALLS,
464
A. TOWNSHEND,
P. ZUMAN
the ethyl derivative (I) no wave for hydroxyi ions was observed (Fig. 5). In 0.01~M sodium hydroxide solution the hydroxyl ion wave disappears at a lower dithiocarhamate concentration, Logarithmic analysis of the wave of I in 540-4 M solution in a &ittonRobinson buffer (PH x0.8) gave a reasonably straight line for the plot of log[(i~ &)/i’j against potential, with a slope of -0.050 V,
Fig. 5. The effect of tllc prcscnce of cttl,yIditlliocnrbrrm~teon the tlnock wtzvc due to hydroxide ions. IO ml 0.x M N&OH. Concentration of cthyldithiocnrbamntc: (x) o; (3) 1.0; (3) 2.0; (4) 0.9; (5) 3.8: (6) 4.8: (7) 5*7; (8) 6.6 (‘10 -4 fW), Curves rccordcd from -0.7 V towrrrclsmore poaitivc potcntinlu (except curve x ; from -0.3 V), li =LZI 50 cm.
The effect of adsorption on the wave i U was proved at concentrations greater than 2 *x0- * M by the linear proportionality of the wave heigflt (i,,) to the height of the mercury column, The linear dependence of (i,+&) and (i,+il +i~) on the square root of ttlc height of the mercury column indicates that these processes are diffusion-controlled, No reduction waves were obtained over the whole PH range (o-x3) studied. Below PH 7 a shift of the current corresponding to hydrogen evolution from the buffer towards more positive potentials indicated that catalytic hydrogen evolution was taking place. However, this current was not separated from the current of the supporting electrolyte and no developed reduction wave was observed in any of the solutions studied. N~ntier of eZ&rom transfemd Comparison of the anodic waves with waves of benzophenone and of nitrobcnzene at PH 9.2 (borate buffer) indicated that the wave 2’in alkaline solutions corresponds to a two-electron process and that the waves ir and is at lower pH values correspond to two successive one-electron uptakes. A comparison of the hci&ts of tfle waves of compounds I and II with the waves of diethyldithiocarbamate in x.95 xo- 4 M solutions at PH 9.2 showed ratios of x.97 and 2.22 respectively. As the total height of the waves of diethyldithiocarbamate corresponds to a one-electron process 11, the transfer of two electrons for monoalkyldithiocarbamates upas verified, l
AMar. Chim. Acta, 40 (rgG8)
459-472
POLAROGRAPHY OF MONOALKYLDfTHSOCARBAMATES
465
Millicoulometrically a value of .tt~~1.87 was found with a 5. x0-4 M solution of the ;rt-butyl derivative (III) in an acetate buffer (PH 5.7) ancln =x.89 with a I *IQ-~ 2M solution of compound III in a phosphate buffer (pr-r 6.9). The millicoulometric determination of $2at high pi values was prevented by the consecutive reaction of the electrolysis product (p. 466). The values found at the limiting current of wave $1 at x ‘10-3 jW were tt =0.70 at 5*x0 -4 M in an acetate buffer (PH 4.7) and n=o,gr in a phosphate buffer (PH 6.9). Str~$urat eSfccts. The behaviour of the methyl (II) and butyl (III) derivatives was completely analogous to that of cthyldithiocarbamate (I), The presence of waves * zo, il, iz and i was proved in all cases. The character of the adsorption pre-waves was analogous, the ratio (in $-it) 52 remaining practically I: : f in all cases. Tile structural changes in this limited group had little effect on the half-wave potential. In 0.1 iW sodium hydroxide solution the approximate values of the half-wave potentials -0.49 V for ethyl (I) and -0~50 V for n-butyl (III) (II), ( -0.47 V for methyl dithiocarbamate, all vs. S.C.E.) are in sequence and show the differences that would correspond to a modified Taft equation AE+= eX,i~* ctX* for reaction constant &%o.z V. It is interesting that even in this case- as for some other anodic processes -the value of the reaction constant is positive.
Controlled-potential electrolysis of solutions of compound III at the potential of the limiting currents of waves il or iz at lower PH[ values resulted in a regular decrease of the limiting- current with time during the electrolysis. No precipitate was observed when the electrolysis was carried out at the limiting current of wave il. On the contrary, in 0.1: &# sodium hydroxide solution, when electrolysis was
Fig. 6. Change of polarographic curves with time during controlled-potcntiat clcctrolysis of sodium n-butyfdithiocarbamatc (I IO- 3 M) in 0.1 N NaOH. Volume of solution - 2.0 ml; tirno of the start of recording of tho curve: (r) o; (2) 55 min; (3) 7 h 25 min; (4) 17 h. Applied potontinf = -0.3 V corresponds to tho limiting current (arrow). Curves rccordcd from - 1.0 V towarcls mart positive potentials I l
Anal. Chim, Ada,
40 (rgG8) 45g-472
D. J. HALLS,
466
A. TOWNSHEXD,
P. ZUMAN
out at the potentiaf corresponding to the limiting current of wave i of compound III, the height of wave i remained practically constant for several hours and only then clecreasecl (Fig. 6). During the electrolysis, a netiv wave i’ appeared at a potential about o.xg V more positive than wave i. The relative height of this wave increased during the electrolysis (compare ‘curves 3 and 4, Fig. 6). A solution of the butyl derivative (III) kept under the same conditions, but without applying the potential, showed that cleavage of III by a chemical reaction in the bulk of solution during electrolysis was insif;nificant. A black precipitate formed during the electrolysis was proved to be mercury sulphide by the action of acid. The electrolysed solution liberated tile characteristic odour of an isothiocyanate. In the solution after electrolysis for 24 11,where the total height of the waves was still about zoo/, of the original height, no ultraviolet bands due to dithiocarbamatc were observed (Fig. 7). Calculation has shown that x0/0of the original amount of compound (III) would still be apparent from the U.V. spectrum. The “end absorption” of the electrolysed solution at 228 nm (curve B, Fig: 7) was shifted compared with sodium hydroxide (2x5 nm, curve C, Fig. 7). Similar end absorption was observed in the presence of carbonate (curve D, Fig. 7). When the efectrolysed solution was acidified the end slbsorption was shifted to a shorter wavelength and a bancl at 240 nm was apparent (curve E, Fig. 7). Similarly, during electrolysis in a borate buffer (PH 9.2) it bfack precipitate was formed and a shoulder was present in the U.V. spectrum at 240 nm. Formation of carbon disulphide WZB demonstrated by the presence of a small band at 320 nm and of a polarographic reduction wave at - IJ V, wbiclr diminislled after bubbling nitrogen through the solution. Rcaclions wilh m-W iotts. Addition of cadmium ions to tile solution of sodium ethyldithiocarbamate (I) in 0.x M sodium liydroxidc solution resulted in a decrease
carried
Fig. 7. U.V. spcctta of clcctrolysis producta. Controftcdwpotcntirrlckctrolysis of x I o-3 M sodium tr-butyldithiacarbamato carriectout for 24 hat --al3 Vin 0.1: M sodium hydroxide. (A) x.25. x0-4 M wbutyidithiocarbamato in o.012 M sodium hydroxide bcfare olcctrolysis (diluted 8 x ) ; (X3) solution after clcctralysis (diluted 2 x ) : (C) 0.05 M sodium hydroxide (diluted 2 x ); (D) soln. as in (C) in prescncc of sodium carbonate ; (E) soln. as in (B) ncidificd by hyclrochlaric acitl, l
Fig. 8. Addition of EL solution of cadmium ions to sodium cthyldithiocarbzrmutc. To ro mt of 5. x0-4 N sudium cthytdit~tiocnrbumrrtoin 0.x M sodium hydroxide 4. x0-3 M cndn~iutn aulphatc was successively added. Currents corrcctcd for dilution. (f) Height of anodic wave (i) of dithiocnrd bamstc; (2) totat height of nnoclic WQVCS. A?&.
Cki?% tl C:iZ,40 (qG8) 459-472
POLAROGRAPHY
OF MONOALKYLDITH10CARBAhlATES
467
of the anodic wave of ethyldithiocarbamate. A white precipitate appeared initially, but when the ratio of cadmium(Il) :dithiocarbamate exceeded I: : 2, the precipitate turned yellow if the PH of the reaction mixture was sufficiently high. Simultaneously a new, small anodic wave formed at more positive potentials (at about -0.3 V). When the ratio of cadmium(H) : dithiocarbamate reached x :I, the slope of the dependence of the first wave on cadmium concentration changed (Fig. 8). With further increases in cadmium concentration the total height of the anodic waves decreased and the anodic waves vanished when the concentration ratio was about z Cdz+: I: dithiocarbamate. No cathodic waves corresponding to soluble complexes were observed. This behaviour can be tentatively explained by the assumption that the white precipitate initially formed is the compound Cd(dithiocarbamate)z. The yellow precipitate is attributed to cadmium sulphide, formed from the compound Cd(dithiocarbamate)z and further cadmium ions. It is assumed that in this reaction the cthylmonothiocarbamate ion is formed, which is responsible for the formation of the anodic wave12 at -0.3 V. Further addition of cadmium ions results in the cleavage of the . monothiocarbamatc with further formation of cadmium suIphidc. This interpretation seems to be confirmed by the result of a large-scale experiment, which showed tflat 50 IO -5 moles of sodium ethyldithiocarbamate react with r .f x0 -4 moles of cadmium ions, In this reaction the formation of 0,94 ‘x0-4 moles of cadmium sulphide was proved gravimetrically. This yellow precipitate, when decomposed by hydrochloric acid, liberated hydrogen sulphide. fn the solution thus formed, x.0x. xo -4 Ad’ cadmium(I1) was found polarographically. The other reaction products were not identified. In attempts to isolate the product assumed to be the x : I complex, the existence of which was indicated by the intersection in Fig. 7, solutions of cadmium(I1) (I 0x0-3 moles) and dithiocarbamate (I GO -3 moles) were reacted in the presence of Z*IO-3 moles of sodium hydroxide. When the white precipitate was extracted with chloroform, yellow cadmium sulphide began to precipitate rapidly from the chloroform layer. The UV. spectrum of the compound remaining in the chloroform layer showed a band at 235 nm, indicating the presence of isothiocyanatcs? Their presence was supported by the pungent odour of the residue after chloroform had been distilled off. Similar results were observed when mercury{ II) ions were added to the solution of ethyldithiocarbamate ions in o. r n/rsodium hydroxide solutions. A white precipitate (attributed to the dithiocarbamate-mercury compound) was formed initially. On further ‘addition of mercury(H) this precipitate turned black, owing to mercury(H) sulphide formation. The decrease in the anodic waves with increasing metal ion concentration was analogous to that found for cadmium, except that the anodic wave decreased practically to zero when the ratio mercury(I.1) :dithiocarbamate was about 3:~. Somewhat different results were observed when lead ions were added to a solution of ethyldithiocarbamate in a borate buffer (PH 9.3). A white precipitate was formed and the anodic wave of ethyldithiocarbamate decreased practically to zero when the ratio of lead(Il) :dithiocarbamate reached I :z. The composition of the white precipitate is therefore Pb(dithiocarbamate)z, Addition of further lead ions slowly turned the precipitate black. l
Anal.
C&n.
Acta, 40 (x968)
45g-472
468
D.
J.
HALLS,
A.
TOWNSHEND,
P. ZUMAX
1)ISCUSSION The two one-electron diffusion-controlled and in can be formulated by scheme (A)-(D): RNHCSS - F= \ ., PKy H+ RNHCSS
RNHCSS =zi+
RNHCSS
- + Hg w 1s1
- +H+
RNHCSS
processes corresponding
to waves il
(A)
-1-c
(r-3)
N$o
S+H+ -6 R-NN=C/ \S_ Pf(Z H&l)
(D) The wave ir at potential El corresponds to processes (A) and (B), and wave i2 at potential 5’2 to processes (C) and (D). The shift of the half-wave potential of wave & indicates that the value of p& for the ethyl derivative I is about 3.5. This is in good agreement witlt the plc value of 3.ro for monocyclohcxyl and 2.95 for isopropyl dithiocarbamates obtained6 from kinetic data. The ]PH independence of the half-wave potential of wave z’t. above pw 4 indicates that the electroactive form is identical with the form of dithiwarbamate predominating in the solution. The shift of the potential of wave in indicates that the value of p& is greater than 12, This is consistent with the alkaline decomposition of monoalkyldithiocarbamates. It is assumed14 that this reaction t&es place with the formation of R-N=CSS=-, and that it can be observed only at PH about 14. The pKa value of the uncornplexed species RNHCSSis therefore greater than x4. Even when a shift toward smaller p& values can be expected for the complex species in reaction (C), the effect seemingly is not sufficient to shift it below 12. The slopes of the pH dependences of the half-wave potentiaXs of wave ix below pw 3 and of wave i2 over the whole pH range are consistent with the values expected for a reversible one-electron process accompanied by a transfer of one proton. The electrode reactions (A) and (B) corresponding to wave ir are analogous to the reaction of metal salts with monoalkyldithiocarbamates at lower E)H values, as was shown for the reaction of lead ions in a borate buffer. Even though the participation of the nitrogen atom in the mercury bonding cannot be excluded, the bond to two sulphur atoms can be considered as more probable. The actual form of the mercury compound resulting in reaction (D) is not known. The experimental results nevertheless indicate that the compound formed is unstable and undergoes further changes in alkaline media. A possible decomposition scheme for a complex of a metal (M) can be formulated as follows:
POLAROGRAPHY
OF MONOALKYLDITHlOCARBAMATES
469
SR-N=C’ WI)
\S_
+
R-N=C=S
$-MS
(E)
SR-N=C=S+-tH-
+
+H+
R-N--C’
u?
‘OSR-N=C’
+M2+ ‘O-
R-N=C WI)
S’ Lo-
scvornl
&
R-N=& BE(H)
P\O_
w
stcpcr
2 or1-
-j. R-NHz+MSi-COP
(HI)
The formation of metal sulphides, isothiocyanates and carbonates as predicted by this scheme was proved. The scfreme (A)-(F) explains the behaviour in controlled-potential elcctrofysis of clerivative I in 0.x M sodium hydroxide solution, The delay in the decrease of the limiting current can be explained by the cteavagc of the electrolytically formed complex [RNCSS+]M(II) accordingto reactions (E) and (F), In the latter, monoalkyl monothiocarbamate is formed, This compound is known to give anodic wave@; these waves will resemble, both in their heights and potentials, the waves of monoalkylclithiocarbamates and hence it would not be surprising if in a mixture of monoand dithiocarbamates, one wave is formed. During the initial stages of electrolysis ’ the rate of formation of the monothiocarbamate is just sufficient to supply enough depolariser to keep the limiting current practically unchanged. As the conversion of dithiocarbamate into monothiocarbamate becomes more complete, the anodic wave decreases. The rate of the interposed chemical reaction (I?) governs the rate of the controlled-potential electrolysis. Reaction (F), as was shown previouslyrb, is first order in hydroxyl ions. Hence at lower E)H values (PH 6.9) the regeneration of an electroactivc species (with similar electrochemical properties to the original compound) cannot affect the controlledpotential electrolysis, which therefore follows a simple course, The actual course of the decomposition may be more complex than shown in reactions (E)-(H). This is indicated by the formation of carbon disulphide during electrolysis at PH 9.2. It was also not established whether the two waves observed during electrolysis in 0.1: M sodium hydroxide (Fig, 6) both belong to the monothiocarbamate or if one belongs to another intermediate. It was necessary to exclude the possibility that the wave i2 corresponds to mercury sulphide formation rather than to a reaction of RN=CSS? Comparison of the half-wave potentials of wave ia with that of sulphide ions indicates that the sulphide wave is 150 mV more negative. Moreover, the su1phid.e wavelo shows a markedly different PH dependence of half-wave potentials to wave i2. Furthermore, Awl.
Cltim
Acta, 40 (x968) qjg-472
470
D.
J. HALLS,
A.
TOWNSHEND,
P.
ZUMAN
the formation of the white precipitate attributed to [RNCSS]M(II) (even when isolation was impossible) indicated that reaction (E) and the subsequent reactions may bc fast, but, uncler the conditions studied, not so fast as to convert the primary product formed after the transfer of the second electron into sulphide ion during the drop-lift. Hence the process differs from that observed recently’7 for thiourea in aciclic solution. The anoclic waves of monoalkyldithiocarl~amates corresponding to a twoelectron over-all process differ in principle from tlre vast majority of organic sulphur compounds, which give one-clcctron anodic waves 1-R. Among other exceptions the waves of thiourcaiH-2” and of tliiobcnzamidc~~ can be mentioned. Tn the latter case bcnzonitrilc, mcrcury(I1) sulphide and some benxamide were isolated from the controlled-potential clcctrolysis products. For a two-electron process at potential Ez, it is possible, apart from the scheme (A)-(D) corresponding to a system of consecutive reactions, to devise another schcmc, in which tllc reaction at potential L;z would be competitive with that depictccl in ecluations (A) and (R). Such a scheme would involve the formation of mercury(I1) and reaction with the anion of clitlliocarbamatc. Tl~c cllange in the wave-height iz during controlled-potential electrolysis at the limiting current of il cannot be used here as a cliagnostic tool 22 because the formation of a slightly solul~lc salt such as the product of (13) will result in a decrease of wave i:! for both competitive and consecutive reactions. On the other hand, a competitive process requires that the same species reacts with uni- and bivalent mercury ions. This in turn would indicate the same shape of the pH-dependcncc of the half.wave potentials. The d’ffcrent shape of this depenclence for waves il and i:! indicates that a competitive process, at least of the consiclerecl type, is less probable than the consecutive process (A)-(D). Two types of adsorption processes affect the polarographic curves. Wave iu, with a limiting current value of about 0.4pA, corresponds to a surface covered by a single molecule of about 50-100 A3 (depending on the composition of the compound). Wave ir\, observed in sodium hydroxide solution, reaches a limiting value of 5 ,YA and corresponds to a surface of s-so A2 per molecule. If the composition of the mercury compound remains the same, it can be concluded that either the wave it, corresponds to a monolayer and i,+ to a multilayer formation or, perhaps more probably, the wave i R corresponds to a film in which the molecules are oriented flat at the surface and the wave i* to a film in which they are perpendicular and only “anchored” by the functional grouping. Such an observation has been made and a similar interpretation offered for waves of other sulphur compounclszz-20. The effect of wave (in +i) on the waves of hydroxyl ions (Figs. z and 5) indicates that the film of mercury compounds with alkyldithiocarbamates in alkaline media is difficult to penetrate. Because the result% obtained for monoalkyldithiocarbamates are in good general agreement with those reported 7.8 for dithiocarbamoyl carboxylic acid, it seems that the observed behaviour is characteristic for a wide range of compounds with the grouping -NHCSS-. The authors thank Professor R. BELCHEIX for his interest. D. J. H. thanks Professor M. STACEY for the provision of a research grant and I?.%. the Science Research Council for a provision of a Senior Visiting Fellowship. A?sal. Cihrr. Actn, 40 (1gG8)
459-472
POLAROGRAPHY
OF
~fOSOALKYLDITHIOCARRh~~AT~S
471
SUMMARY
Monoalkyldithiocarbamates give two anodic polarographic waves, corresponding to a one-electron and a two-electron process. A reaction scheme is proposed. The unusual behaviour observed during controlled-potential electrolysis in sodium hydroxide media is interpreted as a chemical reaction of the primary clcctrolysis product; the species formed gives a wave similar to the original compound. Different adsorption phenomena are probably caused by varying orientation of the mercury compound on the electrode surface. The adsorbed layer formed in sodium hyclroside solution is so close-packed that it prevents hydrosyl ions from penetrating and thus is the giving an nnodic wave. For analytical purposes, 0.1 M sodium hydroside most suitable supporting electrolyte; linear calibration graphs can be obtained over the range 5 axe-5-I .Io-:a M.
Les rnonoalkyldithiocarbamates donnent clcuzc vagucs l”)larograplliques anodiqucs, correspondant h un et A deux 6lcctrons. Un sch6mn dc la rdaction cst propos8. Le comportcmcnt obscrv6 au tours de l’c%xtrolyse zi potentiel contrOld en milieu hydroxyde de sodium est interprdtb: r6action chimique du procluit cl’&lcctrosont probablcment causes par une lyse primnire. Divers pht5nom&nes d’adsorption orientation variable du composc5 dc mercure sur la surface dc l’blectrode. La couchc adsorb& formc!e dans la solution d’hydrosycle de sodium cst si compactc qu’clle ernpCche la p&n&ration des ions hydroxylcs et fournit ninsi une onde anodiclue. Pour l’analyse, l’hydroxyde clc sodium 0.x hr est l’&xtrolyte de base qui convient le mieux. Des graphiqucs lin6aircs de calibrage sont obtcnus dc s*ro-” ri I.Io-:~ M.
ZUSAMMENFASSUXG
Monoalkyldithiocarbamate ergeben zwci anodisclle polarographische Stufen, welche cinem Einelektronenund einem Zweielcktronenprozess entsprechen. Ein lieaktionschema wird vorgeschlagen. Das Verhaltcn, welches man w2hrend cler Elcktrolyse mit kontrolliertcm Potential im Natriumhyclroxid-Medium beobachtet, wird mit Hilfe einer chernischen Reaktion dcs primiircn Elektrolyseproduktes interpretiert. Die gebildeten Spezies crgebcn tine Stufe, die der der ursprtinglichen Verbindung ~hnlich ist. Unterschiede in Adsorptionsphtinomena kijnncn durch variiercndc Orientierung der Quecksilberverbindung auf der Elektrodenoberfllichc erkl5rt werden. Die adsorbierte Schicht welche i.n Natriumhydroxidliisung gebildet wird, is so dicht, dass sie ein Eindringcn der Hydroxylionen verhindert und so ihre anodische Stufc unterdrtickt. Stir analytische Zwecke ist 0.1 iM NaOH cler am mcisten geeignete Grundelektrolyt. Linearc Eichkurven kijnnen im Bercich von 5 *IO-~ bis I * IO -2 M erhalten werden. REFERENCES I I. M.
I
AND
J. J. LINGANE,
Polarogvnplry,
2nd
Eth.,
Intcrscicncc,
Anal.
Chiwt. Acta,
New York, 40 (rgG8)
1952.
459-472
D. J. HALLS,
472
A. TOWNSHENR,
P. ZUMAN
BkeztN~ AND 1’. HUMAN. Polarograpiby i*r Medicine, Z~iochemistvy and Z%avmacy, Rcviscd Eng. Edn., Intorscioncc, Now York, 1958. J, ~-fP,YIcOVSK+ AND J. I<~ITA, Fvincifiks of ~~drcrogvaphy, Czech. Acaclcmy of Scicnccs, Prague. I9ei. M. ~?EDORO~KO, 0. MhrrouWK AND P. %unfA~, ~~ZJe~~~~7~ Gzecii. C&em. Commtrn., 21 (xc)@) 672, with rcfcrcnccs to previous work. R. ZAIrRADNfu AND I?. ZUMAN, Collection Czcclt, Gfrem. Commtcn., 24 (x959) x 132. Rnalysl, if1 prcpnration. D. J, I-JALLS, A, TOWNS~IIGND AND T‘. %UhfAN, R. ZAIIHADN~K AND L. JI~N!~ovw&, Chcm. L&/y, 48 (x954) IX. R, ZA~IRADN~IC, CAonr. Listy, 49 (1955) x002: Cotclccfion Czech. Chm. Commtrr~., 21 (r9sG) 4,!7. D. J. I-JALLS, PVOC. Sot. /J7ml. cj~??ti., 4 (1967) 40. 0, MAN~U&~C, Private communication. AND 1’. %LlbtAN, PJ?tUt. ChiWZ. /?CfU, in prCS!t. 1). , tiALLS, I\. ‘rO\VNStltZNn V. VASAK, Ctaenr. Listy, 48 (rgg.$) 19. v. ii ~IDIVEC AND 'IX,SVATIIK, R. %AIIIUDN~K AND A. KJAER, .4cC~ Cherrr. Sca7tci., 13 (1959) 442. M. WRONSKI, %cszyfy Nauk. Udw. Lodt., Narrlri Ma&.-Pvzyvod., Ser. If, No. G (1959) 121.
2 M.
3 4 5 G 7 8 9 x0 xx I2
13
14
I 5 IX. ZAIIICADNJIC, Collectio?t Czech. C/rem. Corrt7rzr17r., 24 (1959) 3407. IG A, TlZflrONOV, IOU. Z~~ND, Collecliotr Czech. Claem. Commt~71., 25 (19Go) 2313. 22 T. %UhlAN, in A. STKEITWIIESIJR AND R. w. TAFT, I>ro~,wss irt
Vol. 5. Intcrscicnco, 1967. p. 81. 23 I?, ZUMAN, C/rem, Z:isly, 48 (x954)
24
25 26
I?. ZUMAN AND I’<, ~UnXANOVA, .2x (f9gB) 123. 0, IWANOUSIEK AND P. %U&lAN.
20 (1955)
1340.
21
G78.
M, I:LXDORO$JKO (1956)
AND
P. ZUMAN,
rooG; Collection
Physicat
Commw., 20 (1955) 652 ; co~lcclio~b ~SCCk t%?#L
Czcclr. Clrem.
(t955)
Ovgarric
ChC?dStYy,
046.
Ctlem.
&?Gfy.
49
C&?nt.
Listy,
4g (1955)
G68; Collectiorr
G%x%. Chant. ~ommtor.,
Chcnr.
Lisly,
4g (1955)
x484; Collccliorr
Czcclr. Cllem. Commzrr~. .
CO?#?tlZit~.
,
459
Prhtcd (nTttc Ncthcrkmds POLAROGRAPHY PART THEIR
XXI’. ANODIC REACTIONS
OF
SOME
SULPHUR-CONTAINING
COiMPOUNDS
WAVES OF MONOALKYLDl[THIOCARBAMATES WITH HEAVY Ml2TAL.S
AND
Analytical polarographic methods are commonly based on waves corresponding to reduction or oxidation of the clectroactive particle, but anodic waves which correspond to the formation of a mercury compound at the surface of the dropping mercury electrode are also useful l-3. In connection with csrlier work on the anodic waves of sulphur compounds4 as well as studies on monothiocarbamates and dithiocarbamatesb and with respect to the application of this group of compounds as behaviour of some monoalkyldithiocarbamates was pest icidesa, the polarographic studied. As in other practical applications of polarograplly, it is essential to elucidate the course of the electrode process, at least in principle, because only then is it possible to anticipate the role of other components of the supporting electrolyte. A fundamental study was very important in the present case, because the effects of adsorption and the composition of the compounds formed depend on the reaction medium, The types of process governing the various waves and the chemical reactions involved were therefore studied, As the anodic wave corresponds to a chemical reaction between a particular form of the dithiocarbamatc and mercury ions, it was of interest to compare information on the structure of the mercury compound from polarugraphic data with that obtained for reactions with heavy metals, Reports on the polarography of monoalkyldithiocarbamates are rather limited in number and scope; only dithiocarbamates derived from amino acids have been studied74 At high PH values, the compounds give a single wave, which is split into two waves below PH IO. Based only on logarithmic analysis of these waves it was assumed’*8 that the anodic wave corresponds to a one-electron process. Some of the features of the polarographic waves of monoalkyldithiocarbamates and the possibility of using these waves for analytical purposes have been in.dicated in a preliminary report? EXPERIMENTAL
Polarographic *
Part
Xf:
Co&xztion
curves Czech.
Chenr.
were Commwt.,
recorded 22
using
a Pofariter
PO4
(Radiometer,
(1957) QZQ,
On leave from J. Hcyrovuky Xnstitutc of Polarography, Czcchoslovak Academy of Scicnccs, Prihgue. **
AmIf. Claim. ACta, 40 (1968) 459-472
D. J. HALLS, A. TOWIVSHEXD, P. ZWMAN
460
Copenhagen) in conjunction with a Kalougek vessel with a saturated calomel electrode. The dropping electrode used had the following characteristics: out-flow velocity 3;fz=x.59 mg SCC-1, drop-time 11=3.6 see at the potential of the saturated calomel electrode and at a mercury pressure It =68 cm in 0.1 IM potassium cliloride. Controlled-potential electrolysis wlas carried out with a dropping mercury electrode in o+-2.0 ml of solution stirred by the falling drops. A vessel designed by MANOU&K*” with a reference electrode separated by sintcred glass and an agar bridge saturated with potassium nitrate and with a device to keep the level of the fallen mercury constant was used, The potential during electrolysis was kept constant manually, by means of a potentiometer. A Wnicum SP 800 spcctrophotometcr was used to record the spectra. A PyeDynacap pH meter was used with a glass electrode to measure PH. The pH values of the solutions were checked after the polarogrnphic electrolysis had been carried out.
Monoalkyldithiocarbamates I-III (R-NHCSS-1 Na+; I, R=C&It-; IX, R= XII, R =C~HU-) Were prepared by the following procedure. To 0.1 mole of the appropriate amine was adder1 6 E: of sodium hydroxide (o.15 mole) dissolved in the minimum amount of water. The reaction mixture was cooled and 7 ml of carbon disulphide (0. I:I mole) Was acldcd slowly. The solution was heatccl under vacuum to remove most of the water. Acetone was added to extract the dithiocarbamate. To the filtered acetone extract, ether was added to crystallize out the clithiocarbarnate, which was then filtered off, redissolved in acetone or ethanol and again precipitated with ether, These compounds decompose slowly on standing and re-crystallisation was necessary once a fortnight. The water of crystallisation was determined by loss of weight at 78* z’n 21&~2t0.Each compound was a &hydrate. Melting points were: I, gg”; IX, 88”; III, 64-&Y (dccomp.), uncorrected. Aqueous 0.01: M stock solutions of sodium monoalkyldithiocarbamates made up in distilled water, borate buffer (PH 9.3) or 0.01: M sodium hydroxide showed about 10% decomposition after storage at z” for 3 days, This reaction proceeded especially quickly in 0.01 IM sodium hydroxide solution where a white precipitate was formed, probably a disulphide resulting from autosidstion. Hence, small quantities of stock solution in distilled water were prepared daily. Buffer solutions and all other supporting electrolytes were’ prepared from AnalaR-grade chemicals, CHa-;
In polarographic studies, IO ml of the buffer or other supporting electrolyte was cleoxygcnated by a stream of nitrogen, then the appropriate volume (usually 0.5 ml) of the stock solution was added, and, after a brief flush with nitrogen, the current-voltage curve was recorded. For logarithmic analysis, polaropaphic curves were recorded automatically at n slow rate of scanning (xoo mV/min) to reduce the effects of hysteresis of the recording instrument, To study the reaction of dithiocarbamates with metal ions, various amounts of a stock solution (0.004 M) of the metal salt were successively added to buffered n9d.
Chit&
ACta,
40
(ICJfi8)
4SI)-d(72
POLAROGRAPHY
OF MONOALKYLDITHIOCARBA~~AT~S
46x
4.85 010-4 M dithiocarbamate solutions and the i-E curves wsre recorded after each addition. For the millicoulometric determination of the number of electrons transferred (n), o,s-2.0 ml of the solution of the examined compound in a suitable supporting electrolyte was transferred to the cell10 and the capillary was positioned so that its tip was just under the surface of the solution. The solution was deoxygenatcd, the i-E curve recorded and a potential (usually corresponding to the limiting current) selected. The solution was discarded and replaced by the same quantity of a fresh solution. A voltage corresponding to the chosen potential was taken off.‘the potentiometer and the change of current with time at this potential was recorded at a slow chart speed. After IO-ZO~/~ conversion had been achieved by electrolysis (which required 30-60 min), the electrolysed solution was replaced by the supporting electrolyte alone, in which the condenser current was recorded in order to correct the measured limiting current. The logarithm of the current was plotted as a function of time and from the slope (d logildl) of the resulting straight line the value of n was calculated from: tt-
(4 . 5 *IO’@ (ia)o}/{c~
(d log i/at) )
and z, the volume where (id)0 is the limiting current at t--0, c the molar concentration (in ml), The method was tested by determining s for thallium(I) and cadmium(I1). The average reproducibility was better than =f=5. For the identification of the electrolysis products the same basic procedure was followed, but instead of the continuous i-l record, polarographic i-E curves were recorded at intervals. After electrolysis for 24 h, polarographic curves and U.V. spectra were recorcled. RESULTS
Because the polarographic behaviour of afl straight-chain monoalkyldithiocarbamates studied was found to be similar, it was considered appropriate to choose monoethyldithiocarbamate (I) as a representative compound for a more detailed study. Effect of fin. The ethyl derivative (I) shows in hydroxide solutions one welldeveloped anodic wave (i) (Fig. I). At concentrations above 6 -1~0-4M this is accompanied by an adsorption pre-wave in (Fig. z) at a potentia.l only slightly different from wave i. Below PH xx the wave i is accompanied by another adsorption pre-wave ia which is observed at concentrations greater than I *IO-* M, Above PH IX this prewave merges with the wave i. Below phi: 8, the main wave (i) is split into two waves ([it +&J and &), the ratio of which remains constant and is practically I: : x, The sum of the waves (&f-iz) is constant over the entire PH range studied and is equal to the height of the wave i in hydroxide solutions. In aciclic solutions the more positive wave & is indistinct because of its closeness to the current corresponding to the dissolution of the mercury in the supporting electrolyte. Wave i2 is also complicated by adsorption effects. The separation of waves ix and ia: is best understood from the ps-dependence of the half-wave potentials (Fig. 3). The half-wave potential of wave Z’Zshifts with Anal.
Claim, Ada,
40 (1968)
459-472
D. J.
HALLS,
A. TOWNSHEND,
P. ZUMAN
* 1 ’ c . J. A!‘._; ‘9L a ,u. ‘; U” L- . _ . 3: Pig. I. 2%~ cffcct of ptt on the WUVCY of yodi~m.cLthyi~ithiocnrbnmatc. 5. m-4 M YOCKU~ cihj;clithiocarb(rmuto, f3ritton-Robinson buffers (prt shown on polarogrrrm). Curves rccorclcd from - t .o V towrrrd~ more positive potcntialu, Ir - 08 cm. .Y
I
Fig. 2. The effect of concentration on the WQVCSof sodium c?thyIditfriocarbam~tc, o, T M NnOH concentration of sodium cthylclithiocclrbatnatc: (I) o; (2) 2.0; (3) 3.9; (4) 5.8; (3) 9.9; (6) x1.5 ( 010-4 M). Curves rccorcled from - x.0 V towards more positive potcntirrls, k = 50 cm,
increasing PH towards more negative potentials with a slope of 0.067 mV/pH, whereas that of wave z”lremains practically constant, When the half-wave potential of wave ia approaches -0.3 V at which wave $1appears, both waves merge into wave i, the half-wave potential of which is shifted by 0.057 mV/pH to more negative potentials in a similar manner to wave ia. The half-wave potential of wave i. changes with PH only below PH 3 (Fig. 3), and at higher PH values is practically pw-independent, This causes waves iu and i to merge above pi XL A difference in shifts of halfwave potential can also explain why the wave in is not separated from the total wave i below pw 13. Anal. C~liWLACta, 30 (19683)
459-472
POLAROGRAPHY
463
OF n-IOXOALlKYLlXTHlCOCARBAMATES
Whereas the height of the adsorption pre-wave i. increases with increasing concentration only up to about I *IO -4 M and then becomes constant, the sum of (ia+&) ancl of (i,+il+&) are both linearly proportional to concentration (Fig. 4a) -4 M; this was proved in acetate buffer, pi 4.7, and up to a concentration at 8.~0 phosphate buffer, PH 6.9.
Fig. 3. The p;r clepuntlcncc of the half-wave potentials mate. 0 = Britton-Robinson buffers (see Fig. I). ()
of the W~VCY of sodium f-i&C& solutions.
cthytdithiocarba-
=
a
0
Fig, 4. The dcpcndcncc of the height of the wavc~ (a) Phosphate buffer ptr G.9, (b) 0.1 M N;aOW,
2
4
6
of cthytdithiocarbamatc
8
10
on concentration.
The height of wave in limits at about 6.10 -4 M (Fig. 4b) whereas the total height of waves i~+i in 0.x M sodium hydroxide solution is a linear function of concentration of ethyl derivative I up to x.2 x0-3 M. The half-wave potential of the wave i is shifted to more positive vatues with increasing concentration of the dithiocarbamate I. A similar direction of shifts was observed for other compounds3 forming slightly soluble mercury compounds. With increasing concentration of dithiocarbamate a decrease of the anodic current due to reaction of hydroxyl ions with mercury was observed. In OJ MT sodium hydroxide solution, 5.8 10 -4 M in dithiocarbamate I (curve 4, Fig. z), of the hydroxyl wave is similar in height to wave ir ; at still higher concentrations l
l
Anal.
Cirim. Acta,
40 (xgG8) 459-472
D. J. HALLS,
464
A. TOWNSHEND,
P. ZUMAN
the ethyl derivative (I) no wave for hydroxyi ions was observed (Fig. 5). In 0.01~M sodium hydroxide solution the hydroxyl ion wave disappears at a lower dithiocarhamate concentration, Logarithmic analysis of the wave of I in 540-4 M solution in a &ittonRobinson buffer (PH x0.8) gave a reasonably straight line for the plot of log[(i~ &)/i’j against potential, with a slope of -0.050 V,
Fig. 5. The effect of tllc prcscnce of cttl,yIditlliocnrbrrm~teon the tlnock wtzvc due to hydroxide ions. IO ml 0.x M N&OH. Concentration of cthyldithiocnrbamntc: (x) o; (3) 1.0; (3) 2.0; (4) 0.9; (5) 3.8: (6) 4.8: (7) 5*7; (8) 6.6 (‘10 -4 fW), Curves rccordcd from -0.7 V towrrrclsmore poaitivc potcntinlu (except curve x ; from -0.3 V), li =LZI 50 cm.
The effect of adsorption on the wave i U was proved at concentrations greater than 2 *x0- * M by the linear proportionality of the wave heigflt (i,,) to the height of the mercury column, The linear dependence of (i,+&) and (i,+il +i~) on the square root of ttlc height of the mercury column indicates that these processes are diffusion-controlled, No reduction waves were obtained over the whole PH range (o-x3) studied. Below PH 7 a shift of the current corresponding to hydrogen evolution from the buffer towards more positive potentials indicated that catalytic hydrogen evolution was taking place. However, this current was not separated from the current of the supporting electrolyte and no developed reduction wave was observed in any of the solutions studied. N~ntier of eZ&rom transfemd Comparison of the anodic waves with waves of benzophenone and of nitrobcnzene at PH 9.2 (borate buffer) indicated that the wave 2’in alkaline solutions corresponds to a two-electron process and that the waves ir and is at lower pH values correspond to two successive one-electron uptakes. A comparison of the hci&ts of tfle waves of compounds I and II with the waves of diethyldithiocarbamate in x.95 xo- 4 M solutions at PH 9.2 showed ratios of x.97 and 2.22 respectively. As the total height of the waves of diethyldithiocarbamate corresponds to a one-electron process 11, the transfer of two electrons for monoalkyldithiocarbamates upas verified, l
AMar. Chim. Acta, 40 (rgG8)
459-472
POLAROGRAPHY OF MONOALKYLDfTHSOCARBAMATES
465
Millicoulometrically a value of .tt~~1.87 was found with a 5. x0-4 M solution of the ;rt-butyl derivative (III) in an acetate buffer (PH 5.7) ancln =x.89 with a I *IQ-~ 2M solution of compound III in a phosphate buffer (pr-r 6.9). The millicoulometric determination of $2at high pi values was prevented by the consecutive reaction of the electrolysis product (p. 466). The values found at the limiting current of wave $1 at x ‘10-3 jW were tt =0.70 at 5*x0 -4 M in an acetate buffer (PH 4.7) and n=o,gr in a phosphate buffer (PH 6.9). Str~$urat eSfccts. The behaviour of the methyl (II) and butyl (III) derivatives was completely analogous to that of cthyldithiocarbamate (I), The presence of waves * zo, il, iz and i was proved in all cases. The character of the adsorption pre-waves was analogous, the ratio (in $-it) 52 remaining practically I: : f in all cases. Tile structural changes in this limited group had little effect on the half-wave potential. In 0.1 iW sodium hydroxide solution the approximate values of the half-wave potentials -0.49 V for ethyl (I) and -0~50 V for n-butyl (III) (II), ( -0.47 V for methyl dithiocarbamate, all vs. S.C.E.) are in sequence and show the differences that would correspond to a modified Taft equation AE+= eX,i~* ctX* for reaction constant &%o.z V. It is interesting that even in this case- as for some other anodic processes -the value of the reaction constant is positive.
Controlled-potential electrolysis of solutions of compound III at the potential of the limiting currents of waves il or iz at lower PH[ values resulted in a regular decrease of the limiting- current with time during the electrolysis. No precipitate was observed when the electrolysis was carried out at the limiting current of wave il. On the contrary, in 0.1: &# sodium hydroxide solution, when electrolysis was
Fig. 6. Change of polarographic curves with time during controlled-potcntiat clcctrolysis of sodium n-butyfdithiocarbamatc (I IO- 3 M) in 0.1 N NaOH. Volume of solution - 2.0 ml; tirno of the start of recording of tho curve: (r) o; (2) 55 min; (3) 7 h 25 min; (4) 17 h. Applied potontinf = -0.3 V corresponds to tho limiting current (arrow). Curves rccordcd from - 1.0 V towarcls mart positive potentials I l
Anal. Chim, Ada,
40 (rgG8) 45g-472
D. J. HALLS,
466
A. TOWNSHEXD,
P. ZUMAN
out at the potentiaf corresponding to the limiting current of wave i of compound III, the height of wave i remained practically constant for several hours and only then clecreasecl (Fig. 6). During the electrolysis, a netiv wave i’ appeared at a potential about o.xg V more positive than wave i. The relative height of this wave increased during the electrolysis (compare ‘curves 3 and 4, Fig. 6). A solution of the butyl derivative (III) kept under the same conditions, but without applying the potential, showed that cleavage of III by a chemical reaction in the bulk of solution during electrolysis was insif;nificant. A black precipitate formed during the electrolysis was proved to be mercury sulphide by the action of acid. The electrolysed solution liberated tile characteristic odour of an isothiocyanate. In the solution after electrolysis for 24 11,where the total height of the waves was still about zoo/, of the original height, no ultraviolet bands due to dithiocarbamatc were observed (Fig. 7). Calculation has shown that x0/0of the original amount of compound (III) would still be apparent from the U.V. spectrum. The “end absorption” of the electrolysed solution at 228 nm (curve B, Fig: 7) was shifted compared with sodium hydroxide (2x5 nm, curve C, Fig. 7). Similar end absorption was observed in the presence of carbonate (curve D, Fig. 7). When the efectrolysed solution was acidified the end slbsorption was shifted to a shorter wavelength and a bancl at 240 nm was apparent (curve E, Fig. 7). Similarly, during electrolysis in a borate buffer (PH 9.2) it bfack precipitate was formed and a shoulder was present in the U.V. spectrum at 240 nm. Formation of carbon disulphide WZB demonstrated by the presence of a small band at 320 nm and of a polarographic reduction wave at - IJ V, wbiclr diminislled after bubbling nitrogen through the solution. Rcaclions wilh m-W iotts. Addition of cadmium ions to tile solution of sodium ethyldithiocarbamate (I) in 0.x M sodium liydroxidc solution resulted in a decrease
carried
Fig. 7. U.V. spcctta of clcctrolysis producta. Controftcdwpotcntirrlckctrolysis of x I o-3 M sodium tr-butyldithiacarbamato carriectout for 24 hat --al3 Vin 0.1: M sodium hydroxide. (A) x.25. x0-4 M wbutyidithiocarbamato in o.012 M sodium hydroxide bcfare olcctrolysis (diluted 8 x ) ; (X3) solution after clcctralysis (diluted 2 x ) : (C) 0.05 M sodium hydroxide (diluted 2 x ); (D) soln. as in (C) in prescncc of sodium carbonate ; (E) soln. as in (B) ncidificd by hyclrochlaric acitl, l
Fig. 8. Addition of EL solution of cadmium ions to sodium cthyldithiocarbzrmutc. To ro mt of 5. x0-4 N sudium cthytdit~tiocnrbumrrtoin 0.x M sodium hydroxide 4. x0-3 M cndn~iutn aulphatc was successively added. Currents corrcctcd for dilution. (f) Height of anodic wave (i) of dithiocnrd bamstc; (2) totat height of nnoclic WQVCS. A?&.
Cki?% tl C:iZ,40 (qG8) 459-472
POLAROGRAPHY
OF MONOALKYLDITH10CARBAhlATES
467
of the anodic wave of ethyldithiocarbamate. A white precipitate appeared initially, but when the ratio of cadmium(Il) :dithiocarbamate exceeded I: : 2, the precipitate turned yellow if the PH of the reaction mixture was sufficiently high. Simultaneously a new, small anodic wave formed at more positive potentials (at about -0.3 V). When the ratio of cadmium(H) : dithiocarbamate reached x :I, the slope of the dependence of the first wave on cadmium concentration changed (Fig. 8). With further increases in cadmium concentration the total height of the anodic waves decreased and the anodic waves vanished when the concentration ratio was about z Cdz+: I: dithiocarbamate. No cathodic waves corresponding to soluble complexes were observed. This behaviour can be tentatively explained by the assumption that the white precipitate initially formed is the compound Cd(dithiocarbamate)z. The yellow precipitate is attributed to cadmium sulphide, formed from the compound Cd(dithiocarbamate)z and further cadmium ions. It is assumed that in this reaction the cthylmonothiocarbamate ion is formed, which is responsible for the formation of the anodic wave12 at -0.3 V. Further addition of cadmium ions results in the cleavage of the . monothiocarbamatc with further formation of cadmium suIphidc. This interpretation seems to be confirmed by the result of a large-scale experiment, which showed tflat 50 IO -5 moles of sodium ethyldithiocarbamate react with r .f x0 -4 moles of cadmium ions, In this reaction the formation of 0,94 ‘x0-4 moles of cadmium sulphide was proved gravimetrically. This yellow precipitate, when decomposed by hydrochloric acid, liberated hydrogen sulphide. fn the solution thus formed, x.0x. xo -4 Ad’ cadmium(I1) was found polarographically. The other reaction products were not identified. In attempts to isolate the product assumed to be the x : I complex, the existence of which was indicated by the intersection in Fig. 7, solutions of cadmium(I1) (I 0x0-3 moles) and dithiocarbamate (I GO -3 moles) were reacted in the presence of Z*IO-3 moles of sodium hydroxide. When the white precipitate was extracted with chloroform, yellow cadmium sulphide began to precipitate rapidly from the chloroform layer. The UV. spectrum of the compound remaining in the chloroform layer showed a band at 235 nm, indicating the presence of isothiocyanatcs? Their presence was supported by the pungent odour of the residue after chloroform had been distilled off. Similar results were observed when mercury{ II) ions were added to the solution of ethyldithiocarbamate ions in o. r n/rsodium hydroxide solutions. A white precipitate (attributed to the dithiocarbamate-mercury compound) was formed initially. On further ‘addition of mercury(H) this precipitate turned black, owing to mercury(H) sulphide formation. The decrease in the anodic waves with increasing metal ion concentration was analogous to that found for cadmium, except that the anodic wave decreased practically to zero when the ratio mercury(I.1) :dithiocarbamate was about 3:~. Somewhat different results were observed when lead ions were added to a solution of ethyldithiocarbamate in a borate buffer (PH 9.3). A white precipitate was formed and the anodic wave of ethyldithiocarbamate decreased practically to zero when the ratio of lead(Il) :dithiocarbamate reached I :z. The composition of the white precipitate is therefore Pb(dithiocarbamate)z, Addition of further lead ions slowly turned the precipitate black. l
Anal.
C&n.
Acta, 40 (x968)
45g-472
468
D.
J.
HALLS,
A.
TOWNSHEND,
P. ZUMAX
1)ISCUSSION The two one-electron diffusion-controlled and in can be formulated by scheme (A)-(D): RNHCSS - F= \ ., PKy H+ RNHCSS
RNHCSS =zi+
RNHCSS
- + Hg w 1s1
- +H+
RNHCSS
processes corresponding
to waves il
(A)
-1-c
(r-3)
N$o
S+H+ -6 R-NN=C/ \S_ Pf(Z H&l)
(D) The wave ir at potential El corresponds to processes (A) and (B), and wave i2 at potential 5’2 to processes (C) and (D). The shift of the half-wave potential of wave & indicates that the value of p& for the ethyl derivative I is about 3.5. This is in good agreement witlt the plc value of 3.ro for monocyclohcxyl and 2.95 for isopropyl dithiocarbamates obtained6 from kinetic data. The ]PH independence of the half-wave potential of wave z’t. above pw 4 indicates that the electroactive form is identical with the form of dithiwarbamate predominating in the solution. The shift of the potential of wave in indicates that the value of p& is greater than 12, This is consistent with the alkaline decomposition of monoalkyldithiocarbamates. It is assumed14 that this reaction t&es place with the formation of R-N=CSS=-, and that it can be observed only at PH about 14. The pKa value of the uncornplexed species RNHCSSis therefore greater than x4. Even when a shift toward smaller p& values can be expected for the complex species in reaction (C), the effect seemingly is not sufficient to shift it below 12. The slopes of the pH dependences of the half-wave potentiaXs of wave ix below pw 3 and of wave i2 over the whole pH range are consistent with the values expected for a reversible one-electron process accompanied by a transfer of one proton. The electrode reactions (A) and (B) corresponding to wave ir are analogous to the reaction of metal salts with monoalkyldithiocarbamates at lower E)H values, as was shown for the reaction of lead ions in a borate buffer. Even though the participation of the nitrogen atom in the mercury bonding cannot be excluded, the bond to two sulphur atoms can be considered as more probable. The actual form of the mercury compound resulting in reaction (D) is not known. The experimental results nevertheless indicate that the compound formed is unstable and undergoes further changes in alkaline media. A possible decomposition scheme for a complex of a metal (M) can be formulated as follows:
POLAROGRAPHY
OF MONOALKYLDITHlOCARBAMATES
469
SR-N=C’ WI)
\S_
+
R-N=C=S
$-MS
(E)
SR-N=C=S+-tH-
+
+H+
R-N--C’
u?
‘OSR-N=C’
+M2+ ‘O-
R-N=C WI)
S’ Lo-
scvornl
&
R-N=& BE(H)
P\O_
w
stcpcr
2 or1-
-j. R-NHz+MSi-COP
(HI)
The formation of metal sulphides, isothiocyanates and carbonates as predicted by this scheme was proved. The scfreme (A)-(F) explains the behaviour in controlled-potential elcctrofysis of clerivative I in 0.x M sodium hydroxide solution, The delay in the decrease of the limiting current can be explained by the cteavagc of the electrolytically formed complex [RNCSS+]M(II) accordingto reactions (E) and (F), In the latter, monoalkyl monothiocarbamate is formed, This compound is known to give anodic wave@; these waves will resemble, both in their heights and potentials, the waves of monoalkylclithiocarbamates and hence it would not be surprising if in a mixture of monoand dithiocarbamates, one wave is formed. During the initial stages of electrolysis ’ the rate of formation of the monothiocarbamate is just sufficient to supply enough depolariser to keep the limiting current practically unchanged. As the conversion of dithiocarbamate into monothiocarbamate becomes more complete, the anodic wave decreases. The rate of the interposed chemical reaction (I?) governs the rate of the controlled-potential electrolysis. Reaction (F), as was shown previouslyrb, is first order in hydroxyl ions. Hence at lower E)H values (PH 6.9) the regeneration of an electroactivc species (with similar electrochemical properties to the original compound) cannot affect the controlledpotential electrolysis, which therefore follows a simple course, The actual course of the decomposition may be more complex than shown in reactions (E)-(H). This is indicated by the formation of carbon disulphide during electrolysis at PH 9.2. It was also not established whether the two waves observed during electrolysis in 0.1: M sodium hydroxide (Fig, 6) both belong to the monothiocarbamate or if one belongs to another intermediate. It was necessary to exclude the possibility that the wave i2 corresponds to mercury sulphide formation rather than to a reaction of RN=CSS? Comparison of the half-wave potentials of wave ia with that of sulphide ions indicates that the sulphide wave is 150 mV more negative. Moreover, the su1phid.e wavelo shows a markedly different PH dependence of half-wave potentials to wave i2. Furthermore, Awl.
Cltim
Acta, 40 (x968) qjg-472
470
D.
J. HALLS,
A.
TOWNSHEND,
P.
ZUMAN
the formation of the white precipitate attributed to [RNCSS]M(II) (even when isolation was impossible) indicated that reaction (E) and the subsequent reactions may bc fast, but, uncler the conditions studied, not so fast as to convert the primary product formed after the transfer of the second electron into sulphide ion during the drop-lift. Hence the process differs from that observed recently’7 for thiourea in aciclic solution. The anoclic waves of monoalkyldithiocarl~amates corresponding to a twoelectron over-all process differ in principle from tlre vast majority of organic sulphur compounds, which give one-clcctron anodic waves 1-R. Among other exceptions the waves of thiourcaiH-2” and of tliiobcnzamidc~~ can be mentioned. Tn the latter case bcnzonitrilc, mcrcury(I1) sulphide and some benxamide were isolated from the controlled-potential clcctrolysis products. For a two-electron process at potential Ez, it is possible, apart from the scheme (A)-(D) corresponding to a system of consecutive reactions, to devise another schcmc, in which tllc reaction at potential L;z would be competitive with that depictccl in ecluations (A) and (R). Such a scheme would involve the formation of mercury(I1) and reaction with the anion of clitlliocarbamatc. Tl~c cllange in the wave-height iz during controlled-potential electrolysis at the limiting current of il cannot be used here as a cliagnostic tool 22 because the formation of a slightly solul~lc salt such as the product of (13) will result in a decrease of wave i:! for both competitive and consecutive reactions. On the other hand, a competitive process requires that the same species reacts with uni- and bivalent mercury ions. This in turn would indicate the same shape of the pH-dependcncc of the half.wave potentials. The d’ffcrent shape of this depenclence for waves il and i:! indicates that a competitive process, at least of the consiclerecl type, is less probable than the consecutive process (A)-(D). Two types of adsorption processes affect the polarographic curves. Wave iu, with a limiting current value of about 0.4pA, corresponds to a surface covered by a single molecule of about 50-100 A3 (depending on the composition of the compound). Wave ir\, observed in sodium hydroxide solution, reaches a limiting value of 5 ,YA and corresponds to a surface of s-so A2 per molecule. If the composition of the mercury compound remains the same, it can be concluded that either the wave it, corresponds to a monolayer and i,+ to a multilayer formation or, perhaps more probably, the wave i R corresponds to a film in which the molecules are oriented flat at the surface and the wave i* to a film in which they are perpendicular and only “anchored” by the functional grouping. Such an observation has been made and a similar interpretation offered for waves of other sulphur compounclszz-20. The effect of wave (in +i) on the waves of hydroxyl ions (Figs. z and 5) indicates that the film of mercury compounds with alkyldithiocarbamates in alkaline media is difficult to penetrate. Because the result% obtained for monoalkyldithiocarbamates are in good general agreement with those reported 7.8 for dithiocarbamoyl carboxylic acid, it seems that the observed behaviour is characteristic for a wide range of compounds with the grouping -NHCSS-. The authors thank Professor R. BELCHEIX for his interest. D. J. H. thanks Professor M. STACEY for the provision of a research grant and I?.%. the Science Research Council for a provision of a Senior Visiting Fellowship. A?sal. Cihrr. Actn, 40 (1gG8)
459-472
POLAROGRAPHY
OF
~fOSOALKYLDITHIOCARRh~~AT~S
471
SUMMARY
Monoalkyldithiocarbamates give two anodic polarographic waves, corresponding to a one-electron and a two-electron process. A reaction scheme is proposed. The unusual behaviour observed during controlled-potential electrolysis in sodium hydroxide media is interpreted as a chemical reaction of the primary clcctrolysis product; the species formed gives a wave similar to the original compound. Different adsorption phenomena are probably caused by varying orientation of the mercury compound on the electrode surface. The adsorbed layer formed in sodium hyclroside solution is so close-packed that it prevents hydrosyl ions from penetrating and thus is the giving an nnodic wave. For analytical purposes, 0.1 M sodium hydroside most suitable supporting electrolyte; linear calibration graphs can be obtained over the range 5 axe-5-I .Io-:a M.
Les rnonoalkyldithiocarbamates donnent clcuzc vagucs l”)larograplliques anodiqucs, correspondant h un et A deux 6lcctrons. Un sch6mn dc la rdaction cst propos8. Le comportcmcnt obscrv6 au tours de l’c%xtrolyse zi potentiel contrOld en milieu hydroxyde de sodium est interprdtb: r6action chimique du procluit cl’&lcctrosont probablcment causes par une lyse primnire. Divers pht5nom&nes d’adsorption orientation variable du composc5 dc mercure sur la surface dc l’blectrode. La couchc adsorb& formc!e dans la solution d’hydrosycle de sodium cst si compactc qu’clle ernpCche la p&n&ration des ions hydroxylcs et fournit ninsi une onde anodiclue. Pour l’analyse, l’hydroxyde clc sodium 0.x hr est l’&xtrolyte de base qui convient le mieux. Des graphiqucs lin6aircs de calibrage sont obtcnus dc s*ro-” ri I.Io-:~ M.
ZUSAMMENFASSUXG
Monoalkyldithiocarbamate ergeben zwci anodisclle polarographische Stufen, welche cinem Einelektronenund einem Zweielcktronenprozess entsprechen. Ein lieaktionschema wird vorgeschlagen. Das Verhaltcn, welches man w2hrend cler Elcktrolyse mit kontrolliertcm Potential im Natriumhyclroxid-Medium beobachtet, wird mit Hilfe einer chernischen Reaktion dcs primiircn Elektrolyseproduktes interpretiert. Die gebildeten Spezies crgebcn tine Stufe, die der der ursprtinglichen Verbindung ~hnlich ist. Unterschiede in Adsorptionsphtinomena kijnncn durch variiercndc Orientierung der Quecksilberverbindung auf der Elektrodenoberfllichc erkl5rt werden. Die adsorbierte Schicht welche i.n Natriumhydroxidliisung gebildet wird, is so dicht, dass sie ein Eindringcn der Hydroxylionen verhindert und so ihre anodische Stufc unterdrtickt. Stir analytische Zwecke ist 0.1 iM NaOH cler am mcisten geeignete Grundelektrolyt. Linearc Eichkurven kijnnen im Bercich von 5 *IO-~ bis I * IO -2 M erhalten werden. REFERENCES I I. M.
I
AND
J. J. LINGANE,
Polarogvnplry,
2nd
Eth.,
Intcrscicncc,
Anal.
Chiwt. Acta,
New York, 40 (rgG8)
1952.
459-472
D. J. HALLS,
472
A. TOWNSHENR,
P. ZUMAN
BkeztN~ AND 1’. HUMAN. Polarograpiby i*r Medicine, Z~iochemistvy and Z%avmacy, Rcviscd Eng. Edn., Intorscioncc, Now York, 1958. J, ~-fP,YIcOVSK+ AND J. I<~ITA, Fvincifiks of ~~drcrogvaphy, Czech. Acaclcmy of Scicnccs, Prague. I9ei. M. ~?EDORO~KO, 0. MhrrouWK AND P. %unfA~, ~~ZJe~~~~7~ Gzecii. C&em. Commtrn., 21 (xc)@) 672, with rcfcrcnccs to previous work. R. ZAIrRADNfu AND I?. ZUMAN, Collection Czcclt, Gfrem. Commtcn., 24 (x959) x 132. Rnalysl, if1 prcpnration. D. J, I-JALLS, A, TOWNS~IIGND AND T‘. %UhfAN, R. ZAIIHADN~K AND L. JI~N!~ovw&, Chcm. L&/y, 48 (x954) IX. R, ZA~IRADN~IC, CAonr. Listy, 49 (1955) x002: Cotclccfion Czech. Chm. Commtrr~., 21 (r9sG) 4,!7. D. J. I-JALLS, PVOC. Sot. /J7ml. cj~??ti., 4 (1967) 40. 0, MAN~U&~C, Private communication. AND 1’. %LlbtAN, PJ?tUt. ChiWZ. /?CfU, in prCS!t. 1). , tiALLS, I\. ‘rO\VNStltZNn V. VASAK, Ctaenr. Listy, 48 (rgg.$) 19. v. ii ~IDIVEC AND 'IX,SVATIIK, R. %AIIIUDN~K AND A. KJAER, .4cC~ Cherrr. Sca7tci., 13 (1959) 442. M. WRONSKI, %cszyfy Nauk. Udw. Lodt., Narrlri Ma&.-Pvzyvod., Ser. If, No. G (1959) 121.
2 M.
3 4 5 G 7 8 9 x0 xx I2
13
14
I 5 IX. ZAIIICADNJIC, Collectio?t Czech. C/rem. Corrt7rzr17r., 24 (1959) 3407. IG A, TlZflrONOV, IOU. Z
Vol. 5. Intcrscicnco, 1967. p. 81. 23 I?, ZUMAN, C/rem, Z:isly, 48 (x954)
24
25 26
I?. ZUMAN AND I’<, ~UnXANOVA, .2x (f9gB) 123. 0, IWANOUSIEK AND P. %U&lAN.
20 (1955)
1340.
21
G78.
M, I:LXDORO$JKO (1956)
AND
P. ZUMAN,
rooG; Collection
Physicat
Commw., 20 (1955) 652 ; co~lcclio~b ~SCCk t%?#L
Czcclr. Clrem.
(t955)
Ovgarric
ChC?dStYy,
046.
Ctlem.
&?Gfy.
49
C&?nt.
Listy,
4g (1955)
G68; Collectiorr
G%x%. Chant. ~ommtor.,
Chcnr.
Lisly,
4g (1955)
x484; Collccliorr
Czcclr. Cllem. Commzrr~. .
CO?#?tlZit~.
,