E. A. ZAKHAROVA

SOME FEATURES OF THE ANODIC DISSOLUTION OF COPPER IN ALKALINE AND CARBONATE SUPPORTING ELECTROLYTES IN AMALGAM STRIPPING POLAROGRAPHY

(Presented by the scientific seminar of the Department of Physical Chemistry)

For the analysis of substances of high purity, the method of amalgam polarography with accumulation has in recent years become increasingly widespread [1]. However, in a number of cases the assessment of the quality and quantity of the impurity being determined is complicated by the appearance of anomalous double or triple peaks for a single element. Earlier [2] we studied the influence of the concentration of alkali and of Pb ions in solution on the form of the complex anodic peaks of lead. The literature also contains information on the occurrence of double peaks of lead [3] and copper [4] in alkaline and carbonate supporting electrolytes; however, detailed investigations in this direction are lacking. It was therefore of interest to study the anomalous behavior of copper amalgam during its anodic dissolution in alkaline and carbonate solutions, which are often used for the simultaneous determination of several elements. In the present work we investigated the influence on the form of the anodic copper peak of a number of factors—the concentration of copper in solution, the electrolysis time, the nature and concentration of the supporting electrolyte, and the rate of change of potential—with the aim of elucidating the causes of such anomalous behavior of copper amalgams during anodic dissolution in the indicated supporting electrolytes.

Experimental procedure

In the work a self-recording Hungarian polarograph 7-77-4/b was used. The polarographic cell and electrodes did not differ from those described earlier [2]. Standard copper solutions were prepared from weighed portions of metallic copper dissolved in nitric acid and were successively diluted with tridistillate to a final concentration of about \(1 \cdot 10^{-5}\) M. For preparing the supporting-electrolyte solutions, reagents of sp. ch. or kh. ch. grade were used; the Cu \((2+)\) content in the solution under investigation was less than \(1 \cdot 10^{-8}\) M. The number of electrons \(z\) participating in the process of anodic dissolution of copper from the amalgam (per copper atom) was determined by a coulometric method. The quantity of electricity \(Q_p\) associated with the anodic process of copper dissolution was determined by measuring the area under the anodic peak, stopping the change of potential after the peak potential had been reached¹. Measurement of the area

was carried out with a planimeter. These data were compared with the quantity of electricity \(Q_{\mathrm{e}}\) spent on the reduction of copper ions during the preliminary electrolysis. The sought number of electrons \(z\) was determined from the obvious equality:

\[ \frac{1}{n} Q_{\mathrm{e}} = \frac{1}{z} Q_{\mathrm{p}}, \]

whence

\[ z = \frac{n Q_{\mathrm{p}}}{Q_{\mathrm{e}}}. \]

(\(n\) is the number of electrons participating in the electrolysis process per one copper ion; \(n = 2\)).

During one day, work was usually carried out with the same drop of mercury; in the event of a deviation of the copper peak height \(> 5\%\), a new drop of mercury was electrodeposited. The completeness of dissolution of copper from the amalgam was checked by repeatedly obtaining an anodic polarogram after intensive stirring for 1 minute with the circuit open. The absence of a copper peak on the repeated polarogram indicated complete dissolution of copper from the amalgam.

For a comparative study of the electro-dissolution of copper amalgam in different supporting electrolytes, the electrode-transfer technique was used. It consists in carrying out the preliminary electrolysis, with formation of the amalgam, in one supporting electrolyte, and the anodic dissolution of the amalgam in another. In this procedure, the amalgam drop together with the siphon of the auxiliary electrode and the drop of electrolyte wetting them is transferred, with the circuit open, in an atmosphere of inert gas first into an intermediate solution for washing, and then into the test electrolyte, in which the dissolution of copper from the amalgam is carried out. The electrode-transfer technique and the design of the cell have been described in more detail earlier [5]. The losses of copper during transfer, with the technique used by us, did not exceed 4–6%.

In coulometric calculations of the number of electrons using the transfer technique, it is possible to apply the “standard area” method under the anodic peak instead of measuring the electrolysis current in those electrolytes in which it is difficult to measure the electrolysis current because the current efficiency deviates from 100%, or in the case of very low concentrations of the substance being determined in solution, when the electrolysis current is less than the residual current.

For the areas under two anodic peaks obtained in different electrolytes under identical electrolysis conditions, the relation

\[ \frac{1}{z_0} Q_{\mathrm{p}}^{0} = \frac{1}{z} Q_{\mathrm{p}}, \]

is obviously valid, where \(Q_{\mathrm{p}}^{0}\) is the “standard” area under the anodic peak in the electrolyte for which the value \(z_0\) is known. From this equality we determine \(z\) for anodic dissolution of the metal from mercury into a solution of another electrolyte. It should also be noted that the standard-area method makes it possible to determine the number of electrons in the anodic dissolution of the amalgam in the electrolyte under study without carrying out preliminary electrolysis in this supporting electrolyte (Table 6).

Experimental Part

The dependence of the height of the copper peaks on the concentration of its ions in a 0.1 M \(K_2CO_3\) solution is shown in Fig. 1. In the region of copper concentrations

Fig. 1. Dependence of the height of the second copper peak on the concentration of Cu(2+) ions in 0.1 M \(K_2CO_3\) solution. Experimental conditions: \(\tau = 3\) min; \(v = 5\) ml; \(r = 0.04\) cm (\(\tau\)—electrolysis time; \(v\)—solution volume; \(r\)—radius of the mercury drop). The symbol \(\triangle\) denotes the height of the first, more negative, copper peak, appearing at Cu(2+) concentrations higher than \(6 \cdot 10^{-6}\) M.

Fig. 2. Effect of the preliminary electrolysis time on the magnitude of the anodic current of the first (curve 1) and second (curve 2) copper peaks in 0.1 M KOH. Curve 3 is the area under the double copper peaks with potential arrest. Experimental conditions: \(C = 1 \cdot 10^{-5}\) M; \(v = 5\) ml; \(r = 0.04\) cm; \(W = 600\) mV/min. (\(C\)—concentration of copper in solution).

At \(1\cdot 10^{-7}—6\cdot 10^{-6}\ M\), under the experimental conditions, one anodic peak appears on the polarogram with \(\varphi_p=-0.04\ \mathrm{V}\); with a further increase in the copper concentration the peak broadens and splits, the first peak depending on the concentration nonlinearly. For copper in \(0.1\ N\) KOH, in the concentration range studied, \(5\cdot 10^{-8}—1.5\cdot 10^{-5}\ M\), the peak shape is complex (Fig. 3, curve 3), the second peak increasing linearly with concentration, while the first reaches a limiting value at a concentration of \(4.5\cdot 10^{-7}\ M\). The dependence of the anodic peak on electrolysis time for \(9\cdot 10^{-6}\ M\) Cu(2+) in \(0.1\ N\) KOH solution has a similar character (Fig. 2). The area under the peak with a stop in the interval of 1–20 minutes changes linearly (Fig. 2).

The background concentration has a noticeable effect on the shape of the anodic peaks of copper. Thus, in \(0.1\ M\) \(K_2CO_3\) solution at a Cu(2+) concentration of \(1\cdot 10^{-5}\ M\) and two-minute electrolysis, a characteristic double copper peak is obtained on the polarogram (Fig. 3, curve 2). Under the same electrolysis and dissolution conditions, but in \(0.5\) and \(1\ M\) \(K_2CO_3\) solutions, anodic polarization gave one normal copper peak with \(\varphi_p=-0.15\) and \(-0.18\ \mathrm{V}\), respectively (Fig. 3, curve 1). In alkali solutions of different concentration (\(0.1\ N—2\ N\)), at the same initial copper concentration in the drop, double copper peaks are obtained with potentials shifting to the negative side with increasing pH. The influence of the nature of the background on the shape of the anodic copper peak was studied by the electrode-transfer method. In this procedure, electrolysis with accumulation was carried out in KOH, \(H_2SO_4\), and acetate buffer solutions, while electrodissolution was carried out in backgrounds that are: 1) indifferent with respect to Cu(2+) (\(H_2SO_4\), \(KNO_3\)); 2) complex-forming agents (\(NH_4Cl—NH_4OH\)); 3) forming sparingly soluble compounds with Cu(2+) (KOH, \(K_2CO_3\)). As can be seen from Table 1, on all backgrounds except KOH and \(0.1\ M\) \(K_2CO_3\), at the same concentration of copper amalgam, normal copper peaks were obtained. Complex peaks in \(0.1—0.2\ N\) KOH and \(0.1\ M\) \(K_2CO_3\) backgrounds appeared irrespective of whether electrolysis and dissolution were carried out in these solutions or only dissolution (in copper-free solutions) after accumulation of copper in other electrolytes (\(0.36\ N\) \(H_2SO_4\), acetate buffer, \(pH=4.2\), etc.). From Table 1 it can be seen that the sensitivity for copper, \(\frac{I}{C}\) (the magnitude of the anodic peak referred to unit concentration in the solution under identical stirring conditions), differs on different backgrounds. Thus, the largest \(\frac{I}{C}\) is obtained in \(H_2SO_4\) solutions and the smallest in \(NH_4Cl—NH_4OH\).

Fig. 3. Typical shapes of simple and complex copper peaks on backgrounds: 1 — \(0.5\ M\) \(K_2CO_3\); 2 — \(0.1\ M\) \(K_2CO_3\); 3 — \(0.1\ M\) KOH (electrode — mercury drop); 4 — \(0.1\ M\) \(K_2CO_3\) (film electrode).

Table 1

Anodic copper peaks obtained by the method of transferring the electrode into various supporting electrolytes.
Experimental conditions: \(v = 3\ \mathrm{ml};\ C = 1.6\cdot 10^{-5}\ \mathrm{M};\ \tau = 2\ \mathrm{min}\)

¹) The width of the composite peak was conventionally measured at half the height of the more positive peak.
\(I/C\) is the ratio of the peak height to the ion concentration;
\(\delta\) is the half-peak width; \(\varphi_p\) is the peak potential.

Comparison of the values of

\[ \frac{I}{C}, \]

obtained in solutions after transfer from KOH, with the values of

\[ \frac{I}{C}, \]

obtained after carrying out electrolysis in a sulfuric-acid or acetic-acid supporting electrolyte, shows that in the first case lower values of

\[ \frac{I}{C} \]

are obtained. This may be caused either by a decrease in the copper concentration in solution due to hydrolysis, or by a decrease in the electrolysis current owing to passivation of the mercury

drop in the cathodic cycle. Table 1 (column 2) also shows anomalously large values of the half-wave width of copper on KOH and K₂CO₃ backgrounds (220–210 mV). The influence of the rate of change of potential (\(W\)) on the shape of the copper waves on backgrounds of 0.1 M K₂CO₃ and 0.1 N KOH is illustrated in

Table 2

Influence of the rate of change of potential on the height of the double copper waves and on the distance between them. Background: 0.1 N KOH, concentration Cu(+2) = \(9 \cdot 10^{-6}\) M, \(\tau = 3\) min

\(W\) — rate of change of potential (mV/min),
\(I_1\) — height of the first wave (the more negative one),
\(I_2\) — height of the second wave,
\(l\) — distance between the first and second waves.

Fig. 4 and in Table 2. At very low rates (\(W\) less than 100 mV/min) only the first wave, of elongated shape, is obtained. As the rate of change of potential is increased from 200 to 1000 mV/min, the ratio of the magnitude of the first wave to that of the second

\[ \left(\frac{I_1}{I_2}\right) \]

decreases (Table 2), the distance between the waves remains constant, and the shape of the wave becomes simpler (Fig. 4, curve 3). The influence of stopping the potential after the first and second peaks on the shape of the waves was studied (Fig. 5). When the potential was stopped after the first peak (Fig. 5, curve 1) (with the recorder paper tape continuing to move), not all the copper left the drop, since upon subsequent starting of the drum (curve 1, point \(b\)) a second wave appeared at a potential coinciding with the second wave (curve 2). When the potential was stopped after the second wave (Fig. 5, curve 2, point \(a\)), a double copper wave was obtained whose area was equal to the sum of the areas of the individual waves with stopping (curve 1).

Fig. 4. Influence of the rate of change of potential on the shape of the anodic waves of copper. Curves: 1—333 mV/min; 2—500 mV/min; 3—1000 mV/min. Experimental conditions: \(C = 3 \cdot 10^{-6}\) M, \(\tau = 3\) min; \(r = 0.04\) cm.

We used the technique of stopping the potentiometric drum to calculate the areas under the anodic waves, from which the number of electrons participating in the process of anodic dissolution of copper amalgam in KOH and K₂CO₃ solutions was calculated. From Tables 3, 4

and 5 it is seen that, on these supporting electrolytes, copper passes into solution with the participation of two electrons both at the dropping electrode and at the film electrode (for the shape of the double copper peak at the film electrode, see Fig. 3, curve 4). For

Fig. 5. Anodic polarograms of copper recorded with stopping of the potential after the first peak (curve 1) and the second peak (2). Points a and b indicate the moments of stopping and starting the potentiometric drum.

small concentrations of Cu(2+), underestimated values of \(z\) were probably obtained because of the large residual current, which introduced an error into the calculation of the electrolysis current.

Table 3

Calculation of the number of electrons participating in the anodic dissolution of copper from amalgam. Experimental conditions: \(\tau = 3\) min, \(W = 600\) mV/min, \(r = 0.04\) cm, supporting electrolyte—0.1 N KOH

¹ \(Q_\mathrm{p}\) was calculated from the area under the anodic peak with stopping of the potential after the second peak.

Table 4

Calculation of the area and the number of electrons in the anodic dissolution of copper from amalgam. Experimental conditions: supporting electrolyte—0.1 M \(K_2CO_3\), \(\tau = 3\) min, \(r = 0.04\) cm

¹ Area under the double peak with stopping.
² Total area under the first and second peaks with stopping.

To compare the number of electrons involved in the oxidation of copper on different supporting electrolytes at the same initial concentration of copper in the drop, the “standard-area” method mentioned above was used. The validity of the “standard-area” method was checked by us by comparing the areas of anodic dissolution of copper in solutions of 0.36 N H₂SO₄ ($z_0 = 2$) and 1 M NH₄Cl + 1 M NH₄OH. As is seen from Table 6,

Table 5

Calculation of the area and of the number of electrons involved in the anodic oxidation of copper from amalgam. Experimental conditions: supporting electrolyte—0.1 M K₂CO₃, $C = 5 \cdot 10^{-6}$ M, $\tau = 1$ min, $W = 666$ mV/min, film electrode, $S = 0.2$ cm²

in the second electrolyte $z = 1$, that is, the dissolution process proceeds with the formation of monovalent copper ions, which accounts for the low sensitivity of copper determination on the ammonia–chloride supporting electrolyte.

The influence of stirring the solution with a stream of nitrogen during recording of the anodic polarogram was studied. It was found that, under the experimental conditions, stirring did not affect the shape or height of the complex anodic teeth of copper on the 0.1 M K₂CO₃ supporting electrolyte and only slightly changed the shape of the copper tooth on KOH.

Discussion of Results

The results obtained, taking into account some literature data relating to the anomalous dissolution of amalgams, can be explained from several points of view. It may be assumed that the cause of the formation of double teeth is the formation of several intermetallic compounds of copper with mercury, having different standard electrode potentials, with equilibrium between them being established slowly. Confirmation of this assumption would seem to be the experimental fact that amalgams of certain metals (Co, Mn) give several teeth on the polarogram, which, according to the authors [6, 7], are caused by the oxidation of different intermetallic compounds. This view is disputed by Zebreva [8], according to whom free metal atoms formed as a result of the dissociation, at room temperature, of unstable solid compounds of the metal with mercury take part in the electrochemical reaction.

Our experimental data with transfer of the electrode do not support the above assumption. If the cause of the appearance of double teeth were the formation of intermetallic compounds in mercury, it would be expected that double teeth would occur on any supporting electrolytes, regardless of their nature and concentration. However, our experiments show (see Table 1) that on some supporting electrolytes (KOH, K₂CO₃) a double copper tooth appears, whereas on others it does not.

Further, one may assume the presence of a stepwise discharge of copper from amalgam. The process of anodic dissolution of copper amalgam with respect to the valence of the ions entering the solution has not been sufficiently studied. There are literature data [9] on the dissolution of a copper anode in aqueous-

solutions, from which the conclusion is drawn that copper is initially oxidized to the monovalent state, while the final result depends on what happens to Cu(I). Under certain conditions (alkaline and ammoniacal solutions), during the oxidation of a copper amalgam dropping electrode, double waves appear on the polarogram; their equal height leads investigators [10, 11] to the conclusion of a stepwise discharge. The most direct answer to the question of how many electrons participate in the electrochemical reaction should be given by coulometric measurements of the area under the anodic peak. The measurements we carried out (see Tables 3, 4, 5, 6) showed that in all supporting electrolytes (except \( \mathrm{NH_4Cl}—\mathrm{NH_4OH} \)), regardless of whether a normal or a double peak is obtained, the area under the anodic peak corresponds to the two-electron electrodissolution of copper from the amalgam. In the case of \( \mathrm{NH_4Cl}—\mathrm{NH_4OH} \), a one-electron normal copper peak is obtained.

Table 2

Calculation of the number of electrons participating in the oxidation of copper from an amalgam in different supporting electrolytes by the “standard area” method. Concentration of Cu(+2) \(1.6 \cdot 10^{-5}\) M, \(\tau = 2\) min, \(r_k = 0.04\) cm. From measurements of the electrolysis current, \(Q_e = (18 \pm 1.5)\cdot 10^{-6}\) coul was obtained.

Thus, the first two assumptions are not confirmed by the experimental data. The most probable explanation, apparently, is the suppression of the anodic current by surface films formed by the products of the electrode reaction with the anions of the supporting electrolyte. Kemula and co-workers [12, 13] found an inhibiting effect of precipitates of lead and cadmium hydroxides on diffusion currents obtained with a dropping amalgam electrode in alkaline solutions. In this case, sharp breaks appeared at the beginning of the wave on the polarogram, and the waves were deformed. In our experiments, the appearance of double peaks can be associated with the formation of a precipitate of copper hydroxides or carbonates on the surface of the mercury drop. Deformation of the anodic peak may be caused by the simultaneous action of several factors: 1) a decrease in current due to a decrease in the concentration gradient; 2) an increase in the rate constant of the discharge process due to an increase in potential; 3) a decrease in current due to thickening of the film; 4) an increase in current due to dissolution of the film in an excess of ions of the supporting electrolyte. From this point of view, an increase in the concentration of copper in the drop (after the solubility product of the precipitate in the near-electrode layer has been reached) causes an intensification of the action of the third factor, while an increase in the concentration of the supporting electrolyte intensifies the fourth factor—the dissolution of the precipitate.

From the literature data [14] it is known that copper ions in alkali and carbonate solutions can form sparingly soluble compounds, which dissolve in an excess of the supporting electrolyte with the formation of complexes. From a comparison of the solubility products for \(\mathrm{CuCO_3}\), equal to \(2.36 \cdot 10^{-10}\) [15], and \(\mathrm{Cu(OH)_2}\), equal to \(5.6 \cdot 10^{-20}\) [15], it is evident that with

at the same background concentration (0.1–1 M), a much lower copper concentration is required for formation of a Cu(OH)\(_2\) precipitate, which is also observed experimentally: in 0.1 N KOH we were unable to obtain a single sharp copper peak even at low concentrations of Cu(2+). The solubility of carbonate precipitates and of copper hydroxides in an excess of background also differs (it is greater for carbonates). This may explain the disappearance of double copper peaks with increasing K\(_2\)CO\(_3\) concentration, in contrast to KOH solution.

Schematically, the effect of the film can be explained by Fig. 6. It is evident that, with equal area, the normal and double peaks must have different half-peak widths, which is in fact observed experimentally (Table 1, column 2).

Fig. 6. Schematic representation of the effect of the film on the shape of the anodic peak: 1—normal peak; 2—in the presence of a film.

The influence of the rate of potential change on the shape of the peaks, from the standpoint of formation of an insoluble film, is reduced to the creation of different conditions for the formation and dissolution of precipitates (since these processes occur in time).

The fact that stirring the solution in the anodic cycle did not affect the shape of the complex copper peak may be connected with the adsorption nature of the films, as has been indicated in the literature [12].

Conclusions

1. The influence on the character of the anodic copper peak of various factors has been studied: the concentration of copper in solution, the time of preliminary electrolysis, the rate of potential change, stirring of the solution, and the nature and concentration of the background.

2. From coulometric calculations it has been shown that the complex character of the anodic dissolution of copper in the backgrounds studied is not associated with stepwise oxidation of copper.

3. It has been shown that the technique of transferring the electrode from one electrolyte to another can be used: (a) for carrying out the accumulation process in an electrolyte in which there are no complications in the cathodic process; (b) for determining the quantity of electricity expended on the deposition of the metal by the method of the standard area under the peak.

4. Possible causes of the complex character of the anodic dissolution of copper from an amalgam in alkaline and carbonate solution have been discussed, and assumptions have been made concerning the formation of passivating precipitates on the surface of the mercury drop.

Literature

1. A. G. Stromberg, E. A. Stromberg. Determination of ultramicroconcentrations of substances in solution by the method of amalgam polarography on a stationary mercury drop. Zavod. labor., 27, 3, 1961.

2. E. A. Zakharova, L. N. Popova. Influence of the conditions of concentration and dissolution on the shape of the anodic lead peak in the method of amalgam polarography with accumulation. Izv. TPI, 128, 64, 1964.

3. S. I. Sinyakova, I. V. Markova. Determination of ultralow contents ...

lead and copper in alkalis and acids using amalgam polarography on a stationary mercury drop. Zavod. laborator., 27, 521, 1961.

1. W. Kemula, E. Rakowska, Z. Kublik. Analysis of traces of impurities in uranium salts. J. Electroan. Chem., 1, 205, 1959/1960.

2. A. A. Kaplin, M. S. Zakharov, A. G. Stromberg. Accelerated determination of microconcentrations of copper and lead in indium by the method of amalgam polarography with transfer. Methods for the Analysis of Chemical Reagents and Preparations, issue 5–6, IREA, Moscow, p. 92, 1963.

3. W. Kemula, Z. Galus. Application of an electrode in the form of a hanging mercury drop to the study of the system $\mathrm{Mn}^{2+}/\mathrm{Mn}(\mathrm{Hg})$. Roczn. Chem., 36, 1223, 1962.

4. W. Kemula. Advances in Polarography, Perg. Press, V. I, 125, 1960.

5. A. Zebreva. Electrochemical properties of liquid amalgams. Proceedings of the Institute of Chemical Sciences, Academy of Sciences of the Kazakh SSR, 9, 55, 1962.

6. D. Royer, I. Kleinberg, A. Davidson. Anodic behavior of copper in aqueous solutions. J. Inorg. Nucl. Chem., 4, 115, 1957.

7. N. Furman, W. Cooper. Study of the polarographic behavior of the dropping amalgam electrode. J. Am. Chem. Soc., 72, 5667, 1950.

8. K. Zh. Sagadieva. Amalgam polarography of some nonferrous and rare metals. Dissertation, 1962.

9. W. Kemula, B. Behr, I. Dojlido. Polarographic study of the anodic dissolution of lead amalgam in alkaline solutions. Roczn. Chem., 31, 205, 1957.

10. W. Kemula, I. Dojlido. Polarographic behavior of cadmium amalgam in alkaline solutions. Roczn. Chem., 33, 451, 1959.

11. L. Meites. Polarographic study of metal complexes. III. Oxalates and carbonates of copper. J. Am. Chem. Soc., 72, 184, 1950.

12. Chemist’s Handbook. Vol. 3, p. 519, Goskhimizdat, 1952.

1. For brevity, below we shall use the expression: area under the peak with stopping. ↩