Techniques for Increasing the Resolving Power of the APN Method. The Technique of Electrode Transfer (Electrolyte Replacement)

A. A. Kaplin, A. N. Pokrovskaya

(Presented by the scientific-methodological seminar of KhTF)

Resolving power, along with sensitivity and accuracy of determination and duration of analysis, is a principal characteristic of analytical methods.

By the resolving power of the APN method [1] is meant the possibility of simultaneously recording two elements on one polarogram at such a minimum difference between the potentials of the anodic peaks and at such a limiting ratio of the concentrations of the ions being determined in solution that measurement of the heights of the anodic peaks with the required accuracy is still possible. Thus, the criteria determining the resolving power of the method are the difference between the potentials of the anodic peaks ($\Delta \varphi_{r}, b$) and the limiting ratio of the concentrations of the elements being determined, which depends on the value of $\Delta \varphi_a$ and on the width of the anodic half-peak.

The necessary choice of $\Delta \varphi_a$ is achieved by changing the nature and composition of the background, and by using complexing agents. As complexing agents, as in classical polarography [4], inorganic and organic acids and bases and their derivatives are used: halides, sulfates, nitrates, phosphates, hydroxides, oxyacids (tartaric and citric) and their salts, rhodanides, cyanides, acetates, amines, complexones, etc.

The choice of a complexing agent solves two problems.

1. Increasing $\Delta \varphi_a$ for two (or more) elements being analyzed. Selection of the complexing agent is carried out experimentally or theoretically. For reversible waves, for this purpose it is necessary to know the value of the half-wave potential ($\varphi_{1/2}$) for the simple ion and the stability constant for complex ions; for irreversible ones—the equilibrium potential of the system ($\varphi_p$), the standard exchange-current density ($i_0$), and the transfer coefficient $\alpha$ [3].

The minimum required $\Delta \varphi_a$ is usually $0.10$–$0.20$ V, depending on the degree of reversibility of the process.

By changing the concentration of the background—complexing agent—it is possible to separate the potentials of the anodic peaks. For example, by decreasing the concentration of hydrochloric acid from 1.2 to 0.2 M, $\varphi_a$ of bismuth shifts from 0.15 to 0.05 V, and that of antimony—from 0.16 to 0.05 V (relative to the SCE). Accordingly, $\Delta \varphi_a$ of antimony and bismuth increases from 0.05 to 0.1 V [2].

Systematic work of great practical value has been begun on compiling spectra of the potentials of anodic peaks [3], analogous to the spectra of half-wave potentials on various backgrounds [4].

1. Elimination of the interfering influence of the base metal, which is reduced at the electrode with formation of an amalgam and, upon anodic dissolution, masks the anodic peak of the ion being determined. By introducing a complexing agent, the reduction potential of the base metal is shifted into the region of potentials considerably more negative than the reduction potential of the ion being determined. For example, the use of a phosphoric-acid background in the determination of impurities in indium leads to a shift of the reduction potential of indium into the region \((-1.4 \div -1.6\ \mathrm{V})\) (vs. S.C.E.) owing to the formation of complexes

\[ [\mathrm{In}(\mathrm{HPO}_4)]^{(3-2)+} \]

or

\[ \mathrm{In}(\mathrm{H}_2\mathrm{PO}_4)_n^{(3-n)+}. \]

The masking influence of tin \((+4)\) on the anodic peaks of copper, lead, and zinc is eliminated by using an alkaline background \((0.1\ \mathrm{N}\ \mathrm{KOH})\), where the reduction potential of tin is shifted into the region of hydrogen evolution [32; 9]. The interfering influence of one of the elements whose determination is not required under the conditions of the analysis is also eliminated. For example, on a background of \(\mathrm{NH}_4\mathrm{Cl} +\) Trilon B, the reduction potentials of lead, cadmium, and indium are shifted into the region of hydrogen evolution, thereby eliminating the interfering influence of these elements on thallium [6].

An increase in resolving power (an increase in the limiting ratio of the concentrations of the interfering element and the one being determined) can be achieved by selecting the accumulation potential on a background where the masking ion is reduced more irreversibly than the ion being determined. The accumulation potential is chosen from the half-wave potentials of both elements or from the curves of anodic current versus accumulation potential. Thus, the optimum accumulation potential for bismuth (Fig. 1) is a potential of \(-0.4\ \mathrm{V}\) (vs. S.C.E.), at which the limiting reduction current of bismuth is reached, while the anodic current of the interfering element (copper) is minimal, although the concentrations of copper and bismuth in the solution are the same [7].

Fig. 1. Dependence of the depth of the anodic notch of Bi and Cu on the accumulation potential: \(1.5\ \mathrm{N}\ \mathrm{HCl}\); \(C_{\mathrm{Cu}} = C_{\mathrm{Bi}} = 0.15\gamma\); \(V = 2\ \mathrm{ml}\); \(\tau = 2\ \mathrm{min}\); \(r_k = 0.04\ \mathrm{cm}\).

With the minimum value of \(\Delta \varphi_a\) achieved by using a complexing agent, registration of peaks with sufficient accuracy is possible only at a definite ratio of ion concentrations in the solution (for example, \(10:1\) for Sb:Bi [2] and \(4:1\) for Bi:Sb [8]). Increasing the concentration ratio above the limiting value leads to partial or complete overlap of the anodic peaks.

The resolving power can be increased by changing the factors that affect the width of the anodic half-peak \((\delta,\ \mathrm{V})\). The latter depends on the reversibility of the electrode process, the number of electrons in the anodic cycle, the rate of change of potential \((w,\ \mathrm{V/sec})\), and the radius of the electrode \((r,\ \mathrm{cm})\) for a mercury drop electrode. As follows from the equation for the anodic peak [9], in order to increase the resolving power of the method (to decrease the time required for the metal to leave the electrode and, correspondingly, to decrease \(\delta\)), it is necessary to decrease the radius of the electrode and the rate of change of potential [9, 10].

According to E. A. Zakharova [11], for lead, when the rate of change of potential is decreased from 1000 to 100 mV/min \((r_k = 0.04\ \text{cm})\), the width of the anodic half-peak decreases from 108 to 43 mV. A change in the radius of a hanging mercury drop electrode (using cadmium as an example) from \(9.6 \cdot 10^{-2}\ \text{cm}\) to \(1.6 \cdot 10^{-2}\ \text{cm}\) leads to a decrease in the half-peak width from 140 to 60 mV (at \(w = 350\ \text{mV/min}\)) [30]. Replacement of the hanging mercury drop electrode by a thin mercury film, several microns thick, deposited on a silver substrate made it possible, along with a 2–3-fold increase in sensitivity, to increase the resolving power of the method [12, 13]. The half-peak width on a film electrode also depends on the rate of change of potential and on the film thickness \((l,\ \mu)\) within a certain range of \(l\) values. According to V. A. Igolinskii [13], for thallium, at a film thickness of \(4\ \mu\), \(\delta\) does not depend on \(w\) over a wide range of potentials: from 1.67 to 640 mV/sec. From the theoretical calculations of de Vries and Van Dalen [14] it follows that, for films 4–10 \(\mu\) thick, \(l\) does not depend on \(\delta\) (in the range 0.3–2 V/min) and is 40–50 mV (for thallium).

At a film thickness of 75–100 \(\mu\), with an increase in the rate of change of potential from 0.3 to 2 V/min, the half-peak width increases from 40–50 to 90 mV. These calculations have not yet been confirmed experimentally.

A decrease in the rate of change of potential and in the electrode radius, along with an increase in resolving power, leads to a decrease in the depth of the anodic peak, which can be compensated by selecting optimal conditions [33]. In the case of partial mutual overlap of the peaks of the ions being determined, when the currents of the diffusion branch of the more electronegative (I) element and of the kinetic branch of the more electropositive (II) element are summed, their separation is possible by stopping the change of potential at the moment when the maximum is reached on the anodic current–potential curve for element I. After the metal has completely left the electrode, the recording device returns to the residual-current line, the change of potential is switched on again, and the peak of element II is recorded [16, 11, 13]. A more effective method than manual stopping of the potential change is the use of an electronic resolving attachment, first proposed for an oscillographic polarograph [13] and used in developing a method for determining antimony in tin in the presence of copper and bismuth [17]. An electronic attachment to the self-recording polarograph 7-77-4B [18] with linearly varying potential has been designed and assembled. The possibilities of using the attachment are shown by the example of determining lead in the presence of thallium in a hydrochloric-acid medium \((\Delta \varphi_a = 0.06\ \text{V})\). It was established that, when the attachment operates in automatic mode, lead and thallium can be determined in a ratio of 8:1, and in forced mode up to 100:1 without corrections and up to 500:1 with corrections according to the correction table [18]. Forced stopping of the potential at a potential somewhat more negative than the equilibrium potential of the more electropositive element makes it possible to determine the latter even in the case when it is completely masked by a considerable excess of the more electronegative, for example, base metal.

When there is an excess of the more electropositive metal, use of the attachment is inexpedient if the equilibrium potential of the more electropositive element lies more negative than the potential of the anodic peak of the more electronegative element, since stopping leads to partial anodic dissolution of the more electropositive element, which is difficult to take into account even with the use of correction tables. If the anodic peak of the more electronegative element does not appear

because, owing to the summation of the currents of both peaks, the attachment does not operate in the automatic mode. If it is not possible to increase \(\Delta \varphi_a\) beyond \(0.06\)—\(0.10\) V by using a resolving attachment or the other procedures described above, methods of chemical separation are employed: extraction and distillation of the base in the form of volatile compounds. Diisopropyl, diethyl, and isopropyl ethers and alkylphosphoric acids are used as extractants (for example, in the separation of a number of impurities from indium [19, 21]). In determining copper, lead, and zinc in tin, the base was distilled off in the form of tin tetrabromide [21]; germanium in the determination of antimony and bismuth was removed in the form of germanium tetrachloride [2]. Coprecipitation and ion-exchange resins are, as yet, practically not used in the ASA method. The use of chemical separation methods, although it gives good results, nevertheless increases the analysis time, the labor involved in the work, the necessary quantity of reagents and glassware, and increases the danger of introducing contaminants.

A possible way of eliminating the interfering influence of the base (or another interfering element) without chemical separation is the electrode-transfer technique, which consists in separating the processes of electrolytic accumulation of the impurities being determined and anodic dissolution of the amalgams, i.e., in carrying out these processes in different supporting media. By this method, in oscillographic polarography with accumulation, cadmium and lead [23] were determined in a \(1:10^5\) and \(1:10^6\) excess of uranium or iron, and lead in blood and copper in milk [23]. To eliminate the interfering influence of hydrogen and copper on the anodic tin peak, Phillips and Shain carried out electrolysis in a \(1.2\) M HCl supporting medium, and electrodissolution in a medium of \(0.1\) KCl \(+\) \(0.01\) HCl [24]. With the use of a \(0.1\) N KCNS medium for electrodissolution it proved possible to determine traces of zinc, copper, nickel, and chromium in a \(\mathrm{Na_2HPO_4}\) solution [25].

However, transfer of the electrode under ordinary conditions (in air and with the circuit opened) led to losses of the impurities concentrated in the drop, probably because of oxidation of the amalgams.

To eliminate losses during transfer, we developed an electrolyzer design in which transfer of the electrode is carried out without opening the circuit and in an inert medium [26, 27]. Later, a cell was described in which the electrolyte is also changed in an inert medium [28].

In the operation of transferring the electrode after electrolytic accumulation, the amalgam drop, together with the electrolytic key and the film of electrolyte wetting them, is transferred, with the circuit closed and at the accumulation potential, in an atmosphere of inert gas, first into an intermediate solution for washing the electrodes and then into the supporting medium in which anodic dissolution of the amalgam is carried out. Losses under such conditions, as experience shows, are practically absent. The transfer technique was described in detail by us earlier [26, 27].

Let us consider in which cases it is advisable to use the electrode-transfer technique to increase the resolving power of the method.

1. Elimination of the interfering influence of the base metal forming an amalgam. Determination of microimpurities in metals that form an amalgam, without chemical separation of the base, is possible if the metal is sufficiently soluble in the selected supporting medium and the anodic peaks of the impurities and of the base metal can be separated at least by using the procedures described above.

If, after dissolution of the base metal in the supporting electrolyte, the anodic peaks of the impurity and of the base coincide, then for anodic dissolution of the amalgams one can choose such a supporting medium in which the potentials of the anodic peaks of the mutually interfering elements differ significantly due to

by means of complexation or a change in the degree of reversibility. For example, in determining lead in indium without chemical separation of the latter, a background of 1 M $H_3PO_4$ can be used, owing to the good solubility of $In(NO_3)_3$ in it and the considerable shift of the potential of the anodic indium peak toward more negative potentials. When the accumulation and anodic dissolution of lead are carried out on this background, in some cases the anodic peak of indium masks the anodic peak of lead (the difference between the peak potentials is 0.05 V). Also, against a background of 2 M $H_3PO_4$, the potentials of the anodic peaks of lead, cadmium, and thallium are close. By choosing, for anodic dissolution (after transferring the electrode), an alkaline ethylenediamine background, we obtain a difference of 0.3 V in the potentials of the anodic peaks of lead and indium, sufficient for determining lead in a considerable excess of indium. On the other hand, in a background of $KOH +$ ethylenediamine (2 ml) it is impossible to dissolve a weighed portion of indium (0.2 g), because of the low solubility of indium nitrate obtained by dissolving metallic indium in nitric acid.

1. Elimination of the interfering effect of a matrix whose cations are reduced on the mercury drop without formation of an amalgam. The interfering effect of a matrix containing such cations as iron, molybdenum, calcium, vanadium, chromium, etc., consists in an increase of the residual current and in difficulty in recording anodic peaks. The use of large compensation currents is not always possible. It is therefore advisable to carry out accumulation on the background of the matrix, and anodic dissolution on a background free of interfering components. An example is the development of a procedure for determining copper in calcium phosphate (see p. 252 of the present collection) and the determination of copper in the presence of manganese [28].

2. Determination of inorganic cations in organic objects. Accumulation is carried out on an organic background, and dissolution on an inorganic one. This makes it possible to use, in the analysis of organic objects, extensive data on the potentials of anodic peaks, the selection of backgrounds for the separate determination of ions, etc., accumulated by amalgam polarography in the analysis of inorganic objects.

3. The electrode-transfer technique makes it possible to select optimal conditions for obtaining anodic waves of the elements being determined, using one and the same background for accumulation and different backgrounds for the anodic dissolution of amalgams. Thus, for example, during accumulation on a background of $H_3PO_4$, copper, lead, thallium, cadmium, and zinc can be concentrated in the mercury drop. For anodic dissolution, after accumulating repeatedly in the same solution, the following backgrounds are used: for copper and lead, $KOH +$ ethylenediamine; for thallium, $NH_4Cl +$ Trilon B [7], etc.

4. In the simultaneous determination of several impurities from a single polarogram, determination of one of the elements may be hindered owing to irreversibility, formation of a double wave, the interfering effect of a neighboring, usually more electronegative element, or mercury. In this case, after recording the polarogram on the first background, one can, after repeated accumulation, carry out anodic dissolution on a background where this element gives an anodic wave suitable for analytical purposes [29].

Procedures have been described in the literature in which, in determining two elements, first on one background a total anodic wave of both elements is obtained, and then, from another weighed portion on another background, the anodic wave of one of the elements is obtained [20]. The magnitude of the wave of the second element is determined from the difference between the total wave and the wave of the first element. When the electrode-transfer technique is used for this purpose, a single weighed portion and two backgrounds are sufficient. In the first background the total anodic wave is recorded; in the second, the anodic wave of only one element is recorded (after transfer of the electrode). However, in order to obtain unambiguous results it is more convenient to select such backgrounds in which

which, after transfer of the electrode, each element would give separate, easily measurable notches.

The electrode-transfer technique can be used for the determination of electronegative elements on backgrounds where recording the anodic peaks of these elements is difficult because of the interfering effect of hydrogen. Accumulation is carried out on the first background in the region of hydrogen evolution, and anodic dissolution of the amalgams—after transfer of the electrode—on a background on which the interfering effect of hydrogen on these peaks is eliminated (alkaline, neutral backgrounds).

1. In principle, a significant increase in resolving power is possible when recording cathodic polarograms in the method of amalgam polarography with accumulation, using the electrode-transfer technique. For this purpose, after accumulation of impurities on the selected background, dissolution is carried out in the film of solution wetting the electrodes during transfer. The cathodic peaks are then recorded. The small volume of liquid on the electrode surface, approaching the volume of the mercury electrode, makes it possible in the cathodic cycle to deposit completely the previously concentrated metal, while at the same time using a large choice of backgrounds for separating the cathodic peaks.

LITERATURE

1. A. G. Stromberg, E. A. Zakharova. Zavod. labor., 30, 261, 1964.
2. E. N. Vinogradova, A. A. Kamenev. ZhAKh, 20, 1 (1965).
3. A. G. Stromberg, M. S. Zakharov. ZhAKh, 19, 2, 1961.
4. T. A. Kryukova, S. I. Sinyakova, Aref’eva. Polarographic Analysis, Moscow, 1959.
5. T. Takahashi, H. Shirai. Analyst, 5, 195 (1963).
6. N. A. Mesyats, M. S. Zakharov, B. F. Nazarov, A. G. Stromberg. ZhAKh, 19, 959 (1964).
7. A. A. Kaplin. Dissertation, TPI, Tomsk, 1966.
8. L. F. Zaichko, V. F. Yankauskas, M. S. Zakharov. Zavod. labor., XXXI, 3, 265, 1965.
9. V. E. Gorodovykh. Dissertation, TPI, Tomsk, 1964.
10. V. E. Gorodovykh. Nastoyashchii sb., p. 104.
11. E. A. Zakharova. Dissertation, Tomsk University, Tomsk, 1965.
12. A. G. Stromberg, V. A. Igolinskii. In: Methods for the Analysis of Chemical Reagents and Especially Pure Preparations, Moscow, 1963.
13. V. A. Igolinskii. Dissertation, TPI, Tomsk, 1963.
14. De Vries, E. Van Dalen. J. Elektr. Chem., 9, 1961.
15. A. G. Stromberg. Zavodskaya laboratoriya, XXXII, 10, 1175, 1965.
16. W. Kemula, L. Galus, Z. Kublik. Nature, 182, 1228 (1959).
17. V. A. Igolinskii, L. F. Zaichko, V. Z. Bashkatov. Izv. TPI, vol. 128, 1964.
18. A. V. Konkova, A. G. Stromberg. Nastoyashchii sb., p. 147.
19. N. A. Mesyats, L. F. Zaichko. Izv. TPI, Tomsk, 128, 1964.
20. S. I. Sinyakova, I. P. Markova, A. G. Dudareva. ZhAKh, 18, 3, 337, (1963).
21. I. S. Levin, L. V. Mishchenko, L. F. Zaichko. Izv. SO AN SSSR, (in press).
22. A. G. Stromberg, M. S. Zakharov, V. E. Gorodovykh, L. F. Zaichko. Zavod. labor., 26, 5, 1961.
23. R. Kalvoda, Proc. Inter-Symp., Pergamon Press, 288 (1961).
24. S. Phillips, I. Shain—Anal. Chem., 34, 262 (1962).
25. M. Van Sway, L. Deelder. Nature, 191, 241 (1961).
26. A. A. Kaplin, M. S. Zakharov, A. G. Stromberg. In: Methods for the Analysis of Chemical Reagents and Preparations, issue 5–6, p. 92, IREA, Moscow, 1963.
27. A. A. Kaplin, E. A. Zakharova. In: Preparation and Analysis of High-Purity Substances, “Nauka,” Moscow, 1966.
28. M. Ariel, U. E. Eisner. J. Electroanal. Chem., 7, 4, 307 (1964).
29. V. M. Yankauskas, V. M. Pichugina, A. A. Kaplin. Zavod. labor. (in press).
30. B. F. Nazarov, A. G. Stromberg. Proceedings of the Conference of Chemists. Tomsk, 1965.
31. S. I. Sinyakova. Proceedings of the Commission on Analytical Chemistry. Moscow, Nauka, 1965.
32. A. G. Stromberg, M. S. Zakharov, V. E. Gorodovykh, L. F. Zaichko. Zavodskaya laboratoriya, 27, 6, 1961.
33. A. A. Kaplin, A. G. Stromberg. Nastoyashchii sb., 108.