PHYSICAL CHEMISTRY

V. A. LUNENOK-BURMAKINA, A. P. POTEMSKAYA
and Corresponding Member of the Academy of Sciences of the USSR A. I. BRODSKII

INVESTIGATION OF THE MECHANISM OF ANODIC FORMATION OF OZONE FROM SULFURIC ACID SOLUTIONS

Very little is known about the mechanism of anodic ozone formation. By analogy with its formation in a silent discharge, one may assume the gross scheme \( \mathrm{O_2 + O = O_3} \) \((^1)\). The participation of water is more probable

\[ \mathrm{H_2O + O_2 = O_3 + 2H^+ + 2e} \tag{1} \]

with possible intermediate formation of oxygen ions or radicals. Investigation of anodic voltage curves and of the state of the electrodes led a number of authors \((^{2-4})\) to the conclusion that, in anodic formation of \(\mathrm{O_2}\) and anodic oxidation, oxygen participates in the form of surface oxides. In particular, V. I. Veselovskii et al. \((^5)\) suggest, for platinum electrodes at high overvoltages corresponding to the production of ozone, the formation of higher oxides of the peroxide type, which with water yield ozone or oxygen:

\[ \mathrm{PtO + 2H_2O = PtO[O - O] + 4H^+ + 4e;} \tag{2a} \]

\[ \mathrm{PtO[O - O] + H_2O = PtO + O_3 + 2H^+ + 2e;} \tag{2b} \]

\[ \mathrm{PtO[O - O] = PtO + O_2.} \tag{2c} \]

This scheme differs from (1) in that the oxygen evolved at the anode does not directly participate in the formation of ozone. On the other hand, we previously found \((^6)\) that, during electrolysis of \(\mathrm{KHSO_4}\) solutions, the yield of persulfate decreases sharply upon addition of \(\mathrm{H_2O_2}\), and the latter proves to be the main source of anodic \(\mathrm{O_2}\). As the concentration of \(\mathrm{H_2O_2}\) decreases, the persulfate yield increases and, in parallel, the fraction of oxygen from water in anodic \(\mathrm{O_2}\) increases. This made it possible to suppose that the higher surface oxides of platinum are formed not only according to (2a) from water, but also from peroxide compounds, for example

\[ \mathrm{PtO + H_2O_2 = PtO[O - O] + 2H^+ + 2e,} \tag{3} \]

and that these peroxide compounds, or their precursors in the anodic reaction, also participate in ozone formation.

To investigate anodic ozone formation we used the heavy isotope of oxygen, which we had already successfully employed earlier \((^{2,6-10})\) for the study of anodic processes.

Electrolysis was carried out in a vessel connected to a vacuum apparatus for separation of ozone and analysis of the anodic gas. After freezing out water, the latter was passed through a trap with 5% \(\mathrm{KI}\), where the ozone was collected and titrated, while the oxygen was measured by volume. For isotopic analysis, the gas entered another branch of the apparatus, where, after additional drying, the ozone was frozen out, degassed by pumping, purified by distillation into another vessel, where it was thawed and decomposed by heating to \(150^\circ\). The anodic oxygen and the oxygen from ozone were analyzed for \(\mathrm{O^{18}}\) content in a mass spectrometer. The completeness of ozone decomposition was checked by control

experiments on increasing the pressure by a factor of \(1^{1}/_{2}\) (with an accuracy of 5%) after heating the vessel in which the collected liquid ozone had been evaporated. The completeness of separation was also checked: ozone containing no excess \(O^{18}\) was frozen out from a mixture with \(O_2^{18}\). To determine \(O^{18}\) in \(H_2SO_4\) and \(KHSO_4\), they were converted into \(PbSO_4\), which was heated with degassed carbon at \(700^\circ\). The resulting \(CO_2\) was introduced into the mass spectrometer. The content of \(O^{18}\) in water was also determined in the form of \(CO_2\) after exchange with steam \((^{11})\). Heavy \(H_2SO_4^{18}\) was prepared by adding anhydrous acid to \(H_2O^{18}\). From it, \(KHSO_4^{18}\) was obtained by neutralization with KOH. An electrolyte containing 4–10 g of monoperoxysulfuric acid (Caro’s acid) in 60–70 ml of 5–6.5 \(M\) \(H_2SO_4\) solution in \(H_2O^{18}\) was prepared by dissolving 40 g of \(K_2S_2O_8\) powder with 26 ml of conc. \(H_2SO_4\) under strong cooling, after which 50 ml of cooled \(H_2O^{18}\) was added. After standing for 1 h at \(20^\circ\), the liquid was filtered. Electrolysis was carried out at \(20^\circ\), with a current of 4–7 A, at a voltage of 11 V, an anodic current density of 1–1.5 A/cm\(^2\), and a volumetric current density of 0.1–0.2 A/ml. The hollow platinum anode was additionally cooled with running water. Before the experiment it was washed in nitric acid and repolarized with a current of 0.1 A/cm\(^2\). The determination of \(S_2O_8^{2-}\) and \(SO_5^{2-}\) in the electrolyte was carried out iodometrically \((^{12,13})\). Under the experimental conditions, appreciable amounts of \(H_2O_2\) were not formed, and ozone condensed, depending on the experimental conditions, at 2–6 mg/h with oxygen evolution of about 1 l/h. In experiments with 4.5–6.8 \(M\) \(H_2SO_4\), during electrolysis 0.004–0.008 moles of \(H_2SO_5\) (per 40–60 ml) were formed as a result of hydrolysis of persulfate.

During electrolysis of saturated solutions of \(KHSO_4\) in \(H_2O^{18}\) or \(KHSO_4^{18}\) in \(H_2O\), as well as 2.5 \(M\) \(H_2SO_4\) in \(H_2O^{18}\), the ozone had the isotopic composition of the oxygen of water, while in the sulfate the oxygen composition did not change (Table 1, experiments 1–3).

Table 1

Isotopic composition of the electrolyte, oxygen, and ozone

It was found earlier \((^{6,8})\) that oxygen from water does not enter the simultaneously formed persulfate. These data exclude the participation of persulfate (or its possible precursors) in ozone formation. In agreement with this, the addition of persulfate to the electrolyte does not increase the ozone yield \((^{1})\).

In the electrolysis of more concentrated H₂SO₄ in H₂O¹⁸, the anodic oxygen is diluted with sulfate oxygen to a greater extent, the higher the acid concentration, while ozone has an isotopic composition intermediate between O₂ and water. A. N. Frumkin² found the same intermediate composition in the electrolysis of an HClO₄ solution. With increasing acid concentration, the fraction of O¹⁸ in it that passes from the water increases. This is caused not by exchange, which under these conditions proceeds very slowly¹⁴, but by the hydrolysis of persulfate, in which, as we found earlier¹⁰, oxygen from water enters the sulfate formed. This hydrolysis is favored by the high acid concentration. As a result of it, light H₂SO₅ is formed, the anodic decomposition of which dilutes O₂¹⁸ with light oxygen. In experiments 4 and 5 with 4.5 and 6.8 M H₂SO₄, the content of O¹⁸ in ozone is equal to that calculated from (1). For a more reliable verification of this equation, we increased the difference between the isotopic composition of the anodic oxygen and that of water by carrying out electrolysis with additions of Caro’s acid (experiments 6 and 7) or hydrogen peroxide (experiments 8 and 9). In the experiments with H₂O₂, its constant concentration of about 2% was maintained by adding 80% H₂O₂ dropwise to the electrolyte. It is evident from Table 1 that in all experiments the 1 : 2 ratio of oxygen from water and from anodic O₂ in the ozone formed is observed with good accuracy.

From the experiments described it follows that the formation of ozone and persulfate are two independent parallel processes and that ozone is formed according to the schematic stoichiometric equation (1). Taking into account the electrochemical and other data mentioned above, one must consider that in this process oxygen actually participates in the form of higher surface oxides, according to equation (2b), but these oxides are formed not only from water according to (2a), but in the presence of peroxy compounds also (and even predominantly) with the participation of the latter, for example according to (3) in the case of the presence of H₂O₂. From these same oxides, according to (2c), anodic O₂ of the same isotopic composition as theirs is formed.

As indicated above, steps (3) and (2c) also explain the predominance of oxygen from H₂O₂ in the anodic O₂ obtained in the electrolysis of sulfate or sulfuric acid solutions in the presence of hydrogen peroxide. Apparently, our results can be explained only by such a mechanism.

Institute of Physical Chemistry
named after L. V. Pisarzhevsky
Academy of Sciences of the Ukrainian SSR

Received
23 I 1961

CITED LITERATURE

1. E. Briner, R. Halfeli, H. Paillard, Helv. chim. acta, 20, 1510 (1957); E. Briner, A. Yalda, Helv. chim. acta, 24, 1829 (1943).
2. A. N. Frumkin, Intern. Committee Elektrochem., Thermodyn. and Kinetics (Proc. 9th Meeting, Paris, 1957), London, 1959, p. 396.
3. R. I. Kaganovich, M. A. Gerovich, E. Kh. Enikeev, DAN, 108, 107 (1956).
4. V. G. Ginzburg, V. I. Veselovskii, ZhFKh, 24, 366 (1950); K. I. Rozental’, V. I. Veselovskii, ZhFKh, 27, 1163 (1953); G. I. Borisov, V. I. Veselovskii, ZhFKh, 27, 1195 (1953); L. M. Volchkov, L. G. Antonova, A. I. Krasil’shchikov, ZhFKh, 23, 441, 714 (1949); 27, 512 (1953).
5. A. A. Rakov, V. I. Veselovskii et al., ZhFKh, 32, 2703 (1958).
6. A. I. Brodskii, I. F. Franchuk, V. A. Lunenok-Burmakina, DAN, 115, 934 (1957).
7. K. I. Rozental’, V. I. Veselovskii, DAN, 111, 637 (1957).
8. A. N. Frumkin, R. I. Kaganovich et al., DAN, 102, 981 (1955).
9. M. A. Gerovich, R. I. Kaganovich et al., DAN, 114, 1049 (1957).
10. I. F. Franchuk, A. I. Brodskii, DAN, 118, 128 (1958).
11. A. I. Brodskii, S. G. Demidenko et al., Zhurn. anal. khim., 10, 256 (1955).
12. A. I. Brodskii, A. S. Afanas’ev, M. G. Dikova, Zhurn. prikl. khim., 5, 929 (1932).
13. I. A. Bondin, Zav. lab., 7, 1248 (1938).
14. T. C. Hoering, I. W. Kennedy, J. Am. Chem. Soc., 79, 56 (1957).