Effect of the Preparation Method of an Oxide–Cobalt Catalyst on Its Activity in Ammonia Oxidation Under Elevated Pressure

N. P. Kurin, P. E. Bogdanov

(Presented by Prof. L. P. Kulev, Doctor of Chemical Sciences)

It is known that oxide–cobalt catalysts at atmospheric pressure give a high degree of oxidation of ammonia to nitric oxide and are of considerable interest as replacements for expensive platinum catalysts in nitric-acid production [1, 2, 3, 4, 5, 6]. However, the literature contains no data concerning the oxidation of ammonia in the presence of non-platinum oxide catalysts under elevated pressure, except for the indication that under these conditions they become extinguished [3, 4].

The purpose of the present work is to study the influence of the method of preparing $\mathrm{Co_3O_4}$ on its catalytic properties in the oxidation of ammonia under a pressure of $8\ \mathrm{kg/cm^2}$, at which a number of domestic nitric-acid plants operate.

The absence of a sufficiently well-developed theory of catalytic processes does not permit the rational method for preparing active catalysts to be established in advance, and therefore, up to the present time, in developing catalysts for practical use it has been necessary to proceed experimentally.

For the preparation of oxide–cobalt catalysts in the form of $\mathrm{Co_3O_4}$, we used two methods: (a) the method of thermal decomposition of salts and (b) the precipitation method.

The first method excludes contamination of the resulting $\mathrm{Co_3O_4}$ by impurity ions; the second method, while giving a finely dispersed precipitate, does not exclude the possibility of some entrainment by the product of foreign ions from the solution.

The first two samples were obtained by thermal decomposition of chemically pure $\mathrm{Co(NO_3)_2 \cdot 6H_2O}$ and $\mathrm{CoC_2O_4}$ at $550$–$600^\circ$C, followed by grinding and pressing at $1000\ \mathrm{kg/cm^2}$.

The third sample was prepared through cobalt hydroxide. For this purpose, from a hot one-normal solution of cobalt nitrate, a rose-red cobalt hydroxyoxide was precipitated by means of a potassium hydroxide solution. The precipitate was filtered off, washed until the reaction for $\mathrm{NO_3'}$ ions with diphenylamine disappeared, and oxidized with a 10-percent hydrogen peroxide solution to $\mathrm{Co(OH)_3}$. The latter was washed with distilled water, dried at $105^\circ$C, pressed into tablets, and heated at $200^\circ$ for 2 hours, at $400^\circ$ for 2 hours, and at $550$–$600^\circ$ for 3 hours.

In all cases, for charging into the contact apparatus the 1.5–2 mm fraction was selected.

The characteristics of the catalysts are given in Table 1. For cobalt protoxide–oxide obtained by precipitation, the specific surface area was approximately determined by the methylene-blue adsorption method [13]; for the freshly prepared sample it was 50 m²/g, and for the spent sample, 39.8 m²/g.

Table 1

Characteristics of the catalysts

In studying the catalytic properties, an ammonia–air mixture was continuously passed through the catalyst layer, and the degree of oxidation of NH₃ to NO was determined by the evacuated-bulbs method [3].

Results Obtained and Their Discussion

The first to be studied were the catalytic properties of the oxide cobalt catalyst prepared by thermal decomposition of cobalt nitrate. At a temperature of 700°C and a layer height of 35 mm, the maximum degree of oxidation of NH₃ to NO at the optimum space velocity¹) of 20,000 was only 15%.

Subsequently the height of the catalyst layer was increased to 75 mm in the hope of obtaining better results. In this case, at space velocities of 5000–10,000, a temperature of 500°C, and \(C_{NH_3} = 9.2\%\), the degree of oxidation was 50%; at higher temperatures this value was considerably lower, and at 680°C, close to the range of temperatures important for technology, the degree of oxidation at space velocities of 15,000–20,000 remained very low (22–32%).

At a catalyst-layer height of 75 mm, increasing the ammonia content from 9.7 to 13.8% (by volume) at a temperature of 640° causes a further decrease in the degree of oxidation.

In the presence of Co₃O₄ prepared by thermal decomposition of cobalt oxalate, the highest yield of nitric oxide at 8 kg/cm²

¹) By space velocity is meant the amount of gas, in m³, passing through 1 m³ of bulk catalyst volume per hour under normal conditions.

(\(C_{\mathrm{NH_3}} = 10.2\%\)) was obtained at 700°, a space velocity of 9150, and amounted to 68.3%. At a temperature of 600° and the same space velocity, the degree of oxidation did not exceed 60%. A further increase in velocity leads to ammonia slip through the catalyst.

The process of ammonia oxidation in the presence of \(\mathrm{Co_3O_4}\), prepared by the precipitation method, was studied at various temperatures, space velocities, ammonia concentrations, and at three different layer heights. As can be seen from Fig. 1, a, at a layer height of 40 mm the maximum degree of oxidation of \(\mathrm{NH_3}\) to NO at a temperature of 700°C was 65%.

Fig. 1. Dependence of the degree of oxidation of \(\mathrm{NH_3}\) to NO on the space velocity in the presence of \(\mathrm{Co_3O_4}\), prepared by the precipitation method with KOH. Layer height: a — 40 mm; b — 60 mm; v — 75 mm; g — 40 mm + asbestos layer.

With an increase in the layer height to 60 mm (Fig. 1, b), the process of ammonia oxidation at a pressure of 8 kg/cm² becomes more complete: the maximum degree of oxidation of ammonia to nitric oxide at 700 and 720° and an \(\mathrm{NH_3}\) concentration of 10.5% (by volume) rises, respectively, to 92 and 95%. At the same time, the productivity of the catalyst increases sharply, which is characterized by the possibility of attaining the indicated high degrees of oxidation at space velocities \(V = 150000—220000\).

With a further increase in the catalyst layer height to 75 mm at a temperature of 700°C (Fig. 1, v), the degree of oxidation (\(\alpha\)) decreases; however, the curve \(\alpha = f(V)_{C_{\mathrm{NH_3}}, t, P}\) runs almost parallel to the abscissa over a wide range of space velocities, 50000—250000. The maximum degree of oxidation under these conditions is 85%.

The effect of the catalyst layer height (\(h\)) is shown more clearly in Fig. 2, from which it is evident that the curves \(\alpha = f(h)_{C_{\mathrm{NH_3}}, V, t, P}\) pass through a maximum located at \(h = 60\) mm.

Variations of the process in order to determine the optimum oxidation conditions at a layer height of 60 mm showed that at a high space velocity of 218000 and a temperature of 720°C, the optimum ammonia concentration lies within 9.2—9.7% (by volume), and the attained

the degree of oxidation increased to 97–98% (Fig. 3). Along with a high degree of oxidation, the process is characterized by great stability.

Comparing the data obtained in the oxidation of ammonia in the presence of three samples of cobalt oxide, it must be recognized that

Catalyst-layer height, mm

Fig. 2. Dependence of the degree of oxidation of NH₃ to NO on the height of the layer of Co₃O₄ catalyst prepared by precipitation with KOH.

NH₃ content in the ammonia–air mixture, vol. %

Fig. 3. Dependence of the degree of oxidation of NH₃ to NO on the ammonia concentration ($C_{\mathrm{NH}_3}$, vol. %) in the initial ammonia–air mixture at a catalyst-layer height of 60 mm.

the best results in degree of oxidation and productivity were achieved in the presence of Co₃O₄ obtained by the precipitation method. For it, at a pressure of 8 kg/cm², an ammonia concentration of 10.5%, and 700°C, the dependence of the degree of oxidation ($\alpha$) on the space velocity ($V$) can be expressed by the empirical equation

\[ \alpha = 55.63 + 44.4 \cdot 10^{-5} \cdot V - 13 \cdot 10^{-10} \cdot V^2 \ldots \]

Taking into account the difficulties of regulating the temperature regime when carrying out the ammonia oxidation process at a catalyst-layer height of 40 mm and the reduced degrees of oxidation, we suggested that this is connected with considerable heat losses from the upper layers of the catalyst. To retain heat in the upper catalyst layer, experiments were carried out with a layer of thermal insulation made of low-activity asbestos placed above the catalyst (the gas stream passed through the catalyst from top to bottom). The height of the asbestos layer was 40 mm. The experiments showed that, in this case, the process of ammonia oxidation to NO improves. This is expressed in a certain increase in the degree of ammonia oxidation to nitric oxide and in the possibility of operating at considerably higher space velocities (Figs. 1, 2). Thus, at an ammonia concentration of 10.5% and a temperature of 750°C, without thermal protection of the upper catalyst layers, the degree of oxidation and the optimum space velocity were, respectively, 55% and 50,000 (curve 3, Fig. 1a), whereas with thermal protection (curve 1, Fig. 1g), all other conditions being equal, these values increased to 70% and 125,000.

Thus, thermal insulation of the upper layers of the catalyst, when the height of its bed is small, makes the process more stable and permits operation at considerably higher space velocities, with attainment of a higher degree of oxidation of ammonia to nitric oxide.

Two-stage oxide catalyst

D. A. Epshtein and N. M. Tkachenko [7] investigated, at atmospheric pressure, a two-stage catalyst consisting of one platinum–rhodium gauze and an iron–bismuth oxide catalyst. The degree of ammonia oxidation ($\alpha$) was 94–96%.

Later this proposal was used for the oxidation of ammonia at atmospheric pressure [4, 8].

Fig. 4. Contact tube with a two-stage catalyst:
$a$—$Co_3O_4$, prepared by precipitation with KOH;
$b$—$Fe_2O_3$—66.6; $Bi_2O_3$—19.4; $MnO_2$—12% (by weight).

Labels in the figure: ammonia–air mixture; refractory mixture; asbestos; thermocouple sheath; quartz tube; nitrogen gases.

Fig. 5. Dependence of the degree of oxidation of $NH_3$ to NO on the space velocity in the presence of a two-stage oxide catalyst.

Graph labels: degree of oxidation of $NH_3$, %; space velocity: 1000.
Conditions shown: $p = 8$ kg/cm²; $C_{NH_3} = 9.7$ vol.%;
1—($t = 650^\circ$); 2—($t = 700^\circ$); 3—($t = 750^\circ$); 4—($t = 780^\circ$).

Fig. 6. Dependence of the degree of oxidation of $NH_3$ to NO on the ammonia concentration ($C_{NH_3}$—vol.%) in the initial ammonia–air mixture in the presence of a two-stage oxide catalyst.

Graph labels: degree of oxidation of $NH_3$, %; $NH_3$ content in the ammonia–air mixture, vol.%.
Conditions shown: $p = 8$ kg/cm²; $t = 760^\circ$; $v = 17150$.

In the postwar period, two-stage oxidation of ammonia at atmospheric pressure in the presence of platinum and non-platinum catalysts was introduced into production at one of the domestic nitric-acid plants [9, 10]. It is indicated that on such a catalyst the degree of oxidation of ammonia is the same as on a platinum–rhodi-

meshes [10]. Along with this, platinum losses are reduced somewhat (from 0.05–0.55 g/t HNO₃ to 0.04 g/t HNO₃), but they remain high.

In this connection, it was of interest to use, as a charge, the most active cobalt catalyst found by us, placing it over a less active oxide iron–bismuth–manganese catalyst [11]. Onto a layer of iron–bismuth–manganese catalyst (Fe₂O₃—68.6; Bi₂O₃—19.4; MnO₂—12%, 60 mm high, Fig. 4,b), an oxide–cobalt catalyst obtained by the precipitation method was charged (Fig. 4,a). The total height of the oxide two-stage catalyst was 87 mm.

The character of the dependence of the degree of oxidation of ammonia to nitric oxide (α) on the space velocity, the ammonia concentration (\(C_{\mathrm{NH_3}}\)), and the pressure (Figs. 5, 6, 7) in the presence of the two-stage catalyst studied by us remains the same as that observed for the various types of oxide catalysts described in [11].

Fig. 7. Dependence of the degree of oxidation of NH₃ to NO on the change in pressure in the presence of a two-stage oxide catalyst.

At a content in the initial mixture of 9.7% (by volume) NH₃ (the optimum concentration when conducting the oxidation process in the presence of an oxide cobalt catalyst obtained by the precipitation method) and a pressure of 8 kg/cm², the maximum degree of oxidation of NH₃ to NO in the presence of the two-stage oxide catalyst was obtained at 750°C, a space velocity of 50,000, and amounted to 84.8% (Fig. 5).

The effect of the two-stage process of ammonia oxidation in the presence of oxide cobalt and iron–bismuth–manganese catalysts is expressed in an increase in the degree of ammonia oxidation and in the possibility of operation at higher space velocities. Thus, at a pressure of 8 kg/cm², a space velocity of 25,000, a temperature of 700°C, an ammonia concentration of 10.3%, and a bed height of 40 mm in the presence of an oxide cobalt catalyst (curve 2, Fig. 1,a), the degree of oxidation was 54%. Under the same conditions and with a bed height of 87 mm, the oxide iron–bismuth–manganese catalyst [11] provided a degree of oxidation equal to 80%. Two-stage oxidation (curve 2, Fig. 5) under approximately the same conditions increased the degree of oxidation to 83.5%.

The optimum space velocities in the presence of the iron–bismuth–manganese catalyst with 12% MnO₂ [11] were about 20,000. This value for the precipitated cobalt catalyst, 40 mm high, was 50,000. The two-stage catalyst could operate at the same increased space velocity as the oxide–cobalt catalyst (Figs. 1,a and 5).

The contact time, calculated according to the equation of N. P. Kurin,

\[ \tau=\frac{\beta \cdot P \cdot 273[10000-C_{\mathrm{NH_3}}\cdot(125+1.75\alpha)]\cdot 3600}{10000\cdot V\cdot(273+t)}, \]

where \(\beta\) is the fraction of free space in the contact zone (~0.5);
\(P\) is the pressure in the system, in kg/cm²;
\(C_{\mathrm{NH_3}}\) is the concentration of ammonia in the ammonia–air mixture (% by volume);

\(\alpha\)—degree of oxidation of ammonia, in %;
\(V\)—space velocity;
\(t\)—experimental temperature, in °C;
\(\tau\)—time, in seconds, for the iron–bismuth–manganese catalyst with 12% \(MnO_2\), for the cobalt catalyst obtained by the precipitation method, with a bed height of 60 mm, and in the presence of a two-stage catalyst under optimum conditions, proved to be equal, respectively, to 0.158, 0.012, and 0.059 seconds; that is, even in the most favorable case (the cobalt catalyst obtained by coprecipitation), this value is 20–100 times greater than for the platinum catalyst.

The maximum degree of oxidation on the two-stage oxide catalyst, equal to 84.8%, was nevertheless lower than that obtained when the process was conducted on a pure cobalt catalyst under optimum conditions and with a bed height of 60 mm (95–98%).

However, in our opinion this is connected not with the principle of a two-stage system, but with the insufficiency of experiments in developing the optimum operating conditions for the two-stage catalyst. Proper selection of the bed heights of the igniting cobalt catalyst and of the less active iron–bismuth–manganese catalyst, as well as the development of optimum conditions with respect to space velocities, ammonia concentrations, and temperatures, will ensure high degrees of oxidation at lower total catalyst costs. It may be possible to carry out a highly efficient process of ammonia oxidation under pressure in the presence of a multistage contact consisting of separate layers (more than two) of nonplatinum catalysts differing in nature.

As for the mechanism of ammonia oxidation in the presence of cobalt catalysts, we assume that the most important stage of this process is the interaction of adsorbed ammonia with oxygen in the surface layers of the \(Co_3O_4\) crystal lattice, with formation of hydroxylamine and subsequent course of the process according to the scheme described in the communication on iron–bismuth–manganese catalysts [11].

The reasons responsible for the enhanced catalytic properties of the cobalt catalyst obtained by the coprecipitation method are associated with a greater development of its reaction-accessible surface and with a change in its activity in light of the theory of semiconductor catalysis [12], owing to possible occlusion by the \(Co_3O_4\) precipitate of small amounts of potassium ions that are difficult to remove by washing.

Conclusions

1. Three oxide–cobalt catalysts prepared by thermal decomposition of cobalt nitrate and cobalt oxalate and by precipitation with caustic potash were investigated.

2. It was shown that the contact prepared by precipitation with a solution of caustic potash possesses the greatest activity. On this catalyst, under a pressure of 8 kg/cm², a degree of oxidation of ammonia equal to 95% was achieved at a space velocity of 225000, a temperature of 720°, and with 10.5% (vol.) \(NH_3\) in the initial mixture.

3. With an increase in the bed height of the precipitated catalyst, other conditions being constant, the degree of oxidation at first increases and then decreases. The optimum catalyst bed height is 60 mm.

The conditions for oxidation of \(NH_3\) to NO at a small height of the contact layer are improved when the upper layers of the catalyst are protected from heat losses.

1. A two-stage catalyst was studied. When a cobalt catalyst layer is used as the ignition layer, the productivity of the iron–bismuth–manganese catalyst is increased, thereby creating fundamental possibilities for increasing the productivity per unit of catalyst space without the use of platinum.

2. The oxide cobalt catalyst obtained by precipitation with caustic potassium from a solution of cobalt nitrate is highly promising for carrying out the process of ammonia oxidation to nitrogen oxides under a pressure of 8 kg/cm².

REFERENCES

1. Scott, W. W., Ind. Eng. Chem., 16, 75, 1924.
2. Scott, W. W., Leech W. D. Ind. Eng. Chem., 19, 170, 1927.
3. Atroshchenko V. I., Kargin S. I. Technology of Nitric Acid, Goskhimizdat, 1949.
4. Boreskov G. K., Bakhurov V. G. Materials on Technical Information of GIAP, February 1950.
5. Atroshchenko V. I., Zasorin A. P. Proceedings of the Kharkov Polytechnic Institute, 4, 115, 1954.
6. Griesbach R., Reinhardt P. D., Z. anorg. allg. Chem., 5, 241, 1955.
7. Epshtein D. A., Tkachenko N. M. Zh.P.Kh., 11, 5, 1938.
8. Arsen’eva L. Z. Production of Nitric Acid in Germany, GIAP, 1949.
9. Mel’nikov E. Ya. Zh. P. Kh., 30, 1439, 1957.
10. Minovich M. A., Belyaev N. I. Chemical Science and Industry, 1, 661, 1956.
11. Kurin N. P., Bogdanov P. E. Study of the process of ammonia oxidation under pressure in the presence of iron–bismuth–manganese catalysts (article placed in this collection).
12. Roginskii S. Z., Chemical Science and Industry, 2, 138, 1957.
13. Brunauer S., Adsorption of Gases and Vapors, vol. 1, 1948.