PROCEEDINGS
OF THE TOMSK POLYTECHNIC INSTITUTE NAMED AFTER S. M. KIROV,
ORDER OF THE OCTOBER REVOLUTION AND ORDER OF THE RED BANNER OF LABOR

Volume 257
1973

EXPERIMENTAL STUDY OF CARBON DIOXIDE ADSORPTION ON ALUMINUM OXIDE

V. V. Nakhalov, N. F. Stas, G. G. Savel’ev

(Presented by the scientific seminar of the Department
of General and Inorganic Chemistry)

Aluminum oxide readily absorbs carbon dioxide [1]; therefore it can be used as a regenerable adsorbent for purifying gases from $\mathrm{CO}_2$.

Active aluminum oxide is obtained by thermal decomposition of the hydroxide. Depending on the dehydration conditions, the surface of the adsorbent is chemically heterogeneous [2]. As a result, chemisorption processes proceed differently depending on the degree of dehydration.

Studying the IR spectra of $\mathrm{CO}_2$ molecules adsorbed on $\mathrm{Al}_2\mathrm{O}_3$, Fink [3] identified four forms of chemisorption, which at certain temperatures transform into one another:

Form I represents the case of weakly bound $\mathrm{CO}_2$ molecules, which are almost completely removed at room temperature by evacuation. Form II is formed through the binding of $\mathrm{CO}_2$ molecules by $\mathrm{Al}^{+3}$ ions and adjacent oxygen ions; however, the bond is not strong, and in this case carbon dioxide can also for the most part be desorbed by evacuation. Form III is dominant at room temperature; it is formed with the participation of $\mathrm{OH}^{-}$ groups, which are always present on the surface of dehydrated—

... oxide vacuum [4].

The transition of form II to form III at room temperature at the initial moment proceeds at a high rate; however, the state of equilibrium is established only after many hours. Form III is stable up to 325°C at a CO$_2$ pressure of 60 mm Hg and up to 200°C in a vacuum of $10^{-5}$ mm Hg.

Form IV—a surface aluminum carbonate—is formed at temperatures above 200°C. This fact is of particular interest because normal aluminum carbonate is not known under ordinary conditions.

To study the chemisorption of CO$_2$ on aluminum oxide, we used an apparatus based on a MacBain balance with automatic recording of the change in weight [5].

Aluminum hydroxide was obtained from $\mathrm{Al_2(SO_4)_3 \cdot 18H_2O}$ by the usual method [6]. A sample of aluminum hydroxide, thoroughly ground in a mortar, in an amount of 50–100 mg was placed in the reaction vessel and evacuated at room temperature to a pressure of $10^{-3}$ mm Hg.

Decomposition was carried out under nonisothermal conditions at a heating rate of 10–15° per minute in a vacuum of $10^{-3} \div 10^{-4}$ mm Hg. Heating was carried out to 300°C; at this temperature the sample was held until constant weight was reached.

To study the dependence of the adsorption capacity of aluminum oxide on CO$_2$ pressure and temperature, we recorded adsorption isotherms at temperatures of 20, 50, and 80°C (the temperature was maintained constant to within $\pm 0.4^\circ$).

The obtained isotherms are presented in Fig. I.

Fig. I. Adsorption isotherms of CO$_2$ on aluminum oxide;
1 — 80°C, 2 — 50°C, 3 — 20°C.

The isotherms have the typical Langmuir form: as the pressure of CO$_2$ increases, the adsorption value at first rises sharply and then increases ever more slowly. At 760 mm Hg, the adsorption capacity is 76 mg/g at 20°C; with an increase in temperature to 80°C, the adsorption capacity rises to 108 mg/g.

Calculations by the method of Roginskii [7] established that the experimental results give a linear dependence in the coordinates $P/a - f(P)$. This indicates that the isotherms are indeed described by the Langmuir equation

\[ a = a_m \cdot \frac{bp}{1+bp}, \]

where
$a$ is the amount of adsorbed carbon dioxide,
$p$ is the partial pressure of CO$_2$;
$a_m$ and $b$ are constants equal, for 20°, respectively, to 84.8 and 0.0119.

In addition, by the method of Roginskii the isothermal heats of adsorption at 20° and 50°C, corresponding to half-coverage, were calculated.

The values obtained (2090 and 2410 cal/mol, respectively) indicate that the adsorption of CO$_2$ on Al$_2$O$_3$ under these conditions is weakly activated.

Applying to the obtained isotherms the Clausius–Clapeyron equation [7]

\[ Q = \frac{RT_1T_2}{T_1 - T_2}\cdot \ln \frac{P_1}{P_2}, \]

where
$Q$ is the isosteric heat of adsorption;
$P_1$ and $P_2$ are the equilibrium pressures for the same amount of adsorbed gas at temperatures $T_1$ and $T_2$, we determined the heats of adsorption of CO$_2$ on Al$_2$O$_3$ as a function of the degree of coverage $\theta$.

The obtained results (in cal/mol) are presented below:

The results of calculating the heats of adsorption by the Clausius–Clapeyron method and by the Roginskii method ($\theta = 0.5$) are in good agreement. The numerical values of the heats of adsorption indicate, on the one hand, the heterogeneity of the adsorbent surface and, on the other hand, a weakly activated mechanism of CO$_2$ binding.

The kinetic curves for chemisorption of CO$_2$ on aluminum oxide at 20°C are presented in Fig. 2, from which it is seen that the chemisorption rate is greatest at the initial moment: in 30 minutes approximately 50% of the equilibrium amount of carbon dioxide is adsorbed, and equilibrium is reached in 8–10 hours. Evidently, for operation of the adsorbent under real conditions, the initial segment of the kinetic curve is of interest.

Along with adsorption capacity and process kinetics, in the case of regenerable absorbents the regeneration stage is of great importance.

We carried out experiments on vacuum and thermovacuum regeneration on the same aluminum oxide sample. In the first case the adsorbent was regenerated two hours after the start of adsorption. In this case, by vacuum regeneration at a pressure of $10^{-3}$–$10^{-4}$ mm Hg to constant weight, 42.4% of the chemisorbed carbon dioxide is removed. Thermovacuum regeneration with gradual heating to 300°C makes it possible to remove 96.6% of the absorbed CO$_2$.

Fig. 2. Kinetics of carbon dioxide adsorption on aluminum oxide;
1 — on a freshly prepared sample,
2 — after regeneration.

In the second case, regeneration was carried out 14 hours after the start of adsorption. In this case, during evacuation only 18% of the chemisorbed carbon dioxide is desorbed. With subsequent thermovacuum regeneration under the same conditions ($p = 10^{-3} - 10^{-4}$ mm Hg, $T = 300^\circ\mathrm{C}$), the amount of carbon dioxide removed increases to 61%.

The results of the desorption experiments can be explained on the basis of Fink’s ideas concerning the existence of several forms of chemisorption of $\mathrm{CO_2}$ on aluminum oxide.

It is probable that, in the case when desorption is carried out after 2 hours, the predominant form of $\mathrm{CO_2}$ adsorption was form II. In the case when regeneration is carried out after 14 hours, form II apparently passes into the more stable form III, and therefore the amount of desorbed $\mathrm{CO_2}$ decreases by a factor of 2.4 (from 42.4% to 17.9%).

REFERENCES

1. J. Tuul, “J. Phys. Chem.,” 66, No. 9, 1736, 1962.
2. P. Fink, “Z. Chem.,” 7, 284, 1967.
3. P. Fink, “Z. Chem.,” 7, 324, 1967.
4. S. J. Gregg, J. D. Ramsay, “J. Phys. Chem.,” 73, 1243, 1969.
5. G. G. Savel’ev, V. V. Bordachev. Izv. TPI, 176, 147, 1970.
6. D. V. Karjakin, I. I. Angelov. Pure Chemical Reagents. Goskhimizdat, Moscow, 1955.
7. L. S. Kharitonova. Methods for Determining the Heat of Adsorption on Metals. Moscow, 1968.