M. V. Kharitonova.

Irreversible Adsorption of Ba″ by Silicic Acid and Its Relation to Reversible Adsorption.

(From the laboratory of inorganic chemistry).

(The article contains 2 diagrams).

The usual method for determining the amount of adsorbed substance is as follows: to a definite amount of adsorbent there is added a definite amount of adsorptive of known concentration; the mixture is shaken, and then either centrifuged, or simply allowed to stand until the liquid settles, and the concentration of the adsorptive is again determined. The difference in concentrations is the amount of substance adsorbed. Very many investigations of this kind have been carried out in recent years and, although much in this phenomenon still remains unclear, in any case a dependence between the amount of absorbed substance and the concentration of the solution may be regarded as established. In the few works in which the final concentration of the solution was determined after filtration and careful washing of the solid phase, the picture obtained is somewhat different. It must be stipulated that all investigations of the latter type concern works studying adsorption during coagulation, and it is possible that a simultaneous observation of two phenomena is taking place. To this type of work belong the studies of Linder and Picton¹) on the coagulation of arsenic sulfide; they come to the conclusion that the amount of adsorbed substance does not depend on the concentration. Michaelis²), investigating the coagulation of mastic, finds that the amount of substance adsorbed is almost independent of the concentration. Witney and Ober³) find it possible to give the following formula for coagulated arsenic sulfide Ba(As₂S₃)₉₀. Freundlich⁴) considers that the study of adsorption accompanied by washing of the solid phase is incorrect, since the collateral phenomenon of hydrolysis is introduced. But it is possible that those studying adsorption during coagulation without washing observe simultaneously three phenomena: adsorption during coagulation, so to speak internal, and then external adsorption, which can be observed in almost any immersion of a finely divided solid phase in a solution, and which consists of a part of the substance adsorbed and easily washed out—reversibly adsorbed—and a part not washed out—irreversibly adsorbed; for a number of investigators of pure adsorption, working with such adsorbents as charcoal (Wippen⁵)), BaSO₄ (Wohlers⁶)), CaCO₃, MgCO₃, BaSO₄, BaCrO₄ (P. Marc⁷)), SiO₂ (Jordis⁸)) and others, state that it is not possible to wash out the adsorbed substances completely; a small part remains firmly bound.

It may be that this non-washable part represents the first layer of adsorbed substance, held by electrochemical forces, and then this phenomenon must be strictly parallel to reversible adsorption.

1) Cited according to Duclaux “thèse”.
2) Zeit. Biochem. 1907 — 6.
3) Zeit. phys. ch. 1901. 39.
4) Zeit phys. ch. 1909 г. 7.
5) Lieb. Ann. 1846. 59.
6) Zeit. anorg. ch 1908. 59.
7) Zeit. phys. ch. 1910. 75.
8) Zeit. Elect. 1905. 11.

It is possible that this is the result of hydrolysis, as Reychler¹ and Freundlich² suppose, and then the amount irreversibly adsorbed should not increase with concentration.

It is possible that these are two different phenomena, as Weimarn believes: that, when a solid phase is immersed in a liquid one, the former retains from the solution a portion of the substances by chemical forces—physicochemical and physical.

Arrangement of the experiments.

As the adsorber I took silicic acid. The silicic acid was prepared from kieselguhr and caustic soda in an autoclave; the soluble glass obtained was thoroughly washed and then treated two more times in succession with caustic soda and hydrochloric acid. The silicic acid was dried in air, ground in a mortar, sifted through a sieve, and kept in a bottle with a ground-glass stopper.

Its composition: $\mathrm{SiO_2}$ — 71.35%, $\mathrm{H_2O}$ — 28.5%, impurities — 0.15%.

As adsorbent the following salts were taken: $\mathrm{BaCl_2}$, $\mathrm{BaBr_2}$, $\mathrm{Ba(NO_3)_2}$, $\mathrm{Ba(ClO_3)_2}$, $\mathrm{Ba(C_2H_3O_2)_2}$, from Kahlbaum. Barium salts were chosen for the reasons that $\mathrm{Ba^{\bullet\bullet}}$ is a good adsorbent, on the one hand, and has a reliable method for quantitative determination, on the other.

10 g of air-dry silicic acid were placed in a bottle with a ground-glass stopper; from a burette 50 c.c. of a barium salt solution of known concentration were added to it, and it was shaken well by hand at definite intervals of time at room temperature. After 24 hours, in order to determine the reversible adsorption, a portion of the settled liquid was taken with a pipette and its concentration was investigated. To determine the irreversible adsorption, the contents of the bottle were transferred to a filter, filtered off, and washed until the reaction for $\mathrm{Ba^{\bullet\bullet}}$ disappeared. This was always preceded by a blank experiment. In all experiments, the distilled water used was freshly distilled with $\mathrm{Ba(OH)_2}$ to remove carbon dioxide, which affects the results of the experiments.

The data on adsorption of $\mathrm{Ba^{\bullet\bullet}}$ are summarized in the tables below. The calculation is given for 1 g of $\mathrm{SiO_2}$.

I. Experiment with $\mathrm{BaCl_2}$.

The amounts of chlorine before the experiment and after the experiment remained unchanged in the solution. Chlorine was determined by the precipitation method, and, in order to make the influence of adsorption possible with this method negligible, the determination of the concentration of Cl′ in the initial solution and after the experiment was carried out in exactly the same quantitative ratios. 10 cc of solution No. 1, containing 40 milliequivalents of BaCl₂ in 50 cc, were taken; from a burette 33 cc of AgNO₃ were added; 9.4 cc of NH₄CNS were required for back-titration. After the experiment, 10 cc of the liquid that had settled above the silicic acid were taken, 33 cc of AgNO₃ were added; in one case 9.35 cc of ammonium thiocyanate were required, in another 9.4 cc. The experiment was repeated with solutions No. 2 and No. 3: the results obtained were the same. The same was observed by Wohler¹), Boinetien²) and Duclaux³), i.e. cleavage of the salt and adsorption of one of the ions.

At the same time, the solution, neutral before the experiment, became acidic. The acidity was determined by titration with caustic soda, \( \frac{1}{40} N \), in the presence of phenolphthalein. NaOH was prepared under conditions excluding the presence of CO₂.

Witney and Ober⁴), without giving numerical material, state that the concentration of H• corresponds exactly to the quantity of adsorbed Ba••; Duclaux did not observe such correspondence; in my experiments I also did not observe it.

I do not give numerical material because, for the elucidation of this question, such an adsorbent as SiO₂ is of little suitability. Silicic acid itself is soluble in water, and this circumstance could be corrected for, but, as the work of V. Lenher and Truol⁵) has shown, the solubility increases in salts. A special investigation must be devoted to this question.

After the experiment, silicic acid, when shaken with ammonium, potassium, or sodium salts, begins to release the absorbed barium; the same was observed by Witney and Ober.

II. Experiment with BaBr₂.

¹) Loc. cit.
²) Jur. pract. ch. 1902. 42.
³) Thèse.
⁴) Loc. cit.
⁵) J. am. chem. Soc 1916 33.

In the experiments with $\mathrm{BaBr_2}$, the amount of irreversible adsorption is so insignificant that it was necessary to take 20 g of silicic acid. Bromine likewise was found entirely in the solution. The solution after the experiment was acidic.

III. Experiment with $\mathrm{Ba(NO_3)_2}$.

IV. Experiment with $\mathrm{Ba(ClO_3)_2}$.

V. Experiment with $\mathrm{Ba(C_2H_3O_2)_2}$.

Conclusions

From the tables presented above it follows that:

a) Both for reversible adsorption and for irreversible adsorption, the amounts of absorbed barium are different for different barium salts;

I

II

b) For irreversible adsorption one cannot assume the existence of any type of chemical reaction, since, in the case of an exchange-decomposition reaction, all salts would have to behave identically; in the case of hydrolysis, \(BaBr_2\), \(BaCl_2\), and \(Ba(NO_3)_2\) should not have given any great difference.

c) Whereas for reversible adsorption the figures change noticeably with change in concentration, for irreversible adsorption, when it is referred to 1 g of \(SiO_2\), this change in the fourth decimal place is so insignificant that one may say that the concentration “almost” has no effect; thus for \(Ba(NO_3)_2\) we have 0.0024, 0.0019, 0.0017 at the corresponding concentrations of 32, 26, 19.5 milligram-equivalents in 50 c.c. This dependence is indisputable when the absorbed \(Ba^{\bullet\bullet}\) is recalculated for 10 g of \(SiO_2\), as has been done in the appended diagram II.

d) From the appended diagrams it is evident that there is no complete parallelism between reversible adsorption (diag. I) and irreversible adsorption (diag. II).

For reversible adsorption the order of action of the anion will be:

\(C_2H_3O_2 > Br > NO_3 > ClO_3 > Cl\). For irreversible adsorption \(C_2H_3O_2 > NO_3 > Cl > ClO_3 > Br\). Ganguli and Dhar¹, studying the adsorption of cations by manganese peroxide, obtained for \(Cu^{\bullet\bullet}\), \(Ba^{\bullet\bullet}\), \(Ni^{\bullet\bullet}\), \(Fe^{\bullet\bullet\bullet}\), for nitric-acid salts, an absorption considerably higher than for chlorides. The same was observed by Van Bommelen² in studying the adsorption by silicic acid of the salts \(KNO_3\) and \(KCl\), which agrees with my observations.

1. The jour. of phys. ch. 1922. XXVI. 9. ↩↩

2. Loc. cit. ↩↩