On the Influence of Pectin on the Crystallization of Ammonium Chloride¹

I. Ya. Melik-Gaikazyan, D. D. Saratovkin, A. G. Kliss

All substances, according to the nature of their influence on the surface tension of a solvent, are divided into two classes: those that lower the surface tension are called surface-active; those that raise it are called inactive [1]. When we are dealing with the growth of a crystal from a solution, the surface-active substance is concentrated at the interface between crystal and solution, i.e., on the surface of the crystal; this markedly changes the rate of growth of the various faces of the growing crystal. According to Wulff’s law [2], the rate of growth of a face is proportional to its surface tension. The aforementioned lowering of the surface tension must be different for different faces of the crystal. It follows from this that the growth rates of the faces must change in different ways, which leads to a change in the habit of the crystal.

It is also known that crystals are usually bounded by the most slowly growing faces. Rapidly growing faces are displaced by these slowly growing ones. If, for one reason or another, the ratio between the growth rates of the faces changes, then the habit of the crystal also changes, and faces appear on it that were previously absent.

Let us suppose that, in the absence of an impurity, the crystal is bounded by faces A and that there are no faces B on it because they grow more rapidly. If an impurity appears in the solution that can be adsorbed on faces A but not on faces B, then the ratio between the growth rates will immediately change for the following reason: an impurity adsorbed on a face does not hinder the growth of the crystal. Since new molecules (atoms or ions) of the crystallizing substance cannot be deposited beneath the adsorbed impurity molecules, they are deposited on top, and the molecules of the adsorbed impurity become included in the crystal. Thus, the face on which impurities are adsorbed will be able to continue its growth without hindrance. Conversely, those faces on which adsorption does not occur will be impeded by the impurity, since this impurity will accumulate in front of the face in the solution layer. Since the retardation depends on how much impurity is present in this layer, it is not difficult to understand that faces B must inevitably reduce their growth rate to a value smaller than the growth rate of faces A. The absence of faces B means only that, in the corresponding places of the crystal, there are either edges or multi-

edge angles. Both the former and the latter may be regarded as faces with infinitely small surfaces. If the growth rate of these faces is reduced to a value smaller than the growth rate of faces A, the corners or edges prove to be cut off by previously absent faces B. Further growth of the crystal must inevitably lead to its being bounded only by faces B, since they grow more slowly than faces A.

Special attention should be paid to the fact that in this case the crystal assumes a faceting in which the faces capable of adsorbing an impurity disappear, and, conversely, there appear those faces on which adsorption does not occur. Since adsorption on the faces of a growing crystal leads to absorption of the impurity by the crystal, while the bounding of the crystal by faces that do not adsorb the impurity leads to the pushing of the impurity into the layer of solution surrounding the crystal, the following regularity may be formulated.

A crystal is bounded by such faces as enable it to push aside the impurities that contaminate it. It appears that the crystal, as it were, strives to assume a form in which it can grow purified of the impurity.

It should be noted, however, that crystals completely free of impurity can be obtained only if the crystallization rate is sufficiently small and does not exceed a certain critical value, which depends on the conditions of removal, from the surface of the growing crystal, of the impurity that hinders its growth. This removal may be effected both by diffusion of the impurity away from the crystal and by concentration currents. Under conditions requiring rapid crystallization, not ensured by diffusion and concentration currents, the crystal begins to grow in skeletal or even dendritic form, entrapping within itself portions of solution with an increased concentration of impurity. In this case, of course, no purification of the crystal from the impurity occurs.

The question of the crystallization of ammonium chloride attracts the attention of many investigators for a number of reasons.

First, ammonium chloride does not at all give well-formed, large crystals from aqueous solutions, but forms loose dendrites; whereas large crystals of it are needed for various technical requirements [5]. Second, ammonium chloride dendrites, in their outward appearance and in many of their properties, are analogous to the dendrites observed in steel ingots. This makes it possible to model with them a number of phenomena observed during crystallization [6]. Finally, using them as an example it is convenient to solve questions concerning chemical interaction with an impurity, since there are many substances whose chemical interaction with ammonium chloride is well known [6].

Ehrlich [4] found that ammonium chloride in the presence of pectin crystallizes, upon slow cooling, in the form of large transparent crystals; upon more rapid cooling, long needle-shaped crystals are formed. At the same time he observed that, when the ammonium chloride obtained in this way is recrystallized from a solution in pure water, already without the admixture of pectin, the same crystals again grow. Such a result can occur only if the necessary amount of pectin has been included in the crystals.

On the basis of the above considerations, we came to the conclusion that in Ehrlich’s experiments the pectin had been included either between the crystals in crystalline druses or in the crystals together with the mother liquor, entrapped owing to skeletal crystallization. With slow crystallization, which provides the possibility of removing the impu-

from the crystals, owing to diffusion and concentration currents, one may expect the appearance of crystals free of impurities. Recrystallization of such crystals, of course, should give the same result as the crystallization of pure ammonium chloride.

We repeated Ehrlich’s experiments and obtained crystals whose external appearance was exactly the same as those he obtained. These crystals are shown in photographs in Figs. 1 and 2 at natural size. The method for growing the crystals is very simple. A solution of ammonium chloride, saturated at a temperature of 70°, with an admixture of pectin of 0.01% by weight, is poured into a Dewar vessel, which in turn is placed under such conditions as can ensure that its temperature is lowered to room temperature only after several days. We placed it in a bath of hot water, likewise insulated on all sides. Druses of crystals were also obtained (Fig. 3) (when the impurity concentration was increased to 0.1%), containing individual crystals with a large number of clearly expressed faces and edges. In external appearance these crystals very much resemble rock-crystal crystals. Owing to their good transparency it was evident that they contain no more or less noticeable interlayers with mother liquor captured in the process of skeletal growth. Several such crystals, separated from the druse and first cleaned mechanically and then washed with distilled water, were dissolved without the addition of pectin. On crystallization from such a solution, ammonium chloride precipitated in the form of ordinary dendrites, i.e., in the way it crystallizes without an admixture of pectin.

In order properly to evaluate this fact, it should be noted that already 0.00003% pectin is sufficient to convert the octahedral faceting of ammonium-chloride dendrites into cubic faceting. The former is characterized by the perpendicular arrangement of the branches of the dendrites to one another, and the latter by the appearance of angles of approximately 70°, i.e., equal to the angle between the diagonals of a cube. If one compares dendrites of pure ammonium chloride, shown in Fig. 4, with dendrites obtained from a solution with an admixture of only 0.00003% pectin (Fig. 6), it is evident that even such an exceedingly small amount of impurity already changes the character of dendritic crystallization, causing the appearance of angles different from 90°. And when crystals obtained from a solution containing 0.1% pectin are dissolved, crystals then grow from the newly obtained solution such as can be obtained only in the absence of even these three hundred-thousandths parts of pectin. A photomicrograph of such dendrites is given in Fig. 5.

The question of precisely which faces cover crystals that do not contain pectin but grow from a solution with an admixture of it is at the stage of further study. Here there is a large number of small faces representing a combination of several simple forms.

Conclusions

On the basis of experiments with the recrystallization of ammonium chloride obtained in the form of large transparent crystals from solutions containing pectin impurities, one may conclude that the crystal is bounded by those faces that allow it to reject the impurities contaminating it. The appearance of particular faces on the growing crystal, when an impurity is present in the solution, is determined by the difference in adsorption of the impurity on faces with different indices.

In the case where a small addition of impurity is capable of changing the octahedral faceting of ammonium-chloride dendrites to cubic faceting,

Fig. 1. Crystals of ammonium chloride grown from an aqueous solution in the presence of 0.01% pectin. Weight 10–20 g

Fig. 2. Crystals of ammonium chloride grown from an aqueous solution in the presence of 0.01% pectin

Fig. 3. Druses of ammonium chloride crystals grown from an aqueous solution in the presence of 0.1% pectin.

Fig. 4. Dendrites of pure ammonium chloride. ×32

Fig. 5. Dendrites of ammonium chloride grown during recrystallization of the crystals shown in Fig. 3. ×32

the presence of the usual, octahedral faceting of the dendrites may serve as a reliable indication of the absence of this impurity.

Fig. 6. Dendrites of ammonium chloride grown in the presence of a 0.00003% pectin impurity

REFERENCES

1. A. I. Brodskii. Physical Chemistry, Vol. I, Leningrad, 1948.

2. Yu. V. Vul’f. On the question of the rates of growth and dissolution of crystal faces. Selected Works on Crystallochemistry and Crystallography. Gostekhizdat, Moscow–Leningrad, pp. 17–113, 1952.

3. V. D. Kuznetsov. Physics of the Solid State, Vol. 1, Tomsk, 1937.

4. Erlich. Zs. f. anorg. chem., 98, 1931.

5. D. D. Saratovkin. Dendritic Crystallization, Metallurgizdat, Moscow, 1953.

6. Yu. Ya. Til’mans. Crystallization of ammonium chloride and ammonium bromide in the presence of impurities. Journal of General Chemistry, 1941, issue 11, pp. 869–979; issue 1, pp. 4–9; 1948, issue 10, pp. 1752–1754; 1952, issue 3, pp. 384–395.

1. Report read at the inter-institutional conference on dielectrics and semiconductors on February 6, 1958, in Tomsk. ↩