Engineer-Technologist I. V. Gebler.

The Petukhovo Soda Lakes

(In Slavgorod Uezd, Omsk Province).

(With a diagram).

Lakes containing a considerable amount of sodium carbonate, on the basis of which they are called soda or natron lakes, occur in various places on the globe. Besides the well-known Egyptian lakes, lying south of the Nile delta, 80 kilometers from Cairo, and supplying Egyptian soda, or the so-called trona, soda lakes are found in India, Persia, Hungary, California, and Mexico. The largest are the lakes of California, of which Mono Lake is 22.5 kilometers long and about 14.5 km wide, while Owen Lake, better known and still more important, reaches 35.4 kilometers in length and 12.85 kilometers in width. The water of this lake has a specific gravity of 1.076 and contains, per liter, 27.7 g Na₂CO₃; 4.47 NaHCO₃; 10.8 Na₂SO₄; 31.1 NCl (and KCl)¹. Thus, with its substantial dimensions, Owen Lake contains an enormous reserve of natural soda, which is now being exploited.

As for Russia, only insignificant soda deposits have been recorded on its territory; among the lakes mentioned in the literature² is the soda lake in Erivan Province near the village of Burun; the water of this lake contains 6.6% solid residue, the composition of which, according to Abikh’s analysis, is as follows: Na₂CO₃—14.71%; Na₂SO₄—10.36%, and NaCl—74.61%; soda, according to the description, crystallizes in this lake together with sodium sulfate, a circumstance which, however, is not consistent with the concentration of the lake brine, for the quantity of both salts in the solution is in any case insufficient for their crystallization, even taking into account the salting-out action of sodium chloride. Further, soda occurs in the form of efflorescences along the shores of the Caspian and Aral Seas, likewise together with sodium sulfate and sodium chloride. Finally, in Siberia³ similar soda efflorescences are reported in the Baraba Steppe, near Okhotsk, and along the Kurtamysh River, a tributary of the Tobol (Pallas). There are no indications whatsoever of soda lakes in Siberia, and in general such lakes were apparently unknown before the discovery of the Petukhovo soda lakes, which the author visited in the summer of 1925 and to the description of which he now turns.

The Petukhovo soda lakes are located in the Klyuchevskaya dacha of the so-called Northern Gatsky pine forest, 110 km from the city of Slavgorod, 55 km from Kulunda station of the Kulunda Railway, and 12 km from the large village of Klyuchi. The nearest settlements to the lakes are the villages of Severnaya and Redev; from both of these villages, situated at the very edge of the pine forest, to the soda lakes (in different directions) is no more than 7 km. Thus, the lakes described do not lie in such remote places, but comparatively close to populated localities, and one must be surprised that until very recently they remained unknown. This circumstance indicates that in Siberia there are still quite a few such corners, unvisited—

¹ Lyubavin. Technical Chemistry, I, II, p. 230.
² Idem.
³ Nikitinsky. Monograph on Soda Production, p. 98.

…new and unexamined, closer acquaintance with which may yield certain other interesting novelties. Even now—and, moreover, in Slavgorod Uyezd itself—information about them is in most cases very scanty; and in replies from local inhabitants to inquiries about the Petukhovo soda lakes, they usually understand other Petukhovo lakes, located 25 km to the northeast near the large village of Petukhovo.

In one of these lakes, whose brine contains chiefly sodium chloride, common salt is deposited and has been extracted there since olden times¹, as a result of which this Petukhovo group of salt lakes is far better known in the district. We mention this circumstance here in order to eliminate confusion between two groups of lakes of the same name that have nothing in common with one another.

As became clear from inquiries made on the spot, the Petukhovo self-depositing soda had long been known only to the peasants of the nearest villages; they carted it away for their own use and used it in washing linen. In about 1920–21 the soda lakes attracted the attention of the head of the post office in the village of Klyuchi (whose surname could not be ascertained), who sent a sample of crystalline lake soda somewhere for analysis. Apparently at about this same time lake soda was brought by someone to Slavgorod, and here the soda nature of the Petukhovo lakes was first established by a resident of Slavgorod, Master of Pharmacy M. S. Krestovyi (1921); on the basis of this, in January 1922 the lakes were recognized as being of state importance². The first exploitation of the Petukhovo lakes was undertaken by private individuals, Messrs. Kultenev and Kuzmichev, who in 1921 set up a small plant for soda extraction; working under contract with Uzekonot, this plant, as was reported in Slavgorod, existed until the beginning of 1924.

In 1922 the Technical Council of Glavkhim decided to begin construction of a plant for calcining soda at the Petukhovo lakes. The Petukhovo Soda Construction enterprise was created, or so-called “Sodostroi.” The latter carried out a survey of the lakes and performed a number of works of a construction and reclamation character: dwelling houses, storehouses, and drying sheds were built; auxiliary enterprises were equipped in the village of Severnaya: a sawmill, a cottage-type brickworks with a floor furnace, a small smithy and a flour mill; and, finally, dams were built between the lakes. An analysis of the water of the soda lakes apparently was not made; at least among the materials of “Sodostroi” no analytical data were found, except for areometric measurements. The extraction of self-depositing soda undertaken yielded 3,315 poods in 1922, and 10,195 poods of crystalline soda in 1923. Subsequently the extraction of soda was suspended, since, because of the changed market situation and the importation of Ural soda (calcined and caustic), the sale of Petukhovo crystalline soda, which economically could not bear transportation, could not be carried out profitably; calcination of the soda, or its processing into caustic soda, also appeared of little advantage. Therefore “Sodostroi” was liquidated in 1923, while the soda lakes were transferred to the Omsk GSNKh, and later passed from it to the Slavgorod Uispolkom, under whose authority they remain at the present time.

This is the information on the soda lakes which we have considered it appropriate to present, using the “Sodostroi” materials at our disposal, as well as information obtained on the spot, before proceeding to the results of our own observations.

The route to the Petukhovo soda lakes lies from Kulunda station, which can be reached by the Kulunda railway from two directions: from Tatarskaya station of the Omsk Railway (345 versts; two mixed freight-and-passenger trains a week according to the 1924 timetable), and from the town of Pavlodar (100 versts, one train a week).

From Kulunda an excellent steppe road, very convenient for motor traffic, runs all the time southward through the villages of Plotavskaya (26 versts), Klyuchi (17 versts), and Reden (5 versts). From the last village, situated at the very edge of the above-mentioned Northern pine forest, whose enormous pines suddenly rise up amid the monotony of the bare steppe, it is 7 versts to the lakes. The road runs through the pine forest for its entire length and becomes very difficult, since the wheels sink almost to the hubs in the loose sand, and at least 2 hours are required to cover these 7 versts and reach the lakes, the area occupied by them appearing suddenly amid the dense thicket of the surrounding forest. On this area there are three lakes; the larger of them occupies the southern position in the group; being round in shape, its diameter is more than 2.5 kilometers, and its greatest depth, according to fragmentary information, is about 5 meters. According to the data of “Sodostroy,” the area of this lake is 1,369,370 square sazhens, and the mean depth has been taken as 1.443 sazhens, on the basis of which the volume of water is calculated at 1,712,805 cubic sazhens. Along the northern shore of the lake there is a large quantity of gray mud, the thickness of whose layer is about 3/4 meter. From the typical medicinal mud characteristic of many salt lakes, this mud differs by a considerable content of sand, which makes it coarse and non-plastic. With increasing distance from the northern shore the layer of mud becomes smaller, and in the middle of the lake, as well as at its southern end, there is no mud at all—the bottom and shores are firm and sandy. Along the eastern shore of the lake, which closely adjoins the forest, a narrow belt of reeds gives shelter to water birds: broods of garganey duck (Anas boschas) and ruddy shelduck (Tadorna rutila) were encountered here; in the lake water some small crustaceans swim in small numbers. Determination of the density of the water on the spot with an areometer gave 1.75°Bé at \(t = 27.0^\circ\mathrm{C}\); the lake water is distinctly alkaline in taste and foams when rubbed between the palms. From the lake basins of the two small lakes situated nearby to the north, the large lake is separated by low natural barriers only a few meters wide; this neck consists of sandy-loam material with a small amount of clayey substance, at least in the upper layer. In springtime, when the waters rise, all the lakes sometimes join together. Such a phenomenon, from the point of view of the future exploitation of the more concentrated small lakes, would be undesirable, and therefore these necks have been reinforced and strengthened by “Sodostroy”; moreover, owing to the insignificant height of the one of them which separates the first lake from the second (the larger of the small ones), a dam with a simple row of piles has been built on one of its sections. Nevertheless, in view of the permeability of the material of the natural necks, subterranean communication between the lakes apparently does exist, and this circumstance, as we shall see below, must have great importance in the exploitation of the lakes. In the materials of “Sodostroy” there is an indication that south of the Big Lake, not far from it, there is a fresh lake called “Rybnoye,” which in spring has an outflow into the former and in this way dilutes it; how true this is could not be ascertained on the spot. In the same materials, concerning the water of the Big Lake, it is said that it has the smell of hydrogen sulfide and is dark green in color; during our visit to the lakes in the first days of July, however, we observed colorless, completely transparent water, with a faint unpleasant odor, in which, however, hydrogen sulfide was not at all perceptible.

There are two small soda lakes lying side by side with the large one, as already mentioned; the larger of them, which we shall call Lake No. 2, according to the data ...

“Sodostroi,” has an area of 42,886 sq. sazhens, with an average depth of 0.39 sazhen; the volume of water is 16,726 cubic sazhens. The density of the water, determined on site with an areometer, proved to be 12.00° Bé at \(t = 27.8^\circ\text{C}\). The shores of the lake are everywhere covered with an incrustation of soda; a thick layer of mud of the same character as on the Large Lake makes access to the water difficult, and one can approach it only along the existing causeway; otherwise one may sink in the loose mud almost up to the waist. In the water, at least near the shores, there is a multitude of reddish crustaceans; their number is so great that it is impossible, without precaution, to scoop water into a bottle in the form of a stream without its being full of these inhabitants of the alkaline lake. Neither the shore mud nor the lake water has the odor of hydrogen sulfide; the same faint, indefinite smell is perceived as in the water of the Large Lake. The lake water is bottle-green in color with turbidity that does not settle for an indefinitely long time. The lake freezes in winter; the materials of “Sodostroi” state that the thickness of the ice reaches 0.17–0.42 sazhen (evidently being unequal at different depths); the ice is not dense, porous, not fit for consumption, and of course contains an insignificant amount of soda. Because a considerable mass of water turns into ice, which, according to the rule of freezing of solutions, contains only a small quantity of salts, the entire mass of soda remains in the solution, whose concentration greatly increases and causes crystallization of soda. Soda crystallizes chiefly in spring (beginning in March) on the bottom of the lake as a continuous cover several vershoks thick, and also on the inner side of the ice. This is precisely the soda that was extracted in 1922–23. The extraction was carried out as follows: the ice was sawn through, broken open, and from the underside of the overturned ice floes large crystals of sodium carbonate were collected. With the onset of warm weather, after the melting of the ice, extraction continued from rafts and boats: the soda was scooped from the bottom with perforated ladles together with silt, which was then immediately washed off. The productivity of such work was from 15 to 20 poods of washed crystalline soda per worker per day. Extraction continued until the first days of May. Subsequently, with the rise in temperature of the lake water, the crystallized soda dissolved, in connection with which extraction, carried out in this way for \(1 \tfrac{1}{2}\)–2 spring months, ceased. All the extracted crystalline soda, about 15,000 poods, is at present lying in storage on site.

Lake No. 3 is still smaller; according to the data of “Sodostroi” its area is 20,391 sq. sazhens, the average depth 0.24 sazhen, and the volume of water—4,898 cubic sazhens. The areometric determination on site gave—10.00° Bé at \(t = 27.8^\circ\text{C}\). This lake is in all respects analogous to the lake No. 2 described above. Crystallization of soda in it likewise occurs, but in a relatively smaller quantity. Soda has not been extracted from it.

At an insignificant distance from the lakes there are several wells with completely fresh water; this was the first fresh and quite good water that we encountered among the solonchaks of the Kulunda steppe.

Let us now proceed to characterize the chemical composition of the water of the soda lakes on the basis of the analysis I performed. It should be noted that I was nowhere able to find any analyses of the water of these lakes, and, apparently, the water has not, generally speaking, been analyzed in any detail.

In the above-mentioned essay by M. S. Krestov it is said that the water of the Petukhovo lakes contains up to 12.5% soda, but on the basis of what data this figure was obtained is not explained. In the materials of the Soda-Construction Commission that carried out the survey, no analyses are likewise given; only areometric readings are provided for all three lakes in the following form:

On the basis, evidently, of these data alone (for no others are given), the soda content in the water of the lakes was established, and its reserves in the form of crystalline sodium carbonate were calculated in the following figures:

In such a calculation, made on the basis of hydrometric readings, it was evidently assumed that the water of the lakes contains no salts other than sodium carbonate. In fact, this is far from the case; our analysis established a fairly considerable content of extraneous salts in the brine, and nevertheless the amount of soda proved greater; therefore the figures cited are in every respect erroneous.

I analyzed the water of Bolshoe Lake (No. 1) and of Lake No. 2. From Bolshoe Lake the water was taken from a boat 20 meters from the shore on July 7 at 11 o’clock in the morning. The temperature of the water was 26.0°C; the temperature of the air—30.0°C. From Lake No. 2 the water was taken 5 meters from the shore, from the bridges, on the same day and at the same hour; the temperature of the water—27.5°C; the temperature of the air—30°C.

Both water samples, in 2-liter bottles with ground-glass stoppers, were brought to Tomsk and examined in the laboratory of fibrous and dye substances of the Tomsk Technological Institute by the usual methods employed in the analysis of waters; chlorine was determined by the gravimetric method, while total and combined carbonic acid were determined as follows: a certain portion of water was taken directly from the bottle, and by Pettenkofer’s method the total amount of carbonic acid was determined; another portion of water was evaporated in a platinum dish on a water bath, after which the dish with the dry residue was heated for 1 hour on a sand bath at 300°C; according to Lunge, such a procedure ensures removal of the second equivalent of CO₂ of bicarbonates; the residue after this treatment was dissolved in distilled water, and by the same method the carbonic acid corresponding, in this case, to the carbonates present in the water, and the residual bicarbonates was determined, on the basis of which the content of CO₃″ and HCO₃′ was calculated.

Determination of specific gravity was made by means of a pycnometer at 16.0°C. The results of the analysis are presented in the following table in gram-ions and gram-equivalents per 1 liter of water:

Lake No. 1

(Large).

Specific gravity: 1.0160.
Reaction: strongly alkaline.
Transparency: complete.

Color: colorless.
Odor: faint—unpleasant.
Taste: alkaline-saline.

One liter of water contains:

Dense residue (140° C) . . . . . . . . 16.1000.
Loss on ignition . . . . . . . . . 0.7200.

Lake No. 2

Specific gravity: 1.0961.
Reaction: strongly alkaline.
Transparency: turbid.

Color: bottle-green.
Odor: faint—unpleasant.
Taste: alkaline-saline.

One liter of water contains:

Dense residue . . . . . . . . . . . . . . . . . 108.8400.
Loss on ignition . . . . . . . . . . . . . . . . 2.3000.

As is evident from the figures given, the content of sulfate salts in the water of the lakes is insignificant, whereas that of the chlorides, on the contrary, is very large; it is interesting to note that the relative amount of the former and the latter is almost the same in the water of both lakes: the ratio of the quantity of the ions $\mathrm{Cl}'$ to the quantity of the ions $\mathrm{SO_4}''$ in the water of the Large Lake is

\[ \frac{2.924}{0.132}=22.15, \]

and in the water of the Small Lake is

\[ \frac{18.972}{0.889}=21.34. \]

Potassium ions were not found in the lake water; in the large lake, as may be seen, there is an insignificant quantity of magnesian salts, while in the water of the small lake there are no other cations except sodium. In view of this latter circumstance, the carbonate salts can be determined by another method in the water of this lake, namely by titration. It is known that if hydrochloric acid is added to a solution of sodium carbonate, using phenolphthalein as indicator, while cooling to zero and in the presence of NaCl, decoloration occurs at the moment when half of the total amount of carbonate has been neutralized; sodium bicarbonate is neutral to phenolphthalein and may be titrated, if it is also present in the solution, in the presence of methyl orange, together with the other half of the normal carbonate salt. The method of Wurber¹ based on this for determining carbonate and bicarbonate alkali when both are present was applied, for the purpose of verification, in the study of the water of the small lake. The data obtained were as follows:

25 c.c. of water from the Small Lake were diluted to 300 c.c.; the solution obtained was titrated in 25 c.c. portions with hydrochloric acid at $t$ between 0 and $+1^\circ\mathrm{C}$.

The titer $(T)$ of HCl $=0.00276$. The amount of HCl used for decoloring the solutions with phenolphthalein $a = \text{av. }16.75$ c.c. The amount of HCl used for neutralization in the subsequent titration with methyl orange $A = \text{av. }37.75$ c.c. $2a$ c.c. HCl correspond to normal carbonate salt; $A-2$ c.c. HCl correspond to bicarbonate salt.

Hence the content of both salts in 1 liter of water will be:

\[ \mathrm{Na_2CO_3} = \frac{106.1\cdot 2a\cdot T_{\mathrm{HCl}}\cdot 12\cdot 40}{36.5} = 64.4424\ \text{g}. \]

\[ \mathrm{NaHCO_3} = \frac{84.1\,(A-2)\cdot T_{\mathrm{HCl}}\cdot 12\cdot 40}{36.5} = 12.8592\ \text{g}. \]

Consequently, both methods for determining carbonate salts gave quite concordant results, as may readily be seen by comparing the figures obtained.

Neglecting the insignificant content of nitrates and ammonium salts, and further assuming that the magnesium in the water of the Large Lake is present in the form of the most soluble, in the presence of soda solution, bicarbonate, the salt composition of both lakes may be represented, on the basis of the results of the analysis, by the following figures:²

Accordingly, taking the volume of water in the lakes according to the above-cited data of “Sodostroi,” the reserve of soda in them, calculated as anhydrous sodium carbonate, will be as follows:

Comparing the figures obtained with the above-cited data of “Sodostroi,” where the quantity of crystalline soda in the Large Lake is determined as 19946858 poods, and in the Small (No. 2) as 501793 poods, we find that the two do not at all coincide; indeed, recalculating the latter figures as anhydrous sodium carbonate, we obtain that its content in the Large Lake is expressed as 7392891 poods, and in the Small as 185972 poods, which gives a total of 7578863, whereas the same total according to our data is 10746542 poods, i.e., 3167679 poods, or 41.7%, greater; meanwhile, a priori one would have expected the opposite relation, for the figures appearing in the materials of “Sodostroi,” as already indicated above, were obtained on the basis only of areometric determinations, and, evidently, the considerable content of foreign salts was not taken into account, while an exact analysis was lacking, and apparently is presented here for the first time; therefore our figures deserve preference, and their reliability with respect to the total reserve of soda depends only on how far the data concerning the volume of the lakes, obtained on the basis of measurements made by the Soda-Construction Commission, correspond to reality.

Thus the reserve of soda in the Petukhovo lakes, assuming that the uninvestigated lake No. 3 (very small in size) contains in any case not less than 100,000 poods of it, amounts, in round numbers, to 11,000,000 poods. It is interesting to compare this figure for the reserve of natural Siberian soda, leaving aside for the time being the question of the economic feasibility of extracting it, with the production of this product at Russian factories; the corresponding data are found in the book by L. F. Fokin—A Survey of the Chemical Industry in Russia in 1921. It is stated there that the greatest quantity of sodium carbonate (calcined soda) was produced in 1913: 9,721,000 poods; thus, the entire quantity of Petukhovo soda exceeds the production of all Russian soda plants in 1913 by only 11.5%; therefore the soda lakes described, on a nationwide scale, of course cannot have serious significance—for that, the reserves of soda in them are too small—but their local significance is undoubtedly great; for many years these lakes could supply the Siberian market with natural soda, if properly exploited, cheaper than factory soda and soda imported from European Russia, and moreover of remarkable purity. A sample of crystalline soda was taken from its storehouse (there are about 15,000 poods) at the lakes, brought to Tomsk in a hermetically sealed jar, and analyzed; moisture, the content of carbonic-acid salts, were determined, and a test was made for the presence of foreign substances. The work was carried out as follows: a weighed portion of the ground crystalline soda in a platinum crucible was first dried at \(t:110^\circ\text{ C}\), and then heated on a sand bath for an hour at \(t:300^\circ\text{ C}\); in this way all the water and the second equivalent of \(CO_2\) of the bicarbonates were removed; another weighed portion of soda was dissolved in

water, and here carbonic and bicarbonate alkali were determined by titration according to Wurder’s method, and a test was made for foreign anions. The quantity of sodium bicarbonate found in this way made it possible to calculate the loss \((\mathrm{CO_2} + \mathrm{H_2O})\) of the bicarbonate during its decomposition; subtracting this number from the total loss in weight on heating the first weighed portion, we obtain the amount of water of crystallization.

The results of the analysis are as follows:

Petukhovo self-deposited soda:

Thus, this soda contains only traces of impurities and in purity surpasses even ammonia soda, not to mention Leblanc soda, in which the content of foreign substances amounts to 2–3%, and often more; when the jar with soda crystals is opened, a peculiar unpleasant odor is perceived, which completely disappears after the crystals have lain in the air for several minutes. The calcined soda is of a completely white color.

The possibility and success of exploiting the Petukhovo soda lakes depend to a considerable degree on the circumstance mentioned above, namely, whether there is an underground connection between the waters of the Large and Small lakes. The sandy material of the narrow bars between the lakes is, apparently, sufficiently water-permeable and makes such an assumption very probable; indirect evidence in favor of it may be seen in the fact that the relative quantities of chloride and sulfate salts are approximately the same in the waters of the lakes investigated. If this is indeed so, then the Small lakes are natural concentration basins for the more dilute, but possessing large reserves of soda, Large lake, whose waters, filtering through the aforementioned bars, feed these basins. The slight depth of the small lakes causes stronger warming of their water in summer and, consequently, more intensive evaporation; in winter, greater freezing of the lakes and weathering of the ice likewise lead to an increase in the concentration of the lake waters and to their enrichment with soda. Consequently, under such conditions sodium carbonate, together with other salts, gradually passes from the Large lake into the Small ones, where extraction of self-deposited soda may be concentrated in the last winter months; moreover, the very method of this extraction can evidently be the same as that adopted by “Sodostroy,” i.e., breaking up the ice and removing soda crystals from the bottom by scooping them out.

Such a method in the present case must be considered the most expedient and economical, for the construction of special basins for freezing out would be too expensive, and it is hardly necessary, in view of the miniature size of the lake, which by itself gives a good deposition of sodium carbonate.

The question of soda extraction is resolved quite differently if the assumed connection of the lake waters is in fact absent and all the lakes constitute entirely independent basins that do not communicate with one another. Then the Large lake, with the principal reserve of soda, must be exploited directly, and, given the low concentration of its water, this task appears rather difficult, of course, in the economic sense. The total quantity of salts per 1 liter of water of this lake is 16.10 grams; for

in order to concentrate it to the density of the water of lake No. 2, containing 108.84 grams of salts per liter, i.e., to bring it to that concentration at which freezing-out can be carried out and the crystallization of soda ensured—which, generally speaking, is apparently facilitated by the salting-out action of sodium chloride—it is necessary, in order to obtain 1 liter of solution of the required density (108 grams per 1 liter), to remove 6.75 liters of water by evaporation. This problem may be solved in two ways: either by means of evaporating basins, or by graduation. The enrichment effect in either method of operation depends on the meteorological conditions for the given locality; the mean temperature during the summer period, insolation, and air humidity are of great importance here, as are winds and the amount of atmospheric precipitation; and therefore, lacking the relevant data, it is difficult to draw a decisive conclusion as to which of the indicated methods of operation would be more advantageous in the present case.

Generally speaking, solar heat is used with advantage for evaporating brines chiefly in southern countries. According to Lübavin¹, sodium chloride is obtained from seawater in this way on the shores of the Atlantic Ocean up to 48° north latitude. The climate of the Kulunda steppe is fairly dry, and the summers here are sometimes very hot², but all this is insufficiently constant, and the duration of the hot period is not likely to be great; therefore evaporation of lake water by solar heat in the corresponding basins—although in certain periods it may be very intensive—would in the end prove insufficiently productive for choosing this method of enrichment in extracting soda from the Big Lake.

It seems to us that, in the present case, concentrating the brine by means of graduation would prove more advantageous and reliable. To be sure, graduation works have substantial drawbacks: they are very bulky, are relatively expensive, and, most importantly, during the movement of the solution through the graduation works some loss of salts occurs, owing to the carrying away by the wind of fine particles of the solution; this loss is the greater, the more concentrated the saline solution passing through the graduation works, and in the graduation of common-salt solutions, when the concentration is brought to 22% NaCl, it reaches 10% and more³. In the present case, however, this loss can in no way be so great, since the drying of the solution would have to be carried only to the final density corresponding approximately to only 11% of the total salt content. Moreover, the very process of graduation in obtaining a more expensive product, such as soda is in comparison with sodium chloride, can and should be conducted with greater care—for example, by placing the graduation works so that they are not exposed to the action of very strong wind, which is fully favored by the position of the lakes, located in a forest. The resulting lowering of the working effect under such conditions is offset by the lesser degree of concentration of the solution in comparison with the usual practice of graduation in the extraction of common salt.

As for the size of the enrichment installation, it is determined primarily by the amount of soda that would have to be obtained by freezing out the enriched solution during one spring period. The data for this may be obtained on the basis of observations of the deposition of sodium carbonate in the Small Lake (No. 2), taking into account the amount of the latter crystallizing out per unit volume of lake water. For calculating the surface of the graduation works, some guidance may be given by the following figures, cited in Lübavin’s Technical Chemistry (Vol. II, p. 159), from the practice of graduating saline solutions in the extraction of sodium chloride at one of the German

of plants: initially a solution with 5.76% NaCl was raised onto the graduation tower; from each square foot of the surface of the graduation tower, during trickling, 32.9 cubic feet of water evaporated; at the second trickling the solution contained 8.44% sodium chloride—25.15 cubic feet of water evaporated per 1 sq. foot. At the third trickling the salt content was 12.11%, and 21.88 cu. feet of water evaporated; finally, for the fourth time a solution containing 16.53% salt trickled down, and the evaporation amounted to 17.38 cu. feet. As may be seen, as the solution becomes denser, the effect of the graduation tower’s operation decreases; this is explained by the well-known property of saline solutions of lowering the elasticity of water vapor in proportion to the content of dissolved salt. These data, of course, can have only approximate significance, since they refer, first, to a solution of common salt only, and, secondly, to an entirely different locality; nevertheless they are not useless and are suitable for preliminary orientation. Given the weak concentration of Bolshoye Lake, the dimensions of the graduation tower must, generally speaking, prove considerable; but this is mitigated by the circumstance that the lakes are situated in a pine forest, and there is any amount of timber material here, both for the framework of the graduation tower and for the faggot filling, close at hand.

Crystalline soda, which can be obtained from the lakes, will in no case constitute a final product for sale; being a decahydrate, containing 62.92% water of crystallization, it is not convenient for transport even over short distances, and therefore at the place of its extraction it must be converted either into calcined soda or into caustic soda. The demand for the latter in Siberia is greater than for soda ash, i.e., the consumers of caustic soda are the numerous soap-boiling plants, working exclusively by the simple washing method without splitting the fat¹, whereas the very weak development of the chemical industry in Siberia limits the demand for soda ash. However, the conditions for causticizing soda at the Petukhovo lakes prove rather unfavorable; the point is that nowhere nearby are there deposits of limestone; lime can be delivered here only from remote districts, namely from the Irtysh—near the town of Pavlodar, 200 versts—or, of better quality, from Kamen—250 versts—and the transport of burnt lime, when delivered to the lakes by cart road (from Kamen), costs about 70 kopecks, i.e., the price of the lime necessary for causticizing is disproportionately high; therefore the possibility of using it is inevitably connected with regeneration, i.e., with burning the calcium carbonate obtained in the process. Yet even under this condition, the increased fuel consumption and the additional work for regenerating the lime—for which purpose the calcium carbonate, for the convenience of burning, would perhaps have to be briquetted—makes the limestone method of causticizing rather doubtful economically.

More reliable under the given conditions would apparently be Löwig’s ferrite process, which consists, as is known, in heating previously dehydrated sodium carbonate with iron oxide to 700–800°; the sodium ferrite thus obtained, Na₂O·Fe₂O₃, is decomposed by water, giving solutions of caustic soda at 35–40° Bé; the iron oxide precipitates and is used again. For such work one might employ Kuznetsk red iron ore from the Yurman deposit, near the Guryev plant. However, taking into account the absence in the nearest district of mineral fuel, it is apparently still better to abandon the causticizing of Petukhovo soda on site and confine oneself merely to dehydrating the crystalline soda, releasing as the final product anhydrous sodium carbonate (calci-

dehydrated soda). The costs of dehydration can be reduced by weathering the crystalline soda in dry summer days in the air; for this the crystals must be spread in a thin layer on a wooden platform directly in the sun and frequently turned over; weathering under such conditions proceeds very rapidly, and in this way no less than half of the water of crystallization can be removed, and consequently the fuel consumption in subsequent calcination will be considerably reduced.

The demand for soda ash could increase with the establishment in Siberia of fat saponification, which should long ago have been undertaken in order to avoid losses of enormous quantities of glycerin, so necessary and valuable in the production of explosives—(dynamites). Fat-splitting plants, concentrated in the same Kulunda steppe, as the most cattle-breeding district, operating, for example, by the contact method, the installation of which is not associated with particularly complicated equipment, would, along with good glycerin, release fatty acids onto the market, which would serve as material for soapmaking, and then the soapmaker would become a consumer of calcined soda, which, in turn, would lead to a reduction in the price of soap¹.

Concluding the present article, I wish to dwell somewhat on the question of the origin of the soda lakes described.

Regarding the causes of the formation of sodium carbonate in nature, various assumptions have been expressed, and, obviously, this process of soda formation under different conditions is, generally speaking, not the same. Egyptian soda, according to Sikenberger², is formed from sodium sulfate, which is reduced to $\mathrm{Na_2S}$ owing to the activity of filamentous algae (confervae); sodium sulfide, under the influence of carbon dioxide gas developing as a result of the decomposition of organic substances by microbes, passes, with the release of hydrogen sulfide, into sodium bicarbonate, which accumulates in the lake water and, as it evaporates, is deposited in the form of trona, i.e. sodium sesquicarbonate.

Such a process of soda formation, obviously, does not take place in our case, since it is accompanied by abundant release of hydrogen sulfide, whereas the odor of this gas at the Petukhovo lakes—as has already been noted—is not observed at all, and it has not been detected in the lake water.

Prof. Tanatar, on the basis of his experiments³, proves that the formation of soda in nature is possible through the interaction of sodium sulfate, limestone, and carbonic acid; the latter, increasing the solubility of calcium carbonate (calcium bicarbonate is formed), ensures the progress of the exchange reaction between both salts; in order that this reaction should not proceed backward under strong dilution and removal of carbonic acid, rapid separation of the reaction products is necessary, i.e. gypsum must precipitate and soda be rapidly washed away; such a process, Prof. Tanatar concludes, should be favored by alternation of the dry season with abundant precipitation.

P. Melikov, describing in two articles “On the Conditions of Formation of Natural Soda”⁴ his extremely interesting works aimed at clarifying—

…of the conditions under which soda can be obtained proves that the effect of the reaction between sodium sulfate and calcium bicarbonate is governed by the influence of colloidal substances (colloidal hydrate of aluminum oxide), which manifests itself in the fact that, being absorbers, colloidal substances can remove from the sphere of interaction one of the reaction products—the one that is more completely absorbed. The participation of colloids, as is proved by the experiments of P. Melikov, also accounts for the possibility of soda formation in the interaction of sodium chloride and calcium bicarbonate and, finally, of sodium chloride and sodium sulfate with ammonium carbonate, which in turn can form in soils rich in organic, humic substances.

The indicated processes of soda formation must be expressed by the following five schemes:

According to Tanatar:

\[ \begin{aligned} 1)\quad & \mathrm{CaCO_3 + CO_2 + H_2O = Ca[HCO_3]_2}.\\ 2)\quad & \mathrm{Ca(HCO_3)_2 + Na_2SO_4 \rightleftarrows 2NaHCO_3 +} \\ & \mathrm{\qquad\qquad\qquad\qquad + CaSO_4} \end{aligned} \]

For the reaction to proceed in the direction of soda formation, a concentration is necessary that ensures the precipitation of gypsum, which remains in the soil.

According to Melikov:

\[ \begin{aligned} 3)\quad & \mathrm{Ca(HCO_3)_2 + NaCl = 2NaHCO_3 + CaCl_2} \\ & \text{[or: } \mathrm{Mg(HCO_3)_2 + NaCl = 2NaHCO_3 +} \\ & \mathrm{\qquad\qquad\qquad\qquad + MgCl_2}\text{]} \\ \\ 4)\quad & \mathrm{Ca(HCO_3)_2 + Na_2SO_4 = 2NaHCO_3 +} \\ & \mathrm{\qquad + CaSO_4\ [or: Mg(HCO_3)_2 + Na_2SO_4 =} \\ & \mathrm{\qquad\qquad = 2NaHCO_3\ MgSO_4]} \end{aligned} \]

The soda that forms is absorbed and retained by the colloidal hydrates present in the soil more strongly than the chloride or sulfate salts, which are washed out earlier; the soda is carried away with later wash waters and accumulates in the lakes.

\[ \begin{aligned} 5)\quad & \mathrm{(NH_4)_2CO_3 + 2NaCl = Na_2CO_3 + 2NH_4Cl} \\ & \text{[or: } \mathrm{NH_4HCO_3 + NaCl = NaHCO_3 + NH_4Cl}\text{]} \\ \\ 6)\quad & \mathrm{(NH_4)_2CO_3 + Na_2SO_4 = Na_2CO_3 +} \\ & \mathrm{\qquad + (NH_4)_2SO_4\ [or: 2NH_4HCO_3 + Na_2SO_4 =} \\ & \mathrm{\qquad\qquad = 2NaHCO_3 + (NH_4)_2SO_4]}. \end{aligned} \]

Soda is retained by the colloidal hydrates of the soil and is washed out after the ammonium salts; in the lake water there must be present the ion \(\mathrm{NH_4'}\), or the ion \(\mathrm{NO_2'}\), or the ion \(\mathrm{NO_3'}\), or all together in more significant quantities.

In view of the fact that in the water of the lakes, as is evident from our analysis, the ions \(\mathrm{NH_4'}\), \(\mathrm{NO_2'}\), and \(\mathrm{NO_3'}\) are almost absent (traces), it cannot be assumed that the formation of soda in our case proceeded according to schemes 5 and 6.

The process represented by scheme 2 can take place only with strict observance of such a concentration that precipitation of gypsum occurs and it is removed from the sphere of interaction; only under such a condition will the reaction proceed from left to right. The constant observance of this condition in soils, of course, cannot be assumed, and therefore soda formation by this route can have only a more or less accidental character. More constant must be considered the process of soda formation proceeding according to schemes 3 and 4, i.e., from sodium chloride or sodium sulfate and calcium (magnesium) bicarbonate with the participation of colloidal hydrates of aluminum and iron, which in our case may be present in the loamy soil near the Petukhovo lakes. The secondary products of the reactions taking place, i.e., calcium or magnesium salts, must likewise collect in the lakes. However, their solubility, with an excess…

sodium carbonate is so negligible that the corresponding ions in the water of the lakes should either not be detected at all, or only in the most insignificant quantities, as is evident from the analyses given above: a small amount of Mg was found only in the water of the not so concentrated Bolshoye Lake. The Ca ion was not detected in either of the investigated lakes, owing to the lower solubility of calcium carbonate salts.

Studies of the soils near the lakes and of the groundwaters, as well as a more detailed study of the lakes themselves, may yield much interesting information as to whether the processes of soda formation are occurring at the present time, and what the intensity of these processes is.

To the article by I. V. Gebler, “Petukhovo Soda Lakes.”

Composition of the dense residue of the water of the soda lakes.

Lake Bolshoe (No. 1).

Lake Maloe (No. 2).

1. In the summer of 1924, the Slavgorod District Executive Committee, under whose jurisdiction the Petukhovo lakes are located, began organizing a small plant for causticizing soda, making use for this purpose of boilers and cast-iron pans delivered in their time by “Sodostroi,” partly to the lakes, partly to Kulunda station. In a small building of the plant—constructed on the very shore of the lakes—an iron boiler with a capacity of about 500 vedra has been installed, in which causticization of lyes will be carried out, and a thick-walled cast-iron pan for evaporating and melting caustic soda, of which it is proposed to obtain 500–600 poods per month. For the reasons just stated, such production must be considered economically little reliable. ↩↩↩↩↩

2. Chem. Zeitung 1892, p. 1691. ↩↩↩↩

3. S. Tanatar, “On the Formation of Soda in Nature,” Journal of the Russian Physico-Chemical Society, 1896, p. 376. ↩↩

4. Journal of the Russian Physico-Chemical Society, 1896, pp. 307 and 551. ↩