Professor I. F. Ponomarev. Docent M. N. Yakimov.

Effect of Magnesium Sulfate on the Strength of Magnesia Cement.

A staff member of the West Siberian Scientific Research Institute in the city of Tomsk, M. N. Yakimov, carried out work on studying the conditions of evaporation of water from the brine of Lake Bolshoe Yarovoe, which belongs to the series of salt lakes of the Kulunda Steppe.

As a result of this study it was found that under certain conditions solutions of magnesium chloride with a high content of magnesium sulfate are obtained.

There may be very large reserves of such natural brines in the Kulunda Steppe. Thus, for example, the reserve of Lake Bolshoe Yarovoe alone is expressed at about 20 million tons.

Magnesium chloride is one of the constituent parts of magnesia cement, which is also called Sorel cement. It is usually considered that Sorel cement is formed by caustic magnesium and magnesium oxide, i.e. so-called caustic magnesia.

We were faced with the question of whether magnesium sulfate would have an undesirable or harmful effect in the manufacture of magnesia cement. It was necessary to determine, expressed in figures, how magnesium sulfate affects the properties of magnesia cement.

To solve this problem, caustic magnesite was taken, obtained from the Magnesite plant—grade No. 1—and kept in the laboratory for 2 years, unstoppered, in a vessel. To prepare the mixtures, magnesium chloride and magnesium sulfate were used; these substances were taken in the form of chemically pure reagents from Goslaboratoriya.

At first, experiments were made with mixtures of magnesium chloride and caustic magnesite without an admixture of magnesium sulfate. The magnesium chloride and magnesium oxide were taken in a ratio of 1 : 1, while the magnesium chloride was taken in the form of solutions of the following concentrations:

1) 25°; 2) 30°; 3) 31°; 4) 32° Baumé.

The mass obtained from such mixtures had throughout a viscous consistency, close to that of thick plasters. In order to test the effect of the concentration of magnesium chloride on the strength of magnesia cement, figure-eight specimens were made. To prevent the mass from flowing out of the molds during the preparation of the figure-eight specimens, the molds had to be lubricated with thick petroleum jelly. After twenty-four hours the figure-eight specimens, both of pure magnesia cement and with sand, still remained liquid, but after two days they acquired considerable strength. The tests…

were carried out after two and 7 days. The test results are given in the following table:

The highest strength was shown by the briquettes with a concentration of magnesium chloride of 31° Bé; therefore all the work was carried out with the same solution, having a concentration of 31° Bé.

For the investigation a system of 14 mixtures was taken, in which the content of magnesium oxide remained constant throughout—50%, with the aim of maintaining the adopted mixture in which the ratio between magnesium oxide and the soluble magnesia salts is equal to 1:1.

The composition of the soluble salts was subjected to gradual change, while the sum of magnesium chloride and magnesium sulfate remained constant throughout, equal to 50% of the entire mixture. The amount of magnesium sulfate gradually increased, starting from 0.5% in the second mixture to 20% in the 14th mixture.

From these magnesia mixtures, briquettes were made, 6 pieces for each mixture. The briquettes were stored in air and were then subjected to tensile testing on a Michaelis apparatus after the usual periods following gauging: 7 and 28 days.

A total of 84 briquettes were made of neat cement without admixture of sand and 84 briquettes with filler—normal sand, of the kind usually used for making briquettes when testing Portland cement, i.e., bounded by sieves having 144 and 225 openings per sq. centimeter. The mixture with sand was always in the same ratio, 1:3. Below are given the results of 168 determinations, which were made for 14 mixtures of neat cement and for 14 mixtures with sand. From the result of each 6 determinations, the mean value for each mixture was derived.

Test results for 7-day specimens

During storage of the specimens in air it was observed that the rate of increase in strength depends on the air temperature.

This was reflected especially in the specimens of mixtures No. 8, No. 11, and No. 12. During the storage of the specimens of these mixtures, in the room where the cabinet for storing the specimens was located the stove was not heated, as a result of which the air temperature dropped somewhat.

The results of the initial tests of these three mixtures, No. 8, 11, and 12, are presented in the following table:

Owing to the sharp deviation of the results obtained from the general course of the strength curve for seven-day figure-eights, tests were redone for mixtures Nos. 8, 11, and 12. The new results, as is evident by comparison with those given earlier, proved considerably higher.

From this one should draw the very interesting conclusion that the strength of articles made of magnesia cement also depends on the temperature at which the hardening of the magnesia cement proceeds. We were unable to go more deeply into this aspect in our investigation. It may serve as the subject of a special study for examining the influence of temperature on the strength of magnesia cement.

The data obtained for 14 mixtures are plotted as a curve on the diagram, where the strengths of seven-day figure-eights are expressed as a function of the addition of magnesium sulfate.

The curve for pure cement, indicated by the solid line (Fig. 1), shows that when MgSO₄ is added to MgCl₂, magnesia cement without the addition of sand as a filler (ballast) has its tensile strength lowered somewhat—from 57 kg per sq. cm for cement without MgSO₄ addition to 42 kg per sq. cm for cement containing 50% MgO, 30% MgCl₂, and 20% MgSO₄. The curve drops more steeply, by 10 kg per sq. cm, from 0 to 5% MgSO₄, and then the fall in strength proceeds more gently, giving a difference of 5 kg per sq. cm over the interval of 15%.

Mixtures with sand behave differently. In them no influence of the MgSO₄ addition on the tensile strength of the figure-eights is observed; the curve runs almost parallel to the abscissa from 0 to 15% MgSO₄; according to the obtained value of the strength of the mixture with 20% MgSO₄ = 38 kg per sq. cm, the curve rises with increasing MgSO₄ content. Thus it may be considered that, in mixtures with sand, the MgSO₄ addition has no practical significance for the strength of magnesia cement.

To determine the effect of the addition of \( \mathrm{MgSO_4} \) on the strength of magnesia cement during further hardening, 112 eight-shaped specimens were made from 14 seven-day batches.

Fig. 1.

Figure labels:
Vertical axis: Tensile strength in kg/cm²
Horizontal axis: Content of \( \mathrm{MgSO_4} \) (weight %)
Legend: \(x\) — neat cement; \(o\) — with sand 1:3

from mixtures of the same composition as indicated above: 56 eight-shaped specimens of neat cement and 56 eight-shaped specimens with sand in a ratio of 1:3. As a result of each set of 4 determinations for a given mixture, the average value was derived.

Results of tests of 28-day specimens

The data obtained for 14 mixtures are expressed by two curves on the diagram (Fig. 2). The curve for neat cement is drawn as a solid line; that for the mixture with sand is drawn as a dashed line.

The addition of MgSO₄, over a long hardening period, even increases—although only slightly—the strength both of neat cement and of mixtures with sand. For neat cement the curve is almost parallel to the axis of abscissas, having an ordinate equal on average to 53 kg per sq. cm, and only for the mixture with 20% MgSO₄ do we have a somewhat higher strength—57.5. For mixtures with sand, an increase from 27 to 39.5 is observed.

Comparing both curves of tensile strength after 7 days of hardening (Fig. 1) and after 28 days (Fig. 2), the following conclusions may be drawn:

1. Magnesium sulfate has an effect on the strength of neat magnesia cement, slightly lowering the tensile strength: from 57 to 42 kg per sq. cm.

2. This decrease is observed only in the first days of hardening; thereafter, by the one-month period, the strengths of the neat-cement mixtures even out and, on average, remain around 55 kg per sq. cm.

3. In mixtures with sand, no such effect of the addition of \(\mathrm{MgSO_4}\) on strength was found: after seven days of hardening, the tensile strength of the mixtures has an average value of about 32 kg per sq. cm.

4. With further hardening of the mixtures with sand, even an increase in strength is observed; the maximum is in the mixture with 7% \(\mathrm{MgSO_4}\), with an average value of about 43 kg per sq. cm, and it decreases as the \(\mathrm{MgSO_4}\) addition is increased, nevertheless remaining at an average value of about 38 kg per sq. cm.

Fig. 2.

The following table gives the data of the investigation on the basis of which the diagrams were constructed.

Summarizing the interesting results obtained, one may draw the following general conclusion: the addition of MgSO₄ to magnesia cement does not exert such an influence on its properties as would have to be taken into account in practical application. Thus, in obtaining magnesia brine suitable for use as a component of magnesia cement, attention should be paid only to the content of common salt.

In our investigations we separated out the common salt. However, in further studies it is necessary to determine what influence an admixture of common salt has on the properties of magnesia cement. This will have great economic significance, since the cost of brine without the separation of NaCl is lower.

On the basis of these conclusions, an experiment was carried out using not an artificially prepared mixture of MgCl₂ and MgSO₄, but a mixture of caustic magnesite with brine having a density of 31° Baumé. This brine had the following composition:

The common salt was removed.

Eights were made, for the mixing of which natural brine was taken; sand and wood sawdust were used as fillers.

The tensile strength of the eights obtained after 7 days is indicated in the following table:

For mixture No. 9

\[ \text{With respect to } \frac{\mathrm{MgSO_4}}{\mathrm{MgCl_2}} = \frac{3.42}{23.51}\cdot 100 = 14.5\% \]

the ordinary brine taken for the experiment corresponds to the mixture indicated in the table under No. 9 with a 14% ratio.

These experimental results indicate that the brine of the Kulunda lakes can be used for the manufacture of magnesia cement and articles from it with both mineral and organic fillers.

Significance of magnesia salts for the production of building materials

The extensive construction connected with the implementation of the UKK problem requires a colossal quantity of building materials. Magnesia cement can render very great assistance, especially for carrying out the construction of the Kulunda com-

Binat, which is to receive cement from the Yashkino plant, or from Chernorechenskoye, or from the Ural plants.

The planned construction of sugar factories along the Barnaul–Semipalatinsk railway line, the construction of the Barnaul combines, and construction in Kazakhstan will also be the nearest consumers of Kulunda magnesia cement.

The reserves of magnesian salts in the Kulunda lakes make it possible to place on the agenda not only the task of extracting magnesian salts, \(MgCl_2\) and \(MgSO_4\), which are required for the hardening of magnesia cement, but possibly also the production of magnesium oxide.

According to the process already developed for the decomposition of magnesium chloride, when it is acted upon by sharp superheated steam, a very pure magnesium oxide and hydrogen chloride are obtained, which may be expressed by the following reaction:

\[ MgCl_2 + H_2O = MgO + 2HCl \]

According to the calculation of Giprochim, the cost of the magnesium oxide obtained is quite low—about 20 rubles per ton.

But this process encounters a great difficulty in that it is necessary to find an application for the hydrochloric acid obtained as a by-product.

To obtain 1 ton of \(MgO\), about 2.5 tons of \(MgCl_2\) are required, and in the process about 5 tons of 36% hydrochloric acid will be obtained. The question arises: where is such a large quantity of hydrochloric acid to be put?

If the reserve of only one lake—Bolshoye Yarovoye—is taken as 20,000,000 tons of magnesian salts, then, with its exploitation over 50 years, it would be possible to produce annually 114,000 tons of magnesium oxide and 114,000 tons of magnesian brine required for the hardening of magnesia cement. Using sawdust as filler, it would be possible to obtain annually about 300,000 tons of magnesian fibrolite in the form of various fibrolite products, or considerably more—up to 500,000 tons—of magnesian concrete.

The utilization of the enormous quantity of hydrochloric acid obtained must be resolved as a result of extensive investigations on the basis of a number of complex problems.

Major consumers of hydrochloric acid may be the forest-chemical industry and the production of \(AlCl_3\) from clays. Aluminum chloride is required for the cracking process; in the processing of coal into liquid fuel it can be used for the production of metallic aluminum, and it can also serve as an intermediate product for obtaining aluminum oxide by the action of ammonia on \(AlCl_3\). Aluminum oxide, provided its cost is low, may serve to increase the refractoriness of clays in the production of highly refractory materials, in which the metallurgy of ferrous and nonferrous metals, the production of coke, glass, porcelain, etc., are greatly interested. All this must be taken into account in developing the chemical five-year plan for the West Siberian Territory.

In the Kulunda steppe, where the lakes with magnesian salts are located,

...with salts; deposits of dolomitized limestones are known. In the near future, the study of this raw material is planned, with the aim of clarifying whether it can be used to replace caustic magnesite. If this question is resolved favorably, the raw-material base for the production of magnesia cement will be considerably expanded.

Here one can only note that the conditions for the decomposition of dolomite were studied by Academician A. A. Baikov. When fired in the temperature range of 600–700°, magnesium carbonate decomposes, while the limestone remains unchanged. The product obtained does not slake to a fluff when wetted with water. It must be ground, after which it possesses the properties of caustic magnesite. The admixture of ground limestone does not interfere with the hardening process, but the cement obtained is weaker than caustic magnesite made from magnesium carbonate.

The investigation of the effect of $\mathrm{MgSO_4}$ on the strength of magnesia cement was carried out in the Laboratory of Silicate Technology of the Siberian Chemical-Technological Institute; the experimental part was performed by laboratory assistant N. M. Nevzorov and preparator I. P. Starikov.