US789671A - Process of making alkaline fluosilicates. - Google Patents

Description

tity of quicklime would be requisite.

Niren Srares Patented May 9, 1905.

Pare r @rrrce,

PROCESS OF MAKING ALKALINE FLUOSILICATES.

SPECIFICATION forming part of Letters Patent No. 789,671, dated May 9, 1905.

Application filed September 14:, 1904. Serial No. 224,470

To aZZ whom, it may concern.-

Be it known that I, J ULIUS ALoIs REICI-I, a subject of the Emperor of Austria-Hungary, and a resident of II Praterstrasse 32, Vienna, Austria-Hungary, have invented a new and Improved Process of Making Alkaline Fluosilicate, of which the following is a full, clear, and exact description.

This invention relates to a process of making alkaline fluosilicate from the residues resultingduring the conversion ofalkaline silicoliuoric acid from the residues resulting during the conversion of alkaline silico-fiuorids (silico-fluorids of alkaline metals) in the manufacture of caustic alkalies or alkaline carbonates.

' For the manufacture of caustic alkalies or of alkaline carbonates, as is well known, it has been proposed for a long time to decompose alkaline silico liuorids (silico fluorids of alkaline metals) with calcium carbonate or quicklime, the alkaline silico-fluorid being either boiled with at most double the weight of calcium carbonate or with so much quicklime that the fluorin and the silicon are saturated, for which purpose about the same quan- These quantities are to be considered also as ample for the conversion of the alkaline silico-fluorid, as the conversion proceeds substantially according to the following equation:

from which it follows that one hundred and eighty-eight parts, by weight, of sodium silicofluorid or two hundred and twenty parts, by weight, of potassium silico-fluorid, two hundredandtwenty-fourparts,byweight,ofquicklime, or four hundred parts, by weight, of calcium carbonate are necessary. Consequently the weight of the quicklime is substantially the same as that of the alkaline silico-fluorid and the weight of calcium carbonate substantially double the weight of the alkaline silicofluorid. These processes, however, have up to the present had no success in practice, as experience has shown that from the calciferous residues resulting from these processes the silico-fiuoric acid cannot be recovered in a manner suflicient for industrial purposes,

which can alone adapt such a process for practical application. This defect could also be confirmed by the experiments which led up to the present process. If one proceeds according to hitherto-existing instructions, one obtains residues which are not only very imperfectly soluble in hydrochloric acid, but which are also bad and diflicult to wash, these residues on being treated with hydrochloric acid giving insufficient and varying yields of silico-iiuoric acid, as only fifty-four to sixtytwo per cent. of this acid can be recovered, even when hot acid is employed, a quantity which is consequently technically entirely insufficient.

According to the present process residues which are capable of being readily washed and which are almost completely soluble are successfully obtained from which the silicofluoric acid can be obtained or recovered in amounts corresponding with theory, whereby the alkaline silico-fluorid can henceforth be employed with advantage in particular for the industrial manufacture of caustic alkalies in a continuous process, which hitherto was not possible. This success is obtained by intimately mixing the residues, which can be obtained according to the mentioned boiling process, or, if desired, also according to the melting process proposed in recent times, with an additional fairly large quantity, suitably with about just as much lime (in the form of quicklime, milk of lime, hydrate of lime) as was requisite for the complete conversion of the alkaline silico-fluorid or for the complete saturation of the silicon and fiuorin contained therein. On account of the fact that, as already mentioned, about the same quantity of quicklime as alkaline silico-fiuorid is sufficient for this purpose, the resulting residues are mixed and boiled or heated, according to the present invention, once more with about the same quantity by weight of quicklime as the weight of the potassium silico-fluorid employed-that is, once more with as much lime as was requisite for the complete saturation of the silicon and fluorin in the alkaline silicofluorid. This quantity of lime can be suitably added before or during the conversion of the alkaline silico-fluorid, whereby in the latter case the requisite intimate mixing can be cfobvious or expected, and, moreover, it could not be foreseen that a special effect could be obtained thereby.

The following comparative experiments show the peculiar effect of the special surplus of lime.

If, for example, five kilograms of potassium silico-fluorid are boiled or heated with only about five kilograms quicklime, which would be enough to saturate the silicon and fluorin, according to the equation a residue is obtained which can be washed free from alkali only with difficulty with large quantities of water. If the attempt is made to dissolve this residue in hydrochloric acid, a turbid solution is the result, and even when hot hydrochloric acid is employed a great portion remains behind undissolved, which can be filtered and washed only with difiiculty, whereby if it is desired to determine the silicofiuoric acid, for example, by precipitation with baric chlorid exceedingly varying yields are obtained which lie between fifty-four and sixty-two per cent. If on the contrary five kilograms potassium silico-fiuorid are boiled with about ten kilograms quicklime, or if the residue obtained by boiling or heating five kilograms potassium silico-fiuorid with five kilograms quicklime is boiled alone with five kilograms quicklime, a residue is obtained which can be well and easily washed and which is easily soluble even in cold hydrochloric acid, only a small residue being left behind. The yield of recovered silico-fluoriC acid amounts to approximately one hundred per cent.

If five kilograms potassium silico-fluorid are boiled with only about seven kilograms quicklime, the washing of the obtained residue, although perceptible quantities (0.7 49 kilograms) of free quicklime go into solution, proceeds already substantially more slowly.

Its solubility in hydrochloric acid is incomplete and the yield in recovered silico-fluoric acid sinks to about 81.8 per cent, which can hardly be considered as sufficient for practice.

From these experiments it consequently follows that in order to obtain a yield corresponding as much as possible with theory so much lime must be added to the residues as is requisite for the complete conversion of the employed alkaline silico-fluorid or as is necessary for the complete saturation of the silicon and fiuorin contained in the alkaline silicofluorid.

In practice it has proved of advantage in order to save hydrochloric acid not to aim at an approximately theoretical yield, but to employ a somewhat smaller quantity of lime, which when using, forexample, five parts, by weight, of potassium silicofluorid can be taken at about nine parts, by weight, of quickime.

1f, namely, five kilograms potassium silicofiuorid are boiled with about nine kilograms quicklime, or if the residue obtained by boiling or heating live kilograms potassium silicofluorid with five kilograms quicklime is boiled with four kilograms quicklime, the yield of regenerated silico-fluoric acid amounts to about ninety-seven per cent, which, however, may be considered as sufficient for practice, especially as the residue obtained can likewise by means of washing still be easily freed from the alkali formed. If it is desired in certain cases to reduce the yield still further, the addition of lime can naturally be still further lowered in case this still allows a suitable industrial result to be obtained.

Comparative experiments with carbonate of lime turn out substantially in an analogous manner. Nevertheless the result is not so favorable as in the case of the employment of quicklime. If, for example, five kilograms sodium silico-fluorid are boiled with the customary double quantity of carbonate of lime, consequently with about ten kilograms, such voluminous and difficultly-washable residues are obtained that a recovery of the silico-fiuoric acid appears to be in no way feasible. If, on the contrary, five kilograms sodium silicofluorid are boiled with about 5 twenty to twenty-two kilograms carbonate of lime, consequently with about four times the amount, or if the residue obtained by boiling five kilograms sodium silico-fluorid with ten kilograms carbonate of lime is boiled alone with ten to twelve kilograms carbonate of lime, residues are obtained which are indeed substantially more readily washable, but from which, nevertheless, only up to about ninety per cent. of the employed silico-fluoric acid 5 can be recovered.

On account of the fact that, as is evident from the above, far greater quantities of lime are requisite and on account of the fact that, as has proved to be the case, the conversion with calcium carbonate proceeds more slowly it results therefrom that the industrial recovery of silico-fiuoric acid or the manufacture of alkaline carbonates in this manner in itself is not advantageous, but can be effected in a better manner by introducing carbonic acid into the caustic alkali obtained from the alkaline silico-fluorid by means of quicklime, for which purpose the carbonic acid can be employed which is obtained in the manufac- I3 ture of the quicklime requisite for'the conversion of the alkaline silieo-fluorid.

The process can be carried out industrially advantageously, for example, in the following manner: One hundred kilograms potassium silico-fluorid are boiled with about one hundred and eighty kilograms quieklime, or one hundred kilograms potassium silicofluorid are boiled with about one hundred kilograms quickliine, and thereupon the residue obtained in the last case with about eighty kilograms quicklime, for which about onefourth to one-half hour of effective boiling is requisite. The residue obtained is then washed with water in order to free it from the alkali formed, and. it is dissolved in about twelve hundred liters hydrochloric acid of about 12 Baume. The hydrochloric solution is mixed with potassium chlorid in excess and, indeed, with ninety-five to one hundred kilograms, whereby all the silico-fluoric acid is precipitated out as potassium silicofluorid. The alkali solution obtained gives by concentration about forty kilograms solid caustic potash.

From the foregoing specification it is clear that the additional amount or excess of lime cannot effect any reaction, since this additional amount may be added to the residues in order to facilitate the mixing of the lime with the residues. There is only produced a mechanical mixing of the lime excess with the residues, which mixing may be facilitated by boiling the alkaline fluosilicate immediately with the excess of lime. This excess does not, and should not, effect any chemical reaction. Probably such a reaction may take place if in the presence of the said excess of lime the residues are dissolved in hydrochloric acid, but in What manner the residues become fully soluble cannot be said. chlorid formed in stat'zmascendi has a special action on the bodies present in the residues to be brought into solution. If caustic lime is employed, the reaction may be explained by the following equation:

NftzSlFe+ I By dissolving the residues formed in hydrochloric acid the reaction takes place, perhaps, according to the following equation:

3C2LF2+CLSlOs+4CiLO+7TICl The main part of the added acid is neutral- Perhaps the calcium ized by the excess of lime present and by the lime combined with fluorin and silicium, as may be seen from the latter equation. The solution contains, therefore, a large amount of calcium chlorid, so that the formed potassium silico-fluorid is not in a pure acid solution, apart from the fact that potassium chlorid is added in excess, so that a further salt is present in the solution, which salts hinder a solution of a larger quantity of the alkaline silico-fluorid.

What I claim is 1. Process for recovering or obtaining alkaline fluosilicate from the residues formed in the conversion of alkaline silico-fluorid into alkali by means of lime, said process consisting in boiling the said residues with substantially as much lime as that necessary for the complete conversion of the alkaline silicofluorid, treating the precipitate formed during said boiling operation with an acid and precipitating therefrom alkaline fluosilicate by means of a salt of the desired alkali.

2. Process for recovering or obtaining alkaline fluosilicate from the residues formed in the conversion of alkaline silico-fiuorid into alkali by means of lime, said process consisting in boiling the said residues with substantially as much lime as that necessary for the complete conversion of the alkaline silicofluorid, treating the precipitate formed during said boiling operation after washing with an acid and precipitating therefrom the alkaline fluosilicate by means of a salt of the desired alkali.

3. Process for recovering alkaline fluosilicate from the residues formed in the conversion of alkaline silico-fluorid into alkali by means of lime, said process consisting in boiling the said residues with substantially as much lime as that necessary for the complete conversion of the alkaline silico-fiuorid, said boiling being effected during the production of the alkali itself for the sake of obtaining a more intimate mixing with the residues with the lime, treating the residues resulting from said boiling With an acid and precipitating therefrom the alkaline fiuosilicate by means of a salt of the desired alkali.

In testimony whereof I have signed my name to this specification in the presence of two subscribing Witnesses.

. JULIUS ALOIS REICH.

Witnesses:

ALVESTO S. Hoeuu, AUeUs'r FUGGER.