US2954277A - Method of processing magnesium chlorides - Google Patents

Description

Sept. 27, 1960 A. M. THoMsr-:N 2,954,277

METHOD o? PROCESSING MAGNESIUM CHLORIDES Filed .my 15, 1959 IN V EN TOR.

METHOD OF PROCESSING MAGNESIUM CHLORIDES Alfred M. Thomsen, 265 Buckingham Way, Apt. 402, San Francisco 27, Calif.

Filed `Iuly 15, 1959, Ser. No. 827,357

Claims. (Cl. 23--201) The chemistry involved in the decomposition of magnesium chlorides by means of heat, in the presence of oxygen and/ or steam, is admittedly very old. Apparently it was known before 1850, and during the following three or four decades it was spasmodically in use on a limited scale. However, no success attended these trials and it was finally abandoned everywhere. The reason for such abandonment must be found in the technical difficulties involved, so, the instant application is concerned not with improvements in the chemistry involved but with the mechanics of the operation.

The chemical reactions may be represented as follows: MgCl2-f-O=MgO-|-Cl2-9.6 Cal., and

Calories. While both these reactions are endothermic it is obvious that the input of energy is quite small. Economic reasons are also involved in suggesting that this ancient chapter be revived. There is in the Great Salt Lake, Utah, an inexhaustible source of magnesium chloride. In a co-pending application, to wit, Ser. No. 791,- 018, filed Feb. 4, 1959, called Method of Making Magnesium from Salt Lake Brines, I have already called attention to this resource. Another factor is the tremendous modern demand for a high-grade magnesium oxide.

Perhaps an even more potent reason for such revival of interest lies in the commercial chlorine-hydrochloric acid industry. Once safely entrenched behind the enormous caustic soda demand lthe electrolytic soda industry now faces a complete reversal. In the electrolysis of salt, chlorine is far the more important and caustic soda disposal presents a real problem. It is, therefore, of supreme importance to seek means of producing chlorine with a substitute for caustic soda as an additional product. Such a substitute obviously resides in magnesia with a present market of nearly a million tons per annum.

The equations given previously as the reactions whereby such chlorine and/or hydrochloric acid may be produced show only the anhydrous salt. Unfortunately, the natural salt is not anhydrous but crystallizes with six molecules of water of crystallization. Of this water, tive moles may be removed with but little loss of chlorine but to remove the last mole by heat alone results in a copious evolution of HCl and the formation of the Oxy-chloride. To avoid this has been a challenge for many years. In the previously mentioned application I have added an equally `old way, namely, a prior crystallization of the double salt with ammonium chloride, which may be dehydrated without loss. On subsequent heating ammonium chloride is evolved, condensed, and re-cycled, after which the magnesium chloride may be fused with impunity. Another way would be to de-hydrate in an atmosphere of hydrochloric acid, but that is rather hard to contemplate in any commercial opera-tion. But as already mentioned, the commercial production, from the Oxy-chloride, has been operated for years though subsequently it had 'latent Patented Sept. V27', 1960 to bow to the superiority of electrolysis. I have illus -trated my version of how that may, once more, become of commercial merit in the drawing which I will now describe. Subsequently I shall add a few words as to its `applicability to the anhydrous salt as well.

In said drawing, I commence at the bottom, where in a mixer I commingle crystallized magnesium chloride, which melts in its water of crystallization, with an intermediate form of magnesium oxide, still retaining a little chlorine as designated by the x placed beneath the Cl. I use enough of the oxide to produce a solid mass when the reaction is complete. Provisionally, that means somewhat more than an equal amount of magnesium as oxide to that present as chloride, yet rather less than l/z times as much. In elect this mixture is similar to the Sorel cement used in magnesia ooring. This solid material is crushed and classied. An acceptable size is between l5 microns and 100 microns, the dust being separated by air iioating and re-cycled to the mixer.

The accepted fraction is now fed continually into the device called Reactor 1, containing an ordinary fluidized bed of de-hydrated magnesium oxy-chloride. It is maintained in said fluidized state by the upward traverse of air heated to about 550 C. in a discharging heat accumulator placed in the drawing below, left, of said reactor. When the temperature of the air `drops to about 450 C. reversal of ow takes place and the charging accumulator which has meanwhile been heating, it substituted. Above the reactor I have indicated another pair of heat accumulators that function as coolers for the gas leaving said reactor and as preheaters for the combustion air required by the fuel. In this manner, virtually all sensible heat can' be re-cycled, only that heat being required in eifecting a chemical change being required. It Will be obvious that by thus recovering the larger part of the sensible heat in the gas leaving the reactor and transferring it to the air of combustion I have achieved a great economy in heat usage. It will be equally obvious that the addition of 'a small continuous flow of we Oxy-chloride cannot adversely elfect the mechanical coni dition of the de-hydrated, infusible material composing the fluidized bed.

A unit of my system is thus seen to consist of the reactor, which functions continuously in one direction, and four heat accumulators that function, alternately, as receiving and as giving out this stored heat energy. Of course, this is a schematic representation for the sake of simplicity. Inasmuch as it is a copy of the blast furnace stoves of the high iron furnace it is a well known heating technique which I have applied to a new purpose and made regenerative as well. An actual installation would have three accuniulators both above and below the represented reactor. Each side would then show two accumulators discharging and one receiving heat. On reversal of flow, the cooler one of the discharging stoves would be switched to the heating cycle, and the one just heated to its maximum temperature would be used to replace it. In this manner a smooth, uninterrupted flow of hot air would traverse the uidized bed.

In this manner the magnesium chloride is converted to an Oxy-chloride and approximately one-half of the chlorine is evolved as hydrochloric acid together with almost all the water of hydration. Nitrogen, unused oxygen, steam and HCl is then represented as the gases leaving reactor l. I have then shown the increase in bulk of said fluidized bed passing on to reactor 2, where the identical operation is repeated, only, as the offending water has been removed, the passage of hot air now yields chlorine in place of hydrochloric acid. In a continuous process, with rather large storage in the uidized beds, it is impossible to make this cut absolute. A little chlorine will be evolved in reactor 1, and a little HC1 will be evolved in reactor 2, as I have indicated. The surplus increment from reactor 2 is then passed on to reactor 3 for nal removal of all chlorine. This is effected by the deliberate addition `of steam to the traversing air so the gas leaving the reactor again contains its chlorine as HC1, not as free chlorine.

Such use of steam is mandatory if it be `desired to have all chlorine removed without the otherwise unavoidable rise in temperature needed if steam be omitted. The total chlorine of the feed is thus obtained in the following ratio, chlorine 40%, hydrochloric acid 60%. An excellent grade of magnesia is likewise produced if this technique be followed. Collectively, these items are most important. Unless ALL be taken into consideration as objectives of relatively equal importance my application of this old chemistry, in spite of the modern technique and the heat recuperative features, would still remain economically unremunerative.

Throughout all reactors the temperature will be maintained substantially within the range previously specified, to wit, between a low of 450 C. and a high of 550 C. Of course, any deviation from these temperatures is not fatal but within these brackets the reactions proceed smoothly. Similarly, the best place to observe such temperature is by placing the pyrometer within the space occupied by the fluidized bed. Itis Within this bed, where the reaction occurs, that temperature control is vital.

It will be obvious that if anhydrous magnesium chloride be substituted for the Sorel Cement then this three-step process becomes automatically a two-step process. The only purpose served by Reactor 1 is dehydration with consequent evolution of hydrochloric acid. If a start be made with anhydrous material then reactors 2 and 3 will suice, and chlorine will be obtained at once. However, the second step with steam is still mandatory to produce an acceptable magnesia product. In this case the ratio of chlorine to hydrochloric acid will be about l to 1, by volume.

Many minor modifications suggest themselves once the plan of my operation is understood. Thus, once the uidized bed is established in reactor l, it will be plain that no additional water would be introduced if a liquid feed of magnesium chloride were introduced directly as an atomized product. Inasmuch as said hydrated salt melts in its water of crystallization. It is immaterial whether such an atomized spray be directed upon the surface of the bed or simply injected into the space above said bed which would then act as a spray drier, dehydration being completed in the iluidized bed into which such a spray, or rather the solids from said spray, would ultimately arrive.

However, the very reverse of this procedure would work to better advantage. Again, taking advantage of the properties of magnesium chloride, we lind that threefourths of the resident water can be evaporated with a loss of as little as 5% of the resident chlorine which will, of course, be lost as HC1. It would, therefore, be advantageous to partially dry this crystallized salt down to a water content of approximately 20%, grind it to powder, and then introduce it in this form into the fluidized bed of reactor 1. In this manner much water would be prevented from entering into the gaseous materials leaving said reactor which, in turn, would facilitate further work in the recovery of hydrochloric acid therein. The same objective would be reached if the operation in reactor l were two-stage instead of single stage. Let, therefore, this work be performed with a spray of fused salt in the iirst of these reactors, followed by a complete de-hydration in the second. Obviously, the acid from the rst step would only be iit to discard while the acid from the second step would be correspondingly improved. As a iinal refinement fire gas might, be used directly in the rst stage.

`Inasmuch as the decomposiiton of magnesium chloride is a function of time as well as temperature it is obvious that the time of residence within said fluidized bed is also a matter of importance. This must be obtained by analysis of the fraction withdrawn from said bed, but once established it is governed solely by the rate of feed into said bed. Parenthetically it may be stated that in a two-stage dehydration, as previously described it would be advisable to lower the temperature of the rst stage to 300 C., or even less, and thus minimize the loss of hydrochloric acid in said stage. Having thus fully described my process,

I claim:

1. The method of processing anhydrous magnesium chloride which comprises; adding said magnesium chloride in powdered form to a fluidized bed consisting essentially of magnesium oxide with some retained chloride, maintained in a fluidized state by the upward traverse through same of air heated to a temperature between an approximate 450 C. and an approximate 550 C. with attendant evolution of chlorine gas; withdrawing continuously an amount of said uidized magnesium oxide corresponding to the added magnesium chloride and adding said withdrawal to a uidized bed consisting essentially of chlorine-free magnesium oxide maintained in said iluidized state by the upward traverse through same of a current of air and steam heated to a temperature between an approximate 450 C. and an approximate 550 C., thus obtaining substantially pure magnesium oxide with attendant evolution of hydrochloric acid gas; and withdrawing continuously from said fluidized bed of magnesium oxide an amount of said magnesium oxide corresponding to the anhydrous magnesium chloride originally added to the iirst fluidized bed.

2. The method of processing magnesium chloride which comprises; adding magnesium chloride in hydrated form to a fluidized bed of relatively anhydrous magnesium oxychloride maintained in a uidized state by the upward traverse through same of a current of air heated to a temperature between an approximate 450 C. and an approximate 550 C. thus converting the additive hydrated magnesium chloride into substantially anhydrous oxychloride with attendant evolution of hydrochloric acid gas; withdrawing an amount of said oxychloride corresponding to the added magnesium chloride, continuously, and adding said withdrawal to another fluidized bed consisting essentially of magnesium oxide with some residual oxychloride, maintained in a fluidized state by the upward traverse through same of a current of air heated to a temperature between an approximate 450 C. and an approximate 550 C. thus converting the added oxychloride into a product consisting essentially of magnesium oxide, still retaining oxychloride, with attendant evolution of free chlorine gas; withdrawing continuously an amount of said magnesium oxide corresponding to the added oxychloride of magnesium and adding said withdrawal to a iluidized bed of magnesium oxide maintained in said fluidized state by the upward traverse through same of a current of air and steam heated to a temperature between an approximate 450 C. and an approximate 550 C. thus obtaining substantially pure magnesium oxide with attendant evolution of hydrochloric acid gas; and withdrawing continuously from said fluidized bed of magnesium oxide an amount of said magnesium oxide corresponding to the hydrated magnesium chloride originally added to the first iuidized bed.

3. The method of processing magnesium chloride set forth in claim 2, with the added step that the dehydration of magnesium chloride described therein be conducted in two steps with two iluidized beds, separate and distinct from one another, the temperature of the first bed being maintained substantially 200 C. lower in temperature than the second bed, thus performing a partial dehydration with low evolution of hydrochloric acid before transferring to the finishing bed in which the temperature 5 s maintained at between 450 C. and 550 C. as previously specified.

4. The method of processing magnesium chloride set forth in claim 2, with the added step that the gaseous fluid emanating from the fluidized beds be cooled by passage through a heat accumulator; conveying the heat thus stored, on reversal of ow, to the air of combustion required in heating another heat accumulator, the second accumulator, in turn, conveying its stored heat to the uidizing medium on reversal of flow.

5. The method of processing magnesium chloride set forth in claim 1, with the added step that the gaseous fluid emanating from the uidized beds be cooled by passage through a heat accumulator; conveying the heat thus stored, on reversal of flow, to the air of combustion required in heating another heat accumulator, the second accumulator, in turn, conveying its stored heat to the uidzing medium on reversal of ow.

References Cited in the tile of this patent UNITED STATES PATENTS Ebner Apr. 18, 19139 2,413,292 Christensen Dec. 31, 1946 2,694,620 Lathe Nov. 16, 1954

Claims (1)

1. THE METHOD OF PROCESSING ANHYDROUS MAGNESIUM CHLORIDE WHICH COMPRISES, ADDING SAID MAGNESIUM CHLORIDE IN POWDERED FORM TO A FLUIDIZED BED CONSISTING ESSENTIALLY OF MAGNESIUM OXIDE WITH SOME RETAINED CHLORIDE, MAINTAINED IN A FLUIDIZED STATE BY THE UPWARD TRAVERSE THROUGH SAME OF AIR HEATED TO A TEMPERATURE BETWEEN AN APPROXIMATE 450*C. AND AN APPROXIMATE 550*C. WITH ATTENDANT EVOLUTION OF CHLORINE GAS, WITHDRAWING CONTINUOUSLY AN AMOUNT OF SAID FLUIDIZED MAGNESIUM OXIDE CORRESPONDING TO THE ADDED MAGNESIUM CHLORIDE AND ADDING SAID WITHDRAWAL TO A FLUIDIZED BED CONSISTING ESSENTIALLY OF CHLORINE-FREE-MAGNESIUM OXIDE MAINTAINED IN SAID FLUIDIZED STATE BY THE UPWARD TRAVERSE THROUGH SAME OF A CURRENT OF AIR AND STEAM HEATED TO A TEMPERATURE BETWEEN AN APPROXIMATE 450*C. AND AN APPROXIMATE 550* C., THUS OBTAINING SUBSTANTIALLY PURE MAGNESIUM OXIDE WITH ATTENDANT EVOLUTION OF HYDROCHLORIC ACID GAS, AND WITHDRAWING CONTINUOUSLY FROM SAID FLUIDIZED BED OF MAGNESIUM OXIDE AN AMOUNT OF SAID MAGNESIUM OXIDE CORRESPONDING TO THE ANHYDROUS MAGNESIUM CHLORIDE ORIGINALLY ADDED TO THE FIRST FLUIDIZED BED.

Images