US2028390A - Method of producing the alkali metals - Google Patents

Description

51511.21 1 936, I v M, Q N 2,028,390

METHOD OF PRODUCING THE ALKALI METALS Filed Sept. 29, 1955 MPOR fies ssum: Cum .6

MM. fiezssuke no 300 4- '6' 7n 3 [on low 2662655 wz INVENTOR.

Patented Jan. 21, 1936 UNITED STATES METHOD OF PRODUCING 'rnn ALKALI METALS Miles G. Hanson, Flint, Mich.

Application September 29, 1933, Serial No..69l,491

4 Claims.

This invention relates to a process for economically producing metals of the alkali group, particularly lithium. It also describes a modification of the process by which mixtures of the alkali metals with magnesium or calcium may readily be made. Lithium and the other alkali metals are customarily produced by electrolysis of the fused salts. The process is expensive to carry out, particularly in the case of lithium because of its light weight and low melting point, 186 degrees centigrade which make it very difficult to separate from the heated fused salts of higher specific gravity.- Its extreme avidity for oxygen and nitrogen also make recovery by this method hard to accomplish without large losses of the metal.

This process is-based on the well-known fact that the alkali metals may be displaced from their oxygen and chlorine compounds by the reducing action of. magnesium, aluminum or calcium. Potassium and sodium compounds are reduced by the action of iron, although this action does not 'take place with lithium, caesium or rubidium compounds. These reducing reactions take place i with increasing violence as the atomic weight of the element is less, the reaction with lithium compounds being practically explosive. Small amounts of the alkali metals have been formed by conducting the reaction in a sealed container to exclude the air, the reaction being initiated by heating the container and the metal volatilized by the heat of the reaction condensed in a cooled portion of the reaction chamber. The state of the art is well summarized in Mellors Treatise on Inorganic and Theoretical Chemistry, vol. 2, pg. 450. The production of any considerable amount of metal by these reduction processes was found by me to be impossible because the reaction results in the liberation of gases which rapidly vbuild up the pressure in the sealed container so that the vapor pressure of the metal is exceeded and the metal is not volatilized but remains mixed in finely divided form with the other products of thereaction. In the case of. lithium the reaction proceeds so violently that the reduction of more than a few ounces of the material is not possible and practically no lithium metal is distilled from the reacting mass. In order to determine whether the alkali metals, particularlylithium, could be made economically by this process, a study of the factors affecting the process was made. It was realized that inorder to remove the metal set freeby the reducing action the metal must be volatilized and therefore the pressure at which the reaction takes place must not exceed the vapor pressure of the freed metal at the temperature of the reacting mass. Figure I shows the vapor pressures of certain of the alkali metals as they are dependent on temperature. The vapor pressure of magnesium is also included in this figure. Experiment showed that the reaction between any of the alkali metal-oxygen compounds and magnesium resulted in an immediate pressure rise within a sealed container and that the pressure could only be kept low enough to insure complete volatilization of the freed al-- kali metal by continuously removing the gas set free by the reaction. In the case of lithium, for instance, the pressure must be kept below 6 mm of mercury if the temperature of the reactin mass is 900 degrees centigrade.

I have found these reducing reactions to proceed'at a very rapid rate, this being particularly so in the case of lithium compounds reduced with calcium or magnesium. If any quantity of the metal is to be prepared the pressure cannot be kept low enough to insure volatilization of even a small portion of ,the alkali metal that is freed from its compound, even though very high ca-- pacity vacuum'pumps be used. It has been proposed to slow down these reactions by admixing inert materials to the reacting mass, such as calcined magnesium oxide, but the addition of even as large an amount as 50% by weight of the total charge did not reduce the reaction rate enough to be of value in maintaining a. low pressure in the retort. The effect of grain size on the rate of re action was investigated and it was found that the grain sizeiis an important factor. Thus, the reaction between lithium oxide and magnesium, both ground to pass a mesh per inch screen, is extremely rapid, the use of particles of magnesium just able to pass a 25 mesh per inch screen results in a slow reaction taking several minutes for completion. During the slow progress of the reaction with large grain materials a considera-- ble portion of the magnesium is volatilized and the reduction of the oxide is not complete. The condensed metal is found mm a mixture of lithium and magnesium. Thus the control of the grain size of the mixture affords control of the condensed metal composition as well as of the reaction rate.

To secure the pure alkali metals it was found that rapid, complete reduction of the'alkali metal compound was essential, and that the reaction take place while the pressure within the reaction chamber is kept low. A reaction that takes place under such conditions of pressure and temperature as to fail to volatilize the metal formed will not result in eflicient recovery of the metal. .55

- paratus shown in Figure 2 was developed. The

vacuum retort above I50'degrees centigrade over the lower two thirds of its length and is provided with a water cooled cover 2, upon which the volatilized metal is condensed. The pressure is reduced by a vacuum pump 1, of large capacity connected directly to the retort, and the pressure is continuously recorded by a pressure gauge, 3. is provided with a side tube opening into it, 4, and provided with a sliding member entering through a vacuum tight stufling box. This rod is for the purpose of injecting the prepared reacting materials into the vacuum retort as the reaction progresses. The material, consisting of a mixture of alkali metal compound and the reducing metal, is compressed into pellets, 5, which are placed in the side tube. In operation the retort is heated to 750-800 degrees centigrade, the side tube is filled with pellets of the reacting materials and the pressure in the retort is reduced to below 1 mm. of mercury. The reaction is started by forcing the rod, 6, inward to propel a pellet into the retort where reaction takes place. The reaction results in a rise of pressure as indicated by the gauge. The pressure is prevented from rising too far by the continuous operation of the pump, and as each pellet completes its reaction another is pushed into the retort. The size of the pellet is determined asthat which does not cause excess pressure rise with the pump and size of reaction chamber used. As an example a mixture of lithium oxide and magnesium, both powdered to pass a 100 mesh per inch screen, can be charged into a retort of 2 liters capacity and a pump able to remove 5 liters of gas per minute is able to maintain the required pressure if the pellet does not exceed, 10 grams in weight.

After all the material has reacted the retort is cooled to room temperature and the top removed in an inert gas atmosphere. The condensed metal may thus be removed without danger of fire.

During the course of these experiments we have found that either the carbonate, hydroxide,

or oxide of the alkali metal may be used. We prefer the oxide where a final pure product is desired. The carbonate tends to produce an appreciable amount of carbon in the deposit, and the hydroxide has the disadvantage of producing a more explosive reaction, charge that may be employed. While we do not limit ourselves thereto, we prefer to manufacture lithiumfrom the oxide as obtained by the action of hydrogen on molten lithium carbonate. It is preferable to heat the carbonate to 750-800 degrees centigrade in hydrogen for a sufflcient time to reduce the carbonate completely to oxide. This process has the additional advantage that the amount of lithium carried into the retort by each pound of lithium oxide is much greater than in the case of the carbonate or hydroxide, which increased efliciency is claimed as part of the present invention. The carbonate is the most readily available of the compounds of lithium and by the process in accordance with the present invention it is made available for a reduction for which it has not heretofore been used commercially.

While it. is preferable to use magnesium as a reducing agent in the preparation of lithium, caseium or rubidium, either calcium or aluminum may be employed instead of magnesium I, is maintained at a temperature;

The retort thus lessening the v without departing from the scope of the invention.

I have found that the composition of the condensed metal recovered can be predetermined and varied from substantially pure lithium, to yields containing as high as 90% of magnesium by varying the proportion and/or grain size of the alkali metal compound and the magnesium employed in the reaction.

For certain purposes the use of lithium mixed with magnesium is preferable to the pure metal as difliculty and danger in handling are eliminated, this being particularly the case in metallurgical applications where lithium is added to molten metals at high temperatures.

The making of an alloy of lithium with magnesium or calcium by mixingthe molten metals is difficult and dangerous. By the present invention this step is eliminated, as a mixture of the alkali metals with magnesium or calcium can be condensed as the final step in my process, the mixture charged into the retort being properly proportioned.

As an example, one and one half parts of magnesium to one part of lithium oxide by weight will produce metal containing about 20% magnesium. The mixture may be varied to suit the composition desired, or pellets of pure magnesium may be alternated with pellets of the mixture, though in the latter case the product is not as uniform, but tends to be deposited in layers on the cooled portion of the retort.

To render the handling of the alkali metal less dangerous a final pellet of magnesium may be charged into the retort, this coats the alkali metal deposit with a thin coating of magnesium and protects the deposit from air or gases that would react with the alkali metal as the retort is in process of being unloaded.

Having now described the invention, what is claimed is:

1. The process of producing lithium meta l by the reduction of lithium oxide by metallic magnesium or aluminum, the reduction being carried out at a pressure below 6 mm. of mercury, and within a temperature range of 750-900 degrees centigrade, and the metal being recovered by condensation of the vapor on a cooled portion of the apparatus.

2. The process of producing lithium metal consisting in heating a mixture of lithium oxide and not more than an equal amount by weight of magnesium metal to a temperature or 750 to 900 degrees centigrade, the reaction being carried out at apressure of not more than 6 mm. of mercury.

3. A process for producing lithium metal comprising the following steps; production of lithium oxide from lithium carbonate by heating the carbonate in the presence of hydrogen to 750-900 degrees centigrade, reduction of this oxide product by ,metallic magnesium at a temperature of 750-900 degrees centigrade under a pressure of not more than 6 mm. of mercury, the vaporized lithium being condensed on a cooled portion of the container.

4. The process of producing an intimate mixture of finely divided metallic lithium and magnesium, consisting in heating a mixture of lithium oxide and magnesium to a temperature of 750-900 lithium oxide by weight.

MILES G. HANSON.

Images