US2864691A - Method of producing titanium - Google Patents

Description

Dem-$6, 1958 J. J. GRAY METHOD OF PRODUCING TITANIUM Filed Dec. 9, 1955 INVENTOR J1 mes cfokrz Gray ATTORNEYS METHOD OF PRUDUCING TITANIUM James John Gray, Widnes, England, assignor to imperial Chemical industries Limited, London, England, a corporation of Great Britain Application December 9, 1955, Serial No. 552,180

Claims priority, application Great Britain December 10, 1954 5 Claims. (Cl. 75-845) This invention relates to the manufacture of metals and more particularly to the manufacture of titanium metal.

At the present time commercial large scale production of titanium metal appears to be mainly confined to the Kroll process, wherein titanium tetrachloride is reduced by magnesium to give a mixture oftitanium metal and magnesium chloride from which the titanium is recovered by aqueous leaching or vacuum distillation purification methods.

It is, however, well known that titanium can be made by the reduction of titanium tetrachloride with metallic sodium. Thus in 1910, Hunter (J. American Chemical Society 1910, vol. 32, p. 330) prepared titanium by reacting purified titanium tetrachloride with sodium in a steel bomb. On heating the system to redness the reaction took place with explosive violence and was almost instantaneous. These results were confirmed by Lely and Hamburger (Chemical Abstracts 1914, vol. 8, p. 2855) who likewise reduced the tetrachloride with sodium in a steel bomb and then fused the metal in an electric vacuum furnace.

It is obvious that when the reaction is carried out in this manner, the scale of operation is severely limited. In particular, the process does not lend itself to the economic production of titanium on the commercial scale.

In a modification of the process described in F. I. A. T. Final Report No. 798, an attempt was made to overcome this limitation of batch size by reacting a molten layer of sodium, floating on a flux of sodium chloride and potassium chloride at 800 C., with titanium tetrachloride which was introduced into the reaction vessel below the level of the flux in order to avoid choking of the tetrachloride inlet pipe. In order to prevent contamination of the product the flux must be free from oxygen-containing compounds. The titanium product was then recovered by an aqueous leaching process. The metal, however, appeared to be inferior to that produced by the roll process, and there is no evidence to show that it could meet the present rigorous specifications for titanium.

Thus, so far as I am aware it has not been found possible until very recently to devise a commercially satisfactory method for the production of good quality titanium metal by reduction of titanium tetrachloride with sodium. One major obstacle appears to be the difficulty which is experienced in controlling the reaction conditions. The reaction is preferably carried out in the liquid phase so that the temperature must be high enough to maintain the salt that is formed as a byproduct in a molten state and yet not so high as to vaporize the sodium completely. Sodium boils at 880 C. and sodium chloride melts at 804 C. so that there is an operating margin of only about 80 C. The corresponding temperature margin in the Kroll process Where magnesium is used instead of sodium is about 380 C. Moreover, the heat 2,864,691 Patented Dec. 16, 1958 evolved when sodium and titanium tetrachloride react is more than 50% greater than when magnesium is used as the reducing agent. For these reasons Kroll has already given it as his opinion that it would be very difiicult to use sodium alone to bring about the reduction (see Metal Industry 1952, vol. 80, p. 383). He suggests that either an oxygen-free flux must be added or one must use a combination of sodium with another reducing agent such as magnesium to diminish the melting point of the salt that is formed and so increase the gap between the fusion point of the reaction mass and the boiling point of sodium. Kroll also states that the volatility of sodium is a very grave draw-back since vaporisation of some sodium is unavoidable and this gives rise to some vapour phase reaction which forms products such as titanium sub-halides which are pyrophoric.

The fundamental difficulty is therefore to maintain a commercially useful rate of producing titanium metal while avoiding an uncontrolled and potentially dangerous rise in the temperature of the reaction mass.

In view of the statements quoted above, I have been surprised to find that this difiiculty can be overcome in a very simple and effective manner by utilising that property of sodium which has hereto been considered an almost insuperable drawback to the successful exploitation of the reduction of the tetrachloride with sodium on the commercial scale, namely, the volatility of sodium.

In my process the heat balance is controlled by arranging conditions so that part of the heat of reaction is dissipated by vaporising a proportion of the molten sodium, the vapour being subsequently condensed and the liquid metal returned to the reaction zone. Any sudden surge of heat can thus be readily absorbed if there is provided some reserve of cooling capacity to cause refluxing of sodium. This condition I have found can be achieved in a number of ways one of which is to cool by external means the wall of the reaction vessel above the surface of the reaction mixture. This may be conveniently done by employing water-cooled coil welded to the outside of the reaction vessel below the lid. Perhaps the simplest method however, which is surprisingly effective, is the provision of a relatively large cooling surface on the upper part of the reaction vessel, remote from the reaction zone. Conveniently, the lid of the vessel provides the requisite cooling surface and this also operates to return the condensed sodium rapidly to the reaction mass, minimizing any tendency for sodium to collect and perhaps eventually solidify in a layer on the underside of the lid.

It is not sufiicient for my purpose however if in the course of refluxing the vaporised and condensed sodium is allowed to fall back to the reaction mixture with a fairly even spread so that the surface is uniformly cooled in the process.

I have found that the highest temperatures in the reaction vessel are attained at the centre of the surface of the reaction mixture and especially at the point where titanium tetrachloride is fed on to the fused salt. Here local reaction temperatures are likely to be as high as 1050 and even higher and although I make provision for the dissipation of the heat emanating from this point I am concerned to maintain the temperature at the sides of the vessel as low as is compatible with maintaining the reaction mixture in a liquid state. If the temperature on the inside of the wall is allowed to approach the high temperature at the centre there is a probability of the iron walls of the vessel being punctured in the process as there is a tendency for the titanium to form a eutectic with iron at high temperatures in excess of 1000" C. In addition action vessel.

the reaction vessel which will result in contamination of the titanium with iron. Both these tendencies must of necessity be avoided and for this reason my invention is aimed at so condensing the vaporized sodium that it runs down the sides of the vessel and cools the material in proximity with the walls.

In aprocess for the :manufacture of. titanium by. the re action of titanium tetrachloride with metallicsodium in an inert atmosphere and at a temperature z'xbovethemelting point of sodium chloride the present invention comrises Withdrawing from the reaction mixture a portion of heat generated in the reaction while maintaining a: substantially lower-temperature on the. surface of thereaction mixture in proximity with the side. of-thereaction vessel than on the surface of the reaction mixture in proximity I with the centre bynso controlling the rateof the reaction that a part of the sodium is refluxed during thereaction and is returned to the reaction mixture inproximity with the side of the reaction vessel.

I have found that thebest method of ensuring that the bulk of the vaporized sodium is returned near the side of the vessel is to employ a conical-shaped lid to the reaction vessel. By this means condensed sodium is directed down the inner side of the lidand thencedown the wall of the vessel. Thus coo-ling the surface of the meltnear thewall of the vessel.

It must be understood that by aconical-shapedlid I do not wish to restrict my invention to lids which are strictly -of the shape of a geometrical cone or dome but it is intended *to cover diverse geometrical shapes of lids which have conical or dome features always provided that the sides subtend an angle to the horizontal.

I prefer however to employ a conical-shaped lid which subtends an angle to the horizontal in excess of and operating with this angle ensures almost complete canalising of the condensed sodium down the sides of the re- For example, the angle to the horizontal may be in excess of The attached drawing illustrates one arrangement of the reactor and its accessories with which the process of the invention may conveniently be carried out. The reactor 1, placed in a furnace 9, is initially charged with molten sodium to the level 3. Titanium tetrachloride is fed into the reactor through the inlet tube 4 while the inert atmosphere, preferably argon gas is introduced=through the tube 5 before charging the reactor with sodium. Molten sodium is fed through the inlet tube 2. The tube6 is a vent which enables excess gas to be released from-the reaction vessel as required. The temperature of the reaction mixture is measured by the thermocouple It The lid of the reaction vessel is conical in shape and is sealed to the reaction vessel at point 7, the side of the lid subtending an angle of to the horizontal. As the reaction between the sodium and titanium proceeds much'of the vaporised sodium condenses on the inner side of the lid 8 and flows from there down the perpendicular wall of the reaction vessel to the surface of the reaction mixture. In

'the process a considerable quantity of heat'absorbed by the lid is dissipated to the external atmosphere and the relatively cool condensed sodium in flowing down'the side of the reaction vessel prevents any overheating of the vessel walls.

I have mentioned that the lid itself provides the cooling surface for a condensation and return of vaporised sodium but my invention is not limited to cooling based only on the remoteness of the lid from the reaction mixture. "It must be understood therefore that cooling by the'lid may be reinforced by water cooling or air cooling the lid during the reaction. Heat transfer fluids, liquid metals or oilmay also be used to render the cooling of the lid more effective.

'In using the process of the invention I have found no One part by weight of sodium was introduced into a vessel provided with a conical-shaped lid, the air in the vessel having previously been displaced by argon. The

vessel was heated up to a temperature of about 830 C. and an atmosphere of argon was maintained in it during the whole of the reaction. 2.1 parts by weight of titanium tetrachloride were then gradually introduced in the liquid form. Reaction occurred with the evolution of much heat and as a result of this, partof the sodium was vaporised, condensed on the relatively cool lid of the vessel and flowed back into the body of the vessel down the vessel walls. The rate of addition of the titanium tetrachloride was so adjusted that the temperature of the conical lid did not exceed 650 C. During the addition of the tetrachloride the lid was 'cooledby means of a blast of air which was directed on it from a fan. When the reaction subsided additional heat was supplied to keep the reaction mass at a temperature a little in excess of 850 and after completion of the reaction the product was held at this temperature for a further half hour to allow for sintering of the product in accordance with the procedure described in British specificationNo. 720,517.

On completion of the reaction the titanium salt mixture was worked up by aqueous leaching using nitric acid as described in copending U. S. application Serial No. 461,688, filed October 23, 1953, in the name of Ferguson, Howell and myself, to give a pure titanium powder which after arc melting was found to have a hardness of approximately DPN.

.What I claim is:

1. In a process for the manufacture of titanium by heating a reaction mixture of titanium tetrachloride and metallic sodium in an inert atmosphere Within a reaction vessel having side walls and at a temperature above the melting point of sodium chloride, the improvement which comprises withdrawing from the reaction mixture av portion of the heat generated in the reaction while maintaining a substantially lower temperature on the surfaceof the reaction mixture in proximity with the side walls of the reaction vessel than on the surface of the reaction mixture in proximity with the centre, said improvement comprising the steps of maintaining the reaction mixture as a molten pool in said reaction vessel, vaporizing a portion of the sodium from said molten pool, condensing the vaporized sodium on a cooling surfacepositioned above said reactionmixture and disposed-at an angle to the horizontal inv excess of 30 and'flowing substantially all of the condensed sodium along said surface and downsaid side walls whereby substantially all of the condensed sodium is returned to the reaction mixture in proximity to the side walls of said reaction vessel.

2. The improvement according to claim 1 in which the reaction is carried out in the presence of a conicalshaped lid to the reaction vessel.

3. The improvement according to claim 2 in which the lid of the reaction vessel subtends an angle in excess of 45 to the horizontal.

4. A process for manufacturing titanium which comprises providing a molten pool of sodium in an inert atmosphere within a reaction vessel having side walls, maintaining said pool at a temperature sufficient to react with titanium tetrachloride to form titanium metal, introducing titanium tetrachloride into said reaction vessel on the surface of said molten pool of sodium and centrally thereof thereby forming a reaction mixture containing titanium metal and generating heat, vaporizing a portion of the sodium in said reaction mixture by. said generated heat, condensing the thus vaporized sodium on a surface positioned above said reaction mixture and at an angle to the horizontal in excess of 30, and returning' substantially all of the condensed sodium to said reaction mixture by flowing said condensed sodium downwardly along said surface and side walls onto the surface of said reaction mixture in proximity to said side walls thereby maintaining the temperature of the surface of the reaction mixture in proximity to said side walls at a substantially lower temperature than the temperature of the surface of the reaction mixture at the point where said titanium tetrachloride is introduced.

5. The process of claim 4 wherein said reaction vessel includes a conically shaped lid, the vaporized sodium being condensed thereon and flowing downwardly therefrom along said side walls onto the surface of the reaction mixture in proximity to said side walls.

References Cited in the file of this patent UNITED STATES PATENTS FOREIGN PATENTS France Sept. 1, 1954 OTHER REFERENCES Journal of Metals, April 1950, pages 634-640.

Claims (1)

1. IN A PROCESS FOR THE MANUFACTURE OF TITANIUM BY HEATING A REACTION MIXTURE OF TITANIUM TETRACHLORIDE AND METALLIC SODIUM IN AN INERT ATMOSPHERE WITHIN A REACTION VESSEL HAVING SIDE WALLS AND AT A TEMPERATURE ABOVE THE MELTING POINT OF SODIUM CHLORIDE, THE IMPROVEMENT WHICH COMPRISES WITHDRAWING FROM THE REACTION MIXTURE A PORTION OF THE HEAT GENERATED IN THE REACTION WHILE MAINTAINING A SUBSTANTIALLY LOWER TEMPERATURE ON THE SURFACE OF THE REACTION MIXTURE IN PROXIMITY WITH THE SIDE WALLS OF THE REACTION VESSEL THAN ON THE SURFACE OF THE REACTION MIXTURE IN PROXIMITY WITH THE CENTRE, SAID IMPROVEMENT COMPRISING THE STEPS OF MAINTAINING THE REACTION MIXTURE AS A MOLTEN POOL IN SAID REACTION VESSEL, VAPORIZING A PORTION OF THE SODIUM FROM SAID MOLTEN POOL, CONDENSING THE VAPORIZED SODIUM ON A COOLING SURFACE POSITIONED ABOVE SAID REACTION MIXTURE AND DISPOSED AT AN ANGLE TO THE HORIZONTAL IN EXCESS OF 30* AND FLOWING SUBSTANTIALLY ALL OF THE CONDENSED SODIUM ALONG SAID SURFACE AND DOWN SAID SIDE WALLS WHEREBY SUBSTANTIALLY ALL OF THE CONDENSED SODIUM IS RETURNED TO THE REACTION MIXTURE IN PROXIMITY TO THE SIDE WALLS OF SAID REACTION VESSEL.

Images