US2875038A - Method of producing crystalline metal - Google Patents

Description

Feb. 24, 1959 R. s. DEAN 2,875,038

METHOD OF PRODUCING CRYSTALLINE METAL Filed Feb. '16. 1955 3 Sheets-Sheet 1 Tid Feb. 24, 1959 R. s. DEAN 2,875,038

. METHOD OF PRODUCING CRYSTALLINE METAL Filed Feb. 16, 1955 3 Sheets-Sheet 2 Q B x I MBV/V/O/V PUT Feb. 24, 1959 R. s. DEAN METHOD OF PRODUCING CRYSTALLINE METAL 3 Sheets-Sheet 3 Filed Feb. 16, 1955 jog air/spas 7 METHOD OF PRODUCING CRYSTALLINE METAL Reginald S. Dean, University Park,,Md., assignor to Chicago Development Corporation, Riverdale, Md., a corporation of Delaware Application February 16,1955, Serial No. 488,540 '2 Claims. c1. 7544.5

This invention relates "toprocessesfor. producing titanium and zirconium. It is particularly'concerned with processes for forming andreducing the lower chlorides of. titanium andzirconium to. produce pure, massive or coarsely particulate, titanium or zirconiumina form which is not pyrophoric or attacked by water or dilute acid. It has particular but not exclusive application to processes which are in whole or: in part electrolytic.

The problem of producingtitanium or zirconium-in a massive or coarsely particulatearform bye-thereduction of-a solution of itslower chlorides infused baths ofxalka linous=chlorides or-bromides has hadwconsiderable attention from investigators in the metallurgicalfield? The problem has been partially-solved by severalzinventors but only by special means which are intrinsically unsatisfactory and uneconomical. The most successful. ofthese means necessarily involve-the reductionrof the tetrachloride of the metal. 1

I n-this categorymay be mentioned particularly J the process of British Patent No. 479,0 14 f January 28; 193'8. According to this patent, .m'olten'sodium" was zfloatedf on molten: alkali halide; and titanium tetrachloridewas added. The process could be said to encompass titanium chloride broadly, which is in. fact claimed inzthe patent. The product of'this process, whichwas commercially used by theDeutscheCloldu. Silber- Scheide Anstalt', was not cold ductile-a requirement for usefultitaniumi metal; The reason for the failure of this process was'recognized by Kroll and fully set for-thin German Patent No.'674,625 of 'April 18, 1939, namely thereduction of 'TiCll, with alkali metals; is: incomplete and the TiCl -iand liCl formed contaminate the product and cannot-be separated ,frorn it; Y

In my studies of the process I learned that-the difiiculties of alkali metal reduction as experienced by the Scheide Anstalt and by Kr;oll weredue, as stated by Kfroll- .to the incomplete-reduction t of titaniunr chlorides lkali metals orforthat-matterby alkaline earth metals mcluding magnesium unless.aspecial procedure such as that of Kroll is employed. 1 found that this incom- .let'e reduction was due toan equilibrium in which the alkali or. alkalineearth metal could be present in, solution up o, a. fairly high concentration inthe presencev ofititaniumqdichloride andrtitaniumunetal. This fi ndingis exp e d' br-t e r i in'Solutionf in used chlorides t. 'ricn+sis e ,v

The existence ofthis equilibrium with a considerable displacement to the. leftis the-scientific discovery which .I forins the baSiS'Of'; my present invention.

culties duewtothis equilibrium-by-,avoiding the formation Itwill be clearthatithe Kroll process avoids-the atent I tion. This leads to an equilibrium constant for Equation.

earth chloride.

By further'st'udy of the Reactionl above I'found this the equilibrium in this reaction was such that the-higher the concentration of TiCl the higher the concentration of sodium metal solution with which it wouldfbe' in equilibrium.

For the purposes of my invention the reasons for this unique situation were not of any importance. I have found, however, inmore recent studies of the processes ofmy invention that the activity of the alkalimetal solu,-' tion is a negative exponential function of its concentra- 1 of the form 8 'ricl cm' This explanation is given here to enable those skilled in the art to better understand my invention.

In its broadest embodiment my invention consists in bringing a very dilute solution of a lower chloride of titanium or zirconium in a fused alkali or alkaline earth chloride into contact with a very dilute solution of an alkali or alkaline earth metal in a fused alkali or alkaline I have found that the rate of difiusion of the alkali and alkaline earth metalsis much greater than the rate of diffusion of lower titanium and zirconium chlorides dissolved in fused alkali or alkalineearth chlorides. Several embodiments of my invention are therefore based I allow the concentration of the TiCl to attenuate to p a few hundredths of a percent titanium by .weight of the melt. I then add a dilute solution of lithium in the eut ec tic which can be produced by simple solution of the metal in the; eutectic. The lithium solution is of comparable concentration to the titanium chloride. The titanium solution is added at exactly the same rate, stoichiornetrically, as the lithium solution. I find that under these conditions there is formed at the outset pure particulate titanium at the point of additionof thelithium'solution. However, with continuedaddition of the lithium solution the more rapid difiusion of'the lithium causes the'reaction zone to move toward higher concentrations ofjthe titanium chloride which in'accordance with' Reaction 1 kali'or alkaline earth metal solution.

and Equation 3 brings about incomplete reduction oi the titanium'a situation which grows worse like achainreaction. Equation 3 is in fact a mathematicaFeXpres sion of the conditions for a chain-reaction- One embodiment of my invention, therefore, includes methods for preventinguncontrolled diffusion ofith'e al- Briefly these methods includei (i1) Procedures for reaction preferentially with the alkali metal solution; Q

b) Procedureswhich' reduce the rateof diffusion of the aIkaIi or alkaline earthmetal in the fused bath."-

'In the simplest instance the procedure of a above may be followediby adding chlorine at a pointwhere the difi'using alkali-nietal escapes from'the reaction'zone. This chlorine reacts preferentially with the dissolved alkali metal and with proper control need be added in only small amount as only. the dissolved alkali; metal escaping reaction with the titanium dichloride need be arrested. Of course the chlorine added represents: a

' degree of inefliciency in reduction, but this is a small 3 price to pay for obtaining crystalline form. A

While the reaction at the reaction zone is complete and thetitanium formcdis not contaminated. with TiCl or;;'I iCl it is sometimes inconvenient to isolate this reaction zone from the zone of TiCl concentration. In such instances, I may first cease addition of; the alkali or alkaline earth metal and the lower chloride of titanium and thenadd chlorine to convert all the TiCl to TiCl which can be leached from the particulate metal in only slightly acid solutions. Alternatively, I dissolve the salt containing TiCl in an acid solution containing an oxidizing agent, e. g. FeCl which prevents the formation of the insoluble decomposition products of TiCl In place the titanium in particulate of chlorine I may add TiCL, and TiCl which serves the same purpose but less adequately because in suc event some finely divided titanium is formed.

It'must be noted here that my invention is not stepwise reduction of titanium chlorides, which is old as noted in British Patent No. 479,014 andGerman Patent No. 574,625, previously referred to. Stepwise reduction does not accomplish my result except perhaps transiently. My invention consists in preventing the alkali or alkaline earth metal from diffusing beyond the reaction area and reacting with relatively large concentrations of TiCl My invention therefore requires a proper relationship between rate of addition of the reactants, in this instance titanium dichloride, alkali or alkaline earth metal and chlorine, TiCl or TiCl So far there have not been discussed procedures for adding the reactants other than simple addition which will be fully illustrated by examples. a a a Procedure b above encompasses methods of lowering the rate of diffusion of the alkali metal introduced into the fused halide bath. One method which I have found for accomplishing this is to select a reducing metal which is as little soluble as possible in the fused bath chosen. For example, I take a bath of KCl+NaCl and add a small amount of metallic barium which sinks to the bottom and dissolves only very slightly in this bath. I then add sodium to the barium and a chloride of titanium to the fused bath. Sodium being soluble in the barium diffuses slowly and meets the low concentration of diffusing titanium chloride a little above the barium layer. A foraminous barrier may be suitably placed above the barium layer so-that the particulate titanium is deposited on it. Magnesium may be used in place of the barium. Sincethe magnesium floats the geometrical distribution of the foraminous barrier must be different. This will be illustrated by examples.

Other combinations for accomplishing the same thing include: KCl+LiCl+MgCl LiCl-l-SrCl In the following examples it .will be understood that all procedures are carried out in. accordance with accepted practice in the titanium industry. The metal and the fused salt bath are at alltimes protected with an inert atmosphere usually pure argon. It will also be understood thatthe temperatures of operation, which are not particularly critical, may be adjusted to the melting point of the bath to the end that the latter be maintained molten. In describing the constituents of the bath in this application I have used the term alkalinous metal chlorides, which term is meant to include the chlorides as well as the bromides of mag-' arrangement of apparatus for retarding the rate of diffusion of reducer metal through a fused halide bath;

Figure 2 is a diagrammatic representation of a similar form of non-electrolytic apparatus, involving use of a horizontally disposed foraminous barrier member; and

Figure 3 is a diagrammatic representation of an electrolytic cell adapted for use in carrying out the process of the invention.

Exqmple l The reaction vessel for carrying out this example is illustrated in Figure 1. In this figure, 1 is a reaction vessel of steel provided with a cover 4, and inlet for argon 5, and an outlet for argon 6. The entire'pot is surrounded with a furnace to maintain a temperature of 850 C. An open-bottom bell 2 passes through the cover. This bell is formed of titanium. A multiplicity of titanium tubes 3 are spaced peripherally around the bell so that they introduce TiCl into the reaction zone beneathand adjacent the open bottom of bell 2.

. In this reaction vessel I react 46 lbs. of sodium with 188'lbs. of TiCl maintaining the temperature within the range of 400500 C. and so that the reaction proceeds at a moderate rate. This produces essentially TiCl and sodium chloride. The reaction mixture is protected with argon in the closed reaction vessel. When the reaction is complete I add 5 times the amount of NaCl, and melt the mixture at 850 C., out of access to air. I now introduce a saturated solution of sodium in molten NaCl into a restricted area in the vessel defined by said bell, and add sodium within this restricted area. The reaction zone is maintained just outside the restricted area and at such a point that the concentration of sodium in the molten NaCl beneath the bell reaches .8 10 gram equivalents per gram of fused salt. I add TiCl, just'outside the area, that is to say, to the salt bath adjacent the open bottom-of the bell. In this way any dissolved sodium which escapes the reaction zone forms TiCl is pure crystalline Ti and particles of Ti.

' At the completion of the operation I continue to pass in TiCL, until-the sodium is entirely reacted and the residual TiCl is converted to TiCl I then leach the salt from the Ti with very dilute acid.

" Example II I proceed as in Example I except that at the completion of the reaction I do not add excess TiCl but add magnesium to reduce the residual TiCl to Ti. I then wash with dilute acid to removethe salt.

' :Example IV -I proceed as in Example I except that at the completion of the reaction I do not add excess TiCL; but dissolve the salt in a slightly acidified solution of ferric chloride to prevent the precipitation of insoluble titanium compounds.

agglomerates of filiform Example V I proceed as in Example 11 except that I pass chlorine into the reaction zone instead of ZrCl v V 7 Example V! I proceed as in Example Iexcept that I use as the fused bath autectic mixture of KCl-I-NaCl. I then place a layer "of magnesium 'in the bell and add sodium to the magnesiumlayer and TiCl to-the periphery of the bath. I add these at a rate to bring the reaction zone tothe place indicated in Figure 1. Thisis established as a steady The titanium metal formed in the reaction zone state under conditions where the concentrationoi sodium reacliingthe reactionzone is'below gr'am equivalents- Inthisexample Iplacea layer of barium onthe bottom of: a; reaction-vessel and use a -bath of sodium and potassium chloride. I place over this=layer ofbarium a'foraminous barrier. I pass sodium '1'i1 to the barium layer.

ti'qn t a 83 9 equ v? I then ifdclhtie -TiGl-PIifil, in thempper portion of the ture to rdifiuse downward and. rneet.the,;upwardlyditiusing sodiumju'stabove the foraminous barrier. The=titanium collects on.the foraminous barrier which is conveniently made of 'Iji.. At the end of the 'reaotion l add anvexccss oi; TiCl and then drainotf the salt-.fromthe bottomfefthe vessel, leaving the titanium mixedwithf some salt :on the foraminous barrierj The titanium salt. in thisfmixtureds TiCl I remove the salt by washing ,with very dilute HQl.

Figure' 2 illustrates an apparatus for carrying out an embodiment of this inventions 'Ih this'figure, A represents a steel-pot-with a bottom outletiB-tconnected to asimilar pot (not shown) byapipeqQand meanslnot shown) for transfer of liquids between pots. The pot is provided vide an argon.atmosphereawithimthe'pot: Thepot'is filled with NaCl-KCl mixture and this mixture is fused. The bottom outlet B is frozen and a layer of barium Ba placed on the bottom. A foraminous titanium screen G is put in place above the barium layer, and a tube H arranged for feeding sodium into the barium layer. Two tubes] and K are provided through the cover for introducing sodium and T iCl respectively.

The TiClformed is caused to diffuse downward to meet the upwardly diffusing sodium. These reactants are caused to meet at low concentration just above the foraminous barrier G. The location of this reaction zone is regulated by the rate of addition of the sodium to the barium layer and the rate of addition of the TiCL; and Na in the upper part of the bath. When the proper steady state is established the TiCl and total sodium must be in stoichiometric proportions to form Ti.

The amount of barium is also a factor as the sodium alloyed must not cause it to float.

Example IX In this example I set up an electrolytic cell with an elecrolyte of fused sodium chloride.

Example X I proceed as in Example IX except that I use an impure titanium anode containing oxygen and iron and having a Brinell hardness of 250. The recovered titanium is fully refined and has a Brinell hardness of 80.

Example I proceed as in Example IX except that I add Cl 'near the cathode. As a result I increase the current density at which the necessary steady state of concentration is ob- I place the two electrodes at a distance of 1 foot and I add .1% by weight of TiCl to the bath.

a inas nd asses ap s, The resplt so far producing titauium the i i'i 1' ithe ath toconvertTiC 1n the bath toTrQl remove the cathode and dissolve the salt.

: Example XIlr a hi x uplsliu e x rur a rs Pu e i con n iica ho s" KCl; Na,C ;j3Q%f -.I i'pa'aa' rent until theopen. l v f volt lthen a j q t'au i ou isue elsettqlvs w ignf riqd s dd t cnsfifi e; pose anielectrolyticcell' rkethatof n F ure 1 T 1. flfltjB; a er this pot is'-a.liner E ;w W A qsfi asa hq e, fcru etit nium ,1. e t diicsmtall and immediately below h mmd m qds l mej ug wrx t lli e zt aw fr his t iiibt covered" from, thgali v, tai inss isqla M52 t eatme- ,by dissolv g in dilute I-lCl'con- Inrthi mehur d ce auififlhfimmucoutinuous non- P. 9! P. a t vtita iumi on ste y mal ing the. steels a cathode in an, electrolytic; cell h titaniuml anodexand ingtroly 549.76; KGLAQF/h bIaC nd.-:20%r ike t P SSrZ unid c ionalsurre cr hreush th cent-ch1 ampere per square inch until the open circuit voltage of the cell reaches 0.1 volt. I then interrupt the electrolysis, and regenerate the electrolyte by adding chlorine. This cycle is repeated until the desired thickness of plate is obtained.

of impure titanium, the electrolyte is fused MgCl at 850 C. with 5% added TiCl- The cathode is a rotating titanium disc. When a current of 1000 amperes/sq. ft. is'passed through this cell, an open circuit voltage of 0.9 volt is soon found. This clearly corresponds to deposition of magnesium with some titanium at the cathode. I now add C1 until the opencircuit voltage of the cell is less than 0.1 volt which indicates that there is substantially no codeposition of Mg with the Ti, and then proceed withthe electrolysis with the necessary additions of Cl It will be clear that under these circumstances the titanium chloride content, of the bath is increasing. To obviate this I add magnesium periodically. If the magnesium is added in exactly an amount equivalent to the C1 then the titanium chloride in the bath will remain constant and only the MgCl will increase. The magnesium must not be added in such amounts as to form titanium metal but only to reduce TiCl to Ticl This may be balanced against the withdrawal on the cathode. Since the cell of this example is a continuous arrangement, the withdrawn metal attached to the cathode contains MgCland TiCl These salts cannot be effectively removed by water washing so that I remove them by pressing and evaporation as set forth, for example, in my copending application Serial 391,153, filed November 9, 1953,. '-The final product is a pure titaniumingot with a hardness of Brinell 90.

{ Example, XV

added at regular intervals and only in such amounts as to reduce ZrCl to ZrCl and not. to form zirconium metal.

T'In this including the zirconium chlorides may be dissolved from ,the metal-.by.an acidQsolution of ferric chloride. I 5 Y i a I claim:

1. In a process for the production of a crystalline deposit of a metal of a first group consisting of titanium and zirconium by reduction of [a lower chloride of a metal. of saidfir'st group 'ijn solution in a fused bath of at least one salt selected from the group consisting ofalkali metal chlorides, alkaline earth metalchlorides and magnesium chloride by reaction'therewith of a reducer metalselected from the'group consisting or alkali 'metals, alkaline earth metals and magnesium 'dissolved-in a fused bath of at least one salt selected from the group consisting of alkali metal chlorides, alkaline earth metal chlorides and magnesium chloride, the steps which consist in bringing together said solution of the chloride of said first group ,metal and said solution of reducer metal in a reaction zone at a low concentration of both reactants below 10- gram equivalents per gramof said fused salt and thereafter adding said solutionsof reactantsto the reaction zone at rates which maintain such low concentration and in s'toichiometric proportions',.whereby said first group metal is produced in coarsely crystalline form. 1

2. In a process for. the production of a crystalline deposit of a metal of a first group consisting of titanium and zirconium by reduction of a chloride of a metal of saidfirst group by reacting 1) a solution of a lower chloride of said first group metal in-a fused bath of at least onesalt of the group consisting'of alkali'metal chlorides" and bromides-,'-'alkaline metal chlorides andbromides, magnesium chloride and magnesium bromide, with (2). a solution of'a reducer metal of the group consisting of alkali metals, alkaline earth metals and magnesium in a fused bath of at least one salt of the group consisting of alkali metal chlorides and bromides, alkaline earth metal chlorides and bromides, magnesium chloride and magnesium bromide, the step which consists in maintaining in the zone of the reaction a concentration of both reactants at below 10- gram equivalents per gram of said fused salt by adding said solution of lower :chloride of metal of the first group and said solution of reducer metal to the reaction zone at a rate which maintains such concentration. 1

References Cited in the file of this patent t UNITED STATES PATENTS Freudenberg Feb. 21, 1939 7 2,205,854 Kroll z June 25, 1940 2,561,862 Hill July 24, 1951 2,607,674 Winter Aug. 19, 1952 2,722,509 'Wainer Nov. 1', 1955 2,734,855 Buck et al Feb. 14, 1956 12,734,856 Schultz et a1, Feb. 14, 1956 2,783,196 R aney "Feb. 26, 1957 p f FOREIGN PATENTS 682,919 Great Britain ov. 19, 1952 OTHER REFERENCES Sibert et al.: Electrolytic Titanium, Journal of Metals, September 1956, pages 11621168.

Claims (1)

1. IN A PROCESS FOR THE PRODUCTION OF A CRYSTALLINE DEPOSIT OF A METAL OF A FIRST GROUP CONSISTING OF TITANIUM AND ZIRCONIUM BY REDUCTION OF A LOWER CHLORIDE OF A METAL OF SAID FIRST GROUP IN SOLUTION IN A FUSED BATH OF AT LEAST ONE SALT SELECTED FROM THE GROUP CONSISTING OF ALKALI METAL CHLORIDES, ALKALINE EARTH METAL CHLORIDES AND MAGNESIUM CHLORIDE BY REACTION THEREWITH OF A REDUCER METAL SELECTED FROM THE GROUP CONSISTING OF ALKALI METALS, ALKALINE EARTH METALS AND MAGNESIUM DISSOLVED IN A FUSED BATH OF AT LEAST ONE SALT SELECTED FROM THE GROUP CONSISTING OF ALKALI METAL CHLORIDES, ALKALINE EARTH METAL CHLORIDES AND MAGNESIUM CHLORIDE, THE STEPS WHICH CONSIST IN BRINGING TOGETHER SAID SOLUTION OF THE CHLORIDE OF SAID FIRST GROUP METAL AND SAID SOLUTION OF REDUCER METAL IN A REACTION ZONE AT A LOW CONCENTRATION OF BOTH REACTANTS BELOW 10-3 GRAM EQUIVALENTS PER GRAM OF SAID FUSED SALT AND THEREAFTER ADDING SAID SOLUTIONS OF REACTANTS TO THE REACTION ZONE AT RATES WHICH MAINTAIN SUCH LOW CONCENTRATION AND IN STOICHIOMETRIC PROPORTIONS WHEREBY SAID FIRST GROUP METAL IS PRODUCED IN COARSELY CRYSTALLINE FORM.

Images