US2758021A - Process of preparing metal powders by a fluo-solid reduction process - Google Patents

Description

United States Patent TROCESS OF PREPARING MET-AL POWDERS BY O SOLID REDUCTION PROCESS JosefihlEJDrapeau, JrQ,'.Calumet City, Ill.,=and Richard-I. THalsted, Hammond, 'Ind., assignors to The Glidden :CompanyfCleveland, Ohio, a corporation of Ohio :Nm-Drawing. Application Augustll, 1948, Serial No. 43,754

'SClaims. (Cl; 75-.5)

EFhisinventionrrelates I to. ail-improved process for pro ducing metal .powders by-chemical reduction'of comtpoundssof-ithclmetal. -It relates particularlyto improvements in the fiuo-soli'd type -of reduction process wherein a massofi the finelyadivided solidmetal compound ismaintainedin a turbulent or fluid-condition and is simultanet'ously reduced-byi'means ofa current of reducing gas-flowing through'the mass.

'Fluo-soli'd process, so named because the fine solids .are mainta'inedin a fluent,'fluid or fluidized state have "been employed in the past for a variety of chemical reactions where the particular-merits of this type ofreaction oystemhavebeenrecognized. The United States Patent No. 1 ,810,055, issued 'June 16, 193/1, to Georg Muller, describes'the basic features of a duo-solid processand indicates that in such process the finely-divided 'solid reactant-is sustained at a relatively fixed level in the apparatus by the upwardly flowing current ofgas which simultaneously maintains it in a highly turbulent gasborne condition. Under such turbulent'or fluid-condit-ion, reactions whichdepend on'rnutual-surface contacts between thepowder particles and the current of gas occur about as "easily and rapidly as 'liquid phase reactions. Moreover, the conditions promote excellent heat-transfer and-provide low concentrations of gaseous or vaporizable products of the chemical'reaction. Accordingly, the'fluosolid *type-ofprocess is readily capableof bringing about reactions -wl1ichif carried out in a static condition would proceed slowly, if 'at'all, and'with concurrent difficulties due *to localized overheating, incomplete reactions .;in localized areas, andinefiec'tive removal of reaction-depressin'gconcentrations of products of reaction.

Attempts have .been made to utilize the advantages of the 'fluo-solid process in reducing metal compounds to metal powders; but as far asvwe are aware, such attempts "have "heretofore been unsuccessful mainly because the highly turbulent orfiuid condition of the charge is gradua'llylost'as thereduction approaches completion. We have found in our experiments with the process that for some reason .or.reasons not yetrunderstood, the mixture of reduced solid-state metal particles and unreduced solidstatemetal.compoundparticles begins to exhibit the characteristics of a.sticky, plastic mass after a major .part of the metal compoundhas been reduced. The mass becomes sluggish, it adheres to the vertical side walls of the reaction vessel and the current dfgasforms channels through the mass whereupon all turbulence ceases. When an iron oxide or a copper oxide is being reduced with hydrogen, this lossof fluidity occurs when about 95% of the oxide has been reduced to metal, and it occurs even at temperatures .as. low as .750 .F. for iron oxides and260 'F. for copper oxides. The reduction can'be otherwise carried out very efficiently at these low'tem- "peratures up 'tothe time the loss offiuidity occurs, but heretofore could 'not be'carried on economically'beyond that stage. Wehave found, however, "that by *suita'bly "treating-the reduction mass prior to the time'the "loss-of fluidity occurs, in manners hereinafter pointed out, the

ffPatented Aug. 7, 1956 2 mass can be maintained ina freely turbulent and-fluidized condition throughout the entire reduction, and the reduc- :tion can be carried on as'long as desirable or necessary to eifect the complete reduction of the metal compounds, or where mi xturesof compounds'are-involved, to effect the .completereduction of-one or more of the compounds and to partially or completely reduce another of the compounds.

Accordingly, -it is an-object of this invention to eliminate the loss of fluidity which occurs in fluo-solidreductions of.gas=bor-ne, reducible-metal compounds to solidstate metal powders.

it is a further object to effect the completereduction of metal compoundsto metal'p'owder by means ,ofthe duo-solid process.

It is an additional object to produce metal powders of maximum purity by completely reducing compounds of the desired metals under duo-solid conditions "of gaseous reduction;

A further-objectis'to produce alloypowders or mixed or .coa-tedpow'ders by flue-solid reduction "methods.

These and other objects "wil'lbe apparentfrom the'following description-oftheinvention.

Early in thecourse'of our experimentation with 'iiuosolid-reductions of metal compounds'to solid-state metal powder'by means of reducing gases such as hydrogen, carbon monoxide, methane, city gas, etc., and mixtures thereof -with-each other "and/ or'with inert diluents, we

encountered the troublesome problem outlined a'bove,.and

'in seeking ways to overcome the loss of fluidity of the mass,- wefound that by introducing and maintaining small amounts ofiinely powdered relatively-inert refractory material such as bone a sh, 'aluniina'liounsilica flour, graphite andmany "others, the fluidity oftne reduction charge could be readily maintained throughout .the entire reduction. The 'amount'of such refractoryjpowder which. is needed constitutes a minorproportion of the reduction mass, and isgeneral'ly'less than about 15% of'the mass. At'least about "1"% is generally required for the purpose butthe exact minimumquantity which must be used in aparticu- 'lar reduction can best be determined'by experiment since it is not a quantity which is independent of theprecise conditions under which the reduction is carried out. For example, an amount less than 1%by weight of the reduction mass'mig'ht .be sufiicient to prevent the loss of fluidity wheniallof the refractory can be retainedvin the mass, but since a. current of reducing gas is constantly flowing through the mass, 2. variable proportion of the refractory material fails to adhere to the particles .of the reductionmass and is carried out of the mass byvthe current of reducing gas and is ultimatelycarried out of the reductionflcolumn. Such lossesmust, of course, be compensated by introducing corresponding amounts to the fluo solid mass, so as to maintain at'least the minimum amount needed'for thepurpose. The amountwill also depend on the shape and s'izeof theparticles of refractory and on the size and shape .of the unreduced metal compound and/or of the reduced metal powder. Thus, a given weight of refractory in the form .of. platelike particles is apt to .bemore eifective .thana corresponding weight of angular or spherical particles whose ratio of surface to mass'isless than thatof .the-;.plate-like particles. It will-also be readily apparent thatthe size of the refractory particles is alsoa factonsincethe. smaller the particles are thegreateristhe surfacewhichagiven weight-thereof willcover. Ideally, a monomoleculanfilm 'of'the refractory material on eac'lrpaiticle of .-the reduc- 'tion mass'would be about all that lisneeded, and on astatistical'ba'sis only a major fractional part of the amount required to form "a monomolecular layer might'we'll .be used. Under practical conditionsdfoperatiomhowevet, such minute quantities can seldom 'be distributed uniformly enough over the surfaces of the particles of reduction mass to be effective. It is, however, within the ability of one skilled in the art, when guided by the foregoing principles, to readily determine the amount which, under his conditions of operation, is effective for the intended purpose of so coating or neutralizing the particles that loss of fluidity is avoided on a practical basis.

It will be understood that when pure metal powders are to be produced, the refractory material should have chemical characteristics which make it inert with respect to the reducing gas, the metal compound and reduced metal particles at the highest temperature attained during the reduction. That is, it should not react with the reducing gas so as either to be altered by the gas or to alter the gas harmfully, and it should not react with or flux the metal compound so as to alter the metal compound or to be altered by it, and it should not be reduced or dissolved by the metal particles which are liberated by the reduction. Furthermore, the refractory coating material should have physical characteristics which permit it to remain in the solid state at the highest temperature attained during the reduction. Many refractory materials meet these requirements when the reduction temperatures are below about 1800 F., as for example, bone ash, graphite, calcium carbonate, fiuorspar, sodium fluoride, lime, magnesia, silica, talc, thoria, titania, beryllia, fire-clays, fullers earth, magnesite, sillimanite, steatite, etc. Some of these materials are apt to be unsuitable at higher temperatures, or when in the presence of metal compounds which flux or react with the refractory materials. Graphite, of course, should not be used when iron-group metals are being reduced at temperatures which permit them to be carburized appreciably by free carbon. Magnesia is capable of being reduced appreciably by carbon monoxide at elevated temperatures and where such reduction would be harmful, magnesia should be avoided. Likewise, if the decomposition of carbonates, such as that of calcium carbonate to calcium oxide and carbon dioxide, produces harmful concentrations of carbon dioxide, these carbonates should be precalcined or avoided. Persons skilled in the art will recognize these and other precautions which may be advisable in order to avoid side-reactions which may impede the satisfactory carrying out of certain reductions to produce pure powders. As pointed out above, however, the use of the fiuosolid process has the advantage of bringing about reductions at lower temperatures than are usually required for static reductions of the same compounds by the same reducing gases, so that the use of the process aids in avoiding the extremes of temperature which may require F the exercise of such precautionary measures. Thus we have found that copper oxide may be completely and rapidly reduced to metallic copper by means of hydrogen at temperatures as low as 260 F. and iron oxide (FeaOr) may be completely and rapidly reduced by hydrogen to iron powder at temperatures as low as 770 F. Numerous other oxide metal compounds and metal salts can likewise be reduced at unexpectedly low temperatures. Accordingly, at least the more common metals such as iron, copper and nickel can be produced in powder form by using the invention, without requiring any great attention to the suppression of side reactions or decompositions.

Fluo-solid reductions of mixtures of metal compounds, to produce alloy or coated powders, may be carried out in accordance with the principles of this invention. In effecting such reductions, as for example the reduction of a mixture of copper oxides and zinc oxide to make brass powders, or a mixture of copper oxides and tin oxide to make bronze powders, or a mixture of oxidic iron compounds and tin oxide to make tin-coated iron powder, or a mixture of oxidic iron compounds and zinc oxide to make zinc-coated iron powders, a fluidized mass of the mixed oxides may be reduced to the stage at which loss of fluidity begins to become apparent, and inert refractory material may then be introduced after which the reduction of the mixed oxides may be continued to completion. However, in the case of mixed oxides such as those mentioned above, it is also possible to introduce excess zinc oxide or tin oxide as a refractory in place of an inert refractory either in the original mixture or at the time that loss of fluidity begins to occur. When one of the reducible oxides is thus used as the refractory material, enough must be added to take care of such amount as will be reduced to metal and to provide a sufiicient excess to avoid loss of fluidity. Moreover, in the instances mentioned above where copper oxides and zinc oxide or tin oxide are being reduced to form brass or bronze powder, if excess zinc oxide or tin oxide is used as the refractory, then the reduction should be terminated when all of the copper has been reduced and when enough zinc oxide or tin oxide has also been reduced to yield the desired alloy. At this stage, however, enough unreduced zinc oxide or tin oxide should remain in the mass to function as a fiuidizing refractory. Other mixtures of oxides may be reduced similarly to produce alloys, one determining feature of such mixtures being that one of the oxides in the mixture is reduced or reducible at a selected temperature at a slower rate than the other oxides(s) of the mixture. When such differential in reduction rates prevails, the more easily reduced oxides(s) can be completely reduced while a part of the slowest-reducing oxide remains as such to function as a fiuidizing refractory. Thus, in any such reduction, it is essential that a part of the slowest-reducing oxide or metal compound remain unreduced at the conclusion of the reduction, and such unreduced material must have the other attributes of a refractory material which enable it to function as a preserver of fluidity in the fluidized reduction mass. In other mixtures of compounds, the compound which functions as a fiuidizing refractory may not be reduced at all at the temperature used to reduce the primary metal vcompound, but after the primary compound has been reduced, the temperature may be raised enough to effect the partial reduction of the secondary or refractory compound.

Maintenance of fluidity throughout the entire course of the reduction is an important economic factor in making the flue-solid reduction of metals commercially practicable. When the fluo-solid process is carried out without the present invention, the reduction charge, even it brought to a completely reduced state, must be mechanically broken loose from the walls of the reaction column in order to remove it from the column. Furthermore, the reduced mass is then in the form of lumps which are difficult to remove without substantially dismantling the column. On the contrary, when fluidity is maintained by using the present invention, the reduced charge is in a free-flowing condition and by simply turning off the current of reducing gas, the charge drops freely to the bottom of the column and may be drained therefrom through a small opening such as one would drain water from a vertical tube. Thus, the reduced charge is easily removed from the column and the column walls do not need to be cleaned off after each reduction. Accordingly, the column does not need to be dismantled, and instead is ready for another reduction charge as soon as the previous one has been drained out. Consequently, the column can be utilized very efliciently on a time basis where batch operations are carried on, and in factthe time lost between batches becomes so small that the utilization of the column closely approaches the efiiciency of a continuous process. While some of these advantages have been pointed out in the Muller patent as being de sirable characteristics of the fluo-solid process, they have not heretofore been attained in metal powder reductions because of the loss of fluidity which occurs in such reductions. Through our present invention, such reductions can now be carried on with the attainment of high mechanical operating efiiciencies.

In operating a fluo-solid reduction column in accordance with our invention, any reducible metal compound or mixture of reducible compounds which can'be reduced to solid-state metal by means of gaseous reducing agents may be treated. The compounds may be oxidic metal compounds such as the oxides, hydroxides, carbonates, basic carbonates, etc., or metal salts such as chlorides, sulfates, nitrates, acetates, oxalates, etc. When reducing any of such compounds, or others, the refractory material may be mixed with the reduction charge at any time prior to the stage at which loss of fluidity occurs. We generally prefer, however, to introduce refractory material at the time the charge is introduced. Then as soon as the. fluidizing gas is turned on and the charge has been fluidized, the refractory material becomes intimately mixed with the charge and is thereafter. continually kept mixed with it as the reduction proceeds. However, it is equally practicable to carry on the reduction until the onset of loss of fluidity is imminent, and then to introduce the refractory material into the mass directly or into the stream of fluidizing reducing gas which carries it into the charge and mixes it with the charge.

When the reduction of a charge has been completed, instead of cooling the reduced metal particlesto room temperature in the column, we may drain out or otherwise remove the reduced chargefrom' the column as soon as the reduction has been completed, and transfer it to an inert liquid cooling bath or to a separate cooling chamber provided with a suitable protective atmosphere. The'transfer should, of course, be effected in a protective atmosphere or medium also, so that the hot metal particles are protected against oxidation or other chemical alteration while being transferred and cooled. In this way the column is freed promptly for use in treating the next batch of charge. Moreover, since thereduced metal remains in afree-flowing condition,.it needs little, if any, milling to break up lumps or agglomerated particles, and may be sieved directly to desired particle-size brackets.

The fluidizing reducing gas employed in the process may, of course, be introduced as a pure gas; e. g. hydrogen, carbon monoxide, methane, propane, acetylene, other gaseous or volatile hydrocarbons, alcohols, ketones, etc.; or as diluted pure reducing gases, or diluted mixtures of reducing gas, the diluent being preferably a cheap inert gas. Nitrogen is a suitable diluent for many reductions, while carbon dioxide may be suitable for. those carried on at low temperatures. The use of. diluted re ducing gases or diluted mixtures makes it possible to moderate the evolution of heat which occurs in exothermic reductions. By varying the proportions of diluent to reducing components, the reduction rate may be. suitably controlled so that excessive heating of the reduction mass is avoided, and so that the temperature at which the reaction proceeds may be wholly controllable by an operator through the application of external heat. When carbon is used as all or part of the refractory material, it may be deposited in the charge in the column by effecting the in-situ decomposition of carbonaceous materials, such as grain flours, methane, acetylene, propane, etc. Part of such deposited carbon, being very finelydivided, may be blown from the column, but by separating it from the exit gases and recycling it through the column, it can be made effective for fluodizing purposes. After the reduced charge has been cooled to about room temperature, it may, if desired, be treated in any of a number of ways to separate the refractory powder from it. Winnowing or other types of air separation, a leaching operation when the refractory is water or solvent soluble, electrostatic and/ or magnetic separation, gravity separation, or liquid displacement methods, and combinations of these with each other or with numerous other well known separation methods may be used. For some intended uses of the metal powder, the refractory material may not be harmful and need not be removed.

Example 1 Three thousand. grams of cuprous oxide having. a: sieve analysis as follows:

Percent Plus 40 mesh trace Plus mesh. 5 Plus 325 mesh 24 Minus325 mesh- 71 was introduced into a vertical hard glass tube 4 in diamet'er and 6 feet high. The upper end of the tube was plugged except for a half-inch diameter gas outlet. The bottom end was also plugged except for a central gas inlet orifice. A supply of hydrogen under pressure was connected to the inlet orifice. Upon turning on the hydrogen and'adjusting the pressure thereof, the mass of copper oxide gradually became turbulent and upon further adjustment of pressure the turblent mass was elevated to about the middle of the tube, where it thereafter remained, always in a highly turbulent or fluent state. Lighted gas burners were applied to the outside surfaces of the glass tube around the middle section or slightly lower, and a thermocouple was lowered into the turbulent mass from the gas outlet. After a few minutes the temperature of the turbulent mass reached about 250 F. A cold mirror was then placed over the gas outlet so that the exit gases would impinge on it and as soon as a film of moisture was deposited on the mirror, the gas burners were turned very low. The moisture film indicated' that reduction has commenced, and since the reduction reactions are highly exothermic, it was advisable to reduce or stop the application of external heat. The temperature of the'turbulent mass rose fairly rapidly for a few minutes,- but soon settled down to temperatures slightly above 300 F. The gas burners were then adjusted to low levels so that the turbulent mass thereafter remained around 350 F. After an' elapsed period of about one hour, the reduction had proceeded far enough to cause the turbulent mass of reduced copper and unreduced oxide to show indications of stickiness, sluggishness orloss of turbulence. Thereupon, 300 grams of minus 325 mesh bone ash was introduced through the top end of the tube without discontinuing the flow of hydrogen to or from the column. Shortly after the bone ash had been added, the sluggishness of the turbulent mass disappeared and the reduction was continued for about 30 minutes more. At the end of that time, no moisture was condensed on a cool mirror held near the gas outlet showing that reduction was complete. The temperature of the mass was still around 350 F. The gas burners were turned off, and fans were turned against the. tube to promote cooling of the reduced copper powder. The hydrogen was continued during the cooling period to also assist in abstracting heat from the powder. After a few minutes the mass had cooled to room temperature, the hydrogen was thereupon turned off, the bottom plug was removed, and the copper powder was drained from the column into a container and was promptly sealed therein.

Subsequently, the copperpowder with its admixed quantity of bone ash was winnowed in a current of air which blew the bone ash away from the copper. After being .thus purified, chemical analysis of the resulting copper powder showed it to be composed of:

Percent Copper 99.6 Home ash .2

The separated bone ash was retained for subsequent reuse. During the reduction as described above, about 50 cubic feet of hydrogen were consumed.

Example 2 We charged 3,000 grams of ground mill scale showing an iron content of 73.92%. This scale had an apparent density of 2.16 and a flow of 46 seconds. The screen analysis of the milled iron scale prepared for this run was as follows:

+40 mesh +80 mesh +100 mesh +150 mesh +200 mesh +325 mesh -325 mesh l- HLII This milled iron scale was charged into a vertical hard glass tube 4 in diameter and 6 feet high. The upper end of the tube was plugged, except for a /2" diameter gas outlet. The bottom of the tube was also plugged, except for a central gas inlet orifice and a A" glass tube which was used to discharge the sponge iron powder at the end of the reduction experiment. This glass tube was sealed by means of a rubber tube and a clamp. A supply of hydrogen under pressure was connected to the gas inlet orifice and, upon turning on the hydrogen and adjusting the pressure thereof, the mass of iron oxide gradually became turbulent. Upon further adjustment of pressure, the turbulent mass was elevated to about the middle of the tube, where thereafter it remained, always in a highly turbulent or fluent state. Lighted gas burners were applied to the outside of the glass tube around the middle section or slightly lower, and a thermocouple was lowered into the turbulent mass from the gas outlet.

During the raising of the temperature to around 250 F., it was observed that the fluidity characteristics of the iron scale were not satisfactory. However, a fluent state developed immediately after all the moisture that was apparently present in the original iron scale was driven off. The film of moisture on the iron scale appeared to change the surface characteristics and reduced the ability of the iron scale to fluidize freely.

During the raising of the iron scale up to 600 F., carbon dioxide was used as a fluidizing gas. At 700 F. the carbon dioxide was replaced by hydrogen gas. At 770 F., moisture was detected by placing a cold mirror over the gas outlet. The temperature was raised to and held between 900 and 1000 F. for a period of two hours. During this time we had introduced about 100 cubic feet of hydrogen. A small sample was removed from the column by opening up the discharge tube at the bottom of the column. A few grams of iron powder were removed from the column to a glass flask and sealed therein. On cooling, the iron was determined by chemical analysis and found to be 82.8%. The hydrogen continued to flow through the mass of material for six hours. During this six-hour period, we consumed 120 cubic feet of hydrogen.

A sample was removed from the fluidized bed, was collected in an Erlenmeyer flask under a reducing atmosphere, and cooled to room temperature. This sample of iron powder showed an iron content of 92.85%. The temperature of the fluidized bed was raised from l000 to 1100 F. during the last hour of the experiment. During the last hours treatment, we added to the 3,000 grams of iron scale 150 grains of bone ash. As soon as the bone ash was added to the fluidized bed, the bumpy, sluggish condition of the mass disappeared and we obtained a freely fluidized bed. During this period we passed 38 cubic feet of hydrogen through the unit. The heat was removed from the furnace and the charge was allowed to cool in a fluidized bed to room temperature. The hydrogen was then turned off and the iron powder was discharged from the furnace.

The iron powder with its admixed quantity of bone ash was winnowed in a current of air which blew the bone ash away from the iron powder. After being thus purified, the physical and chemical analysis of the iron powder showed:

Iron percent 97.55 Bone ash do 0.12 Apparent density d0 1.82 Flow "seconds" 57 Screen analysis:

+40 mesh percent .1 +80 mesh do 57.4 mesh d0 19.2 mesh do 16.2 +200 mesh do 5.6 +325 mesh do 1.3 325 mesh do .2

Example 3 Three thousand grams of cuprous oxide having a sieve analysis as in Example 1 were introduced into the glass tube described in Example 1, and were brought to a fluidized condition with hydrogen. The mass of oxide was then reduced as described in Example 1., except that 300 grams of zinc oxide were introduced in place of the bone ash of that example. Reduction of the cuprous oxide was completed after 95 minutes at a temperature of 350 F., the end of the reduction being determined by the absence of a moisture film on a cool mirror placed over the discharge outlet. The temperature of the fluidized mass was then raised to about 1000 F, at which temperature a moisture film formed on the cool mirror, showing that reduction of the zinc oxide had commenced. This exothermic reduction reaction heated the mass somewhat and by reducing the amount of heat applied to the outside of the tube, the temperature of the mass was kept between 1000 F. and 1200 F. for two hours. The gas burners were then turned off, and the reduced mass was allowed to cool while being maintained in fluodized condition by the continued flow of hydrogen. When the mass had cooled to about room temperature, the hydrogen was turned off, and the reduced mass was drained from the tube into a container and was promptly sealed therein. Subsequently, the reduced mass was winnowed in a current of air which blew the unreduced zinc oxide away from the metallic powder. Subsequent analysis of the cleaned metal powder showed that the powder was of a brass composition of 94.5% copper and 5% zinc. The powder had a hydrogen loss of 31% showing it to be substantially free of unreduced oxides.

Example 4 Three thousand grams of ground mill scale such as used in Example 2 were mixed with 300 grams of tin oxide in the fluodized column described in Example 1. The mill scale was then reduced by means of hydrogen under fluodized conditions at temperatures between about 1000 F. and 1200 F., all as described in Example 2. After the mill scale had been reduced to iron, the temperature of the fluidized mass was raised to between about 1200" F. and 1400 F. and partial reduction of the tin oxide was accomplished under fluidized conditions. Heating was then discontinued and the fluid charge was cooled to room temperature by continuing the flow of hydrogen. The hydrogen was then shut off and the charge was drained from the tube. After winnowing the product to free it of unreduced tin oxide, the resulting metallic powder was found to be tin-coated iron powder of high purity.

Example 5 Three hundred grams of zinc oxide and three thousand grams of ground mill scale were charged into the column described in Example 1. The mixed oxides were then fiuodized by means of a stream of hydrogen and then Were heated to and maintained between about 1000 F. and 1200 F. while fiuodized until all the mill scale and a part of the zinc oxide had been reduced. Heating was then discontinued and the charge was cooled in the column under hydrogen until cold. The charge was then drained from the column and was winnowed until most of the unreduced zinc oxide had been separated from the metallic powder. Subsequent analysis and examination of the powder revealed it to be zinc-coated iron powder.

In the following claims, the term refractory second metal compound is used to identify one or more solid metal compounds which remain solid at the flue-solid reduction temperature of the selected primary metal compound; which do not flux the primary metal compound being reduced or react chemically with it other than to assist in its reduction by maintaining gas-borne, turbulent conditions; which do not flux, adhere to or react chemically with the metal powder reduced from said primary metal compound at any of the temperatures employed in its reduction; and which is either inert with respect to the fluidizing reducing gas, or if reduced by said gas at the same or higher temperature is reduced more slowly than the selected primary metal compound under the existing reduction conditions.

Having disclosed our invention, what we claim is:

1. In a fluo-solid reduction process for preparing mixed metal powder wherein (1) a mass of finely-divided reducible, iron oxide is sought to be maintained in a fiuodized but substantially non-traversing condition with an elongated vertical reaction chamber by subjecting it to an upward flow of fluodizing gas capable of continuously maintaining the mass in a turbulent gas-borne condition and while so maintained is subjected to sutficient heat below the melting point of the metal reducible therefrom to bring about chemical reaction between the said iron oxide in solid state and gaseous reducing components of said fluodizing gas, whereby the iron of said iron oxide is liberated as powder in solid state, and wherein (2) said liberation of iron powder in solid state induces loss of turbulency in said mass before complete reduction of the iron oxide can be efiected, the improvement which comprises mixing about by weight on the iron oxide of a finely-divided added metal oxide selected from the group consisting of zinc oxide and tin oxide with the said gasborne turbulent mass of iron oxide, before the latter has been reduced to an extent at which loss of turbulency becomes imminent, and thereafter continuously maintaining in said gas-borne turbulent mass a minor amount not less than about 1% by weight on said iron oxide of said added metal oxide in an unreduced condition; thereafter carrying the fluo-solid reduction of the iron oxide to completion in the presence of at least said 1% of unreduced added metal oxide; terminating the reduction and separating the unreduced portions of said added metal oxide from the resulting solid state metal to recover mixed metal powder in a condition substantially free of unreduced added metal oxide.

2. A process as claimed in claim 1 wherein the added metal oxide is zinc oxide, and wherein the mixed metal powder resulting from the terminated reduction is Zinciron powder substantially free of iron oxide.

3. A process as claimed in claim 2 wherein the tluosolid reduction is carried out at temperatures between about 1000 F. and 1200 F. in hydrogen as the fluidizing and reducing gas, and wherein reduction is continued until the iron oxide has been reduced completely and the zinc oxide has been reduced partially to provide zinc-iron powder.

4. A process as claimed in claim 1 wherein the added metal oxide is tin oxide, and wherein the solid-state metal powder resulting from the reduction is tin-iron powder substantially free of iron oxide.

5. A process as claimed in claim 4 wherein the reduction of the iron oxide is carried out in hydrogen as the fluidizing and reducing gas at temperatures between about 1000 F. and 1200 R, where the tin oxide is thereafter reduced by hydrogen as the fiuidizing and reducing gas at temperatures between about 1200 F. and 1400 F., in the presence of the iron powder derived from said iron oxide, thereby to provide tin-iron powder, and wherein the reduction at the latter temperatures is terminated while at least said 1% of tin oxide remains unreduced.

References Cited in the file of this patent UNITED STATES PATENTS 1,196,049 Von Rauschenplat Aug. 29, 1916 1,810,055 Muller June 16, 1931 1,984,380 Odell Dec. 18, 1934 2,216,769 Drapeau, et a1 Oct. 8, 1940 2,254,976 Powell Sept. 2, 1941 2,316,664 Brassert et al Apr. 13, 1943 2,339,137 Berge Ian. 11, 1944 2,478,912 Garbo Aug. 16, 1949

Claims (1)

1. IN A FLUO-SOLID REDUCTION PROCESS FOR PREPARING MIXED METAL POWDER WHEREIN (1) A MASS OF FINELY-DIVIDED REDUCIBLE, IRON OXIDE IS SOUGHT TO BE MAINTAINED IN A FLUODIZED BUT SUBSTANTIALLY NON-TRAVERSING CONDITION WITH AN ELONGATED VERTICAL REACTION CHAMBER BY SUBJECTING IT TO AN UPWARD FLOW OF FLUODIZING GAS CAPABLE OF CONTINUOUSLY MAINTAINING THE MASS IN A TURBULENT GAS-BORNE CONDITION AND WHILE SO MAINTAINED IS SUBJECTED TO SUFFICIENT HEAT BELOW THE MELTING POINT OF THE METAL REDUCIBLE THEREFROM TO BRING ABOUT CHEMICAL REACTION BETWEEN THE SAID IRON OXIDE IN SOLID STATE AND GASEOUS REDUCING COMPONENTS OF SAID FLUODIZING GAS, WHEREBY THE IRON OF SAID IRON OXIDE IS LIBERATED AS POWDER IN SOLID STATE, AND WHEREIN (2) SAID LIBERATION OF IRON POWDER IN SOLID STATE INDUCES LOSS OF TURBULENCY IN SAID MASS BEFORE COMPLETE REDUCTION OF THE IRON OXIDE CAN BE EFFECTED, THE IMPROVEMENT WHICH COMPRISES MIXING ABOUT 10% BY WEIGHT ON THE IRON OXIDE OF A FINELY-DIVIDED ADDED METAL OXIDE SELECTED FROM THE GROUP CONSISTING OF ZINC OXIDE AND TIN OXIDE WITH THE SAID GASBORNE TURBULENT MASS OF IRON OXIDE, BEFORE THE LATTER HAS BEEN REDUCED TO AN EXTENT AT WHICH LOSS OF TURBULENCY BECOMES IMMINENT, AND THEREAFTER CONTINUOUSLY MAINTAINING IN SAID GAS-BORNE TURBULENT MASS A MINOR AMOUNT NOT LESS THAN ABOUT 1% BY WEIGHT ON SAID IRON OXIDE OF SAID ADDED METAL OXIDE IN AN UNREDUCED CONDITION; THEREAFTER CARRYING THE FLUO-SOLID REDUCTION OF THE IRON OXIDE TO COMPLETION IN THE PRESENCE OF AT LEAST SAID 1% OF UNREDUCED ADDED METAL OXIDE; TERMINATING THE REDUCTION AND SEPARATING THE UNREDUCED PORTIONS OF SAID ADDED METAL OXIDE FROM THE RESULTING SOLID STATE METAL TO RECOVER MIXED METAL POWDER IN A CONDITION SUBSTANTIALLY FREE OF UNREDUCED ADDED METAL OXIDE.