US3583864A - Chemical process of producing an iron-copper alloy powder - Google Patents

Abstract

A PARTICULATE ALLOY OF IRON INTIMATELY INFILTRATED WITH FROM ABOUT 1 TO ABOUT 50% BY WEIGHT OF COPPER IS PRODUCED BY MIXING A REDUCIBLE COMPOUND OF IRON WITH AN APPROPRIATE PROPORTION OF COPPER COMPOUND SELECTED FROM THE GROUP CONSISTING OF ELEMENTAL COPPER AND REDUCIBLE COMPOUNDS OF COPPER, THE MIXTURE BEING HEATED UNDER REDUCING CONDITIONS AT A TEMPERATURE BETWEEN ABOUT 1010 AND 1150*C. UNTIL THE REDUCIBLE COMPOUNDS ARE SUBSTANTIALLY COMPLETELY REDUCED TO THE METAL STATE.

Description

A. ADLER 3,583,864

CHEMICAL PROCESS OF PRODUCING AN IRON-COPPER ALLOY POWDER June 8, 1971 2 Sheets-Sheet I.

Filed May 5. 1969 FIG .l MICRON F l G FIG.3

0. l MICRON f WM l0 O MICRONS 10 0 MICRONS ATTORA E) June 8, 1971 A. ADLER CHEMICAL PROCESS OF PRODUCING AN IRON-COPPER ALLOY POWDER Filed May 5, 1969 2 Sheets-Sheet l FIG;5

l. 0 MICRON ATTOPA/F) United States Patent Oflice Patented June 8, 1971 US. Cl. 75--.5BA 21 Claims ABSTRACT OF THE DISCLOSURE A particulate alloy of iron intimately infiltrated with from about 1 to about 50% by weight of copper is produced by mixing a reducible compound of iron with an appropriate proportion of copper compound selected from the group consisting of elemental copper and reducible compounds of copper, the mixture being heated under re ducing conditions at a temperature between about 1010 and 1150 C. until the reducible compounds are substantially completely reduced to the metal state.

CROSS REFERENCE TO RELATED APPLICATIONS This application is a continuation-in-part of application Ser. No. 471,767 filed July 13, 1965, now Patent No. 3,489,548.

BACKGROUND OF THE INVENTION This invention is concerned with alloys of copper and iron, and more particularly with novel iron powders preinfiltrated with copper, and with processes for their preparation.

As employed herein and in the appended claims, an alloy is defined to be a substance having metallic properties and composed of two or more chemical elements of which at least one is elemental metal. It has in the past been recognized that the physical properties of iron are improved by alloying with copper. Since these metals are virtually insoluble in each other at room temperature, and since their mutual solubility is quite limited even at elevated temperatures, various expedients have been employed in an attempt to bring them into intimate association. Thus, copper and iron powders have been blended and molded into finished objects by the techniques of powder metallurgy. In addition, the infiltration process has been employed, i.e. filling the pores of a sintered powder metallurgy part, of iron or steel in this case, with a metal or alloy of lower melting point, e.g. copper.

These prior methods for the preparation of copper-iron alloys are characterized by important problems and disadvantages. Blended copper and iron powders are subject to segregation in storage and shipment. More important, even very finely divided, blended powders do not provide the degree of homogeneity which affords optimum properties. In addition, sintering of compacted copper-iron powder blends results in expansion of the compact, representing a porosity increase which is a major factor in inferior strength, molding flaws and high rejection rate.

A relatively simple procedure has now been discovered for the preparation of novel copper-iron alloys in powder form, offering substantial cost advantages over conventional infiltration techniques. This procedure imparts to the new powders a unique microstructure, characterized by a more intimate admixture of the elements, to which the superior physical properties which result may be attributable. Upon simple pressing and sintering, the new particulate alloys shrink to high-density products of great strength.

SUMMARY OF THE INVENTION In essence, the process of the present invention entails blending a reducible iron compound with copper or with a reducible copper compound, and reducing these reactants to the metallic state at elevated temperature. The relative proportions are selected to yield a product containing from about 1 to about 50% by weight of copper.

DESCRIPTION OF THE FIGURES The distinctive metallographic features of the new alloys will be better understood by reference to the accomf'panying illustrations, which are photographic reproductions of highly magnified electron microscope photomicrographs. Each photograph depicts a portion of a single particle of one of the new alloys, asetched in a 3% solution of nitric acid in alcohol in order to erode the copperrich areas.

FIG. 1 illustrates an alloy of iron with 7% by weight of copper, prepared by the process of the present invention and enlarged approximately 100,000 diameters.

FIG. 2 shows the alloy of FIG. 1 after heat treatment as detailed hereinafter, also enlarged approximately 100,- 000 diameters FIG. 3 depicts an alloy of iron with 20% by weight of copper, prepared in accordance with the present invention and enlarged approximately 3600 diameters.

FIG. 4 shows the alloy of FIG. 3 after heat treatment, also enlarged approximately 3600 diameters.

FIG. 5 shows a portion of the particle depicted in FIG. 4, this time enlarged approximately 17,700 diameters.

The salient features of these illustrations will be referred to hereinafter, in connection with a discussion of the microstructure of the new products.

DETAILED DESCRIPTION OF THE INVENTION Copper sinters at about 825850 C., and it melts at about 1083 C. in the pure state and at about 1095 C. when it contains 3.2% or more of iron In order to achieve the uniquely intimate consolidation of the iron and copper which characterizes the invention, it is essential that the temperature during the reduction step reach at least C. above the sintering temperature of copper. In addition, it has been found that inferior results are achieved at reduction temperatures below about 1850 F. (l0l0 C.). Such reduction conditions produce a powder alloy of low apparent density containing excessive fines and characterized by poor flow and frequently by high shrinkage upon pressing and sintering. Further, and particularly in reduction with carbon (e.g. coke), temperatures below 1850 F. require excessive time for adequate reduction or lead to substantially incomplete reduction. The minimum practical and economic reduction temperature is 1850 F.

However, particularly for those alloys containing in excess of 20% by weight of copper, reduction temperatures at or above the copper melting temperature tend to cause agglomeration of the reduced product to a mass which can be broken down into smaller particles only with difficulty. Accordingly, especially when alloys of such high copper content are prepared, it is best to keep the reduction temperature below the melting point of copper. In all cases, however, reduction temperatures above about 1150 C. are undesirable, since they favor extensive product agglomeration, producing a tough, hard mass which is impossible to grind to metal powder by any economic means. Best results have been achieved at 18502000 F. (10l01093 C.) and especially at 1900- 1950 F. (10381066 C.).

Under the described conditions, the reactant mixture is reduced to a particulate alloy of iron intimately infiltrated with copper, without the need for the copper to flow through the mass of iron to saturate the pores as is the case in conventional infiltration of molded iron compacts. During the new process, the oxide or other reducible iron compound undergoes a solid state reduction. Where the reduction is conducted below 1094 C., but at least 100 above the copper sintering temperature, the copper does not liquify, but the reduced solids undergo grain boundary diffusion, with comparable results. These lower reduction temperatures are especially preferred in the production of alloys of high copper content, as previously discussed, and the high copper volume fractions are believed to favor products of infiltrated nature.

When reference is made in this disclosure and in the appended claims to particulate alloys of iron infiltrated with copper, this expression is intended to describe individual particles containing both copper and iron in intimate mutual dispersion, as distinguished from mere blends of copper powder with iron powder, and as further distinguished from particles of the one element merely coated with the other.

By their nature, the new particulate alloys of the present invention are not subject to segregation into the individual elements in storage or shipment. They may be truly termed preinfiltrated alloys, since they are directly moldable by conventional powder metallurgy techniques to useful parts, without the need for a separate penetration of liquid copper into a molded iron part as practiced in conventional infiltration.

Any reducible iron compound may be employed in the new process, including iron salts or any oxide of iron such as hematite, magnetite, beneficiated magnetite ores, flue dusts, synthetic oxides or reducible mill scales, e.g. from rolling mill operations. By reducible mill scales is meant those which are reducible to the extent of about 99% or better, such as carbon steel mill scale and low alloy steel mill scales.

The source of copper may be elemental copper, such as reduced copper powder, atomized copper, electrolytic copper powder or hydrometallurgical copper powder. However, it will usually be less expensive to employ an oxide of copper, either precipitated or mechanically produced, such as cuprous oxide, cupric oxide, copper mill scale, copper flue dust, or cement copper, a by-product of mine waste water which typically contains about 50- 98% cuprous oxide. Where copper oxides are employed, it is sometimes beneficial to incorporate nitric acid in the reaction charge, to promote a more intimate dispersion. Alternatively, cupric nitrate or other water-soluble copper salt, may be employed in water solution as the copper source.

In addition to copper and iron, nickel, cobalt, molybdenum or tungsten may be included, or reducible compounds of these elements, such as nickelous acetate, cobaltic oxide, molybdic oxide, or tungstic anhydride. Such elements will generally be employed in minor proportion, i.e. up to about 6% by weight of any one element or a total of up to about 12% by weight in the case of several in combination, in order to improve strength. Such elements may be added in the form of soluble salts such as nitrates or acetates, for optimum dispersion, or as compounds in combination with an acid or base solvent, such as ammonia water or nitric acid.

Those reactants are preferred which have particle sizes finer than 250 microns, and especially preferred are iron compounds finer than about 50 microns and copper compounds finer than about 20 microns, since these favor the most intimate interdispersion of the elements.

While not essential, it is often advantageous to blend the reactants with an aqueous dispersion of an organic adhesive, to minimize dusting losses or segregation of the reactants during handling and subsequent processing.

Suitable binders include animal protein glue, methyl cellulose, hydroxyethyl cellulose, carboxymethyl cellulose, sugar and the like. Such compounds are readily decomposed at the reduction temperature and should preferably be low in ash content. Adhesive levels of about 0.5% of the total batch weight are usually entirely adequate, the proportion of water usually being up to about 18% of the total batch.

Where an adhesive is employed, the blending of reactants may be advantageously conducted in a mix muller or chaser, which will form the ingredients into hard pellets up to about an inch in diameter with little or no secondary grinding. These pellets may be charged to the reduction furnace without drying. If no adhesive is employed, the reactants are thoroughly blended before charging to the furnace.

Reducing conditions may be provided within the furnace in the form of a gaseous reducing atmosphere, e.g. hydrogen or carbon monoxide, or sources of such agents including dissociated ammonia, water gas, producer gas and the like. Alternatively, the reducing conditions may be provided by incorporating finely divided carbon within the reactant mixture, and this procedure is preferred for economic reasons. Suitable forms of carbon include lamp black, petroleum coke, anthracite fines, carbon black, bone black or graphite. The minimum carbon levels, being dependent on the particular reactants and reactant proportions, may be estimated from the stoichiometry of the reaction, but optimum levels are best determined by experiment. Carbon levels of about 8-12% are typical. Even where a gaseous reducing atmosphere is provided, it is sometimes beneficial to incorporate a minor proportion of carbon, particularly in large batches, to ensure effective reduction. As is well known in the reduction art, where carbon is the principal reducing agent it is ordinarily desirable to sweep the furnace with a carrier gas, to drive off the volatile reduction products and prevent re-oxidation. Suitable carrier gases include endothermic gases high in nitrogen, exothermic gases, methane, propane, natural gas, or the like. Manufactured gas, producer gas or hydrogen may, of course, also be used if desired.

The reduction is continued until the reducible compounds are substantially completely reduced, i.e. about 93-98% reduced. During reduction, the product particles usually form a lightly sintered sponge, which may readily be subdivided to final particle size specifications by milling. It has been found that a residual oxygen content of 2% or more yields a particularly friable sponge which is broken up more easily than sponges of lower oxygen content. The time required to carry the reduction to this level will vary with the reducing conditions, the batch size, and the particular reactants selected. Ordinarily, reduction is substantially complete on a laboratory-size batch in from about 30 minutes to two hours. It is undesirable to prolong the reduction unnecessarily, e.g. to the point of quantitative oxygen removal, since this may lead to difiiculty in breaking up the sponge.

After cooling, the product may be ground in a hammer mill or pulverizer to final particle size specifications, generally to pass 60 or mesh.

The resulting powders may be molded into useful shapes by conventional powder metallurgy techniques, i.e. by compacting at about 10-60 tons per square inch and then sintering under non-oxidizing conditions at elevated temperature, preferably at or above the melting point of copper for optimum properties. It should be noted that where blended copper and iron powders are molded in this fashtion, it is necessary at high copper content to sinter below the melting point of copper, with attendant strength disadvantages, in order to avoid loss of copper by blistering. This is caused by agglomeration of the copper-rich masses in such blends during sintering. It is also important to note that the sintering of premixed copper and iron powders results in an expansion of the compact as a reflection of the alloying process. In the new particulate alloys, on the other hand, any expansion as a result of alloying occurs during the reduction step and not during the sintering of the compacted particles. The sintering of these new alloys represents a consolidation of the pre-infiltrated or alloyed particles. This results in shrinkage of the sintered compact, which, as previously noted, is more desirable.

Sintered parts prepared from the particulate alloys of the described reduction process possess good physical properties even without further treatment. The degree of shrinkage and the resulting strength depend mainly on the copper content, with maximum shrinkage and maxi mum strength at about 30% copper.

A further important feature of the present invention resides in the discovery of a novel heat-treatment process for the alloy powders, which leads to a striking and unprecedented increase in the density and physical strength of the final sintered compact. This heat treatment represents in effect a densification or preshrinkage, reducing the degree of shrinkage in the final sintering step without, however, leading to the undesirable expansion effect referred to previously.

The ot'y transition temperature of iron lies at about 910 C., but in the case of iron containing over 3.5% copper this transition occurs at about 835-850 C. It has been found that heat treatment of the new alloy powders is most effective at a temperature between the a-q transition temperature of the iron-rich phase and a temperature about 150 C. lower. The best results are obtained by heat treating at a temperature less than 50 C. below the a-q transition, preferably at about 825-845 C. Such conditions lead to development of the highest density, strength and hardness, with minimum shrinkage, upon sintering. Properties begin to drop abruptly from the optimum when the heat-treatment temperature exceeds 845-850 C.

For best results, the powder should be heat treated for at least about minutes before cooling. This should afford sufficient time for the major portion of the excess copper within the supersaturated iron to separate as a fine precipitate. The maximum time is not critical, and periods up to hours may be used. However, there is no advantage to heat treating for more than 4 hours. Of course, some time may be needed to heat the batch to the operating temperature, and this will vary with the size of the batch.

If the reduced powder subjected to heat treatment contains over 2% oxygen, it is best to heat treat under reducing conditions, so as to further reduce oxygen content below 2% and preferably below 1% during this step. If, however, the oxygen content is already 2% or less, an inert atmosphere may be employed if desired.

The powders are discharged from the furnace as loosely sintered masses, which may be re-ground to particles finer than 60-80 mesh in the same manner as after the primary reduction process.

Heat treatment as described greatly enhances the properties of the alloy powders of this invention. After compacting and sintering, the products exhibit outstanding tensile strength, as may be seen from the appended examples. It should be noted that those alloys containing about 16% copper have tensile strengths above 45,000 p.s.i.; while 610% copper yields 60,000 p.s.i. or more; l040% copper yields 75,000 p.s.i. or more; and 40-50% copper yields, tensile strength above 65,000 p.s.i. By way of comparison, sintered parts made from preblended powders have tensile strengths which level off at 4045,000 p.s.i., and this is substantially unaffected by prior heat treatment of the powder blends.

In addition, sintering shrinkage is greatly reduced as a result of heat treatment of the new alloys, density is substantially increased, and porosity reduced. Each of these advantages is illustrated quantitatively in the examples.

Such sintered properties are within the range of or superior to those of the convention infiltrated part, yet they are obtained by the rather simple and cheap pressing and sintering process rather than by the more involved and costly infiltrating technique.

A preliminary insight into the microstructures of the new alloys is provided by their X-ray diffraction patterns. An alloy containing 20% copper displays essentially the same pattern after primary reduction and after heat treatment: both an a-iIOIl-IiCh phase and an e-copperrich phase is detected. The diffraction pattern of a 7% copper alloy after heat treatment is substantially similar, except for the expected indication that a lower proportion of the copper-rich phase is present. However, the 7% alloy after primary reduction exhibits a pattern corresponding to a single a-iron-rich phase. In each of these diagrams, the iron peaks are relatively flattened, the copper peaks sharper.

The failure to resolve a second phase by X-ray diffraction in the 7% copper alloy after reduction is rather surprising, indicating that alloys of copper content below the solubility limit in iron at the reduction temperature (about 8% at 1125 C.) retain the copper in solution after primary reduction. This may be related to the rate of cooling upon removal from the furnace, the temperature dropping to about 490 C. in the first minute in the typical case, thereby quickly traversing the range of maximum solubility change. (The solubility of copper in iron at 500 C.

is only about 0.25% On the other hand, X-ray data show that subsequent heat treatment within the specified temperature range for 5 rrrinutes or more does result in precipitation of an X-ray resolvable copper-rich phase.

A more accurate insight into the microstructures is gained by microscopic examination of the alloy particles, mounted, polished, and etched with nitric acid to erode copper-rich areas and thereby delineate the phases.

In those alloys containing from about 1 to 10% copper the microstructure is seen to be highly uniform, with the copper content up to about 8% dispersed within the iron grains in the form of particles less than about 1000 A.

(0.1g) in diameter. At 250 power magnification, no clearcut grain boundaries can be seen, and it is necessary to turn to the higher magnification of the electron microscope for further delineation. In FIG. 1 is shown a 7% copper alloy of iron after primary reduction, and FIG. 2 shows the same alloy after heat treatment, both views enlarged about 100,000 diameters. The depressions represent areas where copper was removed by the etching acid, and the fineness of the dispersion is readily apparent. When the fields are surveyed further, it is found that the specimen after reduction displays dispersed copper-rich particles about 350 A. in diameter arranged in clustered bands, with an average distance of about A. between copper particles. The clusters are separated from each other by distances ranging from as little as 500 A. to as great as about 0.75 micron. Between the clusters, copperrich particles also averaging about 350A. in diameter are detected, but it is likely that particles smaller than the 50 A. resolution limit are also present. It is likewise possible to detect cracks in the microstructure, about 0.1-0.3 in length by 400-900 A. in width, as well as angular voids 0.1-0.9,u. in diameter. Upon heat treatment, the cracks are no longer visible, and the clusters now appear in bands or rings about l0ni5 in length by 3,1.LIL1 in width. The copper particles in these rings ar found to average about 400 A. in diameter, with an interparticle spacing of about 150 A. Those copper particles between clusters which are resolved average about A. in diameter, with an interparticle spacing of about 550 A.

Microscopic examination of those alloys of the present invention which contain in excess of 10 and up to about 50% by weight of copper reveals that they contain unusually fine and closely packed iron-rich grains, less than about 35 in their major or largest diameter, with an average separation between adjacent iron grain boundaries of 7 less than about 5,14, in a primary copper-rich matrix. These iron-rich grains contain a highly uniform dispersion of finely divided copper-rich particles.

FIG. 3 illustrates an alloy containing 20% copper, after primary reduction, and FIG. 4 depicts the same alloy after heat treatment, each enlarged about 3600 diameters. In these pictures, the prominent insular areas which occupy most of the field of view are plateau-like iron grains surrounded by copper-rich valleys eroded by the acid. There may be a tendency here for the iron grains to appear to the eye as depressions, but this apparent steroscopic reversal is an optical illusion which can be attributed to the lighting. It will be noted that in the specimen after primary reduction, the iron grains are irregular in shape, whereas after heat treatment they have become rounded or spheroidized.

At high magnification, cracks are detected in the reduced specimen, about 2500 A. in length and 850 A. in Width. These may account at least in part for the higher shrinkage which occurs upon sintering, and they are absent in the particles after heat treatment.

The following are the average dimensions observed in the 20% copper alloy after primary reduction at 1125 C., as determined by Zeiss counter measurements:

Iron grains:

Upon close inspection, the iron grains of FIG. 4 present a much rougher texture than those of FIG. 3, the reason for which is revealed by greater magnification. FIG. 5 represents a portion of the field of view of FIG. 4, enlarged about 17,700 diameters. The most prominent feature of the illustration is an elongated valley or copperrich area separating portions of two adjacent iron grains. The pictured area within those iron grains is pitted by erosion of copper-rich particles through the acid treatment, and it is seen that the copper particles within the iron grains fall into two different size groups: primary particles having a diameter of about 01-05 7, and secondary particles having a diameter less than about 0.05 2.

This unexpected and unique structure only appears after heat treatment, and it seems likely that the larger primary particles represent agglomeration of the original copper particles in the reduced specimens, resulting from the difference in solubility of copper in iron at the reduction temperature (about 8% at 1125 C.) and at the heattreating temperature (about 1.4% at 835 C.) It may be theorized that the primary copper particles represent the copper in excess of solubility at the heat treating temperature, which agglomerates during such treatment step. It may further be theorized that the smaller secondary particles represent that copper which dissolved in the iron to saturation at the heat-treating temperature and subsequently separated during cooling to room temperature. The fact that these secondary particles are similar in size to the copper particles present within the iron grains after primary reduction lends credence to this hypothesis. Nevertheless, it should be understood that the present invention is not limited by any theory or hypothesis, however plausible.

The following are the average dimensions observed in the 20% copper alloy after heat treatment at 835 C.

Iron grains: Microns Length 20.09 Width 16.39 Intergranular spacing 1.66

8 Primary copper particles within iron grains:

Microns Length 0.2713 Width 0.2404 Intergranular spacing 1.66 Secondary copper particles within iron grains:

Length 0.0331 Width 0.0254 Interparticle spacing 0.0287 [ron particles in copper-rich areas:

Length 0.1238 Width 0.1122 Interparticle spacing 0.4720

The dispersion of more copper Within the iron grains in the alloys of the present invention may account for their higher sintered strength relative to conventionally infiltrated alloys of equivalent density, whose copper content would appear to be more concentrated between the grains.

Microscopic examination of the novel alloys of this invention after compacting and sintering has also been conducted, and it is found that the alloys as sintered closely resemble the unsintered particles. This can be seen by comparing the average iron grain sizes and intergranular distances in the heat-treated powders and the sintered parts which result, e.g. for the 20% copper alloy;

Distance Iron grain between iron Length, [l Width, p grains, n

Heat-treated powder 20. 09 16. 39 1. 66 Sintered part 22. 09 17. 39 1. 68 Conventional infiltratioiL 56. 66 27. 91 8. 33

Distance Iron grain between iron Length, ,1 Width, ,1 grains, t

Sintered part from heat treated powder 16. 28 14. 00 3. 41 Conventional infiltration 40. 01 22. 66 12. 51

The porosity of sintered iron compacts is ordinarily such that it is necessary to introduce in excess of 10- 15% copper for adequate infiltration by conventional technique. Accordingly, for a standard of comparison for the new alloys of low copper content, it is necessary to turn to sintered parts made from blended copper and iron powders. Sintered bars containing 7% copper and prepared from a blend of mesh copper and iron powders, when microscopically examined, exhibit large angular pores and massive copper areas 30 and more in diameter. The sintered compact prepared from the new particulate alloy containing 7% copper, on the other hand, exhibits a very fine, uniform, close-packed structure.

The excellent physical properties provided by the new particulate copper-iron alloys can be even further enhanced by various techniques, providing tensile strengths as high as 150,000 p.s.i. For instance, the incorporation of minor proportions of graphite before molding and sintering afiords increases of from 30,000 to 60,000 p.s.i. in tensile strength. Graphite levels of about 0.52% are usually adequate. Re-pressing and re-sintering (coining) operations are also beneficial for increasing density and strength, as are various post treatments, such as quenching, drawing and normalizing, as further illustrated in the examples which follow.

Provision of the following examples for illustrative purposes is not intended to restrict the invention, the scope of which is defined by the appended claims.

Example l.7% copper alloy (A) Reduction: Grams Iron mill scale 1247.1 Dried cement copper 77.0 Hydroxyethyl cellulose 5.3 Water 285.0

The iron mill scale of the above formulation is a byproduct of steel blooming or finishing mills, finer than 325 mesh with about 50% i5% coarser than 20 microns. It has an apparent density of 1.8-2.2 grams per cubic centimeter and an analysis as follows:

As FeO 70.50

The cement copper of the above formulation is a byproduct of mine waste water, finer than 20 microns with about 85% finer than 10 microns. It has an apparent density of 0.8-1.5 grams per cubic centimeter and an analysis as follows:

Percent Cu 1 90.80 8.03 Zn 0.22 Fe 0.47 SiO 0.03 Other metals 0.26 Soluble nitrates 0.01 Soluble chlorides 0.08 Soluble sulfates 0.10

A Cu, 2.91; as C1120, 96.72; as CuO, 2.61.

The ingredients are combined and milled into pellets in a mix muller or chaser, which permits intimate admixture with a minimum of grinding action. The resulting .pellets are charged to a reduction furnace at about 1120-1135" C. and held at that temperature in hydrogen or dissociated ammonia for 45 minutes. After reduction the pellets are removed from the furnace and broken up, first in a hammer mill to 4 inch and smaller, and then in a micropulverizer so that all particles are finer than 80 mesh. The product has an apparent density of 2.3-2.5 grams per cubic centimeter and an oxygen content of about 1.6% (obtained by reduction in hydrogen at 1050 C. for 30 minutes) or 2.37% (obtained by Leco meth0d melting in vacuum at 3500 F.). The hydrogen weight loss reflects only reducible oxygen content.

Hematite (Fe O or magnetite (Fe 'O -Feo) in sufficient quantity to provide the same iron content may be substituted for the iron mill scale in the above formulation. In the same way, pure cuprous oxide may be substituted for the cement copper. j

For the reduction step, carbon monoxide may be substituted for hydrogen, or gases rich in carbon monoxide or hydrogen, such as producer gas, may be used.

(B) Heat Treatment:

150 grams of the reduced powder is charged to the reduction furnace, maintained at 825-845 C. for one 10 hour, and cooled. The powder is discharged from the furnace as a loosely sintered mass and reground to powder finer than 80 mesh in the same manner as after the primary reduction. The annealed powder has an oxygen content of 0.3% (by weight loss in hydrogen) or 1.14%

(Leco method).

Example 2.--1 and 2% copper alloys Grams Iron mill scale 1314.2 'Cupric nitrate trihydrate 37.5 or 76.0 Carboxymethyl cellulose 6.2 Water 300.0

The above formulations are reduced at 1150 C. and heat-treated in the same manner as is described in Example 1. Hematite (Fe O or magnetite (Fe O -FeO) in sufficient quantity to provide the same iron content may be substituted for the iron mill scale.

The cupric nitrate may be replaced by equivalent proportions of cupric oxide and nitric acid, or by an equivalent proportion of cupric acetate.

Example 3.-14% copper alloy With 1% nickel Grams Iron mill scale 1153.3 Dried cement copper 154.0 Animal protein glue 6.0 Water 270.0 Nickel nitrate (20.3% Ni) 49.5

This formulation is pelletized, reduced at 1110" C. and heat-treated as described in Example 1. Equivalent quantities of hematite (Fe O and nickelous acetate tetrahydrate may be substituted for the iron mill scale and nickel nitrate. An equivalent proportion of pure cuprous oxide may be substituted for the cement copper.

Example 4. 20% copper alloy Grams Iron mill scale 1072 Dried cement copper 205.2 Cupric nitrate trihydrate 37.1 Lampblack 105.1 Methyl cellulose 7.9 Water -153 Grams Iron IIllll scale 1072.8

Dried cement copper 220.5

Molybdic oxide 15.0 Ammonia water (26 B.) 34.0 Carboxymethyl cellulose 6.0 Water 275 A similar alloy containing 1% tungsten is prepared in the same manner, by substituting 12.6 grams of tungstic anhy dride (W0 for the molybdic oxide in the above formulation. An equivalent cobalt content is provided by substitutmg 14.1 grams of cobaltic oxide (C0 0 for the molybdic oxide. Hematite (Fe O may also be substituted for the iron mill scale by appropriate adjustment in the quantity added.

Example 5.Particle size elfect scales of varying particle size. The resulting reduced powders are compacted at 50 tons per square inch, sintered in Example 9.-Heat-treatment effect The procedure of Example 8 is repeated, this time confining each heat treatment to a -60-minute period while 14 and sintered as in the previous examples, in both the as-reduced and as-heat-treated forms, with results as follows: I

Tensile Elonga- Sintered Linear strength, tion, density, shrinkage,

p.s.i. percent g./cc. percent Hardness 1% cobalt:

Reduced 31,800 1. 9 5. 97 1. 43 B 29. 1 Heat-treated 55, 200 4.0 6. 79 0. 72 B 59. 7 1% nickel:

Reduced 32, 400 1.9 5. 94 1. 54 B 33. Heat-treated 65, 100 2.3 6.77 0.90 B 70.9 1% molybdenum Reduced 35, 300 1.8 5. 90 1.73 B 28. 1 Heat-treated- 65, 500 1. 9 6. 90 0. 69 B 74. 5

more closely exploring the temperature range between 810 and 850 C., with temperatures controlled to i2 C. Six test specimens are molded from each batch, with averg./cc. and the Sintered densities from 6.52 to 6.84 g./cc., for this series of tests.

Example 10.-Heat-treatment effect The procedure of Examples 8 and 9 is repeated, this time subjecting a 7% copper alloy powder to heat treatment at 835 C. for periods of 30, 45 and 60 minutes, with the following results:

Heat treatment for (minutes) 30 45 60 Compact: Green density, gJcc. 5. 91 6. 40 6. 65 Sintered Compact:

Linear shrinkage, percen 1. 21 0. 97 0. 79 Tensile strength, p.s.i 56, 700 58, 100 63, 300

Example 13.-Use of elemental copper The procedure of Example 1 is repeated, this time substituting for the cement copper an equivalent proportion (70 grams) of atomized copper powder finer than 100 mesh. The reduction is conducted at 1000 C. for minutes, with heat treatment at 835 C. for one hour. After compacting and sintering as before, thereduced and heat-treated powders provide the following properties:

- Reduced Heat-treated Tensile strength, psi. 52, 400 65, 800 Elongation, percent" 0. 8 1. 0 Sintered density, g./cc 6. 39 6. 70 Lineanshrinkage, perce 0.85 0.09 Hardness B 61. 3 B 68. 4

Example 14.-Other copper sources The procedure of Example 1 is repeated,substituting for the cement copper equivalent quantities in proportion to their copper content of various other copper sources. After reduction at 1125 C. for 45 minutes and heat treatment at 825-845 C. for one hour, the powders are pressed and sintered as before to yield the following properties:

Example 11.-Reduction temperature and 45 Si d T ntere ensile heat treatment effects Copper source density,g./cc. strength, p.s.i. Samples of each of the reduced powders prepared in Cum; Oxide (833% (mm M1 50,100 Example 7 are heat-treated for one hour at 835 C. The gop er m nl scaleN(88.1 6%576 gm 2. g $53 emen opper o. heat-treated powders are then compacted and smtered as Cement Copper 2 (93.04% 688 68,000 before, and subjected to physical testing, with results was Reduced copper powder 09. 52% Cu 6.75 69,200 follows: 2

Reduction temperature Batch Batch Heat treated powder:

Apparent density, gJec 2. 44 2. 12 2. 10 2. 69 2. 58 2. 57 Flow rate, see/50 g 33.3 38.6 None 25.3 25.7 25.0 325 mesh, percent- 40. 5 53. 9 60. 0 34. 1 40. 2 45. 0 Hz wt. loss, percent 1 0.51 0. 52 0. 67 0. 30 0.22 0.20 Compact: green density, gJec 6. 48 6. 37 6.30 .6. 6.73 6.66 Sintered compact:

Sintered density, g./cc 6. 68 6. 55 6. 52 6. 83 6. 80 6. 76 Tensile strength, p.s.i 53, 500 49, 000 47, 500 63, 300 61,100 62, 600 Elongation, percent 1. 9 2. 0 1.8 3. 1 3. 2 2. 9 Linear shrinkage, percent 1. 49 1.31 1. 53 0. 79 0.77 0.64 Hardness B 60.9 B 59.4 B 63.0 B 63.3 B 58.9 B 60.4

1 1,100 C. for 30 minutes.

Example 12.-Eifect of other metals In accordance with the procedures of Examples 3 and 4, 7% copper alloy powders are prepared, each containing 1% cobalt, nickel or molybdenum. These are compacted with 0.75% stearic acid, compacted at various pressures,

Example 15.Effect of compacting pressure A 7% copper alloy powder, prepared by reduction and heat treatment as described in Example 1, is combined and each compact is sintered at 1120 C. for 45 minutes in hydrogen. The physical properties, as a function of compacting pressure, are found to be as follows:

Compacting pressure (t.s.i.)

Example 16.--Graphite effect with varying compacting pressures Example 15 is repeated, this time incorporating 1% graphite in each heat-treated powder prior to compacting and sintering, with results as follows:

Compacting pressure (t.s.i.)

Compact: Green density, g./cc- 5. 96 6. 33 6. 68 6. 78 Sintered compact:

Sintered density, g./cc 6 0 Tensile strength, p.s.i. Elongation, percent Linear shrinkage, percent- Hardness 7 0.43 6.80 0.01 72,700 91,500 100,100 137,000 1.9 2.0 2.0 0.0 0. 86 0. s4 0. s1 0. 80 B 72.3 B 83.2 B 86.9 B 100.2

Example 17Coining effect The procedure of Examples 15 and 1 6 are repeated, this time subjecting the final sintered piece to re-pressing and re-sintering under the same conditions used in the first pressing-sintering cycle. The properties achieved are summarized below:

Compacting pressure (t.s.i.)

Compact: Green density, g./cc 6. 6.63 6. 76 6. 74 Sintered compact:

Sintered density, g.lce 6. 83 6. 79 6. 92 6.90 Re-pressed density, g./cc 6.86 6. 81 7. 13 7.05 Re-sintered density, g./cc 6. 93 6. 85 7. 20 7. 11 Final tensile strength, p.s.l 77, 200 111, 300 82, 900 123, 600 Final elongation, percent. 3. 0 2. 0 5. l 3. 0 Final hardness B 94. 2 B 85. 3 B 99. 7 Linear shrinkage, percent:

After 1st sintering 0. 79 0. 91 0. 75 0. 78 After 2nd compaction 0. 08 0. 01 0. 06 0. 01 After 2nd sintering 0.07 0. 01 0. 05 0. 01

1 1% graphite additive.

Example 18.-Graphite effect with varying copper content Copper content (percent) Compact: Green density, g./cc 6. 63 6. 68 6. 86 6. 97 sintered compact:

Sintered density, g./ce 6. 76 6. 80 7. 2O 7. 46

Tensile strength, p.s.i 103, 700 109, 100 112, 000 112, 900

Elongation, percent 2.1 2.0 2. 0 1. 9

Linear shrinkage, percent 0. 77 0.81 1. 46 1. 96 Hardness B 88.3 B 86.9 B 97.4 B 101.5

Example 19.Post-sintering treatments Sintered compacts prepared as in Example 18 are subjected to various additional treatments to further enhance physical properties, with results as follows:

Copper content (percent) Without graphite:

Aged density, g./cc 6. 7 6. 78 7. 33 7. 59 Aged tensile strength, p.s.i 73, 300 96, S00 85, 600 77,400 Elongation. percent 2. 0 1. 9 2. 1 2. 0 Linear shrinkage as aged,

0.48 1. 65 2. 47 B 84.4 B 92.0 B 95.9 With 1% graphite:

Aged density, g./ce 6. 72 6. 74 7. 26 7. 51 Aged tensile strength, p.s.i 118, 100 133, 300 147, 000 137, 200 Elongation, percent 1. 1. 9 1. 9 2. 0

Linear shrinkage as aged,

percent Hardness.

1 Quenched in water from sintering and aged at 485 C. in hydrogen for 30 minutes.

2 Normalized in nitrogen at 1,000 C. for 60 minutes after sintering, then furnace-cooled to 370 0., held at 815 C. in hydrogen for IO-minute soak, quenched from this temperature in water, and finally aged at 260 0. one hour in air.

0. 83 0. 63 1. 58 1. 89 B 85.4 B 34.4 B 110.8 B 109.4

Example 20.Eifect of copper content in reduced powders A series of ferrous alloy powders of varying copper content is prepared by the reduction procedure of Example 1A, with appropriate adjustment in the cement copper charged. The reduced powders are blended with 0.75% stearic acid lubricant, compacted at 50 t.s.i. and sintered at 1120 C. for 45 minutes in hydrogen. The physical properties are summarized in the table below:

Percent Percent Sintered Tensile, Linear density, X10- Elongashrink- Copper 0 1 g./ec. p.s.i tion age Hardness 5. 90 37. 8 1. 9 1. 82 B 27. 2 5. 90 38. 0 1. 9 1. 84 B 29. 7 5. 92 38. 5 1. 1 1. 81 B 35. 2 5. 96 39. 6 1. 0 1. 81 B 34. 0 5. 98 39. 7 1. 0 1. B 35. 5 6. 08 40. 4 1. 9 2. 20 B 42. 2 6. 29 45. 9 1. 8 2. 35 B 49. 3 6. 37 48. 9 2. 0 2. 64 B 51. 6 6. 54 51. 6 2. 1 3. 17 B 55. 6 6. 75 56. 6 2.0 3. 61 B 68. 2 6. 87 60. 5 3. 0 3. 68 B 69. 6 6. 81 79. 7 3. 9 8. 13 B 90. 9 6. 88 58. 2 3. 8 8. 48 B 86. 4 7.27 42.0 4. O 8. 44 B 53. 8

- As indicated by percent weight loss in hydrogen at 1,100 C. for 30 minutes.

2 Reduced at 1,050 O. for 30 minutes; sintered at 1,120 O. for 10 minutes.

3 Reduced at 950 C. for 30 minutes; sintered at 1,120 C. for 10 minutes. 4 Reduced at 950 C. for 30 minutes; sintered at 1,095 C. for 10 minutes.

Example 21.Eifect of copper content in heattreated powders Reduced powders prepared as described in Example 20 are heat-treated at 835 C. for 60 minutes in hydrogen, before compacting at 50 t.s.i. and sintering at 1120 C. for

45 minutes in hydrogen. The effect of the heat treatment on physical properties is summarized in the table below:

Percent Percent Sintered Tensile Linear density, Percent 10- Elongashrinkg./cc. Porosity p.s.1. tion age Hardness 6.80 13. 8 49.0 4. 9 0. 77 B 42.2 6. 82 13. 8 51. 6 4. 8 0. 80 B 50.0 6. 83 14.0 58. 3 3. 9 0.75 B 53.9 6. 83 14. 1 63. 3 3. 1 0. 79 B 63.3 6.89 13.4 68.9 4.0 0.93 B 66.1 6. 89 13. 6 82. 8 3. 9 1. 14 B 77.2 7. 02 12. 3 83. 7 4.1 1. 32 B 80.0 7. 12 11. 2 84. 8 4. 1. 55 B 80.9 7. 23 10. 1 85. 9 4. 0 1. 97 B 83.1 7. 38 8. 4 86. 7 3. 9 2. 32 B 84.8 7. 46 7. 8 87. 4 4. 1 2. 58 B 88.1 7. 48 8.8 87.3 3.0 1. 39 B 89.3 7. 55 9. 3 78.9 3.1 2.03 B 79.7 7. 75 8. 1 66. 4 3. 9 2.18 B 76.5

1 As indicated by percent weight loss in hydrogen at 1,100 C. for 30 minutes. 2 Reduced at 1,050 C. for 30 minutes; sintered at 1,120 C. for 10 minutes. 3 Reduced at 950 C. for 30 minutes; sintered at 1,120 C. for 10 minutes. 4 Reduced at 950 C. for 30 minutes; sintered at 1,095 O. for 10 minutes.

Example 22.Copper-iron powder blends Example 24.Reduction temperature For purposes of comparison, a series of copper alloys Grams is prepared by prior art procedures, by blending appropri- Iron mill scale 2376 ate proportions of 100 mesh reduced elemental copper and Copper fine dust 288,7 iron powders for minutes, compacting at 50 t.s.i. and Coke (Cabot) 303.8 sintering at 1120 C. for 45 minutes in hydrogen. The Sugar (Sucrose) 38.1 properties obtained are summarized below: Water 330.7

Percent Sintered Linear density, Percent Tensile, Elongashrink- Percent copper g./cc. porosity l0- p.s.i. tion age 1 Hardness 6.28 20.4 35.6 4.9 0.01 B 25.5 6. 30 20. 7 40.0 2. 9 0. 92 B 21.4 6.37 20.7 42.1 1.6 1. 88 B 39.3 6. 20.2 45.2 2.3 2.04 B 20.2 6.49 19.8 44.3 2.3 2.24 B 19.7 7.02 16.6 26.6 4.5 0.99 B 3.2

1 All samples expand on sintering. 2 Sintered at 1,000 C. for 30 minutes.

Heat treatment of the powder blends by the procedure of Example 21 prior to compacting has no significant effect on the results.

Example 23.Impact strength The impact strengths of the novel alloys of Example 21 are compared with the values for conventional alloys prepared from blended copper iron powders as in Example 22, and with iron compacts infiltrated with copper in the conventional manner, with results as follows:

Impact strength, it./lbs.

Percent copper Process Oharpy Izod Tension 7 Exafinple 21 7 +1 raphite) o .5 Bleraded powders. g g 3 7 (+1 raphitc o 1L??? Example 21 4. 6 15 Conventional 3. 2

infiltration. 25 d0 5. 0

. Example 21- 5. 3

The above ingredients are combined, milled into pellets, and reduced in 600 gram batches at 2000 F., 1850 F. and 1800" F. The gases evolved, containing carbon monoxide and carbon dioxide, are vented and burned. Thus, continuing combustion of vented gas provides an indication that the reduction is continuing. The reduction times required to approach burn out (completion of recation) are:

1 hr. 48 min 2000 2 hrs. 35 min. 1850 7 hrs. 20 min. 1800 The reduced cakes (12% copper) are crushed, ground and heat-treated at 1830 F. for one hour and then at 1535 F. for 30 minutes in dissociated ammonia. The annealed cakes are then crushed, ground and screened to mesh.

The comparative powder properties are:

l Pressing at 40 t.s.i. and sintering at 2,050 F. 35 minutes.

What is claimed is:

1. A process for preparing a particulate alloy comprising iron infiltrated with from about 1 to 50% by weight of copper, said process comprising the steps of mixing a reducible compound of iron with an appropriate proportion of a copper compound selected from the group consisting of elemental copper and reducible compounds of copper, heating said mixture under reducing conditions at a temperature between about 1010 (1850 F.) and 1150 C. (2100 F.), continuing said heating until said reducible compounds are substantially completely reduced to the metallic state having an iron-rich phase, subjecting said particulate alloy to heat treatment by maintaining said alloy for at least about 5 minutes at a temperature between the oz-'y transition temperature of the iron-rich phase of said alloy and a temperature of about 150 C. below said transition temperature and thereafter cooling the alloy.

2. The process of claim 1 wherein the temperature during said reduction step is below the melting point of copper.

3. The process of claim 1 wherein the temperature during said reduction step is between about 1038 (1900 F.) and 1066 C. (1952 F.).

4. The process of claim 1 wherein said reducible mixture is characterized by a particle size finer than about 250 microns.

5. The process of claim 1 wherein said iron compound has a particle size finer than about 50 microns and said copper compound has a particle size finer than about 20 microns.

6. The process of claim 1 wherein said iron compound is an oxide of iron.

7. The process of claim 6 wherein said iron compound comprises hematite.

8. The process of claim 6 wherein said iron compound comprises magnetite.

9. The process of claim 6 wherein said iron compound is a reducible mill scale.

10. The process of claim 1 wherein said copper compound is an oxide of copper.

11. The process of claim 10 wherein said copper compound is cuprous oxide.

12. The process of claim 10 wherein said copper compound is copper flue dust.

13. The process of claim 1 wherein said copper compound is water-soluble and is introduced in the form of a water solution.

14. The process of claim 13 wherein said copper compound is cupric nitrate.

15. The process of claim 1 wherein said copper compound is elemental copper.

16. The process of claim 1 wherein said reducible mixture includes a minor proportion of a substance selected from the group consisting of nickel, cobalt, molybdenum, tungsten, and reducible compounds of said elements.

17. The process of claim 1 wherein said reducible mixture is blended with an aqueous dispersion of an organic adhesive prior to said reduction step.

18. The process of claim 1 wherein said reducing conditions are provided by incorporating finely divided carbon in said reducible mixture.

19. The process of claim 1 wherein said heating is conducted in a reducing atmosphere.

20. The process of claim 1 wherein said heat-treating temperature is less than C. below said transition temperature.

21. The process of claim 1 wherein said heat-treatment is conducted at a temperature between about 825 and 845 C.

References Cited UNITED STATES PATENTS 2,200,369 5/ 1940 Klinker 0.5 2,754,193 7/1956 Graham et a1 750.5 2,754,194 7/ 1956 Graham et al. 750.5 2,754,195 7/1956 Graham et al. 750.5 2,853,767 9/1958 Burkhammer 750.5

L. DEWAYNE RUTLEDGE, Primary Examinet W. W. STALLARD, Assistant Examiner U.S. Cl. X.R.

Images