US3647573A - Method of making a bronze-iron composite - Google Patents

Abstract

A BRONZE-IRON COMPOSITE CONTAINING FROM 5 TO 75 VOLUME PERCENT IRON, WHEN PRODUCED BY; MIXING FINE POWDERS OF BRONZ AND IRON OXIDE, COMPACTING THE MIXTURE, REDUCING THE IRON OXIDE TO METALLIC IRON, SINTERING THE REDUCED COMPACT TO A BILLET, FOLLOWED BY COLD WORKING THE BILLET TO AN AREA REDUCTION IN EXCESS OF 96 PERCENT, EXHIBITS GOOD PERMANENT MAGNET PROPERTIES AS WELL AS EXCELLENT MECHANICAL STRENGTH, AND CORROSION RESISTANCE.

Description

March 7, 1972 E. o. FUCHS ETAL 3,647,573

" METHOD OF MAKING A BRONZE-IRON COMPOSITE Filed June 5, 1969 320 7 2 I60 5.6 x 9.9 v/o Fe H (OERSTEDS) 2400- |4.3 v/o Fe 9.9 V/O Fe FIG. .2

41m (GAUSSES) FIG.

d/do

/NI/ENTORS HS BY .1 112m v ER ATT EV United States Patent O 3,647,573 METHOD OF MAKING A BRONZE-IRON COMPOSITE Edward O. Fuchs, Union, and James H. Swisher, Stirling,

N.J., assignors to Bell Telephone Laboratories, Incorporated, Murray Hill, NJ.

Filed June 5, 1969, Ser. No. 830,721 Int. Cl. B221? 7/00; H01f 1/08 US. Cl. 148101 Claims ABSTRACT OF THE DISCLOSURE BACKGROUND OF THE INVENTION FIELD This invention relates to a metal composite ferromagnetic material of aligned ferromagnetic filaments in a bronze matrix, useful as a permanent magnet material, and to a method of making same.

PRIOR ART Ideal requirements for a permanent magnet material include the combination of high remanent induction (B,.)' and high coercive force (H as well as structural integrity, indicated by good mechanical strength and corrosion resistance. One way to achieve a high coercive force is to substantially eliminate domain boundaries by subdividing the material into single domain or near single domain particles. Further increases in the coercive force can be obtained by elongating these single domain particles, due to the influence of magnetic shape anisotropy. The theoretical prediction that an aligned assembly of ideal elongated single domain particles could have a maximum energy product (BXH) of about and order of magnitude higher than that of any then existing magnets, let to the development of several methods for producing such assemblies. (Journal of Applied Physics, Supp. to vol. 32, p. 1715 (1961).) These methods are in general complex and time-consuming. For example, one method involves eight basic steps, based upon the electrodeposition and growth of the ferromagnetic metal in a mercury cathode under carefully controlled conditions. (Journal of Applied Physics, Supp. to vol. 32, p. 1715 (1961).) Another method involves reducing parallel bundles of copper alloycoa ted iron wires using wire drawing techniques, and requires repeated bundling and redrawing with several intermediate anneals. (Journal of Applied Physics, vol. 31, p. 1469 (1960).) A similar method involves producing a composite billet of ferromagnetic metal particles in a silver or copper matrix, and repeatedly working and annealing the billet after about a two-fold area reduction to form the desired ferromagnetic filaments. For example, a typical schedule would require three swaging operations, each followed by an anneal, followed by three drawing operations each also followed by an anneal, followed by cutting and bundling the resultant wire, and repeating the working and annealing steps. British Pat. 833,089, issued to Handy and Harman, published Apr. 21, 1960. While the resultant product exhibits good permanent magnet properties, the necessity for repeated working and annealing to efiect extensive reductions 3,647,573 Patented Mar. 7, 1972 makes it commercially unattractive. The corrosion resistance and mechanical properties obtained are not outstanding in these materials.

SUMMARY OF THE INVENTION According to the invention, a composite material of ferromagnetic metal filaments in a bronze matrix has been developed which exhibits good permanent magnet properties and which may be produced by subjecting a billet containing ferromagnetic particles to a cold reduction of only about /2 of that previously required for similar materials, with typically only two intermediate anneals required during reduction, and also exhibits good mechanical strength and corrosion resistance.

In a preferred embodiment, the billet is produced by mixing powders of bronze and iron oxide R 0 compacting the mixture, reducing the iron oxide to metallic iron, and sintering the compact.

BRIEF DESCRIPTION OF THE DRAWING FIG. 1 is a representation of a photomicrograph taken at a magnification of 500x of a longitudinal section of a bronze-iron composite wire of the invention, after an area reduction of 99.7 percent;

FIG. 2 is a graph of coercive force H versus wire diameter reduction (d/a' for three different bronze-iron composites of the invention; and

FIG. 3 is a graph of remanent induction (B versus wire diameter reduction (d/d for the same three different bronze-iron composites of the invention.

DETAILED DESCRIPTION The bronze of the composite material is a copper alloy containing from 1 to 10 weight percent in, below which the tin would have little or no beneficial effect, and above which amount the mechanical properties of the bronze would be adversely affected due to the formation of a brittle phase between the copper and tin. The ferromagnetic material may be iron, cobalt, or nickel, or a mixture, compound or alloy of any of these elements.

It has been found that trace amounts of impurities or minor amounts of additives ordinarily found in commercially available materials do not appreciably afiect the achieving of a suitable product, indicating that highest purity materials are not required. In fact, minor additions of some materials such as nickel or zinc, typically up to about 1 weight percent, to the bronze may be beneficial in contributing solid solution strengthening to the alloy. In general, additives up to about 3 weight percent total and impurities up to about 0.02 weight percent total are tolerable. However, sulfur and oxygen should each be kept below about 0.01 weight percent in the composite, to avoid sulfur or hydrogen embrittlement, the hydrogen embrittlement occurring by the reaction between hydrogen and impurity oxygen.

The ferromagnetic metal may be present in the composite in an amount of from 5 to volume percent, below which suitable remanent induction is not obtainable, and above which the workability is seriously impaired. Within this range the amount of ferromagnetic material chosen to be present will in general depend principally upon the magnetic characteristics desired in the final composite. As already stated, it is frequently desirable to have the highest possible remanent induction (B and coercive force (H in a permanent magnet material. In general, B of the composite will increase directly with increasing volume percent (or packing fraction p, defined as volume percent times 10 of ferromagnetic material, while H of the composite will generall decrease with increasing packing fraction according to the relation,

where H is the coercive force of the composite, H is the coercive force of the isolated particles and p is the packing fraction. It may thus be necessary to choose the particular volume percent which optimizes B, or H for the particular use contemplated.

The billet containing the ferromagnetic metal particles which is to be cold worked to deform the particles into elongated filaments, may be produced by any of several methods known in the art, such as by compacting and sintering a mixture of bronze powder and ferromagnetic metal powder, or compacting such a mixture in a metal tube. However, it is preferred for the achievement of optimum magnetic and mechanical properties to form the billet by intimately mixing finely divided powders of bronze and the nonferromagnetic oxide (e.g., (ac-F6 together with a suitable temporary or permanent binder, compacting the mixture into a structurally integrated body, reducing the oxide in the compact to the metallic species, and sintering the compact. The ease with which the subsequent cold deformation procedure may be carried out is at least in part dependent upon the degree of dispersion of the ferromagnetic metal particles in the matrix of the sintered billet. Such dispersion may in large measure be achieved by beginning with finely divided powders of starting materials. For bronze powder having particle sizes of up to about 40 microns, the equivalent of up to about 15 volume percent ferromagnetic metal may be accommodated without serious impairment of workability. With finer bronze powder, up to about microns, the equivalent of up to about 75 volume percent ferromagnetic metal may be accommodated without serious impairment of workability.

The oxide in the compact may be conveniently reduced to the metal by heating the compact at a temperature of from 300 C. to 600 C. in a hydrogen or other reducing atmosphere for at least 16 hours. If reduction is carried out below 300 C. for less than 16 hours, the rate of reduction is impractically slow. Above 600 C. hydrogen embrittlement may subsequently occur due to incomplete reduction. For convenience, and to insure against oxidation of the reduced particles, sintering may be carried out in the reducing furnace before removing the compact therefrom. Sintering may be carried out in the range of from 800 C. to 1000 C., for from 1 to 8 hours. Sintering below this range leads to a billet having insufiicient strength to withstand subsequent processing, while above this range melting may occur.

The composite billet containing ferromagnetic metal particles, however formed, must be cold worked to an area reduction in excess of 96 percent. Such reduction may be accomplished by swaging or drawing or a combination of these. Rolling, while not as effective in achieving of required properties, may be used to achieve medium coercive force materials. Since such reductions are ordinarily most easily accomplished in the final stages by wire drawing, it is convenient to measure the amount of reduction by the ratio of final wire diameter to initial wire diameter of the composite or d/d In general, it may be stated that the cold reductions to a d/d value of at least 0.2 will result in deformation of the ferromagnetic particles into elongated filaments, substantially aligned with one another, and having lengthto-diameter ratios generally in excess of 10 and that increasing the amount of reduction decreases the diameter of the filaments, increases their elongation and degree of alignment, and increases both the coercive force and the remanent induction of the composite. While the composite may be annealed at any stage of the cold reduction, it is an advantage of the invention that annealing is required only at stages of cold reduction representing about a tenfold decrease in composite diameter. Thus,

for example, a reduction of d/d of 10- may be achieved with but two intermediate anneals. Annealing may be at a temperature of from 250 C. to 700 C., for at least 2 hours, below which workability has not been sufficiently recovered, and above which the fibers may spherodize.

Referring now to FIG. 1, there is shown a cross-section of a bronze-iron composite wire of the invention, containing 9.9 volume percent iron, which has been reduced from 0.95 to 0.05 cm. in diameter (d/a equals 0.05 2), exhibiting elongated and aligned iron filaments in 4 a bronze matrix. Most of the filaments have length-to-diameter ratios in excess of 100.

Example 1 a-FegO prepared by the calcination of ferrous sulfate, and having less than 1 micron grain size, was mixed with a 5 percent tin-bronze having less than 44 micron grain size, in the amounts of 7, 12, and 18 weight percent Fe O corresponding to volume percent of iron after reduction of 5.6, 9.9, and 14.3. Three percent water was added as a binder prior to blending. After blending, the mixtures were compacted at 100,000 p.s.i. into rectangular bars 1.4 x 1.4 x 8.9 cm. The Fe O was reduced to iron at a temperature of 350 C. in a hydrogen atmosphere for 16 hours. The temperature of the furnace was then increased to 900 C. and held for 3 hours to sinter the compacts in hydrogen. A few hot swaging passes were used to densify the bars, then they were cold swaged and drawn to a diameter of 0.05 centimeters, corresponding to about 0.052 diameter reduction. They were then annealed at 500 C. for 4 hours, and drawn to final diameters of from 0.002 to 0.005 cm., corresponding to about 0.0021 to 0.0052 diameter reduction.

B, and H were obtained at several different Wire diameters. Results are shown in FIGS. 2 and 3 and Table 1. In FIG. 2, it is seen that H increases with decreasing wire diameter. As expected from equation 1, the H is higher for the 5.6 and 9.9 volume percent materials than for the 14.3 volume percent material. The difference between the 5.6 and 9.9 volume percent materials is not resolvable within the limits of experimental error. The decrease in d/d corresponds to a decrease in filament diameter, as is shown below in Table 1 for the 9.9 volume percent material.

In FIG. 3, it is seen that B also increases with decreasing wire diameter, as well as with increasing volume percent iron. The results thus indicate the control which may be had over H and B by controlling the amount of ferromagnetic material present, the amount of cold working, and the desirability of extending the amount of cold working to increase H and B,.

The hysteresis loop squareness (ratio of remanent induction B to saturation induction B was also determined for the samples. Results are shown in Table 2 for a d/d value of 0.02.

TABLE 2 (B /B squareness These results indicate good square loop characteristics, making the inventive material useful, for example, as a bistable switching element.

EXAMPLE 2 Tensile strength, 0.2 yield strength and percent elongation in 2 inches were obtained both immediately before and after the anneal, on the 0.1 diameter reduced samples of Example 1. Results are shown in Table 3.

TABLE 3 Tensile 0.2 yield strength strength Percent (p s.i.X10- (p s.i.Xlelongation before before (of 2 inches) Volume percent, iron anneal/after anneal/alter before/after These values compare favorably'with available data for plain bronze in strip form, indicating that the composite has good mechanical strength and formability.

EXAMPLE 3 Corrosion resistance of the composite was tested by immersing samples of Example 1 in a 5 percent aqueous solution of NaCl for 14 days. The rates of weight loss were from 3.7x to 7.9 10- grams per square centimeter per hour, indicating that the corrosion resistance of the composite in salt water is comparable to that of plain bronze, which has excellent corrosion resistance.

What is claimed is:

1. A method for producing a ferromagnetic material including aligned filamentary ferromagnetic particles, comprising the steps of forming a composite metal billet of ferromagnetic material uniformly dispersed in a nonmagnetic matrix, said matrix consisting of an alloy of copper containing from 1 to 10 weight percent tin, and working said billet into wire to elongate and align said ferromagnetic particles and to finally reduce the diameter thereof to at least 0.2 of the initial wire diameter, characterized in that the billet is produced by the method comprising intimately mixing powders of the matrix material and nonferromagnetic oxide of at least one member selected from the group consisting of Fe, Co and Ni, to produce a mixturecontaining the equivalent of 5 to volume percent of ferromagnetic metal, compacting said mixture, reducing the oxide in said compact to metal, and sintering said compact into a billet.

2. The method of claim 1 in which said working is accomplished by alternately cold reducing and annealing said billet, said cold reduction being carried out to elfect up to a tenfold decrease in wire diameter before said annealing is carried out.

3. The method of claim 1 in which the grain size of the powdered matrix material is less than 44 microns.

4. The method of claim 3 in which the nonferromagnetic oxide is Fe O and is present in an amount equivalent to from 5 to 15 volume percent iron.

5. The method of claim 1 in which the reduction of the oxide to metal is carried out at temperatures of from 300 C. to 600 C. for at least 16 hours.

References Cited UNITED STATES PATENTS 2,853,767 9/1958 Burkhammer 75-05 3,029,496 4/1962 Levi 148-31.51UX 3,132,022 5/1964 Luborsky et al. 14831.57 X 3,326,677 6/1967 Alexander et a1. 75--0.5

" FOREIGN PATENTS 697,245 11/1964 Canada 148-101 747,078 11/1966 Canada 14s-31.s7 833,089 4/1960 Great Britain 148-101 L. DEWAYNE RUTLEDGE, Primary Examiner G. K. WHITE, Assistant Examiner U.S. Cl. X.'R.

Images