US3078234A - Magnetostrictive ferrite - Google Patents

Description

Feb. 19, 1963 c. M. DAVIS, JR 3,078,234

MAGNETOSTRICTIVE FERRITE Filed April 24, 1958 2 Sheets-Sheet 1 QUENCH (NI O) (CoO? (F: 0) Fe O} .l l a K QUENOH SLOW COOL l l 1 v v 0.00 0.0| 0.02 0 0.03 0.04 0.05 X INVENTOR.

CHARLES M. DAVISJR.

BY QM United States The invention described herein may be manufactured and used by or for the Government of the United States of America for governmental purposes without the payment of any royalties thereon or therefor.

This invention relates generally to a composition of matter which exhibits optimum dynamic magnetostrictive properties and is more particularly concerned with a cobalt substituted nickel ferrous ferrite.

It is one object of this invention to provide a new and useful cobalt substituted nickel ferrite containing a controlled amount of excess iron in divalent form to improve the dynamic magnetostrictive properties.

Still another object of this invention is a process for producing magnetostrictive ferrites useful in the manufacture of transducers for converting electrical energy into acoustical energy or acoustical energy into electrical energy.

A still further object is to provide a new and useful cobalt substituted nickel ferrite which may be operated at much higher frequencies than is possible with nickel and metallic alloys and which virtually eliminates the necessity of laminar construction of cores made with this material.

Yet another object is the provision of a ferrite which contains less than 23% nickel and 1% cobalt and which is superior in many respects to alloys containing 99% nickel, 50% cobalt or large amounts of other critical alloys.

An even further object is a process of fabrication of a ferrite material which includes quenching the material to improve its dynamic magnetostrictive properties.

These and many other objects will become more readily apparent when the following specification is read and considered in the light of the attendant drawings in which:

FIG. 1 is a graph showing the effect of excess iron on the electromechanical coupling coeflicient [k] of cobalt substituted nickel ferrous ferrite;

FIG. 2 is a graph which displays the effect of cobalt on the electromechanical coupling coefiicient of nickel ferrous ferrite;

FIG. 3 is a graph which indicates the effect of excess iron in divalent form upon the dynamic magnetostrictive constant A and the reversible permeability [n for the quenched samples of FIG. 1; and

FIG. 4 is a table showing the dynamic magnetostrictive properties of a typical ferrite made according to the principles of this invention compared with various alloys, including a conventional ferrite.

Magnetostrictive alloys and ferrites are used in electrical transducers. Nickel is most often used for the purpose in spite of its expense and scarcity, although certain ferroelectric and piezoelectric materials are also employed. A ferrite is a metallic oxide prepared in the same manner as a ceramic. The basic procedure conatet 3,078,234 Patented Feb. 19, 1963 sists of mixing the proper quantities of reagent grade oxides and firing the mixture to a temperature sufiiciently high to permit the atoms to diffuse into the ferrite structure. The general formula used to designate the ferrite composition is:

where Fe is trivalent and M represents a divalent metal ion. The formula for nickel ferrite is:

If a portion of the nickel is replaced by some other divalent metal such as cobalt, the formula becomes:

where x represents the mole fraction by weight of NiO replaced by C00. Variations in x do not alter the ratio of divalent to trivalent metal ions.

Generally e ectro-acoustical transducers constructed of metal and metal alloys are useful only at low frequencies. At higher frequencies it becomes necessary to employ a laminar construction of the transducer to minimize eddy current losses. Since there is a practical lower limit to lamination thicknesses, eddy current losses become excessive at very high frequencies even for the transducer of laminar construction. Furthermore, all the known metallic materials useful for the purpose require the use of large amounts of strategic material such as nickel or cobalt. The only exception is the iron aluminum alloys which, however, are difficult to process.

One disadvantage of the ferrite and piezoelectric materials presently employed is that they do not exhibit dynamic magnetostrictive properties or power handling capacities comparable to those of nickel and some of the metal alloys. Ferroelectric materials also exhibit a low Curie temperature and therefore are subject to the additional disadvantage that they cannot be operated at high ambient temperatures.

As used in this specification and the appended claims, the word acoustical is intended to include ultrasonic and subsonic as well as sonic vibrations.

The general equation of the ferrites with which this invention is concerned is:

It being understood that NiO, C00 and FeO are not present as such in the finished ferrite and that the formula designating the ferrite as a collection of oxides is merely a convenient mode of indicating the valence of the Fe, Ni and Co present in the lattice structure of the material. By mixing together proper amounts of oxides or oxalates of iron, nickel and cobalt or by coprecipitating a saturated solution of proper composition according to well known techniques, a mixture may be formed which then may be presintered at about 1150 C. to produce the composition indicated by the above formula. After presintering, the material is powdered and screened, carefully controlling the impurities which might be introduced during the powdering and screening stage. During or after powdering, a binder is added and the test specimens are formed by pressing or extruding in the conventional manner and are then placed in a furnace and heated to a temperature at which the binder begins to cook off.

This temperature is maintained long enough to eliminate all of the binder and is then increased to approximately 1300-1450 C. and maintained for approximately 1 hour at this temperature in order to sinter the compact and to convert some of the Fe O present to FeO. This conversion takes place according to this general equation:

Nil ,co.rer+.o. on

At the sintering temperature, oxygen tends to come off arid it is therefore necessary to perform the sintering operation in a partial oxygen atmosphere. However, the partial pressure of oxygen in the air is sufficient to produce good results. Only a slight improvement in the electromechanical coupling coefficient [less than 10%] is obtained when the ferrous ferrite was heated in an oxygen enriched atmosphere. On the other hand, heating in an inert atmosphere such as helium tends to destroy the magnetostrictive properties of the material.

Referring now to the drawings which show graphically the optimum numerical values for the subscripts x and y referred to in Equation 1, FIG. 1 shows the variation of the electromechanical coupling coefficient k with the amount of divalent iron present in the ferrite for a series of curves of ferrites containing 0.027 mole of C per mole of ferrite. Curves are shown for both slow cooled and quenched samples. Along the abscissa of the graph of FIG. 1 is plotted values of y, moles of divalent iron per mole of ferrite. The value of the electromechanical coupling coefficient [k] for the slowly cooled samples declines from the value of about 0.24 for no excess divalent iron to about 0.23 for 0.06 mole of excess divalent iron. After 0.06 mole divalent iron, the value of k falls off rather rapidly to zero at 0.08 mole.

The coupling coefiicient of the quenched sample rises rather rapidly from a value of 0.21 at no divalent iron to a maximum of 0.35 to 0.06 mole of divalent iron [which may be thought of, for the sake of convenience, as FeO] and then falls off to a value of 0.25 at 0.08 mole of excess iron. A maximum value of 0.35 occurs for the quenched sample at about 0.06 mole of FeO per mole of ferrite, at this point the k for the quenched sample is far superior to the k of the slowly cooled sample which is 0.22 at 0.06 mole of FeO. The graph of FIG. 1 dramatically illustrates the superior electromechanical properties of the quenched samples as compared with the annealed samples. It should be noted with respect to FIG. 1 that the quantity of cobalt was maintained at a constant value with respect to the moles of ferrite while the nickel was proportionately decreased as the amount of FeO was increased in the samples.

FIG. 2 indicates the effect of cobalt on the electromagnetic coupling coefficient of nickel ferrous while the iron oxide concentration is held at its optimum level of .06. As seen in the drawing, the value of k for quenched samples is again much larger than for the slowly cooled samples. At 0.027 mole of C00, the coupling coefficient reaches a maximum value of 0.35.

In the graph of FIG. 3 the dynamic magnetostrictive constant A and the reversible permeability 11. for the quenched samples of FIG. 1 are plotted against the concentration of FeO while the concentration of C00 is kept at a constant value as is the concentration of the Fe O The NiO is decreased proportionately as the FeO is increased. The value of an remains fairly constant for all proportions of excess iron while the value of A goes through maximum at about 0.042 mole of iron oxide. Increasing has the effect of producing a material. with superior power handling capacities according to the formula:

where B=the flux density, and P =the maximum stress in the transducer material. The optimum value of excess iron to produce the highest value of A is approximately the same as the optimum percentage of FeO to produce the maximum electromechanical coupling coefficient k. Moreover, in the range from 0.02 to 0.07 mole of FeO in the final mixture, the electromechanical coupling coefficient is generally above the value 0.30 and the value of )t is in the range from 2.0 10- to 2.5 10- which are very acceptable values formost purposes. Similarly, the value of the coupling coefficient for the quenched samples ranges from about .3 to .35 for samples of the material containing from 0.01 to 0.04 mole of C00 per mole of the ferrite.

Although none of the graphs indicate the effect of zinc impurities, it has been found that deliberate addition of up to .001 mole of zinc per mole of ferrite does not have an adverse effect upon the magnetostrictive properties of the ferrite. Since, reagent grade materials contain less than this percent of zinc impurities, no special techniques are necessary to control the level of zinc impurities in this process.

The following typical examples of processes embodying principles of this invention are given for purposes of illustration only and are not to be construed as limiting the scope of the invention in any matter whatsoever.

A sample of ]oszsf lo.os[ ]0.02s[ a s] was prepared in the following manner:

The oxides were weighed out and placed in a glass dish, the amount of each were:

Grams F6203 NiO 17.3045 COO 0.4692

The sample was mixed with 45 cc. of ethanol and milled at rpm. in a tungsten carbide mill for 4 hours. The amount of ethanol used was just sufficient to form a slurry which was dried in an oven at 100 C. to remove the ethanol.

The tungsten carbide balls were then separated from the sample by a 20 mesh screen and the dry powder was placed in a boat and put into a furnace; the furnace tem perature was raised to about 1150 C. in approximately l1 /2 hours. The temperature was held for 22 /2 hours and then the furnace was turned off and the powder allowed to cool to room temperature.

The sample was then ground in a tungsten carbide mortar until it passed through a 50 mesh screen. About 20 cc. of ethanol and 4% [by weight of presintered sample] of Cermel C was then added to the presintered sample, and the mixture was milled for twelve hours in a tungsten carbide mill and then dried in an oven at 70 C. The tungsten carbide balls were removed from the sample by means of a 50 mesh screen and the sample was then sieved through a 200 mesh screen.

Seven-tenths of a gram of the sample was weighed out and pressed into a mold [0.250 x 0.450" x 0.450" x 0107"] at 10,000 p.s.i. The sample was then placed in a tube furnace and the Cermel C was gradually cooked off by increasing the temperature to about 300 C. in about 4 hours. The furnace temperature was gradually increased to approximately l300 C.-1450? C. 'while maintaining an atmosphere of air and held at that temperature for 1 hour. A portion of the Fe 0 is converted to divalent iron at this temperature according to Equation 2. The compact is also sintered at this temperature. Sintering is a time-temperature function and may be accomplished at lower temperature by heating for a more prolonged period. The sample was quenched by withdrawing it directly from the hot furnace into room temperature air.

Ferrites made according to the principles of this invention exhibit a Curie temperature of approximately 590 C.; they can therefore be operated at much higher temperatures than is possible with either nickel or the ferro electrics. The amount of cobalt required to produce an optimum magnetostrictive material is dependent, of course, on the operating temperature of the material-the larger the amount of cobalt contained, the higher temperature at which the optimum dynamic magnetostrictive properties occur.

Other ferrites containing larger or smaller percentages of FeO were also made. This was accomplished by varying the relative percentages of R2 NiO and C00 in the original powder mixture. The steps of the process need not be varied to alter the composition of the ferrite however. For example, the following chart shows the effect upon the composition of the ferrite of varying the proportions of the oxide powders:

Perrites having the compositions indicated in the fore going table all exhibit very good magnetostrictive properties as indicated by the graphs in the drawings. Furthermore, other properties of ferrous ferrites made according to the principles of this invention appear to be fairly constant within the composition ranges specified.

A comparison of the properties of various magnetostrictive materials is shown in FIG. 4. The materials listed are 99.5% nickel, 40% Ni-Fe alloy, 43.5% Ni-Fe, 50% NiFe, 12.5% Alfenol, colbalt substituted nickel ferrous-ferrite and cobalt-substituted nickel ferrite. In the second column, the maximum values of electrochemical coupling coeilicient are listed. The largest value listed [40% Ni-Fe alloy] is only slightly larger than that of colbalt substituted nickel ferrous ferrite. Column 3 lists the values of the characteristic frequency times the thickness squared. The values obtained for the cobalt-substituted ferrites exceed that of the metallic alloys by 500 to 1700 times. Therefore, transducers constructed of ferrites could be operated at frequencies in the megacycle range without excessive eddy current losses, or at lower frequencies without the need for laminar construction.

P is given in column 4. For cobalt-substituted nickel ferrous ferrite the value of P is more than twice that of the common ferrite having no ferrous iron. One eifect of the addition of ferrous iron to the crystal lattice of the ferrite is to increate its power handling capacity and make it comparable to that of the metallic alloys.

Columns 5 and 6 indicate the observed data for A and #3 respectively. The increase in k in the ferrous ferrite results largely from the increase in )h Columns 7 and 8 show the percentages of nickel and cobalt contained in the materials listed. The ferrites investigated contain less than 23% nickel and 1% cobalt by weight representing a considerable saving of strategic material.

From the foregoing description it should be apparent that l have invented a new ferritic material containing divalent iron which compares very favorably with nickel iron alloys for use as an acoustical electrical transducer and at the same time requires much smaller percentages of critical nickel and cobalt. Furthermore, by quenching the material, its electromechanical coupling coefii-cient [k] is almost doubled. In many cases the quenching of the material makes it superior in many respects to an iron nickel alloy conventionally used for the purpose.

Obviously many modifications and variations of the present invention are possible in the light of the above teachings. It is therefore .to be understood that within the scope of the appended claims the invention may be practiced otherwise than as specifically described.

What is claimed as new and desired to be secured by Letters Patent of the United States is:

1. The process of producing a magnetostrictive ferrite compact which comprises the steps of preparing a compact consisting essentially of R2 0 NiO, and C00 in the molar ratio xCoO, (1x-y)NiO, Where x=0.01-0.05 and y=0.02- 0.07, heating the compact in at least a partial oxygen atmosphere at a temperature of from about 1300 C. to about 1450 C. for about an hour to convert a portion of the F6203 to diva-lent iron and to sinter the compact to produce a cobalt substituted nickel ferrous ferrite compact and thereafter cooling the compact by quenching in air.

2. A cobalt substituted nickel ferrous ferrite compact as produced by the process of claim 1.

References Cited in the file of this patent UNITED STATES PATENTS 1,976,230 Kato et al. Oct. 9, 1934 2,626,445 Albers-Schoenberg Jan. 27, 1953 2,636,860 Snoek et a1 Apr. 28, 1953 2,723,239 Harvey et al. Nov. 8, 1955 2,736,708 Crowley et al. Feb. 28, 1956 FOREIGN PATENTS 510,462 Belgium Apr. 30, 1952 1,048,444. France Aug. 5, 1953 1,071,068 France Mar. 3, 1954 1,086,346 France .Aug. 11, 1954 756,383 Germany Oct. 20, 1952 OTHER REFERENCES Weil: Comptes Rendus, Mar. 24, 1952, p. 1352.

Wijn et al.: Philips Tech. Rev., August 1954, p. 52.

Bozorth et al.: Physical Rev., Sept. 15, 1955, p. 1792.

Gorter: Proceedings of the IRE, December 1955, pp. 1954, 1960.

Katz: RCA Technical Notes, No. 88, l p., recd. Dec. 2, 1957.

Fresh et al.: Proceedings of the IRE, vol. 44, No. 10, October 1956, pp. 1303-1311.

Claims (1)

1. THE PROCESS OF PRODUCING A MAGNETOSTRICTIVE FERRITE COMPACT WHICH COMPRISES THE STEPS OF PREPARING A COMPACT CONSISTING ESSENTIALLY OF FE2O3, NIO, AND COO IN THE MOLAR RATIO

Images