US2499861A - Permanent magnets and alloys therefor - Google Patents

Description

March 7, 1950 J. R. HANSEN PERMANENT MAGNETS AND ALLOYS 'H'IEREFOR Filed larch 16. 1948 Tn: mruuezuc: orfiuucow AND Zmcomum m a ncsrmcrms -rm: GAMMA PHASE or ALmcoYA-r M50 E a 1/ 5.5 1R1 b 3' i E1 45 257V 5.0 0' lo 20 so 40' 50' Tm: AT HSO'Fm mmu'rzs.

INVENTOR. Jo/m A. Hansel? BY Patented Mar. 7, 1950 PERMANENT MAGNETS AND ALI-Y8 THEREFOR John R. Hansen, Pottersvliie, N. 3., aslignor to Crucible Steel Company of America, New York, N. Y., a corporation of New Jersey Application March 18, 1948, Serial No. 15,250 Claims. (Cl. -424) This invention pertains to improvements in ferrous alloys containing aluminum, nickel and cobalt, and optionally also copper, adapted primarily for use in permanent magnets, to permanent magnets thereof, and to processes for producing the same. Alloys of this type are known as "Alnico alloys.

This application is a continuation-in-part of my copending application Serial No. 101,053, filed October 3, 1946.

It is known to make permanent magnets of an Alnico alloy containing from about 6 to 11% aluminum, 12 to nickel, 16 to cobalt, up to about 7% copper, and the balance substantially iron. It is also known to enhance the magnetic properties of such magnets, particularly the (BH)rnax values, by rendering the same magneticaly anisotropic by heat treatment, 1. e., by cooling from above to well below the Curie point while in a magnetic field, the lines of force of which follow the direction of subsequent permanent magnetization of the magnets.

Objects of the invention are to provide a modified analysis of the above alloy havin greatly improved mechanical and magnetic properties, and such as to permit of a greatly simplified heat treatment.

Alnico alloys, including the above analyses, are characterized as a class in being very brittle and lacking in both hot and cold malleability. They are further not machinable to any practicable degree. It is, therefore, the practice to cast permanent magnets thereof essentially to the finished shape, break the castings from their respective sprues, and further process by cutting or grinding or both with suitable abrasive wheels. The brittleness is especially manifested by breai, age and cracking in casting, grain pickout and edge chipping during grinding and cutting-oi! operations. This reflects the lack of ductility, coarse grain size, and intergranular weakness prevalent in these alloys. Many service failures involving the use of Alnico permanent magnets are caused by the dislodgement of grains or crystals from the magnet, which then become affixed in the small air gaps used. This is especially true in applications such as magnetos where the permanent magnet is the rotor or moving componentof the assembly.

Now I have discovered in accordance with one aspect of this invention that the addition of zirconium in conjunction with titanium substantially eliminates brittleness as manifested by grain pickout and edge chipping.

It is known to add titanium to an alloy of the analysis above specified for improving the magnetic properties in certain respects, for imparting, for example, increased coercive force, but from the standpoint of mechanical properties, such additions per se are wholly undesirable as they result in a high percentage of rejects by reason of edge chipping and grain pickout. I have discovered, however, that additions of zirconium together with titanium, in amounts as specified below, effectively eliminates this type of brittleness.

I have further discovered, curiously enough, that additions of zirconium alone do not improve the mechanical properties of the alloy in the respects above noted, but only when employed in conjunction with titanium. The aforesaid immovements in mechanical properties of the alloy, resulting from additions of a combination of the elements zirconium and titanium, is contrary to expectations in view of the above noted embrittling action of titanium alone, coupled with the fact that the aforesaid improvements are not achieved by the addition of zirconium alone.

Another serious defect which is encountered in the production of permanent magnets from Alnico alloys, including the above analysis, is the frequent occurrence of small voids or cavities therein. Such magnets have heretofore been made by casting the alloy, usually in sand molds. As a result, small voids or cavities are frequently found to be present in the casting, especially at or near the surface thus impairing the magnetic properties. After processing by cutting or grinding or both, the presence of these voids on the pole faces is a particularly serious defect as regards performance of the magnet. These cavities are evidently produced by occluded gases entrapped during casting of the molten metal and subsequent solidification thereof. The resulting gas porosity of the magnets necessitates a high percentage of rejects, as shown by the data presented below.

New I have discovered, in accordance with a further aspect of my invention, that this gas porosity may be effectively eliminated by addition of zirconium to the alloy. My investigation show that for eliminating this defect, the addition of zirconium alone suflices, gas porosity being thereby effectively eliminated whether or not titanium is added to the alloy. The mechanism by which the gas porosity occurs is not as yet clearly understood. However, its elimination by addition of zirconium to the alloy suggests that nitrogen is in some manner responsible, as zirconium readily combines with this element.

For substantially eliminating gas porosity, zirconium may be added to the above-mentioned Alnico alloy in amounts ranging from about 0.25 to 5% although 0.25 to 1% ordinarily suffices. In addition to preventing gas porosity, additions of zirconiumwithin these ranges will also eflectively eliminate grain pickout and edge chipping when incorporated in the above mentioned Alnico analysis along with titanium, the latter likewise in amounts ranging from about 0.25 to and preferably about 0.25 to 1%. For eliminating grain pickout, best results are secured by adding zirconium and titanium in the proportion of about two parts by weight of zirconium to about three parts by weight of titanium, and within percentage limits of each not exceeding about 1%, as substantiated by the test results hereinafter set forth.

A further serious drawback to the attainment of optimum magnetic properties with magnets of the above Alnico analysis as heretofore produced, is the susceptibility of the alloy to the formation oi a multl-phase, usually a two-phase, microstructure, whereas best permanent magnet properties are secured with a single-phase microstructure.

With reference to this point, the microstructure of an alloy of the above Alnico analysis is found to be effected by heat treatment as follows: After homogenizin or normalizing at about 2300 to 2400 F., the alloy will be round to embody a single-phase microstructure. This phase has been identified as having the alpha or body centered cubic structure, and is accordingly referred to herein as the "alpha" phase. However, on heating the alloy at about 1500 to 1600 F. (or at lower temperatures for long time periods), the precipitation of a new phase, designated herein as the "beta" phase, occurs, whereby the alloy assumes a two-phase microstructure comprising the solid solution or alpha phase having the beta phase dispersed therein in a tine state oi subdivision. This beta phase appears to comprise a precipitate of intermetallic compounds derived from the various alloying constituents. This twophase microstructure, consisting of the alpha and beta phases, is also observed in the alloy in the "as-cast condition. 0n heating the alloy to a slightly higher temperature of about 1650 F., a single-phase microstructure, consisting only of the alpha phase, is obtained. At higher temperatures, however, of about 1700 to 2250 F., the precipitation of a third phase occurs, referred to herein as the "gamma" phase, which results, on cooling from this temperature range, in a twophese microstructure comprising a mixture, in varying proportions, of the alpha and gamma phases. The gamma phase is so designated because there are indications that it consists of gamma iron at the above stated elevated temperature and undergoes transformation to a martensitic-iike constituent, as observed under.

on cooling to room temperature.

the microscope.

Both of the two-phase microstructures aforesaid have distinctively inferior magnetic qualities as compared to the single-phase structure resulting from heat treating the alloy at 1650' F. or at 2300 to 2400 F. as aforesaid. This is shown by the test data set forth in Table I below, according to which individual test bars made of an alloy of the above character and having the chemical analysis given in the table, were first soaked respectively at the progressively increasing temperatures given in column 1, were thereupon cooled in a magnetic field to below the Curie point and down to a temperature of about 950 to 1000 F., and thereupon drawn or tempered for five hours at 1100 F. Column 2 shows the relative flux values for the individual specimens resulting from the treatment aforesaid; while column 3 shows the resulting microstructure thereof.

TABLE I Chemical Analysis of Specimen 'lcsied, Percent Alloy 0 Mn Si Al Ni Co Cu Fe A 0.02 0.03 0.00 8.7 14.5 24.7 3.3 Ba].

Column 1 Column Column 3. Heat I ig xf Flux Resulting Treatment 1, 2 M inm- Temp, F. Magnetic Field structure I l. 600 76. 0 n+6 is a n+2? g l, 300 51.5 a+lil 0 g l, 1100 44. ii a+25 g 2. 000 29. 5 0+3 9 2, 38. 5 0+ 0 Z 200 B5. 3 11+ 15 o y 2, 300 100. 0 a 2, 370 100. (l a l Nomenclature: a, b and o designate the alpha. beta and gamma phased, respectively.

It will be seen from the comparative flux values given in the above table that optimum flux values were obtained for the specimens quenched from the normalizing temperature range of about 2300 to 2400" E, or alternatively when quenched from about 1650 F., corresponding to the heat treatments resulting in the single-phase microstructure as indicated in the table. For heat treatments resulting in the two-phase microstructures, either that embodying the alpha plus beta phases, or that embodying the beta plus gamma phases, the flux values obtainable are materially less than those obtainable for the single-phase microstructure comprising only the alpha phase. It will also be noted that in the two-phase structure containing the gamma phase. the magnetic properties progressively decrease with increase in percentage of the gamma constituent.

The test results of Table I would appear to indicate that, for purposes of imparting the desired single-phase microstructure to "as-cast permanent magnets of the above-mentioned alloy, and also for imparting the desired anisotropic magnetic qualities thereto, it would suflice to heat the magnets to a temperature of about 1850 F. followed by cooling substantially below the Curie point in a magnetic field. I have found. however, that in the quantity production of such magnets on a commercial scale. this is decidedly not the case, due for example to such factors as: the tendency of the two-phase microstructure of the "as-cast magnet to persist on reheating, the strong gamma-phase forming tendencies of carbon unavoidably present in the alloy, and also to variations in individual melts due to melting tolerances, whereby such gamma-phase forming constituents as nickel and cobalt may be present in disproportionate amounts in individual melts. e

Due to such considerations, I have found it to be necessary with .Alnico alloys of the above-mentioned analysis, first to sub] ect the magnets thereafter to reheat within precisely regulated lower to cooling in the magthe above-mentioned single-phase microstructure and anisotropic magnetic qualities to the magnets. Thus, the alloy, after casting into a magnet and subjecting to pre- 5 rations, such as grinding,

ete., must be given an initial or normalisingheat treatment. consisting in homogenizing or normalising at 2350 to 2400 1". for about iifteen minutes, followed by a rapid cool to room temperature. For imparting the above-mentioned aniso tropic properties to the magnet. it must then be reheated above the Curie point (i. e.. the temperature at which the alloy transforms from magnetic to non-magnetic). and within the very restricted and critical temperature range of 1610 to 1660' F., for about live to sixty minutes until thoroughly soaked at temperature, this to assure substantial retention of the single-phase microstructure on subseouent cooling. Thereupon. the magnet is eooledin amagneticiiemasaforesaid,, until it is suiiiciently below the Curie point to secure the full anisotropic effect of the magnetic field, this cooling being ordinarily carried down to a temperature within the range of about 1000 to 1200' F. The magnet is thereupon removed from the magnetic held and air-cooled to room temperature. It is then reheated or drawn at a temperature somewhat above 1100" I". and

then cooled below 950 F., for purposes of further improving its magnetic qualities.

Now I have discovered, as set 'forth in my copending parent application Serial No. 701,053, above referred to, that by the addition of small but substantial amounts of silicon to the Alnico alloy having the above-mentioned range of analysis, the aforesaid normalizing treatment may be omitted and the desired single-phase microstructure retained on cooling the alloy from much broader temperature ranges below 1800" F. than those aforesaid, and that. in consequence, the heat treatment of such magnets may be greatly simplified. Thus, I flnd that by adding silicon to the aforesaid alloy in the amount of about 0.15 to 1%, and preferably in the amount of about 0.2 to 0.4%, by weight of the resulting alloy, the initial normalizing heat treatment. consisting in homogenizing at 2350 to 2400 I". and quenching therefrom. which I have found to be otherwise required, may be entirely eliminated; and that, furthermore, the heating of the magnet above the Curie point, for imparting the anisotropic properties referred to, may be carried out within a much broader temperature range. vin, 1590 to 1700 F., than the narrow and critical temperature range of 1610 to 1600' F. which I have P found to be requiredwith the alloy omitting the silicon content.

I have further discovered. in accordance with another aspect of the present invention, that this some improvement may be obtained by substituting zirconium for all or part of the silicon. That is to say, I have now found that the addition of small but substantial amounts of either or both zirconium and silicon, inhibits the formation of the undesirable multi-phase microstrueture, and assures the desired single-phase microstructure of the alloy when heat treated from 1590 to 1700 1". as aforesaid. If zirconium is wholly substituted for silicon, it should be added in amount ranging from about 0.15 to 1%. Too much silicon or zirconium cannot be employed to best advantage, for example, more than 1% of each, as this is deleterious to the resulting magnetic properties. As shown by the test data presented below, a combination of zirconium and silicon within the percentage limits above stated is most effective for inhibiting formaton of a multi-phase microstructure. Additions for this purpose of either or both silicon and zirconium completelyeliminates the necessity for normalining the alloy at 2300 to 2400' 1". prior to the anisotropic treatment. as is required without these additions, and thus eliminates the necessity for heat treating the "as-cast" magnets in excess of 1000" l". for imparting the desired single-phase microstructure. as well as for purposes of the anisotropic treatment above referred to.

The permanent magnet alloy in accordance with my invention has the following broad and preferred ranges of analysis:

Carbon-as low as possible, i. e.. 0 to 0.1% max" and preferably under 0.05% max.

Manganese-about 0 to 1% Hickeb-about 10 to Cobalt-about 18 to and preferably over 20% to about 90% Aluminum-about 6 to 11% Copper-about 0 to 1%. and preferably 0 to 3.5% Remainder-Substantially iron, except for the following: (a) For :ninimizing brittleness and grain pick Zirconium-about 0.25 to 5%, and preferably about 0.25 to 1%; together with Titanium-about 0.25 to 5%. and preferably about 0.25 to 1%.

(b) so: eliminating a 'multi-phase microstructure:

Metal of the group silicon and zirconiumabout 0.15 to 2%. and preferably about 0.15 to 1% of each.

It will be understood, in connection with the above. that either or both the additions (a) and (b) may be employed in any particular analysis. That is to say, there may be added either silicon alone or zirconium alone or a combination of these two; or zirconium may be added in conjunction with titanium: or all three of these elements may be present, dependent on the particular improvements desired in accordance with the various aspects of the invention above discussed As illustrative of the improvement obtained with the present invention, the following test results and inspection report data based on quantity production of permanent magnets, are presented.

Table 11 below gives daily inspection reports on the quantity production, by applicant's assignee, Crucible Steel Company of America, of rotary magneto magnets of the approximate analysis: aluminum 8.4%; nickel 14.5%; cobalt 23.5%: copper 3.0%; and balance iron except for residuals and except for titanium and zirconium additions as indicated in the table:

TABLE II Percent Percent Alloy Modiag Rejections Rejections Rift? ficatlon mm Gas Brittle- R 1 Porosity pass I e A. CASTING AND GATE GRINDING INSPECTION 0 z! .45 TI I, 258 5. 5 0 l5. 5 D0 15. 8 0. 38 1B. 2 007 14. 6 0. D16 14. 6 5. 8 0 5. 8 10. 0 0 l0. 0

7. l5 0. B8 7. l1 0. 037 D 0. (B7

INSPECTION 4 55 2. 8 7. 5. 80 U 5. 80 2. 25 i [W 8. 92 2. 70 3. 6. 20 i. 098 0. 53 0. 6B D. 084 0. 468 0. 652

1 lnclndesbmhenmstings, cracking, chipping and grain pickout.

Thus, zirconium additions reduce reiections !or gas porosity to practically zero as compared to total rejections for this defect, r from about 10 to 50% without this addition. Also zirconium in conjunction with titanium reduces total rejections for brittleness, to less than onehalf of a percent, as compared to rejections ranging up to about 7% without these additions.

The rotary magnets of the Table II inspection reports were not subjected to any cutting-off or grinding other than that of gate removal, and therefore do not show abnormally high rejections for brittleness. To illustrate this, Table ma below gives daily inspection report data, base on the production of magnets of the above analysis and of dimensions 2%" x x 3'," ground to a radius of 1?," on the 2" x 1"" edge.

Thus, whereas total rejections for brittleness ranged from about 50 to 70% in the absence of zirconium along with titanium, rejections for this defect dropped to about 10% for the above type of magnet when titanium and zirconium were added in accordance with the present invention. The manufacture of this magnet using a grade with .45 titanium zirconium resulted in over 90% rejections because of grain pickout and edge chipping, thus requiring the discontinuance of this grade.

Table III!) below gives further inspection report data on the effect of zirconium additions, but without titanium additions, in eliminating gas porosity in the quantity production of various sizes of permanent magnets made of an alloy of the approximate analysis: aluminum 8.3%; nickel 13.8%; cobalt 23.6%; copper 3.1%; and balance iron except for residuals and zirconium 9 ing to 1135 tests on alloys A 8 tive test typirying the improvements in magnetic properties of permanent magnets resulting from silicon additions, in accordance with my invention, to the aforesaid previously known Alnico l alloy omitting silicon. In accordance with these tests, heats were made of alloys having the chemical set forth in Table .IV as alloys A to C respectively, these alloys analyzing substantially the same except for silicon which was present therein in the percentages of 0.02, 0.26 and 0.61 respectively, as indicated. For the magnetic tests, %-inch square bars of each analysis were made up and groundto 1" inch square, cut to 2-inch lengths and squared on the ends to 1.875 inches in length. These were no malized in a gas-fired rotary furnace at 1650 F. for flfteen to twenty minutes, and thereupon cooled to 1100 F. in 4.5 minutes, while in a magnetic field. The specimens were then drawn by heat- F. for flve minutes and furnacecooled to 950' F. in sixty minutes, and checked for magnetic properties in the permeameter with the results as given in Table IV.

8' TABLE IV Chemical Analysis of Specimens Tested, Percent Alloy 0 Mn Si Al Ni Co Cu Bai.

'0 A .020 as .02 a? 14.5 24.7 3.3 Fe B .015 02 .M 8.7 l3.5 2&9 3.1 Fe O .015 02 .61 8.4 13.4 H37 3.! Fe

Results of Magnetic Tests Allo A Allo B A110 0 .02738! .26 0 8i .0] 0 Bl H0 i510 12, 3;: 02 80 495 13' 430 Bd 10,!!! 10.800 4.96 4.65

It will thus be seen that the addition oi silicon within the limits above stated has g eatly increased the (BHhm value without sacrifice in other magnetic qualities, the increase amounting to about 20% for the 26% silicon addition.

Table V below gives the results of comparative to C respectively of Table IV, to show the eii'ects of silicon additions in restricting the formation of a two-phase microstructure, on cooling the alloys to progressively lower temperatures from the normalizing temadditions as indicated. perature before quenching; while Table VI be- TABLE DIb Inspection report for gas porosity Percent PM.

Rejections M t N .M ts an Modi- Pat 511 fnspe ii ation 3? Porosity Emmi P. I.-l0l757 0.10 0. -5so1s 4.30 P. I. & o.1.-m2s.. not up P. I.74554 o w] o. 1,-52094 4.00 P. I. & 0.1.14151. 1.40 1.49 P. I.6687 0.00 a C.I.6046 2.80

P. I. a o. 1,-1827--. 0.10 0.10

.I. u cti s. Licensees. Table IV below, gives the results of comparalow shows, for the same alloys. the elects of silicon additions in restricting the formation a two-phase microstructure, on heating the alloy from room temperature to progressively increasing temperatures.

TABLE V [Nomenclaturm oalpha phase, ggamma phase.)

Heat Treatment, F. t 3 3 2400l 5' cooled to:

HOW-Water Quench.. a a a. 2300-l0' Water 2250 Water a+5% gl0' Water a+357 9.. 2l5010 Water o+36 7.. 2l00l0 Water a+35 0.. a 0. 205010' Water a+35% 0.- a+trace g. a+traco a. 2000l0' Water o+357 a c+5% a... a+trace g. 195010 Water +35% g u+traoe 41.

TABLE VI lNomenclature: c-alpha phase, g-gamnia phase] Allo A Alloy 13 A110 0 Heat Treatment, F. lmqgsi 26% Si m 0 s1 170030 Air 0001 n+1 a. 175030 Air 0001 0+1 an n+1 a.. a. 180030 Air Cool 0+1 9.. 0+1 7. a+1% g 185030' Air 0001.. a+10 g o+10 n a. c+1% g 190030' Air Oool. a+i0 1 a+30% 0-. wflgg l95030 Air Oool 04-30 9.. a+25% 0. 0+5 0 a 170060Air Cool c+ 7.. a a.

It will be observed from Tables IV, V and VI that the cheat of silicon is greatly to restrict the formation of the two-phase microstructure, both as regards cooling from the normalizing temperature and as regards heating from room to elevated temperature. Thus, as shown by Table IV, where silicon is present in the alloy in the amounts stated, the single phase microstructure is obtained on normalizing or homogenizing over the relatively wide temperature range of 2l00 to 2425 F. Tests cl 9. similar nature have established that where silicon is present in the amounts stated, the single phase microstructure is also obtainable by the single step of heating the "as-cast magnet within the relatively broad lower temperature range of about 1590 to 1700 F.

It is by reason of these effects that I am enabled in the quantity production of silicon containing permanent magnets in accordance with my invention to obtain the desired single phase microstructure, either by the single step normalizing or homogenizing the "as-cast magnet over the relatively broad temperature range of 2l00 to 2425" F., or alternatively, by the single step of heat treating the "as cast magnet within the lower temperature range of 1590 to 1700 F. The first procedure above-mentioned, viz., normalizing or homogenizing to 2100 to 2425 F., constitutes a modification of the previous practice described, as the magnets are thereupon cooled directly in a magnetic field. My preferred heat treatment, however, is that last mentioned, namely, the complete omission of the normalizing treatment previously required, and the substitution therefor of the single step of heat treating within the temperature range of about 1590 to 1700 F.

By way of demonstrating the improvements resulting from silicon additions in the quantity production of permanent magnets on a commercial scale, 300-pound induction heats of the above alloy were made, directed to similar analyses ex cent for silicon, with respect to which, however. the aim was to produce heats having silicon contents of 0.1, 0.2 and 0.3% respectively. The analyses actually obtained are as given for alloys E, F and G in Table VII below. These heats were then cast into magnets of similar dimensions, which were given the heat treatment aforesaid, including soaking at 1650 F. followed by cooling in a magnetic field below the Curie point to about 1000 to 1200 F'., and subsequently drawn by reheating above 1100 F. followed by cooling below 950 1". The resulting magnets were then tested in the permeameter for (BB) max values with results as shown in Table VII as follows:

TABLE VII Chemical Analyses of Magnets Tested, Percent Alloy C Si Mn Ni Cu A1 00 Fe E 022 .13 .02 13. 65 3. 20 8.36 24. 00 Bal. F .031 .23 .02 13.80 3.24 8.41 23.80 B51. 0 .015 32 .02 13. 3. 20 8. 49 24. 14 Ba].

Results of permeameter tests expressed in cumulative percentage of magnets tested having indicated (BHMM: values It will be seen from the above tabulations that for the lowest silicon content of .l3%, the cumulative percentage of magnets tested having (BHlmx values below any given value between 4.0 and 5.5 million is considerably higher than are the corresponding percentages for the magnets having higher silicon contents within the range of analysis of my invention. This is of importance in the rejection of magnets for low (BH)mnx values in quantity production. Thus, for example, if the lower acceptable commercial limit for (BH)max is set at 4.5 million, the above table shows that the rejections in the case of the 0.13% silicon analysis would be 12.50%, as compared to rejection of only about 1 for the 23% silicon analysis and 0% for the 32% silicon analysis.

A direct comparison is given in the following Table VIII of magnets obtained from two 300 pound induction heats, in one case consisting of an alloy without silicon, and the other with silicon. The alloys are designated as H and I respectively. Alloy H was heat treated from above the gamma region; cooled rapidly to below said region and then subjected to a magnetic treatment from 1650 F. Alloy I was heat treated directly from 1650" F. into a magnetic field. Both were followed by a draw treatment which consisted of heating to somewhat above 1100 F. and cooling to about 950 F. The energy product distribution is expressed in cumulative percentages.

Tables IV to VH1 1:10., but employing zirconium Tm Um in conjunction with silicon for eliminating the aforesaid two-phase microstructure. Table 1:: WWMMMM'M gives the analyses at the various heats melted my up and cast into radio loud speaker magnets.

N on A] Table It gives test results for magnetic properties of the resulting magnets. Table XI shows 3 m 133 mg) a an the influence of the zirconium and silicon addi- 1 .020 .28 10.90 M2 M4 2 -1 tions, in inhibiting iormation or a two-phase microstructure.

TABLE 1::

Chemical Anllyuis, Percent But No. a

A1 e1 '1! 21 Ni On 00 Mn 0 0.00 0.42 0 0.00 12.00 a. 20.01 000 0022 0.000" mm 0.20 0.40 0 0.3 10.00 0.10 20.01 000 {"fi B0. 0.20 0.40 0 0.20 10.00 0.20 20.10 0.02 0.02 0.000" nm. 004 0.01 0 0.21 10.00 0.20 20.70 0.00 0.02 Do. 0.11 0. 01 0 0.21 10.00 0.10 20.01 0.02 000 Do. 040 000 0 0.20 10.00 0.10 20.01 0.00 0.010 0.000" nu. 0.01 044 0 0.27 10.10 0.24 20.01 0.00 0.02 Do. 001 0.40 0 0.21 1000 010 2010 0.02 0010 Do. 001 0.40 0 0.20 10.00 0.20 20.10 000 0.02 Do. 0.10 000 0 0.20 10.10 0.12 20.01 000 0020 1.10" Dis. 020 0.01 0 0.00 10.70 0.20 20.01 0.00 002 no. 000 0.04 0 0 10.00 200 20.42 0.00 0.00 "round. 000 0.00 0 0 1000 0.10 20.00 0.02 0.00 unre. 000 0.20 0.02 0.02 10.00 2.20 20.10 0.00 0.010 .100! D01. 0010 0.12 0.00 0.00 1000 0.20 20.70 0.00 0.02 000 0:12 0.10 0.00 0.01 1 .00 2.70 2010 0.00 0.02 020 0.11 0.40 0.02 14.00 0.20 2001 0.01 0.010 Do.

Results of um)... um ezpreued in 00010100100 weenie of magnets mm TABLE I A a no I Maonetgualitw inspection for 03$, 01 01 maximum energy product m g: 1100010. 353 53% 100: no 0 and Over and Over Total number of Magnets Tested 3,303 0,000 117 0% E Complete permeameter 10m 420 A110 HG BI (Rm-x10 Permeameter test: a 010 12,000 001 I 010 12,100 040 new No 11 a. 11m. x10 10. From the i'oregoing 00010 10 is obvious that improved magnetic properties are obtained with the m 5. 50 mm silicon alloy which are over and above those ob- 000 0.00 10,000 tained when heat treating an alloy without silicon by the best known practice.

Tables 1:: to XI present data. corresponding to TABLE x:

Influence of silicon and zirconium in restricting the "gamma" phase a Time 01 1000'r. Nm 0000mm 12 000s oz a 10 00001 102: 3%?2... g 0:.-. 1 ]8... 0 000000.012: 0 1000 20%;. i 0405!,03821' n o Under 1% g. Under1% 3135033 0 0n n U 21 "31 n n n l 21. .2 21" n a 2% a n dofiin 009 0 l n n 0.01 01,020 n n W31: 1'-

Table x, which gives the magnetic quality inspection results on magnets cast from heats 1 to 11 inc., containing 0.40 to 0.56% silicon, together with 0.25 to 0.30% zirconium, shows that of a total of approximately 18,000 magnets tested, almost 100% had a maximum energy product (1311) 01' at least 4.5 million, while approximately 95% had an energy product of at least 5.0 million.

All of the magnets or the Table I group cast from heats 1 to 11 inc., had a single-phase microstructure as cast," and were therefore suitable for direct low temperature magnetic treatment below 1800 F., i. e., at 1680 to 1750 F. The castings were cooled from these temperatures in a magnetic field for imparting anisotropic properties.

With reference to Table XI, the tests as to the influence of silicon and zirconium additions in restricting formation of the two-phase microstructure, were carried out at 1950 F., because past experience in working with alloys of this type has shown that a maximum of the gamma precipitate occurs in the shortest time at about 1950 F. The results of Table K1 are also depicted graphically in the accompanying drawing. These test results show that with 0.04% silicon and zirconium, a maximum of 35% of the gamma phase is formed in minutes at 1950 F.; with 0.36% silicon and 0% zirconium, in 30 minutes; and with 0.20% silicon and 0.32% zirconium, only 25% of the gamma phase was formed in 45 minutes. On the other hand with 0.45% silicon and 0.25% zirconium, only from 0 to 5% of gamma precipitation occurred after 45 minutes at 1950 F. These results indicate that the silicon-zirconium combination, is necessary for optimum alpha stability. The accompanying drawing also shows the available energy product of the anisotropically treated magnet for various additions of zirconium and silicon, and after various times of treatment at 1950 F. The graph shows that with 35% gamma precipitation, the energy product, BHmax, is only 2 million; whereas for 0% gamma precipitation, the energy product is 5 million.

I claim:

1. A ferrous alloy adapted for use in permanent magnets, containing: from about up to not more than 20% nickel; about 16 to 30% cobalt; about 6 to 11% aluminum; up to about 7% copper; up to about 1% manganese; up to about 0.1% carbon; up to about 5% titanium; from about 0.15 to 5% of metal selected from the group consisting of the following, (a) zirconium, and (b) zirconium and silicon; and the balance substantially all iron. the metal of said group being present in amount such as to impart to said alloy a substantially single-phase microstructure on heating at about 1600 to 1700" F. and cooling therefrom.

2. A ferrous alloy adapted for use in permanent magnets containing: from about 10% up to not more than 20% nickel; about 16 to 80% cobalt; about 6 to 11% aluminum; up to about 7% copper; up to about 1% manganese; up to about 0.1% carbon; about 0.25 to 5% titanium; from about 0.15 to 2% of metal selected from the group consisting of the following, (a) zirconium, and (b) zirconium and silicon; and the balance substantially all iron, the metal of said group being present in amount such as to impart to said alloy a substantially single-phase microstructure upon heating at about 1600 to 1700' I". and cooling therefrom.

3. A ferrous alloy adapted for use in perma- 14 nent magnets containing: from about 10% up to not more than 20% nickel; about 18 to 30% cobalt; about 6 to 11% aluminum; up to about 7% copper; up to about 1% manganese; up to about 0.1% carbon; about 0.25 to 5% titanium; about 0.15 to 2% zirconium; and the balance substantially all iron, said zirconium being present in amount such as to impart to said alloy a substantially single-phase microstructure upon heating at about 1800 to 1700 F. and cooling therefrom.

4. A ferrous alloy adapted for use in permanent magnets containing: from about 10% up to not more than 20% nickel; about 16 to 30% cobalt; about 6 to 11% aluminum; up to about 0.1% carbon; about 0.25 to 5% titanium; about 0.15 to 1 each of zirconium and silicon; and the balance substantially all iron, said zirconium and silicon being present in aggregate amount such as to impart to said alloy a substantially single-phase microstructure upon heating at about 1600 to 1700 F. and cooling therefrom.

5. A permanent magnet made of a ferrous alloy containing: from about 10% up to not more than 20% nickel; about 16 to 30% cobalt; about 6 to 11% alluminum; up to about 7% copper; up to about 1% manganese; up to about 0.1% carbon; up to about 5% titanium; from about 0.15 to 2% of metal selected from the group consisting of the following, (a) zirconium, and (b) zirconium and silicon; and the balance substantially all iron, said magnet having a substantially single-phase microstructure, and being magnetically anisotropic, and having a BHmu value in the principal direction of at least 4.5 million.

8. A permanent magnet made of a ferrous alloy containing: from about 10% to not more than 20% nickel; about 16 to 30% cobalt; about 6 to 11% aluminum; up to about 7% copper; up to about 1% manganese; up to about 0.1% carbon; about 0.25 to 5% titanium; from about 0.15 to 2% of metal nlected from the group consisting of the following, (a) zirconium, and (b) zirconium and silicon; and the balance substantially all iron, said magnet having a substantially single-' phase microstructure, and being magnetically anisotropic, and having a BHlnax value in the principal direction of at least 4.5 million.

7. A permanent magnet made of a ferrous alloy containing: from about 10% to not more than 20% nickel; about 16 to 30% cobalt; about 6 to 11% aluminum; up to about 7% copper; up to about 1% manganese; up to about 0.1% carbon; about 0.25 to 5% titanium: about 0.15 to 2% airconium; and the balance substantially all iron, said magnet having a substantially single-phase microstructure. and being magnetically anisotropic, and having a Blind: value in the principal direction of at least 4.5 million.

8. A permanent magnet made of a ferrous alloy containing: from about 10% to not more than 20% nickel; about 16 to 30% cobalt; about 8 to 11% aluminum; up to about 7% copper; up to about 1% manganese; up to about 0.1% carbon; about 0.25 to 5% titanium; about 0.15 to 1% each of zirconium and silicon; and the balance substantially all iron; said magnet having a substantially single-phase microstructure, and being magnetically anisotropic, and having a BHma: value of at least 4.5 million.

9. In the manufacture of permanent magnets, the process which comprises: casting into the shape of a permanent magnet. a ferrous alloy containing from about 10% up to not more than 18 20% nickel; about, 16 to 30% cobalt: about 0 t0 11% aluminum: up to about 7% copper; up to about 1% manganese; up to about 0.1% carbon; up to about 5% titanium; from about 0.15 to 5% of metal selected from the group consisting o! the tollowing, (a) zirconium, and (b) zirconium and silicon; and the balance substantially all iron, and thereafter heating the so cast magnet within the range of about 1590" to 1700' F. until thoroughly soaked at said temperature, and allowing said magnet to cool from said temperature while in a magnetic field. and to a temperature sumciently below the Curie point to impart magnetically anisotropic properties to said magnet. and thereafter aging to coercive harden.

10. In the manufacture of permanent magnets. the process which comprises: casting into the shape of a permanent magnet. a ferrous alloy containing from about up to not more than nickel; about 16 to cobalt; about 8 to 11% aluminum; up to about 7% copper; up to about 1% manganese; up to about 0.1% carbon; about 0.25 to 5% titanium; from about 0.15 to 2% of metal selected from the group consisting of the following, (a) zirconium. and (b) zirconium and silicon; and the balance substantially all 583,411

iron. and thereafter heating the so cast magnet within the range of about 1590' to 00 1''. until thoroughly soaked at said temperature, and allowing said magnet to cool from said temperature while in a magnetic field. and to a temperature sumcientiy below the Curie point to impart magnetically anisotropic properties to said magnet, and thereafter aging to coercive harden.

JOHN R. HANSEN.

REFERENCES CITED The following references are of record in the iile 01 this De-tent:

UNITED STATES PATENTS Number Name Date 2,158,019 Jonas Apr. 25, 1089 2,185,464 Howell Jan. 2, 1940 2,285,406 Bieber June 9, 1942 2,295,082 Jonas Sept. 8, 1842 FOREIGN PATENTS Number Country Date 446,894 Great Britain May 4, 1936 Great Britain Dec. 18, 1946

Claims (1)

1. A FERROUS ALLOY ADAPTED FOR USE IN PERMANENT MAGNETS, CONTAINING: FROM ABOUT 10% UP TO NOT MORE THAN 20% NICKEL; ABOUT 16 TO 30 COBALT; ABOUT 6 TO 11% ALUMINUM; UP TO ABOUT 7% OF COPPER; UP TO ABOUT 1% MANGANESE; UP TO ABOUT 0.1% CARBON; UP TO ABOUT 5% TITANIUM; FROM ABOUT 0.15 TO 5% OF METAL SELECTED FROM THE GROUP CONSISTING OF THE FOLLOWING, (A) ZIRCONIUM, AND (B) ZIRCONIUM AND SILICON; AND THE BALANCE SUBSTANTIALLY ALL IRON, THE METAL OF SAID GROUP BEING PRESENT IN AMOUNT SUCH AS TO IMPART TO SAID ALLOY A SUBSTANTIALLY SINGLE-PHASE MICROSTRUCTURE ON HEATING AT ABOUT 1600* TO 1700*F. AND COOLING THEREFROM.

Images