[1]United States Patent Office , 3@1382494 a t e n t e d J u n e 2 3 , 1 9 6 4 3,138,4 94 MUF HOD OF ANNEALING MAGNETIC MATE RIALS Robe.- t E. Bur-liet, Brackenridge, and Donald M. Stewart Tarent um, Pa., assignors to Allegheny Ludium Stee' 5 Corpo ration, Brackenridge, Pa., a corporation of Penns ylvania Filed May 1, 1961, Ser. No. 106,780 7 Claims. (Cl. 148-103) 10 This invention relates to improvements obtainable in magne tic materials by a novel techniqiie in annealing, and relates in particular to a method of effecting a magnetic anneal on soft magnetic core materials. It is a well-established fact that for certain soft mag- 15 netic materials, -@Lnnealing in the presence of an externally applie d magnetic field can cause improvements in certain of their magnetic properties. Such treatment is generally referr ed to as "magnetic annealing" and the effects of such a treatment are to superpose on the niaterial an 20 extra or induced magnetic anisotropy uniaxially so that subse quent magnetization is easy in the forward and revers e directions along this axis, but is increasingly difficult at increasing angles from the axis. The materials in service are magnetized along their axes to take ad- 25 vantag e of the improved magnetic properties. Magnetic anneal ing is not effective for all soft magnetic materials and is known to be Liseful only for the treatment of binary and ternary alloys of the three ferromagnetic eleme nts iron, nickel and cobalt, and some of the silicon- 30 iron compositions. Oiae example of magnetic annealing is the beneficial effect obtained by annealing an alloy known as Supermendu r (49% Fe, 49% Co, 2% V), in the presence of an externally applied magnetic field. It is well known 35 that this material, after being processed in a standard manne r and final annealed in the presence of an externally applie d magnetic field under ultrapure atmosphere conditions , exhibits remarkable improvements in its hysteresis loop characteristics. Such characteristics include a 40 low coercive force (low Hc), a low hysteresis loss, a high maxim um permeabflity (high A max.), a high permeability associated with a high flux density, a high remanence (Br) and a rectangular hysteresis loop with a high flux swing from minus remanence to plus saturation. 45 Similar magnetic properties in varying degrees are obtainabl e by magnetic annealing the other iron-nickel, ironsilicon, iron-cobalt and iron-cobalt-nickel soft magnetic materials. The above-enumerated magnetic properties are those 50 desire d and required for wound cores such as are eniploye d in magnetic amplifiers and in instrument and distributi on transformers. In such applications a magnetic materi al that exhibits a substantially rectangular hysteresis loop of the type effected by the above-enumerated 55 prope rties is ideal for the application required. These materi als and their application in the field of magnetic materi als are well known and will be obvious to one skilled in the art of producing soft magnetic materials. The techniques involved in magnetic annealing of the 60 FeCo-V sys,ems are equally applicable to all soft magnetic materials that are amenable to ma.-netic annealing The method of the present invention is particLIarly ap' plicabl e to any such materials which may be treated in a 65 manne r to effect a high residual induction after being ma.- netized. To conduct magnetic annealing, one must possess special equipment that is capable of annealing the mate2 skilled in the art. All of the known methods being used, however, do involve magnetizing the material while it is being annealed. It obviously would be far more convenient and economical to be able to produce the effect of a magnetic field separately ftom the aniaealing treatment. The equipment required to magnetize such core materials as punched El and EE laminations would be relatively simple if the magn-.tic field did not have to be applied while the laminated structures were in a furnace at high temperatures. It has now been found by applying the method of the present invention magnetic materials such as wound cores or stacked laminations may be treated in a manner to effect good magnetic properties without applying an externally created magnetic field during heat treatinent. In general, the present invention is a method wherein magnetic materials are subjected to a magnetic field outside any specific heating zone in such a manner as to have a residual induction such as is ordinarily associated with hard magnetic materials, and are then annealed at a temperature best suited for magnetic annealing so that the lasting or residual induction acts iii a manner similar to that ordinarily applied by means of an external source during heat treatment, and in consequence causes the material to be magnetically annealed to some extent and to exhibit superior magnetic characteristics to those commonly shown by alloys which have been subjected to annealing alone. The invention is also directed to a method whereby soft magnetic materials are annealed ,Lt a temperature which will effect the magnetic properties of high coercive force coupled with high residual induction and are then left at remanence after applying an externally applied magnetic field at substantially room temperature outside any heating zone, and are then subsequently annealed at a temperature below the Curie temperature which will effect the best magnetic properties during ordinary magnetic annealing. It is therefore an object of the present invention to provide a method of magnetic annealing of iron-nickel, iron-silicon, ironcobalt and iron-cobalt-nickel soft magnetic materials in a manner whereby a magnet,ic field from an external source need not be applied during heat treatment to obtain effects approaching those normally obtained by applying an external field during heat treatment It is also an object of the present invention to provide a method wherein high residual magnetism is induced into soft magnetic materials which are subsequently heat freated to effect improved magnetic properties. It is a further o@ject of the present invention to subject soft magnetic materials to a condition wherein they are susceptible to acquiring a substantially high residual magnetism, subjecting the so-treated materials to a magnetic field so as to induce such residual magnetism, and annealing said materials within a temperature range that will effect maximum magnetic properties. Other objects and advantageous features of the present invention will be obvious from the specification and drawings wherein: FIGURE I consists of superimposed diagrammatic plots of direct current hysteresis loops, such as are familiar to those skilled in the art of magnetic materials. The diagram represents the upper half of the loops only, which a-re direct plots of the data given to illustrate the present invention in each of Tables 1, III, IV and VI. The loop A represents a plot of the test results of the magnetic properties of the third example in Table I; loop B is a plot of the test results of the first example set forth in Table III; loop C is a plot of the test results rial at temperatures in excess of about 1200' F. while 7o of the second example in Table IV and loop D is a plot simultaneously supplying a suitable magnetic field. The of the test results of the third example in Table VI. techniques currently employed are well known to one FIG. 2.is a plot of the direct current hysteresis loop
[2]of the first example of the data given by Table V which has been annealed to effect a mallerial of hi,-h residual magnetism, and FIG. 3 is a triaxial diagram on which are plotted the compositions of binary and ternary iron-nickel-cobalt alloys, the shaded areas of the diagram representing the binary and ternary compositions that are known to respond to magnetic annealing. Magnetic alloys are generally regarded as "soft" magnetic materials such as are subject to the treatment of the method of the present invention, or "hard" such as are illustrated by the permanent magnet materials. The differenc6 in properties of these two materials is in their respective ma,-netic characteristics when subject to an induced ma.-netic field or flux. When the flux-inducing electrical current is stopped, the induced ma,anetic field of soft magnetic materials substantially (thoi-igh not entirely) collapses, while the induced field of the hard magnetic materials remains largely intact so as to effect substantially permanent magnetism. The amount or degree of residual magnetism (B,) that remains in soft magnetic materials after the removal of the induced flux, and the ,amount of field in a reverse direction required to overcome the residual magnetism (H,,), as well as other magnetic characteristics, vary in accordance with the physical and structural condition of the magnetic material. For example, loop A of FIG. 1 shows the first and second quadrants of the hysteresis loop of a 2% V, 49% Co, 49% Fe alloy, known as 2V Permendur, that has been annealed at 1385' F. and rapidly cooled. This diagram represents the magnetic flux density (B) as measured in gausses plotted against the magnetizin- force (H) as measured in oersteds. The techniques employed in obtaining this loop and the interpretation of the B, and H, are well known to one skilled in the art. A material is said to have a square hysteresis loop if it exhibits a high ratio of residual induction (B,) to maximum induction (Bm). However, even a visual examination of the loop A of FIG. I or the data of Table I from which this diagram was derived, will show the material does not have a rectangular loop, and also that it has a relatively high coercive force (H,) and, as previously shown for soft ma,-netic materials, a rectangular hysteresis loop coupled with a low coercive force is one of the goals of ma,-netic annealing. Heat treatment of the Fe-Co-V alloy at temperatures of from about 1450' F. to 1600' F. vastly improves the magnetic properties of the alloy insofar as the H,, and li maximum are concerned, but it impairs the rectangularity of thehysteresis loop as illustrated by loop B of FIG. 1 and the data of Table 111. The hysteresis loop C of FIG. I and the data of Table IV illustrate some of the best properties available from such material after re.-Ular niagnetic annealing in the conventional fashion. These data illustrate that annealing in the presence of an externally applied magnetic field can improve the rectangular characteristics of the hysteresis loop and can cause a remarkable improvement in the D.C. (direct current) coercive force (4 to 5 times lower). Thus, for this alloy, a low energy, rectangular hysteresis loop is obtained by annealin.- in the presence of an externally applied field. This same alloy, the nominal 49% Fe, 49% Co, 2% V alloy, can also be annealed to develop a high energy, high coercive force, high Br hysteresis loop suitable for use in self-biasing, magnetostriction type transducers as illustrated by the data in Table V. Thus, the material exhibits a hi,-h ener-,y rectangularhysteresis loop with high B,, hi-h H, and low tt max. These ,)roperties are illustrated by the hysteresis loop shown by FIG. 2 and the data shown in Table V. In the method of the present invention, advantage is taken of the properties of the material illustrated by FIG. 2. The high B, and hi,-h H, are utilized to effect the 3,138,494 ma,-netic anneal without the application of a flux from an exterior source during treatment. The material annealed at 1022' F. was magnetized (at room temperature) and left at B,, and then reannealed at 1350' F. to 1550' F. and slow cooled. The resultant properties as shown by the hysteresis loop D of FIG. 1 and the data shown by Table VI. As will be observed by an examination of the figure and the data, the effects of magnetic annealing are apparent since the high B,/Bm ratio or squareness of the 10 hysteresis loop are apparent and the coercive force (H,) is low. The i)roperties shown by loop D of FIG. I and Table VI are shown to be better than any obtainable by ordinary heat treatment methods, and although these properties are not as good as those obtained by conven15 tional magnetic annealing, they substantiauy approach the level of such properties, and since this practice eliminates the cumbersome necessity of effecting a magnetic field in an annealin-. furnace, the present method affords a significant advance in the field of magnetic annealing. 20 The method of the present invention is not limited to materials susceptible to acquiring the hysteresis loop of FIG. 2 in that residual magnetism left in any of the Fe-Co, Fe-Co-V, Fe-Ni and Fe-Ni-Co systems, regardless of how slight, will improve the magnetic properties 25 of the heat treated material to some degree. In addition, as will be shown by the data of Table VII, unannealed, as-cold-rolled specimens of the same material were subjected to a relatively high magnetizing force, left at remanence and annealed. The results show clearly that 30 highly stressed material may be subjected to a high magnetizing force to effect residual magnetism that will significantly -improve the magnetic properties during subsequent annealing. The required intensity of the imposed room - tempera35 ture magnetic field must, of course, be sufficient to approach magnetic saturation of the material being treated, or to exceed the knee of the curve obtained when the flux density of the magnetic material is plotted against the 40 niagnetizing force. For example, 10 oersteds are adequate for cold rolled Supermendur alloy which has been annealed at 1022' F. to effect high residual magnetization properties. However, it is preferred to employ a high magnetization field (exceeding 100 oersteds) when processing as-cold-rolled material so as to obtain the high 45 B, or remanence. As started above, soft magnetic materials that are susceptible to magnetic annealing are well known in the art as being certain binary and tertiary alloys of the ironnickel-cobalt system plus electrical grades of silicon steel 50 that contain from about 2% to 6% silicon. The known materials that will respond to applicants' treatment are those materials that are known to respond to ordinary magnetic annealing, and these materials are largely shown diagrammatically by the triaxial diagram of FIG. 3 (re55 produced from "Magnetic Properties of Metals and Alloys," published by the American Society for Metals, Cleveland, Ohio). The shaded areas show all the materials known to respond to magnetic annealing except the silicon iron alloys mentioned above, and therefore 60 show all the alloys known to respond to applicants' treatment, with the exception of silicon iron containing 2-6% by weight silicon. The dark shadings are compositions which show the effect very strongly. All of these alloys may, of course, contain small additions of alloying ma65 terials usually present to affect properties other than magnetic properties. For example, Supermendur alloy contains up to 2.5% V. The heat treatment of the present invention is the same heat treatment employ,- d during magnetic annealing. 70 These temperatures must, of course, be below the Curie temperatlire (preferably at least 2' F. below the Curie temperature) of the alloy being treated, since to equal or exceed the curie temperature will erase the effects of any magnetic field present. However, for the material to 75 benefit from the effects of the residual magnetism, the
[3]3,138,494 5 6 temperature should be within about 300' F. of the Curie TA BLE IR temperature. Consequently, an acceptable range of heat Ann ealed at 1550' F. for 4 Hours, Furnace Cooled treatment for all the materials susceptible to magnetic to 1100' F., and Withdrawn annealing as shown by FIG. 3 is from slightly below the Curie temperature to 300' F. below the Curie tempera5 D.C . ture. The optimum time at temperature depends, of course, Heat B at B at B, from H. from B ,/B@ on the exact analyses of the material being treated, the 1 0H, 100H, 1 00H, 100H, A Max. 100H gauge, etc. A time as short as one-half hour may be Gausses Gausses Gausses Oersteds sufficient, while times of froni about 2 to 6 hours are lo @ usual. 23,154----- 20,800 23,000 9, 700 .45 ----- .42 To secure a material of high B, and high H. such as is shown by FIG. 2 for the Fe, Co, V composition which, as shown, is particularly susreptible to the treatment of the present invention, it is necessary to heat treat the 15 as-coldrolled material;at a temperature wherein some recovery occurs but recrystallization does not occur so that the material will exhibit a high energy rectangular hysteresis loop upon subsequent magnetization. As shown, an ideal temperature @to accomplish this for the 20 49% Fe, 49% Co, 2% V composition is about 1022' F. The following specific examples are given to illustrate the method of the present invention, and in no way limit the claims to the specific embodiments set forth. .014" punched ring samples were subjected to direct current 25 tests in various conditions of heat treatment, typifying both the prior art practice and the practice of the present invention. The data shown by Tables I through IV are illustrative of prior ar-t practices: 30 TABLE I TABLE IV Annealed at 1550' F. for 4 Houtis in P.D. H2 in the Presence of an Externally Applied Circumfereiztial Magnetic Field of 5 to 10 Oersteds, Furnace Cooled to 1100' F. in the Pi-esence of the Field, Withdrawn and Cooled to Room Temperature in the Presence of the Magnetic Field D .C. li cat B at B at B @ from H . froul B ,/B@ 1 0EI, I OOH, 1 00H, I OOH, li Max,. I OOH Gausses Gausses Gausses Oersteds 23,1 52----- l 21,8 00 23,1 00 18,7 00 .205 45,0 00 .81 231 154----- 21,4 00 23,0 00 19,4 00 .192 60,0 00 .85 1 1 1 TABLE V Annealed at 1385' F. for 2 Hours in Pure Dry Hydrogen Annealed at 1022' F. for 4 Hours in Pure Dry Hydroand Fast Cooled by Withdrawing the Annealing Box gen and Withdrawn From the Furnace (.014" Stamped From the Furnace (.014" Stamped Rings) Rin gs) 35 D.C. D.C. TTeat H eat B at IOH, B at IOOH B, from H @ from B at B at Br from H @ from Br/B. Gausses Gausses IOOH, 100H, A May. 10H, IOOH, 100H, 100H, u Max. IOOH Gausses Oersteds 4 0 Gausses Gausses Gausses Oerst eds 23,152 ------- 20,600 22,900 14,700 1.29 7,000 2 3,152----- 1,650 21,400 18,100 20.6 735 .845 23,153 ------- 20,800 23,000 13,000 1.18 7,650 2 3,153----- 1,700 21,400 17,900 20.1 717 .837 23,164 ------- 20,500 23,000 12,800 1.01 6,900 2 3,154----- 1,650 21,300 17,900 20.2 726 .837 TABLE VI .014 Ring Samples Annealed at 1550' F. and Cooled as Indicated Samples Magnetized and Left at Remanence Prior to Annealing B at B at B, from ll.from Heat Treatmeiat IOH, IOOH, IOOH, IOOH, B,/B. Max. Gausses Gausses Gausses Oersteds 23,152-- Reannealed I (fast 21,100 23,000 12,600 .412 .55 24,400 (@ool ed). 23,152-- Not Previously An21,050 23,100 8,600 .460 .37 10,300 -ealed (fast cooled). 23,154-- Reannealed I (furnace 20,700 23,100 14,500 .276 .63 34,900 (@ool ed to 11001 F.). 23,154-- Not Previously Am20,800 23,000 9,700 .462 .42 10,800 iaeale d (furnace cooled to 11001 F.). Annealed previously at 10221 F. TABLEII It may be observed from the data of Table I and from -4nnealed at 1550' F. for 4 Hours and Fast Cooled by the diagrainmatic representation of Table I as repreWithdrawing the Box From the Furnace 65 sent ed by the hysteresis loop A illustrated by FIG. 1, the induction (B), in gausses, at a magnetizing force of D. C. IO H and IOOH, in oersteds, is reasonably high for rings Heat B at B at B, from H @ frotn B,/B annealed at 1385' F. and fast cooled. Also, the residual IOH, IOOH, IOOH, IOOH, Max. IOOY 70 induction (B,) from 100H is relatively high@ hbwever, Gausses Gausses Gausses Oersteds the coercive force (H, @) is undesirably high and the maxi mum permeability (1&) is relatively low. The shape 23,152 ----- 21,100 23,000 7,000 .61 9, 090 .30 of the hysteresis loop is not particularly desirable since 23,153----- 20,950 23,100 6,400 .61 8,940 .28 23,154----- 21,000 23,100 6,900 .58 9,200 .30 it cannot be said to be rectangular, and the coercive 1- I 75 force (averageing 1. 14) is high.
[4]3,138,494 17 It is shown by Table 11 that heating at a higher temperature (1550' F.) for a longer period of tin@e effects substantially the same induction at hi.-h fields as the 1385' F. treatment. However, the Iiigher temperature and Ion.-er time anneal advantageously lower the coer5 cive force, but also lower the residual induction (B,) which, as stated above, are undesirable. Also, as illustrated, the ratio of the residual induction (B,) to the maximum induction (B,,,) at IOOH is low. The advanta,-es of magnetic annealing are demon10 strated by the data shown by Table IV and as illustrated by loop C of FIG. 1. The induction levels at 10H and IOOH remain substantially the same wMIe the residual induction has doubled over the similarly aiinealed material shown by Table 111, and the coercive force is less 15 than half that shown by the previous data. As shown by loop C of FIG. 1 and the B,IB. ratio, the hysteresis loop is nearly rectangular. The data shonvn in Tables V throu.-h VIII clearly show the advantages obtained by employing the method of the 20 present invention. mum is very low. The real significance of this data is th.- high r,-sidual induction and very bigh co.-rcive force. Such a loop and its accompanying H,, demonstrate the prop,.rties of a relatively hatd ot permanent magnetic mat,,i,l. It is the residual indiiction that provides the ma.-netic force that the applicants have found to be an adequate substitute for the magnetic field generally externally applied during magnetic annealing. The data shown by Table VI show several samples that have been treated substantially in accordance with the method of the present invention (examples labeled reannealed). These samples were among those tested in the manner shown by Table V (subject to a magnetic field during the measurement of magnetic properties) and then re-annealed and tested, the results of which are shown by Table VI and loop D of FIG. 1. Since the rings possessed considerable residual niagnetism when heated to temperatures below the Curie temperature but above 1500' F., they, in effect, created their own magnetic field for magnetic annealing. The data show B, TABLE Vil .014" Ritzg Sattiples as Cold Rolled Plits 1550- F. ,4ii,ieal and Sloiv C6ol t,7 11001 F. As Cold Rolled Heat B at 1077H, A Max. Gausses 23,152 -------- 23,500 so 2.',153 -------- 23,500 23,1&4 -------- 23,500 85 "I .014" Riizg San7ples as Cold Rolled Plits 1550' F. Aittieal a7id Slow Coot to 1100' F. ANNEALED B at D,, H@, B at B,, H@, B at B,, H,, Ileat 1011, Gausses Oersteds B,/BIO 501T, Gausses Oersteds B@/B5O IOOH, Gausses Oersteds B,/BICO Max. Gausses Gausse Gausses 23,152 ------ 21,100 14,000 .38 .692 22,800 14,500 .38 .64 23,000 14,400 @38 63 31,200 23,153 ------ 20,700 15,100 .36 .729 22,600 1,5,000 .36 .67 22,900 15,300 .37 .67 33,500 23,154 ------ 20,900 14,100 .39 .675 22,700 14,200 .39 .63 23,000 14,200 .39 .62 32,800 TABLE VIII .OJ4" Riizg Saiiiples Aiiizealed at 1022' F., Tested aiid Left at Br, Reanizealed at 1550' F. aiid Slo)v Cooled to 1100' F., Retested AFTER 10221 F. ANNEAL B at B, from H @ from Heat 100H , iooii, iooir, Br/B . Gaus ses Gaus ses Oersteds 23,152 ------ 21,300 18,000 2 0.8 .85 23,153 ------ 21,300 17,800 2 0.4 .84 23,154 ------ 21,200 17,700 2 0.7 .84 1 L@TAGNE@ TIZED+1550- F.ANNEAL B at B,, H@, B at B,, H,, B at B,, H,, Heat ioii, Gausses Oersteds B,/B 10 @OH, Gausses Oersteds B,/B6O 100H, Gausses Oersteds B,/BlOO g Max. Gausses Gausse s Gausses 23,152 ------ 20,900 14,200 .37 .68 22,600 14,200 .37 63 22,900 14,200 .37 62 30,000 23,153 ------ 20,700 15,500 .36. . 75 22,600 15,500 .37 63 22,900 15,500 .37 .68 40,000 23,154 ------ 20,500 14,100 .31 ;69 22,500 14,100 .32 .63 22,800 14,200 .32 .62 36,900 properties that are significantly higher than the samples similarly heated, but not subjected to the prior 1022' F. anneal or a magnetic remanence (examples entitled "Not Previously Annealed"). The B,/Bm at IOOH, and hence the shape of the hysteresis loop, does not indicate a loop The data shown by Table V, and as illustrated by the liysteresis loop shoivn by FIG. 2, illustrates the effect o'L a low temperature anneal (1022' F. for four hours). Although, as shown by FIG. 2 and the B,IB., the hysteresis loop is very rectangular, the permeability maxi- 75
[5]th-,Lt is perfectly rectangular; however, the shape of the loop is superior to any of the prior treated material except that actually magnetically annealed. Additionally, the increase in residual induction at IOOH and the permeability maximum more than make up for the loss in rectangularity over the properties shown by the data of any of the previously treated samples, with the exception of that shown by Table IV. The data shown by Table VII show the magnetic properties of ring samples that were subjected to a magnetic field of approximately 1077H in the cold rolled, room temperature condition. After annealing (1550' F.) and slow-cooling, magnetic properties were measured at IOH, 50H and IOOH. As may be observed, the coercive force is low, the residual magnetism and permeability are high and the hysteresis loop is reasonably rectangular in shape. The data shown by Table VIH show the magnetic properties of additional ring samples annealed at 1022' F. to effect properties similar to those shown by the examples of Table V. At the conclusion of testing at 100H, the samples were left at remanence, annealed at 1550' F. for four hours and slow-cooled. The magnetic properties at IOH, 5,OH and IOOH are shown to be substantially similar to those shown by Table IV and equivalent to those of Tables VI and VII. We
