US3615916A

US3615916A - Manufacture of permanent magnets

Info

Publication number: US3615916A
Application number: US794158*A
Authority: US
Inventors: Stuart Walter Ker Shaw; Derek Jim Palmer
Original assignee: International Nickel Co Inc
Current assignee: Huntington Alloys Corp
Priority date: 1968-01-30
Filing date: 1969-01-27
Publication date: 1971-10-26
Anticipated expiration: 1988-10-26
Also published as: NL6901344A; GB1204586A; DE1903976A1; BE727676A; FR2001000A1; CA939935A; ES363070A1; CH511287A; SE354488B

Abstract

An improved process of forming ingots of iron-aluminum-nickelcobalt-titanium alloys, with or without copper and columbium, with a columnar crystal structure, useful for making magnets, comprising introducing sulfur, aluminum and titanium into a carbon-containing deoxidized melt of the other alloy constituents.

Description

United States Patent Inventors Priority Stuart Walter Ker Shaw Sutton Coldfield;

Derek Jim Palmer, Solihull, both of England Jan. 27, 1969 Oct. 26, 1971 The International Nickel Company, Inc. New York, N.Y.

Jan. 30, 1968 Great Britain MANUFACTURE OF PERMANENT MAGNETS 29 Claims, 1 Drawing Fig.

US. Cl 148/ 103, 75/135,148/31.57

Int. Cl H011 1/04 Field of Search 75/135;

[56] References Cited UNITED STATES PATENTS 3,226,266 12/1965 Jesmont et a1 148/3 3,314,828 4/1967 Harrison 148/31.57 3,350,240 10/1967 l-liguchi et a1.. 148/1.6 3,428,498 2/1969 Heimke 148/101 3,432,369 3/1969 Naastepad 148/101 3,450,580 6/1969 Hadfield et a1. 148/31.57 3,498,851 3/1970 Shuin et a1. 148/101 3,520,677 7/1970 Komaki et a1 75/10 Primary Examiner-L. Dewayne Rutledge Assistant Examiner-E. L. Weise Attorney-Maurice L. Pinel ABSTRACT: An improved process of forming ingots of ironaluminum-nickel-cobalt-titanium alloys, with or without copper and columbium, with a columnar crystal structure, useful for making magnets, comprising introducing sulfur, aluminum and titanium into a carbon-containing deoxidized melt of the other alloy constituents.

MANUFACTURE OF PERMANENT MAGNETS it is known that it is desirable in making permanent magnets of iron-aluminum-nickel-cobalt alloys, with or without copper and columbium, to cast the alloys in such a way that a columnar crystal structure is produced, since this structure offers the greatest benefits of magnetic anisotropy when the alloy is finally magnetized.

Careful adjustment of the composition of the alloys is essential if they are to yield good columnar growth during solidification and if the overall magnetic properties of the alloys are not to be impaired. It has long been known that the presence of titanium leads to an increase in coercivity in these alloys and that columbium may replace part of the titanium present, though in the absence of titanium columbium will not impart coercivity of the high order obtainable with titanium. Titanium and columbium, however, have an adverse effect on the growth of columnar crystals.

it is also known, as described in a paper by J. Harrison and W. Wright published in Cobalt," No. 35 June 1967, pages 63-68 that the addition of sulfur enables a columnar structure to be obtained despite the presence of titanium in amounts up to 8 percent. Nevertheless the amount of sulfur required, namely 1 percent or more, is very detrimental to the magnetic properties of the castings. More recent work described in a paper by Y. Kamata and T. Anbo in Nippon Kinzoku Gakkai Zasshi, 1967 3l 1,053-1057 has shown that the product of the percentages of aluminum and titanium (A l=Ti) is an important factor, and that as the ratio of this to sulfur increases so does the difi'rculty or even the impossibility of obtaining a columnar structure. Kamata and Anbo also showed that if carbon is introduced as well as sulfur, columnar structures can be produced with higher values of (Al=Ti).

Harrison and Wright added the sulfur to the base charge which they melted. Kamata and Anbo said they added the sulfur before the carbon.

An object of the invention is to provide an improved method of making ingots of iron-aluminum-nickel-cobalttitanium alloys with a columnar crystal structure.

Another object is to provide permanent magnets of ironaluminum-nickel-cobalt-titanium alloys having a columnar crystal structure.

Other objects and advantages of the invention will appear from the following description.

The invention is based on the discovery that if appropriate steps are taken in the production of ingots of the alloys it is possible by a simple and industrially practical process to obtain columnar structures in ingots of alloys in which Kamata and Anbo obtained only equiaxed structures, and greater columnar lengths than in ingots produced by the methods described by either Harrison and Wright or Kamata and Anbo.

The accompanying drawing shows a chart depicting the aluminum content times titanium content against the sum of sulfur content plus carbon content for iron-nickel-aluminumtitanium magnet alloys. The alloys in which such columnar structures can be produced are those containing from 5 to ll percent aluminum, from 7 to 25 percent nickel, from 20 to 55 percent cobalt, from l to ll percent titanium, from 0 to 10 percent copper and from 0 to 4 percent columbium, the balance except for impurities being iron. As is well known, commercial sources of columbium generally contain some tantalum, and all references in this specification to columbium are to the total quantity of columbium and tantalum.

in producing an ingot of such an alloy with a columnar structure according to the invention the essential steps comprise forming a carbon-containing deoxidized melt of the iron, nickel, cobalt and any copper and columbium in the alloy from a base charge part of which may be scrap from a previous melt, but without titanium or aluminum except any introduced as scrap, and adding the titanium and aluminum (or the remaining titanium and aluminum) and sulfur to the melt.

The melt may be formed entirely from virgin materials or in part from scrap produced in a previous process according to the invention. in the latter case, it will of course contain both titanium and aluminum. The necessary steps in the process depend to some extent upon the presence or absence of scrap, and the amount of any scrap that is present.

The precise mechanism by which the carbon and sulfur promote the formation of columnar crystals is uncertain, but it appears clear that columnar growth is favored by minimizing I the formation of nuclei, in particular of oxides of titanium and aluminum and of titanium nitride, that' tend to nucleate equiaxed crystals, and by removing such nuclei or rendering them ineffective after they have been formed.

Because aluminum and titanium are readily oxidized, neither is included in the base charge except as scrap, in which form they are not readily oxidized. in adding them to the deoxidized melt, we find it desirable to add the aluminum before the titanium.

We believe that carbon in the melt tends to limit the number of nuclei formed on the addition of either titanium or aluminum. If the melt is partly composed of scrap, the aluminum in the scrap acts as a deoxidant. If the melt is wholly composed of virgin raw materials, oxygen in it is removed by the carbon as gaseous carbon monoxide so that less nuclei are formed when the aluminum and titanium are added. The sulfur exerts an extremely favorable action, possibly by forming titanium sulfide, which may act by enveloping oxide and nitride nuclei and so rendering them ineffective.

Considering first processes in which the melt is formed wholly of virgin raw materials, the preferred procedure is as follows. A base charge of the iron, nickel, cobalt and any copper and columbium is melted, the melt is deoxidized, advantageously by small amounts of aluminum or silicon, and the aluminum and titanium are added, the carbonbeing added to the base charge or to the melt before the addition of the titanium, and the sulfur being added to the deoxidized melt with or after the titanium. The reason why the melt is deoxidized before the main addition of aluminum, instead of allowing the aluminum to effect the deoxidation, is that the amount of aluminum required for the deoxidation is variable and it is desirable to introduce as accurate an amount as possible into the final alloy.

in order to facilitate control of the carbon content, it is best to add the carbon to the melt after deoxidation and before the addition of aluminum.

When the melt is composed partly of scrap, no special step of deoxidation is required. Provided that the proportion of scrap and its carbon content are high enough, all the carbon required may be introduced by way of scrap. Otherwise carbon should be added to either the base charge or the melt.

As indicated above, the invention enables the columnar length in any given ingot to be longer than it would have been if the ingot had been produced by the methods of the prior art. However, our object, of course, is to produce the greatest possible columnar length. The numerous ingots we have produced, a number of which are described in detail below, were made by pouring the treated alloy at a temperature of l,650 C. into a cylindrical refractory mold 6.5 inches high and 6 inches in external diameter with a conical opening leading to an internal cylindrical cavity 4.5inches high and 1.875 inches in diameter and open at the bottom, the mold with this open cavity being placed on a water-cooled copper base. In each case the mold was preheated to l,l50 C., and an exothermic compound was placed on the top of the molten metal immediately after it was poured. Generally speaking, under these particular conditions of casting, the formation of columnar crystals at least 2 inches high is regarded as satisfactory and 2.5 inches or more is very satisfactory.

While it is essential that there is carbon in the alloy poured into the ingot mold, it is known that carbon has an adverse effect on the magnetic properties of the alloys in question, and accordingly there should be as little carbon as possible. For satisfactory columnar crystallization there should be at least 0.02 percent carbon in the melt at the time of casting, which generally requires the addition of at least 0.03 percent carbon to the melt. So far as obtaining columnar length is concerned, the carbon content of the melt may be as high as 0.25 percent carbon, but the length falls off rapidly when this amount of carbon is increased.

When carbon is added otherwise than by way of the scrap, it may be incorporated in the main melt before deoxidation either as a Constituent of the base charge, e.g., as graphite or as an iron-carbon alloy, or as a separate addition to the melt, for example by immersing a carbon rod in the melt until the boil" caused by the reaction with oxygen in the melt has subsided. The latter process generally results in the incorporation of 0.03 to 0.1 percent of carbon in the melt, but it is difficult to control accurately, and preferably the carbon is added to the melt after deoxidation, e.g., as an iron-carbon alloy. Generally speaking, an addition of 0.05 percent carbon after deoxidation is satisfactory. When the carbon is added as part of the base charge, the amount used is preferably increased, e.g. to at least 0.1 percent, to compensate for the greater loss by oxidatron.

The minimum amount of sulfur introduced into the melt whether as part of the scrap or as the added sulfur is 0.2 percent, but this amount is effective only if the combined contents of titanium, aluminum and columbium are not too high, and in particular if the titanium content does not exceed 6 percent. The amount of sulfur required is mainly affected by the present in the base charge the melt is subjected to vacuum to remove the carbon monoxide formed, and the sulfur is added to the melt with or after the titanium. The use of vacuum melting generally has the advantage that less sulfur is required to produce a satisfactory columnar length at any given titanium content.

It will be appreciated that in order to produce a columnar structure it is necessary to correlate all the variables that have been discussed above.

The individual effects of some of these variables will now be described.

First, the effect of varying the carbon content is shown by a series of tests. In these tests the object was to make an alloy (Alloy X) having the nominal composition 5 percent titanium, 8 percent aluminum, 30 percent cobalt, 15 percent nickel and 3 percent copper, the balance being iron. A base charge of the iron, nickel, cobalt and copper was melted in air and deoxidized with silicon, carbon was added as an iron-carbon alloy containing 3.6 percent carbon, the titanium, aluminum and 0.2 percent of sulfur as ferrous sulfide were immediately added, and the melts were held for 10 minutes at l,650 C. and then cast. A similar casting was made with no addition of carbon for purposes of comparison.

TABLE I Chemical composition wt. percent, balance Fe Percent Columnar Ingot No. 0 added length (in.) 0 Co Ti Al Cu Ni S Si 0 0 0. 019 29. 2 4. 8 7. 3. 2 17. 3 0. 19 0. 2 0. 02 1 0. 018 29. 0 4. S 6. Si 3. 1 17. 3 0. 1!! 0. 2 0. 055 2. 0 053 29. 9 5. 0 7. 5 3. 0 15. 2 0. 21 0. 2 0. 115 2 Q. 061 30. 0 4. 9 7. 45 3. 0 15. 1 0. 20 0. 2 0. 16 3 0. 135 29. 3 4. 5 7. 3 3. 0 14. 3 0. 10 0. 2

content of titanium and columbium. Thus to obtain improved columnar crystal growth in alloys containing 8 percent aluminum, at least 0.25 percent of sulfur must be added if the titanium content is 7.5 percent; at least 0.4 percent sulfur must be added if the titanium content is 9.5 percent; and at least 0.6 percent sulfur if the titanium content is 10 percent or more. The amount of sulfur added should be as small as is consistent with obtaining the desired columnar crystallization, since sulfur adversely affects the magnetic properties of the alloy by combining with titanium to form titanium sulfide. Although as much as 1 percent or even l.2 percent sulfur may be added, it is generally unnecessary to add more than 0.8 percent sulfur. The sulfur may be added in the form of any convenient sulfur-containing compound, e.g., is ferrous sulfide.

To ensure that a columnar structure is produced, it is necessary to give time for the various reactions to take place. In general, we find it desirable to hold the melt for a period of at least 5 minutes, and preferably at least 10 minutes, after the final addition before casting the melt. When carbon is added after deoxidation, the melt is also preferably held for at least 2 minutes after the carbon addition and before any aluminum or titanium is added.

Another factor affecting the columnar length is the casting temperature. It is well known that the extent of columnar growth increases with this temperature, and although this may be as low as 1,550 C. it is preferably 1,650" C. Since as the casting temperature increases, so do the wear of the refractory linings and the loss of the more highly reactive elements, and in particular titanium and aluminum, we prefer not to exceed 1,700 C.

The loss of heat through the sides of the ingot mold should be reduced to a minimum, and accordingly the use of exothermic molds is advantageous.

For ease of operation we prefer to carry out the process in air, but the ingots can be produced by vacuum melting. In this case it is desirable to ensure that carbon is present in the base charge, either as such or as a constituent of scrap, and no specific step of deoxidation is required. Thus with carbon The carbon and sulfur analyses were carried out on specimens taken from the top of the cylindrical part of each ingot, and becat-ise of segregation in the ingots the contents given are likely to be higher than in the columnar zone.

The effect of varying the amount of sulfur added is shown by the results of a second series of tests in which melts of an alloy of the same nominal composition were made by air-melting together the iron, nickel, cobalt and copper, incorporating from 0.03 to 0.l percent of carbon by immersing a carbon rod in the melt until boiling ceased, deoxidizing the melts with silicon, adding the titanium and aluminum and varying amounts of sulfur as ferrous sulfide (including a control test with no sulfur addition), holding the treated melts for 10 minutes at l,650 C. and then casting. The results are set forth in Table II.

These results show that with an alloy containing 5 percent titanium there is no need to add more than 0.2 percent sulfur. As the titanium content increases, however, the sulfur content must be increased. This is shown by th results in table Ill. The alloys to which table lll relate all nominally contained 15 percent Ni, 3 percent Cu and 8 percent A]; the Ti and Co contents are given in the table and the balance was iron. As the titanium content was increased the cobalt content was also increased in accordance with normal commercial practice. All the alloys were made as described in connection with table II, the deoxidation being effected with silicon, and they were cast at l,650 C. after holding for 10 minutes.

TABLE III Analyzed composition, percent Percent Columnar S added Ti C A1 S C TiXAl S+C length (in.)

O. 2 9. 0 42.5 7. 65 0.08 0.018 68. 7 O. 098 0 0.35 8. 75 42. 4 7. 35 0.16 0.045 64. 2 0.205 O 0.4 8. 7 42. 4 7. 7 0. 38 0. 043 *67. 0 *0. 423 3.0 0. 8 8. 5 41. 7 7. 25 0. 72 0. 047 61. 6 0. 767 2. 7

It will be seen from the results obtained with Ingot 42 that with enough sulfur satisfactory columnar lengths can be obtained in an alloy containing as much as 10 percent titanium, that is to say, an alloy in which Kamata and Anbo failed to produce a columnar structure.

The addition of columbium to the alloy has an effect similar to increase in the titanium content in requiring the addition of more than 0.2 percent sulfur in order to produce substantial columnar growth. This is shown by results obtained with the same alloy (Alloy X) containing 5 percent titanium to which table I] relates, treated with carbon and sulfur in the same way and cast at 1.650 C. after holding for 10 minutes. The results are shown in table IV.

TABLE IV 45 Sulfur Colum bium Columnar No. added (2) (Babes) The aluminum content also plays a part in determining the minimum sulfur content required to give a columnar structure. in all the examples so far given the aluminum content was nominally 8 percent. On decreasing this content to 7 percent in alloys nominally containing 15 percent and 3 percent Cu with varying amounts of Co, the balance being iron, improved columnar growth is obtained with only 0.2 percent sulfur when the titanium content is up to at least 6.5 percent, the melting and treatment procedure being as described for table II and the alloy being cast at 1,650 C. after holding for 10 minutes. This is shown by table V.

TABLE V lngot Sulfur Ti Co Columnar No. added (/E) ('2) Height (inches) The effect of varying the times for which the melt is held after adding carbon and before adding the aluminum and titanium and for which the fully treated melt is held before casting is shown by the results in table VI. These relate to a series of tests in which melts of the nominal composition of Alloy X were made by air-melting together the Fe, Ni, Co and Cu, deoxidizing with Al or Si, adding 0.05 percent of carbon as an iron-carbon alloy containing 3.6 percent carbon, holding for periods up to 6 minutes, adding the aluminum, the titanium and 0.2 percent of sulfur as ferrous sulfide, again holding for various periods up to 10 minutes, casting at l,650 C.

TABLE VI Holding time before adding Holding time Columnar Ingot Ti, Al and S before castin length No. Deoxidant (min.) (min. (inches) 51 Al 0 0 1.5 A1 0 10 2. 5 2 0 1. 25 2 3 2.0 2 10 2. 6 0 1. 0 6 10 2. 75 0 10 2. 5 1 10 3. 0 2 0 1. 5 2 1 1. 5 2 3 3. 0 2 10 3.0 6 0 1. 75 6 10 3. 0

Table V] shows that it is generally desirable to hold the treated melt for some time after adding the sulfur and before casting. It also shows that better results may be obtained by holding after the carbon addition and before adding the aluminum, titanium and sulfur; as well as increasing the extent of columnar crystallization, this also improves the sharpness of the texture of the columnar crystals obtained, i.e., the closeness with which their direction corresponds with that of the longitudinal axis of the casting.

It is advantageous to make the titanium content high and the fact that satisfactory columnar structures can be obtained in alloys nominally containing 7 percent or more titanium when made by the preferred method according to the invention is shown by table Vll. All the ingots to which this table relates were made by air-melting virgin materials, deoxidizing the melt, adding 0.05 percent carbon as an iron-carbon alloy 5 containing 3.6 percent carbon, holding the melt for 2 minutes,

ing after the final addition to the melt may be unnecessary.

When the process is carried on in a vacuum furnace, it is preferred to effect melting under an inert gas, then subject the melt to vacuum, add the aluminum, titanium and sulfur under 5 an inert gas, and again subject the melt to vacuum. With the use of virgin raw materials the process may advantageously be carried out in detail in the following way.

A charge containing the iron, cobalt, nickel and copper, together with graphite, is placed in a crucible in the vacuum TABLE VII Analysed composition, wt. percent Added Colurnnar Ingot percent length TiX N 0. C CO Ni Cu Ti Al S S (inches) A1 S+C 0. 062 35. 3 15 3. 2 6. 9 7. 5 0. 13 0. 4 2 51. 7 .192 0. 024 36. 8 12. 5 2. 9 7. 2 5. 95 0. 26 0. 4 3. 4 42. 8 284 0. 037 34. 2 15.0 3.1 7. 4 6. 60 0.28 0. 4 3. 2, 3. 7 48. 8 .317 0. 031 32. 8 16. 8 3. 6. 9 6. 85 0. 29 0. 4 2. 8, 2. 9 47. 2 321 0. 015 37. 3 12. 8 3. 2 8. 3 6. 40 0. 41 0. 6 2. 3, 2. 3 53. 2 425 0. 018 36. 2 14. 9 3. 0 7. 9 6. 40 0. 32 0. 6 2. 8, 2. 7 50. 33S 0. 027 36. 0 17. 2 3. 2 8. 2 6. 83 '0. 43 0. 6 2. 8, 3. 2 56. 1 457 0. 030 39. 5 13. 1 3. 0 9. 2 7. 0. 52 0. 8 2. 4, 2. 4 65. 7 550 0. 010 38. 0 17. 3 3. l 9. 0 6. 65 0. 47 0. 8 3. 0, 3. 4 59. i 480 75 0. 071 40 14. 8 3. 1 9. 6 5. 83 0. 89 0. 8 2. S 63. 8 961 Excellent results can also be obtained in processes in which virgin raw materials are used when the carbon is introduced into the base charge before the deoxidation. it is desirable that the amount of carbon so introduced should be at least 0.05 percent and is preferably about 0.1 percent. The results obtained in such processes in which the deoxidation. after the addition of carbon was effected with aluminum, some melts being cast at once, and some being held for It) minutes at l,650 C. before casting, are shown in table Vlll, which also relates to alloys of the composition of Alloy X.

TABLE VIII Holding time columnar 0 added before casting length Ingot No. (percent) (min.) (inches) furnace. The pressure in the furnace is reduced to 2 microns of mercury and argon is admitted to a pressure of 100 mm. of mercury. The charge is melted under this pressure of argon, and then the pressure is again reduced to 2 microns of mercury so as to remove gas from the melt. in view of the difficulty of making any addition under vacuum, argon is next readmitted to a pressure of l00 mm. of mercury, the aluminum, titanium and sulfur are added and the pressure is once more reduced to 2 microns. The melt is held at this pressure for 10- minutes at 1,650 C. before being cast. To avoid contact between the molten metal and oxygen or nitrogen before and during pouring. the casting is effected in the furnace. it is desirable to avoid lateral heat loss from the cast metal, and therefore, unless the furnace is so constructed that a preheated mold can be introduced without the admission of air, the melt is advantageously cast into an exothermic mold with a bottom chill. Argon is admitted to the furnace at a pressure of 700 mm. of mercury before the casting, since otherwise the violent evolution of gas from the exothermic mould in vacuum might destroy the mold. When the metal has been cast, it is desirable to open the furnace as soon as possible and put an exothermic compound on top of the molten metal, which is then left undisturbed until solidification is complete.

Results obtained in some ingots cast as just described are shown in table X. These results when compared with those in table lll show that vacuum melting enables a satisfactory columnar length to be produced with less sulfur at any given titanium content.

TABLE IX Percent Wt. Chemical composition, balance Fc,

scrap percent Columnar wt. percent Ingot carbon length 0. charge added (inches) 0 00 T1 Cu Ni S Al 82 15 0.05 3.3 0.054 39.0 7.3 2.75 14.20 0.4 *7 83 15 Nil 0 0.009 39.1 7.25 2.75 14.35 0.4 *7 84 30 0.05 3.3 0.060 30.8 7.2 2.75 14.35 *0.4 *7 85 30 Nil 0 0.013 39.1 7.0 2.6 14.30 "0.4 *7 86 50 0.05 3.5 0.062 39.5 7.6 3.05 14.50 0.4 '7 87 50 Ni! 3.5 0.029 *8 *3 "14.5 0.4 7

Nominal content. P

TABLE X Percent Chemical composition, wt. percent, Caddcd Percent Colunmar balance Fe Ingot to base S length 0. charge added (inches) C 00 Ti Cu Ni S Al 0.05 0.2 2 0.053 35.8 7.5 1.7 14.9 0.14 7.0 0.10 0.25 3 0.050 40.2 8.0 2.1 14.75 0.25 6.5 0.05 0.4 2 0.046 30.5 10.3 1.0 15.2 0.24 7.0 0.05 0.4 3.3 0.009-395 10.0 2.0 15.1 0.15 7.!)

This table shows that with the particular scrap used very satisfactory columnar lengths can be obtained without any addition of carbon if half the charge is scrap. lf, however, less than 40 percent of the charge is scrap, an addition of carbon is required. These results also show that when scrap is used hold- The invention presents striking advantages. One is the possibility of using scrap of the very expensive alloys in question, and in fact we surprisingly find that once an alloy has been cast into ingots according to the invention further ingots with satisfactory columnar lengths can be nrndurerl I, rnmnhinn scrap without any additional virgin materials.

Another considerable advantage is the possibility of obtaining satisfactory columnar lengths in alloys of such aluminum and titanium contents that Kamata and Anbo found invariably led to equiaxed structures. This advantage is clearly shown by the accompanying drawing, in which (following Kamata and Anbo) the total content of sulfur plus carbon is plotted against the product of the aluminum and titanium contents. Open circles show the compositions having columnar structures, and black circles show compositions in which Kamata and Anbo obtained only equiaxed structures. The line A marks the boundary drawn by Kamata and Anbo between their columnar and equiaxed structures, and the line B shows the approximate boundary of compositions in which columnar structures can be produced according to the invention. The alloys indicated by asterisks in tables Ill, VII and X have compositions in the region between lines A and B.

Metal from the columnar portions of ingots produced in accordance with the invention may be magnetized by any conventional method to form magnets, and the invention specifically includes these.

It is a further advantage of the process of the invention that magnetic properties actually better than described by Kamata and Anbo can be obtained in such magnets. As an illustration, part of Ingot No. 68 was magnetized and tested. This part was produced by removing and discarding the bottom half-inch of the ingot, and then spark-machining a cylindrical piece 0.5 inch in diameter and 1.3 inch high from the columnar zone. This piece was solution-heated for 1 hour at 1220 C., oilquenched, and placed in a magnetic field of 3,800 oersteds and Anbo 12 minutes at 830 C., aged for 32 the titanium increased. a 570 C., and allowed to cool.

The magnetic properties of the piece thus magnetized were ascertained and are compared in table XI with those indicated by Kamata and Anbo as obtainable in an alloy containing 7 percent titanium, a titanium content at which Kamata and Anbo found that the remanence and energy product of their alloys was decreasing as the titanium increased.

Although the present invention invention has been described in conjunction with preferred embodiments, it is to be understood that modifications and variations may be resorted to without departing from the spirit and scope of the invention as those skilled in the art will readily understand. Such modifications and variations are considered to be within the purview and scope of the invention and appended claims.

We claim:

1. A process in which an ingot of an alloy containing from 5 to l 1 percent aluminum, from 7 to 25 percent nickel, from to 55percent cobalt, from 1 to l 1 percent titanium, from 0 to 10 percent copper and from O to 4 percent columbium, the

balance except for carbon, sulfur and impurities being iron, is

produced with a columnar crystal structure, which comprises forming a carbon-containing deoxidized melt of the iron, nickel, cobalt and any copper and columbium from a base charge part of which may be scrap form a previous melt, but without titanium or aluminum except any introduced as scrap, and adding the titanium and aluminum (or the remaining titanium and aluminum) and sulfur to the melt.

2. A process according to claim 1 in which the melt is formed wholly of virgin raw materials, a base charge of the iron, nickel, cobalt and any copper and columbium is melted, the melt is deoxidized, and the aluminum and titanium are added, the carbon being added to the base charge or to the melt before the addition of the titanium and the sulfur being added to the deoxidized melt with or after the titanium.

3. A process according to claim 2 in which the carbon is added to the melt after deoxidation and before the addition of aluminum.

4. A process according to claim 2 in which the melt is held for a period of at least 5 minutes after the final addition before being cast.

5. A process according to claim 4 in which the holding period is at least l0 minutes.

6. A process according to claim 2 in which the carbon is added after the deoxidation and the melt is held for at least 2 minutes before any aluminum or titanium is added.

7. A process according to claim 3 in which the aluminum is added before the titanium.

8. A process according to claim 1 in which a base charge of the iron, nickel, cobalt and any copper and columbium, together with carbon, is subjected to vacuum and the aluminum and titanium are added, the sulfur being added to the melt with or after the titanium.

9. A process according to claim 8 in which the melt is formed wholly of virgin raw materials.

10. A process according to claim 8 in which the melting is effected under an inert gas, the melt is then subjected to the vacuum, the aluminum and titanium and sulfur are added under an inert gas, and the melt is again subjected to vacuum.

11. A process according to claim 1 in which the melt is formed from a base charge composed partly of scrap produced in a previous process according to claim 1 and partly of virgin raw materials, except for virgin titanium and aluminunijfid the remaining tita niu m and aluminum andadditional sulfur are added to the melt thus formed.

12. A process according to claim 11 in which the scrap amounts to less than 40percent of the base charge by weight and carbon is added to the base charge or the melt.

13. A process according to claim 11 in which the addition of sulfur is made after the addition of titanium and aluminum.

14. A process according to claim 11 in which the addition of aluminum is made before the addition of titanium.

15. A process according to claim 1 in which metal from the part of the ingot having a columnar crystal structure is magnetized to form a permanent magnet.

16. A process according to claim 4 in which the aluminum is added before the titanium.

17. A process according to claim 5, in which the aluminum is added before the titanium.

18. A process according to claim 6 in which the aluminum is added before the titanium.

19. A process according to claim 12 in which the addition of sulfur is made after the addition of titanium and aluminum.

20. A process according to claim 12 in which the addition of aluminum is made before the addition of titanium.

21. A process according to claim 1 in which the melt is cast into a mold with a bottom chill.

22. A process according to claim 1 in which the alloy composition contains at least about 13 percent iron.

23. A process according to claim 22 in which metal from the melt is cast into a mold with a bottom chill.

24. A process according to claim 6 in which the amount of carbon added is 0.03 percent to 0.25 percent, the aluminum is added before the titanium, the sulfur addition is made after the addition of titanium and is at least 0.2 percent and the melt is held at least 5 minutes after the addition of sulfur.

25. A process for making an iron-aluminum-nickel-cobalt alloy containing 5 percent to l 1 percent aluminum, 7 percent to 25 percent nickel, 20 percent to 55 percent cobalt, l percent to l 1 percent titanium, up to 10 percent copper and up to 4 percent columbium with the balance except for carbon, sulfur and impurities being iron comprising:

a. forming a melt from a virgin raw material base charge containing the nickel, cobalt, iron and any copper and columbium for said alloy composition;

b. deoxidizing said melt;

c. introducing at least 0.03 percent carbon into the deoxidized melt;

d. holding the melt at least 2 minutes after introduction of the carbon and then adding the aluminum for said composition;

e. thereafter adding the titanium for said composition and at least 0.2 percent sulfur with or after the titanium addition;

f. holding the melt at least 5 minutes after the sulfur addition and then solidifying the melt.

26. A process according to claim 25 in which metal from the melt is cast into a mold with a bottom chill and is solidified to produce an ingot having a columnar crystal structure.

27. A process comprising melting a base charge composed partly of metal produced by the process set forth in claim 25 and partly of virgin raw material to provide a composition as set forth in claim 25 and then adding aluminum and thereafter titanium for said composition, adding sulfur to the melt with or after the titanium, provided that when less than 40 percent of the base charge is metal produced according to the process of claim 25 carbon is added before adding the titanium, holding the melt at least 5 minutes after the sulfur addition and then casting metal from the melt into a mold with a bottom chill and solidifying the cast metal to produce an ingot having a columnar crystal structure.

28. A process according to claim 27 in which the alloy composition contains at least about 13 percent iron.

29. A process according to claim 27 in which a columnar structured portion of the ingot is magnetized to provide a permanent magnet.

mg 5mm STATES lATENT OFFICE CERTIFICATE OF CORRECTION 3,615,916 October 26, 1971 Patent No. Dated Inventor(s)STUART WALTER KER SHAW and DEREK JIM PALMER It is certified that error appears in the above-identified patent and that said Letters Patent are hereby corrected as shown below:

[- Column 1 line 28, for (Al=Ti) read (Al x Ti) .1

line 32, for (Al="ii) read (Al x Ti).

Column 3, line 48, for *is read --as-.

Column 5, line 46, Table IV, insert -Ingot above "No. in first column of heading.; second column of heading, insert after "Sulfur added" and in fourth column of heading, insert -length-- before (inches) and on line 59, after "15%" insert --;-:i-.

Column 6, line 42, after "minutes," insert -and-.

Column 7, line 21, Table VII, llth column of numbers Inqot No. 74-, for "3.4" read -2.4; 12th column of numbers, same Ingot Ho. 74, for "59.i read -59.8--; and line 22, 7th column of numbers, Ingot No. 75 for "6.83 read --6.65-.

Column 9, lines 32 and 33, for "and Anbo 12 minutes at 830C. aged for 32 the titanium increased a 570C. and allowed to cool. read ----for 12 minutes at 830C., aged for 32 hours at 570C. and allowed to cool.-; same column, line 66 Claim 1, for "form" read -from-.

Signed and sealed this 8th day of May 1973 EDWARQM. FLETCHER,JR. ROBERT GOTTSCHALK Attestlng Officer Commissioner of Patents

Claims

2. A process according to claim 1 in which the melt is formed wholly of virgin raw materials, a base charge of the iron, nickel, cobalt and any copper and columbium is melted, the melt is deoxidized, and the aluminum and titanium are added, the carbon being added to the base charge or to the melt before the addition of the titanium and the sulfur being added to the deoxidized melt with or after the titanium.
3. A process according to claim 2 in which the carbon is added to the melt after deoxidation and before the addition of aluminum.
4. A process according to claim 2 in which the melt is held for a period of at least 5 minutes after the final addition before being cast.
5. A process according to claim 4 in which the holding period is at least 10 minutes.
6. A process according to claim 2 in which the carbon is added after the deoxidation and the melt is held for at least 2 minutes before any aluminum or titanium is added.
7. A process according to claim 3 in which the aluminum is added before the titanium.
8. A process according to claim 1 in which a base charge of the iron, nickel, cobalt and any copper and columbium, together with carbon, is subjected to vacuum and the aluminum and titanium are added, the sulfur being added to the melt with or after the titanium.
9. A process according to claim 8 in which the melt is formed wholly of virgin raw materials.
10. A process according to claim 8 in which the melting is effected under an inert gas, the melt is then subjected to the vacuum, the aluminum and titanium and sulfur are added under an inert gas, and the melt is again subjected to vacuum.
11. A process according to claim 1 in which the melt is formed from a base charge composed partly of scrap produced in a previous process according to claim 1 and partly of virgin raw materials, except for virgin titanium and aluminum, and the remaining titanium and aluminum and additional sulfur are added to the melt thus formed.
12. A process according to claim 11 in which the scrap amounts to less than 40 percent of the base charge by weight and carbon is added to the base charge or the melt.
13. A process according to claim 11 in which the addition of sulfur is made after the addition of titanium and aluminum.
14. A process according to claim 11 in which the addition of aluminum is made before the addition of titanium.
15. A process according to claim 1 in which metal from the part of the ingot having a columnar crystal structure is magnetized to form a permanent magnet.
16. A process according to claim 4 in which the aluminum is added before the titanium.
17. A process according to clAim 5 in which the aluminum is added before the titanium.
18. A process according to claim 6 in which the aluminum is added before the titanium.
19. A process according to claim 12 in which the addition of sulfur is made after the addition of titanium and aluminum.
20. A process according to claim 12 in which the addition of aluminum is made before the addition of titanium.
21. A process according to claim 1 in which the melt is cast into a mold with a bottom chill.
22. A process according to claim 1 in which the alloy composition contains at least about 13 percent iron.
23. A process according to claim 22 in which metal from the melt is cast into a mold with a bottom chill.
24. A process according to claim 6 in which the amount of carbon added is 0.03 to 0.25 percent, the aluminum is added before the titanium, the sulfur addition is made after the addition of titanium and is at least 0.2 percent and the melt is held at least 5 minutes after the addition of sulfur.
25. A process for making an iron-aluminum-nickel-cobalt alloy containing 5 to 11 percent aluminum, 7 to 25 percent nickel, 20 to 55 percent cobalt, 1 to 11 percent titanium, up to 10 percent copper and up to 4 percent columbium with the balance except for carbon, sulfur and impurities being iron comprising: a. forming a melt from a virgin raw material base charge containing the nickel, cobalt, iron and any copper and columbium for said alloy composition; b. deoxidizing said melt; c. introducing at least 0.03 percent carbon into the deoxidized melt; d. holding the melt at least 2 minutes after introduction of the carbon and then adding the aluminum for said composition; e. thereafter adding the titanium for said composition and at least 0.2 percent sulfur with or after the titanium addition; f. holding the melt at least 5 minutes after the sulfur addition and then solidifying the melt.
26. A process according to claim 25 in which metal from the melt is cast into a mold with a bottom chill and is solidified to produce an ingot having a columnar crystal structure.
27. A process comprising melting a base charge composed partly of metal produced by the process set forth in claim 25 and partly of virgin raw material to provide a composition as set forth in claim 25 and then adding aluminum and thereafter titanium for said composition, adding sulfur to the melt with or after the titanium, provided that when less than 40 percent of the base charge is metal produced according to the process of claim 25 carbon is added before adding the titanium, holding the melt at least 5 minutes after the sulfur addition and then casting metal from the melt into a mold with a bottom chill and solidifying the cast metal to produce an ingot having a columnar crystal structure.
28. A process according to claim 27 in which the alloy composition contains at least about 13 percent iron.
29. A process according to claim 27 in which a columnar structured portion of the ingot is magnetized to provide a permanent magnet.