US3723098A - Method of manufacturing iron base alloys - Google Patents

Abstract

IRON BASE ALLOYS ARE MANUFACTURED BY ELECTRON BEAM BOMBARDMENT HEATING OF A MOLTEN POOL OF AN ALLOY AT A REDUCED PRESSURE, AND CONTROLLING THE HEAT INPUT AND THE THROUGHPUT RATE OF THE MOLTEN ALLOY. THE PRESSURE IS MAINBEAM IS MAINTAINED IN EXCESS OF ABOUT 20 KW. PER SQUARE FOOT OF THE SURFACE OF THE MOLTEN POOL, AND THE MOLTEN ALLOY IS EXPOSED TO THE ELECTRON BEAM BOMBARDMENT FOR A PERIOD SUFFICIENT TO REDUCE THE VOLATILE METALLIC IMPURITIES TO EXTREMELY LOW LEVELS. AINED NO GREATER THAN 10**1 TORR, THE POWER DENSITY OF THE

Description

March 27, 1973 c. D'A. HUNT 3,723,098

METHOD OF MANUFACTURING IRON BASE ALLOYS Filed June 15, 1970 2 Sheets -Sheet 1 EXAMPLE 3.

225 LONGITUDINAL A EXAMPLE 3,

/ TRANSVERSE EXAMPLE,I

11s 0/ LONGITUDINAL I25 EXAMPLE 2 LONGITUDIPJAL GHARPY V NOTGH IMPACT STRENGTH, FT. LBS.

IOOF

\EXAMPLE, 50 TRANSVERSE Q l l I l l I -aoo -25o -2oo -|50 -|00 -5o 0 +50 +|oo +ls TEST TEMPERATURE F L v I l'fl 1'," I I g EXAMPLES aas ixmns 4 I I E o9 FIG 2 F) 8 m 'z o 0.6 0

HAMMER ENERGY, FT. LBS.

\ Much 21,1973

- b P h wwmmw O o ummm PP PP PPPPPPPPPPP I- United States Patent 3,723,098 METHOD OF MANUFACTURING IRON BASE ALLOYS Charles dA. Hunt, Moraga, Calif., assignor to Airco Inc.

Continuation-impart of application Ser. No. 722,640, Apr. 19, 1968, which is a continuation-in-part of application Ser. No. 636,666, May 8, 1967. This application June 15, 1970, Ser. No. 46,156

Int. Cl. C2lc 5/52; (322d 7/00 Us. 01. 75- 12 11 Claims ABSTRACT OF THE DISCLOSURE This invention relates generally to iron base alloys, and more particularly it relates to a method of manufacturing iron base alloys having substantially improved physical properties. Theinvention is particularly advantageous in the manufacture of iron-chromium alloys, but is of advantage in manufacturing other iron base alloys as well. This application is a continuation-in-part of application Ser. No. 722,640, filed Apr. 19, 1968, now abandoned, which was a continuation-in-part of application Ser. No. 636,666 filed May 8, 1967, now abandoned.

It is generally known that the processing of steel under vacuum conditions can materially enhance various of its properties. For example, the avoidance of oxide inclusions resulting from vacuum processing, such as vacuum induction melting and vacuum arc remelting, results in a cleaner alloy. This may effect a substantial improvement in the workability of the steel and in its fatigue properties. By way of further example, by starting with proper quality materials, stainless steels may be produced which have a low interstitial content, that is, low carbon and nitrogen content. It is generally known such steels have improved corrosion resistance, and are generally easier to fabricate at given tensile strengths than are'similar stainless steels containing higher amounts of carbon and nitrogen. The improved fabricability achieved at low interstitial content is believed to result from the relative freedom from car bide and nitride at the grain boundaries.

Certain materials are present in iron base alloys as im; purities, generally at low levels, for example below about 0.1% by weight. One class of impurities present in all commercially available iron base alloys are the volatile metallic impurities which are defined herein as lead, hismuth, cadmium, sodium, potassium, silver, calcium, magnesium, barium, zinc, and antimony. The volatile metallic impurities may be divided into two classes, lead, bismuth, cadmium, sodium, potassium, silver, calcium, magnesium and barium, which are essentially insoluble in iron, and zinc and antimony which are slightly soluble.

Commercially available iron base alloys contain these impurities in varying amounts, depending upon the raw materials source and the manufacturing process. Of the total volatile metallic impurities present in commercial iron base alloys, there is generally between about and 75 p.p.m..of the insoluble volatile metallic impurities, and between about 100 and 200 ppm. of the soluble volatile metallic impurities. The presence of these impurities in iron base alloys has heretofore not been considered to have a significant effect on the properties of the alloys. The presence of trace amounts of volatile metallics, therefore, has been generally overlooked. However, electron beam microprobe analysis of fractures in bearing steels shows significant zinc concentrations in the region of the fracture, indicating that the presence of soluble volatile metallic impurities may be a contributing factor in causing failure of the alloy.

Iron-chromium alloys containing in excess of about four percent chromium are commonly referred to as stainless steel. As the name implies, a principal use of stainless steel is in corrosive environments. The corrosion protection in stainless steel is imparted mainly by the presence of chromium, and increasing chromium content generally provides increasing corrosion protection. Other alloying agents, principally nickel, are added to stainless steel to obtain desired properties,v and the type and amount of alloying agents must be carefully controlled to obtain the desired properties. As used herein, the term ironchromium alloy is intended to include those alloys commonly referred to as stainless steel, and, in addition, includes iron-chromium alloys which are not commonly referred to as stainless steel.

The crystal structure of iron-chromium alloys has a marked effect upon the properties of the alloy. Three different crystal structures, austenitic, ferritic and martensitic, or mixtures thereof may be formed upon cooling to room temperature depending upon the amount of chromium, nickel and other alloying agents. Nickel, cobalt, manganese, copper, carbon and nitrogen are referred to as austenite formers, because the presence of these materials in stainless steel tends to stabilize the retention of an austenitic crystal structure. Chromium, silicon, molybdenum, tungsten, vanadium, tin, columbium, phosphorus, aluminum and titanium are known as ferrite formers, since these materials tend to stabilize ferritic grain structure in steels. Martensitic crystal structures are formed by the presence of appropriate combinations of both ferritic formers and austenitic formers in the stainless steel. By balancing the relative amounts of austenite formers and the ferrite formers in the alloy, in accordance with generally known principles, a given iron-chromium alloy can frequently be suitably controlled so that it will be austenitic, ferritic or martensitic, or a combination of these types.

Although iron-chromium stainless steel alloys have desirable corrosion resistance, the addition of the alloying agents necessary to obtain corrosion resistance cause stainless steel alloys to be more difficult to fabricate than normal mild or carbon steel. In this connection, austenitic, ferritic and martensitic iron-chromium alloys are all susceptible to surface disruptions, in the form of edge cracking, and streaking when worked to high reduction in areas and also become work hardened. Some ferritic alloys are susceptible to a particularly undesirable surface disruption generally referred to as surface roping. In order to remedy these defects, commercial practices involve several anneals and surface conditioning operations in reducing slab stock to sheet. Ferritic iron-chromium alloys are also notch sensitive, particularly when they contain in excess of 20% chromium, requiring careful cold working conditions to prevent brittle fracture. Austenitic iron-chromium alloys have a higher rate of work hardening than do the ferritic or martensitic alloys and are susceptible to transformation upon cold working to martensite.

Because of the widespread use of stainless steel alloys in environments where surface corrosion, and surface decoration are important, extensive investigations of stainless steel have been made. The principal aim of these investigations has been to improve corrosion resistance, generally through the development of new combinations of alloying agents. Little commercial effort has been directed to improving the fabricability of existing alloys, while retaining the desired corrosion resistance.

The relative ease of fabrication of a metal may be determined by the ease in which it may be deformed, for example, by rolling or swaging. Another indication of the ease of fabrication is the ductility and the tough ness of the metal. One method of determining the toughness of iron-chromium alloys is to determine the ability of the alloy to withstand impact loading. In addition, the impact strength, which is apparently related to the ability to withstand high stress, low cycle strains, is important in environments where shock loads or violent impacts might be encountered.

The impact strengths of commercial austenitic ironchromium alloys do not change drastically over wide temperature ranges, for example, from -320 F. to 200 F., and the longitudinal impact strength of commercial austenitic alloys is generally sufficient for most purposes. However, the transverse impact strength of commercial austenitic alloys is substantially less than the impact strength in the longitudinal direction. Similarly, the transverse impact strengths of commercial martensitic and ferritic alloys are substantially less than the longitudinal impact strength.

In addition, the ferritic alloys typically exhibit a drastic reduction in both longitudinal and transverse impact strength at temperatures around room temperature or higher (the impact transition temperature) and at temperatures below room temperature have very low impact strengths and are very brittle. Thus, the ferritic alloys cannot be employed where they might be subjected to impact loading at temperatures below the transition temperature, i.e., much below room temperature. This has eliminated the use of ferritic iron-chromium alloys in many applications where, although the corrosion resistance of the ferritic alloy is sufficient for the environment, the possibility of impact loading required the use of the relatively more expensive austenitic alloys. This is particularly true in Welded structures, where the ferritic welds are quite susceptible to cracking due to poor ductility at room temperature, as well as carbide segregation in the weld region which occurs during heating.

An object of the invention is to provide an improved method of manufacturing iron base alloys which provides alloys of improved physical and mechanical properties and improved fabricability. A further object of the invention is to provide a method for manufacturing ironchromium alloys of improved ductility, thereby permitting easier fabrication of the alloys. Another object is to provide a method for making ferritic iron-chromium alloys which exhibit superior ductility and resistance to impact loading at room temperature. A still further object is to provide a method for making iron base alloys having improved impact strengths, and particularly improved impact strengths in the transverse direction. It is another object of the invention to provide iron base alloys having improved fatigue properties.

These and other objects of the invention will be more readily understood from the following detailed description and from the drawings of which:

FIG. 1 is a graph depicting impact strength versus temperature for two commercially manufactured austenitic stainless steels and an austenitic iron-chromium alloy manufactured by the method of the invention.

FIG. 2 is a graph depicting true compressive strain versus hammer energy for several austenitic alloys manufactured by the method of the invention.

FIG. 3 is a graph depicting the longitudinal impact transition temperature range for a commercially manufactured cold worked and annealed ferritic stainless steel and the longitudinal transition temperature, both unwo k d and cold o ked and annealed, of ferritic iron- Cir chromium alloy manufactured by the method of the invention.

FIG. 4 is a graph, similar to FIG. 3 depicting the transverse impact temperature range of the commercial ferritic alloy and the ferritic alloy of FIG. 3.

Very generally, the present invention is directed to the manufacture of iron base alloys containing not more than about 6 p.p.m. of insoluble volatile metallic impurities, i.e., lead, bismuth, cadmium, sodium, potassium, silver, calcium, magnesium, and barium, and not more than about 20 parts per million soluble volatile metallic impurities, i.e., zinc and antimony. Many iron base alloys manufactured in accordance with the invention exhibit substantially improved fatigue properties or impact properties or both. Iron-chromium alloys so manufactured have substantially improved physical and mechanical properties, and have increased impact strengths, as compared to conventional iron-chromium alloys. Austenitic ironchromium alloys have increased impact strengths at all temperatures, e.g., at room temperature and above down to liquid nitrogen temperatures (-320" F.). The transverse impact strength of the austenitic and martensitic iron-chromium alloys in accordance with the invention is more than double the transverse impact strength of similar conventional commercial alloys containing volatile metallic impurities. Ferritic iron-chromium alloys in accordance with the invention have substantially improved transverse and longitudinal impact strengths at elevated temperatures (200 F.-300 F.) and, in addition, have transition temperature ranges generally below room temperatures.

The method of the invention includes electron beam bombardment heating a molten pool of the alloy at a reduced pressure, and controlling the heat input of the electron beam and the throughput rate of the molten alloy in order to expose the molten alloy to heating by the electron beam for a sufficient period of time, and at sutficient heating rates, to reduce the content of volatile metallic impurities to below about 6 p.p.m. of insoluble volatile metallic impurities, and below about 20 p.p.m. of soluble volatile metallic impurities.

In the case of iron-chromium alloys, the carbon and nitrogen content is reduced to fairly low levels such that the carbon and nitrogen Will be in solid solution in the iron-chromium alloy at room temperature. Desirably the process is controlled such that the total carbon plus nitrogen content will be below 250 p.p.m., and will generally be between 70 and p.p.m. Larger amounts of carbon plus nitrogen can be tolerated in the austenite alloys, e.g., up to 600 p.p.m., due to their greater solubility in austenitic alloys than in ferritic or martensitic alloys. Substantially improved impact strengths, in both transverse and longitudinal directions for iron-chromium alloys evidences the substantially improved fabricability and workability of the described alloys as compared to similar conventional iron-chromium alloys.

It is believed, although the invention is not considered to be limited thereto, that the improved physical and mechanical properties of the described iron base alloys is achieved as the direct result of the relatively complete absence of volatile metallic impurities, as well as the low carbon plus nitrogen content in the case of the ironchromium alloys. As indicated, there are two classes of volatile metallic impurities which are present in iron base alloys, and in accordance with the disclosed invention, the alloys contain less than 6 p.p.m. of those volatile metallic impurities which are defined as insoluble and less than 20 p.p.m. of those volatile metallic impurities Which have been defined as soluble. Preferably, the insoluble volatile metallic impurities are present in an amount of not more than about 2 p.p.m., and the soluble volatile metallic impurities are present in an amount of not more than about 10 p.p.m.

It is to be understood that the terms insoluble and soluble as used herein, are relative terms and are intendedto denote the difference between the two classes of volatile metallic impurities. These volatile metallic impurity substances tend to be concentrated at, or in the vicinity of, the grain boundaries of the alloy, or at the dendritic interfaces within the grains themselves, and it is believed that it is the concentration of these impurities at or in the vicinity of these regions which cause the ironbase alloys to have lesser mechanical properties in the transverse direction. The insoluble volatile metallic impurities, lead, bismuth, cadmium, sodium, potassium, silver, calcium, magnesium, and barium are essentially completely insoluble in iron base alloys at all concentrations, and the total content of these impurities in the iron base alloy tends to be concentrated at the last-to-solidify regions as precipitates. It is therefore quite important that the alloys be essentially completely free of, i.e., contain less than 6 ppm. of, the insoluble volatile metallic impurities.

Zinc, and antimony, referred to herein as soluble volatile metallic impurities are somewhat more soluble in iron base alloys than are the insoluble volatile metallic impurities and are not as deleterious as the insoluble impurities because they are not present asprecipitates. However, the soluble volatile metallic impurities also tend to be concentrated at the grain boundaries, or at the intragranular dendritic interfaces, and it is believed that their presence at or in the vicinity of these regions may cause the alloys to have lesser mechanical properties in the transverse direction. Accordingly, the iron base alloys should be substantially free of the soluble volatile metallic impurities, i.e., contain less than about 20 p.p.m.

It is to be understood that measurement of the extremely low levels of volatile metallic impurities in alloys is, at best, a very difficult analytical problem. The limits of the volatile metallic impurities set forth herein are considered to be approximate in nature, rather than absolute, and the manufacture of alloys having compositions differing slightly from the alloys set forth herein, but having the improved fatigue and impact strengths and fabricability of the described alloys, is considered to be within the scope of the invention.

As mentioned above and as set forth subsequently in this specification, iron base alloys containing less than about 6 p.p.m. total of insoluble volatile metallic impurities and less than about 20 p.p.m. soluble volatile metallic impurities exhibit substantially improved physical and mechanical properties as compared to conventional alloys. The removal of volatile metallic impurities from iron base alloys at commercially attractive rates, and particularly the manufacture of these alloys from low cost commercial raw material sources, is advantageously effected by ex posure of a molten alloy to electron beam bombardment at a reduced pressure of below about 10" torr. The power of the electron beam bombardment is controlled so that the power input is in excess of about kilowatts per square foot of the molten alloy surface, and the duration of the exposure of the molten alloy to the electron beam is controlled to between about 100 and about 2000 pounds per hour per square foot of the molten alloy surface.

Bombardment heating of the iron base alloys with an electron beam is considered to be advantageous in the manufacture of alloys containing less than 6 ppm. of insoluble volatile metallic impurities and less than about 20 p.p.m. of soluble volatile metallic impurities, and, based upon present information, it is believed that it is neces sary to utilize electron beam bombardment heating in order to economically practice the disclosed process. It is believed that the impact of the electron beam upon the surface of the molten alloy creates an area of localized superheat on the surface of the molten alloy pool, which is of significant importance in obtaining the desired purification and removal of volatile metallic impurities.

The localized area of superheat adjacent to the surface of the molten pool causes thermal imbalances in the 6 molten pool and creates thermal convection currents in the molten pool. The thermal convection currents create turbulence and stirring in the molten pool, and have the effect of continuously supplying fresh molten metal from the lower portions of the pool to the surface of the pool. The continuous supplying of fresh molten metal to the surface of the molten pool and the turbulent stirring of the pool enhances vaporization of the volatile metallic impurities in an electron beam bombardment heating apparatus compared to other types of conventional heating apparatus such as vacuum induction furnaces.

Other means of heating do not result in the desired uneven heat distribution and consequent thermal convection currents. Electron beam heating has the further advantage that it may be carried out in a hearth cooled sufiiciently to provide a solidified skull of the material being processed. Containment of the molten pool in the skull eliminates problems of contamination due to solubilization of the containment crucible. Further there is no contamination of the molten pool by the electron beam heat source, as in the case with are type heating.

In accordance with a preferred embodiment, one or more electron beams are directed upon the surface of the molten pool of iron base alloy flowing along a cooled hearth. The impact area of the electron beam on the molten pool may be stationary, or the electron beam may be swept across the molten pool. Sweeping of the electron beam across the molten pool at a rate slow enough to maintain local superheat has the advantage that the region of turbulent stirring of the molten pool is moved to different regions of the pool resulting in a more uniform stirring of the molten pool, thus permitting lower power input, as compared to stationary beams, for a given degree of purification.

The method may be successfully practiced with various raw materials in the manufacture of iron-chromium alloys containing between about 4 percent and about 40 percent chromium, with or without the usual alloying agents present in iron-chromium alloys, and particularly stainless steel, as described herein. It is contemplated to utilize the method in the manufacture of iron-chromium alloys directly from conventional iron and ferrochrome raw materials. It is also possible to utilize the method in combination with conventional manufacturing practices in which the product of a basic oxygen furnace, electric arc furnace or vacuum induction furnace is exposed to electron beam bombardment at reduced pressures as disclosed herein.

The surface of the molten pool flowing along the cooled hearth is exposed to a reduced pressure, not more than about 10 torr, and preferably below about 10* torr during the electron beam bombardment. Pressures much below 10- torr do not appear to enhance the process although pressures as low as 10'' torr can be employed if desired. In order to maximize the amount of alloy exposed to the evacuated region, the molten pool is preferably contained in a crucible or a hearth which has a large area relative to its depth and the evacuated region preferably extends over the entire exposed surface area of the pool. Pool depths of between about /2 inch and 3 inches are contemplated. Pool depths greater than about 3 inches result in inefficient stirring by the localized area of superheat with the result that the molten material adjacent the bottom of the pool may not be brought to the surface. The method may be practiced in a batch type process or in a continuous or semicontinuous type process wherein a stream of molten alloy is introduced and withdrawn from the hearth.

The molten alloy pool is heated to a temperature sufficient to maintain the alloy molten and to vaporize the volatile metallic impurities, but insufiicient to vaporize substantial amounts of iron and other desired alloying agents as would render the process uneconomical. This involves proper selection of the reduced pressure established above the molten pool and the temperature of the molten pool. Typically, the average temperature of the pool will be between about 50 F. and about 200 F. above the melting point of the alloy. The temperature of the localized area of superheat may be between about 50 F. and about 300 F. above the average temperature of the molten pool of alloy. Below a 50 F. differential between the average pool temperature and the temperature of the localized area of superheat, there is insufficient stirring of the molten pool, while above a 300 F. differential there is excessive vaporization losses of iron and other alloy elements. Temperature differentials of between 100 F. and 200 F. have been found to provide good results.

The particular length of exposure to electron beam bombardment depends upon the initial composition of the alloy, the temperature and beam impact distribution on the pool surface, the pressure in the evacuated region above the molten pool, and the area and depth of the molten pool. With so many variables to be considered, it is usually preferable that selection of exposure time be on an empirical basis. However, the selection of operating parameters should be made to insure that the insoluble volatile metallic impurities will be reduced to below about 6 ppm. and the soluble volatile metallic impurities will be reduced to below about 20 ppm, preferably below about 2 and about 10 ppm, respectively. In addition, the carbon and nitrogen content of an iron-chromium alloy should be reduced sufficiently that all the carbon nitrogen in the alloy will be in solid solution at room temperature. In the case of ferritic and martensitic iron-chromium alloys this is below about 250 ppm. combined carbon plus nitrogen while in austenitic alloys this is below about 600 p.p.m.

For a given set of operating conditions, sufficient exposure to electron beam bombardment heating results in an alloy having improved physical and mechanical properties. In this connection, good results are obtained in a continuous process when a molten iron base alloy is passed over a cooled hearth disposed in a region evacuated to less than 10- torr at a rate of between about 100 and about 2,000 pounds per hour per square foot of hearth. At lower pressures, greater flow rates may be utilized. At a pressure of 10 torr, the fiow rate should be maintained in the low end of the disclosed range, e.g., about 100 to 300 pounds per hour per square foot of hearth to effect the desired results, while at lower pressure, such as l torr, flow rates approaching 2,000 pounds per hour per square foot of hearth can be employed. Generally, for most purposes, flow rates of between about 200 and about 1,000 pounds per hour per square foot of hearth have been found to be most desirable.

As indicated, the electron beam power input to the molten pool is adjusted to maintain proper fiow of the molten alloy and to provide sufiicient superheat at the surface thereof to cause thermal stirring of the molten pool. The electron beam power input desirably exceeds about 20 kilowatts per square foot of hearth and may be as high as 100 kilowatts per square foot of hearth, although this may result in excessive vaporization losses. The maximum electron beam power which may be utilized is determined by balancing costs against increased vaporization losses of iron and, in the case of iron-chromium alloys, of chromium.

The described method can be carried out in an electron beam furnace having a single pressure zone or in a multipressure zone refining apparatus such as is described in US. Pat. No. 3,343,828. The use of multipressure zone refining apparatus is particularly advantageous in the manufacture of iron-chromium alloys directly from ferrochrome raw materials.

For the more common types of commercially available iron-chromium alloys, such as the A151 300 and 400 series stainless steels, satisfactory results are attainable where the molten alloy is passed over cooled hearths in two successive vacuum zones or stages, as is shown in US. Pat. No. 3,343,828. The first stage may be evacuated to a pressure of about 10 torr, and serves to remove dissolved gases and highly volatile impurities as well as partial removal of the volatile metallic impurities. The second stage may be evacuated to a pressure of about 10- torr or less and the volatile metallic impurities and gaseous impurities are removed to the desired low levels in this zone. Electron beam powers of about 30 to kilowatts per square foot of hearth may be utilized in both zones at throughput rates of 200 to 1000 pounds per hour per square foot of hearth.

Each of Examples 3 through 10, 13 through 20, and 23 through 25, set forth below, were manufactured in accordance with the disclosed process in a hearth type electron beam furnace as described in Pat. No. 3,343,828. The pressure in the first zone was maintained at 10- torr, and the pressure in the second zone was maintained at 10- torr. The flow rate of the molten alloy through the furnace varied from about to about 400 pounds per hour per square foot of hearth, and the power density of the electron beam varied from about 40 to about 80 kilowatts per square foot of hearth.

As indicated, iron-chromium alloys, and particularly stainless steel alloys, are generally identified by their crystalline configuration, austenitic, ferritic, martensitic or combinations thereof. Many stainless steel alloys are mixtures of the different crystal configurations, and the properties of the alloys are, to a large extent, dependent upon the relative proportion of the different crystal configurations present in the alloy. Because of these different properties, and the fact that the end uses of heretofore available commercial alloys were dependent upon the crystal configurations of the alloy, each of the three types of crystal configurations is discussed separately. However, it is to be understood that in many instances the alloys will contain portions of the other crystal configurations, and the designation of an alloy as austenitic, ferritic or martensitic is intended only to designate the predominant crystal configuration present in the alloy.

A more detailed description of such alloys is included in copending application Ser. No. 722,640, now abandoned and in copending application Ser. No. 46,443 filed June 15, 1970, which is a continuation-in-part of abandoned Ser. No. 722,640.

AUSTENITIC ALLOYS The austenitic iron-chromium alloys are those alloys which contain sufficient amounts of austenite stabilizers to cause the alloy to have an austenite crystal configuration at room temperature. The principal austenite stabilizer is nickel, and substantially all commercial austenitic iron-chromium alloys contain nickel, usually in an amount between about 6 percent and about 12 percent by weight and chromium, generally in an amount between about 17 percent and about 22 percent. Other austenite stabilizers, such as carbon, nitrogen, manganese, cobalt and copper may replace a portion of the nickel. In the AISI numbering system, the 300 Series stainless steels are commonly austenitic. The presence of nickel in stainless steel iron-chromium alloys is principally to obtain suflicient ductility that the stainless steel alloy can be fabricated into desired shapes, and to permit welding of the alloy.

The austenitic alloys described herein, and manufactured in accordance with the disclosed method, have substantially improved physical and mechanical characteristics when compared against conventional austenitic alloys prepared in accordance with conventional manufacturing techniques, e.g., basic oxygen furnace, electric arc furnace, or vacuum induction furnace processes. It has been determined that the austenitic alloys of the invention may be readily cold and hot worked and welded with better results than conventional austenitic alloys.

10 of the austenitic alloys of Examples 3 to 9 are all in excess of .90, and most are in excess of .95. Conven- TABLE I Percent by weight Parts per million Volatile metalie Ex Cr Ni Al Mn Mo Si Ti V W Other N O S P impurities 17.2 10.2 NA 03 .14 .1 220 NA NA 70 180 50/150. 18.3 9.1 0.1 1.54 3 650 NA NA 240 270 300total. 18.6 9.7 .003 NA 140 110 9 80 170 6/20. 17.6 9.8 .003 NA .12 170 50 9 80 170 6/20. 16.8 10.0 .003 NA .11 100 NA NA 60 170 6/20. 18.5 9.9 .003 0.34 .12 200 98 12 70 180 6/20. 20.5 10.3 .003 03 .1 6O 35 12 60 180 6/20. 16.4 13.4 NA NA 2.2 100 NA NA 90 110 6/20. 17.3 13.0 NA NA 2.0 NA 100 NA NA 90 350 6/20 16.3 9.1 .003 NA .1 .100 NA NA 60 340 6/20 Table I sets forth the composition of several austenitic iron-chromium alloys. Examples 1 and 2 are commercially available austenitic stainless steels which correspond to AISI grade 304. Example 1 is a vacuum induction melt, representing the latest in commercially practicable processing, and Example 2 is an electric furnace melt. Examples 3 to 9 have nominal compositions comparable to commercially available AISI 300 Series stainless steels, except that Examples 3 to 9 each have less than about 6 p.p.m. of insoluble volatile metallic impurities, and less than about p.p.m. soluble volatile metallic impurities. Further, the alloys of Examples 3 to 10 have carbon and nitrogen contents such that the carbon and nitrogen is in solid solution at room temperature, i.e., below about 300 p.p.m. total carbon plus nitrogen. Accordingly, the described alloys have somewhat lower carbon plus nitrogen content than commercial stainless steels of equivalent nominal composition.

The substantially improved properties of the austenitic iron-chromium alloys described herein is illustrated by the graph of FIG. 1 in which there is plotted Charpy V- notch impact strengths (ASTM No. A370- 67), in both the transverse and longitudinal directions, of Examples 1, 2 and 3. It can be readily seen that at room temperature the described austenitic stainless steel of Example 3 has substantially higher transverse and longitudinal impact strengths than does either of the commercial stainless steels of Examples 1 and 2. Further, the transverse impact strength of Example 3 is equal to the longitudinal impact strength at room temperature and at 320 F., exceeds about 100 foot pounds. The substantial improvement in the impact strength of Example 3 will also be seen from the fact that at -320 F. the transverse impact strength is greater than the longitudinal impact strength of the comlmercial stainless steels of Example 2, and is substan tially equal to the longitudinally impact strength of Example 1.

The described austenitic alloys containing less than about 6 p.p.m. of insoluble volatile metallic impurities and less than about 20 p.p.m. soluble metallic impurities have substantially improved longitudinal and transverse impact strengths as compared to the conventional stainless steel alloys of Examples 1 and 2. Examples 4 to 10 have nominal compositions corresponding to several AISI 300 series stainless steels, and, in addition include additional alloying agents. Example 5, which contains 0.16 percent zirconium, exhibited a transverse impact strength reduced by the presence of zirconium in the alloy. However, the impact strengths of Example 5 are superior to conventional zirconium containing stainless steels of similar nominal composition. Examples 4 and 6 contain titanium and manganese alloying agents respectively. These alloying agents do not affect the transverse impact strength of the alloy. Accordingly, it is possible to add manganese, conventionally added to stainless steels to mask the sulphur content thereof, and still obtain decidedly superior transverse and longitudinal impact strengths.

The so called R.A.T. ratio is an indication of anisotropy, and high R.A.T. ratios indicate low anisotropy,

longitudinal and transverse direction. The R.A.T. ratio tional austenitic stainless steels have much lower R.A.T. ratios of .75 or lower.

In addition to the substantially improved impact strengths and R.A.T. ratios, the substantially improved properties of the austenitic alloys described herein is evidenced by other tests and comparisons with conventional austenitic stainless steel alloys, set forth in the aforesaid copending application.

Corrosion tests were also conducted and it was found that the described austenitic alloys were generally better in stress corrosion resistance, intergranular corrosion resistance and massive corrosion than were equivalent conventional stainless steels. Standard ASTM Huey Tests (No. A262-64) in which the corrosion rate in boiling 65 percent nitric acid is measured, indicated that the austenitic alloy of Examples 3, 4 and 6 have superior corrosion resistance in the sensitized form when compared to Example 2 and would compare favorably with stabilized conventional AISI 321 and 347 grade stainless steel for welding operations.

FERRITIC IRON-CHROMIUM ALLOYS The austenitic iron-chromium alloys described herein have substantially improved impact strengths in both the longitudinal and transverse directions, and these impact strengths are improved over the entire temperature range of 320 F. to room temperature and above. It will be seen from FIG. 1 that the impact strengths decline as the temperature is lowered, but this decline is generally linear. Ferritic alloys which contain relatively small amounts of nickel, generally less than 2 percent and most often less than 1 percent, exhibit wholly different impact strength curves than do austenitic alloys. Conventional ferritic stainless steels exhibit impact strengths generally similar to those of conventional austenitic stainless steels at elevated temperatures, i.e., above 200 F. However, ferritic stainless steels generally exhibit a drastic reduction in impact strengths within a temperature range of about room temperature up to F., and at temperatures below room temperature conventional ferritic stainless steels have negligible impact strength.

The temperature or temperature range over which the impact strength of ferritic steel is drastically lowered, is referred to as the transition temperature. At temperatures below the transition temperature, ferritic stainless steels are brittle and are susceptible to fracture, while at temperatures above the transition range the ferritic stainless steels are considered to be tough. The transition temperature range of a ferritic alloy is of importance in considering the usefulness of the alloy. For example, in welded articles such as ship hulls and pressure vessels rapidly propagating brittle fractures caused by impact loading at temperatures below the transition temperature can result in complete structural failure. It is therefore important that the transition temperature of ferritic alloys be below room temperature if this type of material is to find widespread use in welded structures.

It is well known that it is diflicult to designate a specific temperature as the transition temperature in ferritic alloys. Testing of several samples from the same ingot generally results in varying transition temperatures. This is believed to be due to the differences in the machinin of the samples, the particular grain orientation at the point of fracture, and other mechanically induced errors which lead to a spread in transition temperature values for similar ferritic alloys. The ferritic alloys described herein have even wider transition temperature ranges due to the substantially complete absence of volatile metallic impurities and the very low carbon and nitrogen content.

Nevertheless, comparison of impact strengths and the transition temperature range provides a qualitative correlation of the relative notch toughness of fabricated articles under service conditions, and the ability to employ the disclosed ferritic alloys in welded structures at normal ambient conditions.

12 of the hot rolled material of Example 15 and the shaded area B corresponds to the longitudinal transition temperature range of Example 15 after cold working and annealing at 1450 F. The shaded area C corresponds to the longitudinal transition temperature range of Examples 11 and 12.

FIG. 4 illustrates the transverse transition temperature range of a hot rolled and unannealed sample of Example 15 in the shaded portion A, and the transverse transition temperature range of Examples 11 and 12 in the shaded portion C. Accordingly, when transition temperature ranges below 0 F. are desired such can be obtained by simple cold working and annealing.

It can be seen from FIGS. 3 and 4 that the transition temperature ranges for the ferritic alloys described herein are substantially lower than conventional alloys having TABLE II Percent by weight Parts per million Volatile metallic Cr Ni Al Mn Mo Si Ti V W C N O S P impurities .7 28 03 .06 .35 NA NA NA 700 NA NA 190 200 300 total. .3 NA NA .35 NA 33 NA NA NA 700 NA NA 110 180 200 total .4 01 .03 NA .01 .04 002 .02 NA 20 NA 90 310 6/20. 3 02 01 NA 01 06 002 NA NA 20 8 21 90 140 0/20 2 22 NA NA NA .06 NA NA NA 20 40 NA 150 6/20 6 NA NA NA NA NA NA NA NA 50 NA 130 6/20 .3 06 7 03 01 .07 54 02 1 80 20 NA 60 6/20 .8 .05 2. 0 03 01 07 53 .02 1 60 20 NA 60 100 6/20 .4 NA 3. 7 06 .02 .04 66 NA NA 80 30 NA 20 6/20 .6 04 5.6 03 .01 .05 52 02 .1 80 40 NA 80 110 6/20 There is set forth in Table II the composition of a series of ferritic alloys having nominal compositions corresponding to typical AISI 400 series stainless steels. Example 11 is a conventional electric furnace melt and Example 12 is a conventional vacuum induction furnace melt. Examples 13 to 20 are ferritic alloys as described herein containing less than 6 p.p.m. insoluble volatile metallic impurities and less than 20 p.p.m. soluble volatile metallic impurities. Comparison of the impact strengths and transition temperature ranges of the ferritic alloys described herein with those of conventional stainless steel alloys clearly indicates substantially improved physical and mechanical properties which are achieved by the described alloys.

Example ll, an AISI 430 grade stainless steel has a longitudinal impact strength of less than 100 at 200 F. and a transverse impact strength at 200 F. of less than 50. Example 12, a further sample of a conventional AISI 430 stainless steel, has transverse and longitudinal impact strengths of about 50 at 200 F. Examples 13 to 16 which have nominal compositions corresponding to AISI 400 series ferritic alloys have longitudinal and transverse impact strengths at 200 F. which are significantly greater than the impact strengths of Examples 11 and 12.

Further, and of major importance, the alloys of EX- amples 13 to 16 have substantially lower transition temperature ranges than do the conventional ferritic stainless steels of Examples 11 and 12. Example 11, an electric furnace melt, has a longitudinal transition temperature range of between approximately 175 F. and 225 F., and a transverse transition temperature range in excess of 250 F. Example 12, a vacuum induction melt, has longitudinal and transverse transition temperature ranges of between approximately about F. and 225 F. Example 14 has a longitudinal transition temperature range of between approximately 0 F. and 50 F. and a transverse transition temperature range of between approximately 40 F. and 20 F. Example 15 has a longitudinal transition temperature range of approximately 40 F. to 0 F. and a transverse transition temperature range of between approximately 0 F. and 70 F. The downward shift of the transition temperature range of Example 15 as compared to Examples 14 is brought about by cold working to at least about 50 percent reduction followed by grain refinement annealing, e.g., at 1450 F., and is further illustrated in FIG. 3 where the shaded area A corresponds to the longitudinal transition temperature range similar nominal compositions. This property of the disclosed ferritic alloys is quite important, particularly where the alloy is to be welded. As previously indicated, conventional ferritic alloys have transition temperature ranges at or above room temperature, and are subject to weld cracking under impact loading. Conventional iron-chromium alloys containing 25 percent chromium cannot be welded without cracking. Accordingly, the conventional ferritic alloys are seldom used in welded structures. However, the ferritic alloys described herein, containing from 10 percent to about 25 percent chromium, less than about 6 ppm. insoluble volatile metallic impurities, and less than about 250 ppm. carbon plus nitrogen have longitudinal and transverse transition temperature ranges generally below room temperature and are not subject to weld cracking at ambient conditions.

The ferritic alloys described herein are capable of use in environments where weldability and/or impact strength at room temperature is important, and may replace austenitic alloys in the fabrication of structures which are welded or are subject to impact loading. Since corrosion resistance depends primarily upon the chromium content of the alloy, the ferritic alloy described herein can be employed in place of the higher cost nickel containing austenitic alloys, thus lowering manufacturing cost without sacrificing either corrosion resistance or impact strength. R.A.T. ratios of approximately 1.0 are readily obtained with the ferritic alloys. This is a further indication of the excellent ductility of the ferritic alloys, and evidences the extremely low anisotropy of these alloys.

The improved ability to work and fabricate the disclosed ferritic alloys applies to both hot and cold rolling. It is also possible to eliminate box annealing of the ferritic alloys disclosed herein, and grain refinement annealing may be carried out at lower temperatures.

Rolling of conventional ferritic alloys, and particularly AISI 430 grades, may result in a surface imperfection known as roping in which surface ridges or corrugations form when the strip is stretched or drawn into a part. This is quite detrimental and causes conventional AISI 430 grade ferritic stainless steel to be undesirable in certain uses, for example, in the automobile industry. Rolling and forming of the disclosed ferritic alloys of the present invention, including those equivalent to AISI 430, does not result in undesired roping, thus permitting use of the disclosed alloys where conventional AISI 430 alloys are not suitable.

In addition to the more typical A181 400 ferritic alloys, the present invention is also directed to specialty ferritic alloys. Examples 17 to 20 of Table II are directed to ferritic alloys containing aluminum. The alloys of Examples 17 to 20 are low chrome alloys, the aluminum thereof causing hardening and strengthening of the low chrome alloy and improved oxidation resistance. These alloys have excellent corrosion and oxidation resistance and are useful in many moderately high temperature applications. However, the presence of aluminum in the alloy has caused problems in the manufacture of these alloys heretofore, and commercially available ferritic alloys containing one percent or more aluminum have generally been limited to nondecorative uses, such as automobile mufilers and furnace burners. The aluminum containing ferritic alloys disclosed herein are easily manufactured, and can readily be produced in sheet form for use in decorative applications where superior corrosion resistance is desired.

The effect of the addition of aluminum is to raise the transition temperature range of the alloys in both the longitudinal and transverse directions to above room temperature. However, these transition temperature ranges are still well below the usual transition temperatures of conventional aluminum containing ferritic alloys. At 300 F., Examples 17, 18 and 19 containing 1 percent, 2 percent and 4 percent aluminum have impact strengths in both the longitudinal and transverse directions which are in excess of 235. At 6 percent aluminum, the additional aluminum has driven the transition temperature range upwardly to a point where 300 F. is within the transition temperature range and the impact strength at this temperature is not readily ascertained.

The effect of cold working and annealing the aluminum containing alloys manufactured in accordance with the invention is to reduce the longitudinal transition temperature range about 50 F.

The elfect of increased aluminum content of Examples 17 to 20 has a direct effect upon the R.A.T. ratio, and at higher aluminum contents the alloys become quite anisotropic. Although this results in somewhat lesser ductility than the alloys of Examples 13 to 16, the alloys of Examples 17 to 20 are capable of fabrication, either by hot or cold working, and exhibit improved ductility and toughness as compared to conventional aluminum containing alloys.

MARTENSITIC IRON-CHROMIUM ALLOYS The martensitic iron-chromium alloys are generally hardenable by heat treatment, exhibit high tensile strengths and good ductility and toughness and are desirable where high strength and good corrosion resistance is desired. However, from a commercial standpoint, these materials have largely been ignored in the past, principally because of difficulties in fabricating, relatively low impact strengths as compared to austenitic alloys, and poor weldability.

The martensitic iron-chromium alloys have low chromium content, generally between about percent and about 20 percent, usually between about 10 percent and about 14 percent and contain a sufficient amount of austenite stabilizer to cause the alloy to have a martensitic crystalline configuration at room temperature. Depending upon the relative amounts of chromium and austenitic stabilizers, the martensitic alloys may be wholly martensitic, or may be mixtures of martensite plus ferrite, martensite plus austenite plus ferrite, or martensite plus austenite. In order to obtain good weldability it is generally preferred that the martensitic alloys have a relatively low carbon plus nitrogen content and contain less than about 10 percent of ferrite crystals.

One class of commercial martensitic alloys are those which contain between about 10 percent and about 14 percent chromium and between about 2 percent and about 7 percent nickel. Such alloys have a substantially 100 percent martensitic crystalline configuration. These alloys have tensile strengths above 100,000 p.s.i., and when the carbon content thereof is reduced to below about 0.03 percent by weight have good weldability. Higher carbon content, up to about 0.06 percent by weight may be tolerated in welding applications if 0.01 percent by weight titanium is present.

Because the martensitic alloys are air hardenable, measurements of ductility and toughness, such as impact strength and R.A.T. ratios, are significant only when measured at equal strength levels. Accordingly, comparison of the described martensitic alloys with conventional alloys is somewhat more difiicult than are the austenitic and ferritic alloys. Nevertheless, martensitic alloys containing less than about 6 p.p.m. of insoluble volatile metallic impurities and less than about 20 p.p.m. of soluble volatile metallic impurities have been found to have substantially improved physical and mechanical properties as compared to the conventional martensitic alloys.

The impact strengths of the martensitic alloys are generally intermediate those of the austenitic alloys and ferritic alloys. The martensitic alloys have somewhat lower impact strengths at room temperature than do the austenitic alloys, and the martensitic alloys have a transition temperature range between that of the ferritic alloys and the austenitic alloys. However, the transition temperature range of the martensitic alloys is well below room temperature. Conventional 13 percent chromium, 4 percent nickel martensitic alloys have longitudinal impact strengths of between about 30 and 40 foot pounds at F. Alloys having similar nominal compositions but containing less than 6 p.p.m. insoluble volatile metallic impurities and less than 20 p.p.m. soluble volatile metallic impurities, and less than 250 p.p.m. total carbon plus nitrogen content, have longitudinal impact strengths in excess of foot pounds at -95 F. In addition the transverse impact strengths of the martensitic alloys as described herein are substantially improved, and are in excess of 100 foot pounds at room temperature.

A further example of a martensitic alloy is an alloy containing about 16 percent chromium and about 9 percent nickel, less than 6 p.p.m. insoluble volatile metallic impurities and less than 20 p.p.m. soluble volatile metallic impurities. This alloy is a mixture of austenite, martensite, and ferrite crystals and contains less than about 10 percent ferrite. This alloy has tensile strengths in excess of 100,000 p.s.i., and at room temperature, has a longitudinal impact strength of about foot pounds and a transverse impact strength of about 90 to 100 foot pounds. At 320 F. the longitudinal impact strength is about 90 foot pounds and the transverse impact strength is about 40 foot pounds. The R.A.T. ratios of the described martensitic alloys are in excess of .90 and usually in excess of .95 demonstrating the extremely low anisotropy and excellent ductility of martensitic alloys which are free of volatile metallic impurities.

Similar to the ferritic alloys, the weldability of the described martensitic alloys is excellent when the total carbon plus nitrogen content of the alloys is reduced to fairly low levels, i.e., below about 250 p.p.m. The ability to weld the martensitic alloys without fear of weld cracking and embrittlement during service, renders the described martensitic alloys attractive structural materials for use at temperatures below 0 P. where high tensile strengths are required. Further, the excellent ductility of the described martensitic alloys permits ready rolling and fabrication of the alloys into desired shapes.

TABLE III Percent by weight Parts per million Volatile metallic Example Cr Ni Al Mn M Si Ti V W C N O S P impurities .40 2.03 NA .08 NA .306 95 00 120 50 200 total. 03 NA .001 NA NA NA NA NA 400 100 200 total. 43 .001 .08 NA .413 26 13 30 90 G/20. .22 .25 NA NA NA .300 58 16 110 150 6/20. .22 .25 NA NA NA .401 52 12 120 150 0/20.

It will be seen from the foregoing that iron-chromium alloys of substantially improved physical and mechanical properties may be produced by the method of the invention. The superior toughness, ductility, isotropy, corrosion resistance and fabricability of the described alloys is such that they may be employed in environments and in structures where conventional iron-chromium alloys have been heretofore considered to be inappropriate.

The described method for the manufacture of ironchromium alloys is versatile and may be utilized in the manufacture of iron-chromium alloys directly from iron and ferrochrome raw materials. In addition, the described method may be utilized in combination with conventional basic oxygen furnace, electric arc furnace or vacuum induction furnace processes. In such processes it is contemplated that the iron-chromium alloy would be introduced into the electron beam furnace in molten form and processed therein in a single pressure zone or in dual pressure zones. Of course, greater numbers of pressure zones could be utilized, if desired.

OTHER IRON BASE ALLOYS The method of the invention may also be employed in connection with iron base alloys other than the ironchromium alloys described above. Substantial improvement of the properties of many of the other iron base alloys may result by employing the method of the invention. It is believed that significant improvement in the properties can be achieved with regard to any alloy without a persistent second phase present in the as-cast condition. Thus, for example, aluminum and titanium alloys which are precipitation hardening are unlikely to benefit significantly from the process of the invention, since the presence of precipitates is desirable in such alloys. Where, however, precipitates are generally undesirable, it is believed that the removal of the volatile metallic elements can eifect a significant reduction in the tendency for precipitates to form. Where a non-persistent single phase exists in the as-cast condition, or where homogenization of any second phase can be achieved by appropriate heat treating or working, the improved propertics may be less apparent, although some advantage can be gained.

It is believed that the method of the invention can substantially improve the properties of any iron base alloy having a homogeneous solid solution in the as-cast condition, or where segregation resulting from the manner of solidification can be removed by subsequent solid state heat treating or working to cause diffusion from the grain boundaries. This is because the process of the invention reduces the likelihood of harmful segregation occurring during cooling at the liquid-solid interfaces within the grains themselves, as well as at the grain boundaries. Segregates of residual alloys or of impurities which solidify at different times from the remaining material have a tendency to end up in the spaces between the dendritic arms during the solidification process. Such intragranular segregation occurs during solidification as dendrites grow inwardly from the grain boundaries, and leaves segregates at the dendritic interfaces Within each of the grains. Ap

propriate heat treating and working may be able to remove segregates at the grain boundaries, but does not usually remove intragranular segregates, because of the much stronger intermetallic bond of these segregates at the dendritic interfaces. Thus, such segregates do not readily diffuse out with heat treatment or work. The presence of the volatile metallic elements previously discussed, it is believed, contributes to capturing unwanted alloy proportions between the dendrites by forming complex compounds.

Substantial improvement in the properties of the socalled high strength low alloy steels, such as AISI 4340, is achievable through utilizing the process of the invention. Extremely good transverse to longitudinal impact strength ratios are achievable, and the transverse reduction of area properties are significantly improved over commercially available material.

Ingots of 4340 and 4340M alloy were produced in accordance with the invention and compared with the same grade of commercially available air melted alloy. Table III shows the chemical analysis of electron beam 4340 and 4340M, and commercial air melted 4340 and 4340M. Examples 21 and 22 are commercially available air melted alloys 4340M and 4340, respectively. Example 23 is the same 4340M alloy as Example 21 after process ing in accordane with the invention. Examples 24 and 25 are the same 4340 alloy as Example 22 after processing in accordance with the invention. Tests on Examples 21 and and 23 showed that transverse reduction in area percentages were about ten to fifteen percent higher in the case of Example 23. Moreover, Example 23 exhibited a ratio of transverse to longitudinal reduction in area percentages which approached unity, whereas the same property for Example 21 was only slightly over one-half.

Tests on Examples 22, 24 and 25 (4340) showed that transverse reduction in area percentages were at least three times higher in the case of Examples 24 and 25. The ratio of transverse to longitudinal reduction in area percentages exhibited the same improvement for the 4340 alloy proceeds in accordance with the invention as in the case of the 4340M. A significant improvement was also noted in the amount of surface crack-free cold work which could be performed without intermediate anneals.

Significant improvement has also been observed in the properties of bearing steels produced in accordance with the invention. For example, A181 4620 and 52100 bearing steels have exhibited excellent fatigue properties, and bearings which have been produced from 52100 steel processed in accordance with the invention do not fail in conventional long-time tests for bearing life designed to produce failure in even consumable vacuum are remelted materials. Moreover, the ratio of transverse to longitudinal properties is greatly enhanced, and the cleanliness of the steel is far greater than that obtained in commercially available steels.

More particularly, bearing life tests have been performed on samples made from 52100 steel of three types. The first type was processed in accordance with the invention; the second was processed by vacuum arc remelting; and the third by air melting. Bearing samples made of steel processed in accordance with the invention exhibited three times the life of vacuum arc remelt samples and ten times the life of air melt samples when cycled until 10 percent of the samples failed. When cycled until 50 percent of the samples failed, samples made from steel processed in accordance with the invention exhibited about twice the life of vacuum arc remelt samples and ten times the life of the air melted samples.

Tool steels have been produced by the method of the invention and some improvement in their properties has been noted. It is believed that the absence of interdendritic segregates is at least partially responsible for improving the transverse to longitudinal impact ratio such that it approaches unity and for reducing the likelihood of sur- 17 face and edge cracking and other adverse effects during cold working. Moreover, such steels exhibit improved fatigue properties when made in accordance with the invention.

It may be seen, therefore, that the process of the invention is applicable to all types of iron base alloys where a reduction in the presence of interdendritic segregates can have a significant effect on any property of the steel. By controlling the flow rate of the molten alloy on the hearth and the heat input of the electron beam to reduce the level of the insoluble volatile metallic impurities and the soluble volatile metallic impurities to below about 6 p.p.m. total and 20 p.p.m. total, respectively, a significant improvement in the properties of such iron base alloys can be effected.

It is to be understood that the invention has been described with respect to specific iron-chromium and other iron base alloys and specific methods of manufacture in order to sufficiently describe the invention. Alternative embodiments and equivalent practices are considered to be within the scope of the invention.

Various of the features of the invention are set forth in the following claims.

What is claimed is:

1. A method for the manufacture of iron base alloys comprising, bombardment heating the surface of a molten pool of the alloy with at least one electron beam in a region of reduced pressure of not greater than torr, maintaining the power density of the electron beam in excess of about kilowatts per square foot of the molten alloy surface of the molten pool, and exposing the molten alloy to the electron beam bombardment for a period equal to between about 100 and about 2000 pounds per hour per square foot of the surface of the molten pool.

2. A method for the manufacture of iron base alloys comprising, providing the alloy in molten form, flowing the molten alloy across a cooled hearth disposed within a vacuum region maintained at a pressure of not more than 10- torr, bombarding the surface of the molten alloy with at least one electron beam to create an area of localized superheat on the molten alloy surface, maintaining the power input of the electron beam bombardment above about 20 kilowatts per square foot of hearth, controlling the temperature dilferential between the area of localized superheat and the average temperature of the molten pool to between about 50 F. and about 300 F., and flowing the molten alloy across the hearth at a rate of between about 100 and about 2000 pounds per hour per square foot of hearth.

3. A method for the manufacture of iron base alloys in accordance with claim 2 wherein the molten pool has a depth of not more than about three inches.

4. A method for the manufacture of iron base alloys in accordance with claim 3 wherein the molten alloy flows across the hearth at a rate of between about 200 and 1000 pounds per hour per square foot of hearth.

5. A method for the manufacture of iron base alloys in accordance with claim 4 wherein the electron beam power 18 input is maintained above about 30 kilowatts per square foot of hearth.

6. A method for the manufacture of iron base alloys in accordance with claim 5 wherein the temperature differential between the area of localized superheat on the molten alloy surface and the average temperature of the molten pool is maintained between about 100 F. and about 200 F.

7. A method for the manufacture of iron base alloys in accordance with claim 2 wherein the molten alloy flows across a hearth in at least two pressure zones, the first pressure zone being maintained at a pressure of not more than about 10* torr and the second pressure zone being maintained at a pressure of not more than about 10- torr.

8. A method for the manufacture of iron-chromium alloys in accordance with claim 7 wherein the molten alloy is provided by introducing iron and ferrochrome raw materials into the first pressure zone.

9. A method for the manufacture of iron base alloys in accordance with claim 3 wherein the molten alloy is provided from a basic oxygen furnace, an oxygen lance electric furnace or a vacuum induction furnace.

10. A method for the manufacture of iron base alloys comprising, providing the alloy in molten form, flowing the molten alloy across a cooled hearth disposed within a vacuum region maintained at a pressure of not more than 10 torr, maintaining a depth of the molten alloy on the hearth of not more than about three inches, bombarding the surface of the molten alloy with at least one electron beam to create an area of localized superheat on the molten alloy surface, and controlling the flow rate of the molten alloy and the heat input of the electron beam to reduce the level of the insoluble volatile metallic impurities lead, bismuth, cadmium, sodium, potassium, silver, calcium, magnesium and barium to not more than about 6 p.p.m. total, and to reduce the level of the soluble volatile metallic impurities zinc and antimony to not more than about 20 p.p.m. total. 7

11. A method for the manufacture of iron base alloys according to claim 10 wherein the flow rate and heat input are controlled to reduce the level of carbon and nitrogen to amounts such that essentially all carbon and nitrogen will be in solid solution in the alloy at room temperature.

References Cited UNITED STATES PATENTS 3,352,250 9/1967 Treppschuh 10 3,219,435 11/1965 Gruber 75-65 EB 3,005,859 10/ 1961 Candidus 7565 EB 3,183,077 5/1965 Kraus 7512 L. DEWAYNE RUTLEDGE, Primary Examiner P. D. ROSENBERG, Assistant Examiner US. Cl. X.R.

Images