US3445223A - Alloy for addition of columbium to steel - Google Patents

Description

May 20,1969 0. w.- P. LYNCH ALLOY FOR ADDITION OF COLUMBIUM TO STEEL Filed Jan. 6, 1967 FlrdEBlQflbidH D Cb-Mn-Si Alloy Ht.35 1996M" Ferrocolumbium Alloy Average Particle Size in Inches W A. i, .5 W h O m w w a m w m 5 o mucouwm E uE; 523cm INVENTOR. Dunstan W. I? L ync/7 United States Patent U.S. Cl. 75-123 5 Claims ABSTRACT OF THE DISCLOSURE A columbium-manganese-silicon alloy consisting essentially of 20 to 30 percent columbium, 15 to 30 percent manganese, 15 to 35 percent silicon, and the balance iron and incidental impurities.

This invention relates to a novel columbium-manganese-silicon alloy and, more particularly, to a novel alloy of the type described for ladle additions of columbium to steel which is useful in the production of fine grain steels and has a fast dissolution rate and yields high columbium recovery.

The use of small columbium additions to medium and low carbon steels has increased significantly in the past few years. This increase has been predominantly in two types of steels. First, columbium has been added in small quantities to low alloy high strength steels, normally marketed as hot rolled plate, to promote yield strengths higher than normal. Second, columbium has been added in small quantities to steels made by continuous casting to produce a fine grain steel. In conventional ingot casting fine-grained skilled steels are normally produced by the addition of small amounts of aluminum during tapping. However, the use of aluminum in sufficient quantities to produce the grain size required for fine grain steel causes serious problems in continuous casting operations. Columbium, on the other hand, can be used to produce fine grain steels without causing the difficulties in continuous casting resulting from the use of aluminum.

At present, columbium is added to the molten steel in the ladle during tapping in the form of ferrocolumbium. However, there are numerous disadvantages in the use of ferrocolumbium as a source of columbium. First, ferrocolumbium has a density significantly greater than steel and, therefore, sinks to the bottom of the ladle. Second, ferrocolumbium has a relatively slow dissolution rate in molten steel at the normal tapping temperature of steel. Finally, columbium is a reactive metal and is really oxidized. The foregoing properties of ferrocolumbium result in incomplete solution of ferrocolumbium in the steel, inconsistent and low columbium recoveries in the steel, and non-uniform distribution of columbium within the heat. Since the columbium additions are small, usually on the order of .02 to .04 percent, the columbium levels in the final steel are critical in order to impart the desired properties to the steel, it is apparent that the use of the columbium as ferrocolumbium results in many off-grade heats of steel.

My invention provides a novel alloy containing columbium which has a high rate of solution in molten steel, which has a density more nearly that of molten steel than prior columbium alloys, and which enables the alloy to give consistent and uniformly high recoveries of columbium when added to molten steel. In addition, my novel alloy is more effective in producing fine grain steels than ferrocolumbium.

In the single drawing I have shown in graph form the "ice relative solution rate for alloys coming within the scope of my invention and standard ferrocolumbium.

Alloys within the scope of my invention have a composition falling within the range of 20 to 35 percent by weight columbium, 15 to 30 percent by weight manganese, 15 to 35 percent by Weight silicon, and the balance iron and incidental impurities. Within these ranges, I have found it desirable to avoid alloys having both a high columbium and high silicon content, i.e., alloys in which the columbium content is above 25 percent and the silicon content above 30 percent. In alloys of this composition a crystalline compound having a composition of about 35 percent columbium, 17 percent manganese and 32 percent silicon can form. This compound makes the alloy extremely friable, and, if that crystalline compound is present in sufficient quantities, it can cause disintegration of the alloy resulting in a very fine powder which, although satisfactory for addition of columbium to steel, is difficult to handle and therefore is not generally desirable for commercial usages. For the foregoing reasons, I prefer a low columbium alloy having a composition of about 22 to 26 percent columbium, about 23 to 27 percent manganese, about 24 to 28 percent silicon, and the balance iron and incidental impurities, or a. high columbium alloy having a composition of about 31 to 35 percent columbium, about 23 to 27 percent manganese, about 16 to 20 percent silicon, and the balance iron and incidental impurities. The former alloy has an apparent density of about 5.4 grams per cubic centimeter and a melting point of about 2650 F. The latter alloy has apparent density of 6.5 grams per cubic centimeter and a melting point of about 2900 F. The denser, higher columbium alloy, would normally be preferable, However, it gives lower columbium recoveries when producing the alloy, and therefore, makes it somewhat more expensive than the lower columbium alloy which gives much higher columbium recoveries.

I also prefer that the manganese be kept in a definite relationship with the columbium and silicon. I have found that it i desirable that the ratio of columbium plus silicon to manganese should be between 3 to 2 and 3 to 1.

The best results have been obtained when the ratio is about 2 to 1.

The alloys of my invention may be produced by conventional ferroalloy smelting procedures. I have successfully made alloys within the scope of this invention by induction melting of ferrocolumbium, electrolytic manganese and ferrosilicon. Alloys within the scope of this invention have also been produced by silicon reduction in an electric refining furnace. Utilizing this latter procedure, an electric refining furnace was charged with columbium ore concentrate, manganese ore, steel scrap, ferrosilicon and pebble lime. One hundred eighty-seven pounds of an alloy having an analysis of 25.9 percent columbium, 25.87 percent manganese, 26.91 percent silicon,

.94 percent carbon, and the balance iron was produced. The columbium recovery was 86.8 percent.

To demonstrate the effectiveness of my novel alloy, tests were made comparing the rate of solution of alloys within the scope of my invention with the rate of solution of standard ferrocolumbium in A181 1040 steel. The composition of the alloys tested is set forth in Table I. The tests were conducted in an induction furnace with a constant power input to maintain a definite temperature level prior to making the alloy addition. The alloys to be tested were crushed, then sized to close mesh fractions. Sutficient quantities of each alloy to add .06 percent columbium to the steel were poured onto the slag free surface of a molten steel bath held at 2850 F. The time for the alloy to dissolve was recorded with a stop watch. The

solution time for the alloys tested are set forth in Table II and the results are shown graphically in the drawing.

4 a small amount of aluminum had to be added with the standard ferrocolumbium to produce a similar grain size.

TABLE I Approximate Pounds per ton liquidus alloy addition, temperatures, Density, .60% Ch Cb Mn Si Fe F.) gmsJcc. added Ferro Columbium 66. 7 1. 8 1. 6 Bal. 3, 250 8.0 1. 8 CbMnSi-Alloy Ht. 35 34. 1 25. 8 18. 1 Ba]. 2, 900 6. 4 3. 5 CbMnSi-Alloy Ht. 34 22. 7 24. 7 33. 7 Bat. 2, 650 5. 5 5. 3

TABLE III TABLE II.SOLUTION TIME IN SECONDS FOR .0t5% CO- Co1um LUMBIUM ADDITION TO AISI 1040 STEEL AT 2850 F. Analysis in percent bium Ladle addition to 135 ton recovery Grain Average alloy Cb-Mn-Sl alloy Cb-Mn-Si alloy particle size (in.) Ferrocolumbium Ht. 35 Ht. 34 15 beats Cb percent sue 3.71b./T CaSi+.0307 Ob as L042 6 0 FeCb+.025% alum iuumun .027 .018 90 -7 1 1% 3.71b./T CaSi+.Ol5% Cb as :290::::::::::::::::::::::::::::::::: 1535 2 20 From the foregoing it is apparent that I have discovered The results of the solution tests clearly show that under identical conditions my novel columbium manganese silicon alloys dissolve in molten steel in a fraction of the time it takes for ferrocolumbium to dissolve. In addition, as the particle size of the alloy increases, the time required to dissolve ferrocolumbium increase rapidly, While only a slight increase in time of solution is observed for my novel alloy as the particle size increases.

In addition to the improved solution rate, there was a noticeable difference in the appearance of the steel bath when using my novel alloys. After solution tests with the ferrocolumbium alloy, a tight oxide film covered the surface of the bath. When testing the novel alloy covered by this application, the surface of the steel bath was clean and free from any oxide film.

Tests were also made to determine the effectiveness of addition of my novel alloy for grain size control in the production of fine grain steel. The tests were conducted on 135 ton heats of A181 1050 steel produced in a basic oxygen furnace. Two heats were tapped at about 2920 F. During the tapping operation, silicon and manganese additions were made to the ladle to bring the melt within the specification. Thereafter, one ladle was treated with 3.7 pounds per ton calcium silicide plus .03 percent columbium as ferrocolumbium and .025 percent aluminum. A second ladle was treated with 3.7 pounds per ton calcium silicide and .015 percent columbium in the novel alloy of my invention. The results of the tests are set forth in Table 1H. They show clearly that only one half the amount of columbium was needed to produce fine grain steel when added in my novel alloy as when the columbium was added to the same grade of steel using ferrocolumbium. In addition, no supplementary deoxidation with aluminum was needed to produce fine grain steel when the ladle was treated with my novel alloy, Whereas a novel alloy for adding columbium to steel which has a high rate of solution in molten metal and will give consistent and uniformly high recoveries of columbium when r added to molten steel.

While I have described a preferred embodiment of my invention, it may be otherwise embodied within the scope of the appended claims.

I claim:

1. A columbium manganese silicon alloy consisting essentially of 20 to 35 percent by weight columbium, 15 to 30 percent by weight manganese, 15 to 35 percent by weight silicon, and the balance iron and incidental impurities.

2. The alloy defined in claim 1 having a composition of 22 to 26 percent columbium, 23 to 27 percent manganese, 24 to 28 percent silicon, and the balance iron and incidental impurities.

3. The alloy defined in claim 1 having a composition of 31 to 35 percent columbium, 23 to 27 percent manganese, 16 to 20 percent silicon, and the balance iron and incidental impurities.

4. The alloy defined in claim 1 in which the ratio of columbium plus silicon to manganese is between 3 to 2 and 3 to 1.

5. The alloy defined in claim 1 in which the columbium content does not exceed 25 percent when the silicon content is about 30 percent.

References Cited UNITED STATES PATENTS 2,999,749 9/1961 Saunders et a1. 129X CHARLES N. LOVELL, Primary Examiner.

US. Cl. X.R.

Images