FILLED TUBULAR ARTICLE AND METHOD FOR CASTING BORON TREATED STEEL

Technical Field This invention relates generally to casting of boron treated steel, and more specifically to an improved filled tubular article and method of controlled insertion of preselected materials within the filled tubular article into the molten metal as it is being cast.

Background Art

Up to the present time a number of difficul¬ ties have been experienced with the continuous casting of boron treated steels because of the need to add boron, alloying elements, deoxidants and denitriders to the molten metal near the time of pouring. For example, such steels have been typically produced by adding aluminum to the molten metal immediately prior to the introduction thereof into the mold. However, the aluminum in these aluminum killed steels forms oxides and a reaction with the silicates in the melt to plug the nozzles that meter the molten metal to the tundish and to the mold. Titanium also forms oxides and tends to plug the nozzles in much the same way. Such slag formation at the nozzles not only detrimentally affects the controlled flow rate of molten metal into the mold, but detrimentally affects the ratio of external surface area of the poured stream to the total stream cross section so that there is undesirably an increased oxidation and nitriding tendency. The formation of slag also detracts from the amount of residual aluminum available for obtaining the desired grain refinement.

The material additions such as titanium, zirconium and boron, in singular or combined form, which have been made to the tundish or ladle to improve the response of the material to heat treatment have heretofore been relatively ineffective because there has been a reaction with the atmosphere and a fading phenomena as a result of prolonged exposure of the additional materials to the atmosphere at» he elevated temperature. In other words, the article \or billet that is formed has a less homogeneous and coarser structure than it should have for the expense of the material additions and a lower hardenability than is desired. Typically, several feet of the continuously cast billets are not usable because of imperfections, and so such sections are cut off resulting in a waste of time and material.

Another approach to minimizing the exposure of the highly reactive additive materials to the atmo¬ sphere is to shroud the outlet stream and/or to use an inert gas such as argon to isolate the stream as it is being poured into the mold. Obviously, this adds a considerable cost to the process.

Less prevalent in the continuous casting of boron treated steels is the introduction of an aluminum alloy rod near the entry to the mold as is recommended by Kawecki Berylco Industries, Inc. of Reading, Pennsyl¬ vania. This late addition of aluminum, as well as preselected amounts of titanium and boron integrally formed with the aluminum, has been only partially satisfactory. This is in part due to the aluminum alloy rod melting prematurely so that there is an undesirable reaction with the atmosphere that decreases the effec¬ tiveness of the material additions, and in part due to a less than satisfactory selection of the basic elements and their proportions.

Even though ^' the advantageous teachings of U. S. Patent No. 3,991,808 issued to J. R. Nieman, et al on November 16, 1976 have been widely recognized, such teachings have not resulted in the formation of a single additive rod capable of effectively making boron treated steel and while taking into account the diverse chemical reactions that occur. Not only do the reaction capabilities of the materials of the rod itself have to be taken into proper account, but also one must consider the reaction thereof to the composition of the molten metal and to the gases in the atmosphere. Furthermore, the flexibility of the rod must be maintained to permit coiling thereof on a containment reel. Hence, the additive rod cannot be too stiff or brittle, and if particulate material is contained within the rod it should be capable of being drawn for improved densifi- cation thereof.

Thus, what is desired is a single rod, capable of being easily handled and fed into molten steel, and containing such hardenability intensifiers and additives as are precisely tailored for the production of fine grain, boron treated steels in an effective and economi¬ cal manner. As one would suspect, the full boron effect is likely to be obtained only with a rod having a preselected construction and material composition. In this regard, it should be recognized that a substantial number of inoculating rods are known to the industry which are constructed specifically for completely different casting purposes. For example, the following U. S. patents disclose inoculating rod constructions which are representative of the general state of the art:

4 -

3,056,190 to D. S. Chishol , et al on October 2, 1962

3,367,395 to S. I. Karsey on February 6, 1968

3,921,700 to J. G. Frantzreb, Sr. on November 25, 1975

4,174,962 to J. G. Frantzreb, Sr. et al on November 20, 1979

However, such inoculating rods are totally unsatisfact for making boron treated steels.

The present invention is directed to overcom one or more of the problems as set forth above.

Disclosure of the Invention

In accordance with one aspect of the present invention a filled tubular article is provided for con trolled insertion into a molten metal for making boron treated steel, the filled tubular article having an elongate conduit, an elongate, non-particulate member of primarily aluminum material in the conduit, and a master composition including ferroboron particulate material in the conduit.

In accordance with another aspect of the invention, a filled tubular article is provided having an elongate ferrous metal conduit, an elongate substan tially aluminum member in the conduit, and a master composition including ferroboron and ferrotitanium particulate materials in the conduit. Preferably, the master composition also includes ferrovanadium particu late material.

In a further aspect of the invention, a method of casting boron treated steel includes intro¬ ducing aluminum in the form of a non-particulate elongate member and ferroboron particulate material in preselected percentages by weight below the surface of the molten steel in the mold by means of a containing conduit. Advantageously, the instant invention has successfully made boron treated steel in a continuous as-cast round bar manufacturing facility by introducing preselected amounts of ferroboron, ferrotitanium, and ferrovanadium particulate materials and an aluminum rod in a protective conduit and effecting melting thereof at a preselected depth below the level of the molten steel in the mold. Its success has been determined by a study of boron factors, performance criteria, and chemical analyses of the elements of a plurality of heat-treated parts including experimental tests and comparison base tests.

Brief Description of Drawings Fig. 1 is a diagrammatic, elevational view of a continuous casting facility including an apparatus for progressively feeding a filled tubular article constructed in accordance with the present invention into the molten metal in the mold. Fig. 2 is an enlarged, fragmentary and diagram¬ matic elevational view of the upper portion of the casting facility of Fig. 1 with a portion illustrated in section to better illustrate details of the present invention. Fig. 3 is an enlarged, diagrammatic, cross sectional view of the filled tubular article illustrated in Figs. 1 and 2.

Fig. 4 is a graph showing the relationship between actual boron factor and carbon content.

Fig. 5 is a tabular listing of seven experi¬ mental rod members and the additive material ratio additions in each.

Best Mode for Carrying Out the Invention Referring to Figs. 1 and 2 a rotary continuous casting facility 10 is illustrated of the type utilized by the MacSteel Division of Quanex Corporation and located in Jackson, Michigan. Such facility produces continuous as-cast round bars by utilizing a large bottom pour ladle 12 to pour argon-gas-stirred molten steel 13 into a tundish 14. Liquid steel is teemed from the tundish via a bent nozzle 16 having a relativel small outlet opening at 18, for example about 16 mm (5/8") dia., and into a water cooled mold 20 at a precise angle with respect to a central vertical axis 22. The generally cylindrical mold 20 is of copper and has a precisely contoured or tapered internal bore 24 to allow for solidification shrinkage and to maintain mold contact with the solidifying hot bar for optimum cooling.

As is diagrammatically indicated in Fig. 2, the copper mold 20 has an enlarged annular head portion or top end 26 and a lower cylindrical body portion or bottom end 28, and vertically spaced apart seal means 30 are provided between the mold and a suitable support member 32 to define a chamber 34 through which liquid coolant such as water is circulated. Through a mechanis not shown, the mold 20 is oscillated at a rate of about 60 cycles per minute through a range of about 16 mm (0.625") in the direction of the vertical axis 22 on which it is centered, while at the same time it is rotated at a speed of about 60 revolutions per minute as is schematically shown in Fig. 1 by the movement indicating arrows "A" and "B" respectively. The emergin

bar or strand depends -from the mold and passes through a water spray system 35. Thereafter the bar is cut to length by a carriage mounted saw, not shown, that clamps to the bar and travels with it during the cut. In actuality, the continuous casting facility

10 so far described and used during the development of the present invention, is a twin strand unit having side-by-side molds 20 and associated nozzles 16 to allow simultaneous manufacture of a pair of bars. Since the construction and operation of each strand is the same, a description of one can suffice for the other and only one unit need be illustrated in the drawings. The straight bar length or overall height "OH" is about 10 m (33'), and the bar diameter "D" can be varied from, for example, about 100 to 180 mm (4 to 7") .

An apparatus for introducing additives into the casting mold 20 is generally indicated by the reference numeral 36. The additives utilized in the present invention are in the form of a relatively ductile filled tubular article or treating rod 38 having a lower or distal end 40. The filled tubular article is progressively urged downwardly when viewing the drawings by a wire feed mechanism 42 which unreels the article from a rotatable reel 44. A feed rate of about 64 mm/sec. (2 y/sec.) was found to be satisfac¬ tory in one instance. A hollow tubular guide member 46 is located below the feed mechanism, and is generally aligned with a plane through the central axis 22; however, the guide member has a preselected angle of inclination with respect to the axis so that the distal end 40 of the filled tubular article is below a surface 48 of the molten steel 13 in the mold 20 by a preselected distance "L" as indicated in Fig. 2 and so that the distal end is adjacent the central axis thereat. For

further details of the apparatus 36 reference is made to U. S. Patent No. 3,991,808 issued November 16, 1967 to J. R. Nieman, et al.

Thus, it can be appreciated that as the molten steel 13 is added to the mold 20 the filled tubular article 38 melts at its distal end 40 to add preselected materials below the surface 48 simultaneous with reciprocation and rotation of the mold. Moreover, heat is removed from the copper mold by the water in the chamber 34, and progressive solidification occurs at the periphery of the tapered bore 24 so that a cylindrical bar 50 is continuously formed along the axis 22. In the instant example the retraction rate or formation rate of the bar is about 2m/min. (79"/min.). It is to be understood that the central part of the bar does not immediately solidify, but rather the solidifi¬ cation progresses radially inwardly with time and with the downward movement of the bar. This phenomena is graphically or schematically portrayed in Fig. 2 by the tapered phantom solidification demarcation lines desig¬ nated by the reference numeral 52. At a distal end 54 of the downwardly converging lines 52 the liquid center of the treated molten metal has solidified. For example the distal end 54 is typically reached at a distance "H" of about 5.5m (18') from the- top of the mold for a bar diameter of about 140mm (5.5").

With reference now to the cross sectional view of Fig. 3, the filled tubular article 38 can be seen to include an elongate metal conduit 56, an elongat non-particulate member 58 located within the conduit, and a preselected particulate master composition 60 compactly contained within the conduit. Specifically, the master composition 60 includes ferroboron, the non- particulate member 58 is primarily of aluminum material, and the conduit 56 is of preferably a ferrous material

-H

O WI

for for ability. For example, the conduit 56 can have the following composition in percentage by weight:

C About 0.10% Mn 0.25 - 0.50% S About 0.05%

P About 0.01% Fe Balance I have determined that the master composition 60 should preferably include preselected weight percent- ages cf ferrotitanium and ferrovanadium particulate materials intermixed with a preselected weight percent¬ age of ferroboron particulate material. I have found it desirable to compact the master composition 60 within the conduit 56 to a relatively dense state in order to assure rapid internal dissolution of the conduit. For example, the preferred density of the core is equivalent to a degree of compaction in excess of 10% above the tapped density thereof. The term "tapped density" as used herein, refers to the known procedure described in "HANDBOOK OF METAL POWDERS" - Poster, Reinhold Publishing Co., New York, New York, 1966, page 57.

More particularly, I have found that boron treated steel can be made best by progressively inserting a filled tubular article 38 consisting essentially of the following elements in the proportions indicated into and below the surface of molten metal in the mold:

Broad Preferred Most

Range Range Desirable

(Wt.%) * (Wt.%)* (Wt.%)*

1) aluminum 0.015 - 0.070 0.015 - 0.030 0.016 portion of member 58

2) boron 0.0008 - 0.00462 0.0011 - 0.00308 0.0018 portion of ferroboron

Λ-. V.'IPO «

3) titanium 0.038 - 0.150 0.038 - 0.088 0.0 portion of ferrotitanium

4) vanadium 0.022 - 0.147 0.044 - 0.103 0.0 portion of ferrovanadium

*Note: Reflected as a percentage of the molten metal addition to the mold.

I make note that the ferrous metal portions of the protecting conduit 56 and the selected three particulate materials designated immediately above is not significant since such ferrous portions have a negligible diluent influence on the molten metal. Rather this compatibility factor can be utilized with advantage because ferroalloys of boron, titanium and vanadium are available in the marketplace at economical prices and because the reaction thereof is more tame than the reaction of the purer basic element forms.

Another way to state the preferred material relationship is to designate the ratio of the four ele¬ ments as about 9:1:28:24 which reflects the weight analysis ratio of aluminum, boron, titanium and vanadium in the filled tubular article 38.

In the aforementioned preferred composition of the novel filled tubular article 38, low cost aluminu serves as an effective deoxidizer and denitrider and imparts the desired degree of grain refinement in the cast article by removing dissolved gases from the melt. Above a level of about 0.070 Wt.% the ductility of the cast article can be expected to be adversely affected and an undesired amount of inclusions noted therein. Below a level of about 0.015 Wt.% the amount would be ineffective as a grain refiner. It must be present in a non-particulate or non-powder form for the reason

O

Λ, Vv-

that if aluminum were present in the form of relatively small particles an excessive amount of external surface area would be provided; such larger surface area would at least partly oxidize even within the confines of the conduit 56, for example, and result in a marked decrease in effectiveness of the aluminum additive. While I prefer that the aluminum member 58 be present as a cylindrical rod centrally located within the conduit 56, I also contemplate that it could be formed as a coating on the inside surface of the conduit. However, if the conduit itself were made of aluminum such construction would be unsatisfactory because the larger surface area thereof would be exposed to the atmosphere and it would melt too fast so that the aluminum and master composition 60 would be prematurely exposed to atmospheric contamination. The aluminum that I used was over 99% pure since it is. commercially available in that form.

Ferroboron particulate material provides the desired degree of hardenability to the steel article while replacing an appreciable percentage of more expensive alloying ingredients. Above a boron level of about 0.00462 Wt.% an undesirable secondary reaction occurs involving the precipitation of iron borides that tend to embrittle the article. Below a level of about 0.0008 Wt.% there is insufficient boron available to provide the hardenability effect on the heat treat of the article. The ferroboron particulate material that I used had 17 1/2 Wt.% boron. The preferred addition of ferrotitanium serves as a powerful deoxidizer and denitrider. Above a titanium level of about 0.150 Wt.% there is so much titanium that some would be available to link up with the carbon and detrimentally affect the heat treatment capability of the cast article and its hardenability. This is so because titanium is an exceptionally strong

/ OMPI ,. Y.'IFO

carbide former. Furthermore, the formation of stable inclusions can occur that would adversely affect achin- ability. Below a titanium level of about 0.038 Wt.% the effectiveness of the boron addition would be minimized since the boron would tend to link with the available oxygen and nitrogen in place of titanium. The ferrotitanium particulate material that I used had 70 Wt.% titanium.

Lastly, the preferred ferrovanadium addition serves as a somewhat weaker deoxidizer, a stabilizer, a hardenability agent, a grain refiner and a means of increasing the strength of the boron steel article. Above a level of about 0.147 Wt.% vanadium there would be massive carbide precipitation that would result in a loss of hardenability. Below a level of about 0.022

Wt. % there would not be the degree of system stability desired; in other words, there would be an excessively large variation in the microstructure and hardenability of the final product article. Moreover, there would be an undesirable loss in strength if the level is below that recommended. The ferrovanadium particulate material that I used had 54 Wt.% vanadium.

With the aforementioned proportions, about 50% of the total weight of the master composition 60 is ferrovanadium, about 43 1/2% is ferrotitanium, and about 6 1/2% is ferroboron.

Industrial Applicability

Initially, experimental tests were conducted on casting boron steel by introducing a filled tubular article including a metallic sheath containing pre¬ selected additives into the molten metal flowing into a casting mold substantially as set forth in U. S. Patent No. 3,991,808 mentioned previously. One of the objects of the testing was to try to obtain a satisfactory boron factor at a reasonable cost, and without adding

undesirable amounts of the additives to the chemical analysis of the final castings.

In connection with the so-called boron factor, reference is made to the pioneering work of Marcus A. Grossman, such as his AIME Paper of February, 1942 on Hardenability Calculations from Chemical Compositions, and to ASTM Specification A255 relating to a standard method of End-Quench Test for hardenability of steel. The actual boron factor is generally defined as the actual D.I. in inches calculated from Jo iny divided by D. I. in inches calculated from the chemistry (without boron). When steel is properly made, the boron factor, or its contribution to increased hardenability, is an inverse function of the carbon content. The higher the amount of carbon, the lower the boron factor and the less the contribution to increasing hardenability. This is observable by reference to the chart identified as Fig. 4, wherein the actual boron factor is plotted in the vertical direction of the ordinate and the carbon content of the steel is plotted in the horizontal direction of the abscissa. A target value or normal expectancy value is represented in Fig. 4 by the substan¬ tially straight shaded band or region identified by the reference letter A. The further that the actual boron factor is below the target value' in the band when viewing the chart, the more undesirable it is.

Initially, steels having a chemistry similar to 10B30 Mod., 41B30, and 10B30 Mod. with high silicon (0.5 - 0.65 Wt.%) were experimentally poured. It was found that there was no apparent correlation between the boron factor and the micro-alloying content, since the 0.25-0.40 Wt.% Cr and 0.08-0.15 Wt.% Mo in the 41B30 alloy steel didn't influence the boron factor. Furnace or ladle additions of aluminum, with and without additions of ferrotitanium and silicon zirconium, were

made to the melt before pouring and various filled tubular articles or rods were inserted into the casting cavity at the time of pouring. For example, some rods contained particulate ferrotitanium and/or silicon zirconium along with particulate ferroboron. The actual boron factors obtained varied from less than 1.0 to 2.07 and fluctuated too widely as may be noted by reference to the zone designated by the reference letter "B" in Fig. 4. From this and the chemical analyses of the various heats the conclusions were reached that furnace and ladle additions were erratic and wasteful, and that ferrotitanium additions within the rod were highly desirable. Furthermore, while zirconium additions within the sheath exhibited some degree of success on hardenability, the cost thereof was excessive for the effectiveness obtained. It was also learned that boron factors did not appear to relate to boron content.

With this background, further experimental tests were conducted using four different filled tubula articles or rods for comparison purposes. These rods were designated as Nos. 1-4 in the chart identified as Fig. 5, and different quantities thereof were melted by the following 10B38 carbon steel composition in percent age by weight as it was poured at a preselected pour temperature into the casting cavity at an average pour rate of about 11.3 Kg/sec (25 lbs/sec):

C 0.35 - 0.43% Mn About 1.13% S 0.050% Max.

P 0.040% Max. Si 0.15 - 0.30% Fe Balance

Identical generally cylindrical ingots were produced and tested for chemistry, oxygen and nitrogen levels, hardenability and microstructure. From these data the actual boron factors were obtained for each ingot heat. The steel heats associated with rod Nos. 1 and 3 were prepared by ladle additions of a preselected quantity of aluminum pellets generally in accordance with conventional practices, while there was no aluminum addition to the ladle during the pouring of the heats of rod Nos. 2 and 4. Rather, in accordance with one aspect of my invention, the aluminum wire 58 was incor¬ porated within the ferrous metal conduit 56 of rod Nos. 2 and 4 and surrounded by the particulate master composi¬ tion 60 including preselected proportions of ferroti- tanium and ferroboron as indicated by Fig. 5. The total weight percentage of the aluminum addition in each of the various heats associated with rod Nos. 1 and 2 was maintained the same, as was the aluminum addition in the heats of rod Nos. 3 and 4. For continuity the total weight percentage of the boron addition was kept constant throughout this stage of the experimental tests.

The results of the tests of rod Nos. 1 - 4 were enlightening. Particularly, it was noted that the amount of aluminum present in the ingots associated with rod Nos. 2 and 4 generally doubled in comparison with the ingots associated with rod Nos. 1 and 3, indicating that an unexpectedly high recovery rate was exhibited. Since the average boron factor dropped from about 1.8 to about 1.60 in comparable heats of rod Nos. 1 and 2, and the average boron factor dropped from about 1.93 to about 1.61 in comparable heats of rod Nos. 3 and 4, there was cause to believe that this was due at least in part to the presence of excessive amounts of aluminum and that the amount of aluminum and

C.-rl

^■ -

titanium needed within the rod could be reduced substan tially and still retain the desired level of harden¬ ability. The extra titanium in the heats of rod No. 3 when compared with the heats of No. 1 caused a higher boron factor, but not enough greater to justify the added expense. Moreover, the amount of retained titani in some of the ingots was noted to be higher than desired, for example above about 0.10 Wt.%.

Several heats using rod No. 5 were thereafter run to take into account the above mentioned factors.

Rod No. 5 differed from the first four rods by containi a preselected quantity of ferrovanadium. Boron factors in the neighborhood of 1.90 were noted indicating a definite success with that rod as a result of the vanadium influence. However, the chemical analyses of the ingots indicated that a relatively high residual proportion of the additive elements was retained and that the ratio of the elements within the rod was therefor too rich. Rod No. 6 was provided to reduce the amount of aluminum and titanium substantially, while keeping the amount of boron constant. Upon examining the ingots thus produced it was found that the same high boron factors of about 1.90 were observed. Thus, unexpectedly good results were obtained with less additive material, and this time the chemical analysis of the ingots indicated only minimal amounts of the additives present. This was the best rod of the seven listed in Fig. 5. Another rod ₇ rod No. 7 was evaluated, and is reported here to show that while columbium or niobium is in the same general family as vanadium, the direct substitution thereof for vanadium is not productive insofar as hardenability is concerned. Specifically, the boron factor dropped so much that such substitution was indicated to be entirely unsatisfactory.

-"

The experience gained by analyzing the ingot castings made by using rod Nos. 1-7 as discussed imme¬ diately above permitted more extensive testing to be conducted with a reasonable promise of success, Accord- ingly, even though boron treated steel utilizing a minimum of additive material and having a reasonably high boron factor had never been advanced to the desired level in a continuous casting facility, three more experimental tests were conducted using plain or low carbon steel, medium carbon steel and medium carbon alloy steel having some chromium and molybdenum therein in the continuous casting facility described with reference to Figs. 1 and 2.

Specifically, the filled tubular article 38 having the preferred No. 6 rod construction (28 parts titanium; 1 part boron; 9 parts aluminum; and 24 parts vanadium) was inserted into a plain carbon steel having the following element analysis of primary interest in percentage by weight:

C 0.17%

Mn 0.85%

Si 0.25%

S 0.022%

Al 0.008%

P 0.005%

The filled tubular article 38 was inserted into the mold 20 as indicated in Fig. 2 at the rate of about 64 mm/sec. (2 "/sec) , while the steel at about 1530°C (2790°F) ^* was poured into the mold at a rate of about 4.60 Kg/sec. (10.11 lbs./sec) . The filled tubular article at almost 8 mm dia. (5/16" dia.) exhibited a dissolution depth "L" of about 400 mm (16") . This corresponded to a rate of material addition of about 0.020 Wt.% Al, 0.0023 Wt.% B, 0.063 Wt.% Ti and 0.055 Wt. % V, reflected as a percentage of the molten metal addition to the mold. This enabled withdrawal of the

- IB -

cylindrical bar 50 at a diameter of about 150 mm (5.875" at a rate of about 2m/min. (79"/min.) . The result was the formation of an article having a highly desirable actual boron factor of 2.43, as indicated by the letter "C" on the graph of Fig. 4. This was achieved with reasonable levels of the additive ingredients remaining in the article. For example, 0.027 Wt.% Al, 0.0011 Wt.% B, 0.04 Wt.% Ti, and 0.040 Wt.% V were noted in the bar. Secondly, the same filled tubular article 38 was inserted at approximately the same rate into a medium carbon steel having the following element analysi of primary interest in percentage by weight:

C 0.35% Mn 0.82%

Si 0.29% Cr C.06%- S 0.023% Al 0.008% (Est.) P 0.008%

In this second instance the corresponding rate of materi addition was also about 0.020 Wt.% Al, 0.0023 Wt.% B, 0. Wt.% Ti, and 0.055 Wt.% V, and provided an article havin boron factor of 2.03 as indicated by the letter "D" on t graph of Fig. 4. This was achieved at a final chemistry retention level of about 0.03 Wt.% Al, .0020 Wt.% B, 0.0 Wt.% Ti and 0.06 Wt.% V.

In the third instance, the same filled tubular article 38 was inserted into a medium carbon alloy steel having the following element analysis of primary interest in percentage by weight:

/. IP

C 0.34% Mn 1.06% Cr 0.30% Si 0.25% Mo 0.10%

S- 0.020% P 0.012% Al 0.008% (Est.) In the third case the corresponding rate of material addition was reduced to about 0.015 Wt.% Al, 0.0016

Wt.% B, 0.045 Wt.% Ti, and 0.040 Wt.% V as reflected as a portion of the molten metal addition to the mold. The leaner mixture provided an excellent actual boron factor of 2.16 as indicated by the Letter "E" on the graph of Fig. 4 at a final chemistry retention level of about 0.02 Wt.% Al, 0.0012 Wt.% B, 0.03 Wt.% Ti, and 0.035 Wt.% V.

Hence, it can be appreciated that the filled tubular article and method for casting boron treated steel in accordance with the present invention is extremely successful by providing high boron factors, by providing substantially the lowest practical levels of material additions at a late stage to reduce fade and contamination of the melt, and by providing a manufactured article with relatively low chemistry weight percentage levels of the additive elements. The articles thus produced have exhibited an extremely desirable clean microstructure morphology and/or a minimum of nonmetallic inclusions that are often charac- terized as "dirt". This is indicative that the recovery rate is high, and the process economically efficient.

Other aspects, objects and advantages of this invention can be obtained from a study of the drawings, the disclosure and the appended claims.