US3721587A - Low carbon,niobium and aluminum containing steel sheets and plates and process - Google Patents

Abstract

AN IMPROVED MILD CARBON STEEL WHICH EXHIBITS AN UNSUAL COMBINATION OF PROPERTIES INCLUDING SATISFACTORY HIGH YIELD STRENGTHS, GOOD TOUGHNESS AND WELDABILITY, AND SUPERIOR COLD FORMABILITY IN THE HOT ROLLED CONDITION. THE STEEL, WHICH IS CHARACTERIZED BY MICROSTRUCTURES WHICH ARE VIRTUALLY FREE OF ELONGATED SILICATE INCLUSIONS, HAS THE FOLLOWING COMPOSING BY WEIGHT: CARBON: 0.02 TO 0.08% MANGANESE: 0.25 TO 0.80% SILICON: 0.05% MAXIMUM NIOBIUM: 0.025 TO 0.010% PREFERABLY 0.025 TO 0.045% ALUMIMUM: 0.005 TO 0.025%, PREFERABLY 0.008 TO 0.018% IRON AND IMPURITIES, BALANCE.

Description

United States Patent US. Cl. 148-36 14 Claims ABSTRACT OF THE DISCLOSURE An improved mild carbon steel which exhibits an unusual combination of properties including satisfactory high yield strengths, good toughness and weldability, and superior cold formability in the hot rolled condition. The steel, which is characterized by microstructures which are virtually free of elongated silicate inclusions, has the following composition by weight:

Carbon: 0.02 to 0.08%

Manganese: 0.25 to 0.80%

Silicon; 0.05% maximum Niobium: 0.025 to 0.10% preferably 0.025 to 0.045%

Aluminum: 0.005 to 0.025%, preferably 0.008 to Iron and impurities, balance.

The present application is a continuation-in-part of our copending application Ser. No. 749,273, filed Aug. 1, 1968 for Low Pearlite Ferritic Steel Sheet or Plate and Process, now abandoned.

In US Pat. 3,010,822 granted Nov. 28, 1961, for Columbium Containing Steels, Processes for Their Manufacture and Articles Prepared Therefrom, Altenburger et al. describe a silicon semi-killed steel having a carbon content of from 0.05 to 0.30%, a manganese content of from 0.25 to 1.50%, a silicon content of 0.10% maximum and a columbium content of from 0.005 to 0.20%. The novelty of this disclosure was that it showed how it was possible to produce a steel having exceptionally high tensile properties, especially yield strength, in a relatively inexpensive manner. Since its publication, the demand for steels containing columbium or, as it is alternately called, niobium, has risen markedly. For example, even as early as 1965, Metal Progress for August of that year, in a data sheet on page 88, listed numerous steels having compositions within the patented range. However, an important limitation in the practical embodiment of the invention was that the minimum carbon content of all the steels listed was 0.12% rather than 0.05%. Possibly, this reflects the still widely held opinion among many metallurgists, especially in the United States, that the use of lower carbon contents will irreparably impair the tensile properties of the steel. Or perhaps, it is indicative of the then dominant position of the open hearth process in which it becomes increasingly more difficult and thus more expensive to produce steel as the limits on carbon content are decreased below about 0.12%. Yet, in any case, it will be appreciated that with the exception of this limitation, the invention of Altenburger et al. was generally accepted as a practical solution to the problem of economically producing high strength steel.

In the very early days of the development of these steels, it was generally considered that their exceptionally high tensile properties were the result of the significant microstructural improvements which attended the addition of the niobium. For example, a niobium containing steel will normally exhibit a substantialy finer ferritic grain size than an otherwise similary processed steel of the same composition but without niobium. Consequently, since it was widely recognized that fine ferritic grain sizes contribute to improved toughness, it was anticipated that the niobium steels would necessarily exhibit good impact properties. However, practical experience with the steels soon proved that in many applications the opposite was true. The situation was especially critical with respect to the higher strength grades although in applications involving heavy gauges, difiiculties were also encountered even with the lowest strength grade. For example, it was not at all unusual to observe Charpy V-notch impact transition temperatures at the 15 foot-pound level which were well in excess of F. in such cases. (The Effect of Small Niobium Additions to Semi-killed Medium Carbon Steels, C. A. Beiser. It is an American Society of Metals Preprint No. 138, 1959.)

A satisfactory explanation of the apparently paradoxial behavior of the niobium steels with regard to grain size and toughness was made available in 1963 in a report by W. B. Morrison in the Journal of the Iron and Steel Institute for April of that year, entitled The Influence of Small Niobium Additions on the Properties of Carbon- Manganese Steels. In this article, Morrison presented conclusive evidence to show that niobium produces a finely divided dispersion of carbide and nitride precipitates which both harden the ferritic matrix and refine the grain size. By careful analysis he was able to separate the contributions which precipitation hardening and grain size had on mechanical properties. Significantly, his data showed that it was the precipitation hardening effect which was mainly responsible for the substantial increases in tensile properties, especially in the yield strength. Grain refinement was also shown to have a beneficial effect but it was relatively small in comparison with that attributable to precipitation. The further indications of the analysis were that both precipitation hardening and grain size were also contributing to impact properties. In this case, the data again showed that grain refinement was having a beneficial effect but, contrary to the situation with tensile properties, that precipitation hardening was having a significant adverse eifect. Thus, it was evident that the unexpectedly low impact properties of the niobium steels, in spite of their relatively fine ferritic grain sizes, were a consequence of the predominance of precipitation hardening as the major strengthening mechanism, especially in the higher strength steels.

It will be appreciated that knowledge of the cause of a problem does not necessarily suggest a solution. This has been especially true in the case of the niobium steels. Attesting to this is the fact that much of the development of these steels in the last decade, both here and abroad, has been directed toward improving the toughness of the higher strength grades.

Indicative of the work done in the United States is U.S. Pat. 3,102,831, granted Sept. 3,1963 to N. F. Tisdale for Production of Columbium Containing Steels in which it is shown how it is possible to increase impact properties by combining rolling to a finishing temperature in the range of 1550 to 1750 F. with the use of water sprays subsequent to the last rolling pass. In the practical application of this method, some steel producers have gone to great lengths to achieve the maximum improvements possible with it. An example is the installation by a number of producers of the so-called laminar flow water cooling system in which it is reportedly possible to achieve cooling rates as high as 35 F./second in 0.250 inch gauge sheets from finish rolling temperatures in the neighborhood of 1600 F. Interestingly, however, steelmakers in this country have generally tended to avoid major compositional changes and in this connection it is of some significance to note that the previously mentioned lower limit of 0.12% on carbon content is still widely employed.

Considerable development of the niobium steels has also occurred outside of the United States. Possibly most significant has been the work done in Great Britain. The studies there were apparently fostered as much by a desire to increase weldability as to increase toughness. The reason for this is thought to be connected with the fact that in an effort to take advantage of its beneficial effects on toughness, British made steels have traditionally been produced with relatively high manganese contents and thus, as may be deduced from the well known expression for carbon equivalent (C+1/6 Mn), have on occasion, given trouble in welding. In any case, in their approach to developing the niobium steels, an attempt was made to maintain the benefits of manganese and simultaneously improve both weldability and toughness by going to relatively low carbon contents. The studies with this approach eventually culminated in the development of niobium steels consisting essentially of up to 0.10% carbon, 0.85% to 2.5% manganese, 0.001 to 0.03% nitrogen, up to 0.5% silicon and at least one of niobium (0.01 to 0.20%) and vanadium' (0.01 to 0.30%) as described by R. Phillips in British Pat. 1,050,- 146, granted Feb. 8, 1967 and later in US. Pat. 3,472,- 707, granted Oct. 14, 1969. Subsequent to Phillips initial work, Duckworth, Phillips and Chapman in The Journal of the Iron and Steel Institute for November 1965, in an article entitled, Pearlite Reduced Structural Steels, describe how it is possible with steels of the above composition to achieve further improvements in mechanical properties by using methods similar to those employed by Tisdale. Duckworth et al. also showed that there was some advantage to mechanical properties to the use of heavy final reductions in rolling of the order of 30%. In addition, from an analysis of their findings, which included the determination of the mechanical properties of the steels with a variety of compositions and, in particular, based upon test data which they obtained indicating unsatisfactory impact resistance in steels of relatively low manganese contents, Duckworth et al. conclude at page 1114, A manganese content of not less than 1.2% is needed for good properties.

The prior art also contains indications that the Japanese steelmakers have followed along similar lines in the development of the niobium steels as those taken in Great Britain. For example, in British Pat. 1,062,754, granted Mar. 22, 1967, Yawata Iron & Steel Company of Japan, describe a Process for Producing Steel Pipes for Low Temperatures. Although the steels which are described in the specification are aluminum killed and are for use in the heat treated condition, they have compositions similar to those of Phillips in that a low carbon content, limited to 0.10% maximum, and a relatively high manganese content in the range of 1.0 to 1.5% is used.

From a consideration of the prior art, it will be evident that considerable emphasis in the work which has been done in developing the niobium steels has been placed on improving toughness and weldability, but that little if any attention at all has been given to improving their cold forming properties. Perhaps, this apparent lack of interest reflects the fact that formability has at best transient usefulness in steel in comparison with the other properties mentioned. In other words, formability is only useful for the short time it takes to fabricate a part, whereas strength and toughness are useful throughout the life of the part, and even weldability, although also a property of temporary usefulness, may contribute to premature failure in the event that, owing to poor weldability, hidden flaws, such as underbead cracks, occur in the part and thereby weaken it. Thus, it may be that the formability of the niobium steels has been largely ignored because of what logically appear to be more pressing needs elsewhere. It will be appreciated, therefore, that this apparent neglect of the formability properties does not necessarily imply that forming difficulties do not exist. On the contrary, as those skilled in the art are well aware, the fabrication of the niobium steels is subject to a wide variety of problems and limitations owing to deficiencies in their forming characteristics.

To indicate the state of the art with respect to the niobium steels, it will be valuable to note the difiiculties and problems which arise in a particular forming process as opposed to many forming processes. The reason for this is that while the details may dilfer from process to process, the problems which are encountered are generally similar. For example, the steel either deforms locally rather than uniformly, or cracks, thus introducing in either case a defect which cannot be tolerated in the final part. Of the two most common forming processes, drawing and bending, the latter will be discussed since it is the least complex from a technical standpoint.

An important distinction which must be made when considering bending, is the orientation of the steel during forming with respect to the direction of the prior mechanical work of rolling. When the circumferential direction of the bent part coincides with the rolling direction of the steel, that is, the direction of maximum extension in rolling, the steel is said to have undergone longitudinal bending. When, on the other hand, the cir cumferential direction of the bent part is oriented at right angles to the rolling direction, the steel is said to have undergone transverse bending. Since rolling deformation elongates malleable second phase particles, such as, for example, non-metallic silicate inclusions, and since the existence of these particles greatly weakens the matrix against stresses acting transversely to them, hot rolled steels normally exhibit anisotropy with respect to ductility, with the lowest ductility values being observed when the steel is deformed in the transverse direction. Thus, it will be appreciated that transverse bending involves substantially greater ditficulties than longitudinal bending.

In the early days of the development of the niobium steels, it was anticipated that they would exhibit good formability in bending. It was thought that formability, like most of the other mechanical properties of steel, increased With decreasing grain size and, of course, as has already been pointed out, the niobium steels exhibited relatively fine grain sizes. However, these expectations have generally not been realized. Instead, as in the case of toughness, it has been found that the formability in bending decreases markedly with increasing gauge and strength level. Nor have the processing methods which Tisdale used to improve toughness been observed to have much effect on formabiilty.

One exception to this is steel which subsequent to the last rolling pass and prior to coiling, is subjected to the ultra high cooling noted available in the previously mentioned laminar flow cooling system. Such steels have been reported to possess improved formability. However, as a general rule most niobium steels are rather marginal in this respect. An indication of the limits which the industry has set on the formability in bending of these steels is available in an American Iron and Steel Institute Steel Products Manual on High Strength Low Alloy'Steel and High Strength Intermediate Manganese Steel, published in October 1967. At page 6 of the manual, in a section entitled Recommended Fabricating Practices, a table is presented which lists the minimum inside radii which are suggested for use in the bending of Cb or V Bearing High Strength Low Alloy Steel. It is indicated in the accompanying exposition of the practices that these suggested radii refer to transverse bending. An excerpt of the table is shown below. As those skilled in the art will agree, these radii are fairly liberal and thus there is considerable room for improvements. T is the thickness.

Suggested minimum inside radius Grade Grade Grade Grade Grade Grade Grade Thickness of material, inches 42 1 To 0.180 incl 1T 1%1 2T 2%1 3T 4T Over 0.180 to 0.2299 incl. (sheets) 1%1 2T 2%1 3T 3%1 5T 0.500 max. (plates) 2T 2T 2%1 3'1 3%T 4T 61 Over 0.500

1 Minimum yield strength in 1000 p.s.i.

2 Hot forming recommended for angle bending over 0.500 in. thick.

The literature describing the work which has been done in Great Britain and in Japan on the niobium steels contains little or no mention of formability. However, it is thought that the method of reducing the carbon content to improve toughness and welda'bility has very likely resulted in concomitant improvements in formability, yet, in view of the fact that these steels contain relatively high manganese contents and apparently require special processing to develop optimum properties, both of which add substantially to their manufacturing costs, it is doubted that they will find much practical application in situations where formability is the main problem. The reasons for this are primarily economic ones but there are, in addition, certain technical aspects which are also important to bring out.

To exemplify the situation which is likely to prevail, consider a job shop which has been making parts from niobium steels for several years. The fabricator in this instance may be experiencing difficulties in formability but is otherwise satisfied with the strength, toughness and weldability of the steels. Moreover, he may be committed to the use of a steel of a particular composition range and set of mechanical properties by a part specification over which the steel producer has little, if any, control. Thus, to solve this problem it is necessary for purely technical reasons that the new steel should not represent too wide a departure from the prior art except as regards formability. Furthermore and of foremost importance, the new steel must be economic in comparison with the one that it replaces. Generally, this means that it will be necessary to olfer it at identically the same costs or, at best, at a very slightly increased cost. The reason for this is that the cost which the fabricator incurs in mitigating formability problems is generally not what causes him to seek improved formability. Instead, it is essential that he maintain his competitive position by giving reliable service and the unexpected necessity to stress relieve a steel in order to successfully form it, or to repair parts made from it, or to wait for its replacement in the event that repairs are impossible, all represent time losses which make it increasingly more difficult for him to meet his contractual obligations. Thus, what the fabricator really wants is greater dependability and very often all he has or can offer in return for it, is a greater share of his business.

Commencing in 1963, we have, on our own part, undertaken an extensive investigation of niobium steels in an effort to develop improved mechanical properties. In view of the need for improvements in formability as well as toughness and in recognition of the various practical problems which were likely to be involved, we directed our investigative efforts toward the development of a steel with improved formability and toughness which would also be economical to manufacture and which would otherwise exhibit satisfactorily high yield strengths and satisfactory or improved weldability. By use of relatively low carbon and manganese contents and somewhat increased niobium contents, and by using aluminum in the deoxidation practice so as to eliminate, or, at least, greatly reduce the occurrence of elongated silicate inclusions which are detrimental to both formability and toughness, we have determined that it is possible to produce a steel which meets the above requirements. For the sake of economy, the steel is made in the partially deoxidized or so-called semi-killed condition, but it can also be made in fully deoxidized or killed condition. Thus, in an embodiment of our findings, what we provide that is new and different is an economical aluminum containing niobium steel which exhibits a unique combination of properties including satisfactorily high yield strengths, good toughness, exceptional cold formability, and by virtue of its relatively low carbon and manganese contents, superior weldability. Specific objectives of the invention appear below.

A purpose of the invention is to provide an economical steel sheet or plate in the hot-rolled condition which is capable of undergoing exceptionally difficult cold forming without the need for stress relief.

.A further purpose is to provide an economical steel sheet or plate in the hot-rolled condition having the aforementioned characteristic which is also weldable without the need for preheat as indicated by an uncommonly low carbon equivalent of not greater than 0.30.

A further purpose is to provide an economical steel sheet or plate in the hot-rolled condition having each of the aforementioned characteristics which also has a yield strength of 45,000 p.s.i. or higher.

A further purpose is to provide an economical steel sheet or plate in the hot-rolled condition having all of the aforementioned characteristics which also exhibits transverse Charpy V-notch impact values of at least 40 foot-pounds at plus F. and 15 foot-pounds at plus 32 F.

A further purpose is to manufacture a steel having the aforementioned characteristics by using aluminum in the deoxidation practice so as to eliminate or, at least, greatly reduce the occurrence of elongated silicate inclusions which would otherwise detrimentally affect the formability and toughness properties, especially in the transverse direction of the hot-rolled sheet or plate product.

A further purpose is to economize in the production of a virtually silicate free, cold formable, weldable, high strength, impact resistant steel sheet or plate by producing the steel in either the aluminum semi-killed or aluminum treated silicon semi-killed condition having the following composition by weight:

Carbon: 0.02 to 0.08%

Manganese: 0.25 to 0.80%

Silicon: 0.05% maximum Niobium: 0.025 to 0.10%, preferably 0.025 to 0.045% Aluminum: 0.005 to 0.025%, preferably 0.008 to 0.018% Iron, and incidental constituents, balance.

A further purpose is to produce a low carbon niobium containing sheet or plate Which has far superior bendability as set forth more in detail below.

Further purposes appear in the specification and in the claims.

The incidental constituents which may be present in the steel are the normally occurring impurities which arise as a consequence of the basic steelmaking practice, such as sulfur (typically up to 0.030%), phosphorus (not exceeding 0.02%), and nitrogen (between about 0.002% and 0.01%) depending upon the type of furnace employed in making the steel. Copper (suitably up to 0.25%) may also be purposely added for corrosion resistance or may occur as a residual in somewhat lower concentrations as a result of the use of low grade or inexpensive scrap in the steelmaking process.

To achieve the optimum properties possible with the steel, it must be manufactured correctly and this requires a basic understanding of the purposes and effects of certain of the practices which are used. For example, the effects of the aluminum deoxidation practice are possibly not widely known, at least in the context of niobium steels. In addition, one of the novel characteristics of the steel is that its response to finish rolling practice is contrary to what has generally been observed in the prior art. Thus, it will be evident that it will be necessary to discuss the manufacturing process with special reference to certain practices in order to elucidate properly the nature of the steel.

The steel, according to the invention, may be derived from a basic oxygen furnace, an electric arc furnace, or an open hearth furnace as desired, although its manufacture may be less economic with the latter two than with the former. The alloying additions may be made partly to the furnace and partly to the ladle, or entirely to the ladle, or partly to the ladle and partly to the molds. Without exception, the manner of the additions, as to when and where they are made, is not important provided the final composition is in accordance with the invention. Yet, it is appreciated that certain practices are better suited to one type of furnace than to another. Thus, the following practices which are suggested to be used in the deoxidation of the steel are particularly applicable to the basic oxygen furnace and while they can also be used with either of the other two melting processes mentioned, it will be evident to those skilled in the art how it may be possible to achieve the same results with somewhat different practices.

The steels, according to the invention, are made semikilled with the deoxidation being accomplished either by the use of aluminum or by the use of a combination of aluminum and silicon. If only aluminum is used, then an object of the practice should be to achieve a total aluminum content in the steel of between 0.005% and 0.025% and preferably between 0.008% and 0.018%. The addition should be made entirely to the ladle, but either aluminum pig, ferroaluminum or any of the various cold compacted aluminum-manganese composites may be used as preferred. As an example, satisfactory results have been achieved, assuming a recovery of 25% and using aluminum pig in the amount of slightly less than 1.5 pounds per ton, as for instance, 200 pounds per 140 ton heat.

If both aluminum and silicon are to be used, then an object of the practice should be to achieve a silicon content in the range of 0.03% to 0.05%, independently of the manganese content, and a total aluminum content in the range of 0.005% to 0.015%. In this case, the silicon additions should be made entirely to the ladle, using any suitable source of silicon, such as ferrosilicon or silico-manganese. The aluminum additions, on the other hand, may be made entirely to the ladle, using any one of the aforementioned sources or entirely to the molds using either aluminum shot or any equivalent finely divided source of commercially pure aluminum.

The advantage to using a combination of aluminum and silicon is that the aluminum recoveries are increased, thus affording a greater control of the final aluminum content, and thereby, of the state of oxidation of the steel. The best results in this connection are achieved when the aluminum is added to the molds. For example, steels according to the invention have been satisfactorily produced with both aluminum and silicon with the aluminum being added entirely to the molds in the amount of 6 ozs./ton. Slightly greater aluminum additions of /2 lb./ton are suggested for the case where the additions are to be made entirely to the ladle.

The objective of making the steel semi-killed is to avoid the relatively uneconomic alternative of producing the steel in the completely deoxidized or killed condition. When a steel is cast in the latter condition, it is necessary to employ special practices involving the use of so-called hot tops and hot topping compounds, in order to minimize the depth of the centrally located shrinkage cavity which arises on solidification. These practices have the effect of increasing the ingot to slab yield, by reducing the amount of steel which has to be discarded for the reason that it is lacking in soundness. Yet, even with these practices, the ingot to slab yields which are obtained are generally inferior to those which can be obtained when the steel is made semi-killed. Thus semikilled steels are more economical.

In a semi-killed steel, the volume which would otherwise be associated with the shrinkage cavity is distributed throughout the body of the ingot in the form of relatively small isolated cavities which can be eliminated by subsequent rolling. This condition is present by virtue of the partially deoxidized state of the steel at the time of casting, which causes the evolution of carbon monoxide during solidification.

The state of oxidation of a semi-killed steel is critical in that the volume of the carbon monoxide which is evolved should be just sufiicient to offset the volume of the shrinkage cavity which would otherwise form. If the steel is underdeoxidized, then too much carbon monoxide will form. One way in which underdeoxidation may manifest itself is in the formation of porosity in or near to the surface of the ingot which eventually causes unsightly and unacceptable defects in the surface of the rolled product. If, on the other hand, the steel is overdeoxidized, then the practice will again suffer due to relatively low ingot to slab yields. The reason for this is that too little carbon monoxide is formed and the ingot solidifies with a large central shrinkage cavity which cannot be completely eliminated by rolling. The ingot to slab yields are especially low in' the event that the cavity in some way becomes exposed to the atmosphere. This causes its various surfaces to be oxidized and thus removes any possibility that they may, at least in part, be welded together during subsequent rolling.

In practice, there will always be a certain number of steels which are made either underdeoxidized or overdeoxidized. Fortunately, the condition of the steel with respect to its state of oxidation, is evident within a few minutes of the completion of the teeming of the first ingot and, if the necessity arises, it is possible to take remedial measures to improve the situation in the teeming of the succeeding ingots. For example, if the steel is underdeoxidized, then additional aluminum can be added to the molds. If, on the other hand, the steel is seen to be overdeoxidized, then the steps which can be taken will depend on whether a split practice involving mold additions is being used or whether aluminum additions have already been made to the ladle. In the event of the former practice, the additions can be reduced, although the amount added should never be less than 3 ozs./ ton. If the aluminum were added to the ladle, then there is little that can be done to correct the state of oxidation of the steel. However, the relatively low ingot to slab yields which arise as a result of the central shrinkage cavity being exposed to the atmosphere, can be avoided by water capping the ingot. Water capping ensures that the shrinkage cavity will be covered by a thick layer of sound metal. This is accomplished by flooding the top of the mold with water to a depth of from 4 to 6 inches immediately after the top surface of the ingot has frozen over. Since this practice is also beneficial in preventing bleeding of the ingot top, which is another detrimental manifestation of underdeoxidation, and otherwise has no harmful effects when the steel is satisfactorily deoxidized, all steels according to the invention are made water capped.

It is generally recognized that semi-killed steels are more troublesome to manufacture when deoxidized with aluminum than with silicon. Thus, if it were only a matter of economics, the steels of the invention would be made silicon semi-killed. Ho e er he use of alu i m derives from our findings which show that its presence in the steel is necessary to achieve the optimum properties. In the experimental phase of our development, circumstances were such that it was more convenient to conduct experiments with aluminum killed steels than with any other kind. Thus, We accumulated a great deal of experience with these steels. Eventually, however, it was necessary to develop the steel in the semi-killed condition. As may be anticipated from the foregoing remark concerning the difiiculty of using aluminum as a deoxidizer in a semi-killed practice, the initial trials in this connection were with a silicon deoxidized steel. The results of subsequent forming tests with this steel, however, were not entirely satisfactory. Although the steel performed quite well in comparison with higher carbon, silicon semi-killed, niobium steels, its performance was nonetheless inferior to that which had been observed in earlier trials with aluminum killed steels. In concurrent research on a higher carbon, silicon killed, non-niobium steel, it was determined that chronic poor formability was due to the presence of elongated silicate inclusions which were situated just below the steels surface. In addition, it was also determined that these inclusions could be virtually eliminated from the steel, even though it contained in excess of 0.25% of silicon, simply by the addition of as little as 0.02% of aluminum. Upon examination of the microstructure of the above silicon semi-killed steel, similar subsurface silicates were observed. Thus, it was decided to apply the knowledge which was gained with the higher carbon silicon killed steel and the foregoing aluminum deoxidation practices were developed. As will be evident from examples yet to be cited, the presence of aluminum in the steels according to the invention very substantially improves their forming characteristics.

The processing following teeming and water capping of the ingots, is completely normal in all respects. Accordingly, the steel is permitted to solidify and when solidification is complete, and generally, but not necessarily, shortly thereafter, the ingot mold is stripped and the ingot is reheated to a temperature in the range of 2200 to 2450 F. When the ingot has attained the desired temperature throughout, it is reduced to the slab or billet form or it may be carried through to the final product form by rolling. If it is reduced to the slab or billet form, then after rolling and some additional procesing in which the unusable portion is removed and the remainder is cut hot to the desired sizes, the product of the ingot is allowed to cool to ambient temperature. The several pieces are then examined and further conditioned to remove surface defects and so forth. Subsequently, the pieces are heated for final rolling to a temperature in the range of 2200 to 2350 F. In rolling the steel to either plate or strip, the rolling schedule should be so arranged that the final rolling pass is taken at a temperature below about 1800" F. This is the only requirement which has to be met to achieve the mechanical properties according to the invention. Otherwise, any drafting practice is applicable and there is no need for heavy final reductions and so forth. For example, final reductions of the order of to are satisfactory. Subsequent to rolling, the steels may be permitted to cool to ambient temperatures naturally or may be subjected to enforced cooling using water spray as desired. As with the rolling practice, the use of water sprays subsequent to rolling is not mandatory to produce the steel of the invention.

In consideration of the foregoing, it will be evident to those skilled in the art, that an advantage of the present steel is its ease of manufacture in that it can be produced using identically the same equipment and the very nearly the same practices as have been used to produce niobium steels since their inception. This characteristic of the steel did not arise by chance. On the contrary, in developing the steel we steadfastly strived for economy in its manufacture. In particular, one criterion which we employed in this regard was that the steel must be producible without the need for special processing practices or special equipment to effect the same. Included in the category of special practices and equipment, were holding tables needed to effect low finishing temperatures, uncommonly powerful mills needed to effect heavy final reductions, and special water cooling facilities such as the previously mentioned laminar flow system to effect ultrarapid cooling subsequent to rolling. Of course, it is not meant to imply that these practices would not be beneficial to the steel of the invention. Rather, it is to emphasize the fact that while steels exist which are comparable to the present steel in that they claim similar mechanical properties or impose similar compositional limits on certain alloying ingredients, these other steels are, without exception, dependent for their properties upon the use of some special practice and are thus inherently less economical than the present steel. As concrete examples, contrast the processing needed to manufacture the steel of the invention with the heavy final reductions of the order of 30% which Duckworth et al. suggest to be used with the steels initially developed by Phillips or the need to quench and temper the steels described by the Yawata patent.

Thus, the steel of the invention was developed so that it could be produced using the same equipment as the steels which it was designed to replace. Yet, within the limits of existing processing capabilities, We have unexpectedly discovered that the steel is advantageously and, in regard to certain properties, uniquely responsive to the use of water sprays subsequent to rolling. In particular, an advantage of the steel of the invention is that by merely varying the spray practice after rolling, it is possible to alter the yield strength of the hot rolled product by as much as 10,000 p.s.i. In addition, we have also discovered that the observations of Tisdale, that water cooling from rolling will improve toughness, are not applicable to the steel of the invention. Instead, we have found that low temperature impact values are either not affected or are affected in a slightly adverse manner by the use of water sprays from finish rolling. As will become evident from subsequent examples, the indications of our findings are that these two effects derive from the relatively low carbon and manganese contents of the steel.

The steel according to the invention is quite ductile as is evidenced by the magnitudes of the various parameters which are normally measured in the more common laboratory tests, such as the tensile test or the impact test. The ductility of the steel is also reflected in its improved formability and vice versa. Thus, prior to citing specific examples to support the assertions of improved formability, it will be of value to indicate the relationship of these parameters to formability in a quantitative manner. Since the significance of a particular parameter will vary according to the forming process, it has been decided to develop this background again using the bending process as a specific example. To do this properly, it will be necesary to discuss certain of the theoretical aspects of bending.

In bending, there are three distinctly different modes of deformation depending upon the geometry of the piece being formed. In cylindrical coordinates, the geometrical parameter which is important in this regard is the ratio of the length of the piece in the axial direction to that in the radial direction.

When the geometry of the piece is such that this ratio is small, the upper limit not being known exactly, the steel undergoes a mode of deformation analogous to that which occurs in the bending of a prismatic beam. R. Hill at page 81 in The Mathematical Theory of Plasticity, Oxford University Press, London, 1950, has shown that although fibers on the tension side which are located at successively greater distances from the axis of bending necessarily undergo greater amounts of deformation, the mode of deformation in any particular fiber is identical to that which occurs in a simple tensile test. Thus, the ductility parameters which can be measured in a tensile test are indicative of the formability to be expected of the steel when subjected to bending of this type. Of special importance in this regard is the so-called uniform elongation which corresponds to the onset of local deformation in the tensile test and thus represents a natural measure of the forming limit of the steel in prismatic bending.

The second mode of deformation in bending occurs when the ratio of the length of the part in the axial direction to that in the radial direction is equal to or greater than about 8. In this situation, plastic deformation only occurs in the circumferential and radial directions and the steel is said to be deformed under conditions of plane strain. This type of bending differs from the foregoing in that local deformation, as characterized by necking in tension, is not known to occur under conditions of plane strain. Thus, the major mode of failure in this case is cracking which is typified by the occurrence of axially oriented, normally straight rather than ragged, cracks in the tension surface of the bent part. Since plane strain deformation is not the same as that which occurs in sim ple tension, the ductility parameters which can be measured in a tensile test are at best qualitative indicators of formability in this type bending. However, of these parameters, the reduction of area, rather than either the uniform elongation or the total elongation appears to be the better indicator of formability. In addition, since the major mode of failure in this type of bending is ductile fracture, tests which measure the ease or difficulty of the same also serve as good qualitative indicators of formability. Thus, the Charpy shelf energy or the Charpy impact resistance at the temperature at which the bending is likely to be carried out, is often cited.

The third type of bending arises when the ratio of the axial to radial lengths of the part is less than 8 but greater than the upper limit commensurate with bending as a prismatic beam. The mode of deformation in this case is such that the bent part becomes saddle shaped during forming. Thus, bending in this mode has little commercial importance.

Of the three, bending under conditions of plane strain is by far the most important. In fact, with the possible exception of drawing, it is probably the most widely used method of cold forming currently in existence.

The foregoing assertions of improved mechanical properties in the steel of the invention were based on the production and testing of steels with a variety of compositions, and in various forms, including plate and strip rolled from 30 pound induction melted ingots, commercial size open hearth melted ingots, and eventually 140 ton basic oxygen furnace heats. The following examples cited to support these assertions, however, are based primarily upon experiments which had the objective to consolidate the indications of previous findings with regard to the effects on mechanical properties and microstructure of various compositional constitutents including carbon, manganese, silicon and aluminum and of certain processing variables including the state of deoxidation and the use of water cooling from finish rolling. The experiments consisted of producing and processing six ingots weighing approximately four tons each which were modified by mold additions to a silicon semi-killed basic oxygen heat of the following composition by weight:

The modified ingots were designated to indicate their compositional differences in relation to the base heat. These designations and the compositions corresponding to them ar listed below n Ta e 1.

TABLE 1.-COMPOSITIONS OF THE EXPERIMENTAL STEELS Percent Designation 0 Mn P 8 Si Nb Al High managnese silicon killed steel 0.04 1. 24 0.003 0.019 0.271 0.033 0.003 High manganese silicon semi-killed steel 0.05 1.21 0.006 0.020 0.040 0.038 0.001 Low manganese silicon killed steel 0.05 0.67 0.005 0.020 0.378 0.036 0.003 Low manganese silicon semi-killed steel 0.05 0.72 0. 008 0.020 0.033 0.039 0.001 Aluminum treated silicon semi-killed steel 0.05 0.67 0.007 0.020 0.031 0.038 0.010 High carbon silicon semi-killed steel 0.14 0.68 0.008 0.024 0.031 0.038 0.001

The killed steels were hot topped and the semi-killed steels were water capped. Processing of the ingots subsequent to teeming was in general as described elsewhere n the specification. Specific details with regard to proc- :ssing will be described where relevant. In addition, it is valuable to note that the balance of the heat was aluminum treated in the molds and subsequently processed for high formability applications.

EXAMPLE 1 The ingots, initially measuring 33" x 26" x 64", were rolled to strip mill size slabs, and cut into four pieces each measuring 3%" x 14" x 120". After rolling, a sample from a position corresponding to the approximate center of the ingot' was taken from each of the the six steels. Subsequently, a smaller longitudinally oriented sample was prepared for micro-structural examination and quantitatively rated for the presence of elongated silicate inclusions. The method of rating the samples involved the determination of three factors as follows:

(1) The determination of the total number of silicate inclusions with lengths in excess of 0.0025 inches in a randomly selected circular area having a diameter of l centimeter;

(2) The determination of the average length of the inclusions in units of 0.005 inch; and,

(4) A qualitative estimate at a magnification of of the background cleanliness involving four possible cleanliness levels denoted by the letters A, B, C and D corresponding to very good, good, fair and poor.

The inclusion rating is thus stated in terms of the results of these three determinations. A typical rating, for example, may be 50-2-B: the first number being the number of inclusions; the second, their average length in multiples of 0.005 inch; and the third, the estimate of the background cleanliness.

It is valuable to note that this method is similar to that described in American Society for Testing and Materials Standard Designation: E45-63. In fact, with regard to metallographic procedure, the two methods are identical. The chief differences are as follows:

(1) In E45, only those inclusions which have lengths in excess of 0.005 inch are counted; and

(2) In E45, two inclusions which lie in the same plane are counted as a single inclusion if the distance separating them is less than 0.005 inch. In the present work, this critical distance is reduced to 0.001 inch.

A reason for these differences is that B45 was designed for use with finish rolled steels which generally exhibit much longer inclusions than steels in the semi-finished or slab form.

The silicate inclusion ratings of the six steels which are of interest are listed in Table 2 below. It will be immediately evident from these results that the addition of a small amount of aluminum to the steel had a very pronounced effect in eliminating the silicate type inclu- SIOIIS.

TABLE 2.SILICATE INCLUSION RATINGS OF THE EXPERIMENTAL STEELS Steel: Rating High manganese silicon killed steel 90-1-C High manganese silicon semi-killed steel 951C Low manganese silicon killed steel 42-l-C Low manganese silicon semi-killed steel 304-1-D Aluminum treated silicon semi-killed steel 0B High carbon silicon semi-killed steel 87-l-C In reviewing the above results, it will be appreciated that they only reflect the situation which exists in a very small portion of the steel, and thus are not to be considered as more qualitative indicators of the cleanliness of the steel as a whole. Considerable experience with the present rating method has shown that if the number of inclusions is less than 10, then it is a fairly safe assumption that the steel is clean, whereas if the number is greater than 50, that it is dirty.

EXAMPLE 2 Two slabs from each ingot were subsequently hot rolled on the strip mill to plate having a thickness of 0.375 inch. The slabs were divided into two sets. One set was subjected to water sprays after the final rolling pass, whereas the other set was allowed to cool naturally after rolling. In other respects, all of the slabs were processed similarly. They were all finished between 1700 and 1750 F., with the reduction in the final rolling pass being of the order of The water sprays which were employed were fairly powerful, being capable of cooling rates in the neighborhood of 30 F./sec. However, the spray banks were only about 110 feet long, so that at the speeds at which the plates left the mill, approximately 700 ft./min., application of the sprays to any particular section of the plate did not exceed about 10 seconds.

The longitudinal tensile properties of the six steels in the no sprayed and full sprayed conditions are shown below in Table 3 and Table 4 respectively.

TABLE 3.-LONGITUDINAL TENSILE PROPERTIES IN THE NO SPRAYED CONDITION Percent Yield, Tensile Percent reducstrength, strength, Elong. tion of Steel p.s.i. p.s.i. in 2" area High manganese silicon killed ste 62, 500 78, 600 23. 0 73. 3 High manganese s con semi-killed steel 57, 300 70, 140 24. 5 68. 4 Low manganese silicon killed steel 60, 100 71, 300 23. 5 77. 4 Low man anese silicon semi-kit ed steel 50, 900 63, 700 25. 5 75. 9 Aluminum treated silicon semi-killed steel 49, 500 62, 700 24. 5 76. 6 High carbon silicon semikllled steel 66, 500 71, 900 25. 5 63. 1

A careful examination of these data will show that the tensile and yield strengths of all the steels were favorably affected by the use of the water sprays but that the most significant improvements, in the neighborhood of 5,000 to 10,000 p.s.i., were effected in the case of the low manganese silicon semi-killed and the aluminum treated silicon semi-killed steels. Thus, apparently owing to their relatively low carbon, manganese, and silicon contents, these two steels were substantially more sensitive to processing than any of the other steels listed.

TABLE 4.LONGITUDINAL TENSILE PROPERTIES IN THE FULL SPRAYED CONDITION Percent Yield Tensile Percent reducstrength, strength, Elong. tion of Steel p.s.i. p.s.i. in 2" area High manganese silicon killed steel 66, 500 83, 000 21. 0 73. 3 High manganese silicon semi-killed steel 59, 700 73, 800 21. 5 75. 4 Low manganese silicon killed steel 63, 900 71, 700 22. 5 71. 3 Low manganese sili n semi-killed steel 55, 700 65, 900 23. 5 73. 3 Aluminum treated silicon semi-killed steel 57, 68, 700 22. 5 74. 9 High carbon silicon semi-killed steel 57, 500 72, 900 24. 5 66.5

Further analysis of the data will show that the low carbon, low manganese semi-killed steels were also ad vantageously affected in comparison with the other semikilled steels listed. For example, both the high carbon and the high manganese semi-killed steels, when compared to the low manganese semi-killed and aluminum treated steels, exhibited higher tensile strengths in both the no sprayed and full sprayed conditions but appeared to offer no advantage in regard to yield strength insofar as the same yield strength could be achieved with either of the latter two steels simply by the use of water sprays subsequent to rolling. It is also to be noted that the longitudinal reduction of area value of the high carbon steel was lower than that of any other steel listed. Thus, in any case, it will be appreciated that the low carbon, low manganese steels were not only more sensitive to the use of water cooling after rolling, but they were also advantageously affected thereby, in the sense that it was possible to achieve a range of yield strengths with the highest strengths being equal to those obtainable in steels having either higher carbon or higher manganese contents.

The tensile and yield strengths of the low carbon, low manganese steels were not superior or even equivalent to those of the silicon killed steels which, probably owing to their relatively high silicon contents, exhibited the highest strength values overall. However, as will be apparent from data to be presented in subsequent examples, the killed steels also had relatively poor impact properties and poor formability.

Finally, the fact that the behavior of the low manganese silicon semi-killed steel was almost identical to that of the aluminum treated steel, is only as would be expected in view of the similarities in their respective compositions. Subsequent examples, however, will show that of the two, the aluminum treated steel of the invention is by far the superior one.

EXAMPLE 3 TABLE 5.-LONGITUDINAL IMPACT PROPERTIES IN THE NO SPRAYED CONDITION Energy in Energy in Energy in ft.-lbs. at ft.-lbs. at it.-lbs. at

Steel +75 F. +32 F. 0 F.

High manganese silicon killed steel. 79. 5 28. 5 12. 0 High manganese silicon semi-killed steel DNB DNB 7. 5

Low manganese silicon killed steeL. DNB DNB Low manganese silicon semi-killed steel DNB DNB 10. 5 Aluminum treated silicon semikilled steel DNB DNB 8. 0 High carbon silicon semi-killed steel 39. 5 12. 8

TABLE 6.LONGITUDINAL IMPACT PROPERTIES IN THE FULL SPRAYED CONDITION TABLE 8.TRANSVERSE TENSILE PROPERTIES IN THE FULL SPRAYED CONDITION Energy in Energy in Energy in ft.-lbs. at it.-lbs. at ft. lbs at Steel +75 F. +32 F. F

High manganese silicon killed steel. 47. 0 15. 8 8. 3 High manganese silicon semi-killed steel DNB DNB 6. 8 Low manganese silicon killed steel. 68. 61. 2 Low manganese silicon semi-killed steel DNB D NB 8. 3 Aluminum treated silicon semikilled steel DNB DNB 8. 5 High carbon silicon semi-killed steel 70. O 5. 3

From an examination of the data in these two tables, it will be evident that with the exception of the high carbon silicon semi-killed steel, the results show a trend which is contrary to the observations of Tisdale that water cooling from finish rolling is beneficial to impact resistance. Instead, the results show that the use of the water sprays either had no effect or had a slightly detrimental effect. The opposite trend to this in the case of the high carbon steel is explained, of course, by the fact that this is the type of steel with which Tisdale worked.

In addition, it is worth noting that, almost without exception, the low carbon steels in both the no sprayed and full sprayed conditions, exhibited superior impact values at every temperature to the high carbon steel. It is also to be noted that the values of the aluminum treated steel were among the highest and that, outside of the high carbon steel, the values of the silicon killed steels were among the lowest.

EXAMPLE 4 Exclusive of the high carbon silicon semi-killed steel, the longitudinal tensile properties listed in Table 3 and Table 4 give little indication that there is any difference in ductility between the various steels. Although there is a slight trend towards decreased ductility with the use of water cooling from finish rolling, both the elongation and reduction of area values of the various low carbon steels are all about the same. Even the elongation values of the high carbon steel fail to indicate a difierence in ductility.

Significant differences in ductility, however, do exist. The existence of these differences will be immediately evident from an examination of the transverse tensile properties of the various steels in both the no sprayed and full sprayed conditions as shown in Tables 7 and 8 respectively. The ductility differences are best indicated by the reduction of area values. Apparently, the total elongation is relatively insensitive to the factors which are causing the differences.

TABLE 7.-TRANSVERSE TENSILE PROPERTIES IN THE NO SPRAYED CONDITION Percent Yield Tensile Percent reducstrength, strength, elong. tron of Steel p.s.i. p.s.i. in 2" area High manganese silicon killed steel 63, 300 78, 300 20. 5 61. 2 High manganese silicon semi-killed steel 59, 100 71, 300 18. 5 62. 7 Low manganese silicon L killed steeli 60, 200 70, 800 22. 0 65. 0

ow manganese s1 con semi-killed steeli .Siii 53, 300 64, 900 24. 0 65. 7 Aluminum treate con semi-killled stleel i. 49, 600 63, 100 23. 5 73. 3 Hi h car on si con sem k illed steel- 55, 300 71, 300 21. 5 47. 4

Percent Yield Tensile Percent red'ucstrength, strength, Elong. tion of Steel p.s.i. p.s.i. in 2 area High manganese silicon lulled steel 66, 84, 600 19. 5 57. 7 High manganese silicon semi-killed steel 63, 900 79, 600 21. 5 57. 7 Low manganese silicon lled ste 62, 100 72, 300 21. 0 55. 6 Low manganese silicon semi-killed steel 55, 500 63, 700 22. 5 60. 7 Alugfiifilfil lgreatceld silicon se e s e 58, 700 68, 900 23. High carbon silicon semi 0 70 5 kille eel 58, 500 73,300 21. 5 52. 4

It will be apparent from these data that the aluminum treated steels exhibited substantially higher reduction of area values and even slightly higher elongation values than any of the other steels listed. This is considered to be of the utmost significance in that it indicates the superior formability of the aluminum treated steels. It is thought that the results are attributable to the effect which the aluminum has in reducing the elongated silicate inclusion content. The reasoning underlying this is as follows. A comparison of the longitudinal and transverse reduction of area values of the various steels will show that the longitudinal values were without an exception, higher than the corresponding transverse values. This type of behavior is generally attributed to the presence of elongated second phases in the steel. These phases interrupt the continuity of the steel matrix and because they present a more pernicious type of discontinuity when the specimen is oriented in the transverse direction than in the longitudinal direction, they \effect decreased transverse ductility values. Any second phase which is not isotropically disposed within the matrix can have this effect. Common examples of anistropic second phases are pearlite and silicate and sulfide inclusions. Reducing the volume of such phases reduces the defect area associated with them and thus effects improvements in transverse ductility properties. An excellent example of this in the present work is the ductility properties of the high carbon steel in comparison with those of the various low carbon steels. In this case, reducing the carbon content reduced the pearlite content and substantially improved the ductility in the longitudinal as well as the transverse directions. Thus, in view of the indications of Example 1 that the aluminum treated steels had significantly reduced elongated silicate contents, it will be appreciated that a similar explanation is applicable to the improved ductility values which the above data show for these steels.

In reviewing the tensile properties in Tables 7 and 8, it is worthwhile to note that they show analogous trends to the properties listed in Tables 3 and 4 and thus support the assertions of Example 2.

EXAMPLE 5 The transverse impact properties of the experimental steels in the no sprayed and full sprayed conditions are shown in Tables 9 and 10 respectively. As in the case of the longitudinal properties, the tests were conducted using three-quarter size Charpy V-notch specimens and the various values corresponding to trials at plus 75 F. and plus 32 F. are averages based on at least two and, in most cases, three determinat ons.

TABLE 9.TRANSVERSE IMPACT PROPERTIES IN THE NO SPRAYED CONDITION Energy in Energy in Energy in TABLE 10.-TRANSVERSE IMPACT PROPERTIES IN THE FULL SPRAYED CONDITION Energy in Energy in Energy in it.-lhs. at it.-los. at ft.-lbs. at Steel +75 F. +32 F. F.

High manganese silicon killed steel- 16. 9. 0 6. 3 High manganese silicon semi-killed steel 18. 0 ll. 5 7. 5 Low manganese silicon killed st l 20. 0 20. 0 10. 5 Low manganese silicon semi-killed ste 21.0 10. 8 5. 5 Aluminum treated silicon semikilled steel 47. 0 19. 8 5. 5 High carbon silicon semi-killed steel 10.0 4. 8 2. 0

A review of these data will show that the aluminum treated steels exhibited substantially higher values at plus 75 F. in both the no sprayed and full sprayed conditions than any of the other steels listed. As in Example 5, this is thought to be significant since it shows improved resistance to ductile fracture and thus indicates improved formability in the aluminum treated steel. The explanation of the improvements is also similar to that given in Example 5. In other words, it is thought they reflect the improved cleanliness of the steel with respect to the presence of elongated silicate inclusions.

It is important to note that formability is usually correlated with the Charpy V-notch shelf (or maximum) energy which represents the maximum resistance to ductile fracture, rather than with the impact resistance at a particular temperature. Thus, the use of the impact values at plus 75 F. to indicate formability represents something of a departure from existing theory. However, it is thought to be a reasonable departure since the values at plus 75 F. represent the ductile resistance of the steel at a temperature which is near to the ambient temperatures in most forming processes whereas this may not be even remotely the case with the Charpy shelf energy.

The Charpy V-notch values at plus 32 F. are also im portant to indicate the steels resistance to brittle fracture. In this instance, the aluminum treated steel was not the best but was among the best. The values corresponding to both the no sprayed and full sprayed conditions were each in excess of 15 foot-pounds.

Finally, it should be noted that the transverse impact properties show essentially the same trends with regard to processing and composition as the longitudinal poperties which were discussed in Example 3 and thus support the assertions of Example 3.

EXAMPLE 6 The experimental steels were laboratory tested to indicate formability in bending. To do this, a special bend test was employed. An unusual feature of this test is that the specimen is prepared with a square notch which is oriented in the direction parallel to the axis of bending. The advantage of the test, which derives from the presence of the notch, is that it is possible to uniformly deform a fairly large section of the specimen under conditions of plane strain and to carry out the deformation, both continuously and in a completely controlled manner, to relatively high strain values or correspondingly, to very small inside radius values. For example, strains slightly in excess of 60 percent, which correspond to inside radius values of considerably less than /2T, are possible (T being the thickness).

Specific details with regard to specimen preparation and test procedure are briefly as follows. A carefully squared rectangular section of the steel is cut. If the steel is to be tested in transverse bending, for example, the long dimension of the rectangular section is oriented transverse to the rolling direction. The long dimension of the specimen may be made to any convenient length but the short dimension should be at least ten times the thickness of the notched cross section to ensure the existence of plane strain conditions during bending. The specimen is notched at mid-length and the axis of the notch is made parallel to the rolling direction. The depth of the notch should be approximately one third the thickness of the plate and its width should be approximately equal to the thickness of the plate. Scale adhering to the section of the specimen which will undergo deformation may be removed in any convenient manner provided that the method produces a reasonably smooth surface which is free from scratches in the direction parallel to the axis of the notch. To enable the results of the test to be stated in terms of strain values, the surface of the specimen should be prepared with gage marks. For example, in the present work, a grid with unit squares measuring A of an inch on a side was electrolytically etched into the surface of the specimens. Prior to testing, the specimen is ordinarily prebent. This consists of deforming the specimen by loading it as a simple beam with the machined surface of the notch in compression. After prebending, the specimen is bent by loading it as a column with the axis of the notch at right angles to the direction of loading. The test is terminated either when a crack appears having a length in excess of one-quarter inch, or upon the appearance of a multiplicity of many fine, relatively short cracks, or in the event that failure does not occur when the opposite edges of the notched section are forced together. The results of the test are normally stated in terms of the type of failure and the percent elongation in the tension surface as determined at a position corresponding to the approximate center of the deformed section.

The results of notched transverse bend tests of the six experimental steels in the no sprayed and full sprayed conditions are shown in Tables 11 and 12 respectively. As a matter of interest, the bend angles are also listed to show that it is possible to state the results of the test in terms other than elongation.

TABLE 11.TRANSVERSE BENDING PROPERTIES IN THE NO SPRAYED CONDITION TABLE I2.TRANSVERSE BENDING PROPERTIES IN THE FULL SPRAYED CONDITION Percent Angle of elongation bending, in A a Steel deg. inch Type of failure High manganese silicon 39 22.6 Single long crack.

killed steel. High manganese silicon 11 6. 1 Do.

semi-killed steel. Low manganese silicon 20 13. 1 D0.

killed steel. Low manganese silicon 69 37. 9 Many light semi-killed steel. cracks. Aluminum treated silicon 52. 5 Do.

semi-killed steel. High carbon silicon 45 25. 5 Single long crack.

semi-killed steel.

It will be immediately apparent from an examination of these data that the transverse bending properties of the aluminum treated steel were far superior to those of any of the other steels listed. In the no sprayed condition, the specimen of the aluminum treated steel did not fail at all and in the full sprayed condition, it failed at a very high strain value of 52.5% which corresponds to an exceptionally small inside radius value of slightly less than /2T.

In reviewing the results, it is important to be aware of the fact that it has usually been possible to associate failures due to single long cracks to the presence of equally long subsurface inclusions, Whereas failures due to many light cracks have not been correlated with any particular defect and hence are probably indicative of a general breakdown of the metal in the surface of the specimen. Thus, in any case, it will be appreciated that the indications of the data are that most of the steels failed because of the presence of subsurface inclusions which, in consideration of previous examples, were in all probability elongated silicates. In fact, in the case of the full sprayed high manganese silicon semi-killed steel, there was no doubt that the failure was attributable to a silicate since the inclusion was exposed to view in the preparation of the specimen.

EXAMPLE 7 It was mentioned earlier that the uniform elongation as determined in a tensile test serves as a natural measure of the forming limit in processes such as bending as a prismatic beam where the major cause of failure is the occurrence of localized or non-uniform deformation. Thus, the uniform elongation values of the experimental steels in both the no sprayed and full sprayed conditions and in both the longitudinal and transverse directions were determined along with the other tensile properties as listed in the previous examples. An analysis of these findings, however, did not reveal any definite differences. There was a trend in the data suggesting a decrease in uniform elongation with increasing alloy content but the magnitudes of the various differences were not felt to be large enough to be considered significant. Subsequently, however, additional evidence to verify the existence of this trend was inadvertently obtained. In concurrent studies of steels as described in a copending application, Ser. No. 758,862, filed Dec. 18, 1968, a heat was made in which one of the essential alloying ingredients was mistakenly left out. The ladle analysis of the heat, in percent by weight, was as follows:

Percent C 0.08 Mn 1.20 P 0.008 S 0.021 Si 0.07 Nb 0.053 A1 0.100

Of course, the fact that an essential ingredient had been left out was not discovered until long after the ingots has been cast. Consequently, according to plan, the steel was further modified in the molds using additions of manganese, silicon, and carbon with the'result that a series of steels became available which were, in essence, natural extensions of the steels of the preceeding examples to higher alloy contents. Thus, in an effort to salvage something of value from the experiment, it was decided to roll the steels and compare their properties to those of the steels of the present work.

Three different steels, having compositions as shown in Table 13 below, were rolled to plate having thicknesses of 0.375 inch each. The plates were finished with water sprays and were otherwise processed using practices as indicated elsewhere in the specification.

The uniform elongation values of these steels and of those of the comparably processed low carbon steels of the preceeding examples, are shown in Table 14. For convenience, the steels are listed in order of increasing alloy content. From an examination of these data, it will be immediately evident that there is a very definite trend towards decreased uniform elongation values with increasing alloy contents. Hence, it will be appreciated that the steel of the invention with its relatively low alloy content exhibits superior performance as compared to the other steels listed, when subjected to forming processes in which non-uniform or local deformation is to be avoided. It is worthwhile to note in this connection that, in addition to bending as a prismatic beam, drawing is also such a process.

TABLE 14.EFFECT OF ALLOY CONTENT ON UNIFORM ELONGATION Steel Percent uniform elongation Aluminum treated silicon semi-killed steel 14.50 Low manganese silicon semi-killed steel 15.50 Low manganese silicon killed steel 14.75 High manganese silicon semi-killed steel 13.75 High manganese silicon killed steel 13.25 High Alloy l 11.50 High Alloy 2 11.00 High Alloy 3 10.50

EXAMPLE 8 A steel according to the invention was made aluminum semi-killed with the following composition by weight:

Percent C 0.06 Mn 0.66 P 0.006 S 0.025 Si 0.01 A1 0.018 Nb 0.043

The steel was straight rolled to plate of a thickness of 0.250 inch following the practices previously outlined in the specification. The yield strength of the as-rolled plate was determined to be in excess of 55,000 p.s.i. Subsequently, the steel was subjected to transverse bending trials in a brake press under normal production conditions. A number of pieces of the steel were successfully formed to successively smaller inside radii, starting at a radius of 2T and finishing at 1T. At this point, it was decided to determine if the steel would fail when subjected to the most severe conditions possible with the tools which were available. Thus, a piece was bent using a so-called knife edge to form the inside radius. The radius of this tool was no greater than about of an inch. At first the steel was bent in small increments because it was felt that the tool was so sharp that it might cut the steel, rather than bend it, if the formation was attempted in a single operation. However, the steel formed successfully and eventually, it was decided to attempt to make the bend in a single operation. Again the steel formed successfully. However, as opposed to shaping the steel, the tool actually brinelled the inside surface of the bend to a depth of about 4, of an inch with the result that the average inside radius was no less than about /2T. Nonetheless, it will be appreciated that this was still an extremely severe formation.

Of course, a steel would never be subjected to forming under such adverse conditions in a practical situation nor is the foregoing example meant to imply that the steel of the invention should be expected to perform under such conditions. However, as a result of the production and testing of a number of experimental heats within the scope of the invention, it has been established that improved forming limits in bending are to be expected under practical conditions. A listing of these limits in terms of the minimum inside radii applicable to transverse bending is shown in Table 15. The various values in this table should be contrasted with those presented earlier at page 13 of the specification which indicate the recommendations of the industry in regard to steels of essentially the same thickness and yield strengths.

TABLE 15.MINIMUM INSIDE BENDING RADII APPLICA- BLE WITH THE STEELS OF THE INVENTION Material thickness, inches To 0.180 incl Over 0.180 to 0.250.- Over 0.250 to 0.500-. Over 0.500 2 All percentages given herein in respect to composition are percentages by weight.

In view of our invention and disclosure, variations and modifications to meet individual whim or particular need will doubtless become evident to others skilled in the art to obtain all or part of the benefits of our invention without copying the process and composition shown, and we, therefore, claim all such insofar as they fall within the reasonable spirit and scope of our claims.

Having thus described our invention what we claim as new and desire to secure by Letters Patent is:

1. The method of producing a steel sheet or plate of high yield strength, good toughness, good weldability andsuperior formability in the hot rolled condition, which comprises casting a steel ingot having the following composition by weight:

Carbon: O-O20.08%

Manganese: 0.250.80%

Silicon: 0.05% maximum Niobium: 0.025-0.10%

Aluminum: 0.005-0.025%

Iron and impurities: balance. and hot rolling the steel so cast.

2. The method of claim 1, in which the niobium content is between 0.025 and 0.045%

3. The method of claim 2, in which the silicon content is 0.05% maximum.

4. The method of claim 3, in which the aluminum content is between 0.008 to 0.018%

5. The method of claim 1, in which the aluminum content is between 0.008 to 0.018%

6. The method of claim 1, in which the aluminum content is 0.008 to 0.018%.

7. The method of claim 1, in which the silicon content is 0.03 to 0.05% and the aluminum content is 0.005 to 0.015%.

8. A hot rolled steel sheet or plate having the following composition by weight:

Carbon: 0.02-0.08%

Manganese: 0.250.80%

Silicon: 0.05% maximum Niobium: 0.025-0.10%

Aluminum: 0.005-0.025%

Iron and impurities: balance. and having a yield strength of at least 45,000 p.s.i., transverse Charpy V-notich impact resistance of at least 40 foot-pounds at F. and of at least 15 foot-pounds at +32 F., having a carbon equivalent of at most 0.30 and having super cold formability as exemplified by minimum radii in the least favorable direction of bending as follows:

MINIMUM INSIDE BENDING RADII APPLICABLE WITH THE STEELS OF THE INVENTION Minimum yield strength in p.s.i.

9. A steel of claim 8, in which the niobium content is as follows:

Niobium: 0.025-0.045%.

10. A steel of claim 9, in which the silicon content is as follows:

Silicon: 0.5% maximum.

11. A steel of claim 10, in which the aluminum content is as follows:

Aluminum: 0.008-0.018%.

12. A steel of claim 8, in which the aluminum content is as follows:

Aluminum: 0.008-0.018%.

13. A steel ofclaim 8, in which the aluminum content is as follows:

Aluminum: 0.008-0.018

14. A steel of claim 8, in which the silicon content is 0.03 to 0.05% and the aluminum content is 0.005 to 0.015%.

References Cited UNITED STATES PATENTS 3,208,844 9/1965 Kato et a1. 75--49 3,303,060 2/1967 Shimizu et a1. 75l24 3,328,211 6/1967 Nakamura et a1. 75l24 X OTHER REFERENCES Transactions of ASM, vol. 61, 1968, Gray et al., pp. 255468.

Rudarsko-Metalurski Zbornik, Podgornik, 1967, pp. 143-152.

CHARLES N. LOVELL, Primary Examiner US. Cl. X.R. 29528; 75124 3 793 STATES PATENT \s '09 h -H a .wprwnl v C *FACATE Or COkmeiwN 3,721,587 W I Dated March 20, 1973 Patent No.

fl Alfred G. Allten and Frecerick. J. Semel It is certified that error appears in the above-identified patent and that said Letters Patent are hereby corrected as shown below:

(3011mm iine insert than after "more Column 22, line 32-, "0.5%" should read 0.05%

Signed and sealed this 20th day of November 1973.

(SEAL) Attest:

EDWARD M.FLETCHER,JR. RENE D. 'IEG'BMEYER Attesting Officer Acting Commissioner of Patents