US1794068A - Process of making open-hearth steel - Google Patents

Description

Patented Feb. 24, 1931 UNITED STATES PATENT ALEXANDER. I... FEILD, OF CANTON, OHIO, ASSIGNOR, :BY MESNE ASSIGTS, TO REPUBLIC STEEL CORPORATION, OF YOUNGSTOWN, OHIO, A CORPORATION F NEW JERSEY PROCESS OF `MAKING OPEN-HEARTH STEEL Application' led July 2,9,

The presentinvention comprises a procedure for the manufacture of steel by the open hearth process. In accordance with this invention, the removal of carbon during refining is caused to take place at that rate which is most advantageous from the stand- I points (l) of obtaining a product characterfcombination of advantages sought, for modir fying the slag so that it displays this advantageous composition and for maintaining the desired composition while the refining reactions are in progress.

In what follows, discussion is directed in particular to steel-making in the basic-lined open hearth furnace. However, the inventionis not considered to be restricted to the basic process;

It has hitherto been c`ommon practice in basic open hearth melting to proceed more or less in accordance with a certain standardized program, subject howeverito considerable fiexibility, even in the case of a fixed charge composition. This procedure in its detailed aspects is familiar to those skilled in the art and need not be described here. Attention is directed, however, to the fact that the process is complex and that it is characterized by wide variability even when conducted by melters of acknowledged skill and experience in the art, and in furnaces of modern design. For instance, in the case of thirty-eight 10G-ton heats of plain carbon steel designated for a product whose carbon content was fixed by specification between the limits of (LOG-0.12% and made b ordinary practice in a single furnace rom a charge mixture containing 45% pig iron and steel'scrap of reasonably uniform grade and analysis, the amount of ore used varied from 0- to 9000 lbs., the melting time from 7 hrs. 15 mins. to 12 hrs. 20 mins., and the iin- 1926. Serial No. 125,794." I ishing time from 35 mins. to t hrs. 40 mins. The carboncontent of the metal bath at melting varied from 1.40% to about 0.25%, and the carbon content of the finished steel from 0.06 t0 0.11%.

During the progress of a heat, the melter has for guidance in timing the iron oxide or ore additions and in deciding upon the weight of each ore addition, the following factors; the intangible factor termed personal experience, (2) the carbon content of the metal bath at arbitrarily determined intervals, (3) in visually estimated (or measured) temperature of the slag, (4) the temperature of the metal beneath the slag `as estimated by any one of a number of approximate methods, all of which involve a large `element of judgment, (5) the rate of carbon oxidation at any selected time, as estimated from the vigor and extent of evolution or bubbling of gas (CO and CO2) through the slag. )Vhen necessary, analysis `of the metal bath for manganese, phosphorus, sulphur, and other elements is made to determine whether or not the steel will meet specifications. The presence of an excess of iron oxide in the slag at the end of the refining stage is a fruitful source of the objectionable non-metallic inclusions, above referred to, and

this objection can be obviated by my method.

I have discovered that the open hearth process may be more carefully controlled and uniformity in product insured by a proper adjustment of the iron oxide content of the slag during the various stages of the finishing or refining period. .This adjustment cannot be madefuntil theinitial iron oxide content of the slag at melting is determined by chemical analysis or by other suitable met-bod. If this iron oxide content is lower than the value which has been determined to be advantageous by means of the novel method to be described below,.a sufiicient amount of iron ore or other oxid'eof iron is .added to bring the iron oxidecontent up to thel desired value. If desired, the ore may be added in successive small additions. ,If the initial iron oxide content is already too high, no addition of ore is made until it falls to the desired percentage or to some value slightly below this.

During the heat, analysis of the slag at frequent intervals for iron oxide is desirable and the smaller and more frequent the individual ore additions, the closer to the desired percentage the iron oxide in the slag can be held.

Continuous feeding of iron ore or of other oxidizing agent is within the scope of the invention as is the adjustment of the oxygen content, free or combined of the furnace gases, provided in both instances the purpose of such procedures is to maintain the iron oxide content of the slag at some predetermined value or values. i

The novel method by which the percentage of iron oxide appropriate in the slag for any given furnace and product constitutes a necessary and vital part of my new process for the manufacture of open hearth steel. This method involves the employment of a fundamental relationship which I have found to obtain between rate of carbon elimination, the F e0 content of the slag (total iron calculated to FeO) and the carbon content of the meta-l bath. If

Q dt

l represents the rate of carbon elimination (in per cent carbon per minute) fromA the metal bath at the time t, 01:.the coincident percentage of carbon in the metal bath, and fwf that of Fe() in the slag, then this fundamental relationship may be expressed in the form of the following equation:

ltakw (1.08 @0715) (I) where la is aconstant, a is a numerical factor derived from the equilibrium constant of the reaction FeO-tC-)CO-FF e, andmay for practical purposes be assumed to have a value of 0000757, but is known to vary slightly with the temperature of the metal bath. This equation, when integrated between the limits m2 and w1, representing initial and final carbon contents, yields the elapsed time equation:

26,080 1.08am,- 1 kw3 10g' (1.08wx1- 1 ing a furnaceconstant, i. e. it varies from furnace to furnace, even at times when the' furnaces in question are ostensibly of the same design and capacity. The value of k is undoubtedly controlled not only by hearth 7 area, depths of metal and slaglayers, but also by the rate atl which oxygen is transferred from the furnace gases to the slag layer, where the reaction 4Fe0+ O2=2Fe208 is constantly occurring. The oXygIeIn in question may be free oxygen, CO2, or 20 in the furnace gases. The FegOs thus formed acts like the FegO3 introduced by the iron ore additions in reducing the carbon content of a metal bath. This vrate of oxygen transfer is a characteristic of flame temperature and combustion conditions and is therefore included within the furnace constant 7c. A

Vcan be readily obtained graphically from the carbon-time curve for any selected heat. An alternative method for arriving at k is to employ Equation (II), substituting observed values for m1, m2, T, and w. In this case` however, the method is notexact unless fw remains substantially constant during the period of observation or unless the square root of the mean square of a number of values of w taken at substantially uniform small intervals is used.

In actual practice tables or curves may be used, derivedl from Equations (I) or (II) or both, thus avoiding all unnecessary calculations while the heat is in p-rogress.

From the' equations or from such above mentioned tables o curves, it is readily possible to choose for any particular carbon content desired at the finish of the reining process, such a value of FeO in thegslag as will da: (H) at said fimshing time 'offrom 0 to,say, 0.004% carbon per minute. The lower the rate the lower will be the iron oxide in both slag and metal and the smaller the amount of dissolved iron oxide and gases which have to be removed from the metal by de'oxidizers or by effervescence (in the case of open steels) for the production of a clean steel. The practical value of. ob-

yield a rate of elimination taining in the furnace a metal bath contain-1 ing iron oxide and oxygen-containing gases v in' the minimum practicable quantity before adding ferromanganese, ferrosilicon, etc., will be apparent to those skilled in the art.

' I have found that in the case of a number of 100-ton stationary basic' open-hearth fur- Time Time Peggy Pegg? required required X X when when Minutes Minutes 0.50 0.40 68 29 0.50 0.30 138 59 0.50 0.20 213 90 0.50 0.15 256 107 0.50 0.10 289 125 0.50 0.08 330 133 0.50 0.06 368 143 0.50 0.05 411 149 0.50 0.04 w 158 0.50 0.03

The times as calculated were found to agree closely with those actually occurring in practice.

This table may be written to indicate rate of carbon elimination, as follows Pcggt reo 20% Feo 30% 0. a0- .20 .0016 .0035 .20- .15 .0015 .0035 .15- .10 .0014 .0035 .10- .os .0014 .003s .0s- .00 .0012 .003s .00- .05 .0007 .0020 .05- .04 0 .0829

Suppose it were required to make steel containing 0.04 C. at the finish of the refining operation. Evidently from the above tables such a steel could not be made at all with a slag containing 20%.Fe0. The table shows that with a 30% FeO slag it would be feasibleY to produce a 0.04' C. steel, and that the rate of carbon elimination is relatively rapid at the time this carbon percentage is reached.

Let it be assumed that a rate of .0020 instead of .0029 is decided upon as the proper rate of carbon elimination for minimizing the non-metallic inclusions, then the lvalue of w for this particular condition can be determined from Equation (I).

ObviouslyI the requisite information as to various percentages of carbon content, rate of elimination and FeO percentage in slag can be conveniently tabulated or graphically shown in curves. y

Examples of such curves are shown in the l accompanying drawing and are self-explanatory.

The melter can Aby observation of the rate of carbon elimination (which is dependent y upon the iron oxide content of the slag) as oxide content -of the slag and act accordingly. Such a procedure is obviously within the scope of my invention.

In carrying out the process the melter will kiow in advance the desirable percentage of FeO-in the slag durin the refining stage, and will know that the st results canbe obtained by maintaining this .percentage until the desired carbon content is obtained.

Preferably the percentage of Fe() in the slag is determined at the close of the melting stage, brought to the desired percentage if necessary, and maintained thereafter, but obviously it can be obtained during the refining stage and modified as may be required.

It is a common mistake in practiceto add, too much iron oxide (overoreing) so that at the time of finishing the heattoo much iron oxide is present in the slag, and the melter has up till my invention had no means of knowing how to control the conditions. It will be understood that the heat is` finished when the desired carbon point is reached and that further elimination of carbon by iron oxide in the slag is then stopped or slowed up by the addition of deoxidizers such as ferromanganese or ferrosilicon.

If too little oxide of iron is present toward. y

the close of the reining stage, the elimination of carbon and the duration of the heat are unduly prolonged. When this condition is reached, it is necessary to add more iron oxide, which at the stage of the operation is undesirable in any event, and usuallytoo much is added with the detrimental results above referred to.

By carrying out the process according tov my invention, these difculties are obviated.

It has hitherto been common practice in basic open-hearth melting to proceed more or less in accordance with a certain standardized program, subject, however, to considerable flexibility even in the case of a fixed charge composition. Where the initial metal charge consists, as it usually does, of scrap the scrap first, lon top of the limestone which has been previously laid on the bottom of the hearth, and then to charge-the pi iron after the scrap has started to melt. Aigter the pig iron, either solid or molten, isv added the 11o steel and plg iron, 1t 1s customary to charge lime boil sets in vigorously'and contlnues as a rule for 2 to 4 hrs. This boil is dueto evolution of carbon dioxide from the lime`- stone (CaCO3) in the charge. 'a

Due to increasing hydrostatic pressure of the metal and thermal dissociation of CaCOa, lime is disengaged from the bottomand rises to the top of the slag where it continues to o into solution in the slag.- By the time t e' surface and floats thereon, may not be complete until the lapse of several hours following the melting of the metal charge. The slag is not entirely shaped up until the floating lime lumps have dissolved in the slag, although toward the end of lime solution the few scattered lumps which remain have only a very small effect on the shaping up of the slag.

When the metal charge is melted, a spoon or small hand ladle is introduced beneath the slag and a test sample of metal removed from the bath. This is poured into a small castiron mold and allowed to solidify. It is removed from the mold while at a red or yellow heat and cooled in water more or less slowly, but at a more or less fixed rate. The test piece is then broken .and the carbon content estimated from the appearance of the fracture. The test piece may, if desired, be sent to the chemical laboratory for a rapid carbon analysis for check purposes. The carbon content of the metal at melting down is iinportant, since it is used by the melter in decid- 1n how heavy an initial addition of iron oxide (usually as iron ore) is proper under the circumstances. In case the heat melts at a relatively'low carbon content, oreing may be dispensed with entirely and, in extreme cases, recourse may be had to a recarburizing addition of pig iron or spiegel. It is however, generally agreed that good practice requires a carbon content at melting sufficiently". high to require a certain amount of oreing,

It is at the conclusion of this melting stage that the so-called refining or finishing stage of the heat begins.

According to my invention, it is at this time that the iron oxide (FeO) content of the slag is determined by chemical analysis, and successive analyses are made from time to time, during the refining stage, and the iron oxide content of the slag accordingly so adjusted as to keep it substantially of a constant percentage, as decided upon and determined by the methods above described. It is desirable to make carbon determinations coincident with the ironl oxide analyses. While it -is desirable to correctlycadjust the tion is that the melter, instead of carryin out the refining by more or less haphazar methods based largely on his skill and exper'ience, is furnished with an additional guiding factor of control whereby better final results may be obtained.

I claim:

1. The process of making open-hearth steel which comprises melting the charge and thereafter refining the steel in the presence of a sla in which there is maintained a substantialIy constant percentage of iron oxide, the said constant ercentage of iron oxide being maintained by determining the iron oxide content ofthe slag from time to time in the course of the refining, and adjusting such content if necessary to bring it to the constant percentage.

2. The process of makin steel which comprises melting gli refining the steel after the melting in the presence of a slag having a substantially constant percentage of iron oxide determined according to the equation open-hearth 92: .0000757 kw (108i) in which zg d:

is the desired rate of carbon elimination, c'is a constant, w is the percentage of FeO in the slag and l.fr the coincident percentage of carbon in the bath at time t, the said constant percentage of iron oxide being maintained by determining the iron oxide content of the slag from time to time in the course of the refining, and adjusting such content, if necessary, to bring it to the said constant percentage.

In testimony whereof, I af'lix my signature.

ALEXANDER L. FEILD.

slag composition at the beginning of the refining stage, obviously this may effect.- ed later, as long as sufiicient time is permitted for the reactions, above described, between the metal `and the slag, to take place at the desired rate of carbon elimination.

It is to be understood that the refining of the steel is otherwise carried out in the usual way including the maintenance of the proper fluidity of the slag, and the addition, at the proper time, of deoxidizers such as ferromanganes'e or ferrosilicon, so that the metal bath may be cleansed of dissolved and suspended oxides before it is tapped from the furnace.

The result of the application of my invene charge, and

Images