Search Images Maps Play YouTube News Gmail Drive More »
Sign in
Screen reader users: click this link for accessible mode. Accessible mode has the same essential features but works better with your reader.

Patents

  1. Advanced Patent Search
Publication numberUS3816720 A
Publication typeGrant
Publication dateJun 11, 1974
Filing dateNov 1, 1971
Priority dateNov 1, 1971
Also published asCA975565A1, DE2253480A1, DE2253480B2, DE2253480C3
Publication numberUS 3816720 A, US 3816720A, US-A-3816720, US3816720 A, US3816720A
InventorsBauer E, Dokken R, Hard R, Malhotra U
Original AssigneeUnion Carbide Corp
Export CitationBiBTeX, EndNote, RefMan
External Links: USPTO, USPTO Assignment, Espacenet
Process for the decarburization of molten metal
US 3816720 A
Abstract
A process for making steel by refining a predetermined mass of molten metal having a composition containing at least carbon and iron comprising: injecting oxygen into such mass with a gaseous diluent at a first dilution ratio and calculating from the activity coefficient of each element in the mass a theoretical equilibrium partial pressure of carbon monoxide for the oxidation reaction of each element; calculating the absolute maximum partial pressure of carbon monoxide; selecting the lowest partial pressure and calculating a new metal analysis for the mass; changing the dilution ratio upon meeting a given criterion and repeating the process until the carbon level of the mass has dropped to a predetermined level.
Images(8)
Previous page
Next page
Description  (OCR text may contain errors)

United States Patent 1191 Bauer et al.

[ PROCESS FOR THE DECARBURIZATION June 11, 1974 3,561,743 2/l97l Schroeder et al 75/60 CF MOLTEN METAL 3,565,606 2/l97l Carlson et al 75/60 3,594,155 7/1971 Ramachandran 75/60 [75] Inventors: Edwin F an is Bau g a s: 3 ,6 2 2 1 7 11/1971 Carr et al 235/151.12 Roger Nels Dokken, Katonah; 3,614,682 10/1971 Smith 235/ 151.12 Robert Amos Hard, Lewiston; 3,619,174 11/1971 Fujii et al 75/60 Umesh Kumar Malhotra, 3,754,895 Ramachandran et al all of NY.

- 1 Primary ExaminerFelix D. Gruber [73] Asslgnee' 3:11: ggi Corporation New Attorney, Agent, or Firm-Eugene Lieberstein 22 H d: N 1, 1971 21 A N 1;: 454 [57] ABSTRACT 1 pp A process for making steel by refining a pxedetermined mass of molten metal having a composition US. Cl 444/] containing at least arbon and iron comprising; inject- [51] Int. Cl. CZld 3/04, G06f 15/46 ing oxygen into uch mass with a gaseous diluent at a F ield of Search 151-1, first dilution ratio and calculating from the activity co- 235/ 151.l2; 444/1 efficient of each element in the mass a theoretical equilibrium partial pressure of carbon monoxide for [56] Referen e Cite the oxidation reaction of each element; calculating the UNITED STATES PATENTS absolute maximum partial pressure of carbon monox- 3 04mm M972 ide; selecting the lowest partial pressure and calculat- 3II69'O5S M965 ing a new metal analysis for the mass; changing the di- 3,252.790 5/1966 lution ratio upon meeting a given criterion and repeat- 3,329,495 7/1967 ing the process until the carbon level of the mass has 3,377,158 4/1968 dropped to a predetermined level. 3,463,63l 8/1969 Vayssiere et al.....' 75/60 3,528,800 9/1970 Blum et al. 235/151.1 45 Claims, 8 Drawing Figures new our 01mm. coMPuToR AUTOMATIC. 26

ACTUATOR oxvcsu oxvesu +1: FLOW CONTROL wear GAS "3'? Fmw CONTROL PATENTEDJIIII 1 I I974 SHEET 2 BF 8 INITIAL INPUT DATA METAL ANALYSIS METAL WEIGHT TEMPERATURE 0XYGEN& ARGOH FLOW RATE CALCULATE P2 FOR THE CARBON nsoucnou 0;: F609 3102 9 Mn 0304 SELECT LOWEST 93,

LALCULATE P F08 Ar CALCULATE CARBON OXIDIZED AND METAL OXIDIZED CALCULATE CARBON F02 TO NEXT LOWER sE'rnuaJ TO NEXT HIGHER 551T" ING PRINT COMMAND AND TIME CONTINUE OLD FLOW RATES ELEMENT OF INTEREST PERIODICALLY PRINT TIME TEMPERATURE METAL ANALYSIS ESTABLISHED NEW SLAG ANALYSIS CALCULATE NEW M ETA L ANALYSIS NEW METAL WEIGHT LIMIT START NEXT CYCLE YES CALCULATE HEAT as REACTIOI.)

HEAT LOSS NEW TEMPERATURE INVENTORS EDWIN F. BAUER RT A. HARD ROGER N.DOKKEN UMESH K.MAIOTRA ATTO NEY PATENIEUJUN 1 1 m4 3816L720 SHEET 3 OF 8 A G E F CALCULATE SCRAP ADDITION PRINT COMMAND AND TIME.

7 canons ANY COMMAND TO SWITCH F0 AND FAr PRINT COMMAND AND TIME. W

c AP fi'i, 8, OVER RIDE OXYGEN 03m Ag? m 51 c gfizz FLOW INFORMATION 0:. 'NFORMAT'ON \NFORMATION g AUTOMATIC SSAI ACTUATOR v I I l OXYGEN FLOW ARGOH Fww SCRAP CONTROL CONTROL ADDITION INVENTORS FIG. 2 a EDWIN F. BAUER ROBERT A.HARD ROGER N. DOKKEN UMESH K.MAL. TRA

A O EY SHEET I UT 8 FRACTION OF THE AOD VESSEL WORKING LINING AFFECTED BY THERMAL VARIATIONS so WEIGHT OF WORKING LINING Xi0 L55,

I l I l l I 0 2 1" n m '1 0 INVENTORS EDWIN E BAUER ROBERT A. HARD ROGER N. DOKKEN UMESH K. MA ;1ITRA BY 1 M M [/Mzm 4Q. M Q I A RNEY PATENTEifi N m4 3.816.720

SHEET 8 0F 8 ACCEPTABLE TEMPERATURE BAND FOR REFINING 0F AOD HEATS 2000\ l 1950- T MAX. CURVE o I m 1900- a: 3-. Q: E 1850- T MIN. CURVE 2 LL p..-

l I l l l I l CARBON FIG.5

INVENTORS EDW/ N F BAUER ROBERT A. HARD ROGER N. DOKKEN UMESH K. MAL TRA PATENIEBJUW 4974 SHEET 8 BF 8 INPUT DATA METAL. ANALYSIS METAL. WEIGHT same? LOWEST Pg cmcuwra CARON Aw MEYM. amazes CALCULATE CARBON QMIDIZED IS CAREON N0 H\GHER THAN W CALCULM'E.

NEW warm. ANALYsts NEW METAL- wasswr W New LAfi waewr cmcwwm'a HAT 6F REACTION Hem" L055 NEW TNIPQRM'UE INVENTORS EDWIN F. BAUER PROCESS FOR THE DECARBURIZATION OF MOLTEN METAL BACKGROUND OF THE INVENTION This invention relates to a method of decarburizing molten metal in the refining of steel.

When steel is decarburized with oxygen, equilibrium is established among the metals, the carbon and oxygen at the particular temperature and pressure. Such equilibrium determines the extent to which carbon can be removed, without oxidizing metallics, chromium in particular, from the melt. It is now well recognized that the thermodynamic activity in the bath and the equilibrium established among the elements and the evolved gaseous atmosphere may be modified by dilution of the oxygen with an inert gas.

The limit to which the temperature may be elevated is based upon both practical and economic considerations relating to the effect of high temperatures on the life of the vessel lining as well as on the life of the tuyeres which are necessary for injecting the oxygen and inert gas directly into the melt from beneath its surface. On the other hand, dilution of the blowing oxygen with inert gas also involves exonomic considerations since it decreases the rate of steel production in proportion to the percent volume of inert gas introduced, not to mention the cost of the inert gas itself, which can be substantial.

Heretofore, the principle concern was to reduce the carbon content of the molten bath to a specified level while minimizing the oxidation loss of other elements. Other than the desire to prevent the rapid deterioration of the refractory, by careful temperature surveillance,

time the oxygen combines with carbon until sufficient pressure of carbon monoxide is generated that further combination of oxygenwith carbon is precluded by the preference of oxygen for that metallic element the reduction of whose oxide produces the lowest equilibrium partial pressure of carbon monoxide. From such assumption a theoretical mathematical model has been developed for calculating, at given selected intervals of time, the lowest equilibrium partial pressure of carbon monoxide from which a new mass composition, weight and temperature of the melt can be determined and from which, either automatically or through an operathe process was conducted at the expense of relatively 3 slow steel production rate and without regard to the quantity of inert gas utilized. The apparent lack of concern with economics was justified on the basis that it was impossibile to sample the bath periodically during a production run to determine the extent of metallic oxidation and actual bath temperature. It was considered far wiser to insure an acceptable end melt composition than to risk making an off grade heat by increasing the rate of production and conserving inert gas. At the very best, certain generalized guidelines were empirically established, through actual experience, for changing the dilution ratio at fixed time intervals and at an assumed temperature condition for a specific melt using a given refractory lining composition.

OBJECTS OF THE INVENTION It is an object of the present invention to provide a process for decarburizing a mass of molten metal in terms of achieving from knowledge of only the initial mass composition, weight and temperature, a predetermined end composition under the most favorable economic conditions.

It is another object of the present invention to provide a method for optimizing, in terms of combined performance and economics, the decarburization of a mass of molten metal to consistently yield final melt composition and temperature which conforms to a desired specification for such melt and at a relatively high rate of production.

SUMMARY OF THE INVENTION The process of the present invention is based on the theoretical assumption that for any small interval of tor, readjustments may be made in the dilution ratio to optimize the process. The mathematical model simulates the complicated natural processes occuring in the refractory lined vessel in terms of the reaction thermodynamics, heat and mass balances and thus predicts the process of a heat. Although, the mathematical calculations may be made by a person; in practice, it could not be done quickly enough to be useful, and therefore the present process requires a computer for its practical implementation.

In its broadest aspects, the present invention is a method for refining steel by controlling the decarburization of a predetermined mass of molten metal comprising carbon and iron contained within a refractory lined vessel having means for injecting oxygen and a diluting gas into the mass and adjustable gas flow control means for varying the flow rate of the gases; wherein the process comprises:

a. setting the adjustable control means to establish a first flow rate greater than zero for oxygen and a first flow rate for the diluting gas;

b. utilizing a computer to carry out the following sequence of steps:

1. calculating, a plurality of activity coefficients, from data corresponding to the initial flow rate setting of each gas respectively and from the initial composition, weight and temperature of the mass of metal, that define the thermodynamic activities of each element in the mass as a function of the composition of the mass, each coefficient reflecting the activity of each element in terms of the percentage of the element and the dependency of the element on the percentage of the other elements in the mass, with the activity of iron being equal to its mole fraction and wherein the activities of the oxides of each element have predetermined values;

2. calculating a theoretical equilibrium partial pressure of carbon monoxide for the oxidation reaction of each element by means of the coefficients at the given temperature;

3. calculating the absolute maximum partial pressure of carbon monoxide assuming all the oxygen injected reacted solely with the carbon;

4. selecting the lowest theoretical equilibrium partial pressure of carbon monoxide from (2) and comparing such with the absolute maximum partial pressure of carbon monoxide from (3);

5. make a determination as to whether the lowest theoretical equilibrium partial pressure has a magnitude greater than the absolute maximum partial pressure and from said determination if such theoretical partial pressure is greater than the absolute maximum partial pressure than further calculate using said first gas flow rates, the

amount of carbon oxidized and a new metal analysis and temperature for the composition.

6. make a determination as to whether the lowest theoretical equilibrium partial pressure has a magnitude of less than such absolute maximum partial pressure and from said determination if such theoretical partial pressure is of a magnitude lower than the absolute maximum partial pressure then further calculate from the theoretical equilibrium partial pressure of carbon monoxide and from said first gas flow rates the amounts of carbon and metal oxidized, a new metal analysis and temperature for the composition.

7. providing an indication of said new carbon content in said mass;

8. comparing the amount of oxidation of a specific signal element in the mass with a preestablished limit of oxidation for the specific element;

9. providing an indication for resetting the adjustable gas flow control means to increase the proportion of diluting gas should the amount of oxidation of the specific element be at least equal to the preestablished limit ofoxidation;

c. resetting the adjustable gas flow control means in accordance with the indication provided in step (9) d. repeating the sequence from step (1) at predetermined time intervals of less than 2 minutes until the carbon content indicated in step (7) has at least decreased to a predetermined level.

BRIEF DESCRIPTION OF THE DRAWINGS Advantages of the invention will be apparent from the following detailed description of the invention taken in conjunction with the following drawings in which:

FIG. 1 is a general schematic diagram of a decarburization system which utilizes the present invention;

FIGS. 2 and 2a combined represent the logic diagram of the preferred program for carrying out the present invention;

FIG. 3 is a graph, empirically established which defines the factor F for that portion of the refractory lined vessel which participates in the heat capacity of the system;

FIG. 4 is a graph showing the steady state temperature loss in the bath as a function of the size of the bath;

FIG. 5 is a graph, empirically established, which defines an acceptable band of bath temperature values for corresponding carbon contents.

FIG. 6 is a graph, empirically established, showing an alternative acceptable temperature range for corresponding carbon levels in the bath.

FIG. 7 is a simplified logic diagram of the preferred program shown in FIGS. 2 and 2a illustrating an alternative embodiment for changing the dilution ratio of oxygen in response to satisfying a given one of a number of different criteria.

DETAILED DESCRIPTION OF THE PROCESS Shown diagrammatically in FIG. 1 is a simplified refining system for decarburizing steel consisting of a refractory lined refining vessel 10 charged with a predetermined mass of molten metal 12. The starting composition of the mass may represent essentially iron and carbon where the refined end product is to be a low carbon iron. Alternatively, the starting composition for a stainless steel may be composed essentially of iron, carbon and chromium with nickel in some cases and with additional minor alloying constituents such as, for example, silicon and manganese. Moreover, for a nickel based alloy the starting composition may be composed essentially of carbon, iron, nickel. chromium, molybdenum and niobium. This invention is not to be construed as limited to any specific steel starting composition nor is the original percentage of any ingredient critical to the invention.

Oxygen is passed from a source (not shown) through an oxygen flow control 18 which regulates the flow rate of oxygen before injection into vessel 10. Likewise, a diluting inert gas is passed from a separate source (not shown) through a flow control 20 which regulates the flow of inert gas. The gas flow controls 18 and 20 respectively are conventional controllers which may be either automatically or manually operated as will as explained hereafter. The gases are combined in the mixing valve 16 and injected directly into the melt l2 preferably through a tuyere assembly 14. Other suitable gas injectors may be employed such as ceramic tubes, conduits, nozzles and the like so long as such means can withstand the bath temperatures involved without introducing undesirable contaminants. Separate means (not shown) may be used to control pressure in the respective flow lines.

The diluting gas may be any gas inert with respect to decarburization and preferably one selected from the group consisting of helium, neon, argon, krypton and xenon, or mixtures thereof. Nitrogen may also be used but with caution because of possible side effects. Argon is the most preferred gas.

The production rate as measured by the time required for decarburization is primarily affected by the temperature and gas flow rates and may be maximized by blowing the heat at the maximum total gas flow rate obtainable for the refining vessel and heat size, which is roughly 1,000 to 2,000 cubic feet per hour of total gas flow per ton of metal refining capacity for the vessel, and by keeping the flow rate of oxygen high relative to the flow rate of inert gas until the refractory is threatened by high temperature or until oxidation of the constituents in the melt other than carbon exceed predetermined levels. The temperature must also be kept above a minimum allowable temperature so that the heat can be finished, tapped and teemed into ingot molds without danger of solidifying prematurely. The present invention does not contemplate the use of temperature control elements or devices. Increases in temperature are due to the oxidation reactions occurring in the bath whereas reductions in temperature are due to absorption of heat by the applied inert gas, bath additions, and from the steady state heat loss occurring as a function of bath size, the heat capacity of the refractory and the vessel surroundings.

The method of the present invention whereby the gas flow controls 18 and 20 are operated, either manually or automatically, through the use of the digital computer to achieve optimum performance at maximum efficiency will now be described. The computer 22 is a general purpose computer which is programmed to operate in accordance with the logic diagram of FIGS. 2 and 2a representing the preferred embodiment of the present invention. Any programmer skilled in such art can rapidly prepare a computer program in the language receptive to the computer utilized to operate in accordance with such logic diagram. In order to carry out the program the computer 22 must have stored in memory data representative of the mathematical equations taught herein as well as data representative of Appendixes A, B, and C and FIGS. 3, 4, and 5 or 6 respectively. The only input information necessary for any given heat is the starting metal composition, temperature, metal weight, the selected initial oxygen and inert gas flow rates and the final desired carbon content. From steelmaking experience it is known that an oxygen flow rate more than ten times greater than the inert gas flow rate will be harmful to the vessel tuyere assem bly 14. Hence, as a practical matter the initial oxygen to inert gas flow rate should be less than 10 1.

Using the initial information the computer 22 calculates a plurality of coefficients that define the thermodynamic activities of each constituent element in the bath as a function of the composition of the bath. Employing such coefficients the computer 22 calculates the carbon monoxide partial pressure in equilibrium with carbon and the various metallic elements and their oxides. The reaction which produces the lowest equilibrium partial pressure of carbon monoxide is assumed to be favored and to proceed for some small increment of time. This increment of time is approximated for purposes of the present invention to be a period necessarily greater than zero but less than 2 minutes with a preferred range of between 3 to 30 seconds. Hence, the computer 22 without feedback can provide updated simulated sampling of the progress of a heat to determine whether and to what extend readjustment is to be made in the dilution ratio of oxygen and inert gas. The new conditions determined by the computer are automatically fed into memory and used in place of the previous conditions for the next iteration.

Upon injection, oxygen reacts immediately with some metallic species which, in turn, will react with dissolved carbon to form carbon monoxide according to the general reaction:

Where: M is any metallic element in the bath l2such as Fe, Cr, Mn, Si, etc; C is Carbon: O is Oxygen and x and y are integers which represent the chemical formula of the metallic oxide in question.

The equilibrium constant for this general reaction can be calculated from the following equation:

Where: K equilibrium constant; AF= standard free energy of the reaction; R is the gas constant and T= temperature ("K). The general chemical reaction represents reactions involving, for example, silicon, manganese, chromium, iron, nickel and/or other metallics which may be present in the mass. The appropriate equilibrium constants are calculated for each reaction in accordance with the above equation using the data from Appendix A as shown below.

APPENDIX A Standard Free Energies of the Various Reactions of Interest FeO-i-C=Fe+CO(g) 24.175 21.99T

The theoretical partial pressure of carbon monoxide in equilibrium with carbon and the other ingredients in the bath may now be calculated from the following equation using the reactions given in Appendix B as set forth hereinbelow:

K .rIu PCOE/ MIIIII y ac Where: a activity of the specific element; a activity of dissolved carbon; and Pag theoretical equilibrium partial pressure of carbon monoxide.

APPENDIX B ACTIVITIES OF THE VARIOUS COMPONENTS OF INTEREST a (activity of dissolved carbon) 5 %C X f f0 I fc fc fc fc fc log f 0.22 X %C log f 0.024 X %Cr log fd 0.012 X %Ni log f 0.106 X %Si log f 0.012 X %Mn a (activity of dissolved chromium) %Cr X f fCr I farfCr logf =-0.12 X %C log f O a (activity of dissolved silicon) %Si X f fSi fSi fSl logf X log f 0.112 x %Si a (activity of dissolved manganese) %Mn X f a (activity of iron) Mole Fraction where: f f J' f represent the activity coefficients of carbon chromium, silicon and manganese respectively, and f J B ,f f represent interaction parameters with supercripts a s C, Cr, Ni, Si, Mn [3 5 C, Cr and y i C, Si The activities appropriate for nickel based alloys can be determined by the methods described in, for example; H. Schenck and M. G. Frohberg, Steelmaking: The Chipman Conference, The M.I.T. Press, Cambridge, Mass, 1963, and H. Schenck and E. Steinmetz, Special Report No. 7, Stahleisen Sonderberichte, 1966, Publisher Stahleisen M.B.H., Dusseldorf.

The computer is programmed as shown in the logic diagram of FIG. 2 to select the lowest theoretical equilibrium partial pressure of carbon monoxide from which the computer then calculates the amount of carbon oxidized if all of the oxygen burns carbon and if not then the amount of carbon and metal oxidized. This is done by comparing the equilibrium partial pressure of CO with that which would be generated if all the oxygen burned carbon.

An alternate definition for the equilibrium partial pressure of CO is:

co co PS/FIG c Where: F C0 flow of CO generated (an unknown) F flow of inert gas introduced P, system pressure. If all of the oxygen burns carbon, the CO partial pressure can be calculated simply from the following equation:

Where: F flow of oxygen introduced. Now then, if the lowest selected P t P calculated from the equation directly above with all parameters known, then only carbon will be oxidized and in the amount C 2 F 12/387; Where: C pounds of carbon oxidized per unit time, 12 is the molecular weight of carbon and 387 is the molar volume in cubic feet at 70 F and one atmosphere pressure.

On the other hand, should the lowest selected P be less than P then the amount of carbon burned is calculated directly from Equation (1) as follows:

00" en/ 00+ FIG) PS multiplying out P F co m rearranging F (Ps P P F then and in terms of pounds of carbon per unit time reduces to C: (PCU'3 1P: Pcg FIG The remainder of the oxygen, then, must oxidize metal. This is derived from the following formula:

Where: M pounds of the metal oxidized per unit time; Wm molecular weight of the metal lb./lb. mole; V= valence of the metal. The amount of metal oxidized is subtracted from the original total metal weight and the total weight of the oxide resulting is added to the slag. The weight of the carbon oxidized is also substracted from the original metal mass and new composition percentages are calculated. The new composition and weight are stored in memory and used in place of the old composition and weight at the start of the next cycle.

In order to establish the most favored of all the possible reactions from thermodynamics, the temperature must be known. The heat generated or absorbed by the reactions is determined from thermodynamic data and from the heat of reactions given in Appendix C shown below. These are translated into temperature by dividing them by the combined heat capacities of the metal, the vessel and-the slag. The heat capacity of the metal in the case of steel is taken to be 10.5 lb.cal/lb. molC/54 lb./lb. mol X W lbs. metal 0.195W lb. cal/C Where: W= total metal weight in pounds. The heat capacity of the vessel (mainly the refractory) is taken to be Where: W, total working lining refractory weight in pounds and F empirical factor for the portion of the refractory lining which participates in the heat capacity of the system. The heat capacity of the slag is taken to be 0.55 W, lb. cal/C Where: W, slag weight in pounds. Thus, the heat ca pacity for the system at steelmaking temperatures is 'y 0.lW+ 0.27FW, 0.55), lb. cal/"C The factor F has been determined by trial and error methods of comparing actual heat data with program results. This factor is shown in FIG. 3 indicating the proper value to be used as a function of refractory weight. The refractory lining weight is known from the vessel design and the metal and slag weights are calculated at the end of each time interval.

APPENDIX C HEAT OF REACTION FOR THE VARIOUS OXIDATION REACTIONS Heat of Reaction Ar (25C) Ar (1,650C) AH= 8,072 lb.cal/lb. mol

and the temperature change due to each cubic foot of argon introduced is:

Argon A T= 20.9/A

Finally, the steady state heat loss of the system due to its surroundings, which occurs regardless of the rections within the vessel, is determined by means of empirical data presented in FIG. 4. These data result from extensive trial and error comparison of computer calculations and actual experience. The new temperature of the bath is calculated by summing the various contributions and is stored in memory in addition to the new composition and weight data to'replace the previous conditions at the start of the preparatory next succeeding time interval.

Once the computer has computed the new bath composition and temperature a determination is made as to whether the ratio of oxygen to inert gas should be adjusted. This determination is made for each cycle, i.e., each time the computer irterrogates the progress of the heat. As explained earlier, this should be done at intervals of less than 2 minutes apart to conform to the theoretical assumption of a single reaction predominating during such time interval. It should be understood that the computer interrogates the progress of a heat solely from initial conditions without feedback information from the vessel. Hence, the computer can be operated simultaneously during the actual decarburization of the molten mass or establish beforehand the optimum flow conditions for such decarburization. In addition, the computer can conduct the decarburization operation automatically or, alternatively, provide detailed instructions for controlling the operation manually through an operator. For a completely automatic operation, as shown in FIG. 1, the output of the computer 22 is fed into the actuator 30 which in turn regulates the gas flow controls 18 and 20 respectively. The automatic actuator 26 takes the digital output information from the computer and converts such information to appropriate analog electrical signal levels representing the new gas flow rates. The flow controllers 18 and 20 respectively, when set for automatic operation, will respond to the new levels thereby adjusting the ratio of gas flow in accordance with the computer command. Actuator 26 is manually preset to establish initial flow conditions. Maintaining accurate fractional ratios, however, between the flow of oxygen and the inert gas requires highly expensive and complex gas flow controller systems. In those instances where the tonnage of molten mass to be decarburized is small, a tradeoff is suggested between a totally optimized operation and the high cost of a totally automatic system. In such cases the gas flow controls may be simplified to permit merely whole number ratio adjustments such that, for example, the oxygen to inert gas flow ratios are varied only in a monotonically descending order such as: 4:1, 3:1, 2:1, 1:1, 1:2, etc. Here as well, manual operation by an operator in response to computer instructions would be preferred. The program would be run through the computer with the computer readout 24 providing detailed instructions as to when to make adjustment to the gas ratio. Instruction would be based upon selecting in sequence the next closest whole integer ratio although the computer can instruct the operator to skip the next closest ratio and go to a subsequent one or, of course, to maintain the same ratio.

The selection of each succeeding gas ratio is determined in the preferred embodiment illustrated in FIGS. 2 and 2a by comparing the amount of oxidation of a specific element during each time interval with a predetermined critical limit for such element. The selection may also be made based on alternative methods which will be discussed hereafter in connection with the simplified flow chart of H6. 7.

In the case of low carbon iron the specific element of interest is iron and when using the criterion of limited oxidation for switching, a limit of oxidation is established for each time interval by dividing the total number of time intervals by the maximum acceptable limit of oxidized iron after decarburization. Thelimit for each time interval need not be the same, e.g., a lower limit may be used for a first given number of intervals and the limit then raised for the remaining intervals.

In the case of nickel based alloys the specific element of interest may be any single element selected from the group consisting of chromium, molybdenum and niobium. The limit of oxidation would be established as in the case of low carbon iron.

in the case of iron based alloys the specific element of interest may be either chromium, molybdenum or manganese. The limit of oxidation would be established as in the case of low carbon iron.

In the case of stainless steel the specific element of interest is chromium and the limit of oxidation may be established as in case of low carbon iron or alternatively and preferably as taught in U.S. Pat. No. 3,046,107 issued July 24, 1962, in the name of Edward C. Nelson et al. As taught in such patent the volume oxygen percent injected in the bath is dictated by the following relationship which is intended to apply for steel containing 3 to 30 percent chromium:

Percent 0 PercentO2=Amount 02l02+lnert Gas- ,1 5

Where: percent Cr is the lower limit of chromium content in weight percent at the end of a time interval; T is the temperature of the melt in K at the end of a time interval; percent C is the percent of carbon in the melt at the end of a time interval; and percent oxygen is the oxygen volume percent for such time interval.

The above formula may also be used to directly calculate the percent 0 that should be employed for the next time interval. This may be done merely by using for percent Cr the calculated chromium content in weight percent at the end of the time interval. This would then determine whether the preceding flow of oxygen should be changed and to what extent.

In order to converse refractory the temperature of the bath, for each given time interval, should not be permitted to exceed a predetermined maximum value. Likewise, to prevent premature solidification the temperature of the bath should not be permitted to fall below a predetermined minimum value. Hence, the computer 22 following a subroutine in the logic flow chart of FIG. 2 and 2a will override any command to alter the flow rate of oxygen and diluting gas should the temperature fall below a preestablished acceptable temperature band. A functional relationship exists between the carbon content of a heat and the temperature. An acceptable temperature band corresponding to discreet carbon levels can therefore be established for any given heat. FIGS. 5 and 6 are two examples of empirically established acceptable temperature bands for stainless steel. This invention should not be construed as limited to any specific acceptable temperature band.

It the temperature for a given carbon level lies below the acceptable band, the computer will direct the operator to ignoreany instructions to effect a change in the gas flow control in response to excess oxidation of the element of interest and instead to cause an increase in temperature by heating the molten mass. Such instruction may take the form of command to reset the gas flow control to establish a given flow rate, which may be the same as the existing flow rate or one with a higher proportion of oxygen, in order to prolong or intensify the exothermic reactions occurring in the bath during such time interval thus providing additional heat. The process cycle will then be renewed and the new temperature again evaluated to determine whether it lies within the acceptable band. This will continue until the temperature has risen to or above the acceptable lower limit. Operation of the process in such manner in order to regulate temperature requires however, a tradeoff in increased metallic oxidation. An altemative method for increasing the temperature of the molten bath is to add a deoxidizer to the mass such as by the addition of one or a combination of the following: silicon, aluminum, manganese, ferrosilicon, ferrochromium and ferromanganese. The addition of a deoxidizer may be used in combination with the first procedure to achieve a quicker response.

if the temperature for a given carbon content lies above the acceptable band, the computer 22 will direct the operator to cool the heat. The preferred method of cooling the heat is by the addition of scrap. The added scrap material must obviously be compatible with the molten metal composition. in order to avoid piecemeal scrap addition until the bath has cooled to a point in the acceptable band it is preferable that the operator add scrap in accordance with the following formula:

S (scrap addition) ]T'' maa' mtn/ l Where: T temperature of bath in K; T maximum acceptable temperature at the carbon level of the bath in K; T minimum acceptable temperature at the carbon level of the bath in "K and C is a constant. Other fluid coolants may also be used such as steam and carbon dioxide. Another alternative method of cooling, which may be used in combination with the above, is to inhibit the flow of oxygen and diluting gas for a predetermined period of time.

Although temperature override as described above is desirable it is not critical in practicing the present invention. The process may be carried out to achieve optimum economics without holding the temperature to within a predetermined temperature band. When the process is used without compensating for temperature the program illustrated in the logic diagram of FIGS. 2 and 2a effectively reduces to that shown in FIG. 7 where the block labeled criterion for changing gas flow control would be equivalent to the block labeled element of interest oxidized more than established limit. Moreover, it is possible for some common starting metal composition to determine starting flow conditions which for such compositions will not cause the temperature to fall outside acceptable limits.

The selection of each succeeding gas ratio may be determined as stated hereinbefore by alternative methods to yield substantially the same result as the preferred method described above. These alternative mehtods are based on information regarding the carbon content, the temperature and the processing time respectively which are regularly monitored and stored by the computer. Temperature override is not necessary for these methods which are schematically described in the flow sheet shown in FIG. 7. Each method is based on a different criterion for changing the gas flow control. In the method based on carbon content, discrete gas ratios are predetermined to correspond to carbon content ranges. Thus, when the computer finds that the carbon content has progressed from one range into a different range, it will render an indication to that effect and signal the operator to switch into a corresponding gas ratio in the flow control sequence or may perform this function automatically. In the method based on temperature, discrete gas ratios are predetermined to correspond to certain bath temperature ranges. Thus, when the computer finds that the temperature has progressed from one temperature range into the next, it will render an indication to that effect and signal the operator to switch to the corresponding ratio in the gas flow control sequence. In the method based on time, discrete gas ratios are predetermined to correspond to certain time periods relating to periods of progress in the refining operation. Thus, after a certain time period has elapsed, the operator is ordered to switch to the next gas ratio in the sequence. In each of these methods, the computer issues a final printout indication that the refinement is complete based upon the calculation that the carbon content is below a predetermined level. However, operation of the process by such alternative methods may not render the optimized economics offered by the preferred embodiment.

Appended hereto is a number of illustrative examples of steelmaking computer readout instructions for manual control of the decarburization operation as well as a printout of a complete program for directing the operation of the computer.

EXAMPLES Grade: Pure iron, 15 tons Instructions: (Time)(min)* 0, Flow rate- 0, Flow Ar F 2! R ate. (NCEH) Analyses: %C %Cr %Si %Mn TempF Starting 0.44 0,2l 0.30 0.05 2710 Computed Final 0.001 0.22 0.00 0.00 3064 Actual Final 0.001 0.10 0.00 0.01 3082 Actual Treatment Time=34 min. Gas ratios were switched at the times (min) mentioned above to obtain the aim carbon.

Grade: 436,17 tons Actual Treatment Time=50 min. Gas ratios were switched at these mentioned time in minutes to obtain the aim carbon.

Grade: 321. 23 tons Instructions: (Time)(min)' 0, Flow Rate 0, Flow Ar Flow Rate Rate (NCFH) Analyses: 9K. 20 945i 2M5 TempF Starting 0.52 19.30 0.28 L 2768 Computed Final 0.05 l6.30 0.08 0.50 3207 Actual Final 0.05 l6.45 0.09 0.60 3200+ Actual Treatment Time=44 min. Gas ratios were switched at these mentioned time to obtain the aim carbon.

PROGRAM 23 SHORT TON STAINLESS STEEL AR Argon flow rates in NCFH during various steps (NCF is 70F and 1 atm) (7,600, 8,850, 17,700) OX Oxygen flow rates in NCFH during various periods (22,800, 17,700, 8,850)

ARS Argon flow rate in NCFH through shroud 1,200/two tuyeres) HL Steady state heat loss expressed as C /min.

AP=Argon pressure in p.s.i.g.

OP=Oxygen pressure in p.s.i.g. To be established OT=Oxygen inlet temp "F in the plant AT=Argop inlet temp F FE percent lron SIOZ Oxide of Si CR 3 04 Oxide of Cr MNO Oxide of Mn FEO Oxide of Fe T02 Total cubic ft of Oxygen TA Total cubic ft of Argon Z2 Time of oxygen blow AFEO Activity of FeO CRl percent Start Chromium AMNO Activity of MnO AR(l) Argon correction for temperature and pressure OX(l) Oxygen correction for temperature and pressure Z Calculation time (3 secs) PT Print time in min. ASIO, Activity of SiO AQR04= Activity of Cr O SlW Weight of Si in metal in lbs.

MNW Weight of Mn in metal in lbs.

CRW Weight of Cr in metal in lbs.

FEW Weight of Fe in metal in lbs.

NlW Weight of Ni in metal in lbs.

CBR Carbon removed as CO in lbs.

CRR Chromium oxidized in lbs.

SlR Silicon oxidzed in lbs.

MNR Manganese oxidized in lbs.

FER lron oxidized in lbs.

TM Total moles of metals XFE Mole fraction of iron AKCR Equilibrium constant of Chromium and carbon reaction ACR Activity of chromium AC Activity of carbon PCR Partial pressure of CO with respect to Chromium reaction AKSI Equilibrium constant of Silicon and carbon reaction AS] Activity of Silicon P5! Partial pressure of CO with respect to Silicon reaction AKMN Equilibrium constant of Mn and carbon reaction AMN Activity of Mn PMN Partial pressure of CO with respect to Mn reaction AKFE Equilibrium constant of Fe and C reaction AFE Mole fraction of Fe PFE Partial pressure of CO with respect to Fe reaction A Argon in ft lmin.

0 Oxygen in f t /min.

ACTOX percent Oxygen in mixture of oxygen and argon L Lb moles of metals CALOX percent Oxygen calculated by Nelson Griffing formula ECB Temp rise due to carbon oxidation in C ECR Temp rise due to Cr oxidation in C EMN Temp rise due to Mn oxidation in C E8] Temp rise due to Si oxidation in C EFE Temp rise due to Fe oxidation in C AHL Temp drop due to heating of Argon in C/min.

E Temp rise of system in C TMIN & TMAX define the temperature band SW Weight of slag 8510; SiO in slag SFEO FeO in slag Cl' O in Slag SMNO MnO in slag SCRAP Amount of scrap to cool the bath S Amount of scrap at any instant CS Carbon in scrap in lbs.

SIS Silicon in scrap in lbs.

MNS Manganese in scrap in lbs.

CRS Chromium in scrap in lbs.

FES Iron in scrap in lbs.

AODNG 1085 1699 l 1 l6 1 125 11135 1 1 M 1,160 l 179 1 l8?) 1 196 126B 121% 1.220 123% [24G 1255 1266 1272 128% 296 13152] 1310 1320 1330 134$ 185% 1.360 1376 1380 1396 1466 1416 1429 143?! 144% 1459,! 146% 1475 148M 149% 1562 [51% 152B 154% 1550 1560 I576 1586 CONTINUED AODNG CONTINUED [FCC-'61 15; l l: l l

I l CONZ'INUE IF(TIMB-PT)33: 34; 34 33 CONTINUE GO T0 17 34 TIME=G GO TO 17 STOP END

19 IF(C-- I 5) 1'. 17, 17

GO TO 9'7 97 CONTINUE 3 RETURN END What is claimed is: l. A process for making steel by refining a predeter a. setting said adjustable control means to establish a first flow rate greater than zero for said oxygen and a first flow rate for said diluting gas;

b. utilizing a computer to carry out the following sequence of steps:

1. calculating a plurality of activity coefficients, from data corresponding to the initialflow rate setting of each gas respectively and from the initial composition, weight and temperature of said mass of metal, that define the thermodynamic ac- SLAG CR304 FEO MNO CAO PRINT 38:94: S'vb 5012304; SFEO; SMNO; CAO: 55102 38 FORMATCI; 16; F7- hFS. b131 6. I)

SUBROUTINE Fri-(P102181 A, T, 0BR) IF(P- 2*02*l.3*B/(2*O2+A)))[9,2,2

PRINT [8; Z2; l -8*T-460);C: CR; FE:SI:MN:(P*76G) l8 FORMAT(F6. I; 16; 1 .7.3; 2F7.2; 21 6.2; I5)

tivities of each element in said mass as a function of the composition of said mass, each coefficient reflecting the activityof each element in terms of the percentage of said element and the dependency of said element on the percentage of the 4. comparing the absolute maximum partial pressure ofcarbonmonoxidefrom (3) with the ,lowesttheo-

Patent Citations
Cited PatentFiling datePublication dateApplicantTitle
US3046107 *Nov 18, 1960Jul 24, 1962Union Carbide CorpDecarburization process for highchromium steel
US3169058 *Nov 18, 1960Feb 9, 1965Union Carbide CorpDecarburization, deoxidation, and alloy addition
US3252790 *Nov 18, 1960May 24, 1966Union Carbide CorpPreparation of metals and alloys
US3329495 *Sep 22, 1964Jul 4, 1967Yawata Iron & Steel CoProcess for measuring the value of carbon content of a steel bath in an oxygen top-blowing converter
US3377158 *Apr 28, 1965Apr 9, 1968Jones & Laughlin Steel CorpConverter control systems and methods
US3463631 *Nov 30, 1964Aug 26, 1969Siderurgie Fse Inst RechMethod and arrangement for determining the oxidation reactions during refining of metals
US3528800 *Feb 14, 1966Sep 15, 1970Leeds & Northrup CoOptimized blowing control for basic oxygen furnaces
US3561743 *Oct 17, 1967Feb 9, 1971Gen ElectricUse of stack gas as oxygen potential measurements to control the bof process
US3565606 *Jun 27, 1967Feb 23, 1971Westinghouse Electric CorpMethod for controlling carbon removal in a basic oxygen furnace
US3594155 *Oct 30, 1968Jul 20, 1971Allegheny Ludlum SteelMethod for dynamically controlling decarburization of steel
US3614682 *Jun 22, 1970Oct 19, 1971Firestone Tire & Rubber CoDigital computer control of polymerization process
US3619174 *Jun 6, 1969Nov 9, 1971Sumitomo Metal IndMethod for controlling the carbon content in and/or the temperature of the steel
US3621217 *Sep 9, 1969Nov 16, 1971Gulf Research Development CoSpecial-purpose process control computer
US3754895 *Jan 27, 1971Aug 28, 1973Allegheny Ludlum Ind IncProcess for decarburization of steels
Referenced by
Citing PatentFiling datePublication dateApplicantTitle
US4113469 *May 2, 1977Sep 12, 1978British Steel CorporationRefining molten metal
US4187541 *Jun 8, 1978Feb 5, 1980Institut Kibernetiki Akademii Nauk Ukrainskoi SsrDigital analyzer for determining liquidus temperature of metals and alloys
US4190888 *Jun 8, 1978Feb 26, 1980Institut Kibernetiki Akademii Nauk Ukrainskoi S S RDigital device for determining carbon content in iron-carbon melts
US4198679 *Jun 28, 1978Apr 15, 1980Institut Kibernetiki Akademii Nauk Ukrainskoi SsrMethod and device for discriminating thermal effect of phase transformation of metals and alloys in the process of their cooling
US4512802 *Mar 21, 1983Apr 23, 1985Nippon Yakin Kogyo Kabushiki KaishaComputer optomized
US4647019 *Apr 1, 1986Mar 3, 1987Union Carbide CorporationVery small refining vessel
US4708738 *Nov 18, 1986Nov 24, 1987Union Carbide CorporationMethod for refining very small heats of molten metal
US4711430 *Apr 1, 1986Dec 8, 1987Union Carbide CorporationSide-injected metal refining vessel and method
US5327357 *Dec 3, 1991Jul 5, 1994Praxair Technology, Inc.Method of decarburizing molten metal in the refining of steel using neural networks
US5442570 *Dec 15, 1993Aug 15, 1995Nippon Steel CorporationMethod of controlling heat input to an alloying furnace for manufacturing hot galvanized and alloyed band steel
US5522915 *Sep 2, 1994Jun 4, 1996Heraeus Electronite Japan, Ltd.Method and apparatus for sequentially and continuously determining concentrations of carbon, hydrogen, and nitrogen in molten steel, and method and apparatus for rapidly determining trace amounts of carbon in molten steel
US6093235 *Oct 14, 1996Jul 25, 2000Mannesmann AgProcess for decarbonising a high-chromium steel melt
US6830606 *Jul 1, 2002Dec 14, 2004Nippon Steel CorporationMethod for decarbonization refining of chromium-containing molten steel
US6854573Oct 25, 2001Feb 15, 2005Lord CorporationBrake with field responsive material
DE3311232A1 *Mar 28, 1983Oct 11, 1984Nippon Yakin Kogyo K KMethod for decarburising of molten metal
DE4324528C2 *Jul 21, 1993Dec 12, 2002Siemens AgRecheneinrichtung zur Bestimmung von Entschwefelungszusätzen für eine Stahlschmelze und Stahlschmelze mit Entschwefelungszusätzen
EP0160375A2 *Mar 15, 1985Nov 6, 1985Allegheny Ludlum Steel CorporationSystem and method for producing steel in a top-blown vessel
EP0545379A1 *Dec 2, 1992Jun 9, 1993Praxair Technology, Inc.Method of decarburizing molten metal in the refining of steel using neural networks
WO2002075003A2 *Mar 21, 2002Sep 26, 2002De Angelis VinicioArgon oxygen decarburisation converter control method and system
Classifications
U.S. Classification75/382, 266/83, 700/29, 75/557, 266/80, 700/68, 700/36, 700/271, 75/384
International ClassificationC21C5/34, C21C7/072, C21C5/30, C21C7/068
Cooperative ClassificationC21C7/068
European ClassificationC21C7/068
Legal Events
DateCodeEventDescription
Dec 26, 1989ASAssignment
Owner name: UNION CARBIDE INDUSTRIAL GASES TECHNOLOGY CORPORAT
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST.;ASSIGNOR:UNION CARBIDE INDUSTRIAL GASES INC.;REEL/FRAME:005271/0177
Effective date: 19891220
Oct 8, 1986ASAssignment
Owner name: UNION CARBIDE CORPORATION,
Free format text: RELEASED BY SECURED PARTY;ASSIGNOR:MORGAN BANK (DELAWARE) AS COLLATERAL AGENT;REEL/FRAME:004665/0131
Effective date: 19860925
Jan 9, 1986ASAssignment
Owner name: MORGAN GUARANTY TRUST COMPANY OF NEW YORK, AND MOR
Free format text: MORTGAGE;ASSIGNORS:UNION CARBIDE CORPORATION, A CORP.,;STP CORPORATION, A CORP. OF DE.,;UNION CARBIDE AGRICULTURAL PRODUCTS CO., INC., A CORP. OF PA.,;AND OTHERS;REEL/FRAME:004547/0001
Effective date: 19860106